National Academies Press: OpenBook
« Previous: 3 Problem Formulation: Sources, Settings, and Ecological Receptors
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

4

Fate, Transport, and Potential Exposure in the Environment

This chapter comprises six sections. The first section includes a brief overview of how exposure information is used in risk assessment. The second section provides an overview of fate characteristics of the UV (ultraviolet) filters as a foundational understanding of these characteristics is necessary for understanding exposure regimes and for interpreting the available environmental measurements that may be used to evaluate exposures in time and space. Recognizing the need to consider the physical features of environments that influence exposures, and that those features vary across systems, the third section provides an overview of the range of hydrological conditions that influence the fate and transport of introduced UV filters. The fourth section reviews the data on environmental exposures with respect to measured concentrations in water and sediment, and links this back to environmental settings and the fate characteristics of the UV filters. The fifth section covers analytical chemistry considerations. The final section summarizes findings and conclusions from the chapter.

HOW EXPOSURE INFORMATION IS USED IN ECOLOGICAL RISK ASSESSMENT

Ecological risk assessments (ERAs) utilize information on exposure (i.e., concentrations in water or sediment) and effects (i.e., toxicity or hazard) to characterize the potential for risks to ecological receptors (e.g., animals, plants, microbes). The ERA framework for the assessment is set in “Problem Formulation” (described in Chapter 3) and that guides the types of information that are sought for evaluating exposure, and for establishing quantitative exposure estimates. Depending on whether the ERA is a screening or higher-level assessment, exposure estimates may be conservative and high bounding (for screening level), or may reflect measured or modeled exposures with spatial and temporal variation relevant to the receptors (for higher level). Because exposure information must be coupled with effects information to assess risk potential, several considerations are important. These include alignment of units of exposure (usually as internal or external concentrations), considerations of time scales (short term versus longer term), and understanding locations in relation to releases and receptors. The lattermost consideration is influenced by the ecology of the receptors at relevant life stages that determines where they may be located in relation to a source, as described in Chapter 3. The selected ERA approach depends on the assessment program (i.e., screening level versus higher level), availability of information, and uncertainties associated with adequately representing exposure regimes.

Because of the spatial and temporal variations associated with sources and releases of UV filters to the aquatic environment, a combination of fate considerations, modeling, and measurements are appropriate for estimating

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

and characterizing exposures. Prospective environmental exposure models will typically give bounding estimates and can address UV filters for which there are little measurement data. Where available, measurements of UV filters in water, sediments, and biota can indicate how current releases of UV filters are manifested as exposure levels. However, measurements in water are highly variable and may not be useful in defining what an organism’s average exposure is over time, challenging comparisons to toxicity studies where concentrations are consistent or reported as an average over a certain time frame. Uncertainties associated with models and measurements need to be recognized and addressed in a risk assessment.

In all of the exposure scenarios described in Chapter 3, Problem Formulation, the ultimate exposures for any receptors will be significantly impacted by the interaction of the UV filters with the natural environment (environmental fate) of the UV filter chemicals once released. In turn, the relevant fate processes are largely driven by the physico-chemical properties of chemicals. The importance of these properties are well recognized in the conduct of environmental risk assessments and explains why they are generally required prior to conducting risk assessments.

FATE CHARACTERISTICS OF UV FILTERS

The physico-chemical properties of chemicals largely drive environmental behavior (i.e., fate and effects) in ecological systems. These properties, in combination with environmental factors, define where chemicals may reside, how long they will persist, and whether they will degrade into other products. A chemical simply detected in the environment does not directly equate to an effect on organisms. Effects depend on exposure levels (concentration), duration (persistence), the extent and nature of interaction between the chemical and the organism (bioavailability) and hence uptake by the organism through one or more exposure pathways (i.e., absorption from water, uptake via filtration, and/or uptake through diet or indirect ingestion of sediment, organic matter, and/or other organisms). Even the uptake and presence of a chemical in an organism does not equate to an effect; they may be excreted without harm. An effect requires transport to a biological receptor and, in some cases such as photoreactive chemicals, further interaction with the environment. Once released to the environment, chemical contaminants like UV filters can partition into different environmental compartments. For example, volatile chemicals released into a water body could partition out of the water body and into the atmosphere, thereby reducing exposures for aquatic organisms. However, chemicals may also accumulate in phases (e.g., suspended organic particles including phytoplankton as well as sediments) or at interfaces between phases (e.g., air–water interface, water-surface microlayer, and possibly external mucus layers of corals) that could lead to elevated exposures at those locations. In addition, UV filter chemicals may also partition from the water or sediment into biota (absorption) through membranes such as gills, permeable skin, or the intestinal system; the bioaccumulation of UV filters into biota is discussed in Chapter 5. The relevance of these types of exposure processes, in so far as they have been documented in the literature for inorganic and organic UV filters, are described in the following sections.

Physico-Chemical Parameters of Inorganic UV Filters Impacting Fate

Inorganic UV filters are metal oxide particles that tend to aggregate with other inorganic and organic particles or colloids in water, including bacteria, algae, clays, and mineral oxides (Keller et al., 2013). They are used in both nano (< 100 nm) and macro sized forms in sunscreens (there is no distinction in their regulation by FDA). While the exact proportion of sunscreens that now use nanoparticles is unclear, there is a known increase in their use. Researchers have found a predominance of nano sizes in sunscreen formulations under study and nano sizes can also be found mixed in with microsized forms (Labille et al., 2010; Lu et al., 2015; Nohynek et al., 2007; Smijs and Pavel, 2011). Inorganic UV filter particulates may settle out of water columns individually or as aggregates. While electrostatic interaction and hydrodynamic models accurately describe aggregation rates in model or synthetic waters, most environmental fate and transport models include more electrostatic empirical parameters or affinity coefficients to quantify the efficiency of aggregation (i.e., number of collisions between particles that lead to formation of larger aggregated particles) (Geitner et al., 2019; Praetorius et al., 2014). A variety of modeling platforms for the environmental fate of nanomaterials and other particle sizes have been developed at watershed to

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

regional scales (Liu and Cohen, 2014). Figure 4.1 illustrates the general mechanisms and processes that influence titanium dioxide (TiO2) or zinc oxide (ZnO) fate in water. Probabilistic modeling approaches have been undertaken, including for TiO2 from multiple sources (Zheng and Nowack, 2021). Such approaches have estimated (for 2016) that 50 percent of the nanosized TiO2 particles entering wastewater came from the portion of nanosized TiO2 particles in pigments. This approximation highlights the importance of surface coatings in controlling the release and fate of TiO2 particles into the environment. Models have been developed to evaluate the fate of nanomaterials in the environment. These include material flow analysis models (MFAMs), multimedia compartmental models (MCMs), and spatial river/watershed models (SRWMs) (Baalousha et al., 2016; Nowack, 2017; Suhendra et al., 2020; Williams et al., 2019). These models are fairly robust but have only been validated in a limited number of freshwater and estuary aquatic environments (e.g., Meesters et al., 2014; Praetorius et al., 2012; Pu et al., 2016).

Studies of existing environmental exposure models for engineered nanomaterials (ENMs) have been conducted. The sub-micro fate and transport models covered in a review by Suhendra et al. (2020), as well as work by Reed et al. (2012), resulted in a number of high-level outcomes. For example, ZnO undergoes dissolution to release zinc ions to a greater extent than TiO2, which is relatively insoluble in water (Suhendra et al., 2020). This outcome builds on the work of Reed et al. (2012), who found that the dissolution rate of ZnO is impacted by the particle size and morphology as well as the media of the experiment or environment. When 100 mg of ZnO 25 nm was dissolved in nanopure water, Roswell Park Memorial Institute medium, Dulbecco’s modified Eagle’s medium (DMEM), and moderately hard water, the impact of media was very important to dissolution. Dubelcco’s media contained 34 mg/L zinc compared to nanopure water which had 7.40 mg/L, and suggests that in biological

Image
FIGURE 4.1 Inorganic UV filter fate processes in surface waters. SOURCE: Adapted from Suhendra et al., 2020.
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

systems significant dissolution may happen rapidly. When another particle shape that had less aggregation was considered, dissolution was instantaneous in DMEM media (Reed et al., 2012). Once zinc ions dissolve from ZnO, the fate of zinc ions in water columns, sediment or organisms become unrelated to the fate of ZnO itself. Once dissolved from ZnO, zinc ion can complex with dissolved ligands (e.g., hydroxide ion, sulfide ion, natural organic matter, etc.) (Cheng and Allen, 2006; Holmes et al., 2020; Miao et al., 2010; Figure 4.1). However, zinc hydroxide (Zn(OH)2(s)), zinc sulfide (ZnS(S)), or carbonate (ZnCO3(s)) are fairly insoluble in water at common pH and temperature levels in surface waters (Ma et al., 2013b; Miao et al., 2010; Wu et al., 2019a,b). However, at low pH levels (< 5) zinc ion is fairly soluble (Miao et al., 2010; Youn and Choi, 2022). Zinc ions can adsorb to soil, sediment, and suspended particles in water (Chen et al., 2019; Waalewijn-Kool et al., 2013; Walaszek et al., 2018). Overall, pH emerges as one of the most critical factors controlling zinc ion solubility in water, with higher soluble zinc concentrations at lower pH levels.

Suhendra et al. (2020) and Ma et al. (2013b) also found that ZnO can develop insoluble-coating layers on the surface (e.g., sulfides) that decrease the potential dissolution of zinc; this is most likely to happen in anaerobic conditions, such as sediments. Engineered coatings of silica, alumina, or polymers (PEG) on ZnO or TiO2 UV filters can degrade and transform over time (Philippe et al., 2018). Natural organic matter and other dissolved chemicals can adsorb onto the surface of inorganic UV filters, changing their surface charge characteristics. Inorganic UV filters can aggregate with each other (homoaggregation) or with other suspended solids in a water column or in sediment/pore waters; aggregation leads to larger particles that can settle out of the water column. UV filters with net negative surface charges, which are impacted by degradation of engineered surface coatings and/or interactions with NOM or other solutes in water, tend to aggregate more slowly and stay longer in the water column. Sunlight can photocatalytically react with inorganic UV filters in the water column and produce reactive oxygen species, which may initiate toxicity to organisms or transformation of chemicals in the water column. The lifetime of inorganic UV filters in the water column is determined by their size, density, surface charge, and tendency to aggregate with other particles in water, which then promotes settling to sediments.

Physico-Chemical Parameters of Organic UV Filters Impacting Fate

The physico-chemical properties of the organic UV filters necessary for conducting environmental risk assessments include solubility, volatility, acid dissociation constants (if appropriate), and affinity for solid phases (often approximated as hydrophobicity). Some of these properties can be estimated from other, similar chemicals or predicted from computational chemistry models employing quantitative structure activity relationships (QSARs) which, in the context of environmental partitioning, are often referred to as linear free-energy relationships (LFERs; Schwarzenbach et al., 2016). Table 4.1 summarizes the available data with respect to these fundamental properties for the UV filter chemicals. Advanced computational models have been developed to predict more accurate partition distribution coefficients; these rely on a small set of chemical descriptors that encapsulate all relevant interactions of organic chemicals with different phases. These chemical descriptors, known as Abraham’s parameters, do not appear to be readily available for many of the organic UV filter chemicals (Ulrich et al., 2017). However, a recent study (Sobanska, 2021) suggests that these computational approaches may enable more accurate predictions of environmental fate parameters for a number of the organic UV filters of interest. Figure 4.2 illustrates the processes influencing the fate and disposition of organic UV filters that govern exposure.

Solubility

The solubility of organic UV filters in pure water can be used as a baseline predictor for other more environmentally relevant fate parameters. A chemical compound’s aqueous solubility can also be used to estimate relevant dosing levels in aquatic toxicity tests, as it is uncommon that environmental exposure exceeds solubility except in certain scenarios such as pure chemical spills. For organic chemicals, solubility in pure water can generally be considered an upper limit estimate, as solubility in waters of high ionic strength (i.e., brackish or marine waters)

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

TABLE 4.1 UV Filters Used in the United States and Their General Phyisco-Chemical Properties

UV Filters CAS # Canonical SMILES Solubility in Ultrapure Water (mg/L) Henry’s Constant (atm-m3/ mole) Octanol Water Partition Coefficients (Log Kow) Vapor Pressure (Pa) pKa Log Dow @ pH 8.1 (Assuming No Anionic Partitioning)
Aminobenzoic acid 150-13-0a C1(C(O)=O)=CC=C(N)C=C1 6,110a 1.5×10−10b 0.96a 0 @ 25°Ca 4.65a –2.49
Avobenzone 70356-09-1a O=C(c1ccc(cc1)C(C)(C)C)
CC(=O)c2ccc(OC)cc2
0.027a 2.0×10−10b 6.1a 7.7×10−8 @ 25°Ca
Cinoxate 104-28-9b CCOCCOC(=O)/C=C/
C1=CC=C(C=C1)OC
127.4b 5.1×10−9b 2.65b 4.5×10−2 @ 25°Cb
Dioxybenzone 131-53-3a COC1=CC(=C(C=C1)C(=O)
C2=CC=CC=C2O)O
0.013a 2.0×10−9b 2.33a 2.6×10−5 @ 25°Cb 6.78c 0.99
Ecamsule 92761-26-7a CC1(C2CCC1(C(=O)
C2=CC3=CC=C(C=C3)
C=C4C5CCC(C4=O)(C5(C)C)
CS(=O)(=O)O)CS(=O)(=O)O)C
> 600,000a 2.1×10−23b -1.84a 8.4×10−21 @ 25°Cb
Ensulizole 27503-81-7a C1=CC=C(C=C1)
C2=NC3=C(N2)C=C(C=C3)
S(=O)(=O)O
109a 1.3×10−14b -1.42a 9.8×10−13 @ 25°Cb
Homosalate 118-56-9a CC1CC(CC(C1)(C)C)OC(=O)
C2=CC=CC=C2O
0.0912a 1.9×10−5b 6.27a 1.5 ×10−2 @ 25°Ca 8.1b 5.859
Meradimate 134-09-8b CC1CCC(C(C1)OC(=O)
C2=CC=CC=C2N)C(C)C
0.074b 6.9×10−8b 6.28b 5.0 ×10−4 @ 25°Cb
Octinoxate 5466-77-3a CCCCC(CC)COC(=O)
C=CC1=CC=C(C=C1)OC
0.051a 1.8×10−6b > 6a 30 @ 154°Ca 3.0×10−3 @ 25°Cb
Octisalate 118-60-5a CCCCC(CC)COC(=O)
C1=CC=CC=C1O
0.074a 3.3×10−5b 5.94a 5.2×10−2 @ 25°Ca 8.13a 6.074
Octocrylene 6197-30-4a CCCCC(CC)COC(=O)C(=C
(C1=CC=CC=C1) C2=CC=CC=C2)C#N
0.04a 3.0×10−9b 6.1a 4.2×10−7 @ 25°Ca
Oxybenzone 131-57-7a COC1=CC(=C(C=C1)C(=O)
C2=CC=CC=C2)O
6a 1.5×10−8b 3.45a 1.1×10−3 @ 25°Ca 7.1c 2.409
Padimate O 21245-02-3a CCCCC(CC)COC(=O)
C1=CC=C(C=C1)N(C)C
0.11a 4.0×10−6b > 6.2a 5×10−4 @ 25°Ca
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
UV Filters CAS # Canonical SMILES Solubility in Ultrapure Water (mg/L) Henry’s Constant (atm-m3/ mole) Octanol Water Partition Coefficients (Log Kow) Vapor Pressure (Pa) pKa Log Dow @ pH 8.1 (Assuming No Anionic Partitioning)
Sulisobenzone 4065-45-6a COC1=C(C=C(C(=C1)O)C
(=O)C2=CC=CC=C2)S(=O)
(=O)O
300,920a 7.0×10−15b 0.515a 1.79×10−9 @ 25°Ca 0.022a –7.563
Titanium dioxide 13463-67-7b O=[Ti]=O NA NA
Trolamine salicylated 2174-16-5b C1=CC=C(C(=C1)C(=O)O)
O.C(CO)N(CCO)CCO
2240b 7.1×10−13b 7.3×10−9b 2.06b 4.8×10−4 @ 25°Cb 1.1×10−2 @ 25°Cb
Zinc oxide 1314-13-2b O=[Zn] NA NA

a ECHA (https://echa.europa.eu/information-on-chemicals/registered-substances?p_p_id=dissregisteredsubstances_WAR_dissregsubsportlet&p_p_lifecycle=1&p_p_state=normal&p_p_mode=view&_dissregisteredsubstances_WAR_dissregsubsportlet_javax.portlet.action=dissRegisteredSubstancesAction).

b EPI Suite™, v. 4.11 (www.epa.gov/tsca-screening-tools/epi-suitem-estimation-program-interface).

c PubChem (https://pubchem.ncbi.nlm.nih.gov).

d Note that the UV filter is an ionic pairing of monoethanolamine and salicylate, which are modeled separately.

NOTE: Log Dow calculations were only applied for acids; for neutral compounds and bases, the committee assumed that they would sorb at least as strongly as their Kow would indicate.

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Image
FIGURE 4.2 Factors influencing the fate and environmental distribution of organic UV filters. SOURCE: Adapted from Schwarzenbach et al., 2016.

is generally lower than in pure water due to salting-out effects.1 However, solubility for some organic chemicals can also vary with temperature. Reported solubilities (pure water, 25°C) for the organic UV filters of interest are provided in Table 4.1. Broadly speaking, the reported solubilities range from relatively low values (dioxybenzone at 0.013 mg/L [13 μg/L]) to quite high values (sulisobenzone at 300,000 mg/L [300,000,000 μg/L]). However, testing conditions for “freshwater solubility” or “saltwater/marine solubility” are not standardized at the national or international level and the literature reports high variability in these predictions and/or measurements for some UV filters (see Mitchelmore et al., 2021). It is rare, especially for poorly studied chemicals such as UV filters, to have solubility measurements reported for environmentally relevant aqueous geochemical conditions (e.g., seawater). Standardized measurements of solubility are made in deionized water at 25°C, meaning that results are of maximum solubility in water. Actual solubility would be less in freshwater and usually much less in saltwater. Complicating the measurement of solubility (and its relevance to environmental fate predictions) is the fact that for ionogenic organic chemicals—chemicals that can exist as ions in solution—the solubility can be impacted by the charge state of the chemical, which is in turn dependent on parameters such as the acid-dissociation constant(s) for the chemicals (i.e., pKa values) and the pH of the aqueous phase. Importantly, when an organic chemical is in its charged (ionic) form, it is much more soluble in water than when in its neutral (uncharged) form. Thus, knowing the

___________________

1 In theory, during salting-out, an increase in salt concentration results in some of the water molecules being attracted by the salt ions, decreasing the number of water molecules available to interact with the organic chemical protein. Scientists do not completely understand the exact reason this happens, but it is commonly observed.

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

pKa of the organic chemical and the pH of the water into which the chemical is dissolving is critical to assessing its solubility. These parameters underscore the importance of collecting data under real-world conditions. However, these data, which can include natural components (i.e., DOC) data on the influence of co-released chemicals (e.g., other UV filters, other components of sunscreen formulations) that may impact chemical fate, are lacking.

Volatility

Volatility is related to the propensity of some organic chemicals to transfer from a condensed phase (solid, liquid) to the gas phase at lower temperatures. More volatile compounds (higher vapor pressures, lower boiling points) are more likely to be found as a vapor and their environmental distribution is increasingly via the atmosphere. Volatility in environmental assessments is often described by a compound’s vapor pressure, the concentration of a chemical in the gas phase when in equilibrium with its pure condensed phase. Available vapor pressure data are summarized in Table 4.1. In general, the vapor pressure of most organic UV filters are quite low at ambient temperatures. For aquatic environments, Henry’s Law constants are used to reflect the propensity of an organic chemical to transfer from water to air. Henry’s Law constant is the equilibrium ratio of the concentrations of a chemical in the gas phase versus water. Because Henry’s Law constants reflect both vapor pressure and solubility of organic chemicals and can be calculated from these data, the constants are prone to the same potential confounding factors as solubility and vapor pressure; they are often dependent on temperature, salinity (ionic strength), and in some cases, pKa and pH. Highly volatile compounds are prone to distribution via the atmosphere with subsequent deposition at colder temperatures (e.g., at the poles or in the mountains) with environmental occurrence being unassociated with direct human presence at the location. The phenomenon is known as “global distillation” and well known for persistent organic pollutants (POPs) such as the highly chlorinated pesticides DDT and Dieldrin (http://www.pops.int). These compounds are often nonbiodegradable, highly lipophilic, and highly volatile. For ionogenic organic chemicals, which often have high solubilities and low vapor pressures (only the neutral form is expected to volatilize), the pH of the aqueous system also complicates Henry’s Law constants. However, in light of the generally low vapor pressures and low/moderate to high solubilities, losses to the atmosphere may be negligible for the organic UV filters of interest in this report (Pintado-Herrera and Lara-Martin, 2020).

Acid Dissociation Constant

Acid dissociation constant (pKa) of an organic chemical in water provides insight into the tendency of a chemical to be charged or uncharged at environmental pH. The toxicity of ionizable compounds is often dependent on whether the molecule is ionized or neutral. The neutral (uncharged) form is usually the more toxic, and knowledge of the pKa informs whether an organic chemical will be neutral in a given water with a specific pH. Unsurprisingly, bioaccumulation follows the same pattern, with the neutral form of many organic chemicals being more bioaccumulative. Reflecting the diversity of chemical structures, the pKa values of organic UV filters can be quite low (e.g., sulisobenzone at 0.022, indicating its almost universal existence as a negatively charged anion; Table 4.1) to relatively moderate (e.g., octisalate, at 8.13, reflecting its existence as both neutral molecules and anions at environmentally relevant pH values; Table 4.1).

Hydrophobicity and Accumulation on Solid Phases

For many organic chemicals, releases to aquatic environments can result in sorption to solid particles, some of which can settle out under non-turbulent flow conditions. Sorption, the tendency of a chemical to physically interact with solids, is a key process impacting many environmental organic contaminants. The degree to which many organic chemicals sorb (i.e., stick to) these solid phases can be predicted based on the organic carbon content of the solid phase and the hydrophobicity of the chemical as well as the soil adsorption coefficient. The hydrophobicity of a chemical is arguably the most important physico-chemical parameter relevant for environmental risk assessment. Hydrophobicity (the tendency to partition out of water) and the related lipophilicity (the tendency to

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

partition into fat, lipid, or oil) is often expressed as the octanol:water partition coefficient, or Kow, often expressed in log10 form (i.e., log Kow).

Higher log Kow compounds partition preferentially into natural organic matter (such as that associated with environmental solids) and into lipids (such as those present in aquatic organisms). Thus, a compound’s log Kow can be used as a key parameter to predict its sorption to environmental solids as well as its bioaccumulation potential in the environment. Hydrophobic organic chemicals (high log Kow values) are often also lipophilic compounds. They tend to have lower aqueous solubilities and as such may sorb, or stick to, solid particles present in water. High Kow compounds are therefore critical to assess in terms of long-term or chronic aquatic toxicity, as acute toxicity may not be expressed at the limit of solubility and longer exposures at these lower concentrations may elicit effects. Log Kow is the most common parameter used to develop QSARs for ecotoxicity and bioaccumulation, and is a driver for most environmental fate models where understanding the amount of chemical in solution versus that sorbed to solids is critical.

Expression of sorption behavior for purposes of environmental exposure modeling is usually given as a solid-water distribution coefficient (Kd value), which quantitatively relates the amount of a chemical found on solids to that dissolved in the water phase. For many organic chemicals and environmental solids, the Kd value can be normalized to the fraction of the solid phase composed of organic carbon (foc; as this is the primary sorbent in environmental scenarios) and is given as the Koc. A compound’s Koc value, which is often considered relatively independent of the identity of the sorbing phase, is commonly predicted from its Kow value. For ionogenic organic chemicals, one can attempt to correct for the charge state of the organic chemical as a function of the pH of the system (i.e., Dow values, which are pH dependent), though predictions for sorption of charged organic compounds are complicated by uncertainty (Tülp et al., 2009). Ultimately, the affinity of a chemical for a solid phase is critical not only for predicting accumulation in freshwater and marine sediments, but also for predicting the rate of transport in groundwater systems (i.e., when contaminants are released to the subsurface, such as through on-site wastewater treatment).

With respect to organic UV filters, only a few studies have examined the sorption of these compounds to environmental solids, though several studies have observed accumulation of these chemicals in sediments (described later in this chapter). In a study of the sorption of avobenzone and padimate O to freshwater sediments, Li et al. (2016c) observed highly linear sorption of these two compounds to four different sediments and derived experimental log Koc values of 3.25 ± 0.04 and 4.41 ± 0.03 for these two chemicals, respectively. The authors also concluded these were in relatively good agreement with those predicted from Kow using standard relationships (Schwarzenbach et al., 2005), though the Kow employed for avobenzone may not have been accurate. Others (Pintado-Herrera and Lara-Martin, 2020) have estimated log Koc values for oxybenzone (3.103), octinoxate (4.08), homosalate (4.03), octisalate (3.933), and padimate O (3.41), though again, the accuracy of such predictions is predicated on accurate Kow input values. Field-derived Kd values are very limited, with only one study (He et al., 2019a) reporting any values (to date). This study did not observe apparent Kd values to be associated with the compounds hydrophobicity, suggesting that either hydrophobicity (i.e., Kow) was not the primary driver of sorption to solids for the UV filters, or that the water and sediment samples from which these values were derived were not in equilibrium with each other. In another study (Fagervold et al., 2019) a field-based Kd value was derived for oxybenzone, but could not be derived for avobenzone due to sporadic detection. While this study did derive laboratory-based Kd values for sediments for these two chemicals, the lack of information as to the organic carbon content of the sorbents limits the broader applicability of these data.

In addition to accumulating in solid phases such as sediments or soils, organic UV filters may also accumulate on non-settling particles (i.e., suspended solids including phytoplankton) in the water column as well as the air-water interface or surface microlayer (SML), though the data supporting the latter is much more limited and variable. With respect to suspended particles, a few field studies have documented significant contributions of suspended particles to total measured aqueous concentrations. Whether determined by removing particles through filtration (0.22 μm filter; Tovar-Sánchez et al., 2013) or by desorbing particulate-bound organic UV filters through sonication (Benede et al., 2014), inclusion of the particulate-bound fraction of organic UV filters resulted in higher total aqueous concentrations, though the extent to which this was evident was somewhat variable. Most recently, a few field-based studies have documented the presence of some organic UV filters (particularly homosalate,

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

octinoxate, octocrylene, oxybenzone, and avobenzone) on microplastic particles, indicating that these seemingly ubiquitous particles may afford an additional phase onto which organic UV filters can partition. The relative affinities of organic UV filters for microplastics (i.e., partitioning constants) remains largely unexplored, though higher log Kow compounds would be expected to partition more strongly. As both the dissolved and particle-bound (including colloid-bound and microplastic-bound) fractions of UV filters can potentially contribute to exposures for aquatic organisms, consideration of both fractions is likely warranted.

The SML has been increasingly recognized as a unique habitat and potential exposure medium for aquatic species (Cunliffe et al., 2013). Organic chemicals—including organic UV filters—have the potential to partition into the SML (Guitart et al., 2004), likely due to the presence of high concentrations of naturally derived lipids, fatty acids, and proteins (Wurl and Obbard, 2004), for which organic UV filters would be expected to have a higher affinity (depending on their hydrophobicity). For some chemicals, this enrichment (relative to the bulk subsurface water) may be 500-fold (Wurl and Obbard, 2004). Interest in the SML likely has arisen due to reports of an oily sheen sometimes observed on the surface of heavily used, slow, or nonmoving recreational water bodies; the extent to which inert components of sunscreens contribute to the SML remains unexplored. The limited data with respect to organic UV filter enrichment in the SML does indicate some enrichment in the SML. In one field-based study (Tovar-Sánchez et al., 2013), slightly elevated (< 2 fold) levels of oxybenzone were observed in the SML, whereas a more recent study (Fagervold et al., 2019) observed enrichment of avobenzone and oxybenzone in freshwater and coastal bay SMLs. However, another study in the U.S. Virgin Islands (Bargar et al., 2015) was inconclusive, primarily due to the infrequent detections of organic UV filters in paired SML/bulk water samples. Nevertheless, consideration of the SML and measurements or estimations of the extent to which the SML is enriched (relative to bulk water) may be important depending on the life cycles of sensitive organisms, some of which such as embryonic corals and fish may spend time near or within the SML.

UV filters may also be adsorbed onto and absorbed into external mucus layers of animals. This may be an exposure pathway for corals via feeding on organisms entrapped in the mucus. The nature and role of mucus for corals is described by Brown and Bythell (2005).

Biodegradation and Microbial Metabolism

Biodegradation of xenobiotics in the environment is an important component of understanding their environmental risks. Biodegradation can ultimately limit the concentration and spread of a chemical, can return elements to their natural biogeochemical cycles, can transform chemicals to generally less toxic metabolites, and—when coupled with sorption—is a key component of environmental exposure modeling in risk assessment. Studies of biodegradation in wastewater treatment plants (WWTPs) following OECD standard test guidelines are described in Chapter 3 as part of the discussion about potential sources into the environment. Based on the tests described in Chapter 3, avobenzone, dioxybenzone, octocrylene, ensulizole, and ecamsule have been determined to be nonbiodegradable with little to no evidence of biotransformation.

In the case of UV filters, environmental compartments other than waste treatment are relevant due to the routes of exposure into the environment being both indirect (rinse-off, down-the-drain to WWTPs, land runoff) and direct (rinse-off in the absence of wastewater treatment and direct rinse-off during use, such as at the beach). The following section describes additional evidence related to biodegradation of UV filters in the aquatic environment. Notably, nearly all of the organic UV filters lack biodegradation data for soils and marine waters.

Evidence of Microbial Degradation in Nonstandard Tests or Conditions

It is well known that biodegradation and biotransformation are important in exposure scenarios beyond wastewater treatment, including surface (fresh) water, soil, sediment, and marine conditions. Compared to wastewater, these other environmental media are highly understudied, although some effort is emerging to devise improved marine biodegradability studies driven by the recent interest in microplastics and polymers at the international level (Ott et al., 2020a,b). One test guideline series, OECD 314, provides designs to quantify kinetics of degradation in a wide range of nonwastewater media, but there are no OECD 314 test guideline studies on UV filters.

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

A range of nonstandard research investigations have been conducted on various UV filters. These are summarized in Chapter 3, Table 3.4. Evidence from nonstandard test systems is limited with the bulk of research focused on oxybenzone and its metabolites. Rodríguez-Rodríguez et al. (2012) and Badia-Fabregat et al. (2012a,b) confirm the biodegradability of oxybenzone in sewage sludge as shown in Table 4.2. Interestingly, metabolism of UV filters has been researched frequently using metabolically competent fungi. Under enriched conditions (perhaps not environmentally relevant at this time), the fungus Trametes versicolor was found to virtually eliminate parent oxybenzone with identifiable metabolites in the mixture.

Liu et al. (2012) found anaerobic degradation a more favorable condition for oxybenzone degradation than aerobic environments, suggesting more advanced WWTP designs incorporating anaerobic digestion are needed

TABLE 4.2 Evidence of Microbial Biotransformation of UV Filters in Non-Wastewater Contexts

UV Filter Degraders Medium Biotransformation and Biodegradation Observations Source
Oxybenzone Soil column 12% of load degraded Janzen et al., 2009
82–96% eliminated from pore water due to sorption
Mixture of oxybenzone, octocrylene, padimate O, octinoxate, benzophenone-1a and 4-DHBa Fungus, Trametes versicolor Sewage sludge at 10 mg/L chemical per 5 g/L of fungus Removals of parent were 22%, 58%, 70%, 79%, 100%, and 1%, respectively Rodríguez-Rodríguez et al., 2012
Oxybenzone Fungus, Trametes versicolor Sewage sludge at 10 mg/L ~100% degradation of parent, metabolites verified, high laccase activity Badia-Fabregat et al., 2012a
Oxybenzone Wastewater microbial community Liquid mixture of 10% activated sludge and digested sludge under oxic and anoxic conditions Anaerobic biodegradation concluded to be more favorable; a 1 mg/L exposure was eliminated to ~0 by 28 d in states, although anaerobic conditions were faster. Metabolites formed in both Liu et al., 2012
Benzophenone-1a Fungus, Trametes versicolor Sewage sludge at 10 mg/L ~98% degradation of parent, high laccase activity; 4HB and 4DHB (metabolites) were slowly eliminated Badia-Fabregat et al., 2012a
Benzophenone-1a Staphylococcus aureus and Enterococcus faecalis (gram +) Salmonella typhimurium and Serratia rubidae (gram −) Seeded on solid phase nutrient medium 2.5–26.7% removal; 1–24% removal of the metabolite 4HB dosed alone Chiriac et al., 2021
Octocrylene Microbial consortia from a landfill site and enriched on octocrylene (Mycobacterium agri and Gordonia cholesterolivorans) DSMZ culture medium for parent degradation and metabolite formation M. agri – 19.1% biodegradation including a number of metabolites; Suleiman et al., 2019
G. cholesterolivorans – no parent loss but biofilm grew
Padimate O Natural marine sediment communities Marine sediments in aerobic and anaerobic reactors Kinetic rate (first order) constants under aerobic and anaerobic conditions similar. Rate constants increased over time up to the value of 0.039/d. Removal was up to 90% parent. Volpe et al., 2017

aIncluded in Table 4.2 as an expected degradate.

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

for higher removal from wastewater. Oxybenzone metabolites were more metabolized by Trametes versicolor (Badia-Fabregat et al., 2012a) than purified strains of several gram positive and negative bacteria (Chiriac et al., 2021). Badia-Fabregat et al. (2012a) proposed a metabolic biotransformation pathway for oxybenzone (Figure 4.3) and noted that, as in mammals, phase I and phase II enzymes participate in the degradation of oxybenzone and benzophenone-1 by T. versicolor. For the UV filters studied by the authors (including derivatized camphors used on the European market), the main metabolites identified were the conjugated forms.

Suleiman et al. (2019) and Volpe et al. (2017) studied octocrylene and padimate O degradation, respectively. Suleiman et al. (2019) enriched soil leachate from a contaminated site for octocrylene-competent degraders by culturing microbial consortia on the UV filter as the sole source of carbon. Mycobacterium agri retained competency in culture and biodegraded 19.1 percent of added octocrylene along with formation of metabolites. Octocrylene was nonbiodegradable in wastewater but showed biotransformation in sediment conditions (Suhendra et al., 2020). More targeted investigations are needed to understand the biodegradation profile of this UV filter under natural conditions. Volpe et al. (2017) investigated the ready biodegradable UV filter padimate O in natural marine sediment communities returned to the laboratory. High levels of biodegradation (90 percent) in aerobic and anaerobic environments were found (Table 4.2).

Abiotic Transformation Processes

UV filters can undergo transformations in the absence of specific biological activity after release into the environment. Broadly speaking, rapid abiotic transformation, such as through hydrolysis, is unlikely as the organic UV filters would need to be stable enough for use in products such as sunscreens. In addition, reactions between UV filters and chemical oxidants—such as chlorine added to freshwater (Sakkas et al., 2003) and seawater (Manafsi, 2017) swimming pools—can lead to transformation of many UV filters to less-studied transformation products that can exhibit increased toxicity (Vione et al., 2015).

Image
FIGURE 4.3 Degradation pathways of oxybenzone (BP3) by the fungus Trametes versicolor. SOURCE: Badia-Fabregat et al., 2012a.
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

Of interest for environmental exposure assessments are photochemical transformations. This is not particularly surprising, as UV filters are intentionally used for their ability to intercept light at wavelengths that can damage cellular function. Photolytic transformation of UV filters in aquatic environments can occur both through direct photolysis, where the UV filter itself absorbs light and subsequently transforms, and indirect (or sensitized) photolysis, where the UV filters may react with a reactive species that is generated from other chemicals (photosensitizers) present in the water column. In many cases, phototransformation can be described as photodegradation, where the concentration of the parent chemical declines. However, photoisomerization, in which the three-dimensional structure of the UV filter chemical is altered, can also occur for some UV filters (i.e., octinoxate; MacManus-Spencer et al., 2011). This can complicate the term “photochemical half-life” as the three-dimensional structure of the molecule changes, but it may be a reversible reaction. Whether photoisomerization is sufficient to reduce potential risk (if, for example, only one isomer imparts toxicity) and therefore relevant will depend on the individual UV filter. In most cases, complete photochemical transformations, in which an irreversible reaction occurs and photochemical transformation products are formed, are the most relevant for environmental risk assessments.

Examples of the existing literature-derived aquatic photochemical half-lives of organic UV filters are included in Table 4.3, though these values should be considered with extreme caution: the photochemical half-life values may be affected by the wavelength(s) of irradiation, irradiation dose, and the presence of additional organic molecules (or molecular aggregates; Hanson et al., 2015). Importantly, as noted in Chapter 2, there is some evidence to suggest that the presence of some UV filters can impact the phototransformation of other UV filters (e.g., octinoxate in the presence of avobenzone; Sayre, 2005). Further, a recent study on avobenzone photochemical degradation (Hanson et al., 2020) revealed that avobenzone photostability is highly dependent on the molecular-scale local environment, with higher photostability of avobenzone when present in protic environments, even when primarily contained within aqueous surfactant micelles (further discussion of the role of molecular-scale interactions is found in Box 4.1). The extent to which similar behavior occurs for other UV filters remains largely unexplored, particularly in aquatic environments where macro-molecular assemblages with other anthropogenic organic compounds (i.e., other ingredients of sunscreen formulations) or dissolved organic matter could result in similar molecular-scale impacts to photochemical transformations. While detailed mechanistic photochemical transformation assessments (taking into account varying solar irradiation, quantum yields for direct photochemical transformations, steady-state concentrations of other photosensitizers, etc.) may ultimately be needed, even well-controlled empirical aquatic photostability assessments are not readily available for many UV filters.

As is clear from this compilation, there are a broad range of reported aquatic photochemical half-lives for UV filters, with some UV filters likely transforming rapidly (e.g., ensulizole), whereas others appear to be very photostable (e.g., dioxybenzone). However, aquatic photostability data for organic UV filters are limited and there are some chemicals where no data appear to be available. Of particular interest is the broad range of potential half-lives for some chemicals (e.g., oxybenzone). This may be due to the range of environmental conditions investigated, as indirect photochemical half-life estimations are very sensitive to the type and concentrations of photosensitizers (i.e., dissolved organic carbon) present in a water body. Further, while some studies have attempted to differentiate which photochemical sensitizers (i.e., OH) are resulting in UV filter transformation, these mechanistic studies are relatively rare (Zhang et al., 2022). Photostability in sunscreen formulations specifically (which may differ significantly from photostability in the aquatic environment) is described in Chapter 2. Most studies on UV filter photostability have been performed in organic solvents, as this is more relevant to sunscreen formulations; photostability data for UV filters in dilute aqueous environments is significantly limited.

As with the transformation products formed due to chemical oxidants such as chlorine and bromine, there is the potential for toxicity associated with photochemical transformation products both of parent molecules (e.g., octinoxate; Butt and Christensen, 2000) and metabolites (e.g., oxybenzone; Vuckovic et al., 2022). Chlorination transformation products themselves might warrant further evaluation, as they may be present not only in WWTP discharges (which are often chlorinated), but also in discharges of recreational pools or other settings. For both photochemical transformation products and chlorination transformation products, unlike biological transformation products, which tend to exhibit reduced biological impacts than the parent chemical, transformation product toxicity evaluations may be warranted.

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

PHYSICAL FACTORS AND SPATIAL RELATIONSHIPS

As established in Chapter 3, the physics of a water body heavily influences the concentration of a pollutant. Although this is the case for any aquatic pollutant, given the diversity of sources (from rinse-off of swimmers to fixed point sources such as WWTP effluent) consideration of the hydrography of any given waterbody will be essential to ascertain risks to aquatic biota. The hydrography of a water body includes measures of the physical features (e.g., size of the water body, degree of water mixing, retention time of the water/flushing) and will vary considerably from location to location (Figure 3.8). The influence of the hydrography of a given location in addition to types and input rates of the UV filters (e.g., number of people for recreational sources and the influence of other sources) is important for determining their transport, fate, and potential exposure to aquatic organisms.

In general, due to the wide range of aquatic habitats that UV filters may enter via recreational activities, and the varied inputs discussed in Chapter 3, assessing risk will require inclusion of the unique dynamics of the water bodies considered. These dynamics may include turnover time (or residence time) of the waterbody, rates of mixing and directionality (i.e., prevailing currents and downstream transport), vertical stratification of the water column, and sediment type and patterns of sediment movement.

UV filters may enter both freshwater and marine environments from point sources of municipal wastewater treatment systems (Chapter 3). Figure 4.4 illustrates a map of the continental United States where in-stream flows and WWTP design flows are used to estimate “dilution factors” in different river reaches (Rice and Westerhoff, 2017). Such factors are also developed for marine systems. This and similar approaches are used to account for dilution of trace organic chemicals in WWTP effluents and to estimate ranges of environmental concentrations. Despite the large dilution potential of lakes and marine environments, it is possible for elevated concentrations of UV filter near local point-sources (e.g., septic systems, onsite WWTPs) and nonpoint sources (e.g., bathers, beach wash-off, stormwater). However, this is likely to be temporally variable given changes in inputs over hours, days, and seasons. Thus, exposure calculations, and especially monitoring efforts, need repeated and replicated measures in space and time for an accurate exposure assessment.

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

TABLE 4.3 Example Reported Aquatic Photochemical Half-Lives for the Organic UV Filter Chemicals Included in This Study

UV Filter Direct Photolysis Half-life (Hrs) (Source) Indirect Photolysis Half-life (Hrs) (Conditions) (Source)
Aminobenzoic acid 5.2 (Allen et al., 1998)
Avobenzone 17 (Allen et al., 1998)
Cinoxate Photoisomerization only (Morlière et al., 1982)
Dioxybenzone Not measurable (Allen et al., 1998)
99 (Kotnik et al., 2016)
Ensulizole 1.23 to 7.45 (Ji et al., 2013)
0.5 (Westphal et al., 2020)
Meradimate 5.9 (Allen et al., 1998) 0.002 to 0.14 (synthetic conditions) (Lanzafame et al., 2017)
Octinoxate 21 (Rodil et al., 2009)
0.8 (trans isomer) (MacManus-Spencer et al., 2011)
1.3 (cis isomer) (MacManus-Spencer et al., 2011)
< 1.4 (summer sunny days) (Vione et al., 2015)
0.4 (trans only, in presence of SWRA) (MacManus-Spencer et al., 2011)
120-216 (ECHAa)
Octisalate Not measurable (Allen et al., 1998)
Octocrylene > 72 (Rodil et al., 2009)
96 (Allen et al., 1998)
Oxybenzone Not observed (Li et al., 2016c)
> 72 (Rodil et al., 2009)
Not measurable (Allen et al., 1998)
76 (Kotnik et al., 2016)
10,392 (Liu et al., 2011)
13 (seawater) (Y. Li et al., 2016c)
6.5 (freshwater) (Y. Li et al., 2016c)
5208 (10 mg/L HA) (Liu et al., 2011)
72 (near surface) (Semones et al., 2017)
Padimate O 20 (Rodil et al., 2009)
31 (Allen et al., 1998)
1.6 to 27 (Sakkas et al., 2003)
6.3 to 39 (DI with HA to seawater) (Sakkas et al., 2003)
Sulisobenzone 357 (Semones et al., 2017) 96 (near surface) (Semones et al., 2017)

a https://echa.europa.eu/registration-dossier/-/registered-dossier/15876/5/2/4.

NOTES: There are no data available for the organic UV filters homosalate and trolamine salicylate. In some cases, the reported half-lives may be for photoisomerization of the molecule and not for complete photodegradation (i.e., irreversible bond-cleavage). In addition, given the role of molecular-scale environments on photostability (e.g., Hanson et al., 2020), some values may be biased by the addition of organic additives to enhance UV filter solubility.

ESTIMATED AND MEASURED CONCENTRATIONS IN WATER AND SEDIMENTS

Exposure assessments for chemicals in aquatic environments typically include estimates or measures of chemical concentrations in water, sediments, and/or internally in biota (i.e., tissue residue levels). This section of the report covers estimates and measures in water and sediments while Chapter 5 provides information on concentrations in biota. Estimated or modeled concentrations are given first, followed by a review of measured concentrations of UV filters in water and sediments. A combination of both methods is appropriate for assessing exposures that align with the information on effects presented in Chapter 6.

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Image
FIGURE 4.4 Color-coded map for dilution factors for streams across the United States based on average streamflows. Stream segments gray in color represent stream segments that are not impacted by publicly owned treatment works wastewater discharges. Dilution factors decrease during lower flow periods. SOURCE: Rice and Westerhoff, 2017.

Estimating UV Filter Concentrations

Concentrations of UV filters in the aquatic environment vary due to numerous and diverse sources, dilution, and environmental fate processes. Each of these factors affects the levels of UV filters to which an organism might be exposed, with the duration of exposure also critical in determining effect. Thus, to estimate the exposure of aquatic organisms to UV filters, it is helpful to understand the possible concentrations that may occur and the likelihood of their occurrence. To estimate these concentrations, a number of complementary approaches are often useful. The first approach is to model the concentrations in the environment based on input rates, which include factors such as the number of people applying sunscreen, rates of rinse-off, their recreation habits, concentrations and removal rates of UV filters in WWTPs, amount of effluent discharged, and more (see earlier discussions in Chapters 2 and 3). Ideally, inputs are then used in combination with environmental factors such as the size of the waterbody, hydraulic residence times, and information on other fate processes to estimate the concentration potentially present. These provide useful bounding estimates that may not capture the range of variability but serve to indicate whether there is a potential for exposures that exceed toxicity thresholds.

The second approach that complements these models involves measurements of the occurrence of UV filters, ideally with frequent and replicated sampling over space and time (see the section Measuring UV Filter Concentrations). Measurements using discrete (space and time) samples (i.e., “grab” samples) represent a “snapshot” in time of the concentration at the specific location sampled, while replicated samples over a larger area reflect variability of location and time collected. Due to variability in both rates of UV filter inputs and environmental conditions, the concentrations of any chemical may vary substantially over time, even at the scale of hours, and evidence to date shows variability over the day, month, and season. To obtain exposure estimates over time, a few different

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

methods have been employed. The concentration of contaminants in the tissues of resident biota (e.g., a sessile organism or a non-migratory fish) can be used to reflect exposure concentrations over time; however, using biota for this purpose can be complicated as a compound’s bioavailability and uptake depends on the specific organism (and exposure pathways). Additionally, an organism may rapidly excrete or metabolize the compound of interest or uptake or loss may occur based on life history traits (e.g., reproduction and spawning). Artificial passive samplers (e.g., POCIS, SPMDs) (Booij et al., 2016; Burgess, 2012; Burgess et al., 2016; Gong et al., 2018; USGS, 2010) along with other methods such as continuous low-flow aquatic monitoring samplers (CLAM) (Ensminger et al., 2017) have been developed to try to overcome these issues, as they “sample” the environment over longer periods of time and thus are expected to provide an indication of the average longer-term dissolved (or “free”) concentration of a particular chemical. These types of samplers are continually undergoing development. While passive samplers deployed in surface waters may provide an indication of average exposure levels, they would likely not be able to characterize the short-term variations or pulses in exposure levels in the water. For sediments, temporal variations in UV filter concentrations are likely dampened relative to water concentrations, and passive samplers could provide a longer-term measure of exposure of organisms living on or within the sediments; an example is the evaluation carried out by Muz et al. (2020).

UV Filter Exposure Modeling

A few publications model inputs, transformations, and/or exposure concentrations of UV filters in the environment based on inputs and environmental conditions. For example, a model estimated the mass loading of four UV filters (octinoxate, octocrylene, avobenzone, and oxybenzone) into Texas beaches along the Gulf of Mexico (Sharifan et al., 2016). The only measured input variable of the compounds in this study was from recreational swimming. The model incorporated the average concentration of UV filters in sunscreens (from Poiger et al., 2004), estimates of applied sunscreen, the surface area of swimmers’ bodies, the solubility of the UV filters, and the number of swimmers (annually) visiting recreational, coastal beaches in Texas. This model provides useful insights into the mass loading of UV filters (in kg/year) that each county in Texas may have from recreational swimming. Annual mass loadings at all of the beaches studied were reported as 1,873.8, 1,249.2, 1,015, and 624.6 kg/year for octinoxate, octocrylene, avobenzone, and oxybenzone, respectively. This type of model of mass loading does not allow for estimates of concentration at a given location, but instead provides a perspective on the mass of UV filters that may enter surface waters from recreational activities.

An additional study in Swiss lakes combined a series of survey approaches to estimate loading of UV filters (Poiger et al., 2004). Researchers conducted surveys of UV filters found in the region, the number of swimmers in a region, and UV filter use. This study calculated the UV filter inputs based on the above survey information and estimated concentrations in the lakes given their discharge (water volume). Using this information, the authors estimated a worst-case scenario of 966 and 4.2 kg of UV filters entering Lake Zurich and Hüttnersee, respectively, and predicted concentrations of less than 1 up to 266 ng/L (0.001–0.266 μg/L) depending on the UV filter and lake. The study also took samples of the concentrations in these lakes and deployed semi-permeable membrane devices (SPMDs) as integrated passive samplers. The concentrations measured in the lakes ranged from nondetectable (< 2 ng/L) to up to 125 ng/L (0.125 μg/L), depending on the compound. The results from deploying the SPMDs found that the concentrations of UV filters in Lake Zurich were two to four times higher than Lake Greifensee, but they found that the opposite pattern was the case for pharmaceuticals and personal care products that are typically associated with WWTP effluent. These findings suggest that recreational inputs were a significant source of UV filters in this system. However, concentrations of UV filters found in water samples were much lower than those predicted from modeling recreational activities. The authors postulate that these differences may be due to an overestimation of the release from the skin or removal processes in the environment such as sorption to sediments or breakdown processes (Poiger et al., 2004).

A handful of other studies employ similar approaches by coupling surveys of UV filter use and estimates of inputs with measurements of concentrations in water bodies (e.g., Balmer et al., 2005; Labille et al., 2020b; Schaap

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

and Slijkerman, 2018). For example, a detailed study of recreational bathing practices, UV filter use, and measurements of UV filters at three beaches in France (Labille et al., 2020b) provides an example of the value of combining these measurements to understand the scope of potential exposure to aquatic organisms. From detailed counts of bathers and UV filter use surveys, the estimated use of UV filters (per 3,000 beach goers) was approximately 1 kg/day for each UV filter ingredient and a total of 15 kg/day for all UV filter ingredients potentially released into the seawater at these popular bathing sites (Labille et al., 2020b). The authors then calculated the likely concentration in the water or the predicted environmental concentration (PEC) based on bathymetric measurements of the waterbodies and residence times where the recreational bathing occurred. These PECs were compared to measured concentrations. In all cases, the measured concentrations fell below predicted concentrations (about 30 to 49 percent of the PEC for inorganic UV filters and 0 to 3 percent of the PEC for organic UV filters [Labille et al., 2020b]). The authors attribute these differences to factors such as the assumption that all of the UV filters applied would enter the water column and to various fate processes such as interactions with sediments. The authors note that other studies have shown that organic UV filters are absorbed in the skin (Matta et al., 2019) and hypothesize that the amount available for rinsing off of organic UV filters was greatly reduced as a result (Labille et al., 2020b).

Although predicted estimates are typically above concentrations of UV filters measured in surface waters, modeled and measured aquatic and sediment concentration data provide potential ranges, including upper bounds, and probabilities of exposure levels of UV filters to organisms. These estimates provide a foundation for risk assessment. Two such examples of that approach can be found in Carrao et al. (2021) and Burns et al. (2021), in which models were developed for releases at three beaches in the Great Lakes and down-the-drain releases to U.S. streams, respectively. In Carrao et al. (2021), the PECs varied significantly (e.g., for oxybenzone the range was 95,870 ng/L [95.87 μg/L] under the worst-case scenario to 50 ng/L [0.05 μg/L]) depending on the complexity of the assumptions made with respect to sunscreen use, releases, fate and transport processes, and more. The Burns et al. (2021) study, in which wastewater-derived oxybenzone loads (and subsequent concentrations) were estimated for freshwater receiving waters of the United States, predicted a range of PECs (i.e., for oxybenzone the range of 25th, 50th, 75th, and 90th percentile concentrations were 0.004, 0.01, 0.04 and 0.15 μg/L, respectively) that were in general agreement with measured environmental concentrations. However, the Burns et al. (2021) model was aimed at predicting constant wastewater-derived loads to U.S. streams. Whereas most measurements in U.S. streams note significant episodic signals associated with recreational releases (i.e., Rand et al., 2020; Reed et al., 2017), comparisons of PECs with measured data can be difficult. Modeling efforts are most useful when a direct comparison can be made between the PECs and the measurements, as was done by Labille et al. (2020b). With additional efforts to refine the input variables, these models may prove more useful for characterizing potential exposures. Importantly, they also can provide the basis for measuring the effects of these compounds at environmentally relevant concentrations. UV filters introduced to coastal environments present challenges for exposure modeling due to the temporal variations (pulses) associated with recreation as well as the three-dimensional nature of mixing away from the area where the UV filters are introduced. While simpler models can be used for bounding exposures in early-tier exposure estimates, more sophisticated models may be warranted for reducing the uncertainty in exposure estimates.

In aquatic ecosystems where a number of sources of UV filters are likely (e.g., surface waters with a combination of municipal releases and recreational activities are present), models will need to be developed to encompass these inputs to the system (e.g., Balmer et al., 2005). Moreover, measurement of UV filters, to ascertain environmental concentrations should be extensive and conducted over time and space to account for variability in multiple factors including variation in recreational activities and other sources, dilution, currents and water exchange, and chemical breakdown and sorption dynamics of the UV filters compounds. For these reasons, additional confirmation through the use of integrated samplers may provide useful information.

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

Measuring UV Filter Concentrations

The measurement data presented in the following sections and in Appendix C are not intended for direct use in risk assessment as the construction of reliable exposure estimates will need to take into account the variable factors described in this section. To that end, the committee compiled data on UV filter concentrations in water and sediments and investigated their potential relationship with site-specific characteristics related to (a) source strength, (b) proximity to potential sources, and (c) residence time of receiving waters from specific locations in the environmental setting (lower versus higher energy systems). These settings were used to gain perspective on the variability in measured concentrations among locations and times of sampling. They were also chosen to align with the conceptual view of converging extent and type of sources and environmental conditions that would lead to higher or lower exposure concentrations as described in Environmental Settings within Chapter 3.

As described later in this chapter, there are a number of sampling design and analytical challenges to measuring UV filters in water and sediments. The sampling and analytical methods have changed and presumably improved over time. Publications vary with respect to methodological detail, including regarding Quality Assurance and Quality Control (QA/QC). The papers relied upon were published within a 16-year period (2005 to 2021). Some studies are very robust with repeated samplings over time and space while others might consist of a few grab samples. In a few cases, results differed from similar sampling situations and these are pointed out in the text. Nevertheless, the variability in sampling designs and analytical capabilities introduces uncertainty into the data sets especially if single, unreplicated samples are used. In addition to the committee’s compiled data, statistical summaries have also been developed by others (Díaz-Cruz, 2020; Hopkins and Blaney, 2006; Mitchelmore et al., 2021; Tsui et al., 2014b). Collectively the committee’s data compilation and other statistical summaries indicate that concentrations of individual organic UV filters range between detection levels and 10 μg/L; values above 10 μg/L have been reported for oxybenzone, ZnO, and TiO2. Temporal and spatial variations in UV filter concentrations at a sampling location can vary over orders of magnitude depending on the factors explored in this chapter and this will be a critical factor for assessing exposures.

The compilation of values gives insight into spatial and temporal variability, reflects the differential fate characteristics of the UV filters, and reflects patterns of sunscreen usage. The latter may be a confounding factor in the interpretation of measured concentrations as there are regional and temporal differences in usage that will affect observed concentrations (e.g., as one UV filter is less used or is banned, others may become more prevalent). While interpretations made from the data set may not reflect patterns in the future as there may be changes in the use of different sunscreen products, observing changes before and after bans could provide insights into their effectiveness (e.g., if concentrations did not decrease after a ban). However, the available data provide insights into exposure patterns and potentially identify site-specific locations that may experience higher exposure concentrations of UV filters. As described in earlier chapters, UV filters occur in the manufacture and use of other non-sunscreen products and occurrences in the environment will reflect to varying degrees the influence of such manufacture and use.

The compilation capitalizes on sampling from around the world. As many of the products used in the United States are used elsewhere, the measurements of these chemicals in surface waters and sediments provide insights into their behavior and possible exposures in U.S. waters. Where possible, sampling locations within a study were disaggregated to better understand spatial and temporal variability. Thus, while the inventory of samples is inclusive, there are many more individual measurements that are embedded in reported medians, means, or as ranges. Because many studies were not designed to characterize the spatial and temporal characteristics of the UV filters at sampling locations, many of the entries provide snapshots in time and/or space and are not robust statistical representations of the presence of UV filters. Furthermore, without replicates, individual samples provide insufficient evidence that the sample is representative at that location and time; a single sample can be contaminated during or after collection, at any step during extraction and analysis, or the chemical can be lost to the collection vessel or at any point thereafter. The need for appropriate QA/QC steps (e.g., matrix and recovery spikes) is essential to provide accurate exposure data as described in greater detail later. The statistical distributions of data sets for individual UV filter data were explored by plotting the highest values reported from various studies. This reflects a measured upper bound from available data for possible exposure concentrations. For graphing purposes, chemical values looked for, but not detected in a study were assigned a value of 0.005 μg/L as this was representative of the

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

lower range of reported levels of detection. While there are scientific techniques for increased rigor when assigning numerical values, those methods go beyond the intended purpose of this graphical representation. The goal of these figures is to give a high-level view of the UV filter distribution in water, demonstrating the importance of the environmental setting for measured UV filter concentrations.

Organic UV Filters in Water

Studies of measured UV concentrations vary on the inclusion or targeting of UV filters. Oxybenzone was the UV filter most often included for measurement (122 studies in Appendix C, Table C.1) and detected in water (107 studies for the data extracted from papers). The frequency of measurements and occurrences of oxybenzone in water likely reflect the following: (1) oxybenzone is one of the more commonly used UV filters in sunscreen products in the United States and elsewhere because it offers effective protection in UVB and UVA2 ranges; (2) oxybenzone has raised environmental concerns regarding exposures to aquatic biota and has received frequent media attention; and (3) oxybenzone is relatively more soluble in water compared to the other most commonly measured UV filters (i.e., octocrylene and octinoxate; see section on Fate Characteristics of UV Filters; Table 4.1). The distributions of oxybenzone concentrations were sorted by the committee as being from marine beach areas (at or adjacent to bathing areas), coastal areas (beyond swimming areas immediately off beaches, further offshore, or at dive sites), and rivers and lakes (Figure 4.5) to reflect the different physical processes affecting exposure.

As Figure 4.5 shows, higher oxybenzone concentrations were found in marine beach areas than in coastal areas or in rivers and lakes. The highest reported values for beaches are from beaches at embayments, in shallower water, where water residence times are likely greater than for more open coastlines, and where there is higher use by people (hence higher input to water volume and flushing location). Oxybenzone and other UV filters can be rinsed off the skin during entry into the waters; if showers are installed at a beach, UV filters can be rinsed off in

Image
FIGURE 4.5 Log-scale distributions of oxybenzone concentrations in water (ng/L) for three broad types of sampling areas. Each data point is the maximum value reported in the study for a sampling location and event and some sampling consists of only one grab sample. Averages or medians would be lower. All non-detect or limit of quantifications are included at their stated values. Exceptionally high non-detect values that may be associated with matrix effects during analysis are included in the range of other non-detect values (around 5 ng/L) and do not influence the overall shape or information content of the distributions.
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

that manner as revealed by elevated levels of oxybenzone in sand at a beach shower (Downs et al., 2021). Marine beach areas have also been sampled more frequently than other environments and, to some extent, the occurrence of higher values may reflect this higher sampling intensity and/or high inputs to shallow water areas with potentially lower water volumes. Examples of sampling off beaches within embayments include Trunk Bay and Hawksnest Bay in the U.S. Virgin Islands (Bargar et al., 2015; Downs et al., 2016) and Hanauma Bay on Oahu, Hawaii (Downs et al., 2021). Hydrodynamic modeling indicates the residence of time of water in Hanauma Bay can be as long as 50 hours (Downs et al., 2021), which allows for accumulation over a few days. However, substantial variation in oxybenzone concentrations among studies for the same sampling location can also be found. For Trunk Bay, Bargar et al. (2015) measured and reported oxybenzone concentrations for April and June 2013 with the highest concentration of 4,643 ng/L (4.6 μg/L) for a nearshore composite sample in June 2013. In the following year, Bargar et al. (2015) evaluated the spatial distribution of oxybenzone in Trunk Bay in June 2014 and obtained a concentration of 6,073 ng/L (6.07 μg/L) for a composite of three samples taken along the shoreline where swimmers are most likely to be and at a time of day (around noon) when most swimmers will be present. Downs et al. (2016), as part of sampling of Trunk Bay a little before noon in April 2007, reported two single samples with values of 580,000 ng/L (580 μg/L) and 1,395,000 ng/L (1,395 μg/L); swimming activity was reported as more than 180 swimmers in the water and about 130 sunbathers on the beach within 100 m of the two sampling sites. The two studies also reported different concentrations for Hawksnest Bay for which the highest concentration observed by Bargar et al. (2015) was about 2,000–3,000 ng/L (2–3 μg/L) (interpolation from a graph in log scale) in the bathing area in June 2014. Downs et al. (2016) report values of 75,000 and 95,000 ng/L (75 and 95 μg/L) for the two samples collected in April 2007 where more than 230 people entered the water with swimmers coming within 20 m of the coral reefs. Reported water concentrations of oxybenzone around Oahu, Hawaii, also span several orders of magnitude and highlight variability within and between locations, distance from shoreline, and time of day (Downs et al., 2016, 2021; Mitchelmore et al., 2019). Because of the analytical challenges that can arise in the analysis of UV filters (see section on Analytical Chemistry Considerations) and the large differences in values—including the highest reported values for oxybenzone in water—it is advisable to resample these bays for analyses of all UV filters using replicate samples over multiple timepoints (e.g., hours, days, and seasons). It is possible that concentrations will have changed over time as a result of changes in use of sunscreens with oxybenzone.

As with any UV filter, the fate of oxybenzone once released into the environment is governed by partitioning and resistance to degradation. However, the environmental settings of the releases are also important to UV filter fate, as described in Chapter 3. Given the larger dataset available for oxybenzone, the committee examined reported concentrations of oxybenzone in water while also considering select components of environmental settings (see Appendix C). The three factors considered for environmental settings include the strength of the source, the proximity to the source, and the relative residence time of the water into which the release occurred. The analysis is qualitative and based on committee judgment. For example, the relative strength of the source was judged by reported intensity of beach use and the influence of wastewater effluents was judged by the size of the population served; in some cases, papers provided specifics on these factors. The proximity to source areas was judged based on distance from the source area (e.g., a swimming beach or outfall) and information presented in the publications. The relative residence time was judged based on information presented in the publication, the physical features of the receiving environment (e.g., an enclosed embayment, an open beach with high surf, or the presence of weak or strong currents), maps and Google Earth images, and information on tidal ranges at the location. Qualitative descriptors were used to characterize these factors on a relative basis. Each of these factors for environmental setting was judged using the considerations given in Table 4.4.

For each published study and discrete location and time event, source strength, proximity to source, and residence time of water were characterized with a short narrative (see Table C.2 of Appendix C) and color coded—orange, yellow, or blue—to indicate higher, moderate, or lower influence on contributing to or sustaining oxybenzone concentrations based on the general judgment-based criteria given in Table 4.4. To examine the combined influence of these factors on observed oxybenzone concentrations, the studies were sorted and ranked from high to low in terms of maximum measured oxybenzone concentrations in water. To this end, the studies were sorted into 4 bins reflecting orders of magnitudes of concentrations: A (> 1,000 ng/L), B (100–999 ng/L), C (10–99 ng/L), and D (< 10 ng/L [> 1μg/L to < 0.01μg/L]).

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

TABLE 4.4 Factors for Environmental Settings That Contribute to or Sustain Elevated Levels of Oxybenzone in Water

Relative Contribution Relative Source Strength Relative Proximity to Source Relative Residence Time
Higher Documented high beach activity or popular beach during in-season; documented wastewater inputs and reports of associated pollution; metrics indicating high density of people in a watershed Located in the immediate vicinity of the source such as bathing area of a beach or immediately downstream of an outfall Semi-enclosed embayments with long turnover times; reported weak currents and low tidal amplitudes; slow moving river systems or lakes with limited dilution
Moderate Documented lower beach activity or indeterminant or off-season; lower influence of population density and activity in watershed Located further from the source areas. For beach or coastal areas this is beyond the swimming areas; for rivers this is downstream of population centers and outfalls Open beaches but with reported low currents and tidal amplitudes; larger rivers affording more dilution
Lower Documented little or no beach activity; no reported wastewater discharges; no or low populations occupying the watershed and reports of areas as being pristine and under little or no influence Remote from source areas either because they are well offshore or because they are well downstream High energy areas that would foster dispersion and advection based on information about tides, currents, and wave action; very large rivers with high dilution relative to sources

NOTE: Fate of oxybenzone is also influenced by partitioning and resistance to degradation.

For studies within each of the four bins (A through D), the numbers of higher (orange), moderate (yellow), and lower (blue) influences were tallied. This yielded sums for higher, moderate, and lower influence classifications for source strength, proximity, and residence time for each of the four bins. For each bin, the total numbers of higher, moderate, and lower classifications were calculated and converted to percentages for that bin. These percentages reflect combined influences and are shown in Figure 4.6. For example, Figure 4.6 shows that as the combined influence of the factors decrease, observed oxybenzone concentrations decline by one or more orders of magnitude (see Table C.1 in Appendix C for details on individual studies). Specifically, Figure 4.6 shows that all measurements of 1,000 ng/L or higher were taken in places with high or moderate influences from source strength, source proximity, and longer water residence time. The convergence of these three factors at “higher” degrees of influence were found in 80 percent of the measurements of higher levels of oxybenzone (> 1,000 ng/L [> 1 μg/L]). The outcome of this qualitative examination suggests that source strength, proximity to source, and residence time converge to varying degrees from place to place to yield higher or lower concentrations of oxybenzone in water. This reflects spatial and temporal variations that bear on evaluating exposures and are discussed further below. As noted earlier, samples collected to date have varied considerably even at a single location, so this analysis can be viewed as providing qualitative insight into factors governing exposures to UV filters in water as represented by oxybenzone.

The next most commonly reported UV filters in water were octinoxate (92 studies2 with 73 detections) and octocrylene (79 studies with 62 detections). Distributions for these two UV filters in water are shown in Figures 4.7 and 4.8. As with oxybenzone, concentrations for octinoxate and octocrylene tended to be higher in beach areas, including marine and freshwater environments. Several values derived from the reviewed studies for the compounds exceed 1,000 ng/L (1 μg/L) and are most often found at beaches. Again, differential sampling intensity across environments will influence the observed range in concentrations across studies for various types of environments.

Other UV filters marketed in the United States for which there are more than 10 location/event sets of observations extracted from reviewed publications (see Table C.1 in Appendix C) include: homosalate (46 reported

___________________

2 One study is considered a set of measurements at an individual location from a single publication. One publication with measurements taken at multiple locations was counted by the committee as multiple studies.

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Image
FIGURE 4.6 Oxybenzone concentrations are arranged as four bins from high to low (left to right) in relation to the combined influences of source strength, proximity to source, and water residence time.
Image
FIGURE 4.7 Log-scale distributions of octinoxate concentrations in water (ng/L) for three broad types of sampling areas. Each data point is the maximum value for a sampling location and event and some sampling consists of only one grab sample. Average or median values would be lower. All non-detect or limit of quantifications are included at their stated values. In the absence of information, non-detect values are included in the range of other non-detect values (around 5 ng/L) and do not influence the overall shape or information content of the distributions.
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Image
FIGURE 4.8 Log-scale distributions of octocrylene concentrations in water (ng/L) for three broad types of sampling areas. Each data point is the maximum value for a sampling location and event and some sampling consists of only one grab sample. Average or median values would be lower. All non-detect or limit of quantifications are included at their stated values. In the absence of information, non-detect values are included in the range of other non-detect values (around 5 ng/L) and do not influence the overall shape or information content of the distributions.

observations, 38 detections, maximum value of 2,812 ng/L [2.8 μg/L] at a beach in Hong Kong Harbor [Tsui et al., 2014b]; avobenzone (32 reported observations, 25 detections, maximum value of 1,770 ng/L [1.77 μg/L] for a beach in the Canary Islands, Spain [Sánchez Rodríguez et al., 2015]); dioxybenzone (36 observations but all either non detected or at very low levels [e.g., Kung et al., 2018; Tsui et al., 2014b]), octisalate (35 observations, 25 detections, maximum value of 1,030 ng/L [1.03 μg/L] at a beach in Hong Kong Harbor [Tsui et al., 2014b]); padimate O (49 observations, 18 detections, maximum value of 191 ng/L [0.19 μg/L] for a beach in South Carolina [Bratkovics et al., 2015]); and sulisobenzone (30 observations, 20 detections, maximum value of 339 ng/L [0.34 μg/L] for a beach in Hong Kong Harbor [Tsui et al., 2014b]). Few studies searched for all compounds and many collected only a single sample and no replicates at a single, discrete point in time. Therefore, the rankings may not reliably reflect the frequency of the compounds in water or reflect close to average concentrations over a day or other temporal scale. Some compounds were found above 1,000 ng/L (1 μg/L) and this appears to be a useful value for demarcating relative concentrations. Organic UV filters with some concentrations exceeding 1,000 ng/L (1 μg/L) include oxybenzone, octocrylene, octinoxate, homosalate, avobenzone, and octisalate. The highest concentrations are typically observed in recreational areas including marine and freshwater beach environments. Several studies reported a total for organic UV filters. The two locations with the highest reported total concentrations of organic UV filters were Folly Beach in South Carolina with 7,806 ng/L (7.8 μg/L; representing seven UV filters; Bratkovics et al., 2015) and Enogerra Reservoir in Australia with 7,330 ng/L (7.33 μg/L; representing ten UV filters; O’Malley et al., 2021). Given the limited number of reported totals, the values may be greater as additional data are developed. Additionally, it may be possible to make total estimates by extrapolating from the current data set with respect to individual UV filter concentrations relative to totals.

In addition to the committee’s analysis, individual studies show variation in UV filter concentrations across spatial as well as temporal scales. Spatial variation is strongly influenced by the nature of the source and proximity to the source as well as environmental setting. Temporal variability occurs seasonally as well as diurnally or in

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

response to short-term events. For example, studies conducted over time in the vicinity of recreational areas show distinct seasonal variations (Bargar et al., 2015; Bratkovics et al., 2015; Langford and Thomas, 2008; Sánchez Rodríguez et al., 2015; Sankoda et al., 2015; Schaap and Slijkerman, 2018; Tsui et al., 2014b, 2017). Studies conducted over time scales of hours at recreational bathing areas indicate that higher concentrations occur when bathing is happening and generally diminish between such times (Labille et al., 2020; O’Malley et al., 2021b; Picot Groz et al., 2018; Tovar-Sánchez et al., 2013). The same beach at a popular time of the year can exhibit much lower concentrations of UV filters in the water depending on the weather (e.g., rain or sun; Bargar et al., 2015). Most studies that include transects from beaches toward offshore report a decline in UV filter concentrations with distance (e.g., Bargar et al., 2015; Labille et al., 2020; Mitchelmore et al., 2019; among others). While this was not seen in some locations—possibly due to local hydrology and/or additional nonrecreational sources (Mitchelmore et al., 2019)—overall, these patterns support the premise that bathing activities are a source of localized elevations of UV filters in water. Offshore coral reef environments typically exhibit lower concentrations of UV filters as compared to fringing reefs that are present in areas adjacent to beaches. Furthermore, the vast majority of studies have measured only surface waters given that these are easier to sample from and likely more conservative values. However, many organisms (e.g., mobile, pelagic, and/or benthic organisms including corals) are present in deeper waters and very few studies have addressed concentration changes with depth. Tsui et al. (2017) suggested that surface levels were 40 times higher than those at depth. However, these observations are from data collected at the same location but at two different times (Tsui et al., 2014b, 2017).

Recreational use of beaches appears to be a major factor influencing UV filter concentrations. Beaches and coastal areas adjacent to beaches with higher use exhibited higher concentrations of UV filters than beaches that were remote or had smaller degrees of visitation due to seasonal variation in visitation (e.g., Bargar et al., 2015; Bratkovics et al., 2015; Mitchelmore et al., 2019). Beach visitation from local populations and tourists appears to be an important driver influencing the concentration of UV filters in nearshore aquatic environments (Casas-Beltrán et al., 2021).

Organic UV Filters in Surface Microlayer (SML)

The presence of UV filters at the water surface and potentially associated with the SML was examined in several studies. Fagervold et al. (2019) investigated the relative concentrations of UV filters including oxybenzone and avobenzone in the SML and surface water of the main beach in Banyuls-sur-Mer and a nearby artificial lake in Villeneuve-de-la-Raho; the authors note that these two water bodies in France receive highly increased tourism pressure during the summer months. They also note an “oily layer” at the surface when many swimmers were present during the summer (potentially indicative of sunscreen inactive ingredients). The clearest distinction in concentrations between surface water and SML were in the lake, perhaps because of less wind and wave disturbance (Fagervold et al., 2019). Winds and waves may have mixed the waters, diluting concentrations and diminishing differences between the SML and surface water. For lake sampling conducted in August 2016, avobenzone reached a peak of 531 ng/L (0.531 μg/L) in the SML, which was almost 40 times as high as was found in surface water. In July 2016, avobenzone was approximately 20 ng/L in the SML which is about two times as high as in the surface water. Oxybenzone was present at lower concentrations in the SML (about 5–8 ng/L [0.005–0.008 μg/L]) and approximately two to three times as high as in the surface water of the lake.

Schaap and Slijkerman (2018) sampled the SML in waters adjacent to Sorobon Beach in Bonaire in August 2016. Octocrylene concentration was 1,950 ng/L (1.95 μg/L) in the SML, which was 2.4 times as high as it was in the water column. For oxybenzone, the SML concentration was 900 ng/L (0.9 μg/L) which was lower than the water column measurement of 1,540 ng/L (1.54 μg/L). For the August 2016 sampling, concentrations were lower in the SML and surface waters at locations distant from the beach including a mangrove area and a reef.

Bargar et al. (2015) collected surface water and SML for bays in the U.S. Virgin Islands in December 2013. However, because of limited data, statistically significant differences were found only for comparisons for octinoxate at Trunk Bay. Based on comparison of bar charts in the Bargar et al. (2015) paper, octinoxate is about seven times higher in the SML as compared to surface water; on a qualitative basis, octinoxate appears higher in the SML as compared to surface water for the other two bays, Hawksnest and Cinnamon.

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

The relative degree of organic UV filters in the SML as compared to the surface water likely reflects their physical chemical properties with more lipophilic and lower water-soluble compounds—avobenzone and octinoxate—showing relatively higher partitioning to the SML than the more water soluble oxybenzone.

Organic UV Filters in Sediment

Data on UV filters in sediments are compiled in Table C.3 (Appendix C) and maximum reported values for locations and sampling events are summarized in Figure 4.9.

There are fewer observations for UV filters in sediments than for UV filters in water. However, the UV filter data in these media share some characteristics. The most commonly observed UV filters in sediments (i.e., oxybenzone, octinoxate, and octocrylene) are the same as those for water which reflects to some degree UV filters that have been included in studies. Mitchelmore et al. (2019) did not find spatial correlations between water and sediment concentrations. Small-scale variability as well as variations in organic content of sediments and depositional patterns likely confound obscure relationships involving pairs of water and sediment data. It is expected the spatial patterns of the more hydrophobic UV filters (e.g., octinoxate and octocrylene) will reflect the extensive literature on what is known for other hydrophobic chemicals introduced to aquatic environments; predominant factors are proximity to sources, dispersion patterns for suspended sediments, and depositional environments that have higher levels of organic content, and smaller particle sizes (i.e., fine sands and silts). The fate properties described for the UV filters earlier in this chapter provide insight into UV filter specific characteristics that will govern their distribution in the water column and sediments. The expected distribution patterns of UV filter in sediment are evident in studies conducted by Barón et al. (2013), Combi et al. (2016), Fagervold et al., 2019, and Mizukawa et al. (2017). Domínguez-Morueco et al. (2021) has noted that elevated concentrations of octinoxate and octocrylene on suspended sediments (> 1,000 ng/g in some cases in surface water and up to 7,300 ng/g [octinoxate] and 33,000 ng/g [octocrylene] for wastewater) indicates the potential for these UV filters to distribute to sediments.

Image
FIGURE 4.9 Distributions of UV filters in sediments (ng/g DW). Values are the maximum reported concentrations for individual location and event sampling from Table C.1 in Appendix C. Values reported as non-detected are assigned a concentration of 0.1 ng/g which approximates the lowest reported values. Their inclusion is not expected to influence the overall shape or information content of the distributions but provides an overall picture of the available data.
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

When comparing water to sediment in various papers and the tables in this report, it is important to recognize that commonly reported sediment and water concentrations are not equivalent on a mass per mass basis. The highest concentrations of UV filters in sediment (measured in ng/g) are 10–100 times as high as the highest concentrations of UV filters in water (measured in ng/L) on a mass per mass basis. This reflects the tendency of some UV filters to partition to sediment or particulate matter as well as the influence of dispersion and transport. There is also an important distinction between actual occurrence data and the data that may be expected based on environmental fate characteristics. Occurrence data reflect not only the various fate processes, but also the actual environmental releases. One could easily have a highly hydrophobic chemical that should only ever be found in sediments, but is never found there due to its low use or the environmental sampling not actually being representative of the whole sediment compartment.

The variable physico-chemical properties and fate characteristics of the UV filters appear to influence the relative concentrations of the chemicals in sediments compared to water. For example, all measurements of oxybenzone are less than 100 ng/g (0.1 μg/g) DW in sediment (Figure 4.9). In contrast, values of up to 2,200 ng/g (2.2 μg/g) DW for octocrylene were reported for the Ebro River in Spain (Gago-Ferrero et al., 2011a) and up to 447 ng/g (0.45 μg/g) DW for octinoxate were reported for Hong Kong Harbor in China (Tsui et al., 2015). These values may relate to the higher solubility for oxybenzone (hence partitioning to the water rather than sediment) versus octocrylene, which has a higher Kow and likely partitions to the sediment. Based on available data, only octocrylene was observed above 1,000 ng/g (1 μg/g) DW in sediments (i.e., a ppm in familiar language) and most were less than 100 ng/g DW.

Octocrylene appears to have a widespread propensity to be found in sediments, likely due to its widespread use in products, lipophilicity, and potential persistence as described earlier. Muz et al. (2020) utilized passive samplers to examine the distribution of bioavailable chemicals in sediments across several European countries and Australia. Octocrylene was prevalent in freshwater, estuarine, and marine sediments and exhibited elevated concentrations near industrial areas. Muz et al. (2020) noted the bioaccumulative potential of octocrylene and potential for contributing to ecological risk.

Inorganic UV Filters in Water and Sediments Including the Sea Surface Microlayer (SML)

In comparison to organic UV filters for which there are observations about concentrations in surface waters, data are limited on the incremental influence of inorganic UV filters on concentrations of titanium and zinc in the environment. These metals occur naturally in water and sediments and any measurements need to be considered in the context of ambient levels as advised by EPA’s Framework for Metals Risk Assessment (EPA, 2007). Because zinc and titanium vary regionally throughout the United States, ambient levels are region-specific, adding further complexity.

Labille et al. (2020b) report TiO2 along Mediterranean beaches in France. The authors sampled the surface layer (SML) and the water column. Concentrations of TiO2 were higher in the surface layer (100–900 μg/l) compared to the water column (20–50 μg/l). For ZnO, Labille et al. (2020b) observed concentrations of 10–15 μg/L in the surface layer and 1–3 μg/L in the water column. Labille et al. (2020b) observed that concentrations of TiO2 and ZnO were lower beyond the bathing zone. Tovar-Sánchez et al. (2013) observed a similar pattern of higher concentrations of Ti in the SML relative to the water column at popular beaches on Majorca Island in Spain. For the SML, they measured midday concentrations of 12.1–37.6 μg/L for the popular beaches and 23.7 μg/L for a control location; in contrast for the water column, Ti was reported as non-detected. For Zn in the SML, they observed mid-day concentrations of 3.3 to 10.8 μg/L and 0.8 μg/L at a control location; in contrast, Zn concentrations were reported to be 0.1 to 1 μg/L in the water column. Gondikas et al. (2014) examined the possible influence of sunscreens on Ti levels in a recreational lake in Vienna, Austria. They sampled throughout the year and observed elevated levels during the bathing season when Ti concentrations were observed between approximately 1 and 4 μg/L. During this period the Ti/Al ratio also increased supporting the premise that the Ti was introduced. While these studies relate to bathing beaches and receiving waters where measurements have been made, estimates of concentrations in WWTP effluents provide an indication of upper bounds for in-stream concentrations associated with wastewater inputs. Keller (2021 presentation to the committee) indicated that—based on inspection of bar

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

charts—ZnO in treatment effluents could range between approximately 8 and 11 μg/L and that TiO2 could range between approximately 4 and 70 μg/L. These discharges will undergo varying degrees of dilution and settlement of aggregated particles in the receiving waters. A substantial portion of the ZnO and TiO2 in wastewater partitions to the sewage solids within the treatment system yielding sewage solid concentrations of 10s to 100s μg/g.

Nano-sized TiO2 and ZnO are expected to aggregate in the water column and eventually partition to sediments (Pintado-Herrera and Lara-Martin, 2020). Their partitioning may be affected by environmental conditions related to pH, dissolved ions, salt and natural organic matter (NOM) concentrations. Specifically, neutral pH and anion concentration over 5 × 10−3 M cause aggregation and settling (Labille et al., 2010). Nano-TiO2 without coatings can remain in suspension in fresh water for a short time when NOM concentrations range from 5 to 10 mg/L but in seawater will quickly aggregate and settle (Botta et al., 2011; Klaine et al., 2008; Petosa et al., 2010). Partitioning is generally recognized as the fate of nano-particles released to aquatic systems with organic and inorganic particles (Keller et al., 2013). Pintado-Herrera and Lara-Martin (2020) refer to a study that measured concentrations of Ti and Zn in sediments but note that it is virtually impossible to separate incremental loadings from sunscreens from other sources, including deposition and redistribution from natural sources.

Occurrence distributions similar to the above collections for organic UV filters in different source waters is not currently available. A main factor influencing the lack of reported data on ZnO and TiO2 inorganic UV filter concentrations is the analytical difficulty in distinguishing sunscreen formulation and uses of ZnO and TiO2 from other anthropogenic uses of these materials or natural sources of these metals and minerals. Past attempts to fingerprint TiO2 sources and attribute sunscreens as a source of TiO2 in bathing areas used trace elements present with titanium. From these limited studies, TiO2 concentrations are likely in the 1 to 10 μg/L range in areas with high swimmer activities. A recent report has advanced single particle time of flight mass spectroscopy techniques, which utilize size and elemental fingerprinting techniques along with machine learning to differentiate TiO2 sources in water and sediment/soil (Bland et al., 2022). Such efforts may be useful in the future to attribute differing ZnO or TiO2 sources, including from marketed sunscreen products.

ANALYTICAL CHEMISTRY CONSIDERATIONS

Organic UV Filters

For many common environmental organic contaminants or groups of chemicals standard protocols exist outlining how samples are collected, processed, and analyzed (e.g., EPA method 1668a-c for polychlorinated biphenyls and EPA methods 8081 and 8082 for polycyclic aromatic hydrocarbons). As organic UV filters are an emerging class of environmental contaminants, there are currently no standard analytical methods. Environmental monitoring programs have typically measured multiple UV filters in a sample (e.g., from 3–14 UV filters; see Mitchelmore et al., 2021, and Appendix C), which is challenging given the diversity of physico-chemical properties exhibited by UV filters.

A variety of analytical methods, typically either gas chromatography–mass spectrometry (GC-MS) or liquid chromatography–tandem mass spectrometry (LC-MS/MS) based methods have been used in the literature with varying levels of detail reported and QA/QC components included, which can challenge comparisons across studies and potentially the reliability and relevance of the results for exposure analyses and toxicological thresholds (see Mitchelmore et al., 2021, for a review). Employing method blanks, sample replicates (particularly in exposure monitoring studies), and analytical replicates would greatly enhance the reproducibility and reliability of the reported data and help account for any losses or contamination that could occur in a single sample. How a sample is collected and processed influences the concentration of the UV filter reported. Thus, providing information on analytical spike recovery rates is also important, as are assessments into the potential losses that could occur in collecting, extracting, and processing samples including the addition of matrix recovery spikes to assess influences due to the specific composition of the sample. Typically, matrix recovery spikes and other QA/QCs for losses during processing and extraction are not conducted. These are essential to include given the observations of significant losses to various materials that are UV filter independent (e.g., significant loss of octocrylene to glass [Saxe et al., 2021]; losses at multiple processing stages for oxybenzone including to plastic [Conway et al., 2021]).

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

While concentration data are useful, the ability to compare across studies will be limited if detailed methods are not provided. An example of this is whether the water sample is pre-filtered before solid phase extraction (SPE) or not. EPA Method 1694 (EPA, 2007) for organic pesticides recommends filtration if there are particulates visible, with the two phases (i.e., dissolved and particulate/suspended fractions) measured separately. Determining the concentration of UV filters in each of these two fractions may have ramifications for the route of exposure to an organism and implications regarding the reporting of toxicity effects thresholds. Furthermore, the use of unfiltered samples during SPE increases the likelihood that an unknown percentage of the total concentration is extracted via SPE due to the blockage of sites within the SPE media as well as incomplete extraction of particle-bound chemicals. Of critical importance in reporting environmental occurrence data are the levels of a chemical below which reliable data could not be generated (i.e., the lower limit of quantitation; LLOQ). While differences may exist in the ability of a method to detect a compound (i.e., a method detection limit) and reliably quantify that compound (i.e., an LLOQ), the significant uncertainty surrounding any values falling between these two numbers suggest that only values above an LLOQ should be relied upon for quantitative risk assessments. Many studies have not reported LLOQ and other basic parameters that could provide insights into environmental occurrence (see Mitchelmore et al., 2021).

The discrepancies observed between nominal (expected) and measured concentrations from toxicity test exposures highlight some of these issues. There may be additional losses in these test systems that are not relevant to sampling and analytical concerns (i.e., uptake and metabolism by organisms; see Chapter 6). Analytical verification of toxicity test exposure solutions is required to derive accurate toxicity thresholds. How and when these samples are collected during the test is important to ensure accurate representation of test exposures, especially when toxicity test designs do not use a continuous flow-through exposure regime but rather a static-renew design whereby exposure solutions are made and replaced over the course of the experiment, typically daily (see Chapter 6). Once an exposure solution is made, changes can occur over time; the chemical may sorb to containers, degrade via any number of abiotic/biotic processes, be taken up by the organism, or bind to particles and/or precipitate out of solution. Thus, the method used and timing of sampling for analytical verification is key to understanding the fate and exposure of UV filters. However, this component is not often detailed in studies (Mitchelmore et al., 2021). Significant losses in several UV filters over time were observed in a number of studies (e.g., Fel et al., 2019; He et al., 2019b,c; Wijgerde et al., 2020). Losses (i.e., measured compared to nominal concentrations expected) over the time period at sampling were reduced to < LOD–2 percent in the He et al. (2019b,c) studies and losses of up to 91 percent of avobenzone were observed in the Fel et al. (2019) study. Similarly, Wijgerde et al. (2020) reported a measured value of 0.06 compared to the nominal target of 1 μg/L for their oxybenzone exposure solutions. These reduced concentrations may reflect the degradation/loss of the analyte in the exposure chambers and hence the actual concentration an organism is exposed is indeed lower. However, they may also be due to analytical losses during sampling, processing, analysis, or a combination of each. If the latter is not assessed, this analytical verification could be under-reporting the actual exposure concentrations resulting in overestimations of toxicity. Losses during sample processing were highlighted in a recent coral oxybenzone toxicity study demonstrating the importance of matrix recovery and QA/QC spikes in addition to analytical recovery spikes (Conway et al., 2021). It is also possible that contamination of the sample can occur at any of the sampling and processing steps as can interaction with specific matrix components that may result in lower or higher values than expected. Figure 4.10 details some of the ways in which losses/gains in UV filter concentrations may occur during sampling, extraction, and analysis.

Inorganic UV Filters

Analytical determination of inorganic UV filters can be challenging. Without differentiating sources, existing analytical methods include:

  • Mass concentrations of bulk elemental (e.g., Ti, Zn) analysis performed by chemical (e.g., strong acids, peroxides) addition, thermal digestion with analysis by ICP-MS (Inductively coupled plasma mass
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Image
FIGURE 4.10 Overview of some of the ways that analytical measurements may not accurately reflect the concentrations in environmental exposure or toxicological exposure solutions.
  • spectrometry) or ICP-OES (Inductively coupled plasma–optical emission spectrometry) (Lomer et al., 2000; Szakal et al., 2014).
  • Mass concentration using size separation with membranes, size exclusion chromatography, or field-flow fractionation followed by bulk elemental analysis using ICP-MS, or comparable analytical methods.
  • Number and mass concentrations of different particle sizes using single particle ICP-MS or TOF-MS (time-of-flight mass spectrometry) (Azimzada et al., 2020; Helsper et al., 2016; Mozhayeva and Engelhard, 2020; Venkatesan et al., 2018). A benefit of sp (single particle)-TOF-MS is the ability to quantify multiple elements simultaneously in single nanoparticles.

Figure 4.11 summarizes the general approach toward analysis of UV filters. Figure 4.12 provides examples of their application to inorganic UV filters at different stages of their use (i.e., life cycle). Similar analytical schemes have been described well for nanoparticles in other matrices (e.g., Baalousha and Lead, 2015; Singh et al., 2014). The summarized methods have detection limits of approximately 1 μg/L in water or other aqueous solutions. Mass concentrations in water using colorimetric methods have been published (Gokdere et al., 2019), but are not widely utilized, and detection limits on environmental samples remain unclear. Biological tissues and cells can be readily digested to release and quantify Ti or Zn composition. Enzymatic digestates can maintain the size and shape of inorganic UV filters when applied to biological samples. Electron microscopy imaging with elemental analysis techniques can be applied to characterize the size, shape, and composition of UV filters. The most significant analytical challenge regarding inorganic UV filters is the lack of a robust analytical protocol to differentiate inorganic UV filters from other sources of similar minerals/elements, and lack of high throughput analytical methods to image (e.g., electron microscopy) the size, shape, and number of particles in water or biological samples (Singh et al., 2014).

A wide variety of analytical schemes have been applied to quantify nanoparticles in various sources. For example, single-particle ICP-MS has been conducted on over-the-counter sunscreen samples after diluting water containing 1 percent surfactant solution (Triton-X) and then sonicated until aggregates were broken up. Analysis indicated most TiO2 in five different sunscreen samples were in the size range of 20 to 70 nm, assuming spherical particles, and contained 60,000 to 200,000 particles per mL of sunscreen (Dan et al., 2015). The spherical

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Image
FIGURE 4.11 Method development strategy to detect or quantify UV filters in complex matrices. SOURCE: Szakal et al., 2014, ACS AuthorChoice.
Image
FIGURE 4.12 Summary of analytical methods for inorganic UV filters across their life cycle from manufacturing through use and into the environment.
*Common digestion methods for water, filtered particles, sediment, or tissues include boiling or microwave digestion of samples added to solutions containing strong acids (nitric, sulfuric, hydrochloric, hydrogen fluoride). **High organic content matrices, cells or tissues will require hydrogen peroxide, strong bases or enzymes prior to acid digestion. ***Common chemical analysis of after digestion include ICP-OES, ICP-MS, GFAA. + Dichloromethane was more effective than acetone or acetonitrile. SOURCE: Courtesy of committee member Paul Westerhoff.
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

assumption and particle sizes from the sp-ICP-MS study appear consistent with TEM images of solvent-extracted TiO2 from other sunscreen products (Hanigan et al., 2018). The assumption of spherical particles may not be acceptable for all inorganic UV filters (e.g., ZnO), as some are present in higher aspect ratio or rod-shaped materials (Hanigan et al., 2018). Single particle ICP-MS and TOF-ICP-MS are emerging as the most well developed analytical techniques to study TiO2 and ZnO UV filters in environmental samples.

Developing methods to advance particle analysis in various matrices is an active field of research. Liquid-liquid separation techniques such as cloud-point extraction have been explored to separate and concentrate particles from water (Duester et al., 2016) or sediments (El Hadri and Hackley, 2017) to facilitate their imaging or chemical analysis (Mortada, 2020). Cloud-point extraction utilizes surfactants to attach to particle surfaces, and then at near freezing water temperatures the surfactants form a separate phase containing particles that can be collected and analyzed. Colorimetric methods to detect nanoparticles in water is an area of emerging development (Bi et al., 2018; Corredor et al., 2015), but these largely remain the purview of academic studies for TiO2 or ZnO (Bulbul et al., 2018). These methods rely upon color changes in redox-active dyes, to indicate relative surface area and then mass concentrations of equivalent (standard) nanoparticles. Several government agencies have identified “nanoparticle sensors” as a research need (e.g., NSTC, 2020).

FINDINGS AND KNOWLEDGE GAPS

Fate and exposure of UV filters in aquatic environments depend on the characteristics of the chemicals and on the features of the environment including receiving water characteristics (e.g., water flow, mixing, energy, residence time, flushing). Exposures to ecological receptors may involve UV filters that have partitioned into the SML, water column, organic particulates in the water column, and/or sediments. Exposures to aquatic organisms can include direct exposures to UV filters in the dissolved phase in water as well as exposures to these compounds via ingestion of particles to which the filters are adsorbed. The water solubility and lipophicity of the compounds will govern the partitioning between water, organic matter, and accumulation into tissues of organisms. Accumulation into biota and potential for trophic transfer can occur from any of these compartments and is described in Chapter 5. The nature and degree of this partitioning is governed by fate processes associated primarily with water solubility and sorption potential, which influence partitioning (i.e., dissolved and particulate fractions in water and to sediment), and degradation processes (i.e., photolysis and biodegradation), which influence persistence of the chemical or formation of degradation products. These processes differ among the various organic and inorganic UV filters with the latter being influenced also by aggregation processes.

Exposures can be estimated from models and evaluated from available measurement data. Equally important is that the measurement data can be used to verify and improve the models. Exposure estimates from models can be integrated with measurement data to characterize exposure to an ecological receptor. Modeled estimates are typically designed to provide ranges, including upper estimates, of possible exposures based on estimates of loadings to the environment combined with fate characteristics (including adsorption and treatment within wastewater treatment systems for down-the-drain sources) and environmental fate and transport processes in the receiving water environment. This chapter points out that models typically yield higher estimates than what is found when UV filters are measured for the same systems. Differences likely reflect assumptions made as well as lack of knowledge for modeling, measurement, and analytical purposes. These differences indicate the value of taking a combined approach of modeling and measurements for estimating exposure.

This chapter shows that there is variability among the UV filters with respect to sources, fate processes, and occurrences in the environment. Variability also reflects the design limitations of environmental monitoring programs conducted to date. This underscores the importance for exposure assessment to consider UV filters individually or as groups as well as the key variations among sources and environmental settings.

Fate Characteristics

Finding: Organic UV filters are expected to display an array of environmental behaviors based on their range of measured physico-chemical properties. With the notable exceptions of ensulizole, aminobenzoic acid, trolamine

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

salicylate, and sulisobenzone, the organic UV filters are generally hydrophobic and thus would be expected to partition to organic fractions, including particles and sediments. Oxybenzone is moderately water soluble, with less distinct partitioning between aqueous and organic sediment compartments.

Finding: There is a wide range of UV filter behaviors with respect to biodegradation and photostability. Avobenzone, dioxybenzone, ecamsule, ensulizole, and octocrylene have been shown to have low biodegradability, while oxybenzone and sulisobenzone appear to be relatively photostable in laboratory settings. However, the laboratory settings may not accurately recreate environmental conditions.

Finding: The existing fate and transport data suggest that aggregation of TiO2 and ZnO with other particles in the water column leads to formation of larger sized particles as aggregates that settle out of the water column and accumulate in river, lake, or estuary sediments. Increasing salinity leads to more rapid aggregation of particles. Whereas dissolution of TiO2 has been reported, it is likely to be a very slow process, and water quality drivers for this process remain ill defined. ZnO does dissolve to zinc ions in water, after which the fate of zinc ions is influenced by pH, redox, solids, and anions present in the water.

Knowledge Gap: Most understanding of physico-chemical parameters impacting the fate and subsequent bioavailability of UV filters is derived from laboratory or pure water experiments and modeling platforms. While these models are fairly robust, they have been validated in a limited number of freshwater and estuarine aquatic environments. For example, the sorptive behavior of organic UV filters may be enhanced in marine environments (as compared to behavior in fresh or pure water). Given the potential for particle-bound exposures for filter feeding aquatic organisms, such data could be considered critical for risk assessments.

Knowledge Gap: The phototransformation of UV filters in aquatic environments is significantly understudied. Given the relative importance of molecular-scale interactions for UV filter photostability, data on the photostability of UV filters in representative aquatic environments as well as the formation of phototransformation products could be considered critical for risk assessments.

Knowledge Gap: Limited data exists on the role of some common coatings (e.g., aluminum, silica, polydimethylsiloxane) applied to inorganic UV filters on aggregation or dissolution of TiO2 and ZnO, which is important to understand to illuminate exposure routes and bioavailability of UV filters.

Estimated and Measured Concentrations

Finding: For several organic UV filters—oxybenzone, octocrylene, homosalate, avobenzone, and octinoxate—the available data show that the highest measured environmental concentrations in water are in the range of 1 to 10 μg/L, though most measurements for these and all measurements for other organic UV filters are below 1 μg/L.

Finding: Except for octocrylene and octinoxate, which have maximum recorded concentration values between 0.1 and 2.4 μg/g DW, all other UV filters exhibit maximum recorded concentrations in sediments below 0.1 μg/g DW.

Finding: Occurrence measurements of UV filters reflect the spatial and temporal variability in the local setting (e.g., magnitude of sources associated with rinse-off from recreation, wastewater discharges, and potential nonpoint sources; proximity to sources; and the residence time of the receiving water). The highest measured concentrations of most UV filters occur in shallow waters, near/within recreational areas (e.g., swimming beaches), and during the day.

Knowledge Gap: Refined exposure models are needed that can account for pulse inputs of UV filters associated with temporal variations in recreational use and subsequent spatial and temporal variations in exposures of ecological receptors.

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×

Knowledge Gap: Most data for UV filters in river water come from outside of the United States and there is very limited data for the United States. While more data are available for marine waters as compared to freshwater of the United States, these data usually come from single studies and are not of wide coverage.

Knowledge Gap: There are limited data to suggest enrichment of some organic and inorganic UV filters in the SML, which could be verified through further study.

Knowledge Gap: The concentrations of UV filters in sediments are not well characterized or documented with data.

Knowledge Gap: Assessments that account for background concentrations of metals (naturally occurring or other sources of engineered nanomaterials beyond UV filters), the form (dissolved, colloidal/nanoform, particulate) of the metal, and the bioavailability of the metal are needed for assessing risk from inorganic compounds from sunscreens.

Knowledge Gap: Robust monitoring programs are needed to better reflect the concentrations over time and space. This may include minimum replicates and reliable, standardized methods developed to collect, extract, and process samples so that they accurately measure UV filters from environmental samples. This would also include minimum QA/QC procedures (e.g., analytical and matric recovery spikes) and minimum reporting requirements (e.g., method details).

Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 69
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 70
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 71
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 72
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 73
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 74
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 75
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 76
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 77
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 78
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 79
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 80
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 81
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 82
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 83
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 84
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 85
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 86
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 87
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 88
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 89
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 90
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 91
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 92
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 93
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 94
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 95
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 96
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 97
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 98
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 99
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 100
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 101
Suggested Citation:"4 Fate, Transport, and Potential Exposure in the Environment." National Academies of Sciences, Engineering, and Medicine. 2022. Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health. Washington, DC: The National Academies Press. doi: 10.17226/26381.
×
Page 102
Next: 5 Bioaccumulation and Measured Concentrations of UV Filters in Biota »
Review of Fate, Exposure, and Effects of Sunscreens in Aquatic Environments and Implications for Sunscreen Usage and Human Health Get This Book
×
Buy Paperback | $60.00 Buy Ebook | $48.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Regular use of sunscreens has been shown to reduce the risk of sunburn and skin cancer, and slow photoaging of skin. Sunscreens can rinse off into water where people are swimming or wading, and can also enter bodies of water through wastewater such as from bathing or showering. As a result, the ultraviolet (UV) filters - the active ingredients in sunscreens that reduce the amount of UV radiation on skin - have been detected in the water, sediment, and animal tissues in aquatic environments. Because the impact of these filters on aquatic ecosystems is not fully understood, assessment is needed to better understand their environmental impacts.

This report calls on the U.S. Environmental Protection Agency to conduct an ecological risk assessment of UV filters to characterize the possible risks to aquatic ecosystems and the species that live in them. EPA should focus on environments more likely to be exposed such as those with heavy recreational use, or where wastewater and urban runoff enter the water. The risk assessment should cover a broad range of species and biological effects and could consider potential interacting effects among UV filters and with other environmental stresses such as climate change. In addition, the report describes the role of sunscreens in preventing skin cancer and what is known about how human health could be affected by potential changes in usage. While the need for a risk assessment is urgent, research is needed to advance understanding of both risks to the environment from UV filters and impacts to human health from changing sunscreen availability and usage.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    Switch between the Original Pages, where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

    « Back Next »
  6. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  7. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  8. ×

    View our suggested citation for this chapter.

    « Back Next »
  9. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!