3.1. ALIPHATIC LOW CARBON RANGE FRACTION: C5–C8 (EC5–EC8)

The aliphatic low carbon range fraction includes straight-chain, branched, and cyclic alkanes and alkenes; examples include n-pentane, n-octane, 2-methylpentane, cyclohexane, and 1-hexene. Toxicity assessment and surrogate selection for the aliphatic low carbon range fraction is detailed in the PPRTV assessment for this fraction (U.S. EPA, 2022a). This section provides a summary of the approach and results; further detail is available in the PPRTV assessment.

Toxicity values were identified for seven aliphatic low carbon range compounds and one mixture. Tables 1 and 2 provide summaries of the oral and inhalation noncancer toxicity values, critical effects, and key studies. In February 2018 and again in August 2021, literature searches were conducted using a multistep process for the mixtures and individual compounds with toxicity values and for other mixtures and compounds that are relevant to the fraction. The primary toxicological endpoints identified for the fraction were neurological, hepatic, body weight, gastrointestinal [GI], respiratory, and developmental effects. Among members of the fraction that have undergone in vivo toxicity testing, the data available to assess consistency in effects are limited for effects on endpoints other than body weight. In addition to the scarcity of developmental toxicity data for members of the fraction, an important data limitation is the lack of chronic systemic toxicity information for all but three members of the fraction. Only cyclohexene, methylcyclohexane, and commercial hexane have been tested in comprehensive systemic toxicity studies in animals exposed for at least 1 year, all by the inhalation route of exposure. Furthermore, most of the oral toxicity studies are <13 weeks in duration, and few examined comprehensive endpoints.

Table 1

Available RfD Values for Aliphatic Low Carbon Range Fraction (C5–C8 [EC5–EC8]).

Table 2

Available RfC Values for Aliphatic Low Carbon Range Fraction (C5–C8 [EC5–EC8]).

Available oral and inhalation toxicity data for aliphatic low carbon range compounds did not show much consistency across fraction members in terms of toxicological effects or potencies. Thus, there was no basis to identify a surrogate mixture or compound that is representative of the effects and potency of the fraction as a whole, so the most potent component compounds and mixtures were considered as the basis for indicator chemical selection.

Two options are presented for assessment of oral noncancer effects for this fraction. The first is for use when available analytical chemistry data do not identify concentrations of individual chemicals composing this fraction. In this case, the subchronic and chronic provisional reference doses (p-RfDs) (0.05 and 0.005 mg/kg-day, respectively) for cyclohexene are recommended as the indicator chemical for the aliphatic low carbon range fraction. The p-RfDs for cyclohexene are based on hepatic toxicity. The available oral toxicity data for aliphatic low carbon range compounds do not demonstrate significant consistency across fraction members in terms of toxicological effects or potencies. Therefore, there is no basis to identify an indicator chemical or mixture that is representative of the effects and potency of the fraction as a whole. Cyclohexene, among the most potent component compounds and mixtures considered in this fraction, is the selected indicator chemical (see discussion of method in Section 1.3.1). Although the RfDs for cyclohexene are not the lowest available, the subchronic and chronic p-RfD values for n-heptane (0.003 and 0.0003 mg/kg-day, respectively) are not recommended for the following three reasons. First, the n-heptane p-RfDs are screening values based on an read-across analysis and therefore carry additional uncertainty associated with the analogue approach. Second, the analogue upon which the values are based (n-nonane) is outside (C9 [EC9]) the carbon range of the fraction. Third, the chronic p-RfD for n-heptane is highly uncertain, derived with a composite uncertainty factor (UF_C) of 10,000. Evaluation of available data [see U.S. EPA (2022a) for further detail] suggests that use of the cyclohexene p-RfD values is reasonably anticipated to be protective for effects associated with exposure to other constituents of the fraction. These toxicity values are shown in bold in Table 1 to indicate their selection as the indicator chemicals for the fraction.

If the available analytical chemistry data quantify the concentrations of n-hexane, methylcyclopentane, cyclohexene, n-heptane, or 2,4,4-trimethylpentene separately from the remainder of the low carbon fraction, it is recommended that HQs for the individual chemicals with analytical data be calculated and an HI for the mixture be developed using the calculated HQs.

For subchronic oral exposures, the following subchronic p-RfDs can be used as the denominator in the HQ equations: n-hexane (0.3 mg/kg-day), methylcyclopentane (0.4 mg/kg-day), cyclohexene (0.05 mg/kg-day), n-heptane (0.003 mg/kg-day), and 2,4,4-trimethylpentene (0.1 mg/kg-day). In this alternative approach, the subchronic p-RfD (0.05 mg/kg-day) for cyclohexene is recommended for use with the remainder of the fraction, including any other fraction members analyzed individually.

For chronic oral exposures, the following chronic p-RfDs can be used in the denominator of the HQ equations: cyclohexene (0.005 mg/kg-day), n-heptane (0.0003 mg/kg-day), and 2,4,4-trimethylpentene (0.01 mg/kg-day). In this alternative approach, the chronic p-RfD (0.005 mg/kg-day) for cyclohexene is recommended for use with the remainder of the fraction, including any other fraction members analyzed individually.

As with the oral noncancer assessment, two options are presented for inhalation noncancer assessment of this fraction. If available analytical chemistry data do not identify concentrations of individual chemicals composing this fraction, the lowest subchronic and chronic provisional reference concentrations (p-RfCs) among the compounds in this fraction, for n-hexane and n-heptane, respectively, are recommended for the aliphatic low carbon range fraction. These toxicity values are shown in bold in Table 2 to indicate their selection as the indicator chemical for the fraction.

In cases where the available analytical chemistry data quantify the concentrations of n-pentane, n-hexane, cyclohexane, or n-heptane separately from the remainder of the low carbon fraction, it is recommended that HQs for the individual chemicals with analytical data be calculated and an HI for the mixture be developed using the calculated HQs.

For subchronic inhalation exposures, the following subchronic p-RfCs can be used as the denominator in the HQ equations: n-pentane (10 mg/m³), n-hexane (2 mg/m³), cyclohexane (18 mg/m³), and n-heptane (4 mg/m³). In this alternative approach, the subchronic p-RfC for n-hexane (2 mg/m³) is recommended for use with the remainder of the fraction, including any other fraction members analyzed individually.

For chronic inhalation exposures, the following chronic p-RfCs can be used as the denominator in the HQ equations: n-pentane (1 mg/m³), n-hexane (0.7 mg/m³), cyclohexane (6 mg/m³), cyclohexene (1 mg/m³), and n-heptane (0.4 mg/m³). In this alternative approach, the chronic p-RfC for n-heptane (0.4 mg/m³) is recommended for use with the remainder of the fraction, including any other fraction members analyzed individually.

Few data with which to assess the carcinogenic potential of compounds and mixtures in the aliphatic low carbon range fraction are available. No human or animal studies examining carcinogenicity were located for any compound or mixture other than commercial hexane, n-hexane, cyclohexene, and 2,2,4-trimethylpentane. Only the data for commercial hexane were considered adequate to assess carcinogenic potential, resulting in a weight-of-evidence (WOE) descriptor of “Suggestive Evidence for Carcinogenic Potential” and a provisional IUR (p-IUR) of 2 × 10⁻⁴ (mg/m³)⁻¹ for combined pituitary adenomas and adenocarcinomas in female mice (U.S. EPA, 2009e). None of the mixtures or constituents in this fraction had an OSF from the IRIS database, PPRTVs, HEAST, MassDEP, or TPHCWG. Thus, a provisional OSF (p-OSF) was not derived for the fraction. The only available IUR for members of the aliphatic low carbon range fraction is the screening value for commercial hexane (U.S. EPA, 2009e); this p-IUR is selected to assess inhalation carcinogenicity for this fraction. Table 3 shows the recommended cancer risk estimate for the aliphatic low carbon range fraction.

Table 3

Available Cancer Risk Estimates for Aliphatic Low Carbon Range Fraction (C5–C8 [EC5–EC8]).

3.2. ALIPHATIC MEDIUM CARBON RANGE FRACTION: C9–C18 (EC > 8–EC16)

The aliphatic medium carbon range fraction includes n-nonane, n-decane, and longer chain n-alkanes; a few n-alkenes (e.g., tridecene); branched chain alkanes and alkenes; and alkyl-substituted cycloalkanes. Toxicity values for compounds in this fraction are not available from the U.S. EPA’s IRIS database, or from HEAST, ATSDR, MassDEP, or TPHCWG; PPRTV assessments for n-nonane and n-decane are available. Limited toxicity data are available for n-undecane (TERA, 2004). ATSDR toxicological profiles and inhalation Minimal Risk Levels (MRLs) are available for various jet fuels and kerosene, but these mixtures have a substantial aromatic content and are therefore not suitable to represent the toxicity of this fraction. The toxicity of this fraction may be better represented by dearomatized hydrocarbon streams³ and solvents that fall within this carbon range and have minimal (<1.0%) aromatic content.

A PPRTV assessment for mid-range aliphatic hydrocarbon streams was prepared (U.S. EPA, 2009j) to synthesize the findings of these mixture studies and additional supporting toxicity studies on similar mixtures. Complete descriptions of the studies, as well as details of the derivation of toxicity values for the mixtures, are provided in the PPRTV assessment.

Tables 4, 5, and 6 list the available RfDs, RfCs, and cancer assessments for compounds or mixtures in this fraction. The mixture data are considered preferable to single component data, as previously discussed. The toxicity values for the mid-range aliphatic hydrocarbon stream mixture are the recommended values for this fraction and include subchronic and chronic p-RfCs. In addition, Table 4 contains screening oral toxicity values for mixture data that may be useful in evaluating this fraction, developed in Appendix A of U.S. EPA (2009j). Because the toxicity data based on the three unpublished studies (Anonymous, 1990, 1991a, b as cited in U.S. EPA, 2009a) are not peer reviewed, only screening chronic or subchronic p-RfDs are available for the mixture. The surrogate mixture and oral and inhalation noncancer toxicity values selected to represent the fraction are shown in bold in Tables 4 and 5.

Table 4

Available RfD Values for Aliphatic Medium Carbon Range Fraction (C9–C18 [EC > 8–EC16)^, .

Table 5

Available RfC Values for Aliphatic Medium Carbon Range Fraction (C9–C18, EC > 8–EC16)^, .

Table 6

Available Cancer Risk Estimates for Aliphatic Medium Carbon Range Fraction (C9–C18 [EC > 8–EC16]) of Total Petroleum Hydrocarbons^, .

As Table 6 shows, quantitative cancer risk assessments were not available for individual components of the fraction. The mid-range aliphatic hydrocarbon stream mixture data were considered adequate to develop a quantitative estimate of cancer risk from inhalation exposure. However, because the WOE indicates “Suggestive Evidence of Carcinogenic Potential,” there is some uncertainty associated with the quantification. Appendix A of the PPRTV assessment document on the mid-range aliphatic hydrocarbon streams contains a screening p-IUR (U.S. EPA, 2009j). The screening p-IUR is listed in Table 6 (U.S. EPA, 2009j).

3.3. ALIPHATIC HIGH CARBON RANGE FRACTION: C19–C32 (EC > 16–EC35)

The aliphatic high carbon range fraction includes longer n-alkanes, such as eicosane, and branched and cyclic alkanes. Toxicity values are not available for the individual compounds. A search for toxicity information on eicosane in particular was desirable because MassDEP (1994) suggested it as a reference compound for this fraction, but data supportive of derivation of toxicity values were not identified. Food- and medicinal-grade mineral oils are pure (aromatic-free) mixtures of aliphatic hydrocarbons that correspond to this carbon range fraction and have data suitable for toxicity value derivation. Literature searches on mineral oils were performed and the medical literature on mineral oils was consulted. Subchronic and chronic p-RfDs as well as a cancer assessment, including a WOE of “Inadequate Information to Assess the Carcinogenic Potential” for white mineral oil, were derived in a PPRTV assessment (U.S. EPA, 2009s). Table 7 summarizes the resulting oral noncancer values (a quantitative cancer assessment was not performed). These toxicity values are recommended for assessment of this fraction using a surrogate mixture approach.

Table 7

Available RfD Values for Aliphatic High Carbon Range Fraction (C19–C32, EC > 16–EC35)^, .

3.4. AROMATIC LOW CARBON RANGE FRACTION: C6–C8 (EC6–EC < 9)

This fraction contains aromatic hydrocarbons in the C6–C8 range: benzene, toluene, ethylbenzene, and o-, m-, and p-xylenes (commonly referred to as BTEX) and styrene. It is unclear, however, whether styrene is a constituent of petroleum products. For example, styrene is not reported as a constituent of any of the petroleum mixtures including gasoline, kerosene, jet fuels, diesel fuel, fuel oils, lubricating and motor oils, and crude oil in Potter and Simmons (1998). Gustafson et al. (1997) lists styrene as a constituent for only one mixture, diesel, at a very low percentage of <0.002% (by weight), which may mean that it was detected but was below the quantitation limit. The reference provided for that information is a personal communication prepared for British Petroleum; thus, the information cannot readily be confirmed. Given the uncertainty as to whether styrene is likely to exist in sites of petroleum contamination, it was not considered in the assessment for this fraction.

Tables 8, 9, and 10 list U.S. EPA RfD assessments, RfC assessments, and a cancer assessment, respectively, that are available on the IRIS database for the individual compounds (BTEX) in this fraction. In addition, provisional toxicity values were derived for subchronic oral and inhalation exposure to BTEX (U.S. EPA, 2009b, d, g, t). Because BTEX components are routinely analyzed individually at sites of aromatic hydrocarbon contamination and noncancer toxicity values are available for these components, the recommendation for assessing the noncancer hazard associated with this fraction is to assess the BTEX components individually using an HI approach and their compound-specific toxicity values. For cancer assessments, benzene serves as an indicator chemical, because it is the only chemical in this fraction with IRIS OSF and IUR estimates. The OSF ([1.5 × 10⁻²–5.5 × 10⁻² mg/kg-day]⁻¹) and the IUR ([2.2 × 10⁻³–7.8 × 10⁻³ µg/m³]⁻¹) for benzene (U.S. EPA, 2003b) are used as indicators to estimate cancer risks for this fraction from exposures through the oral and inhalation routes, respectively.

Table 8

Available RfD Values for Aromatic Low Carbon Range Fraction (C6–C8, EC6–EC < 9).

Table 9

Available RfC Values for Aromatic Low Carbon Range Fraction (C6–C8, EC6–EC < 9).

Table 10

Available Cancer Risk Estimates for Aromatic Low Carbon Range Fraction (C6–C8 [EC6–EC < 9]).

3.5. AROMATIC MEDIUM CARBON RANGE FRACTION: C9–C10 (EC9–EC < 11)

Constituents of the aromatic medium carbon range fraction include longer chain and multi-substituted benzenes (e.g., cumene [isopropylbenzene], n-propylbenzene, methylethylbenzenes, and TMBs). Toxicity assessment and surrogate selection for the aromatic medium carbon range fraction is detailed in the PPRTV assessment for this fraction (U.S. EPA, 2022d). This section provides a summary of the approach and results; further detail is available in the PPRTV assessment.

Toxicity values were identified for eight aromatic medium carbon range compounds and one mixture. Tables 11 and 12 provide a summary of the noncancer toxicity values, critical effects, and key studies. Literature searches, OECD SIDS, and the Petroleum HPV Testing Group website yielded relevant toxicity data for four additional compounds and one additional mixture⁴ for use in hazard identification for the fraction. The primary toxicological endpoints identified for the fraction were neurological, hepatic, renal, body weight, hematological, endocrine, and developmental effects. The data available to assess consistency in effects across members of the fraction are limited for effects on endpoints other than body weight. There are no reliable human or animal data for three members of the fraction (n-propylbenzene, and tert- and sec-butylbenzene).⁵ There are body-weight data for 11 members, and there are neurotoxicity data for 9 members. For all other primary toxicological endpoints, there are oral or inhalation data for 5–7 members of the fraction. Most of the animal data are from inhalation toxicity studies. Comprehensive systematic toxicity was evaluated in rats and mice in subchronic and chronic inhalation studies for one member of the fraction (isopropylbenzene). In general, studies for other members of the fraction ranged in duration from 4 to 18 weeks; several of these studies (e.g., diethylbenzenes and TMBs) evaluated only neurological endpoints. Developmental inhalation toxicity studies were available for four members of the fraction (isopropylbenzene, 1,3,5- and 1,2,4-trimethylbenzene, and high flash aromatic naphtha [HFAN]).

Table 11

Available RfD Values for Aromatic Medium Carbon Range Fraction (C9–C10 [EC9–EC < 11]).

Table 12

Available RfC Values for Aromatic Medium Carbon Range Fraction (C9–C10 [EC9–EC < 11]).

The data available to assess consistency in critical effects across members of the fraction are limited for effects on endpoints other than body weight. The potencies are comparable with RfDs being within 1 order of magnitude of one another. Given the limited data, the compounds that resulted in the lowest RfDs for these effects and target tissues were considered as the basis for indicator chemical selection. The subchronic and chronic p-RfDs (0.04 and 0.01 mg/kg-day, respectively) for TMBs are recommended as indicator chemicals for the aromatic medium carbon range fraction. The RfDs for TMBs are based on neurological effects (decreased pain sensitivity). While toxicological data from mixtures such as HFAN might be preferred in some cases, the p-RfD for HFAN is based on a screening value, and the Agency has more confidence in EPA’s IRIS TMB oral assessments as the indicator chemical.

Options for oral noncancer assessment of this fraction are presented based on available analytical chemistry information.

If available analytical chemistry data do not identify concentrations of individual chemicals composing this fraction, the subchronic and chronic p-RfDs (0.04 and 0.01 mg/kg-day, respectively) for TMBs are recommended for the aromatic medium carbon range fraction (U.S. EPA, 2016b). Evaluation of available data suggests that use of the p-RfDs for TMBs is reasonably anticipated to be protective for effects associated with exposure to other constituents of the fraction. The indicator chemical and oral noncancer toxicity values selected to represent the fraction are shown in bold in Table 11.

If the available analytical chemistry data quantify the concentrations of TMBs, n-propylbenzene, n-butylbenzene, sec-butylbenzene, tert-butylbenzene, or isopropylbenzene separately from the remainder of the aromatic medium carbon range fraction, it is recommended that HQs for the individual chemicals with analytical data be calculated and an HI for the mixture be developed using the calculated HQs.

For subchronic oral exposures, the following subchronic RfDs or p-RfDs can be used as the denominator in the HQ equations: TMBs (0.04 mg/kg-day), n-propylbenzene (0.1 mg/kg-day), n-butylbenzene (0.1 mg/kg-day), sec-butylbenzene (0.1 mg/kg-day), and tert-butylbenzene (0.1 mg/kg-day). In this alternative approach, the subchronic RfD for TMBs (0.04 mg/kg-day) is recommended for use with the remainder of the fraction, including any other fraction members analyzed individually.

For chronic oral exposures, the following chronic RfDs or p-RfDs can be used as the denominator in the HQ equations: TMBs (0.01 mg/kg-day), isopropylbenzene (0.1 mg/kg-day), n-propylbenzene (0.1 mg/kg-day), n-butylbenzene (0.05 mg/kg-day), sec-butylbenzene (0.1 mg/kg-day), and tert-butylbenzene (0.1 mg/kg-day). In this alternative approach, the chronic RfD for TMBs (0.01 mg/kg-day) is recommended for use with the remainder of the fraction, including any other fraction members analyzed individually.

In some cases, toxicological data from mixtures such as HFAN might be preferred; however, the p-RfD for HFAN is based on a screening value. The Agency has more confidence in an HI approach as an alternative to the indicator chemical approach than for the surrogate mixture approach for this fraction.

Critical effects and values of RfCs for fraction members show consistency across the fraction with respect to the toxicological effects exerted (most frequently, neurological and developmental effects). The data show that an indicator chemical identifying effects on these targets would be reasonably anticipated to be representative of the effects of the fraction as a whole. Therefore, the compounds that resulted in the lowest RfCs for these effects were considered as the basis for surrogate selection.

As with oral noncancer assessment, two options for inhalation noncancer assessment are presented. If available analytical chemistry data do not identify concentrations of individual chemicals in this fraction, the subchronic and chronic p-RfCs (0.2 mg/m³ and 0.06 mg/m³, respectively) for TMBs (U.S. EPA, 2016b) are recommended as an indicator chemical for the aromatic medium carbon range fraction. The RfCs for TMBs are based on neurological effects (decreased pain sensitivity), and available data generally support the nervous system as a target of the aromatic medium carbon compounds. Use of these values is anticipated to be protective for exposure to other constituents based on available information. The indicator chemical and inhalation noncancer toxicity values selected to represent the fraction are shown in bold in Table 12.

Previously, in the PPRTV TPH mixtures document (U.S. EPA, 2009f), the HFAN subchronic and chronic p-RfCs were recommended for assessing noncancer hazards associated with inhalation route exposures to this fraction, based on a 2009 PPRTV assessment (U.S. EPA, 2009h). In 2016, the U.S. EPA IRIS Program published TMB subchronic and chronic p-RfCs of 0.2 and 0.06 mg/m³, respectively (U.S. EPA, 2016b) that are lower than the respective HFAN values of 1 and 0.1 mg/m³ (U.S. EPA, 2009h) (see Table 12). Because these are IRIS values rather than PPRTVs, these IRIS single chemical values should be used in the indicator chemical approach rather than HFAN-based surrogate mixture approach. The 2009 TPH mixture assessment indicates that the HFAN toxicity values are similar to values for other individual compounds in the fraction, which supports using HFAN as a surrogate for the fraction; however, the 2016 TMB values are much lower than the HFAN values and that logic is not applicable.

If the available analytical chemistry data quantify the concentrations of TMBs, n-propylbenzene, or isopropylbenzene separately from the remainder of the aromatic medium carbon range fraction, it is recommended that HQs for the individual chemicals with analytical data be calculated and a HI for the mixture be developed using the calculated HQs.

For subchronic inhalation exposures, the subchronic RfCs or p-RfCs for TMBs (0.2 mg/m³) or n-propylbenzene (1.0 mg/m³) can be used as the denominator in the HQ equations. In this alternative approach, the subchronic RfC for TMBs (0.2 mg/m³) is recommended for use with the remainder of the fraction, including any other fraction members analyzed individually.

For chronic inhalation exposures, the following chronic RfCs or p-RfCs can be used in the denominator of the HQ equations: TMBs (0.06 mg/m³), isopropylbenzene (0.4 mg/m³), and n-propylbenzene (1 mg/m³). In this alternative approach, the chronic RfC for TMBs (0.06 mg/m³) is recommended for use with the remainder of the fraction, including any other fraction members analyzed individually.

Previously, in the PPRTV TPH mixtures document (U.S. EPA, 2009f), the HFAN subchronic and chronic p-RfCs were recommended for assessing noncancer hazards associated with inhalation route exposures to this fraction, based on a 2009 PPRTV assessment (U.S. EPA, 2009h). By definition, HFAN mixtures must contain a combined total of 75% TMB and ethyltoluene isomers (of which at least 22% is ethyltoluene and at least 15% is TMB) (U.S. EPA, 2009h). As noted previously, in 2016, the U.S. EPA IRIS Program published TMB subchronic and chronic p-RfCs of 0.2 and 0.06 mg/m³, respectively (U.S. EPA, 2016b) that are lower than the HFAN values of 1 and 0.1 mg/m³, respectively (U.S. EPA, 2009h) (see Table 12). Because these are IRIS values rather than PPRTVs, the U.S. EPA has more confidence in using these IRIS single chemical values in a hazard index approach rather than the HFAN values in surrogate mixture approach. The 2009 TPH mixture assessment indicates that the HFAN toxicity values are similar to values for other individual compounds in the fraction, which supports using HFAN as a surrogate for the fraction; however, the 2016 TMB values are much lower than the HFAN values and that logic is not applicable.

Few data are available to assess the carcinogenic potential of compounds and mixtures in the aromatic medium carbon range fraction. No human data were identified. Animal data are limited to studies of 1,2,4-trimethylbenzene and isopropylbenzene. Several limitations were identified in the only carcinogenicity study of 1,2,4-trimethylbenzene reported in U.S. EPA (2016b); these limitations included the use of one rodent species, treatment at a single dose level, and lack of quantitative mortality data. Only data from a newly identified study for isopropylbenzene (NTP, 2009) are considered adequate to sufficiently assess carcinogenic potential. This recently identified study was a 105-week chronic toxicity/carcinogenicity study of isopropylbenzene in rats and mice (NTP, 2009). Statistically significant increases in the incidence of respiratory epithelial adenomas of the nose in both sexes and renal adenoma or carcinoma (combined) in males were observed in rats. Increased interstitial cell adenomas were also reported in the male testis; however, the NTP report stated that these are possibly related to isopropylbenzene exposure. While the incidence of interstitial cell adenomas reported in the highest dose group in the male rats was significantly increased compared to the control group and there was a positive trend in the incidences reported among all exposed groups, the incidence in the high-dose group was within the range for historical chamber controls when studies with all exposure routes were considered. Interstitial cell hyperplasia and adenoma are common proliferative lesions in F344/N rats (i.e., the test species) and reportedly will develop in nearly all male rats of this strain that are allowed to complete their natural life span (NTP, 2009). In mice, the incidences of alveolar/bronchiolar adenomas were significantly increased in both sexes; increased incidences of hemangiosarcomas and follicular cell adenomas in males (possibly related to exposure) and hepatocellular adenomas or carcinomas in females were also noted. Based on these data, the study authors indicated that there was clear evidence of carcinogenicity in male rats and male and female mice, and some evidence of carcinogenic activity in female rats.

None of the mixtures or constituents in this fraction had an OSF or IUR from the IRIS database, PPRTVs, HEAST, MassDEP, or TPHCWG. At this time, the U.S. EPA has not formally evaluated the NTP (2009) study and has not estimated the cancer potency associated with the study results. Thus, a p-OSF or p-IUR was not derived for the fraction.

3.6. AROMATIC HIGH CARBON RANGE FRACTION: C10–C32 (EC11–EC35)

The aromatic high carbon range fraction contains PAHs (e.g., naphthalene, anthracene, BaP, BeP, dibenzo[def,p]chrysene) and benzenes with larger aliphatic substituents (e.g., n-hexylbenzene, phenylcyclohexane). This fraction is further subdivided for the purposes of this document. Unsubstituted PAHs consist of aromatic hydrocarbons comprised of two to six fused aromatic hydrocarbon rings and exclude all compounds with alkyl or other substituents on the ring as well as compounds with anything other than carbon and hydrogen in their composition (i.e., exclude heterocyclic compounds). Substituted PAHs (subPAHs) include alkyl-substituted PAH derivatives such as 1,4-dimethylphenanthrene, 1-methylnaphthalene, and 5-methylchrysene. Carcinogenic fraction members that cannot be classified as either PAH or subPAH include all other aromatic hydrocarbons within the C10–C32 and EC11–EC35 ranges that occur in petroleum contamination, such as 1,1-biphenyl. Noncancer toxicity assessment and surrogate selection for the aromatic high carbon range fraction is detailed in the PPRTV assessment for this fraction (U.S. EPA, 2022c). This section provides a summary of the approach and results; further detail is available in the PPRTV assessment.

Noncancer toxicity values were identified for 10 aromatic high carbon range compounds. Tables 13, 14, and 15 provide summaries of the toxicity values, critical effects, and key studies. Literature searches and searches of reviews, OECD SIDS and the Petroleum HPV Testing Group website yielded relevant toxicity data for five additional compounds and three defined mixtures⁶ for use in hazard identification for the fraction. In addition, limited toxicity data that were not sufficient to derive a toxicity value are available in the PPRTV assessment for phenanthrene (U.S. EPA, 2009o). Critical effects identified with existing toxicity values were developmental effects (neurodevelopmental changes, fetal skeletal anomalies), respiratory effects (pulmonary alveolar proteinosis), increased liver weight, decreased red blood cells (RBCs), renal effects (nephropathy, decreased kidney weights, renal papillary mineralization), clinical signs of neurotoxicity, and decreased body weight. Additional potential targets identified based on literature searches include the adult and developing reproductive system and the GI system.

Table 13

Available Subchronic RfD Values for Aromatic High Carbon Range Fraction (C10–C32 [EC11–EC35]).

Table 14

Available Chronic RfD Values for Aromatic High Carbon Range Fraction (C10–C32 [EC11–EC35]).

Table 15

Available RfC Values for Aromatic High Carbon Range Fraction (C10–C32 [EC11–EC35]).

Human data and inhalation data for animals are scarce. Animal oral data to assess consistency in effects across members of the fraction are widely available for body-weight effects and moderate for other endpoints. Chronic systemic toxicity information is lacking for all but five members of the fraction: naphthalene and 1,1-biphenyl have been tested in comprehensive 2-year systemic toxicity studies in animals (inhalation and oral, respectively); 1- and 2-methylnaphthalene have been evaluated in comprehensive 81-week oral studies; and BaP was evaluated in a 2-year cancer bioassay with limited reporting of nonneoplastic findings.

Based on review of the available data [see (U.S. EPA, 2022c) for further details], there is evidence to suggest consistency in body-weight changes, neurological effects, hepatic effects, and hematological effects of some aromatic high carbon range fraction members, but not enough to indicate consistency across the entire fraction. Available data indicate that the kidney and bladder are particularly susceptible to 1,1-biphenyl toxicity, with data from other compounds generally showing increased incidence of age-related nephropathy. There is little evidence to indicate respiratory tract effects following oral exposure for compounds other than 1- and 2-methylnaphthalene (for which pulmonary findings are confounded by inhalation exposure via volatilization from feedstock), although there is limited evidence to suggest consistency in respiratory effects following inhalation exposure across compounds with lower carbon numbers (no data for fraction members with higher carbon numbers; C13–35). The available data are not adequate to provide confidence in an assessment of the consistency in effects for GI tract, reproductive toxicity, or developmental toxicity endpoints (including neurodevelopment and reproductive development).

The lowest oral subchronic and chronic RfD among the compounds in this fraction that is not a screening value is the chronic RfD for BaP (see Table 14); this value is recommended for chronic exposures to the aromatic high carbon range fraction if available analytical chemistry data do not identify concentrations of individual chemicals composing this fraction, and an indicator chemical approach is implemented. Although a subchronic toxicity value is not available for BaP, the chronic RfD is based on a developmental exposure, so the RfD value is applicable to subchronic exposures as well, if an indicator approach is implemented. In addition, extensive chronic toxicity information has been reported for BaP and the developmental endpoint is the most sensitive. Subchronic and chronic toxicity values for several other PAHs, except for those of BeP, which is a screening value, are considerably higher (several orders of magnitude in some cases) than the chronic RfD for BaP, raising the question of whether use of BaP as the indicator chemical for the fraction may be toxicologically relevant. However, emerging information on mixtures and other compounds shows effects at exposures comparable to (or even lower than) levels at which BaP induces toxicity, suggesting that use of BaP values for the whole fraction may be more appropriate than implied by comparisons limited to compounds with toxicity values. For example, recent studies suggest that other PAHs in this fraction may induce altered reproductive tract development (Kim et al., 2011), neurodevelopmental effects (Crepeaux et al., 2014; Crepeaux et al., 2013, 2012), transgenerational changes in immune function (Chu et al., 2013) or adiposity (Yan et al., 2014), or lethal transplacental carcinogenesis (Madeen et al., 2016; Benninghoffand Williams, 2013; Shorey et al., 2013; Shorey et al., 2012; Castro et al., 2009; Castro et al., 2008c; Castro et al., 2008a; Castro et al., 2008b) at very low exposure levels. These newer studies support the selection of BaP as the indicator chemical because it is the only indicator chemical candidate with an oral toxicity value that will be toxicologically relevant for most of these effects. However, users of the indicator chemical method should understand that there could be more uncertainty associated with the application of this toxicity value to the aromatic high carbon range fraction than for its applications in assessments of BaP as an individual chemical in U.S. EPA (2017).

If the available analytical chemistry data quantify the concentrations of naphthalene, 2-methylnapthlalene, 1-methylnapthalene, 1,1-biphenyl, acenaphthene, fluorene, anthracene, pyrene, fluoranthene, or BaP separately from the remainder of the aromatic high carbon range fraction, it is recommended that HQs for the individual chemicals with analytical data be calculated and an HI for the mixture be developed using the calculated HQs.

For subchronic oral exposures, the following subchronic RfDs or p-RfDs can be used as the denominator in the HQ equations: naphthalene (0.6 mg/kg-day), 2-methylnaphthalene (0.004 mg/kg-day), 1,1-biphenyl (0.1 mg/kg-day), acenaphthene (0.2 mg/kg-day), fluorene (0.4 mg/kg-day), anthracene (1 mg/kg-day), pyrene (0.3 mg/kg-day), BeP (9 × 10⁻⁵ mg/kg-day), and fluoranthene (0.1 mg/kg-day). Additionally, the chronic RfD for BaP (0.0003 mg/kg-day) can be adopted for subchronic exposures because it is based on a developmental study (as discussed above). In this alternative approach, the chronic RfD for BaP (0.0003 mg/kg-day) is recommended for use with the remainder of the fraction, including any other fraction members analyzed individually.

For chronic oral exposures, the following chronic RfDs or p-RfDs can be used as the denominator in the HQ equations: naphthalene (0.02 mg/kg-day), 2-methylnaphthalene (0.004 mg/kg-day), 1-methylnaphthalene (0.007 mg/kg-day), 1,1-biphenyl (0.5 mg/kg-day), acenaphthene (0.06 mg/kg-day), fluorene (0.04 mg/kg-day), anthracene (0.3 mg/kg-day), pyrene (0.03 mg/kg-day), fluoranthene (0.04 mg/kg-day), BeP (9 × 10⁻⁵ mg/kg-day), and BaP (0.0003 mg/kg-day). In this alternative approach, the chronic RfD for BaP (0.0003 mg/kg-day) is recommended for use with the remainder of the fraction, including any other fraction members analyzed individually.

The lowest RfC among the compounds in this fraction is the chronic RfC for BaP (see Table 15)⁷; this value is recommended as the indicator chemical for chronic exposures to the aromatic high carbon range fraction if available analytical chemistry data do not identify concentrations of individual chemicals composing this fraction. Although a subchronic toxicity value is not available for BaP (the IRIS program did not develop subchronic values), the chronic RfC is based on a developmental exposure, so the RfC value is applicable to subchronic exposures as well. In addition, extensive chronic toxicity information has been reported for BaP and the developmental endpoint is the most sensitive. Several subchronic and/or chronic toxicity values for other PAHs are considerably higher (>2 orders of magnitude) than the chronic RfC for BaP, raising the question of whether use of BaP as the indicator chemical for the fraction may be overly conservative. However, emerging information (Crepeaux et al., 2014; Yan et al., 2014; Chu et al., 2013; Crepeaux et al., 2013, 2012) on mixtures shows neurodevelopmental effects at exposures lower than levels at which BaP induces toxicity, suggesting that use of BaP values for the whole fraction may be more appropriate than implied by comparisons limited to compounds with toxicity values.

If the available analytical chemistry data quantify the concentrations of 1,1-biphenyl, naphthalene, BeP or BaP in the air separately from the remainder of the aromatic high carbon range fraction, it is recommended that HQs for the individual chemicals with analytical data be calculated and an HI for the mixture be developed using the calculated HQs.

For subchronic inhalation exposures, the subchronic p-RfCs for 1,1-biphenyl (0.004 mg/m³) and BeP (2 × 10⁻⁶ mg/m³) and the chronic RfC for BaP (2 × 10⁻⁶ mg/m³) can be used as the denominator in the HQ equations; as discussed above, use of the chronic BaP value is appropriate because it is based on a developmental study. In this alternative approach, the chronic RfC for BaP (2 × 10⁻⁶ mg/m³) is recommended for use with the remainder of the fraction, including any other fraction members analyzed individually.

For chronic inhalation exposures, the following chronic RfCs or p-RfCs can be used in the denominator of the HQ equations: naphthalene (0.003 mg/m³), 1,1-biphenyl (0.0004 mg/m³), BeP (2 × 10⁻⁶ mg/m³), and BaP (2 × 10⁻⁶ mg/m³). In this alternative approach, the chronic RfC for BaP (2 × 10⁻⁶ mg/m³) is recommended for use with the remainder of the fraction, including any other fraction members analyzed individually.

Table 16 shows the available cancer risk estimates for components of the fraction. If analytical chemistry data do not identify concentrations of individual chemicals composing this fraction, an indicator chemical approach should be used. In this case, the BaP OSF should be integrated with an estimate of the oral exposure rates for the aromatic high carbon range fraction to estimate the oral cancer risk. The IUR should be estimated with the concentration of the fraction in the air to estimate the inhalation cancer risk. Table 17 shows the available RPF values for seven PAHs, with BaP serving as the IC. If analytical chemistry data identify individual concentrations of any of these seven PAH composing this fraction, an RPF approach should be used. In this case, the BaP OSF and IUR estimates can be integrated with estimates of the individual PAH exposure rates to estimate the oral or inhalation cancer risk associated with exposure to the fraction. If analytical chemistry data identify concentrations of individual of PAHs, subPAHs, and other carcinogenic fraction members with cancer risk values, an integrated addition approach should be used. The integrated addition approach assumes that the carcinogenic MOAs of the PAHs are independent of those of subPAH, 1-methylnaphthalene, and the other carcinogenic fraction member, 1,1-biphenyl. In this case, the RPF approach can be used to estimate cancer risk associated with the PAH portion of the fraction, and p-OSF values for 1-methylnaphthalene and 1,1-biphenyl can be integrated individually with their corresponding exposure rates. Response addition can then be used to sum risks across the three similarity groups (i.e., PAH, 1-methylnaphthalene, and 1,1-biphenyl) to estimate the oral cancer risk associated with exposure to the fraction. Because IURs (or p-IURs) were not identified for either 1-methylnaphthalene or 1,1-biphenyl, the integrated addition approach is only applicable to estimating oral cancer risks at this time.

Table 16

Available Cancer Risk Estimates for Aromatic High Carbon Range Fraction (C10–C32 [EC11–EC35])^, .

Table 17

RPFs in the U.S. EPA’s 1993 Provisional Guidance.