ARCHIVED - Draft Screening Assessment of Hexabromocyclododecane (HBCD)

A Level III fugacity (steady-state) box model based on the Level IV multispecies model described by Cahill et al. (2003) was applied for estimating aquatic exposure to HBCD in the pelagic and benthic compartments. An important feature of the Cahill et al. model is its ability to model the fate of transformation products in addition to that of the parent chemical. For HBCD, degradation to 1,5,9-cyclododecatriene (CDT) is considered an important fate process and this degradation product was included in the model as an additional species. CDT was not included in the risk quotient analysis for HBCD but is considered with respect to the overall persistence of the parent substance.

Figure 2-1 provides a conceptual overview of the fugacity model. The model is a mass balance system consisting of 10 downstream boxes each with water and sediment compartments.  For modelling purposes, the river is assumed to be a straight channel of uniform and rectangular cross-section with little or no vegetation present in the watercourse or along the banks. Release from the outfall is considered to be continuous from a steady vertical point source.

For each box, the fugacity (f) of both HBCD and the potentially persistent degradation product, CDT, is modeled in each compartment (water, sediment). Fugacity, in units of Pascal (Pa) represents the “partial pressure” of a chemical species in a particular medium and is analogous to concentration, C (mol/m³), normalized to the relative affinity of the chemical for a particular medium (also known as the “fugacity capacity”, Z [mol/m³.Pa]). Thus, f = C/Z (Mackay 1991).

Aside from mass loading (which is a known discharge rate [mol/h]), the mass transport associated with each process (mol/h) is represented as the product of a fugacity rate coefficient (D, in units of mol/h.Pa) and f (Pa) for other compartments/species (for input processes), or of the modeled compartment/species (for output processes). Transformation of HBCD to CDT is included in the reaction terms. For a detailed review of equations of this model, the reader is referred to the supporting working document for this assessment (Environment Canada 2009).

The main assumptions of the model:

1. chemical release to water only
2. volatilization or air/water intermedia transport is negligible
3. surface water consists of pure water, suspended sediment and biota phases
4. bottom sediment consists of pure water and sediment solids phases
5. first order reaction processes
6. complete instantaneous mixing within boxes
7. equilibrium between phases (pure water, sediment solids and biota) within a particular compartment

Model Parameters

The parameter inputs for the model include chemical properties (e.g., log K_ow, K_oc, degradation rates), substance release rates, receiving river conditions (e.g., river discharge and flow rates), and generic environmental parameters (e.g., organic carbon content of sediments and sediment deposition rates). Environmental parameters were chosen to represent rivers of southern Ontario based on parameters from ChemCan (Webster et al. 2004), the Cahill et al. (2003) model and plausible physical characteristics for similar river systems (considering values summarized in Chapra 1997 and Gobas et al. 1998). For this assessment, the model extended downstream 5000 m, split into 10 boxes. The length of the first and last boxes was set at 100 m each, and the length of the middle 8 boxes was set at 600 m each. 

Loading Estimates and Model Scenarios

Loading estimates for the model were determined using quantities reported in the section 71 notice (Environment Canada 2001), default emission factors recommended by OECD (2004a) and default emission periods recommended in the European Communities Technical Guidance Document (TGD; European Communities 2003). Based on information provided in the section 71 notice, annual import volumes for the year 2000 were in the range of 100 000 to 1 000 000 kg. Furthermore, it was estimated that annual HBCD use at an individual facility in Canada would range from 10 000 kg/year to 100000kg/year. Two release scenario groups were developed to represent the types of HBCD-related activities most likely to be taking place in Canada : raw materials handling (Scenario Group 1), and compounding (Scenario Group 2). The OECD (2004a) defines raw materials handling as the handling of raw materials from their arrival on site to their addition to polymers, including manual handling of bags and sacks, conveyer belts and pneumatic or pumped transfer from bulk storage vessels. Compounding is then the process by which additives such as HBCD are incorporated into materials (e.g., plastics) during polymer production and includes processing and final conversion (OECD 2004a). The two activities of raw materials handling and compounding were separated in order to estimate the predicted incremental risk from each activity. HBCD is not produced in Canada and it is likely that any facility involved in compounding would also need to be involved with raw materials handling. For these facilities, the predicted incremental risks from raw materials handling and compounding would be additive.

Scenario Group 1 applied an emission factor of 0.6% based on OECD (2004a) and emission periods of 200 days for usage of 100 000 kg/year and 60 days for usage of 10 000 kg/year (based on Table B2.8 of Appendix I of the TGD). For each usage rate, three possible levels of sewage treatment were applied (none, primary, and secondary) with removal rates estimated using EPIWIN (2000). The combination of two usage rates and three potential levels of sewage treatment yielded six possible emission scenarios for raw materials handling (Scenarios 1a–1f). Scenario Group 2 applied an emission factor of 0.055% based on OECD (2004a) and the same emission periods and levels of sewage treatment as Scenario Group 1, again resulting in six possible emission scenarios for compounding (Scenarios 2a–2f). Note that the OECD and TGD emission parameters were established by means of expert judgement and tend to the worst-case situation.

All release scenarios were assumed to describe industrial activities at a generic facility located in southern Ontario. Generic scenarios were employed to provide estimated release quantities in the absence of site-specific information. The generic facility was situated in southern Ontario as this region is associated with substantial industrial activity and might therefore be expected to have processing and production plants that utilize HBCD. The river dimension characteristics for these scenarios have been chosen to represent an average “medium-sized” river for the industrialized Lake Erie/lowland region of southern Ontario (i.e., the average of the middle 33% of rivers located in this region, based on Environment Canada’s Hydat database). The river discharge rate was based on the 25th percentile discharge rate for these rivers.

The release scenarios were entered into the fugacity box model and the results obtained were used to estimate potential water column exposure concentrations for pelagic organisms. For each scenario, the dissolved concentration of HBCD predicted to occur in the first 100 m from the point of discharge, termed C_max, was considered to represent a reasonable and conservative exposure concentration in the river and was selected as the predicted environmental concentration (PEC). This concentration is equivalent to that which would result from instantaneous complete mixing of the substance in the first 100 m following discharge to the river.

The major characteristics and model input parameters for each scenario are summarized in Table 2-1.

Model Results and Risk Analysis

Prior to calculation of risk quotients for the benthic and pelagic compartments, the scenarios and model-predicted concentrations were evaluated for their degree of “realism” with respect to expected actual HBCD release conditions in Canada. Scenario 1a resulted in a maximum predicted HBCD concentration in the water column which exceeded the measured water solubility for HBCD (refer to Table 1). Furthermore, upon review, it was judged that direct release of HBCD to watercourses without primary or secondary sewage treatment would not occur under normal operations of processing facilities. Based on these considerations, the scenarios with no sewage treatment (i.e., “none”) were excluded from the risk characterization (i.e., risk quotients were not calculated).

Pelagic Organisms

Table 2-2 summarizes the risk quotient results obtained for pelagic organisms under the retained scenarios. Risk quotients ranged from 0.071 to 3.75 for an annual usage quantity per facility of 10 000 kg/yr and from 0.179 to 10.7 for a use quantity of 100 000 kg/yr. Predicted dissolved water concentrations of HBCD exceeded the predicted no-effect concentration (PNEC) for all raw materials handling scenarios (Scenario Group 1), except for low-volume (10 000 kg/yr) facilities utilizing secondary wastewater treatment. For the compounding scenarios (Scenario Group 2), predicted dissolved water concentrations of HBCD were below the PNEC for all scenarios except for high-volume (100 000 kg/yr) facilities using primary treatment.

Based on the risk quotient results, it is concluded that concentrations of HBCD in surface waters resulting from activities associated with raw materials handling and compounding have the potential to cause adverse effects in populations of pelagic organisms in Canada . Application of secondary treatment processes to wastestreams originating from HBCD processing facilities greatly reduces the potential for risk; however, predicted exposure values still exceed those of minimum effects levels for scenarios associated with large production quantities (e.g., 100 000 kg/yr) and/or use of primary wastewater treatment. It should be noted that although HBCD concentrations are predicted to decrease with distance, the potential distance of impact downstream (i.e., distance with risk quotients greater than 1) is expected to be significant (> 5000 m).

Benthic Organisms

Table 2-3 summarizes the risk quotient results obtained for benthic organisms under each retained scenario. Results for benthic organisms generally paralleled those for pelagic organisms. Risk quotients ranged from 0.051 to 2.37 for an annual usage quantity per facility of 10 000 kg/yr and from 0.152 to 7.11 for a use quantity of 100 000 kg/yr. Predicted bulk sediment concentrations of HBCD exceeded the PNEC for scenarios associated with large-volume raw materials handling (Scenarios 1b and 1c) and smaller-volume raw materials handling with only primary wastewater treatment (Scenario 1e). Predicted bulk sediment concentrations of HBCD were less than the PNEC for all compounding scenarios (Scenario Group 2), suggesting that current volume estimates for this activity should not result in bulk sediment concentrations that exceed minimum effects levels in organisms.  It should be noted that although HBCD concentrations are predicted to decrease with distance, the potential distance of impact downstream (i.e., distance with risk quotients greater than 1) is expected to be significant (> 5000 m).   

Table 2-1: HBCD Emission Rates, River Characteristics and Release for Fugacity Modelling Release Scenarios

¹ Environment Canada 2001
² OECD 2004a
³ European Communities 2003
⁴ Primary wastewater treatment
⁵ Secondary wastewater treatment
⁶ From STPWIN (EPIWIN 2000)
⁷ Discharge estimates were made considering Southern Ontario streamflow data from the HYDAT streamflow database (National Water Data Archive, Environment Canada), and generally represent the 25th percentile of observed discharge rates.
⁸ Channel geometry and hydraulic parameters were estimated using equations derived specifically for southern Ontario (Boivin 2005).

Table 2-2: Model Output and Risk Quotient Analysis for Pelagic Organisms

¹ Primary wastewater treatment
² Secondary wastewater treatment
³ Cmax represents the dissolved HBCD concentration in the first 100 m of river downstream of the emission point.
⁴ C5000 represents the dissolved HBCD concentration at a distance 4900-5000 m downstream of the emission point.
⁵ Predicted dissolved HBCD concentration exceeds measured water solubility (refer to Table 1).
⁶ Risk quotient not calculated because the "no treatment" scenarios were considered unrealistic.
⁷ Not applicable as the predicted exposure concentration was less than the estimated no effect level.

Table 2-3: Model Output and Risk Quotient Analysis for Benthic Organisms

¹ Primary wastewater treatment
² Secondary wastewater treatment
³ Cmax represents the sediment HBCD concentration in the first 100 m of river downstream of the emission point.
⁴ C5000 represents the sediment HBCD concentration at a distance 4900-5000 m downstream of the emission point.
⁵ Risk quotient not calculated because the "no treatment" scenarios were considered unrealistic.
⁶ Not applicable as the predicted exposure concentration was less than the estimated no effect level.