Risks, Release and Concentrations of Engineered Nanomaterial in the Environment

For frequently used engineered nanomaterials (ENMs) CeO2-, SiO2-, and Ag, past, current, and future use and environmental release are investigated. Considering an extended period (1950 to 2050), we assess ENMs released through commercial activity as well as found in natural and technical settings. Temporal dynamics, including shifts in release due to ENM product application, stock (delayed use), and subsequent end-of-life product treatment were taken into account. We distinguish predicted concentrations originating in ENM use phase and those originating from end-of-life release. Furthermore, we compare Ag- and CeO2-ENM predictions with existing measurements. The correlations and limitations of the model, and the analytic validity of our approach are discussed in the context of massive use of assumptive model data and high uncertainty on the colloidal material captured by the measurements. Predictions for freshwater CeO2-ENMs range from 1 pg/l (2017) to a few hundred ng/l (2050). Relative to CeO2, the SiO2-ENMs estimates are approximately 1,000 times higher, and those for Ag-ENMs 10 times lower. For most environmental compartments, ENM pose relatively low risk; however, organisms residing near ENM ‘point sources’ (e.g., production plant outfalls and waste treatment plants), which are not considered in the present work, may be at increased risk.


Engineered nanomaterial use/production amounts Current use/production amounts
The large variance in the data in Table S1a and Table S1b does not stand for natural variability but for the large uncertainty in the assessment. Normal production amount probability distributions were modeled, which have emerged from the data and their variability including specified weighting and natural limits of non-zero and non-negative values. Specifications greater than 1 were e.g. interpreted as 1.25 (a quarter over the specified number). The values in S1 a, b refer to aggregated/agglomerated ENM. The highly dispersed fraction is covered as well, however, cannot be quantified based on the current available data.

-ENM in t/a CeO 2 -ENM in t/a Ag-ENM in t/a
The share of the German annual market volume of the investigated ENM was calculated by multiplying the global annual production of the respective ENM by the ratio of German domestic demand (2014) and the worldwide gross domestic product (GDP) (in market prices for 2014).

Trends in use/production amounts
We modeled the use data trend in time running in relation to the 2015 use volumes by orienting ourselves on our survey results and combining them with other very rough time based growth dynamics given elsewhere (Ricardo Energy & Environment 2016). It is impossible to make any precise predictions or retrospective indications on the use trend. Thus, we restricted ourselves on very rough trend estimations without providing any complex time dependent use computations based on differentiated time dynamics, which are mostly just reflecting highly controversial non-validated market development data. A median annual growth of 5% (with 50% uncertainty/variability range on each side) was used for the time period after 2000 that best combines our own survey results with the ones for established materials given in elsewhere (Ricardo Energy & Environment 2016). A median value of 1% (with 50% uncertainty/variability range on each side) annual growth was assumed for single engineered nanomaterial applications for the time before 2000 by only roughly orienting ourselves on some silver ENM application trends discussed for medical use (Sun, Bornhöft et al. 2016).  Table S3-5: Engineered nanomaterial environmental application fields described as product categories. The given values represent our best mass fraction estimations of a particular engineered nanomaterial used for a particular product category. These values were decreased and enlarged by 50% leading to triangular and symmetrical probability distributions. The distributional symmetry may be limited in the Monte Carlo (MC) based computations by highest or lowest possible fractions of 1 and/or 0.

Engineered nanomaterial applications
Mass fractions of SiO 2 -ENM were adopted from a study of the European Centre for Ecotoxicology and Toxicology of Chemicals (ECETOC) (JACC 2006) on synthetic amorphous silica and complemented with "lubricants, greases, release agents" according to  and a predicted share based on own investigation.
For product categories of CeO 2 -ENM, mass fractions are based on estimates for global consumption of CeO 2 -ENM which are derived from information given in literature and databases. The global mass of CeO 2 -ENM in catalytic converters is based on the worldwide automobile production of cars and commercial vehicles for 2013 (OICA 2015), 80 g CeO 2 per converter (Bleiwas 2013) and a share of 85 % for cars and commercial vehicles equipped with a catalytic converter (own assumption).
The estimate for worldwide consumption of CeO 2 -ENM as diesel-fuel additive was calculated as a mean of high and low concentrations of CeO 2 -ENM in diesel fuel (2-5 mg/l according to Johnson and Park (2012)) with a fuel consumption of 1000 l per year and a global number of 5 Million cars with a fuel born catalyst system (Rocher, Seguelong et al. 2011).
The global mass of CeO 2 -ENM for use in exterior coatings was derived from (Gottschalk, Nowack et al. 2015). After scaling from the Danish to the German market (according to the ratio in population numbers), global consumption for exterior coatings was calculated by a factor which reflects the ratio of the worldwide gross domestic product (Worldbank 2016) to the German domestic demand (Statistisches Bundesamt 2016).
Global mass of CeO 2 -ENM for use as polishing agent is based on numbers given by Goonan (2011) for glass polishing and Reed, Cormack et al. (2014) for chemical mechanical polishing (CMP). The worldwide amount of CeO 2 for use in NiMH-Batteries is based on information given in Goonan (2011).
Mass fractions of product categories containing Ag-ENM are based on the ratio of the overall market volumes of these categories and the mean share of Ag-ENM in corresponding products.  , and information from a producer of lubricants.   Assumptions for the product lifespan are oriented on a classification given in Ricardo Energy & Environment (2016). ENM release during their production, formulation and manufacturing processes (Table S6) occurred before entering into our USE and EOL release phases. The release factors in Table S6 are own assumptions that loosely follow release factors during production and manufacturing used in Gottschalk, Lassen et al. (2015) and for the allocation of this total release into different emission vectors we loosely used values given elsewhere (Ricardo Energy & Environment 2016).

Engineered nanomaterial environmental release
7

Engineered nanomaterial fate in technical systems
The tables S 10-12 contain data on the engineered nanomaterial fate in technical systems represented in simplified form by Transfer Coefficients (TC). Those TCs stand for the annual fraction of ENM transport, transformation/elimination and/or deposition in or between technical systems. Single raw values were varied by ±50% for computing triangular and symmetrical probability distributions. Such symmetry may be affected by considering the limits of highest or lowest possible fraction values of 1 and 0. A value of 1 means that all ENMs are transported, transformed/eliminated and/or deposited and a value of 0 means that no ENM is involved in such processes.
In cases with several raw data values or for exceptions on single value sources, specific information on the probabilistic modeling procedure is given in each case. Since data on ENM product export is not available, we did not integrate such export by performing a conservative release and environmental exposure modeling. Dito, for consumer electronics 0.07, medtech 0.5, metals 1, for other categories 0.  (Walser and Gottschalk 2014) that confirm earlier experimental findings (Walser, Limbach et al. 2012). Our parameters refer to figure 3 and 4a (Walser and Gottschalk 2014

Landfills (LAN) and LAN influents and effluents
Parameter Transfer coefficient Comments and data source Fraction of solid waste ending up in landfills (LAN) 0.27 See comments on data and data sources for these LAN parameters in Table S 10.
Release from landfills to groundwater and surrounding surface waters 0 Dito.

Engineered nanomaterial fate in natural compartments
The table S 13 provides information on raw data or data gaps for the engineered nanomaterial fate in natural environments. The Transfer Coefficients (TC) represent as said before the annual fraction of engineered nanomaterial transport, transformation/elimination and/or deposition. Our best estimations were decreased and enlarged by 50% by means of triangular and symmetrical probability distributions limited by the highest or lowest possible fraction values of 1 and 0.  ).
Not accounted for because of lack of knowledge (Gottschalk, Lassen et al. 2015) Accounted for in STP effluents but not quantified for the retention time in natural waters (Sun, Bornhöft et al. 2016).

Dissolution and other reactions in soils and in sediments.
We stopped our model after the ENM enters the sediment and soil compartments. Currently there are no usable data available to subsequently perform such a highly complex environmental fate analysis. Thus, soils and sediments represent final sinks in the natural environment comparable to landfills in the technosphere. Insofar, we have to emphasize that our modelled concentrations in soils and sediments do not account for any nanomaterial dissolution or transformation that will likely be very significant for all three materials (Ag-ENM, CeO 2 -ENM (Gottschalk, Lassen et al. 2015), and SiO 2 -ENM  ).

Geographic, aquatic and waste management data
Predicted environmental concentrations (PEC) have been derived by dividing the computed engineered nanomaterial flows and deposition volumes by the mass/volume of a particular environmental compartment (bulk material or technical compartment) receiving nanomaterial.

ENM in nature for use (USE) and end of life (EOL) material phases.
The following material input (I) functions refer to the ENM already and persistently out there in nature or technical sinks summing up to a certain time point when material is not anymore in consumption or in release processes. The first equation refers to USE release, the second is an equivalent example for EOL modeling.

Predicted environmental concentrations
Predicted environmental concentrations (PEC) for 2017, 2030 and 2050 are shown for SiO 2 -ENM, CeO 2 -ENM and Ag-ENM. We show on the one hand the modeled mode values, the mode values that only reflect concentrations from EOL (end of life ) releases, the modeled ranges that represent values with "some probability". Such "some probability" simply refers to the results range of our main modeling where we did not per se force all applications running at the lowest or highest production levels and environmental release levels. Such simultaneous scenarios of "all engineered nanomaterial applications running at the same time" at their lowest and in another case at their highest production, use and release levels were modelled separately and provided the total min-max range. We have rounded the model output (3 places after the decimal point). Nevertheless, those values after the decimal point should not suggest any accuracy of individual values taken from the analysis of probability distributions. Natural soils** ng/kg 314. 693 7,051,049.131 1,395.131 3,925,579.580 -10,884,369.732 29,500,863.984 Urban soils* ng/kg 20.074 449,777.216 87.216 197,524.825 -878,978.535 1,881,822.755 Urban soils** ng/kg 503.508 11,281,678.210 2,232.210 6,280,414,991.572 47,201,409.996 Sludge treated soils* ng/kg 164.996 4,372,365.208 513.208 1,832,101,406.555 18,693,550.469 Sludge treated soils** ng/kg 4,873.694 129,152,203.226 16,495.226 52,309,558.260 -291,015,328.100 552,175,564