TiO2 Emissions from Sunscreens in the Old Danube River

1. Introduction Engineered nanomaterials (ENMs) are expected to reach natural waters after normal use of consumer products and there is growing concern about their potential impact on environmental quality [1]. However, there are almost no methods available for detecting them in real world systems. Such methods should be able to identify ENMs against a plethora of natural particles of the same size range [2]. In this respect, the detection of metallic ENMs whose constituents are rarely encountered in the environment, such as silver and gold, poses an advantage over materials with high natural abundance, such as titanium dioxide and silica. The use of elemental ratios has been proposed for overcoming the challenge of detecting the latter [3]. Naturally occurring minerals contain “impurities” that result from geochemical processes, such as co-precipitation, elemental substitution, adsorption, etc. On the contrary, the elemental composition of ENMs is typically pristine in respect to their major components. In this work we attempt to identify the release of TiO2 ENMs from sunscreens in the Old Danube Lake by means of the Ti/Al elemental ratio in suspended particulate matter (SPM). Samples were also collected from sediments (SED), sediment cores (COR), and soils (SOIL) in order to evaluate factors, other than the release of sunscreen, which may affect Ti/Al ratios in SPM. Samples from the Lobau Lake, located at approximately 10 km distance from the Old Danube Lake were also collected for comparison. The Old Danube Lake is a recreational area and receives approximately 1.5 million bathers annually, while the Lobau Lake is a national water park and bathing is limited. Possible sources of particulate matter that could alter the Ti/Al ratio in the SPM of the Old Danube are natural processes, such as atmospheric deposition, rain water runoff, transport of soil into the lake by animals and from anthropogenic sources, such as the release of sunscreen TiO2 from bathers’ skin. We hypothesize that the majority of TiO2 released by the use of sunscreens will associate with the natural SPM [4]. 2. Material and Methods Samples were collected from the Old Danube Lake and the Lobau Lake; both lakes are located in Vienna and the distance between them is less than 10 km. Suspended particulate matter (SPM) samples were collected by pressure filtration through cellulose acetate membranes, from May, 2012 through June, 2013 with a frequency of 2 – 4 days per month during the spring and summer seasons and 1 – 2 during the fall and winter seasons, as described previously [5]. Sediment and sediment core samples were collected from the Old Danube Lake during two sampling campaigns on March 7, 2013 and April 11, 2013 and from the Lobau Lake on April 2, 2013. Sample locations on the Old Danube Lake are shown in Figure 1. Sediment samples were collected by means of an Ekman Bottom Grab (opening 15 x 15 cm, total volume 4.5 L, HYDRO-BIOS Apparatebau GmbH, Kiel-Holtenau, Germany). Water depth at each sample point was measured by submerging a weight attached on the one end of a measuring tape. Sediment samples were placed in polypropylene holders and stored at -7 ˚C. Sediment cores were collected using a sediment corer (UWITEC, Mondsee, Austria). Cores were sliced into 2 cm thick pieces immediately after sample collection, upon return to the lab; the pieces were placed in polypropylene holders, and stored at -7 ˚C. Sediment and core slices were freeze dried (Alpha 1 – 4 LSC, Martin Christ Gefriertrocknungsanlagen GmbH, Osterode am Harz, DE) and stored at -7 ˚C before further analysis. Core slices were digested and analyzed for elemental composition with no further sample treatment [5]. Sediment samples were sieved using a 250 μm sieve to remove gravel and stones and then milled (MM400, Retsch GmbH, Haan, DE) before digestion and elemental analysis. Soil samples were collected from eleven points around the Old Danube Lake on December 6, 2012 and the Lobau Lake on April 2, 2013 by scooping approximately 10 grams of soil into plastic bags. The soil samples were oven dried at 105 ˚C for 72 hours and sieved using a 250 μm sieve to remove gravel and stones before milling, acid digestion, and elemental analysis. Figure 1: Areal picture of the Old Danube Lake, with sampling locations of sediments (points 1 – 23), sediment cores (points 25 – 28), and suspended matter (point 4). 3. Results Titanium concentrations of SED, COR, and SOIL samples are plotted against Aluminum concentrations in Figure 2. The data show a relatively good correlation between Ti and Al for SED, COR, and SOIL samples. The measured Ti and Al values for the Old Danube are in the lower range of values reported for stream sediments across the European continent (Figure 2). This places the Old Danube Lake in a unique position for detecting TiO2 ENM release, due to the combination of relatively low Ti and Al background and intense bathing activities (approximately 1.5 million bathers in a total volume of 3.5 × 106 m³). Figure 2: (left) Ti and Al concentrations of sediment, core slice, and soil samples collected from the Old Danube Lake; (right) Ti and Al concentrations of stream sediment samples collected from various locations across Europe (source: Forum of European Geological Surveys – FOREGS); the concentration range measured in sediment samples of the Old Danube Lake is marked with a red square. 95% confidence intervals and prediction intervals are marked with dashed lines and dotted lines, respectively. Overall, the elemental ratio of Ti/Al in SPM was 0.048 ± 0.016 (N = 89) and exhibited seasonal variations (Figure 3). The ratio was 0.068 ± 0.011 (N = 23) in SED samples, 0.065 ± 0.006 (N = 28) in COR samples, and 0.056 ± 0.011 (N = 11) in SOIL samples (Figure 4). The variation of Ti/Al ratio in SPM samples is explained by the fact that SPM is highly dependable on seasonal factors, while sediments and soils consist of material that has been accumulated over the time span of several years. In order to determine whether there is a release of TiO2 from sunscreens during the bathing season and its potential effect on the Ti/Al ratio in SPM, it is important to define background values. Since the elemental composition of SPM is sensitive to biogeochemical factors, such as plant and animal activity, as well as weather conditions we used Ti/Al values from the spring season as background for detecting changes during the summer/bathing season. Figure 3: the evolution of Ti concentration and Ti/Al ratio in the SPM from April, 2012 through June, 2013. The bathing season is marked with dashed lines. In an attempt to identify the major source of changes in the elemental composition of SPM, we studied the following parameters: 1) transport of soil in the lake from humans and animals, 2) atmospheric deposition, 3) rain water runoff and paint wash off, 4) re-suspension of sediment during bathing activities, 5) release of sunscreen TiO2. Soil transport was evaluated by measuring Ti/Al ratios in soils collected from the Old Danube Lake, which did not differ significantly from SPM (Figure 4). Atmospheric deposition was evaluated by measuring Ti/Al in sediment cores from the Old Danube and Lobau Lakes; while the ratio increases in sediments deposited recently in the Old Danube Lake, the same does not hold true for sediments in the Lobau Lake, where the ratio appears to be relatively constant (Figure 5). Rain water runoff and paint wash off are triggered by rainfall events. We therefore compared Ti/Al ratios in SPM with the intensity of rainfall and found no correlation. Finally, re-suspension of sediment from bathing activities was evaluated by combining Ti/Al ratios in sediments collected in the vicinity of the SPM sampling point, the location of bathing areas, and sediment depth of the collected samples. Figure 4: Titanium to aluminum elemental ratios in suspended matter (N = 89), sediment (N = 23), and sediment cores (N = 28) collected from the Old Danube Lake from May, 2012 to August, 2013. Also shown are ratios of soil samples (N = 11) collected in the vicinity of the lake, on December 6, 2012. Error bars indicate one standard deviation of replicate samples. Figure 5: Ti/Al elemental ratios of sediment core slices collected from the Old Danube and Lobau Lakes. Note the increase of Ti/Al ratio as depth decreases in cores from the Old Danube Lake, while the ratio is relatively stable in cores from the Lobau Lake. 4. Conclusions and Outlook The elemental composition of SPM may be affected by transport of soil in the lake from humans and animals, atmospheric deposition, rain water runoff and paint wash off, re-suspension of sediment during bathing activities, and release of sunscreen TiO2. Soil transport into the lake may result from human and bird movement between the beach and the water. However, such transport would have little effect on the Ti/Al ratio in SPM, since the values for SPM and SOIL are not statistically different, as shown in Figure 4. Atmospheric deposition is a result of aerosol material transported over a long range of distances. The type of aerosols deposited on the Old Danube Lake is most likely to be similar in composition to the aerosols deposited on the Lobau Lake, since the distance between them is only 10 km. While the Ti/Al ratio in sediments remains constant over the years in the Lobau Lake, it is increasing with time in the Old Danube Lake, as shown by sediment core analysis (Figure 5). The lack of correlation between precipitation events and Ti/Al ratios in SPM, indicate that the increase of Ti/Al in SPM during the bathing season was not caused by water runoff or paint wash off. The Ti/Al ratio in sediments collected from the bathing areas were above the background values for SPM. Therefore, we expect an influence of the TiO2 release from e.g. sunscreens on the Ti/Al ratios of top layer sediments (supported by the core samples) and we cannot rule out the possibility that the increase of Ti/Al in SPM during the bathing season is also slightly influenced by re-suspension of sediments from bathing activities. It is worth noting that the Ti/Al measured in the deepest point of the area is rather low, which in combination with the modest increase of Ti/Al in SPM during the bathing season, may indicate that TiO2 ENMs are either aggregating and settling fast, they accumulate on the air-water interface, or they are scavenged by plants and algae. The application of Ti/Al ratios improves the detection of manufactured TiO2 nanoparticles in surface waters over the simple use of total Ti concentrations. However the release of TiO2 particles from sunscreens is not sufficient to enable the detection above the natural background with this methodology. The calculated maximum release of TiO2 from sunscreens to remain under the detection limit arising from the variation of Ti/Al ratios by natural processes equals 5 % of the TiO2 applied on the skin by bathers in the Old Danube Lake. This value requires of course further validation by direct release studies from human skin. 5. Literature 1. Wiesner, M.R., et al., Assessing the risks of manufactured nanomaterials. Environmental Science & Technology, 2006. 40(14): p. 4336-4345. 2. von der Kammer, F., et al., Separation and characterization of nanoparticles in complex food and environmental samples by field-flow fractionation. Trac-Trends in Analytical Chemistry, 2011. 30(3): p. 425-436. 3. von der Kammer, F., et al., Analysis of engineered nanomaterials in complex matrices (environment and biota): General considerations and conceptual case studies. Environmental Toxicology and Chemistry, 2012. 31(1): p. 32-49. 4. Kiser, M.A., et al., Titanium Nanomaterial Removal and Release from Wastewater Treatment Plants. Environmental Science & Technology, 2009. 43(17): p. 6757-6763. 5. Gondikas, A., et al., Nachweis und Charakterisierung von TiO2-Nanomaterial in Oberflächengewässern. Vom Wasser, 2013. 111(3): p. 89-91.