Presence of multi-walled carbon nanotubes (MWCNT) in the environment : release from polymer/MWCNT composites and their interaction with the biocide triclocarban (TCC)

Hennig, Michael; Schäffer, Andreas (Thesis advisor); Hollert, Henner (Thesis advisor)

Aachen (2019, 2020)
Dissertation / PhD Thesis

Dissertation, RWTH Aachen University, 2019


Embedding of multi-walled carbon nanotubes (MWCNT) as nanofillers in neat polymeric plastic matrices enhances various properties of such composites. At the end of a life-cycle, such products may be further disposed of, degraded e.g. chemically or by UV-light exposure, and finally lose their structural integrity. Hence, a release of MWCNT and polymer/MWCNT fragments in the environment is very likely. Once these materials enter the environment, they may also act as sorbents for several hydrophobic organic chemicals. Furthermore, the released materials are bioavailable for aquatic and terrestrial biota, leading to an incorporation and ingestion of such sorbates. The combined exposure of both may then cause an altered bioconcentration of the adsorbed chemicals and adverse effects in the organisms. This work delivers an overview of the environmental impacts caused by MWCNT and polymer fragment release from polypropylene (PP) and epoxy resin (E) nanocomposites (Chapter 2) and what influence the released MWCNT material in combination with the biocide triclocarban (TCC) may exhibit afterwards in the aquatic food web (Chapter 3).In the first part of my study, the release rates for MWCNT from self-produced PP (2.5% w/w) and E (0.25% w/w) nanocomposites were quantified by use of radiolabeled MWCNT ($^{14}$C-MWCNT). One part of the plates was subjected to simulated sunlight irradiation (+SSR, 50 W m$^{-2}$, 3 months), the other part served as control plates (-SSR). Both types were exposed to manual stress, artificial fresh- and sea water, and incubated in a quartz sand-water system or soil. SSR led to surface degradation of the nanocomposites, enhancing the formation of micro-cracks, rough structures and voids. Released material consisted of polymer fragments with embedded or protruding MWCNT; also single tubes were found. Further degradation of the nanocomposites was observed after exposure under the abovementioned conditions. The irradiated PP samples exhibited a significantly higher $^{14}$C-MWCNT release compared to the control groups in all five scenarios with about 0.2% of initially embedded MWCNT (IEM) after manual stress (36-fold higher) and about 0.02% of IEM in the environmental scenarios (6-fold higher). This corresponded to the release of about 26 mg m$^{-2}$ and 2.2 mg m$^{-2}$ $^{14}$C-MWCNT from the mechanical and the environmental stressors, respectively. For the E plates, a 23-fold higher release after mechanical stress was demonstrated with a released percentage of 0.11% of IEM, corresponding to 3.4 mg m$^{-2}$ $^{14}$C-MWCNT. For the environmental stressors, an average of 0.04% of IEM were set free, which is about 11 times stronger than from the -SSR plates. This percentage corresponded to about 1.2 mg m$^{-2}$ $^{14}$C-MWCNT. The energy input and the stressor intensity are the main factors of enhancing the release. Ionic strength diminished MWCNT release from polymer nanocomposites due to agglomeration and re-arrangement of exposed MWCNT on the surface plate. Moreover, a mineralized amount of 1.7 µg $^{14}$C-MWCNT was determined in soil after 188 d (initial amount: 2.5 mg). Released material is bioavailable for aquatic invertebrates. Daphnia magna, Artemia salina and Lumbriculus variegatus incorporated up to 4% of the fragments. This material did not exert any acute toxic effects on the organisms and can be rapidly eliminated again. However, indirect effects on a Daphnia magna population was demonstrated after exposure to 100µg L$^{-1}$ MWCNT and 40 mg L$^{-1}$ weathered fragmented E/MWCNT product. Due to the steady exposure by these materials for 22 d, a food subsequent reduction led to a significant abundance decline in the following sampling days compared to an untreated control group. Furthermore, sublethal effects, like shortened antennae, internal vacuoles or red colored maxillary nephridia were observed.In the second research part of my work, interactions between MWCNT (1 mg L$^{-1}$) and TCC in different environmental media and resulting alterations on TCC bioavailability to Desmodesmus subspicatus, Daphnia magna and Artemia salina were taken into account. Aqueous MWCNT dispersions formed agglomerates and sediment faster in artificial sea water than in fresh water (sedimentation rates of 0.027 h$^{-1}$ and 0.017 h$^{-1}$). Resulting from this, fast agglomeration has a strong influence on the adsorption of TCC on MWCNT. The higher the ionic strength in a medium is, the faster MWCNT agglomerates were formed, providing less adsorption sites for TCC. However, in all tested media, a similar high log K$_{MWCNT}$ of about 7.6 was determined. Furthermore, modelling sorption kinetics for the scenarios, the Dubinin-Ashtakhov model provided the best fit with a coefficient of determination of >0.96. Toxicity of TCC alone was determined and resulted in median effect concentration (EC50) of 19, 21 and 16 µg TCC L$^{-1}$ for D. subspicatus (72 h), D. magna (48 h) and A. salina (48 h), respectively, which is slightly higher than typical environmental concentrations of about 6 µg TCC L$^{-1}$. The bioconcentration factors (log BCF) for TCC (10 µg L$^{-1}$) in the three organisms amounted to 4.6, 4.1 and 3.4, showing a moderate to strong bioconcentration. As a consequence, TCC-MWCNT sorbates are potentially bioavailable for biota, leading to an altered bioavailability of TCC. For D. subspicatus, a toxicity reduction of TCC was observed, since the sorbates are not bioavailable for the algal cells due to retention processes by cell wall structures. Hence, no proper EC$_{50}$ value could be determined. The log BCF amounted to a slightly higher value of 4.7. However, the internal TCC amounts in the algae cells were lower in presence of MWCNT (91 mg kg$_{dw}$$^{-1}$ vs. 55 mg kg$_{dw}$$^{-1}$ after 24 h), indicating a bioconcentration reduction. Concerning both tested invertebrates, an enhancement of TCC toxicity in presence of MWCNT was observed with lower EC$_{50}$ values of about 2 µg TCC L$^{-1}$ and 0.4 µg TCC L$^{-1}$ for daphnids and brine shrimps, respectively. Same increase occurred for the log BCF to 4.4 and 4.1 in presence of MWCNT. For the invertebrates, the internal TCC amounts were also lower in presence of MWCNT (D. magna: 78 mg kg$_{dw}$$^{-1}$ vs. 7 mg kg$_{dw}$$^{-1}$ after 72 h, A. salina: 8 mg kg$_{dw}$$^{-1}$ vs. 7 mg kg$_{dw}$$^{-1}$ after 48 h), eventually assuming a TCC bioconcentration reduction. Both invertebrates ingest TCC-MWCNT sorbates, but depending on the TCC concentration only slow desorption in the gut occurs.In conclusion, all presented studies reveal new insights in the release behavior of MWCNT containing plastic products after disposal: For the first time it could be shown, that released micro- and nano-sized material is bioavailable for pelagic and benthic organisms and also have adverse effects on population level. Additionally, combined effects of MWCNT and TCC led to a reduced toxicity towards green algae and an enhanced toxicity towards fresh water and sea water invertebrates. The applied MWCNT concentrations were about six orders of magnitude higher than the environmental relevant ranges (ng L$^{-1}$). Obtained insights close several knowledge gaps and lead to a better understanding of the potential strong influence of MWCNT in the aquatic environment. However, significant more research is inevitable to assess all possible risks of an estimated release of MWCNT polymer composites and pristine MWCNT to aquatic and terrestrial biota.