end of life cycle assessment for carbon nanotube (cnt)...
TRANSCRIPT
End of life cycle assessment for carbon nanotube
(CNT) containing composites: Release of CNT and
ecotoxicological consequences
Von der Fakultät für Mathematik, Informatik und Naturwissenschaften der RWTH Aachen University zur Erlangung des
akademischen Grades eines Doktors der Naturwissenschaften genehmigte Dissertation
vorgelegt von
Diplom-Biologe Stefan Rhiem
aus Düren
Berichter: Universitätsprofessor Dr. Andreas Schäffer Universitätsprofessor Dr. Henner Hollert
Tag der mündlichen Prüfung: 14. Juli 2014
Diese Dissertation ist auf den Internetseiten der Hochschulbibliothek online verfügbar
ii
iii
iv
v
“homo hominis lupus”
Titus Maccius Plautus (ca. 254–184 v. Chr.)
vi
vii
Summary
An end of life cycle assessment for carbon nanotube (CNT) holding polymeric composites was performed
comprising the evaluation of (1) the release of CNT material from composites during environmental
degradation, (2) ecotoxicological effects on primary producers (green algae) and (3) secondary effects of CNT
presence on the impact of the xeno-oestrogen 17α-ethinylestradiol (EE2) on zebrafish. This work should deliver
an overview of the impacts that may cause CNT release from CNT composite products and what hazardous
impact to the environment released CNT material may exhibit afterwards.
CNT are considered as one of the promising materials in nanotechnology. They consist of rolled up graphitic
sheets with a diameter in the nano-dimension and a length up to several µm. Due to their structure, they
exhibit unique properties, i.e. high electrical and thermal conductivity, very low density, flexibility and
extremely high tensile strength. CNT transfer their properties to neat polymeric matrices, when compounded
together, as they e.g. make plastic composites less flammable, enhance their electric conductivity and improve
the mechanic stability and flexibility. As several applications for nanocomposites, such as construction,
aerospace, medical, water cleaning or solar-harvesting, were discovered, a production scale up for the future is
estimated. This may lead to increasing CNT release in the aquatic environment.
In the present study, for the first time, release rates for CNT from polycarbonate nanocomposites (containing 1
weight % CNT) were quantified by use of 14C-labeleled CNT material. Nanocomposites were produced by
dispersing 14C-CNT in chloroform and adding polycarbonate pellets afterwards in order to produce CNT holding
PC composites. After evaporation of the chloroform and surface improvement by hot pressing, round samples
with a diameter of 18 mm were produced. Two different sample types were prepared for release
quantification, composite plates previously degraded by artificial sunlight (1000 MJ/m2) and fresh prepared
plates. Both were subjected to different degradation scenarios; i.e. water, heat, frost-thaw, humic acids, acid
rain, disposal site effluent and abrasion, in series. The pre-degraded samples exhibited a much higher CNT
release compared to the non-degraded ones (1% of the CNT material released compared to 0.03%). Release of
the pre-degraded samples amounted to 18 mg per m2 and year regarding to Florida irradiation mean. Scanning
electron microscopy pictures (SEM), Infra red (IR)- and X-ray photoelectron spectroscopic (XPS) investigations
of the pre-degraded surfaces showed an uncovering of CNT material and the formation of a dense CNT network
at the surfaces. This surface decelerated the degradation rate of the matrix material beneath. Hence, release of
CNT material from nanocomposite products could be estimated to occur in the environment by different
degradative impacts.
In the second part of this work, CNT were shown to be bioavailable for the unicellular algae Desmodesmus
subspicatus. Cells were exposed to 14C-CNT suspensions (1 mg/L) within their growth media to determine
uptake and association, as well as elimination and dissociation in clear media, afterwards. Attenuated total
reflection Fourier-transform infrared spectroscopy (ATR-FTIR) was used to determine effects on algae
biochemical composition. Differences between the spectra of control and CNT-exposed cells were detected.
CNT-cell interactions were visualized by electron microscopy and related to alterations in their cell
composition. A concentration factor (CF) of 5000 L/kg dry weight was calculated. However, most of the
material agglomerated around the cells, but single tubes were detected in the cytoplasm as well.
Computational analyses, i.e. principal component analysis (PCA) and subsequent linear discriminant analysis
(LDA), of the ATR-FTIR data showed that CNT treated algae differed from the corresponding controls at all
sampling times. No growth inhibition was observed, but CNT exposure changed the biochemical composition of
cells. The bioavailability of CNT to green algae and the influence of the material on the cells are important
findings with regard to possible food chain transfer and environmental risk assessment of CNT.
In the third research part of the present study, interactions between CNT and 17α-ethinylestradiol (EE2) and
resulting alterations on EE2 bioavailability to zebrafish (Danio rerio) was taken into account. EE2 is a highly
viii
relevant xeno-estrogen, as it is used in livestock farming for contraception usages, in human birth control pills,
and for other medical reasons. It has been detected in the ng/L range in surface waters of various streams.
Ethinylestradiol has a high effective potency at its environmentally relevant concentration range and therefore
poses a risk for aquatic organisms. The reported hazardous impacts to organisms range from direct toxicity to
endocrine disruptor impacts on higher organisms, such as fish. Adsorption of the environmental relevant
pollutant EE2 to CNT was observed in the water phase over time. This interaction might have been influenced
by CNT reagglomeration and possible surface modifications as detected in TGA measurements, caused by the
ultrasound dispersion of our samples. Further tests should be performed, to draw an entire picture of the
interactions, as non predictable results were observed for low EE2 concentrations. Experiments with zebrafish
showed the bioaccumulation of EE2 in males exposed to a 24 h aged mixture of 1µg 14C-labelled EE2 and
1 mg CNT per litre for one week. Maximum EE2 concentration in the fish was reached after 48 h leading to a
BCF of 280 L/kg fish wet weight. In a second experiment, with the same amounts, uptake of EE2 to different
compartments of the fish, i.e. bile, gut, gills, skin and liver, was quantified. In this experiment uptake and
effects of EE2 only and of CNT and EE2 in combination (24 h pre-aged dispersions) were evaluated. However,
only slight differences in the uptake and elimination behaviour of fish exposed to EE2 with and without CNT
material in the media could be outlined. Slightly increased uptake of EE2 was detected in the presence of CNT
material after 48 h, but also better elimination within 72 h when fish were transferred to clear media. A last
experiment aimed at the vitellogenin (VTG) induction in male zebrafish which is a reliable marker for estrogenic
endocrine activity. EE2 alone induced a high level of VTG production at 6.6 ng EE2/L, but no induction could be
observed in the fish-blood after they were exposed to 2.2 ng EE2/L alone. Remarkably, for both EE2 treatments
(2.2 and 6.6 ng EE2/L, exposed for 14 days), the VTG induction was low and in the same range in presence of
CNT material in the media. However, to our knowledge, for the first time combinational effects of CNT with
relevant environmental EE2 concentrations were reported, which showed novel effects at low pollutant
concentrations. These results raise the question what combinational potency CNT material may exhibit in the
environment. More complex studies should be performed to assess possible risks of CNT at environmentally
relevant concentrations of pollutants.
To conclude, the present work delivered new insights at the fate of CNT holding composite products at the end
of their life cycle: Moreover, new perspectives for CNT ecotoxicology were outlined and secondary effects of
CNT material and an environmentally relevant pollutant were investigated. These insights may lead to a better
understanding of the impact which nanotubes may have to aquatic organisms. However, more research is
needed to assess all possibly risks of an estimated release of CNT holding products or raw CNT material to the
environment, as our results show big gaps in the knowledge of CNT material behaviour and its possible
hazardous impacts.
ix
Zusammenfassung
In der vorliegenden Arbeit wurde eine Bewertung des Endes des Lebenszyklus von Kohlenstoffnanoröhren
(CNT)-haltigen Polymerkompositen durchgeführt. Die Studie besteht aus drei Teilen: (1) Bestimmung der
Freisetzung von CNT-Material aus Kompositen durch umweltrelevante Einflüsse, (2) ökotoxikologische Effekte
von CNT auf Primärproduzenten (Grünalgen) und (3) CNT bedingte Sekundäreffekte auf die Wirkung des xeno-
Östrogens-Ethinylestradiol (EE2) auf Zebrafische. Diese Arbeit identifiziert Umwelteinflüsse, welche zu einer
Freisetzung von CNT aus CNT-haltigen Kompositmaterialien führen können. Außerdem werden mögliche
schädliche Auswirkungen von freigesetztem CNT-Material in die Umwelt bestimmt.
CNT werden als eine der vielversprechendsten Materialien in der Nanotechnologie betrachtet. Sie bestehen
aus aufgerollten Graphitplatten mit einem Durchmesser im Nanometerbereich und einer Länge von bis zu
mehreren Mikrometern. Aufgrund ihrer Struktur weisen sie einzigartige Eigenschaften auf, z.B. hohe
elektrische und thermische Leitfähigkeit, sehr niedrige Dichte, hohe Flexibilität und extrem hohe Zugfestigkeit.
Zudem können Kohlenstoffnanoröhren ihre Eigenschaften auf Kompositmaterialien übertragen, wenn sie z.B.
Polymeren beigemischt werden. So machen sie Kunststoffe beispielsweise weniger entzündlich, verbessern
ihre elektrische Leitfähigkeit, ihre mechanischen Stabilität und ihre Flexibilitätseigenschaften. Es gibt bereits
zahlreiche mögliche Anwendungsbereiche für Nanokompositmaterialien wie z.B. im Bau, in der Luftfahrt- und
Raumfahrt, in der Medizintechnik oder in der Solartechnik. Da stetig neue mögliche Anwendungen entdeckt
werden, ist ein Produktionsanstieg für die Zukunft sehr wahrscheinlich. Daher ist eine erhöhte Freisetzung von
CNT in die aquatische Umwelt zu erwarten.
Mit Hilfe von 14C-markierten Nanoröhren wurde in dieser Arbeit zum ersten Mal die Freisetzung von CNT-
Material aus Nanokompositen (1 gew. % CNT) quantifiziert. Zur Herstellung der Nanokomposite wurden 14C-
CNT in Chloroform dispergiert und anschließend Polykarbonat-Pellets hinzugegeben. Nachdem das Chloroform
abgedampft war, wurde die Kompositoberfläche noch mit Hilfe einer Heißpresse optimiert. Es wurden runde
Proben mit einem Durchmesser von 18 mm hergestellt. Für die Quantifizierung der CNT-Freisetzung wurden
zwei verschiedene Probentypen eingesetzt. Zum einen wurden Proben mittels künstlichem Sonnenlicht
bewittert (1000 MJ/m2), zum anderen wurden nicht vorbehandelte Proben verwendet. Beide Probentypen
wurden dann einer Serie von Degradations-Szenarien unterzogen. Im Einzelnen wurden die Proben
Behandlungen mit Wasser, Hitze, Frost-Tau-Wechseln, Huminsäuren, saurem Regen, Deponieabwasser und
Abrieb ausgesetzt und parallel die CNT-Freisetzung bestimmt. Die vorbehandelten Proben zeigten hierbei eine
deutlich höhere CNT-Freisetzung im Vergleich zu den unbehandelten Kompositen. So wurden insgesamt 1% des
CNT-Materials aus den vorbehandelten im Vergleich zu 0,03% des CNT Materials aus den unbehandelten
Proben freigesetzt. Bezogen auf den Mittelwert der Jahressonneneinstrahlung in Florida betrug die Freisetzung
aus den bestrahlten Proben 18 mg pro m2 und Jahr. Rasterelektronenmikroskopische Aufnahmen (REM),
Infrarotuntersuchung (IR) und Röntgenphotoelektronen-spektroskopische (XPS) - Untersuchung der
Kompositoberflächen nach der Bestrahlung zeigten eine Freilegung von CNT-Material an der Oberfläche und
die Bildung eines dichten Geflechts aus Nanoröhren. Dieses CNT-Geflecht verzögert die Abbaugeschwindigkeit
des darunterliegenden Matrixmaterials signifikant. Somit konnte eine Freisetzung von CNT-Material aus
polymeren Nanokompositen detektiert werden, bedingt durch verschiedene umweltrelevante Faktoren.
Im zweiten Teil dieser Arbeit wurde die Bioverfügbarkeit von CNT für die einzellige Alge Desmodesmus
subspicatus untersucht. Die Algen wurden in Wachstumsmedien mit 14C-CNT (1 mg/L) exponiert, um die
Aufnahme von CNT an und in die Alge zu bestimmen. Weiterhin wurde die Abgabe von CNT-Material von der
Alge zurück ins Medium untersucht, wenn die Algen nach der Exposition in ein frisches Medium transferiert
wurden. Um den Einfluss von CNT-Exposition auf die biochemische Zusammensetzung der Algen zu bestimmen,
wurde „attenuated total reflection Fourier-transform“-Infrarotspektroskopie (ATR - FTIR) verwendet. Es
wurden Unterschiede zwischen exponierten und nicht exponierten Algen im Bezug auf verschiedene
Biomarkerregionen bestimmt. Elektronenmikroskopische Aufnahmen zeigten, dass das meiste CNT-Material
von außen an die Algenzellen angebunden war. Es wurden jedoch auch einzelne CNT im Zytoplasma detektiert.
Es konnte ein Konzentrationsfaktor von 5000 L/kg Trockengewicht für CNT bei D. subspicatus berechnet
x
werden. Eine computergestützte Hauptkomponentenanalyse (PCA) und anschließende lineare
Diskriminanzanalyse (LDA) der ATR - FTIR-Daten zeigten, dass sich die mit CNT behandelten Algen bei allen
Probenahmen zu den unbehandelten Kontrollen unterschieden. Es wurde keine Wachstumshemmung der
Algenzellen bei den eingesetzten CNT-Konzentrationen beobachtet,. aber die CNT-Exposition veränderte die
biochemische Zusammensetzung der Zellen. Die Bioverfügbarkeit von CNT auf Grünalgen und der Einfluss des
Materials auf die Zellen sind wichtige Erkenntnisse mit Hinblick auf eine mögliche Weitergabe in der
Nahrungskette und der Umweltrisikobewertung von CNT.
Im dritten Forschungsteil der vorliegenden Studie wurde die Wechselwirkungen zwischen CNT und 17α-
Ethinylestradiol (EE2) und die daraus resultierenden Veränderungen auf die EE2-Bioverfügbarkeit für
Zebrafische (Danio rerio) untersucht. EE2 ist ein umweltrelevantes synthetisches Östrogen, das z.B. innerhalb
der Nutztierhaltung und der Antibabypille zur Empfängnisverhütung verwendet wird. Es kann in verschiedenen
Oberflächengewässern in Konzentrationen im ng/L-Bereich gefunden werden. EE2 hat in seinem
umweltrelevanten Konzentrationsbereich eine effektive Wirksamkeit und stellt daher eine Gefahr für
Wasserorganismen dar. Die bekannten Auswirkungen von EE2 auf Organismen reichen von direkter Toxizität
bis hin zur endokrinen Wirksamkeit bei höheren Organismen wie z.B. Fischen. In einem ersten Experiment
wurde eine Adsorption von EE2 an CNT-Material in der Wasserphase beobachtet. Diese Interaktion wurde
beeinflusst durch CNT-Reagglomerierung und die Oberflächenmodifikationen an den CNT, die wir in
thermogravimetrischen Messungen (TGA) festgestellt haben. Experimente mit Zebrafischen zeigten die
Bioakkumulation von EE2 bei männlichen Fischen, welche einer 24 Stunden gealterten Dispersion bestehend
aus 1 µg 14C-markierten EE2 und 1 mg CNT pro Liter für eine Woche ausgesetzt wurden. Eine maximale EE2-
Konzentration im Fisch wurde nach 48 Stunden erreicht. Daraus ließ sich ein Biokonzentrationsfaktor (BCF) von
280 L/kg Fisch-Nassgewicht ableiten. In einem zweiten Experiment in dem die Fische den gleichen
Konzentrationen ausgesetzt wurden, wurde die Aufnahme von EE2 in verschiedenen Kompartimenten des
Fisches, also Galle, dem Darm, Kiemen, Haut und Leber, quantifiziert. In diesem Experiment wurde zusätzlich
ein Versuchsansatz mit nur EE2-Material angesetzt, um die Aufnahme (nach 48 Stunden) und die anschließende
Abgabe (nach 72 Stunden) von EE2 aus den Fischen zu bestimmen. Allerdings konnten nur geringe
Unterschiede in dem Aufnahme-und dem Eliminationsverhalten der Fische zu EE2 mit und ohne CNT-Material
in den angesetzten Medien detektiert werden. Es wurde eine leicht erhöhte Aufnahme von EE2 in Gegenwart
von CNT-Material festgestellt. Gleichzeitig war aber auch eine bessere Eliminierung der Substanz aus den
Fischen nach 72 Stunden zu beobachten, wenn die Fische in frisches unbelastetes Medium gegeben wurden. Im
letzten Versuch wurde die Induktion von Vitellogenin (VTG) in männlichen Zebrafischen bestimmt. VTG-
Induktion ist ein zuverlässiger Marker für die östrogene Wirksamkeit einer Chemikalie. Nachdem die Fische für
14 Tage 6.6 ng/L EE2 ausgesetzt worden waren, konnte eine hohe VTG-Produktion im Fischblut festgestellt
werden. Bei einer Konzentration von nur 2.2 ng EE2/L wurde jedoch kein VTG im Blut detektiert. Wenn CNT im
Medium vorhanden waren, fand man bei beiden EE2-Konzentrationen eine etwa gleich hohe Induktion der
VTG-Produktion, welche jedoch niedriger war als die Induktion bei 6.6 ng/L ohne CNT. Nach unserer Kenntnis
wurden zum ersten Mal Kombinationswirkungen von CNT mit den realistischen Umweltkonzentrationen von
EE2 untersucht. Dabei wurden neuartige Effekte bei niedrigen EE2-Konzentrationen beobachtet, welche so
nicht bekannt waren. Diese Ergebnisse werfen die Frage auf, was für kombinatorische Effekte CNT-Material in
der Umwelt hervorrufen kann. Es müssen komplexere Studien durchgeführt werden, um mögliche Risiken von
CNT in umweltrelevanten Konzentrationen von Schadstoffen beurteilen zu können.
Zusammenfassend gibt die vorliegende Arbeit neue Einblicke in das Schicksal von CNT-haltigen
Kompositprodukten am Ende ihres Lebenszyklus. Darüber hinaus wurden neue Perspektiven für CNT in der
Ökotoxikologie aufgezeigt und die Auswirkung von CNT-Präsenz auf einen umweltrelevanten Schadstoff
untersucht. Diese Erkenntnisse können zu einem besseren Verständnis der Auswirkungen, die
Kohlenstoffnanoröhren für Wasserorganismen haben, führen. Jedoch ist mehr Forschung nötig, um die Risiken
von CNT-Freisetzungen aus Nanokompositen oder purem CNT-Material für die Umwelt zu beurteilen. Unsere
Ergebnisse zeigen große Lücken in der Kenntnis des Materialverhaltens von CNT und CNT-haltigen Produkten,
sowie deren mögliche schadhafte Auswirkungen auf die Umwelt, auf.
xi
xii
xiii
Funding
Exchange to University of Lancaster for ATR-FTIR measurements was funded by the German
academic exchange service (DAAD).
This research has been financially supported by the CarboLifeCycle Project of the Inno.CNT frame
funded by the German Federal Ministry of Education and Research (BMBF) (FKZ 03X0114).
xiv
xv
Acknowledgments
First of all, I would like to thank my Doktorvater Prof. Dr. A. Schäffer, head of the Chair of
Environmental Biology and Chemodynamics (UBC), for supporting my whole education and work
process and for giving me the opportunity to perform my research independently.
I want to thank Prof. Dr. H. Hollert, director of the Department of Ecosystem Analysis (ESA), for co-
supervising and supporting my research.
I cordially want to thank Dr. H. Maes, postdoc in the CarboLifeCycle project, my supervisor and after
all a friend, for her tremendous patience when correcting my science English, the numerous
productive discussions and all the help during my whole work in Aachen. Dank U wel, Hanna!
Moreover, I want to thank A.-K. Barthel, Dr. V. Wachtendorf and Dr. A. Meyer-Plath from the Federal
Institute for Materials Research and Testing (BAM) in Berlin for their tireless efforts to introduce me
to material sciences and for great IR, SEM and XPS measurements.
I want to address thanks to Dr. M. Riding, Dr. F. Martin, Prof. K. Semple and Prof. K. Jones from
Lancaster University for being our partner within the DAAD project and for preparing great ATR-FTIR
measurements. I also want to consider Prof. Dr. W. Baumgartner, A. Weth, from the Cellular
Neurobionics at RWTH Aachen University, and Dr. M. Heggen, Ernst Ruska-Centre of the Research
Centre Jülich, for preparing great TEM pictures and helping me with all electron microscopic issues.
All project partners in the CarboLifeCycle project deserve my thanks for their great willingness to
help me with any question during my research. Many thanks go to Dr. Oliver Schlüter from Bayer
Technology Services for his support and guidance during the synthesis of 14C-labelled nanotubes. I
thank Prof. Dr. C. Popescu and B. Liebeck from the Interactive Materials Research division of DWI at
RWTH Aachen e.V. for their support with thermogravimetrical analysis. Furthermore, I do not want
to forget to thank Sebastian and Michael, who supported me in the lab.
Special thanks go to the whole staff and students of the Institute for Environmental Research for the
great working atmosphere and their willingness to help me anytime.
I would also like to thank my dear colleagues Anne S., Anne W., Felix, Jochen, Andrzej and Matthias
for their permanent willingness to help me with laboratory or office issues, as well as for all the
pleasant lunch and coffee breaks.
I would particularly like to thank my parents, my brother and all my friends for their ongoing support
and friendship during my whole education. You are the people who make my life worthwhile!
Finally, I would like to thank my beloved Katherina. You gave me balance during stressful times and
you offered me such great support during the final days of this work. You have the ability to make me
a better man!
xvi
xvii
Content
Summary .............................................................................................................................................. vii
Zusammenfassung................................................................................................................................. ix
Acknowledgments ................................................................................................................................ xv
Content .............................................................................................................................................. xvii
List of abbreviations ............................................................................................................................ xxi
Chapter 1 Overall introduction ............................................................................................................... 1
1.1 Nanotechnology ........................................................................................................................... 3
1.2 Carbon nanotubes (CNT) .............................................................................................................. 4
1.2.1 General information .............................................................................................................. 4
1.2.2. At the end of the CNT life cycle ............................................................................................ 5
1.2.3 Environmental fate ................................................................................................................ 5
1.2.4 Bioavailability ........................................................................................................................ 7
1.2.5 Nanoecotoxicology ................................................................................................................ 7
1.2.6 Interactions of CNT with pollutants and ecotoxicological consequences ............................ 12
1.3 17α-ethinlyestradiol (EE2) .......................................................................................................... 13
1.3.1 General information ............................................................................................................ 13
1.3.2 CNT-EE2 interactions ........................................................................................................... 14
1.4 CNT containing polymer composites .......................................................................................... 15
1.4.1 General information ............................................................................................................ 15
1.4.2 Environmental plastic debris problem ................................................................................. 16
1.4.3 Environmental issues of CNT containing polymeric composites.......................................... 17
1.5 Objectives and structure of the thesis ........................................................................................ 18
Chapter 2 Quantification of CNT material release from 14C-labelled CNT-polycarbonate composites.. 23
2.1 Introduction................................................................................................................................ 25
2.2 Material and Methods ................................................................................................................ 28
2.2.1 Carbon nanotube synthesis ................................................................................................. 28
2.2.2 Preparation of CNT-containing polycarbonate composites ................................................. 29
2.2.3 Exposure stresses ................................................................................................................ 31
2.2.4 Analysis methodologies ....................................................................................................... 35
2.2.4.1 Liquid scintillation counting (LSC) ................................................................................. 35
2.2.4.2 Morphology analysis (light microscopy, SEM) .............................................................. 35
2.2.4.3 ATR-FTIR spectroscopy ................................................................................................. 36
xviii
2.2.4.4 X-ray photoelectron spectroscopy (XPS) ...................................................................... 36
2.2.5 Data evaluation ................................................................................................................... 36
2.3 Results ........................................................................................................................................ 37
2.3.1 Composite production ......................................................................................................... 37
2.3.2 Light microscopy .................................................................................................................. 38
2.3.3 SEM pictures of the exposed composites ............................................................................ 39
2.3.4 IR-spectroscopy ................................................................................................................... 40
2.3.5 XPS ....................................................................................................................................... 42
2.3.6 Results of the sample degradation processes ..................................................................... 43
2.4 Discussion ................................................................................................................................... 45
Chapter 3 Interactions of multiwalled carbon nanotubes with algal cells: quantification of association,
visualization of uptake, and measurement of alterations in the composition of cells .......................... 53
3.1 Introduction................................................................................................................................ 55
3.2 Material and Methods ................................................................................................................ 57
3.2.1 Synthesis of unlabelled and 14C-radiolabelled carbon nanotubes ....................................... 57
3.2.2 Preparation of CNT test suspensions for exposure of Desmodesmus subspicatus .............. 57
3.2.3 Quantification of CNT accumulation by algae ..................................................................... 59
3.2.3 Visualization of CNT uptake in cells ..................................................................................... 60
3.2.4 Effects of CNT on the algal cell biochemistry ....................................................................... 61
3.3 Results ........................................................................................................................................ 64
3.3.1 Accumulation of CNT by algae ............................................................................................. 64
3.3.2 Visualization of CNT uptake in cells ..................................................................................... 66
3.3.3 Effects of CNT on the algal cell biochemistry ....................................................................... 67
3.4 Discussion ................................................................................................................................... 71
Chapter 4 The influence of carbon nanotubes (CNT) to the bioavailability of 17α-ethinylestradiol to
zebrafish (Danio rerio) .......................................................................................................................... 77
4.1 Introduction................................................................................................................................ 79
4.2 Material and Methods ................................................................................................................ 81
4.2.1 Carbon nanotubes ............................................................................................................... 81
4.2.2 17α-ethinylestradiol ............................................................................................................ 81
4.2.3 Filtration experiments ......................................................................................................... 82
4.2.4 Thermogravimetrical analysis (TGA) .................................................................................... 83
4.2.5 Experiments with zebrafish (Danio rerio) ............................................................................ 84
4.2.5.1 Test organism ............................................................................................................... 84
4.2.5.2 Biodistribution experiments ......................................................................................... 84
xix
4.2.5.3 Accumulation & elimination experiments .................................................................... 85
4.2.5.4 Effect test (vitellogenin induction) ............................................................................... 86
4.2.6 Statistics .............................................................................................................................. 90
4.3 Results ........................................................................................................................................ 90
4.3.1 Filtration Experiments ......................................................................................................... 90
4.3.2 TGA ...................................................................................................................................... 92
4.3.3 Fish tests .............................................................................................................................. 94
4.3.3.1 Biodistribution .............................................................................................................. 94
4.3.3.2 Accumulation and elimination ...................................................................................... 96
4.3.3.3 Effect tests (VTG) .......................................................................................................... 99
4.4 Discussion ................................................................................................................................. 100
Chapter 5 Overall discussion .............................................................................................................. 111
References ......................................................................................................................................... 121
List of publications ............................................................................ Fehler! Textmarke nicht definiert.
xx
xxi
List of abbreviations
µg microgram
ANOVA analysis of variance
ATR-FTIR attenuated total reflection Fourier-transform infrared spectroscopy
BCA bicinchoninic acid assay
BPA bisphenol A
Bq Becquerel
CCVD catalytic chemical vapor deposition
CF concentration factor
CNT carbon nanotubes
d day
DDT Dichlorodiphenyltrichloroethane
dw dry weight
E720nm spectrophotometrical measure for the extinction of light with a wavelength
of 720 nm
EC50 50% effective concentration
EE2 17α-ethinylestradiol
ELISA Enzyme linked Immunosorbent Assay
EPS extracellular polymeric substances
EU European union
g gram
h hour
HA humic acids
HNO3-MWCNT nitric acid functionalized multiwalled carbon nanotubes
HOC hydrophobic organic chemicals
IR infra-red
k1 uptake rate constant
k2 elimination rate constant
ke true elimination constant
kg kilogram
KOW octanol:water distribution coefficient
L litre
LDA linear discriminant analysis
LOEC lowest observed effect concentration
LSC liquid scintillation counting
xxii
M molar
mg milligram
min minute
mL millilitre
MWCNT multiwalled carbon nanotube
nC60 fullerene
ng nanogram
NOEC no observed effect concentration
NOM natural organic matter
NP nanoparticles
OECD Organisation for Economic Co-operation and Development
PAH polycyclic aromatic hydrocarbons
PBS phosphate-buffered saline
PBT persistent bioaccumulative and toxic substance
PC polycarbonate
PCA principal component analysis
PEC predicted environmental concentration
PFC perfluorochemicals
pg pictogram
PNEC predicted no-effect concentrations
POM polyoxymethylene
R² coefficient of determination
ROS reactive oxygen species
sec second
SEM scanning electron microscopy
SiO2 silicon dioxide nanoparticles
SWCNT single walled carbon nanotube
TEM transmission electron microscopy
TiO2 titanium dioxide
TR total amount of CNT material released during all degradation treatments
UV ultraviolet
VTG vitellogenin
ww wet weight
XPS X-ray Photoelectron Spectroscopy
ZnO zinc oxide nanoparticles
xxiii
xxiv
Chapter 1
Overall introduction
Chapter 1
2
Chapter 1
3
The introduction section of the present work is subdivided into the following general introduction,
and a separate introduction at the beginning of each research chapter (Chapter 2, 3 and 4). The latter
ones will provide a more detailed introduction to the topics dealt within each chapter.
1.1 Nanotechnology
Nanomaterials display a novel field in research and technology. The European Union (EU) has defined
them as materials and particles smaller than 100 nm in at least one dimension (2011/696/EU)1. Due
to their unique physical and chemical characteristics, these particles provide novel capabilities for
many fields in technology. Several different materials and chemicals have been classified as
nanoparticles (NP), such as inorganic metaloxide based NP (titanium dioxide; ZnO; SiO2 etc.), carbon-
based NP (e.g. fullerenes, carbon nanotubes, carbon black) or metal NP (e.g. nano-scaled Fe or Ag)
and there are still more to be discovered in the future2-5. The production scale of nanoparticles is
expected to increase in the future as many promising applications are currently under development3,
6-12. Hence, intentional (e.g. water cleaning applications)12-14and unintended release of these novel
particles in the environment is highly expected15-18. Up to now, there is still a lack of information on
the fate, behaviour and effects of NP in the environment and due to their novel properties and small
scale, detection methods and available characterisation for several NP are not sufficient and still
under development3, 4, 7.
To estimate potential release and end of life cycle fate of engineered nanomaterials (ENM), an
increasing number of modelling studies is available well reviewed by Gottschalk et al. (2013)19. As in
most cases no reliable release rates are available, all predicted environmental concentrations (PEC)
are assumptions based on comparable materials or production rates. The urgent need for realistic
release rates and scenarios is great, as production of nanoparticles is increasing and their release to
the environment highly estimated, although predicted no-effect concentrations (PNEC) were much
higher than modelled PEC20. Therefore, most studies do not outline a high risk for ENM to cause
great environmental impact18-21. However, a more detailed link between effect studies4, 22-24 and
Chapter 1
4
realistic exposure scenarios has to be drawn, as there are many uncertainties in the occurrence and
behaviour of ENM in the environment19, 20, 25, 26. Moreover, less investigated secondary effects, e.g.
interactions with pollutants, of ENM are estimated to occur at the environment20, 27, 28.
1.2 Carbon nanotubes (CNT)
1.2.1 General information
Carbon nanotubes (CNT), first described by Iijimain in 199129, consist of cylindrical rolled graphene
sheets with a nanoscale diameter9, 16, 29, 30. There are different types of nanotubes, i.e. single walled
CNT (SWCNT), with one graphene layer, and also multiwalled CNT (MWCNT), with up to 30 walls or
more30. In the literature MWCNT were described as “Russian dolls”, because smaller tubes were
continuously embedded in bigger ones6. The dimensions of CNT range from 1.4 nm (SWCNT) to 50
nm (MWCNT) in diameter, depending on the number of walls, with up to several micrometers in
length7, 29, 30. The structure of CNT leads to unique characteristics, such as high mechanic stability,
good electric conductivity and very low density6, 7, 9, 30. CNT also exhibit a large surface area due to
their high surface to volume ratio9.
Their extraordinary properties make CNT one of the most promising and innovative materials in
nanotechnology. A wide range of applications has already been discovered, such as use in batteries,
for drug delivery or even bone recovery in health section, and usage in the material section, mainly
as property improving additives, but still more is expected for the future6, 16, 30, 31.
Single nanotubes were identified in 10,000 year-old ice cores, estimated to originate from
vulkanism32. Nevertheless, CNT are an anthropogenic product with no environmentally relevant
occurrence. CNT are synthesised by different methods, i.e. carbon arc discharge, laser-vaporization,
electrolysis and mostly by catalytic chemical vapour deposition (CCVD)30, 33, 34. The latter method was
used to produce MWCNT in the present work. It includes round graphene sheets growing around a
metal catalyst (Fe, Ni, Co etc.). Depending on the carbon source (e.g. methane; ethylene; benzene),
Chapter 1
5
catalyst, substrate (graphite, zeolite, CaCO3 etc.), carrier gas (H2, N2) and temperature setting,
different types (SWCNT, MWCNT) and qualities of nanotubes can be produced16, 30, 33. After synthesis,
the raw product is purified, in order to remove remaining catalyst particles or amorphous carbon
regions. Purification takes place by washing the CNT in acid or by electrolysis35, 36. Industrially
produced CNT from different companies as well as batch to batch variations show different purities
and qualities34, 37, 38. This outlines the importance for the development of reliable characterisation
techniques and standards34.
1.2.2. At the end of the CNT life cycle
To estimate the potential end of life cycle fate of raw CNT material as well as CNT material processed
in products, an increasing number of studies dealing with this topic are available19, 39-41. For carbon
nanotubes, first modelled PEC-values for surface water amounted to 0.0005 µg CNT/L17, and
sediment concentrations were estimated to be in µg/kg dry weight range17, 42. Release studies
estimated CNT to be released primary to wastewater treatment facilities, landfills and soils17. Life
cycle studies dealing with CNT exhibit large uncertainties, especially at the end of the life cycle43 as it
is not clear what happens with CNT holding products during disposal. Nanotubes, especially if
embedded in composite holding products, may persist combusting processes during incineration and
could thus be deposited afterwards41, 44, 45. The need to fill the gaps is emerging and research is still
going on to perform a reliable risk assessment.
1.2.3 Environmental fate
Nowadays, there is still a big lack of data to estimate the possible distribution of nanotubes in the
environment. Due to the estimated production scale up, fate and behaviour in the environment of
this novel material have to be intensively assessed further.
Chapter 1
6
Air and water were identified as main media to distribute CNT in the environment46. Whereas, CNT
transport in soil is estimated to be hindered47-49. There are several toxicological studies available
dealing with airborne particles, considering possible asbestos like impact of CNT, as both exhibit a
similar structure50-52. However, the present study is dealing with aquatic organisms and therefore,
airborne particles and soils will not been discussed any further.
CNT entering lakes, ponds or rivers outline a potential risk for aquatic organisms as chemicals and
materials could distribute more easily in the aqueous phase, compared to ashore impacts. Though,
high hydrophobicity makes raw CNT poorly dispersible in water. Consequently, CNT instantly form
agglomerates and show fast sedimentation inside the water phase or deposition at environmental
surfaces53-55. Strong attachment of CNT to natural organic matter (NOM) and sediment particles was
observed9, 53, 56-61. Binding to NOM is known to stabilize CNT in the water phase, whereas adsorption
to sediment particles displays a sink for nanotubes. Petersen et al. (2010)62 investigated the octanol-
water distribution of raw CNT, expecting them to act like hydrophobic organic chemicals (HOC).
However, they found CNT to only sparsely cross between the two phases (octanol-water), when
dispersed in one of them. They estimated the length of the nanotube particles to be responsible for
this particularly behaviour. Hence, determination of a log KOW is not applicable for nanotubes, and
evaluation of CNT behaviour based on similarities to other chemicals is difficult.
CNT are estimated to be very persistent in the environment because of their unique structure.
However, when surface failure or oxidations are present, enzymatic CNT degradation and
mineralization processes up to carbon dioxide was reported in first studies63-67. Co-metabolism was
identified to be the main process in bacteria CNT degradation as there were only degraded in the
presence of another carbon source68. In the future it has to be evaluated if this also occurs in the
aquatic environment, e.g. within sediments.
In general, different types of nanotubes as SWCNT, MWCNT or tubes with functionalized surfaces
show different behaviour in the environment, e.g. oxidized surfaces make CNT better dispersible69.
This raises the need to gain detailed information on the bioavailability, effects and behaviour of all
types of CNT. Moreover, the development of applicable characterisation and detection techniques
Chapter 1
7
for CNT in environmental samples is highly required. This information is important for performing
reliable risk assessment for all kind of carbon nanotubes.
1.2.4 Bioavailability
The bioavailability of CNT for organisms is the basis in assessing its potential hazard and effects to
organisms. This means that CNT are able to directly or indirectly interact with organisms, e.g.
entering their cells. No or only low bioaccumulation behaviour of CNT for aquatic and sediment
organisms was reported up to now70, 71, whereas uptake of CNT from the aqueous phase to different
aquatic organisms has been repeatedly reported. An uptake of CNT in Stylonychia mytilus, an
unicellular protozoan, was detected as CNT were rapidly ingested and agglomerated inside the
organisms72. Afterwards, some of the aggregates were excreted or found attached to the cell
membrane and the micronuclei. Kennedy et al. (2008)53 described functionalized and raw CNT
material to be taken up in the gut of Ceriodaphnia dubians. Also, ingestion of MWCNT to the gut of
Daphnia magna was observed during another study73, but no entrance of nanotubes in any tissue
cells was detected. However, daphnids are able to modify ingested nanotubes, as biomodification of
lipid coated SWCNT was observed74. The daphnids used the lipids as food source and excreted
precipitated SWCNT to the medium. Hence, the daphnids were able change the chemical and
physical properties of CNT in aqueous dispersions.
It can be summarized that functionalisation of nanotubes, the biomodification by organisms and the
bioavailability has to be considered for toxicological testing. Uptake of CNT in organisms might lead
to food chain transfer, secondary effects and transport of adsorbed pollutants in aquatic organisms.
1.2.5 Nanoecotoxicology
Due to their special properties and thus challenging issues during the determination of nanoparticles
ecotoxicological potential, the term “nanoecotoxicology” was formed by Kahru and Dubourguier in
Chapter 1
8
200975. This term should outline the special circumstances and challenges for nanoparticle ecotoxic
evaluation. Handy et al. (2012)76 extensively reviewed the practical considerations, which have to be
mentioned, when using NP in ecotoxicology test systems. In general, NP behave differently compared
other chemicals and therefore special attention has to be drawn towards different parameters, e.g.
dispersion process and agglomeration state in the test systems. This is the main reason, why
heterogeneous results can actually be found in the literature dealing with CNT nanotoxicity4, 39, 77-79.
Direct toxicity is directly related to several factors, for instance the dispersion state, which causes
single CNT fibres or agglomerates in the test system. CNT dispersion state changes over time as
nanotubes tent to reagglomerate. Moreover, in most cases, a difference between SWCNT and
MWCNT toxic impact could be observed, which may contribute to the higher surface area and
smaller size of SWCNT compared to multiwalled ones. However, all these characteristics and the
behaviour of CNT have to be considered for nanotube toxicity testing. In the following, an overview
on the existing literature of CNT toxicity in the aquatic environment will be given.
As water was identified to be a main media for CNT (Chapter 1.2.2), effects of CNT have been studied
towards many different aquatic organisms. In the most cases, no acute hazardous effects of
nanotubes on aquatic organisms could be detected. Although an ingestion of MWCNT to the gut of
D. magna was observed, no acute mortality or absorption were detected73, 80. Only a MWCNT
clogging at the organisms surfaces, as agglomerates stuck to the carapace of the daphnids and
inhibited the daphnid movement, could be observed for CNT concentrations (≥ 0.4 mg MWCNT/L).
This effect was also described for SWCNT74. Once ingested by organisms, mostly food provision or
sediment particle ingestion were able to eliminate the main amount of CNT material from the
organisms’ guts73, 81-84. Using mainly microscopic techniques, it was observed that CNT precipitated
on the gills of rainbow trout85 after exposure of the organisms via aqueous dispersion of CNT. It
remains to be clarified whether nanotubes are bioavailable, i.e., that they are resorbed and even
accumulated in exposed organisms. Incorporation of CNT material through absorption across
external tissues and the gut epithelium has not been reported so far39, although it is the search for
the (nano-)needle in the haystack.
Chapter 1
9
Nevertheless, CNT were shown to be present in cells of different unicellular organisms such as
protozoan and bacteria held in laboratory cultures72, 86, 87. Escherichia coli and Stylonychia mytilus,
were reported to incorporate CNT from their surrounding medium72, 88, and several studies showed
unicellular algae to interact with CNT89-92. Inhibition of algal growth following to CNT exposure was
repeatedly related to light absorption by the nanomaterial causing shading of the cells and to
agglomeration and physical interactions of algae with CNT rather than to a specific mode of toxic
action89, 93. Most effects were only observed at high CNT concentrations (>10 mg/L), which are not
assumed to be environmentally relevant as in modelling surveys, surface water concentrations lower
than ng/L were predicted5.
Disturbance of the metabolic activity of Escherichia coli was detected after cellular uptake of CNT88.
Therefore, CNT may have direct or indirect influences on the chemical reactions in a cell and may in
this way cause sublethal or chronic effects. Uptake by membrane perturbation or during cell division
were two possible ways of CNT to enter cells and cause cytotoxic effects78, 94. Further possible
uptake-routes are currently unknown and have to be investigated in the future.
Kennedy et al. (2008)53 observed different effects for different nanotubes (functionalized and raw
CNT) exposed to the cladoceran Cerodaphnia dubia. They found CNT stabilized by natural organic
matter to show higher toxic effects as functionalized ones. Blaise et al.(2008)95 performed a battery
of standard tests (algae, bacteria and micro-invertebrates) with SWCNT, but they only observed the
green alga Pseudokirchneriella subcapitata to be affected by nanotube exposure. At concentrations
of about 1 mg SWCNT /L its growth was inhibited, but all other exposed microorganisms did not
show acute toxicity towards SWCNT. In most studies SWCNT were identified to exhibit a larger toxic
potential as MWCNT, suggested to be related to their higher surface area and smaller size96, 97.
Since most described effects could not be linked to distribution of NP and/or agglomerates to the
interior of exposed organisms, the underlying mechanisms of rarely observed CNT toxicity remains
unknown. It seems to be possible that physiological changes are a result of mechanical interactions
between the nanomaterial and the outer tissues. For example, Smith et al. (2007)85 found SWCNT in
the ventilation rate of rainbow trout in a dose dependent manner. Furthermore, pathologies in gill
Chapter 1
10
tissues and vascular lesions in the brain of exposed fish were detected. It is not unlikely that
respiratory stress caused by gill irritation may also trigger the observed brain injury due to oxygen
deficit. Adherence of aggregates on the external surfaces and accumulation of CNT in the gut of
Daphnia magna were observed. External adherence led to immobilization through the water column,
ultimately leading to mortality74. Whereas uptake of CNT in epithelial tissues of multicellular
organisms could not be shown, internalization of nanotubes in mammalian cells was repeatedly
observed86, 98, 99. With regard to higher organisms, a delay in hatching of zebrafish (Danio rerio)
embryos was observed, but no effect on the development was detectable after the exposure to
CNT100. Due to agglomeration and the high affinity of the hydrophobic CNT to sediments, the
material tends to settle down to the bottom and remains in the sediment phase of aquatic systems17,
20, 91. Additionally, tests with sediment assays were carried out using Lumbriculus variegatus81 and the
amphipods Hyalella azteka & Leptocheirus plumulosus53. These tests figured out that CNT strongly
adsorb to sediment particles and pass the gut of sediment ingesting organisms without being
distributed to the organism tissues leading to no direct toxic effects.
Several studies about effect on the cellular level, i.e. cytotoxicity, have been carried out in the past to
address possible hazardous effect of CNT to human tissues. Many publications report a toxic impact
to different cell types, such as human macrophage cells or unicellular microorganisms88, 101, whereas
others did not observe any influence to the CNT-cell interaction 102. SWCNT and MWCNT seem to
have different, but related toxicities where SWCNT are pronounced to be more toxic, probably due
to its larger particle surface area and smaller size88, 101. Moreover, the aggregation state and peptide
coating seem to influence the toxic impact97, 103. CNT were shown to be able to enter cells and to
accumulate inside by several research groups78. CNT entering cells were found to distribute
differently inside, depending on their characteristics. Different modified nanotubes were estimated
to accumulate in different parts of the cell, such as cytoplasm, nucleus or the organelles, leading to
different effects10, 102, 104. However, CNT uptake into cells embodies foreign material inside the cell.
The discussion of the cytotoxic mode of action is still ongoing. Some research groups reported metal
impurities at the CNT to stimulate toxic effects in the cells97, whereas others could not underline a
Chapter 1
11
difference between purified and unpurified CNT87. However, bioavailability of impurities is greatly
enhanced by ultrasonication treatment of CNT material and this may give a possible explanation for
this gap105.
Due to their properties, CNT are able to absorb a wide range of pollutants, and may therefore act as
possible carriers for pollutants into cells, too (further referred to in chapter 1.2.6). Moreover, CNT
have been repeatedly reported to be possible carriers for different molecules, as e.g. peptides or
DNA (deoxyribonucleic acid) fragments, to cells102, 106-108. Beside, a possible delocalisation of
molecules from inside the cells to its external environment should also be investigated in the future.
There is evidence from in vitro and in vivo experiments that CNT induce the production of reactive
oxygen species (ROS), which may elicit oxidative stress at the cellular level86, 88, 109-111. Oxidative stress
means the increased production of ROS, such as oxygen ions or peroxides, which can induce various
forms of damage to cells. Reactions of ROS with biological molecules or membranes can result in
irreparable damage to the cell structures and lead to cell necrosis and even cell death112. Subsequent
indirect consequences as lipid peroxidation causing disruption of cell membranes and DNA damage
were previously observed following exposure of CNT to different cells52, 113-115. Membrane damage is
also connected to nanotube cytotoxicity, such as possible perturbation or piercing of nanotubes
through the outer cell membrane brought novel transport ways for pollutants from the periphery
into the cell or molecules out of the cell116, 117. For instance, MWCNT pierced root cap cells were
reported to take up phenantrene from the surrounding medium, while unpierced cells did not show
any significant uptake116.
Summarising, CNT often did not show acute toxic effects to organisms, but uptake process and
detailed mode of action in the organism cells and tissues are still largely unknown. This indicates that
CNT cannot be treated as harmless material and research on its toxicity has to be continued,
particularly as CNT are to be used in the health sector. Last but not least, effects due to chronic
exposure to nanotubes are also a possible pathway for CNT toxicity which has to be studied more
detailed in the future, as up to now most studies only assess the acute toxic potential for this
nanomaterial. One study reported small concentrations of MWCNT to influence the composition of
Chapter 1
12
benthic communities, at environmental relevant concentrations, which cannot be extrapolated from
in vitro test systems118. This may be a first hint that the toxic potential of CNT is not entirely
determined and much more complex test systems, environmental realistic settings and secondary
side effects have to be evaluated.
1.2.6 Interactions of CNT with pollutants and ecotoxicological consequences
Nanotubes exhibit a large adsorption potential, caused by their large surface area9. Many
hydrophobic chemicals (e.g. dichlorodiphenyltrichloroethane (DDT), polycyclic aromatic
hydrocarbons (PAH), antibiotics, endocrine active substances as 17α-ethinylestradiol (EE2), and
several more are known to adsorb to nanotubes15, 119-122. The adsorption takes place at the outer CNT
surface and not at the inner surface or between the walls upon a MWCNT15. Different nanotubes
ware also investigated to act as good absorbers for metals, such as Ni, Cd or Pb and, therefore, usage
of CNT in water pollutant removal is being investigated 9, 15, 122-124. Due to their properties, CNT are
discussed as adsorbents for remediation and water clearing applications57, 121, 123-126. However,
chances for success are doubtful, as costs of CNT are high up to date and in some cases activated
carbon delivered better results as the nanoparticles.
In general adsorption behaviour of CNT can be influenced by several parameters, e.g. pH,
temperature, and other materials59, 119. The presence of humic acids (HA) in the media can alter the
adsorption behaviour of chemicals to carbon nanotubes as HA decreased the sorption of pyrene to
CNT119. Furthermore, in this study the authors reported the CNT dispersion state to play an important
role in adsorption behaviour, as agglomerates exhibited a lower adsorption affinity than well
dispersed individualized CNT119. Another study showed CNT to exhibit differed adsorption behaviour
towards PAH at low concentrations127. It can be concluded that different parameters may alter the
adsorptive potential of CNT and therefore have to be minded.
Possible consequences of adsorption of pollutants to NP may be an altered ingestion and
bioavailability of these materials for organisms. Towell et al. (2011)128 found CNT to reduce the
Chapter 1
13
extractability and mineralization of 14C- labelled PAH in soil when exposed together for two weeks.
Moreover, the authors suggested a reduced bioaccessability of adsorbed PAH to organisms. MWCNT
were described to reduce the aqueous concentration of perfluorochemicals (PFC)129. As a
consequence, the uptake rate constant to sediment living larvae Chironomus plumosus was reduced
and therefore their bioavailability. However, it was suggested that ingestion of NP-pollutant
complexes may increase bioaccumulation and a need for further research was addressed. In another
study, increased bioavailability of different PAHs to sediment living larvae of C. plumosus was
estimated due to ingestion of CNT adsorbed material130. Furthermore, CNT were reported to
significantly retain PAH in soil, which may also lead to higher concentrations in organisms131. A very
interesting publication reported the desorption of PAH aligned to CNT surfaces in simulated
gastrointestinal fluids132. Thus, pollutant-loadings may become bioavailable when digested by
organisms after uptake of CNT-pollutant complexes.
However, only few studies dealing with CNT influence on pollutant bioavailability are available up to
now and due to the relevance of this topic. Much more efforts have to be spent on research in this
area, with a special focus on secondary impact of nanotubes to organisms, as only direct toxicity is
used for risk assessment nowadays.
1.3 17α-ethinlyestradiol (EE2)
1.3.1 General information
17α-ethinylestradiol is a highly relevant xeno-estrogen, as it is used in agriculture for contraception
usages, in human birth control pills, and for other medical reasons133, 134. Up to now, it is not
technically possible to fully remove EE2 in sewage treatment plants. The runoff from agricultural
used areas represent another route for this substance to enter the environment135. Thus, it can be
detected in the ng/L range in surface waters of various streams136-139. Moreover, due to its structure,
Chapter 1
14
it is persistent in river systems with a dissipation time ranging from 20 to 40 days140 and only slow
biodegradation was reported141, 142.
Maes (2011)143 reported a quiet high bioaccumulation and biomagnifications potential for EE2 for
several aquatic organisms (i.e. algae, daphnids, fish). Moreover, ethinylestradiol has a high effective
potency at its environmentally relevant concentration range and therefore poses a risk for aquatic
organisms144. PNEC-value for EE2 for long termed exposure in surface water amounts to 0.1 ng/L145.
The reported hazardous impacts to organisms raise from direct toxicity to endocrine disruptor
impacts at higher organisms, such as fish143, 146. Fish exhibit a low effective concentration for EE2, as
induction of estrogenic responses in juvenile rainbow trout within 14 days of exposure to EE2 ranging
from 0.85 to 1.80 ng/L144, which is in accordance with the literature. Especially zebrafish (Danio rerio)
were reported to exhibit a large sensitivity to EE2143, 147. Ethinylestradiol exposure to D. rerio was
described to adversely affect the survival, growth, reproductive functions, sex differentiation, and
breeding success of these organisms148, 149. Estrogenic impact on these organisms could be reliable
detected by induction of the egg yolk precursor protein vitellogenin (VTG) in male zebrafish. Under
normal conditions VTG is synthesized by the liver of female fish. The sensitivity for VTG induction at
low EE2 concentrations is in the range of ng/L, i.e. EC50 value of 2.51 ng/L during 8 days exposure150,
and one study performed for 14 days reported a low observed effect concentration (LOEC) for
plasma VTG induction in male zebrafish of 5 ng/L151. As a lipophilic compound, EE2 tends to
accumulate in biota, a bioconcentration factor (BCF) for male zebrafish was reported to be 960 L/kg
wet weight152.
1.3.2 CNT-EE2 interactions
Due to its hydrophobicity, EE2 tends to adsorb to organic material such as natural organic matter
(NOM) or soil and sediment particles145, 153. In a study of Pan et al. (2008)122, EE2 was described to be
adsorbed to MWCNT by π- π electron donor-acceptor interactions. The graphene surface of CNT
offers hydrophobic surfaces and the potential to interact with chemicals having π-electrons in their
Chapter 1
15
structure28, 154. Adsorption of EE2 to organic material will probably influence its bioavailability and
bioaccumulation behaviour. On the one hand, binding of EE2 could decrease its bioavailability to
aquatic organisms, but on the other hand an increase in bioavailability may occur after the
adsorption by concentrating EE2 in fish guts. Hereafter, it may get to the blood cycle of the fish. A
study of Park et al. (2010)155 showed nC60 fullerenes to significantly hinder bioavailability of EE2 to
zebrafish when combined and fed by brine shrimp dietary compared to EE2 exposed brine shrimp.
Schwab et al. (2013)156 showed that diuron adsorption to CNT agglomerates was more toxic to algae
compared to free diuron as they estimated a locally accumulation of the pollutant near the algae
surface. The same effect might occur in the fish gastrointestinal tracts, too, as one study also
described the ability of digestive fluids to release adsorbed pollutants from the CNT surface132. These
results may give a first hint what to expect for a combinational exposure of EE2 and CNT to
organisms.
1.4 CNT containing polymer composites
1.4.1 General information
A novel branch of plastic products are polymer composites containing CNT157. Carbon nanotubes
were added to a polymeric matrix (e.g. polycarbonate, polyethylene, epoxy resin) by melt mixing, ball
milling, shear mixing, injection moulding or other techniques with different concentrations (mainly
low CNT percentages as 0.5 to 7.5 weight percent are commonly used)8, 157-164. Due to their unique
properties, i.e. high electrical and thermal conductivity, very low density, flexibility and extremely
high tensile strength, CNT transfer their properties to neat polymeric matrices, when compounded
together, as they e.g. make plastic composites less flammable, enhance its electric conductivity and
improve the mechanic stability and flexibility162, 164-168. An important parameter is the particle
dispersion in the matrix material, as better dispersed CNT forming a network inside the polymeric
material enhanced its properties more compared to embedded CNT agglomerates169-171. CNT
Chapter 1
16
containing polymeric composites are increasingly used for construction172, aerospace173, medical 31,
174, 175, water cleaning176, solar-harvesting177 and several more applications. Regarding to the
increasing amount of available literature, many applications will appear in the future, at it is one of
the most promising applications for CNT up to now.
1.4.2 Environmental plastic debris problem
The impact of plastic waste in the environment is a problem of growing concern. Since the last
decades, several studies dealing with environmental sampling showed alerting data, as plastic waste
of different size and shape was found in nearly every aquatic system around the globe, especially in
the water phase and sediments of oceans and shorelines178-186. Polymers originate from
anthropogenic waste, disposal sites, and fell off during shipping, and all kinds of plastic particles
could be found. Polymers are degraded in the environment by several chemical and biological
factors, but these processes are mainly slow with durations up so several years in salt water, as
reported for example for plastic bags184, 187-190. Plastic debris is known to have a hazardous impact to
the environment whereas accumulation in several compartments is reported and, still worse,
ingestion and effects to organisms were described181, 191-195. Impacts were found all over the aquatic
food web and, therefore, particles ingestion was described e.g. for mussels, fish, seabirds or seals192,
194, 196-200. Moreover, small microplastic fragments with only several µm in diameter or even smaller,
were detected in several aquatic organisms and seabirds worldwide192, 193, 201. Once ingested,
particles could be hardly excreted by the organisms and therefore induces toxic impacts as stomach
stuffing observed in seabirds202. Furthermore, leaching of additives as bisphenol A (BPA) from plastic
particles to organisms was described203, 204. This may cause a toxic impact as BPA endocrine activity.
Due to their hydrophobic surfaces, polymeric particles were described to be good adsorbents for a
wide range of environmental pollutants, e.g. persistent bioaccumulative and toxic substances (PBTs)
and metals205-210. Accumulated with pollutants from the water phase or even from their production,
Chapter 1
17
plastic particles were found to be possible transport vectors for pollutants to organisms as worms or
fish193, 206, 211, 212. José Derraik (2002)178 offered a great extended review on the plastic debris problem.
Several studies dealing with plastic debris are currently ongoing and it can be concluded, that the full
picture of this problem is not yet drawn in research. Moreover, with causing production scale up,
persistent plastic debris will be a big future issue to be solved.
1.4.3 Environmental issues of CNT containing polymeric composites
Due to the expected production scale up for the future (Chapter 1.4.1) and therefore more CNT
holding polymeric composite waste, a release of these products to the environment at the end of its
life cycle can be estimated, as already present for polymeric products (Chapter 1.4.2). Once released,
CNT holding composites will underlie several degradative influences which will change the
morphological appearance of the composite surfaces or the release of CNT particles, or small scaled
composite particles213, 214. CNT polymeric composites outline a special hazard for the environment,
as, additionally to the impacts of the matrix material, CNT can be potentially released or at least
uncovered at the composite surfaces. Petersen et al. (2011)39 expected several influences during
usage or disposal, such as biodegradation, mechanical, washing, diffusion, matrix degradation by
photo-, thermo- or hydrolytic impacts, and incineration to cause matrix material impact and
therefore potential CNT release. Uncovering of nanotubes at polymeric surfaces by ultraviolet (UV)
radiation from simulated sunlight was already described before215-217. A dense entangled CNT layer
uncovered from matrix material at the surface could be detected during dry UV exposure, but no
released CNT. Particle release due to wash off was detected to be supported by rains simulation
during artificial weathering218. Also abrasion of small composite particles was observed during
different treatments, which may also play a role in the environment41, 219, 220. Due to the fact that
composite products with embedded CNT are durable products potential CNT release may occur over
a long time period40. Hence, potential release of CNT has to be estimated to evaluate the potential
risk of this product. Furthermore, toxic potential of small composite particles with entangled CNT at
Chapter 1
18
its surface should be investigated intensively, as they surely will be ingested by organisms and then
cause a possible toxic impact.
A probable secondary effect of CNT polymeric particles is the sorption of pollutants to uncovered
CNT particles at their surface. Adsorption to the polymeric material itself was also described
before193, 206. Hence, small hedgehog like particles containing uncovered CNT at the surface and a
degraded matrix material core may outline a possible risk to aquatic organisms. These particles might
be loaded with pollutants in the water phase and then ingested by fish or sediment organisms and
intoxicate them, as they led to higher body burdens and can possibly release the pollutants due to
digestive fluids132.
Concluding, several environmental impacts of released CNT containing polymer composites at the
end of their life cycle can be estimated. More attention should be spent on ecotoxicological studies
and risk assessment of this material as it may outline a potential hazard material for the
environment, not only due to potential nanoparticle release.
1.5 Objectives and structure of the thesis
This thesis is structured into three research chapters beginning within the ongoing chapter. Each
chapter has its own individual introduction, methods, results and discussion section. In a last chapter,
a overall discussion and conclusion section is provided. All references cited within the chapters have
been listed in a separate section towards the end.
The aim of this work is to determine the fate of CNT and CNT holding products at the end of their life
cycle. In a first step, potential CNT release from CNT containing polymeric products is determined
(Chapter 2). Secondly, uptake, elimination and effects of CNT to primary consumers in the aquatic
food chain (green algae) are investigated (Chapter 3). In a third step, secondary effects of CNT in
combination with the environmental pollutant 17α-ethinylestradiol (EE2) on higher organisms
(zebrafish) are determined (Chapter 4). Finally, an overall discussion dealing with the main results of
all previous chapters is drawn at the end (Chapter 5). This study shall provide information and
Chapter 1
19
coherences of possible CNT release and effects towards the aquatic environment. An entire end of
life cycle assessment and ecotoxicological impacts on the aquatic environment is presented. To our
knowledge this is the first study dealing with the end of the life cycle beginning with product
degradation combined in one work.
In the following the aims of each chapter will be described to allow a better overview of the thesis:
Chapter 2 - Quantification of CNT material release from 14C-labelled CNT-polycarbonate composites
The aim of this chapter is to quantify the release of CNT material from polycarbonate composites
after outdoor exposure. Therefore, radiolabelled CNT (14C-CNT) with a high specific radioactivity of
1.89 MBq/mg were used since they are able to track CNT at very low amounts (ng range per sample).
Most other analytical methods are less sensitive, particle unspecific or only available for qualitative
analysis. First of all, simulation of the exposure of a CNT containing polymer composite towards
global radiation, i.e., the part of solar radiation in UV-, visual- and infra-red (IR) radiation reaching the
earth’s surface was investigated. Subsequently, several other environmental influences such as
moisture, chemicals, rapid changes in temperature as well as mechanical impacts were conducted.
While synergistic as well as antagonistic effects might play a role in the resulting effects of the
individual exposure factors considered, it was decided to study the effects individually. As a first step,
the composites were exposed to the sunlike spectrum of a xenon-arc lamp in order to induce
photochemical degradation. IR spectroscopy, X-ray photoelectron spectroscopy (XPS) and scanning
electron microscopy (SEM) were used to analyse the morphological surface changes during the
artificial irradiation step. Subsequently after irradiation, CNT composites were exposed to
environmentally relevant stress conditions, i.e., mechanical force, rapid temperature changes, water
shaking, and chemicals mimicking disposal site effluents. At each step, the resulting respective
additional release of 14C-CNT was quantified by radioactive analyses.
Chapter 1
20
Chapter 3 - Interactions of multiwalled carbon nanotubes with algal cells: quantification of
association, visualization of uptake, and measurement of alterations in the composition of cells
The aim of this second research chapter is to answer the question if CNT were bioavailable for
organisms at the basis of the aquatic food chain, i.e. green algae.
Uptake and elimination balancing of CNT in Desmodesmus subspicatus were performed after the
exposure of algae cells to a concentration of1 mg CNT/L. Therefore, the interactions of CNT and D.
subspicatus were examined using a new method of density gradient centrifugation to separate CNT
agglomerates from algae cells in the aqueous phase. Secondly, transmission electron microscopy
(TEM) micrographs were produced to obtain a closer look at the interactions in the biotest.
In a third step, alterations in the cell composition after exposure to CNT were detected by attenuated
total reflection Fourier-transform infrared spectroscopy (ATR-FTIR). This method was previously
shown to allow analysis of subcellular alterations in the chemical and biochemical composition of
microorganisms due to stress by toxic chemicals, even at subcytotoxic concentrations.
Chapter 4 - The influence of carbon nanotubes (CNT) to the bioavailability of 17α-ethinylestradiol
to zebrafish (Danio rerio)
Aim of this research chapter is to investigate the impact of CNT to the biodistribution of EE2 at
zebrafish (Danio rerio).
At first, the interactions of EE2 and MWCNT were determined to gain further information on the
sorption of EE2 to the CNT in our test systems. Therefore, filtration experiments as well as
thermogravimetrical analysis (TGA) were performed by use of 14C-labelled EE2, to determine the
amount of EE2 material aligned to the CNT material in the later fish test systems. In a second step,
uptake and elimination experiments with zebrafish Danio rerio were performed by using 14C labelled
EE2 (1 µg/L) in combination with and without 1 mg CNT/L. Additionally, to determine possible
influences of the interactions between the nanomaterial and EE2 on the effect level, the vitellogenin
(VTG) induction potential of both substances alone and in combination were determined at low,
environmentally relevant ethinylestradiol (2.2 and 6.6 ng/L) concentrations.
Chapter 1
21
Chapter 5 - Overall discussion
Finally, in this chapter an overall discussion is provided, dealing with the main conclusions from
chapter 2, 3 & 4. An entire picture of the fate of CNT holding plastic composites at the end of their
life cycle is drawn. Moreover, possible ecotoxicological consequences of released material and
potential secondary effects of CNT are discussed. The conclusion addresses further research needs
and knowledge gaps which should be investigated in the future to allow a reliable risk assessment of
CNT products.
Chapter 1
22
Chapter 2
Quantification of CNT material release from 14C-labelled
CNT-polycarbonate composites
A journal article with parts of the presented results is under preparation:
Rhiem S, Barthel A-K, Maes HM, Wachtendorf V, Meyer-Plath A, Sturm H,Schäffer A:
Quantification of CNT material release from 14C-labelled CNT-polycarbonate composites
Chapter 2
24
Chapter 2
25
2.1 Introduction
The impact of plastic waste in the environment is a problem of growing concern since the last
decades, as several studies with environmental sampling showed alerting data178-184. Due to waste
and disposal, all kinds of polymeric particles are nowadays detected in the environment. Plastic
debris was detected in several organisms such as fish or seabirds and is assumed to have enormous
health impact 196-198.
A novel branch of plastic products are polymer composites containing carbon nanotubes (CNT)157.
Due to their unique properties, i.e. high electrical and thermal conductivity, very low density,
flexibility and extremely high tensile strength, CNT transfer their properties to neat polymeric
matrices, when compounded together, as they e.g. make plastic composites less flammable, enhance
its electric conductivity and improve the mechanic stability162, 164-168. CNT composite materials are
already used in products such as tennis rackets, bicycle frames or automotive bumpers4, 16 and much
more are to come in the future as production is scaling upwards15. As polymer composites will be just
as likely disposed in the environment in their lifecycle, it can be suspected that also nanoparticles
contained in plastic composites will reach the aquatic environment in the future and may cause toxic
impact.
Several research groups tried to build models and to identify potential CNT release pathways from
polymer composites by different life cycle approaches. Most of them identified possible CNT
emission during production process, usage and especially end of life cycle disposal40, 43, 221, 222.
However, CNT release rates calculated in these models suffer from high uncertainty as no reliable
release rates are known to date, which makes the models and risk assessment vague.
First qualitative analytical studies already investigated the possible nanotube exposure during usage
of CNT composites, as for example during abrasion or sanding processes 218-220. One of them found
free CNTs as well as CNT agglomerates in the released dust 220 while others did not, or found only
composite particles with connected nanotubes216, 218, 223. The question of the released size and form is
not only of academic concern, but at least as much of importance for the estimation of the
Chapter 2
26
associated health risk. Therefore, it has to be considered in which form CNT material is released from
composites and which relevance small composite fragments with embedded or surface exposed CNT
might have. Nevertheless, it has to be mentioned that in all of these studies, no real release
quantification could be made due to missing quantitative analytical methods. Only qualitative but no
quantitative identification of CNT fibers when using electron microscopy analyses.
Waste disposal of composite materials is expected to be the most important pathway for CNT
particles to enter the environment as released from the matrix material40. For example, polymer
nanocomposite waste is estimated to pose a possible risk for CNT release in municipal solid waste
incineration plants44. In this case, CNT may be released from solid residues as airborne CNT but they
are supposed to be filtered out. Moreover, it is not clear if CNT are fully gasified during the
combustion process or if they can still be released from solid residues after the burning41, 44, 45. Own
experiments revealed that combustion of CNT containing material in a biological oxidizer - used to
quantify radioactive residues in insoluble material - at 900 °C for 4 minutes did not convert 100% of
the CNT material to carbon dioxide even in presence of an combustion catalyst (data not shown).
Disposal of CNT composite products at landfill sites or inadvertent release of CNT composites to the
environment during their production or use could lead to matrix degradation or transformation and
therefore to CNT emission40. In general, release of CNT from composite products is supposed to
occur over a long period of time as the matrix material degrades slowly 40. Release is difficult to
evaluate16 and different composites with different CNT contents and matrix material surely differ
from each other.
The polymeric matrix material can be expected to be degraded under outdoor conditions in the
environment by several degradation pathways, such as photooxidative degradation by ultraviolet
(UV) radiation from the sun or/and hydrolysis by moisture214. Water during irradiation apart from
chemical hydrolysis or physical effects such as swelling or extraction is known to stimulate potential
particle release due to wash off at the degraded composite surface218. Also other impacts may have
an influence on the degradation or transformation of the composite material, such as temperature or
chemicals224. Degradation has to be examined for each individual composite mixture, as degradation
Chapter 2
27
rates for different polymeric materials differ from each other and possibly varying stabilizer additives
or nanotube materials and respective contents are used in polymeric products214. It also has to be
taken into account that CNT may alter the degradation of the matrix material as a slower matrix
degradation was observed during first UV degradation studies218, 225. Due to uncovering of the CNT,
degraded nanocomposite surfaces may present potential hot spots for CNT release into the
environment during product usage or disposal.
Once released to the environment, CNT may have hazardous impacts on organisms as investigated by
several studies of the ecotoxicological potential of CNT. Petersen et al. (2011)39 give an overview of
the most known hazardous impacts of CNT to organisms with CNT reportedly interacting with e.g.
cells and aquatic organisms and having shown in some cases toxic impacts. Moreover, combinational
toxicity and secondary effects have to be considered as CNT are known to adsorb environmental
toxicants as bisphenol A, diuron or ethinylestradiol and thus alter their fate and biological effects in
the environment13, 28, 122, 128, 156, 226. Hence, potential sources of CNT release to the environment have
to be identified and quantified to fill the gaps in CNT distribution models for the environment with
realistic data and to perform a risk assessment for these nanomaterials.
The aim of the present study was to sensitively quantify the release of smallest quantities of CNT
from polycarbonate (PC) composites after simulated outdoor exposure. Therefore, radiolabelled CNT
(14C-CNT) with a high specific radioactivity were used since they are to track in very low amounts (ng
quantities). Such precise release quantification is not possible with other methods at the moment.
Most other analytical methods are less sensitive, particle unspecific or only available for qualitative
analysis39. The release scenario followed in this study was to simulate exposure by global radiation,
i.e., the part of solar radiation in UV, VIS and IR reaching the earth’s surface, as well as other
environmental influences such as moisture, chemicals, rapid changes in temperature as well as
mechanical impacts. While synergistic as well as antagonistic effects might play a role in the resulting
effects of the individual exposure factors considered it was decided to study the effects individually.
As a first step, the composites were exposed to the sunlike spectrum (≥300 nm) of a xenon-arc lamp
in order to induce photochemical degradation. Subsequently, CNT composites were exposed to
Chapter 2
28
environmentally relevant stress conditions, i.e., mechanical force, rapid temperature changes, water
shaking, and chemicals mimicking disposal site effluents. At each step, the resulting respective
additional release of 14C-labelled CNT was quantified by radioactive analyses. To our knowledge, our
report is the first presenting a complete mass balance for the emission of CNT from UV light
irradiated and non-irradiated composite materials during several sequential environmental
degradation conditions. For this purpose, polycarbonate composites containing 14C-CNT were
produced. Additionally to this, infra-red (IR) spectroscopy, X-ray photoelectron spectroscopy (XPS)
and scanning electron microscopy (SEM) were used to analyse the morphological surface changes
during the artificial irradiation step.
2.2 Material and Methods
2.2.1 Carbon nanotube synthesis
14C-labelled multiwalled CNT (MWCNT) were synthesized by means of catalytic chemical vapour
deposition (CCVD) using 14C-benzene (1850 MBq/mmol, Hartmann Analytic GmbH, Braunschweig,
Germany) diluted (1:10 v:v) with unlabelled benzene as precursor in H2 at 700°C in presence of cobalt
as catalyst and N2 as carrier gas. The process was carried out in cooperation with Bayer Technology
Services GmbH (BTS, Germany) using a small scale fluidized sand batch reactor. The obtained 14C-CNT
material was labelled at the carbon framework and had a high specific radioactivity, i.e.
1.89 MBq/mg CNT. After synthesis, CNT were washed with 12.5% hydrochloric acid to remove excess
catalyst, leading to a product with a C-purity of more than 95%. Characterization of the material by
means of electron microscopy showed that agglomerates had a median size of 500 µm and consisted
of single tubes with 3 to 15 walls (4 nm inner and 5 to 20 nm outer diameter) and a length of several
µm (data not shown).
Chapter 2
29
For the tests with unlabelled CNT, Baytubes® C150P (Bayer Technology Services, Leverkusen,
Germany) were used, as they were similar in shape and manufacturing process to the produced 14C-
CNT as shown by electron microscopy (data not shown).
2.2.2 Preparation of CNT-containing polycarbonate composites
10.0 mg CNT material (Baytubes® C150P or 14C-CNT prepared as described above) was transferred
into a 25 mL bottle together with 0.20 g polycarbonate pellets (Makrolon® 2805, Bayer Material
Science, Germany). Subsequently, 16.0 g Chloroform (Trichlormethane DAB9 pure, Riedel-de Haën,
Germany) was added to the vessel, as the latter acts as a good dispersant medium for CNT 171, 227 and
it also solves polycarbonate. The covered bottle was transferred to an ultrasonication bath (Bandelin
RK 100, 80 Weff, Germany) for 10 minutes. Afterwards, the cooled dispersion was sonicated for two
times 5 minutes (90% energy, “pulse” mode; 0.2 sec. pulse, 0.8 sec. pause) using an ultrasonication
tip unit (Sonoplus HD 2070, 70 W, Bandelin, Germany). For the evaluation of the dispersion quality,
transmission electron microscopy (TEM) samples were prepared by diluting several drops of the
dispersion in 5 mL chloroform and transferring a drop on a TEM grid.
Subsequently, another 0.8 g of polycarbonate pellets were added and the closed vessel was again
treated in an ultrasonication bath for 2x 10 minutes to dissolve the PC. Fresh dispersions were then
transferred to round cylindrical stainless steel forms with an inner diameter of 18 mm and a wall
height of about 3 cm. Four samples could be gained from one batch with 3.0 g CNT-PC dispersion
each. Afterwards, the forms were covered with a stainless steel cap containing a 1.0 mm hole in its
middle to allow chloroform evaporation. Samples were dried at room temperature under a fume
cupboard for about one week until chloroform was fully evaporated and solid composite samples
were left. To obtain a smooth composite surface, dried samples were pressed for about 60 seconds
in a hot pressing unit with a tool temperature of about 300°C. Kapton® polyimid-foil (HN 125 µm, Du
Pont, Germany) was placed above and beyond the sample in the mould as a high-temperature
resistive separation layer. After lifting the stamp, the form was immediately cooled down by air.
Before further treatment, the composite plates were washed in isopropyl alcohol (2-Propanol
Chapter 2
30
anhydrous, 99.5%, Sigma Aldrich, Germany) in an ultrasonication bath for 5 minutes to clean the
surfaces. After drying the composites at room temperature (20°C), they were weighed on a
microbalance (MYA 5.3Y Radwag, Poland; accuracy 1 µg). Such composite plates contained 1% by
weight CNT and were black in colour. Next to CNT containing composites, pure PC plates without CNT
were produced in the same way to serve as controls.
Composites containing 14C-CNT material for balancing the release and composites with non-labelled
CNT material (Baytubes® C150P) for surface analytical characterization were produced. Also samples
containing only PC material without CNT material were produced the same way, which were used for
IR and XPS analysis.
Chapter 2
31
2.2.3 Exposure stresses
The PC-CNT composites were subjected to different types of stress, which were serially applied to
determine their individual contribution to nanotube release. Four plates were placed in a sample
Figure 1: Picture of the irradiation chamber containing a T-quartz glass with the sample holder and collection bottle; 150W lamp is on the right hand and the connections for the water cooling system on the left hand
chamber attached to a commercially available irradiation device (Unnasol800, Unnasol, Germany) to
pre-degrade their surface by irradiation (further referred to as ‘irradiated samples’); four other
composites were directly subjected to the subsequent degradation steps (further referred to as
‘control samples’): all plates were then sequentially exposed to seven different types of
environmental influences: 1) shaking in water; 2) high temperature; 3) frost-thaw cycles; 4) soaking
in alkaline humic acid; 5) soaking in artificial acid rain; 6) soaking in artificial disposal site effluent and
7) gently wiping the surface by use of a tissue paper (details see below).
Chapter 2
32
Irradiation
For the sunlight irradiation simulation experiments, samples were placed inside a sample holder in a
self-constructed closed sample chamber (Figure 1) attached to a commercially available irradiation
device. The chamber enclosed a circular shaped sample holding unit for four identical samples that
are mounted behind circular masks of 17 mm in diameter. This sample holder was vertically placed
inside a T-shaped quarz glass tube (KF Tees, Allectra, Germany). Beyond the sample holder, a flask
was placed to collect released material. The sample holder was mounted via the T-shaped tube to
the irradiation device (type US800, UnnaSol GmbH, Germany) that contains a 150 W Xenon arc light
source (XBO 150W/1 OFR, Osram, Germany), of which the spectral distribution corresponds to that
of the sun’s global radiation). The UV-irradiance at the sample holder was about 220 W/m2 at the
E280-400nm range and had a cut-off at 300 nm (Figure 2). This is about 4 times the maximal UV
irradiance of global radiation in Central Europe, but no unrealistic wavelength are present in the
spectra.
Figure 2: Irradiation spectra at the sample holding unit in the weathering chamber (Figure 1), blue: spectra observed in the chamber (previously normalized for a better comparability); green: global sun spectra; E280-400nm = 220 W/m2
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1
260 300 340 380 420 460 500 540 580 620 660 700 740 780
E ln
orm
alis
ed
Wavelength (nm)
E 280 -400nm= 220 W/m²
Chapter 2
33
Behind the sample holding unit, a water cooling system was placed to ensure a sample temperature
of less than 50°C during irradiation. The temperature was regularly checked by use of an IR
thermometer (IR 260-8S, Voltcraft, Germany). Preliminary experiments with covered irradiated
composites showed that a temperature of 50°C for the same time had no degrading effects on the
sample surfaces (data not shown).
14C-CNT holding composite plates were irradiated until a UV radiant exposure of 1000 MJ/m2 (about
1400 h constant irradiation) was reached. Afterwards, samples were taken out of the irradiation unit,
weighed and transferred to further analysis. Samples with the non-labelled CNT material were
irradiated until 500 and 1000 MJ/m2 were reached respectively and subjected to surface analysis by
light- and electron-microscopy, XPS and IR spectroscopy.
After irradiation of 14C-CNT containing composites, the weathering chamber was cleaned with
chloroform soaked tissues, which were then analysed by means of Liquid Scintillation Counting (LSC).
Afterwards, the plates were taken with a pair of tweezers that was tapped against the edge of a Petri
dish. The dish was then cleaned with a chloroform soaked tissue as well to analyse the amount of
released material.
Water exposure
Irradiated as well as non-irradiated control samples were put in 15 mL millipore water (Milli-Q
filtered deionised water) in a closed glass vessel. The samples were shaken with an orbital shaker
(GFL 3017, GFL, Germany) for 21 days at 180 rpm, to ensure that the plates were covered with water.
During that time, the plates were six times transferred to fresh water. At those time points, the old
medium was divided in two liquid scintillation counting vials to be measured with LSC. The surfaces
of the plates were cautiously dried with a dry tissue piece during water changes to soak adherent
water from the surface which was then quantified for radioactivity. Furthermore, the glass flask was
thoroughly cleaned with chloroform soaked tissue pieces, which were also subjected to LSC.
Temperature
In a second step, samples were stored in Petri dishes at 70±1 °C in a compartment dryer to analyse
the influence of heat on CNT release from the composites. This is the temperature freely exposed
Chapter 2
34
black coloured plastic samples are usually assumed to be able to reach under sunlight exposure in
the environment. After one week, the plates were brought back to room temperature and
transferred to 15 mL water as described above. After being shaken for 1 h, the water was subjected
to LSC to measure 14C-CNT release due to heat
Frost-thaw
In a third step, samples were placed in a freezer at -22±1°C in Petri dishes and brought back to room
temperature once a day for 60 minutes to simulate a winter freeze-thaw cycle. After five days, the
plates were shaken in 15 mL water as described above for LSC measurement.
Humic acid
In a fourth approach, samples were shaken in an alkaline humic acid solution. Therefore, 25.0 mg
humic acid (Fluka, Germany) were diluted in 1 L de-ionized water representing relevant
concentrations in surface water228. The pH was set to 11.3 via 1M sodium hydroxide solution
solution. Samples were shaken in 15 mL of the solution for one week with three solution renewals
and sampling times.
Artificial acid rain
In a fifth step, samples were shaken in artificial acid rain for one week with again three sampling
points. Acid rain mixture was produced according to ISO 29664:2010b229 by a diluted mixture of
hydrochloric acid, nitric acid and sulphuric acid (all Roth, Germany) achieving a pH of 2.55.
Disposal site effluent
To simulate the influence of disposal site effluent from landfills on the CNT release from composite
material, especially organic solvents, a mixture consisting in equal amounts of nine relevant solvents
(methanol, acetone, tetrahydrofurane, iso-octane, touluene, xylene, trichloroethylene,
trichlormethane, and dichlormethane; all Riedel-de Haën, Germany) adapted from Kalbe et al.
(2002)230 was produced. The mixture was diluted to 1% in purified water because higher amounts
would surely dissolve the composites. Samples were shaken in 15 mL of this solution for one week.
The solution was refreshed and sampled three times. Analyses were performed as described for the
previous steps.
Chapter 2
35
Abrasion
Abrasion was tested by gently wiping a water soaked tissue over the sample surface to determine the
amount of CNT material that was only loosely connected to the plate after the previous degradation
treatments. The tissue was afterwards subjected to LSC.
Balancing
Finally, samples were weighed and afterwards pictured under a microscope (Chapter 2.2.4.2) before
resolving them in 40.0 g (27.0 mL) chloroform. The chloroform solution was shaken on an orbital
shaker for 30 min shaking (140 rpm). Afterwards, samples were placed in an ultrasonication bath for
30 min, to fully resolve the composite plates. Subsamples were taken from the solution and
measured with LSC for balancing.
2.2.4 Analysis methodologies
2.2.4.1 Liquid scintillation counting (LSC)
During all treatments, the release of CNT material was determined by radioanalysis. Therefore, liquid
samples or wet tissue samples were placed in 20 mL liquid scintillation counting (LSC) vials (VWR®
Scintillation Vials, 20 mL, Polyethylene, VWR, Germany). Afterwards, scintillation cocktail
(LumasafePlus™, Perkin Elmer, Germany) was added and the amount of radioactivity inside was
measured in a LSC counter (LS 5000 TD, Beckmann Instruments GmbH, Germany) after storing the
samples in a fridge (4°C) overnight.
2.2.4.2 Morphology analysis (light microscopy, SEM)
To study the surface morphology of irradiated and control samples, a Zeiss A1 AxioScope and a
KEYENCE VHX-500F light microscope were used. For a detailed topographic survey of the aged
surfaces, scanning electron microscopy (SEM) was performed with a Zeiss EVO MA 10 microscope.
The samples were coated with a 40 nm gold layer in case of electrostatic charge and electron beam
damage.
Chapter 2
36
2.2.4.3 ATR-FTIR spectroscopy
The chemical characterisation of the treated surfaces was followed by Attenuated Total Reflectance
Fourier transform Infrared (ATR- FTIR) spectroscopy. The IR spectra were recorded on a Nicolet 8700
FTIR spectrometer (nominal resolution of 4 cm-1, 64 scans accumulation) using a Germanium ATR
crystal (n=4, Θ=45°), ideal for highly absorbing or scattering samples. The spectra were detected via a
LN2-cooled MCT detector in the range of 4000 – 675 cm-1.
2.2.4.4 X-ray photoelectron spectroscopy (XPS)
X-ray photoelectron spectrometry (XPS) enables a detailed analysis of the elemental composition and
binding information on the irradiated sample surface. By measuring the energy of photoelectrons a
depth of about 6-10 nm can be achieved. Spectra were measured with a Specs Sage 150
spectrometer (SPECS GmbH, Germany) using a 200 W Al-Kα radiation. For evaluation, CasaXPS
(2.3.16) software was used. The element peaks were corrected using a Tougaard underground
correction and a peak fitting modulation was according to Beamson and Briggs (1992)231.
2.2.5 Data evaluation
Data were processed by use of Microsoft Excel® (Microsoft Office Professional 2007, USA) and
GraphPad Prism (Graph Pad Prism 6, USA). Differences between treatments were examined by use of
a Student T-test (α=0.05), after checking the data for normality with Shapiro-Wilk’s test and for
variance homogeneity with Levene’s test.
Chapter 2
37
2.3 Results
2.3.1 Composite production
TEM pictures of the CNT-PC dispersion after the mictrotip sonication step clearly showed that both
single CNT and small CNT agglomerates already connected with polycarbonate material were rather
homogeneously distributed (Figure 3). The micrographs also prove that CNT were not shortened or
damaged during the ultrasound treatment in a perceptible way contrary to what is reported in
another publication232.
Composites weighed on average 138 mg per plate before degradation; CNT containing composites
were composed of 1 w% CNT.
Figure 3: TEM micrographs of the CNT dispersion in chloroform during sample preparation showing single CNT (A), long CNT structures of up to µm range (B) and CNT agglomerates containing polycarbonate in its middle (C)
Chapter 2
38
2.3.2 Light microscopy
Irradiated PC-CNT composites showed larger cracks than not irradiated samples (Figure 4) It can be
observed that the irradiated sample had a more greyish and less reflecting surface than the control
Figure 4: Microscopically pictures of a pre-irradiated sample (A) and a control sample (B) after
sequence of degradative steps
sample. Also a dark edge could be observed at the irradiated sample where the sample holder was
placed. The irradiated sample had several holes in its surface, where originally air bubbles were
enclosed in the composite, which cracked during irradiation. A bubble which was about to burst was
zoomed in and can be identified by this knocked in region on the right hand. The surface of the
radiated sample was rougher than the control sample surface. Big scratches at the surface were
produced manually during the abrasion process.
Chapter 2
39
2.3.3 SEM pictures of the exposed composites
SEM samples of 500 and 1000 MJ/m2 irradiated composite surfaces showed CNT material to be
uncovered from the matrix material at the surface (light contrasted) (Figure 5). The polymer material
Figure 5: SEM samples of different irradiation degraded PC-CNT composites and controls with no CNT material; (A) PC+0% CNT untreated; (B) PC+0% CNT after 500 MJ/m² irradiation, (C) PC+0% CNT after 1000 MJ/m² irradiation; (D) PC+1% CNT untreated, (E) PC+1% CNT after 500 MJ/m² irradiation and (F) PC+1% CNT after 1000 MJ/m² irradiation.
A
2 µm
D
2 µm
2 µm
B
2 µm
E
2 µm
C
2 µm
F
Chapter 2
40
was degraded due to irradiation. The effect was most distinctive after 1000 MJ/m2 sun like
irradiation.
Krause et.al. (2011)171 proofed Baytubes® C150P to be good individualized in PC composites with 1
weight percent CNT material. These results correlate with our SEM observations, as we have only
have slightly changed composite recipe and dispersion methodology as they used a
microcompounder for mixing and different polycarbonate in their experiments.
2.3.4 IR-spectroscopy
The main damage according to the spectra sensitivity of polycarbonate was caused by high energy of
the ultraviolet radiation of the xenon arc lamp spectra. Wypych (2008)214 declares two types of
irradiation induced degradation mechanisms. Wavelength below 330 nm causes damage by the
Photo-Fries-reaction scenario. Radiation with λ> 330 nm, initiated defect structures by breaking the
methyl side chains in poly- bisphenol A. By using sun like irradiation between 300 nm – infra red
radiation both mechanisms were causing the polymer degradation in the current work. To follow the
oxidative process the carbonyl peak (C=O) at 1700 – 1770 cm-1 and the broad hydroxyl peak at 3600-
3200 cm-1were observed as a measure of oxidation on the polycarbonate chain, while the vibration
band of the methyl/methylene group at 2850 – 2970 cm-1 was taken as an indication of a
rearrangement of the methyl side chain in polycarbonate.
FTIR spectroscopy showed an ongoing chemical changes on the surface during the irradiation
process. The photo-oxidation of λ > 300 nm lead to the formation of different oxidation species in the
carbonyl region (1770 – 1650 cm-1) in the neat polycarbonate as well as in 1 wt%- polycarbonate
(Figure 6). The formation of the peak at 1720 cm-1 was significant. According to this, the original peak
at 1770 cm-1 describes the carbonate group in the polycarbonate structure. The degradation of neat
PC seemed accelerated, compared to the CNT-filled PC, as the peak intensity still increased between
Chapter 2
41
4000 3500 3000 2500 2000 1500 1000
0
1
2
3
4
5
6
4000 3500 3000 2500 2000 1500 1000
0
1
2
3
4
5
6
PC0%_ 000 MJ
PC0%_ 500 MJ
PC0%_1000 MJ
Extinction
Wavenumber (cm-1)
PC1%_ 000 MJ
PC1%_ 500 MJ
PC1%_1000 MJ
Figure 6: FTIR-spectra of (Top) PC without CNT (0%), untreated (black) and after irradiation doses of 500 MJ/m² and 1000 MJ/m²; (bottom) FTIR-spectra of PC+ 1% CNT, untreated (black) and after irradiation doses of 500 MJ/m² and 1000 MJ/m²; all spectra are normalised at 1016 cm-1
500 and 1000 MJ/m². In the hydroxyl region an increase of absorption band was observed after
photolytical treatment, especially in neat PC. The broad hydroxyl band was centred at 3230 cm-1.
Attention was given to the decrease of the methyl/methylene group at 2850 – 2970 cm-1 in both
samples and the shift of the methyl/methylene peaks at 2960 cm-1and 2870 cm-1in PC+1%CNT
sample. This indicated an ongoing side chain oxidation and the shift might be occurred due to
exposed CNT on the surface. Correlating to the methyl side chain reaction, changes at the bending
vibration at 1450- 1420 cm-1 could be observed (Figure 6). In both PC samples the phenyl ring
vibration at 1600 cm-1 was observed gaining in intensity, indicating the formation of low molecular
phenyl species. Hence, photo-oxidation processes of the matrix material can be observed
differencing between filled and unfilled samples.
Chapter 2
42
Figure 7: XPS graphs showing binding energy (eV) vs. intensity (cps); (A) PC 1%CNT untreated; (B)
PC 1%CNT 500 MJ/m2 irradiated; (C) PC 1%CNT 1000 MJ/m2 irradiated; (D) neat CNT material
2.3.5 XPS
To follow the state of degradation XPS analyses of PC+1% CNT samples was carried out comparative
to the SEM investigations. SEM pictures indicated a high exposition of CNT on the polymer surface
after 500 MJ/m² due to the degradation of the surrounding PC. For XPS analysis the energy changes
of carbon 1s photoelectrons, that reflect the chemical bonds, were investigated (Figure 7). For the
untreated CNT samples and the reference neat PC without CNT the position and the line shape of the
C1s peak was identical and unaffected by the presence of CNT, that appeared only as a small peak at
284 eV in the untreated CNT sample. The line shape was dominated by the PC and the binding states
in the PC molecule. The characteristic peak for the sp²-binding state of C=C bonds in CNT appeared at
284,0 eV and it overlapped with the sp³- binding state of the C-C bond in PC (284,7 eV, Figure 7)218
216. Due to the decomposition of the polymer matrix by irradiation after 500 and 1000 MJ/m², the
276 278 280 282 284 286 288 290 292 294 296 298 300
0
500
1000
1500
2000
2500
3000
3500
4000
Inte
nsity [
CP
S]
Binding Energy [eV]
PC1%_000MJ_C1s
CNT:sp²
PC:C-C
PC:C-H
PC:C-O
PC:COOH_a
PC:Oxidation peak
Background
Peak model
Area sp²-CNT: 7%
PC1%CNT untreated
B.E. CNT
276 278 280 282 284 286 288 290 292 294 296 298 300
0
500
1000
1500
2000
2500
3000
B.E. CNT
Area sp²-CNT: 67%
Inte
nsity [
CP
S]
Binding Energy [eV]
PC1%_500MJ_C1s
CNT:sp²
PC:C-C
PC:C-H
PC:C-O
PC:COOH_a
PC:Oxidation peak
Background
Peak model
PC1%CNT after 500MJ/m²
276 278 280 282 284 286 288 290 292 294 296 298 300
0
500
1000
1500
2000
2500
3000B.E.CNT
Area sp²-CNT: 76%
Inte
nsity [
CP
S]
Binding Energy [eV]
PC1%_1000MJ_C1s
CNT:sp²
PC:C-C
PC:C-H
PC:C-O
PC:COOH_a
PC:Oxidation peak
Background
Peak model
PC1%CNT after 1000MJ/m²
276 278 280 282 284 286 288 290 292 294 296 298 300
0
1000
2000
3000
4000
5000
6000
7000
8000
B.E. CNT
Area sp²-CNT: 92%
Inte
nsity [
CP
S]
Binding Energy [eV]
neat CNT_C1s
CNT:sp²
Background
Peak model
CNT:shakeup-1
CNT:shakeup-2
neat CNTC
A B
B
D
Chapter 2
43
sp³-C-fC peak of PC decreased and the CNT peak at 284 eV became dominant in the spectra. The C1s
line shape was then determined by the CNT peak (Figure 7).
2.3.6 Results of the sample degradation processes
No radioactivity was found in the weathering chamber, except for a small amount at the sample
holder. This surely originates from the fact that the samples were fixed and therefore slightly
damaged at the edges. Hence, weathering by irradiation did not result in release of detectable
amounts of CNT material.
Table 1: Release of CNT material from the irradiated and control samples during the degradation sequence, mean and standard deviation gained from 4 replicates; effl. = effluent
irradiated samples (µg) % release control samples (µg) % release
knocking 0.014±0.015 0.1±0.1
21 days water 2.611±1.648 19.0±13.7 0.117±0.101 35.7±26.8
7 days 70°C 0.067±0.042 0.5±0.4 0.004±0.004 1.4±1.4
7 days frost/thaw
0.766±0.572 5.0±3.4 0.002±0.002 1.0±1.3
7 days humic acid
1.866±0.757 12.5±4.1 0.029±0.025 7.4±3.0
7 days acid rain
2.502±0.851 16.9±5.0 0.042±0.014 13.7±9.8
7 days disposal site
effl.
2.753±2.565 18.2±15.6 0.186±0.271 33.6±30.8
tissue wiping 4.121±1.533 27.8±8.6 0.019±0.017 7.2±9.3
Total 14.700±1.857 100 0.400±0.266 100
Weight loss of the composite plate during the whole degradation process was on average
2.82±0.64 mg, equal to about 2% of the plates weight, for the irradiated samples and 2.02±2.86 mg,
about 1.4% of the plates weight, for the control samples.
Chapter 2
44
Table 1 summarize the results of the 14C-CNT material release during the degradation treatments
performed with the 1000 MJ/m2 irradiated samples and the not irradiated control samples. Only a
small amount of CNT was released at the sample holder and during tapping the samples in the Petri
dish, i.e. 0.014±0.014 µg (0.1±0.1 percent of the total amount released during all treatments; further
referred to as ‘TR’).
During the treatment in water, about 2.611±1.648 µg material (19.0±13.7% TR) was released from
the irradiated samples but only 0.117±0.101 µg (35.7±26.8 % TR) from the not irradiated control
samples. Different quantities of material loss were detected during the different sampling points for
the irradiated (photooxidized) samples and the untreated controls during 21 day of water treatment.
The water was renewed seven times during the 21 days of exposure. The highest material loss after
water storage was found for the irradiated samples at the first sampling point. For the not irradiated
samples, the highest loss was detected at the last measurement point, meaning after the longest
storage in the water. Heating (70 °C) did not cause an increased release of material, but frost-thaw
cycles showed enlarged release, i.e., about 5.0±3.4% TR at the irradiated samples compared to only
0.5±0.4% TR after 70°C temperature treatment. For the control samples release rates were only
about 1% TR after both treatments. 7 days of shaking the irradiated samples with alkaline humic acid
solution, acidic rain, and artificial disposal site effluent resulted in release of 12.5±4.1%, 16.9±4.0%,
and 18.2±15.6% TR, relative to the amounts released during all treatments, respectively. For the
control samples corresponding data were 7.4±3.0%; 13.7±9.8% and 33.4±30.8 %, respectively.
Irradiation results in the release of total 7.12±2.71 µg of 14C-CNT material compared to only
0.26±0.30 µg in not irradiated controls. By gentle wiping with a paper tissue a large amount of CNT
material was detached from the irradiated samples, i.e., about 27.8±8.6% TR, but only 7.2±9.3% TR
from controls.
After all, the release from the irradiated samples was about 14.7±1.9 µg CNT material per sample
and 0.40±0.27 µg CNT from control samples. With a radius of 0.9 cm, it can be calculated that about
0.08 µg/cm² was released from control samples during all treatments. For pre-irradiated samples, it
has to be taken into account that only one site of the composite was exposed to light and there was
Chapter 2
45
a small non-degraded frame at the edge (0.5 mm) of the plate so that the irradiated area was
2.27 cm2 per sample plate. When subtracting the amount released from control plates (0.4 µg),
6.30 µg/cm² was released from the irradiated surface. Hence, about 80 times more CNT were set free
in previously irradiated samples compared to controls that were not weathered beforehand.
Referring to the calculated start amount of 1 weight percent CNT per sample, the balancing showed
that about 1.03±0.20% of the CNT were released from radiated samples and only 0.03±0.02% from
control samples after all degradation scenarios. Recovery rates of radioactivity were 133.3±32.2 %
for the radiated and 134.9±39.8 % for the control samples. Recovery was calculated by summarizing
the radioactivity in the plates determined by LSC of the redissolved composite, and the amount of
radioactivity released after the various treatments. Recoveries above 100% might be explained by
underestimation of the starting amount of radioactivity in the plates because the CNT material could
not be perfectly distributed within the composite material although electron microscopy revealed
rather homogeneous dispersion in chloroform.
Assuming a strongly limited penetration of light in the plate and considering the amount of CNT in
the resulting irradiated volume, much higher relative release rates of CNT material would be
calculated. However, since the penetration depth of sunlight in black bodies is unknown this
calculation remains rather vague.
2.4 Discussion
In our study, we found degradation of 1 weight percent of released CNT material from PC after
sunlight irradiation when referring to the composite plates weight. Damage of CNT holding
composites by irradiation has already been described in the literature for other matrix materials such
as epoxy resin or polyoxymethylene (POM)216, 218, 233. Nguyen et al. (2011)233 reported that CNT were
uncovered from matrix material during irradiation and thus protruding CNT formed a dense network
at the weathered surface similar to other studies dealing with a different polymeric matrix material
Chapter 2
46
(i.e. elastic-polyurethane)216, 218. Wohlleben et al. (2011)223 found a layer of 3±1 µm of naked CNT
material at the composite surface after 9 month outdoor equivalent irradiation.
The light microscopic pictures showed a clear degradation of the surfaces due to light irradiation
(Figure 4). Air bubbles enclosed during plate production process were uncovered during light
irradiation and other degradative steps. Compared to the control samples the surface was less
shining at the area where the simulated sunlight hit the surface. Also surface seems more raw at the
irradiated samples which might be due to uncovered CNT material at the surfaces observed in the
SEM images (Figure 5). SEM images proved the increased uncovering of CNT material at the surface
of the composites over irradiation exposure time.
The IR measurements proofed the presence of photooxidative processes during irradiation by
changes of the FTIR spectra in neat PC and PC with 1 wt.-% CNT (Figure 6). The polymer degradation
under irradiation led to the formation of various oxidation products in both PC samples. The
degraded polymer matrix in CNT-filled PC leads to an exposure of CNT on the surface. The addition of
CNT decelerated the PC oxidation process. Nguyen et al. (2011)233, applying IR spectroscopy of the
weathered surfaces epoxy composites, also observed that CNT degrade slower compared to pure
matrix material under similar degradation conditions as in the present study. Furthermore, one study
described the addition of MWCNT to composite coatings to cause better durability towards UV
degradation and therefore to improve the product lifetime.
For the XPS measurements (Figure 7) a comparison of both irradiation exposure times (500 and 1000
MJ/m2) showed that the CNT uncovering on the surface is nearly completed after 500 MJ/m².
Accordingly to the degradation of PC via oxidative processes the peak at 288 eV raised due to
irradiation, indicating oxidative species that were build on the sample surface out of the matrix
polymer. The polycarbonate is decomposed due to irradiation and the CNT were unembedded of the
matrix and exposed on the samples surface, also observed at the SEM and IR measurements.
To our knowledge, no study reported release of CNT material from surfaces degraded by dry UV
degradation218, 233. One author suggested the formation of a CNT network to prevent the CNT tubes
from surface release233. This is coincident to our results as no released CNT material could be
Chapter 2
47
detected in the weathering chamber except some small amounts at the sample holder which may
originate from sample fixation.
We subjected the irradiated composite plates to further environmental stress scenarios because it is
likely that the uncovered CNT at the surface are more easily released217. Wohlleben et al. (2011)223
sonicated irradiated POM-CNT composites for 1 h in water and observed no free particles. The
authors suggest that the detection limit of particle detection method used may not be sufficient. In
our study we used 14C-labelled CNT material with a high specific radioactivity, which resulted in a
detection limit in the low ng CNT material range (limit of detection about 1 ng per LSC sample). This
is much lower than the detection limit of the methodologies used in other studies216, 218, 223.
Comparing irradiated CNT with the control composite much higher amounts were released from the
UV-degraded samples as from the freshly prepared control composite plates (Table 1). During the
first two water treatment steps (for 2 and 7 days) a large amount of radioactivity was released from
the irradiated composite plates (data not shown). This may be explained by wash off of the loose
uncovered material from the oxidized sample surfaces influenced by the mechanical strain due to
sample shaking. Wohlleben et al (2013)218 found a 3fold faster removal of matrix material from the
irradiated CNT composite surface when weathered in moist atmosphere compared to dry conditions.
This confirms our findings, as no radioactive material was found in the weathering chamber after
irradiation under dry conditions, but immediately afterwards during water shaking radioactive
material was released. Polycarbonate is known to be slowly hydrolyzed214. The photo-oxidized
surface contains increasing amounts of hydrophilic functional groups according to the IR spectra
(Figure 6). During water treatment for 21 days the amount of released radioactive material
decreased over time indicating that the degraded surface material had been washed off and
hydrolysis rates may accelerate. For the dark control samples emission of CNT material was very slow
but also increased with incubation time in water.
In our experiment temperature (70°C) had only a minor influence on release of CNT from irradiated
samples, however frost-thaw cycling strongly increased release rates. Temperature treatment with
70°C should not have any influence on the matrix material as PC is known to be stable up to 120°C214
Chapter 2
48
over a prolonged time period. Liu et. al (2002) 224 reported cryo- (-180°C, liquid nitrogen) and heat
treatment (70°C in water bath) of CNT polymer composites to cause differences in their flexural
strength due to the fact that matrix contraction may change the frictional force between CNT and
matrix. This may have played an important role in our experiments as well. We assume that the
temperature changes in our experiment, although not that severe, and the different expansion- and
contraction rates of the CNT and the matrix material loosen the CNT at the irradiated surface
resulting in an increased release of CNT material (Table 1). The control plates showed no increased
release of CNT material under temperature stress as all CNT fibres are still covered with matrix
material. The following degrading scenarios (alkaline humic acid, acid rain and disposal effluent) all
showed nearly the same low CNT material release rates probably because of the fact that the
amount of loose material has already been released during the previous treatments. However,
wiping the surface of the sunlight degraded plates with a tissue a large amount of radioactivity was
detached. Also, during treatment with artificial disposal site effluent treatment large fragments were
released from both irradiated and control samples. This might be due the presence of organic
solvents, although in diluted form (1 % aqueous solution), which are able to dissolve PC and destroy
the connecting matrix material ‘bridges’ of the composite plate. The main degrading factors for the
non irradiated control plates seems to be hydrolysis and abrasion. Heating and frost-thaw treatment
resulted only in very little release. The biggest release was detected during 21 day water shaking and
disposal site effluent (Table 1).
It has to be taken into account that the sequence of all degradation treatments also may work in a
synergistic or antagonistic way on the sample degradation, just like in the environment, where
plastics are definitely exposed to several different degrading conditions and chemicals in
combination or sequentially. During the individual degradation steps large standard deviations were
calculated, but after all treatments the overall released CNT amount was nearly the same for all
replicates of each treatments (Table 1).
When the matrix material is degraded and CNT material was uncovered, release of CNT from the
composite surface could be anticipated even if we were not able to analyze in which morphological
Chapter 2
49
form this release took place either as single CNT, as agglomerates or with remaining (covering)
composite material. Hirth et al. (2013)216 showed CNT to be released during irradiation in simulated
run-off water embedded in matrix material. They also found free released single CNT after UV
degradation followed by ultrasonication treatment, which represents high shear forces, but more
than 95% of the CNT material remained in the composite. The UV degrading experiments indicated a
release of mg fragments per m2 surface per year for polymers without UV stabilization. In our
experiment we quantified for the first time a release in the same range amounting to
6.3 µg CNT material/cm², i.e. 63 mg CNT material/m². Given the fact that sunlight irradiation was
responsible for the degradation of the polymer and taking all other sequential treatments into
account, we can estimate a CNT release rate of 18 mg per m² and year, regarding to the yearly dose
of sunlight energy for the wavelengths between 295 and 385 nm in Florida (285 MJ per m2 and
year)217. Wohlleben et al. (2011)223 estimated 10 µg CNT mass*cm-2*year-1 to be uncapsulated at
POM-CNT composites with about 5 weight % CNT material embedded. In the present study, we
measured the released content of CNT material and not only the uncovered amount. Still, values in
the same range were obtained. It is quite sure that an amount of CNT remained uncovered at our
composite surface after all treatments. Furthermore, we used a different polymer material and a
different embedding process which might give different results.
CNT material that was released from the composites might have exhibited an altered surface
functionalisation compared to the material originally brought to the composite plates during
production. Released oxidized CNT material may have another toxicity compared to untreated CNT
material as some studies report79, 234. Furthermore, oxidized CNT may be subject of further
degradation. Single and multiwalled CNT were repeatedly reported to be degraded by bacteria and
different enzymes63, 65, 68, 235. In the work of Zhang et al. (2012)68, functionalized (oxidized) MWCNT
were degraded to different metabolites and CO2 at environmentally relevant conditions in microbial
communities. The authors identified co-metabolism to be responsible for degradation, as an
additional carbon source was required. This might be important for CNT material released from
composites as we expect it to be partially oxidized due to degradation by UV and the production of
Chapter 2
50
reactive ozone. Surfaces with uncovered CNT material may exhibit antimicrobial activity regarding
the cytotoxic potential of CNT97, 115. Hence, detailed investigations of the possible microbial
degradation of degraded and non-degraded CNT polymer composites materials present a challenging
task for the future. Recently, the degradation half life of various CNT by extracellular enzymes, i.e.
peroxidases, has been estimated to about 80 years67.
Only a few studies dealt with the possible risks of released fragments from nanotube polymer
composites. Wohlleben et al. (2011)223 found that released particles from abrasive and sanding
treatment of different nanocomposites did not cause enhanced toxicity in cell assays. Another study
from the same workgroup examined the cytoxicity of fragments gained by sanding from
polyurethane nanocomposite that contained 3 wt% CNT. They neither found released CNT in their
tests nor induced toxicity218. Ging et al. (2014)236 also did not observe toxicity of degraded CNT
containing epoxy film fragments which were bruised and then exposed to Drosophila. In most studies
the toxicity of free CNT material was examined, whereas the possible negative effects of released
CNT polymer fragments have not been investigated. Especially, released fragments of composite
material with protruded CNT at their surface may pose possible risks for the environment. Petersen
et al. (2011)82 for example found polyethyleneimine polymer coated CNT to be more toxic to
Daphnids than pristine CNT.
The released CNT matrix material hybride particles or microparticles with possibly uncovered CNT at
their surface may be subject of further degradation processes as they exhibit a large surface area.
This may result in further release of CNT material into the environment. Furthermore, the
distribution and fate of these particles in the environment might differ from those of pure CNT
material. Microplastic fragments (several µm diameter and smaller) were already found in several
aquatic organisms or seabirds all over the world 192, 193, 201. Released micro-fragments from
nanocomposites may show similar behaviour in the environment. Polymeric particles were described
to adsorb pollutants and exhibit possible transport vectors into organisms193, 206, 211. Thus, small
degraded composite fragments with protruded uncovered CNT at their surface and a polymeric
composite material core may pose a large adsorbing surface for pollutants. CNT itself are already
Chapter 2
51
known to bind several chemicals to their surfaces28, 121, 237, 238. Therefore, released degraded
composite fragments may be able to locally concentrate environmental pollutants in aquatic
organisms. Ingested in the digestive tract of those organisms the pollutants might be resolved from
the CNT surface and therefore bioavailable for the organisms239. Schwab et al.(2013)156 reported
diuron adsorbed to nanotubes to cause a higher toxic impact to algae as the substance alone. Due to
adsorption of high amounts of diuron to CNT agglomerates, the chemical came in close vicinity to
cells of Chlorella vulgaris. Therefore, composite surfaces might also represent a possible toxicant ‘hot
spot’.
To conclude, we report for the first time release rates for CNT from PC-CNT composites with 1% CNT.
Different degradation scenarios were studied and CNT release was quantified. It could not be
investigated whether the released material concerned single, agglomerated CNT, or hybrid material
containing matrix and CNT material. All these products may have a toxic potential and should be
investigated more detailed in the future. A link has to be built between the plastic debris problem
and possible risks of particle release from polymeric nanocomposite materials. However, it was
shown that irradiation may uncover CNT material at the composite surface, which could then be
released during different environmentally relevant degrading scenarios.
Chapter 2
52
Chapter 3
Interactions of multiwalled carbon nanotubes with algal
cells: quantification of association, visualization of uptake,
and measurement of alterations in the composition of cells
Submitted in a slightly modified form as
Rhiem S, Riding MJ, Baumgartner W, Martin FL, Semple KT, Jones KC, Schäffer A, Maes HM
(2014) Interactions of multiwalled carbon nanotubes with algal cells: quantification of
association, visualization of uptake, and measurement of alterations in the composition of
cells;
submitted to Environmental Science & Technology
Chapter 3
54
Chapter 3
55
3.1 Introduction
Carbon nanotubes (CNT) are considered promising materials in nanotechnology. Due to their unique
properties, such as high electrical and thermal conductivity, very low density, flexibility and
extremely high tensile strength, CNT elicit a lot of commercial expectations in different
manufacturing sectors, for example electronic and medical applications, composite material
development, aerospace technology, and energy storage4. Investment in nanotechnology research
and development as well as nanoparticle production volumes are increasing rapidly worldwide15, 240.
Based on the assumption that this will lead to increasing CNT release in the environment, a lot of
research has been dedicated to investigating the possible (eco)toxicity of such nanomaterial39. In
contrast to the large number of reported effect studies, little data are available on the
bioaccumulation of CNT by organisms. This is consistent with the limited number of analytical
methods available to differentiate carbon-based nanoparticles from organic matrices. For this reason
special analytical techniques have to be applied to detect these materials and to measure their
concentration in biological samples2, 25, 241. Using mainly microscopic techniques, it was observed that
CNT clogged the filter apparatus and covered the carapace of daphnids74, precipitated on the gills of
rainbow trout85, or filled the gastrointestinal tract of several aquatic invertebrates73, 74, 242 after
exposure of the organisms via aqueous dispersion of CNT. It remains to be clarified whether
nanotubes are bioavailable, i.e., that they are resorbed and even accumulated in exposed organisms.
Incorporation of CNT material through absorption across external tissues and the gut epithelium has
not been reported so far39.
Since most described effects could not be linked to distribution of nanoparticles and/or agglomerates
of nanoparticles to the interior of exposed organisms, the underlying mechanisms of rarely observed
CNT toxicity remains unknown. It seems possible that physiological changes are a result of
mechanical interaction between the nanomaterial and outer tissues. For example, Smith et al. (2007)
found a single walled CNT (SWCNT) concentration-dependent rise in the ventilation rate of rainbow
trout. Furthermore, pathologies in gill tissue and vascular lesions in the brain of exposed fish were
Chapter 3
56
detected85. It is not unlikely that respiratory stress caused by gill irritation may have also provoked
the observed brain injury, due to oxygen deficiency. Adherence of aggregates on the external
surfaces and accumulation of CNT in the gut of Daphnia magna prevented the organisms from
moving through the water column, ultimately leading to mortality74. Whereas uptake of CNT in
epithelial tissues of multicellular organisms could not be shown, internalization of nanotubes in
mammalian cells was repeatedly observed86, 98, 99.
In clinical studies, CNT are investigated for their biocompatibility when used as drug delivery agents
on the one hand, and for their risk to provoke (pulmonary) toxicity on the other. There is evidence
from in vitro and in vivo experiments that CNT induce the production of reactive oxygen species
(ROS), which may elicit oxidative stress at the cellular level86, 109, 110. Subsequent indirect
consequences as lipid peroxidation causing disruption of cell membranes and DNA damage were
previously observed following exposure of CNT to different cells 52, 113-115. CNT were shown to be
present in cells of different unicellular organisms such as protozoan and bacteria held in laboratory
cultures72, 86, 243. Escherichia coli and Stylonychia mytilus, were reported to incorporate CNT from their
surrounding medium72, 88, and several studies showed unicellular algae to interact with CNT89-91.
Inhibition of algal growth following to CNT exposure was repeatedly related to light absorption by
the nanomaterial causing shading of the cells and to agglomeration and physical interactions of algae
with CNT rather than to a specific mode of toxic action89. Most effects were only observed at high
CNT concentrations (>10 mg/L), which are not assumed to be environmentally relevant. In modelling
surveys, surface water concentrations in the ng/L range were predicted. Due to agglomeration and
the high affinity of the hydrophobic CNT to sediments, the material tends to settle17, 20, 91.
The objective of the present study was to link CNT-cell interactions to CNT specific effects in algae. To
our knowledge, we quantitatively measured the CNT accumulation in unicellular algae for the first
time. Thereto, the green alga Desmodesmus subspicatus was exposed to radiolabelled CNT (14C-CNT)
to determine algal uptake and association of CNT with cells, as well as elimination and dissociation of
the material from the algae over time. CNT-cell interactions were visualized by means of transmission
electron microscopy (TEM). The results were related to alterations in the cell composition detected
Chapter 3
57
by attenuated total reflection Fourier-transform infrared spectroscopy (ATR-FTIR). The latter method
was previously shown to allow analysis of subcellular alterations in the chemical and biochemical
composition of microorganisms due to stress by toxic chemicals, even at subcytotoxic
concentrations244-246. The combined results of the present paper give new insights on the
bioavailability of CNT for D. subspicatus.
3.2 Material and Methods
3.2.1 Synthesis of unlabelled and 14C-radiolabelled carbon nanotubes
Multiwalled CNT (MWCNT) were synthesized by means of catalytic chemical vapour deposition
(CCVD) of 14C-benzene in H2 at 700°C using cobalt as catalyst247 and N2 as carrier gas. The process was
the same as described in chapter 2.2.1. The obtained 14C-CNT agglomerates were labelled at the
carbon framework and had a slightly less high specific radioactivity, i.e., 1.3 MBq/mg CNT. Moreover,
unlabelled benzene was used to obtain MWCNT for the algae tests. Characterization of the material
by means of electron microscopy showed that agglomerates had a median size of 500 μm and
consisted of single tubes with 3 to 15 walls (4 nm inner and 5 to 20 nm outer diameter) and a length
of several μm (data not shown).
3.2.2 Preparation of CNT test suspensions for exposure of Desmodesmus subspicatus
A total of 1.4 mg of MWCNT agglomerates were weighed on a microbalance (0.0001 mg readability;
Mettler-Toledo UMX2, Mettler-Toledo GmbH, Germany) and transferred to a beaker containing 0.7 L
of the suggested medium for testing the toxicity of chemicals to algae in OECD Guideline 201 (OECD,
2006248). The nanomaterial was suspended in water by means of ultrasonication with a micro tip
(Sonoplus HD 2070, Bandelin, Germany) set to a rhythm of 0.2 seconds pulse at 70 W followed by 0.8
seconds pause for 5 minutes. The procedure was repeated four times or until no more agglomerated
CNT material was macroscopically visible. This method resulted in the presence of small
agglomerates and single tubes of 0.2 to 1.0 µm mean tube length (both referred to as CNT) in the
Chapter 3
58
medium, as was visualized by TEM (Figure 8). In advance of experiments with 14C-CNT, the
concentration and homogeneity of the test dispersion were verified by measuring the amount of
radioactivity in replicate samples directly after sonication. Six times 4 mL were mixed with 16 mL of
scintillation cocktail (Insta-Gel PlusTM, Perkin Elmer, Germany) and subjected to liquid scintillation
counting (LSC; LS 5000 TD, Beckmann Instruments GmbH, Germany).
CNT dispersion (50 mL) was added to an algae suspension of the same volume in a gas wash flask, in
order to obtain an exposure concentration of 1 mg (14C-)CNT/L. Therefore, D. subspicatus cells were
harvested from an in-house stock culture, kept in Kuhl medium at 20±1°C under continuous
illumination (light intensity: 70 µE/m²s), by non-destructive centrifugation (10 min; 350 g; Labofuge
A, Heraeus Christ, Germany). The pellet was resuspended in OECD test medium. Then, the extinction
of light with a chlorophyll specific wavelength (720 nm) was measured in an aliquot of this
suspension using a spectrophotometer (UV-Vis 100-40 Spectrophotometer, Hitachi, Germany). This
surrogate parameter (E720nm) was used to derive the algal density based on dry weight (dw) and cell
number from previously plotted calibration curves (data not shown). The algal suspension was finally
diluted to 2*106 cells/mL to obtain an initial density of 106 cells/mL in the CNT-algae suspension. The
bottles were placed in a climate chamber (20±1°C) at a light intensity of 70 µE/(m² s). The test
medium was aerated using Pasteur pipettes connected to an air-blowing pump (Hailae ACO-961,
10 W, Hailea Group, Germany) to provide the algae with CO2 and to prevent their sedimentation. To
reduce evaporation, the flasks were covered with paper tissue balls.
Figure 8: Transmission electron microscopy micrographs of synthesized MWCNT dispersed in test media; left: single CNT; right: a small CNT agglomerate present in the dispersion
Chapter 3
59
3.2.3 Quantification of CNT accumulation by algae
Algae were exposed to 1 mg 14C-CNT/L for 24, 48, and 72 h. For each sampling time, four flasks were
prepared, as described above. A method to separate algae cells from CNT was developed in advance.
Filtration was no option due to the presence of CNT agglomerates of similar size as the algae cells in
suspension. Good results were however obtained using density gradient centrifugation. Therefore, a
colloidal silica suspension (Ludox® TM-40: 40% in distilled water; Sigma-Aldrich, Germany) was used,
of which two mixtures with tap water were prepared: one of 3:2 v:v to separate compact CNT
agglomerates from the algae and one of 2:3 v:v to isolate the cells from less dense agglomerates. The
density of Ludox, algae, and MWCNT agglomerates amounts to 1.3 kg/L (at 25°C), less than 1.3 kg/L,
and between 1 and 2 kg/L, respectively249, 250. Preliminary experiments with radiolabelled CNT proved
that using this method, algae can be efficiently separated from unbound CNT material in the media
without inhibition of the cell viability.
At each sampling time, the wash bottles were briefly put on a shaker, after which 10 mL of the CNT-
algae suspension was added to 50 mL of the Ludox mixture in a centrifugation flask. After 10 minutes
centrifugation at 220 g, the upper layer could be clearly distinguished from the water column below
by its green colour due to the algae at the surface. CNT were divided between the layers underneath
with the most compact agglomerates lying at the bottom of the vessel. The first layer containing the
algae was pipetted to 50 mL of a 2:3 silica:water mixture and again centrifuged for 10 minutes at
220 g. In this gradient, algae cells settled down and were separated from CNT agglomerates of lower
density present in the water layers above. The bottom layer with the algae was pipetted from the
vessel and filtered through a glass fibre filter (1.2 μm pore size; Whatman GF/C), after taking samples
for fluorometric cell concentration determination.
During the tests, algae cell numbers were determined by use of a fluorescence reader (Infinite M200,
Tecan, Männedorf, Switzerland) and the software Magellan 6.5 (2008). Samples were measured in a
transparent tissue culture test plate (96 well/flat transparent, TPP, Trasadingen, Switzerland) with
triplicate 100 µL algae samples. CNT and Ludox, both, did not influence the measurements.
Chapter 3
60
The filter was placed in an LSC vial containing 18 mL of scintillation cocktail (LumasafePlus™, Perkin
Elmer, Germany). Destruction of the algal cells was facilitated by placing the vial in a sonication bath
for 5 minutes, before the sample was finally subjected to LSC to measure the amount of radioactivity
associated with the organisms. To estimate the algae growth, spectrophotometric measurements
were performed at each sampling time in parallel setups, to which unlabelled CNT were applied.
From the fluorometric values, the algal dry weight was derived, as described above to estimate the
CNT concentration per algae dry weight.
To examine CNT elimination, algae were exposed for 72 h in a large beaker (700 mL, 1 mg 14C-CNT/L),
and transferred to uncontaminated medium, afterwards. Thereto, two times 15 mL of the algae-CNT
suspension were subjected to density gradient centrifugation in separate flasks, as explained above.
The algae layers of both vessels were again combined in 50 mL of fresh OECD medium. After final
centrifugation for 20 min at 350 g, the supernatant was carefully discarded with a Pasteur pipette
and the algae pellet was resuspended in 90 mL OECD medium to start the elimination phase. The
initial amount of radioactivity was measured with LSC in two 5 mL samples and the cell density was
determined fluorometric. Four replicates were sampled after 24, 48 and 72 h. Cell viability was
checked in the control treatment by light microscopy. No deformation or destroyed cells were found.
The change in concentration of CNT in the algae over time was calculated using a one compartment
model251. The amount of CNT associated with the algae at different time points was compared using
one-tailed Student’s T-Test.
3.2.3 Visualization of CNT uptake in cells
To investigate whether the interaction of CNT with algal cells concerns internalization and/or
attachment of CNT at the cell surface, exposed algae were fixed and dehydrated in resin for
qualitative assessment with TEM. Algae were exposed to unlabelled CNT for 48 h, exactly as
previously described. Then, aeration was stopped and the algae were allowed to settle at the bottom
of the test flask. The supernatant was discarded and the remaining concentrated algal suspension
was transferred to a crimp top vial. The cells were fixed overnight at 4°C in a sodium cacodylate
Chapter 3
61
buffer (0.05 M; pH 7.2; Sigma-Aldrich, Germany) containing 2 % glutaraldehyde (25 % in H2O; Merck
KGaA, Germany). Glutaraldehyde was subsequently removed by washing the sample three times
with sodium cacodylate buffer. Post-fixation was performed with osmium tetroxide (1% OsO4; Sigma-
Aldrich, Germany) in sodium cacodylate for 1 h at room temperature. After washing it another three
times with fresh sodium cacodylate buffer and once with distilled water, the sample was gradually
dehydrated with durcupan (Component A; Sigma-Aldrich, Germany). First, a 1:1 mixture of durcupan
and water was added. Thirty minutes later, it was exchanged by a solution of higher durcupan
content, i.e., from 70 over 90 to 100% in three successive steps, and the reaction time was gradually
increased to 40 minutes. The sample was then stored overnight at 4°C in fresh durcupan. To ensure
complete dehydration, durcupan was renewed one last time and allowed to react with the sample
for 40 min at room temperature, before the sample was buffered in LR white resin (middle viscous;
Plano, Germany). After 2 h, the resin was renewed. Thirty minutes later, the fixed and water free
algae cells, present at the bottom of the crimp top vial, were embedded in beem capsules (Plano,
Germany) by 2 days of polymerization at 60°C. Ultrathin slices of about 80 nm thickness were cut off
with a glass knife (Plano, Germany) on an Ultracut OmU-4 microtome (Reichert-Jung, Austria). An EM
10 (Zeiss, Germany) and a CM 20 (Philips, Germany) electron microscope, were used to examine the
samples at 80 and 120 keV.
3.2.4 Effects of CNT on the algal cell biochemistry
Algae were exposed using the exact same setup and exposure durations as in the uptake phase of the
bioaccumulation test). For all exposure durations, 100 mL CNT-free cultures of the same algal density
were prepared in four additional gas wash flasks and incubated under equal conditions. At sampling
time, suspensions were centrifuged at 350 g for 20 minutes. To fix the algae cells, the pellet was
resuspended in 10 mL of a 70:30 ethanol:water solution, which was refreshed twice after repeated
centrifugation for 5 minutes. The final pellet was kept in ~1 mL supernatant and the rest discarded. A
Chapter 3
62
sub-portion of the sample was transferred to a Low-E-reflective glass slide (1 cm x 1 cm) and stored
in a desiccated environment until examination.
CNT-induced alterations in the composition of algal cells were tracked by comparing the output of
attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy from control and
contaminated samples, as previously described246, 252, 253. In brief, spectra were collected from 10
locations on each slide, using a Bruker TENSOR 27 spectrometer equipped with a Helios ATR diamond
crystal (Bruker Optics Ltd., UK). Biomolecules absorb infrared (IR) with wavenumbers (1/λ) between
1800 cm-1 and 900 cm-1, termed the biochemical-cell fingerprint region. The obtained spectra were
individually cut to this range, showing 235 absorbance intensities (spectral resolution: 3.84 cm-1),
baseline corrected, and normalized to the Amide I absorption (1/λ: 1650 cm-1) using the software
OPUS 5.5 (Bruker Optik GmbH, 2005). In samples of exposed algae, it was targeted at areas
containing CNT agglomerates, to increase the opportunity that cells were in contact or associated
with CNT. In slides obtained from CNT dispersions without algae, no peaks were observed in the
biochemical-cell fingerprint region. Hence, the nanomaterial did not influence cell composition
measurements. Differences between the spectra of control and CNT-exposed cells were detected by
means of principal component analysis (PCA) and subsequent linear discriminant analysis (LDA).
These multivariate analyses were applied to the data using software programmed in MATLAB r2008a
(The MathWorks, Inc., US; http://biophotonics.lancs.ac.uk)253-255. PCA was applied as a data reduction
technique to identify 10 new variables, which were linear combinations of the original 235 variables
(wavenumbers). These so-called PCA factors accounted for more than 99% of the total variance in
the data set, and therefore, represented important biomarkers. By performing LDA on the PCA
output, the variation in the spectral data of one treatment (intra-class) was minimized, while the
inter-class variation was maximized253-255. The 40 spectra (4 replicates x 10 measurements/sample) of
one class, e.g., 24 h CNT-exposed algae, form one cluster, as was presented in score plots. To draw
these graphs, one, two, or three relevant factors were chosen as Cartesian coordinates (x, y, or z) of
the PCA-LDA factor space (1D, 2D, or 3D), in which each IR spectrum is represented by one point.
Differences between treatments were identified using the cluster vector approach253, 254, 256. In order
Chapter 3
63
to simplify the detection of CNT-induced biochemical alterations relative to the corresponding
vehicle control, the centre of the control cluster itself was moved to the origin of the PCA-LDA factor
space253. The extent of changes in the algal cell composition due to CNT exposure compared to the
corresponding control was then proportional to the deviation of the centre of the treatment cluster
away from the origin. Segregations of clusters in score plots were displayed in loadings plots or
cluster vector plots, presenting the entire IR spectra, and thus, allowing visualization of differences in
absorbance at specific wave numbers (biomarkers)253. The related functional groups of biomolecules
in algae have been previously reported257-260, and hence, dissimilarities between exposed and control
algae could be linked to CNT-induced alterations in the cell composition.
To determine the significance of treatments relative to the corresponding control, a pair-wise
comparison was carried out on the PCA-LDA scores (Student’s t-test) after checking the data for
normality (Shapiro-Wilk’s test) and variance homogeneity (Levene’s test). For the evaluation of
statistically significant fluctuations over time, Dunnett’s t-test was applied when variances were
homogeneous and Welch’s t-test with Bonferroni adjustment in case of heteroscedasticity. In all
tests, it was checked whether the classes had the same cluster mean (null hypothesis) with a
significance level (α) of 0.05253. To determine the significance of separation distance in the 1D PCA-
LDA scores plots, a one-way analysis of variance (ANOVA) was conducted on plot means, and showed
highly significant (P <0.0001) differences.
Chapter 3
64
3.3 Results
3.3.1 Accumulation of CNT by algae
CNT associated with the algae over time (Figure 9A). After 24, 48 and 72 h of exposure, the mean
algae CNT concentrations amounted to 1.30±0.42, 2.11±1.76 and 4.98±1.63 µg CNT/mg algae dry
weight, respectively. The absolute amount of radioactivity connected to the algae was higher from
time point to time point (data not shown), indicating more association of CNT with the algae than
dissociation of the material from the cells within this period. Algae growth was similar in both control
and CNT treatments with an average specific growth rate (µ) of 0.0045 h-1 (0.11 d-1; Figure 9C). This
shows that CNT did not influence algal growth during the accumulation phase. After transfer of CNT
exposed algae to clear water, a time-dependent release of CNT material from the cells was observed
(Figure 9B).
Figure 9: Concentration of CNT associated with the D. subspicatus over time after: (A) accumulation of 14C-CNT out of the water; and, (B) elimination and desorption of radioactivity after transfer of exposed algae to clear water. The associated growth curves are plotted under the corresponding graphs for both the accumulation (C) and clearance (D) phase. The whiskers indicate the standard deviation on the mean of the replicates. (dw: dry weight; k2: elimination rate constant; R²: coefficient of determination)
Chapter 3
65
The algae CNT concentration decreased from 0.85±0.12 to 0.16±0.01 µg CNT/mg algae dry weight
within three days. The algae density data deviated from the exponential growth curve, especially at
48 h (Figure 9D). This can be explained by sorption of algae to the CNT material, making the cellular
suspension inhomogeneous for fluorescence/light extinction measurements. Using light microscopy
during sampling, it was observed that CNT were attached to the algae cells (data not shown). The
fluctuations in the algal density reveal that the cells were released and the population grew during
the first day, after which they were again adsorbed to the nanomaterial (Figure 9D). This is reflected
in the algae CNT concentration (Figure 9B). The algae CNT concentration did not decrease during the
second day, resulting in a deviation of the 48 h data from the one compartment model. Using this
model, an elimination rate constant k2 of 0.018 h-1 was calculated (Figure 9B). When including the
average specific growth rate during elimination of 0.010 h-1 (Figure 9D), it can be calculated that the
true elimination constant (ke) amounted to 0.008 h-1 and that growth dilution accounted for 55% of
the decrease in the algae CNT concentrations over time261.
With k2, an uptake rate constant (k1) of 85 L*kg-1*h-1 was calculated251. Dividing k1 by k2, a
concentration factor (CF) of 4722 L/kg is obtained. It has to be mentioned that for k1 it cannot be
distinguished if CNT are taken up inside the cells or adsorbed to them. Also for k2 elimination from
cells and desorption from the algae surface could not be distinguished. It can be observed, however,
that except for one replicate, all 72 h algae CNT concentrations were higher than predicted by the
uptake model (Figure 9A). Since no significant difference between the 48 and 72 h CNT
concentrations during the accumulation phase was observed, equilibrium can be assumed, giving a
72 h CF of 4980 L/kg based on dry weight.
Chapter 3
66
3.3.2 Visualization of CNT uptake in cells
CNT like structures were found in TEM slides of CNT exposed algae, whereas no CNT were found in
slides of control algae. It was visualized that both single tubes and agglomerated CNT material
interacted with algal cells. Nanotubes were found attached to the outer cell wall (Figure 10A), the
inner cell membrane (Figure 10B), lying in the cytoplasm (Figure 10C), and piercing cells (Figure 10D)
of D. subspicatus.
Figure 10: Transmission electron microscopy photomicrographs of algae cells exposed to CNT for 48 h
Chapter 3
67
3.3.3 Effects of CNT on the algal cell biochemistry
Table 2: Tentative wavenumber assignments for Figure 11 B, D, F ordered by peak peculiarity (biomolecule association gained from several references257-260)
Sample Wavenumber (cm-1)
Tentative assigned biomolecule
24 h MWCNT
1732 Lipids & fatty acids primarily v(C=O) stretching of esters
1273 Phospholipids/ Nucleic acids
1593 Protein Amide I band mainly v(C=O) stretching
1578 Protein Amide II band mainly δ(N-H) bending and v(C-N) stretching
1196 Nucleic Acid (other phosphate-containing compounds) vas(>P=O) stretching of phosphodiesters
1674 Protein Amide I band mainly v(C=O) stretching
1539 Protein Amide II band mainly δ(N-H) bending and v(C-N) stretching
1134 Carbohydrate v(C-O-C) of Polysaccharides
48 h MWCNT
1736 Lipids & fatty acids primarily v(C=O) stretching of esters
1157 Carbohydrate v(C-O-C) of Polysaccharides
964 Carbohydrate v(C-O-C) of polysaccharides
1277 Nucleic Acid (other phosphate-containing compounds) vas(>P=O) stretching of phosphodiesters
1481 Protein Amide II band mainly δ(N-H) bending and v(C-N) stretching
1362 Protein δs(CH2) and δs(CH3) bending of methyl Carboxylic Acid vs(C-O) of COO- groups of carboxylates Lipid δs(N(CH3)3) bending of methyl
1088 Carbohydrate v(C-O-C) of polysaccharides Nucleic Acid (and other phosphate-containing compounds) vs(>P=O) stretching of phosphodiesters
1547 Protein Amide II band mainly δ(N-H) bending and v(C-N) stretching
72 h MWCNT
984 Carbohydrate v(C-O-C) of polysaccharides
1501 Protein Amide II band mainly δ(N-H) bending and v(C-N) stretching
1076 Carbohydrate v(C-O-C) of polysaccharides Nucleic Acid (and other phosphate-containing compounds) vs(>P=O) stretching of phosphodiesters
1555 Protein Amide II band mainly δ(N-H) bending and v(C-N) stretching
1632 Protein Amide I band, mainly v(C=O) stretching
1678 Protein Amide I band, mainly v(C=O) stretching
1038 Carbohydrate v(C-O-C) of polysaccharides
1207 Nucleic Acid (other phosphate-containing compounds) vas(>P=O) stretching of phosphodiesters
From the one-dimensional PCA-LDA scores plots, it can be derived that CNT treated algae differed
from the corresponding controls at all sampling times (Figure 11A, C, E). The distinction was found to
Chapter 3
68
be statistically significant at each time point (P <0.0001), but segregation of spectral points of 48 h
exposed algae from unexposed cells was most obvious (Figure 11C). The distance between the
cluster means was smallest in the 72 h setups (Figure 11E). This is reflected in the cluster vector plots
(Figure 11B, D, F), showing more pronounced wave number associated alterations in the 48 h than in
the 24 h and most clearly than in the 72 h treatment compared to the according control vector (at
zero absorbance). Highest peaks were observed at 1732 and 1736 cm-1 in the slides from 24 and 48 h
exposed algae, respectively, indicating biochemical alterations associated with lipids in CNT treated
compared to control algae (Table 2)(biomolecule association gained from several references257-259,
Figure 11: One-dimensional PCA-LDA scores plots and corresponding PCA-LDA cluster vectors plots. One-dimensional scores plots for: (A) 24 h control; (C) 48 h control; and, (E) 72 h control. Line markings in the centre of each dataset represent the mean value. PCA-LDA cluster vector plots for: (B) 24 h control; (D) 48 h control; and, (F) 72 h control; the zero vector represents the control sample; tentative wavenumber allocations are available in Table 2.
Chapter 3
69
262). Contrastingly, algae exposed to CNT for 72 h showed no difference to the control algae in the
lipid IR- region (i.e., no distinctive peak could be observed at ~1730 – 1740 cm-1). This indicates that
the
Figure 12: (A) One-dimensional scores plots for all control samples alone. One-way ANOVA conducted on 1D plot means indicates highly significant (P <0.0001) differences. (B) PCA-LDA cluster vectors plot for all control samples (C) Two-dimensional PCA-LDA scores plot for all control samples
observed alterations associated with lipids were reversible over time. Similarly, peaks were observed
in the nucleic acid region of 24 and 48 h exposed algae (1273 and 1277 cm-1), but not of 72 h treated
cells (Table 2). Highest spectral alterations in loadings plots of the 72 h treatment could be observed
at 984 cm-1, which represents polysaccharides. No peak in this region was visible in 24 h, but clearly
visible in 48 h exposed algae (at 964 cm-1, Table 2). Therefore, effects associated with the
polysaccharide composition seem to have been time dependent. In the two dimensional scores plot
(Figure 13), it is shown that most of the variance is explained by the duration of the experiment (LD1,
x-axis) in both treated and control algae. Hence, the cellular composition is influenced by growth of
the algae, the uptake of nutrients from the medium, and/or the production, storage, and release of
Chapter 3
70
photoassimilates over time. The second linear discriminant (LD2, Figure 13) shows CNT induced
deviations, i.e., between 24 and 48 h treated algae and their corresponding controls. Seventy two-h-
exposed cells are not separated from
their control by the y-axis. Again, it can be
observed that 48 h-exposed algae showed
the largest difference from untreated
algae, since there is no overlap between
the corresponding clusters, indicating
that algae incubated for two days were
most influenced by CNT (Figure 13). The
susceptibility of algae for CNT induced biochemical alterations after 48 h might be explained by the
second linear discriminant displayed on the y-axis in the two dimensional score plots presenting the
control groups only (Figure 12C). At that time, the algae show an obvious difference in protein
composition compared to 24 and 72 h cultured algae (Figure 12B). When merging the scores for
untreated algae to eliminate the time
factor in the controls, the 24 h exposed algae showed the largest distance from that control group
(Figure 14A, C). These differences are mainly explained by alterations in the composition of nucleic
acids and carbohydrates, as can be derived from the corresponding cluster vector plots (Figure 14B).
Figure 13: Two-dimensional PCA-LDA scores plot of each single control and CNT treated sample
Chapter 3
71
Figure 14: (A) One-dimensional PCA-LDA scores plot for merged control samples. One-way ANOVA conducted on 1D plot means indicates highly-significant (P <0.0001) differences; (B) PCA-LDA cluster vectors plot. The zero vector represents the control sample. (C) Two-dimensional PCA-LDA scores plot. All plots generated using merged control samples.
3.4 Discussion
We showed that dispersed multiwalled CNT interact with the alga D. subspicatus and influence the
cell composition. Exposure of algae to 14C radiolabelled CNT and separation of cells and nanomaterial
by density gradient centrifugation with a colloidal silica:water gradient allowed quantification of CNT
accumulation. The material associated with the algae over time, giving a CF of about 5000 L/kg based
on dry weight. Next to accumulation of agglomerates around the cells, single tubes were detected in
the cytosol and connected to the outer cell wall and inner cell membrane with TEM. Transfer of CNT
exposed algae to uncontaminated medium led to dissociation of the CNT material, resulting in a
reduction of the algae CNT concentration by 80% within three days. It was previously reported that
CNT could attach to and penetrate algal cells89, 90. Schwab et al. (2011) presumed that irreversible
Chapter 3
72
hydrogen bonds between components of the algal surface and CNT are formed at defects on the
nanotube surface91. During the uptake phase, initially the well-dispersed CNT may become attached
to and pierce algae cells. Over time, agglomerates are formed and a rapid increase of the amount of
CNT associated with the algae was observed. Agglomerates might both connect to the algal surface
as such or develop from CNT already bound to the cell wall. This process can be supported by
sorption of extracellular polymeric substances (EPS), excreted by and surrounding the algae, to the
CNT. EPS were previously shown to stimulate agglomeration of nanoparticles, like e.g., Ag+ and
quantum dots263-265. It is likely that the bioavailability of CNT is largely influenced by the presence of
these and other organic substances in the medium. Since radioactivity stayed attached to the cells
during density gradient centrifugation in CNT-free media, it can be stated that the CNT material is
strongly bound to the cell surface. Nevertheless, rather high amounts of CNT detached from the
algae cells in the depuration phase. This might appear during cell division and growth that was shown
to account for about 55% of the reduction in algae CNT concentration over time. It remains to be
clarified whether internalized CNT can be eliminated by the algae. The fact that CNT adsorb to algae
and that small amounts of CNT may remain associated with the cells presents a pathway of food
chain transfer of this nanomaterial.
The ability of nanotubes to enter cells has been reported for various CNT materials and a wide
variety of different cell types86, 94, 104, 115, 243, 266. Possible uptake routes are still not known in detail.
Active uptake processes like phagocytosis and endocytosis are repeatedly described86, 266 as well as
internalization during cell division72. Functionalized MWCNT-NH3+ were shown to be able to cross the
plasma membrane of human macrophage cells and get either wrapped or remain unwrapped with
the plasma membrane98. As was mentioned above, piercing of cells is reported several times116, but
this process does not necessarily involve entrance of CNT in the cells although it is proven that this is
a way for the material to reach the cytoplasm89. CNT uptake in cells was suggested to be cell type
specific86 and different routes of internalization were found for the same cell type98. It was previously
assumed that piercing of algae cells by CNT may lead to cell wall damage, and result in further toxic
effects89. This may explain the observed differences between treated and control algae detected by
Chapter 3
73
ATR-FTIR spectroscopy. For example, the polysaccharide peak might originate from alterations in the
cell wall composition (Table 2, Figure 11). Another explanation for this peak might be the increased
excretion of EPS by CNT treated algae, an effect of nanoparticle exposure observed in several
studies263-265. Mainly polysaccharides and proteins are released as products of photosynthesis, which
contributes to detoxification mechanisms263-265. Changes in the photosynthetic activity of algae due
to exposure to carbon nanotubes were previously reported90, 91. Wei et al. (2010) found
functionalized MWCNT (f-MWCNT) damaged algae cells and assumed that the direct contact
between f-MWNT and the cell surface was responsible for reduced PSII functional cross section and
oxidative stress during exponential growth90. Schwab et al. (2011) discussed alterations of the cell
wall structure by CNT and shading of the cells due to attached nanomaterial91. In relation to
alterations in the IR biomarker region of polysaccharides, differences in the composition of lipids
between exposed and untreated algae were detected in our investigations (Table 2). Changes in this
region were stronger at 24 and 48 h exposure than at 72 h, indicating that the cells adapt to the CNT
impact over time, and thus, that effects were reversible (Table 2, Figure 11B,D,E). Lipid peroxidation
as a result of the formation of reactive oxygen species (ROS) in the presence of MWCNT was
repeatedly described89, 90. Riding et al. (2012) reported CNT to induce changes in unicellular
organisms by production of ROS. The alteration of the fatty acid composition of phospholipids was
demonstrated by changes of IR signals of the cell membrane246. In other studies residual metal
catalyst impurities of CNT were shown to be responsible for ROS formation110, 267-269.
Internalized CNT (Figure 10) might have interacted with cellular components leading to peaks in the
cluster vector plots of different regions. For example, it has been previously shown that nanotubes
interact with different biomolecules, such as DNA107, 108 or proteins106. As a consequence of nanotube
exposure even DNA damage was observed52. This might explain the observed peaks in the nucleic
acid region (Table 2, Figure 11). Furthermore, CNT were reported to act as carrier for different
molecules. For instance, differently modified CNT were shown to deliver chemicals to different parts
in a cell102, 106, 108 117, 270. Wild and Jones (2009) showed MWCNT to be able to pierce plant cells and to
enable a flow of phenanthrene from the outside to the inside of the cells116. Alterations in FTIR
Chapter 3
74
spectra have been used in the past to identify effects in algal cells at different nutrient levels 260, 271-
274. It is possible that the observed effects are partially caused by a nanotube induced or hindered
flux of nutrient and EPS through the cell wall. Again, these processes might be influenced by the
agglomeration behaviour of CNT material and the growth of the algal population over time.
It can be concluded that CNT are bioavailable for the alga D. subspicatus since a low amount of single
nanotubes was incorporated in the cells. To our knowledge, we report first quantitative data for the
association and dissociation of MWCNT with algae cells. A concentration factor of about 5000 L/kg
algae dry weight was calculated. We found an impact of CNT at subcytotoxic concentrations of
1 mg CNT/L. As also reported by Schwab et al. (2011), no inhibitory effect on the growth of the algae
was observed at this CNT level25. Nevertheless, changes in their biochemical composition, based on
changes in the infrared spectra in the biomarker region, were observed. The underlying processes
have to be more profoundly investigated. Our data stress the importance of further studies dealing
with food chain transfer of CNT, since the material gets attached to or incorporated in algae cells.
These examinations are highly necessary to fully understand the impact of CNT on the aquatic
environment.
Chapter 3
75
Chapter 3
76
Chapter 4
The influence of carbon nanotubes (CNT) to the
bioavailability of 17α-ethinylestradiol to zebrafish
(Danio rerio)
A journal article with parts of the presented results is under preparation:
Rhiem S, Prodöhl L, Liebeck B, Popescu C, Schäffer A, Maes HM: The influence of carbon
nanotubes (CNT) to the bioavailability of 17α-ethinylestradiol to zebrafish (Danio rerio)
Chapter 4
78
Chapter 4
79
4.1 Introduction
Carbon nanotubes (CNT) are one of the most promising materials in nanotechnology and a possible
release to the environment can be expected as a production scale up is highly anticipated for the
future5, 16, 18, 19. Due to their high surface area and hydrophobicity, CNT exhibit good adsorbing
properties for a broad range of organic chemicals28, 154, 275-277. Hence, CNT applications for usage in
water cleaning and removal of pollutants as e.g. the xeno-estrogen 17α-ethinylestradiol (EE2) from
aquatic systems are currently under development13, 14, 122. Once released to the environment, CNT
might interact with pollutants and possibly change their ecological fate and behaviour. CNT are
known to be taken up by organisms like fish without exhibiting direct toxic effects278. On the one
hand, CNT could act as carriers for transport of higher EE2 doses into fish; on the other hand, CNT
could reduce the bioavailability of EE2 if the adsorption of the compound to the CNT surface reduces
the substance resorption by fish.
EE2 is a highly relevant xeno-estrogen, as it is used in birth control pills and therefore excreted by
women133. EE2 is not fully removed by wastewater treatment plants, and was detected in the ng/L
range in the environment136, 137, 279. EE2 has a high estrogenic potency, and is active at this
concentration144. Furthermore, it is persistent in river systems with a dissipation time of about 20-
40 days140. Higher organisms such as fish exhibit a low effective concentration for EE2. For example,
estrogenic responses were induced in juvenile rainbow trout within 14 day of exposure to EE2
concentrations between 0.85 and 1.80 ng/L144. Also zebrafish (Danio rerio) were reported to be very
sensitive towards EE2147. Exposure of D. rerio to EE2 was described to result in adverse effects on
survival, growth, reproductive functions, sex differentiation and breeding success148, 149. Furthermore,
the induction of the egg yolk precursor protein vitellogenin (VTG) was reported to be a reliable
endpoint for the estrogenic impact of EE2 in male zebrafish at concentrations in the ng/L range. A
median effect concentration (EC50) of 2.51 ng/L after 8 days of exposure was previously reported150.
As a lipophilic compound, EE2 tends to accumulate in biota (bioconcentration factor in male
zebrafish 960 L/kg wet weight152) and to adsorb to organic material. In this regard, Pan et al. (2008)122
Chapter 4
80
reported EE2 to be adsorbed to mulitwalled CNT (MWCNT) by electron donor-acceptor interactions.
The graphene surface of CNT offers hydrophobic surfaces and the potential to interact with
chemicals having π-electrons in their structure 28, 154. As was mentioned above, substances adsorbed
to CNT may be more or less bioavailable for organisms. It was for example shown that phenanthrene
adsorbed to CNT was slower mineralized by bacteria as the free compound280. The influence of CNT
on the bioavailability and bioaccumulation of pollutants was repeatedly reported 129, 130, 226. Schwab
et al. (2013)156 showed diuron adsorbed to CNT agglomerates to be more toxic to algae compared to
free diuron as they estimated a local accumulation of the pollutant near the algae surface. The same
effect might happen in fish gastrointestinal tracts. It was described that digestive fluids are able to
release adsorbed pollutants from the CNT surface132.
The aim of the present study was to investigate the impact of CNT on the biodistribution of EE2 in
zebrafish (Danio rerio). At first, the interactions of EE2 and multi-walled CNT (MWCNT) were
determined to gain information on the sorption of EE2 to CNT. Therefore, filtration experiments and
thermogravimetrical analyses (TGA) were performed by use of 14C-labelled EE2, to determine the
amount of EE2 material aligned to the CNT material in the later fish test systems. In a second step,
uptake and elimination experiments with zebrafish Danio rerio were performed by use of 14C-
labelled EE2 material (1 µg/L) in combination with and without a non toxic concentration of MWCNT
(1 mg/L)278. The concentration of 1 µg/L EE2 in the tests was necessary to enable reliable
radiodetection in the different fish compartments, although EE2 has been described to induce
endocrine effects in a lower concentration range150. In a third step, the vitellogenin (VTG) induction
potential of both substances alone and in combination were determined at low, environmentally
relevant, ethinylestradiol (2.2 and 6.6 ng/L) concentrations, to determine possible influences of the
interactions between the nanomaterial and EE2 on the effect level. The scope of this study was to
deliver data for a better understanding of potential secondary effects of carbon nanotubes due to
interaction with an organic pollutant, EE2.
Chapter 4
81
4.2 Material and Methods
4.2.1 Carbon nanotubes
Baytubes® C150P (Bayer Technology Services, Leverkusen, Germany) were used as nanotube
material for the tests. The CNT have a purity higher than 95%. The tubes have 3-15 walls, an outer
diameter of about 13-16 nm and a length up to 1-10 µm. Tubes were delivered in dry &
agglomerated form.
4.2.2 17α-ethinylestradiol
Ring labelled 14C-EE2 with a specific radioactivity of 5.54 MBq/mg EE2 was provided by Shering
Isotope Chemistry (Germany). Stock solutions were prepared in methanol (MeOH, ≥99% for
synthesis, Roth, Germany) and ethanol (EtOH, HPLC grade, Roth, Germany).
Before applied to the biotests, purity of the stock solution was checked by high pressure liquid
chromatography (radio-HPLC, 1100, Agilent/Hewlett Packard, Germany) using a Nucleosil 100-5 C18
HD reverse phase column (250 mm x 4 mm, Macherey Nagel GmbH, Germany). A liquid scintillation
flow radiodetector was attached to the HPLC unit to detect labelled substances. Analysis was
performed by use of a HPLC program developed by Maes (2011)143, using changing gradients of water
(Milli-Q filtered deionised water) and acetonitrile (HPLC grade, Carl Roth, Germany) over 40 min.
Analysis showed no sign of impurities (Figure 15).
Unlabelled 17α-ethinylestradiol (EE2; CAS no. 57-63-6; chemical purity > 97%) in powder form was
purchased from Sigma Aldrich (Germany) and dissolved in a suitable solvent similar to the labelled
substance.
Chapter 4
82
Figure 15: HPLC chromatogram of 17α-EE2 stock solution, radioactive counts per second (cps) vs. retention time (min)
4.2.3 Filtration experiments
500 µg MWCNT were added to 500 mL OECD artificial water (OECD 305)281 and dispersed by means
of ultrasonication (ultrasonication tip, Bandelin, Sonoplus HD 2070, 70 W, ‘pulse’; 0.2 sec pulse,
0.8 sec pause) for 10 min. 49,1 µL from a EE2 stock solution in EtOH (500 ng 14C-labelled EE2) were
spiked to each vessel, which were incubated at 26±1°C for 2.5, 24, 48 and 72 h. For each sampling
point, four replicates were prepared. After incubation, samples were filtered through glass fibre filter
papers (GF/C, 1.2 µm particle retention diameter, 47 mm diameter, Whatman, Germany). The filter
paper was replaced by a fresh one when half of a sample was filtered through, since it was clogged
with CNT material. Afterwards, filters and 5 mL aliquots of the filtrate were subjected to liquid
scintillation counting (LSC; LS 5000 TD, Beckmann Instruments GmbH, Germany) in plastic vials
containing 15 mL cocktail (LumasafePlus™, Perkin Elmer, Germany). Filtrate samples were taken to
determine the amount of free EE2. The radioactivity associated with the filter paper was interpreted
0,00 5,00 10,00 15,00 20,00 25,00 30,00 35,00 min
14C
0,00 5,00 10,00 15,00 20,00 25,00 30,00 35,00 min
0
50
100
150
200
250
300
350CPS
BKG1
L 001-001
L 002-002
L 003-003
L 004-004
L 005-005
L 006-006
L 007-007
L 008-008
L 009-009
L 010-010
L 011-011
L 012-012
L 013-013
L 014-014
L 015-015
Chapter 4
83
as the amount of EE2 that was adsorbed to the CNT material. It was shown on beforehand by using
radioactive CNT (14C-CNT, Chapter 3.2.1; 1.3 MBq/mg) that about 7-15% (depending on the
incubation time) passed through the filters (data not shown). To correct for CNT present in the water
phase, this amount was subtracted from the EE2 content in the filtrate. Furthermore, the amount of
free EE2 material absorbed to the filter paper could be determined by just subjecting EE2 spiked
water without CNT material on filter papers (EE2 filter value).
Further experiments with 0.1 µg EE2/L and 1.0 mg CNT and with 0.1 mg CNT and 1.0 µg EE2/L were
performed to analyze the CNT-EE2 interactions more detailed. For experiments with 0.1 µg/L EE2 and
1.0 mg CNT/L, 100 mL of the filtrate were extracted with 20 mL ethyl acetate (EtOAc; ROTISOLV®
≥99.8%, Roth, Germany) by shaking both in a 250 mL glass bottle (Shott, Duran®, Germany) for 5 min.
Afterwards, ethyl acetate supernatant was collected and evaporated in a rotary evaporator until only
about 1 mL was left, which was then subjected to LSC.
4.2.4 Thermogravimetrical analysis (TGA)
A second approach to determine the adsorption potential of EE2 to CNT in the test systems was TGA.
For this purpose, suspensions with higher CNT concentrations (5 and 10 mg CNT/L) were prepared, as
described above, and 100 & 1000 µg unlabelled EE2/L was added. TGA was namely expected to be
less sensitive compared to radio analytical methods. Furthermore, CNT suspensions (5 and 10 mg
CNT/L) without EE2 were prepared as control samples. After 24 h exposure, the whole treatments
were split to 100 mL vials and centrifuged for 15 min at 350g (Sigma 2K15, Sigma, Germany). The
obtained CNT pellet was collected in a Petri dish and dried at 80°C over night. Afterwards, the
residue was collected and transferred to a TGA unit (Perkin Elmer STA 6000, USA). The samples were
heated in a constant N2 flow to a temperature of 800°C (heating rate 5°C per minute) and the weight
loss was determined. Additionally to the combinational samples, pristine Baytubes® C150P material
and unlabelled 17α-ethinylestradiol were subjected to TGA.
Chapter 4
84
4.2.5 Experiments with zebrafish (Danio rerio)
4.2.5.1 Test organism
Zebrafish (Danio rerio) are cultivated in an in-house culture at the Institute for Environmental
Research (Bio V) at RWTH Aachen University. This culture was obtained from the Fraunhofer-Institute
for Molecular Biology and Applied Ecology in Schmallenberg (IME, Germany) from the westaquarium-
wild type stock (wildtyp/WT Westaquarium). Populations are cultivated in aquaria (100 L) at a
constant temperature of 26±1°C and a light:dark photoperiod of 16:8. The aquaria are aerated by
membrane pumps to provide fresh oxygen to the fish. The water is exchanged once a week and
regularly checked for its quality. Fish are fed twice a day, once with dry food (TetraMin®, Tetra
GmbH, Germany) and once with living food (Artemia salina nauplii) which is daily freshly prepared.
For all experiments, adult male zebrafish were used. All experiments were conducted in accordance
with the Animal Welfare Act and with permission of the federal authorities (Landesamt für Natur,
Umwelt und Verbraucherschutz NRW, Germany),registration number 84-02.04.2012.A015.
4.2.5.2 Biodistribution experiments
For biodistribution experiments, individual fish were exposed to 1.0 mg CNT/L and 1.0 µg 14C-EE2/L in
combination to determine the distribution of EE2 to different compartments of the fish in presence
of CNT. They were incubated for time periods of 6, 24, 48, 72 and 168 h. 1.0 µg/L EE2 was described
to not exert acute toxicity to zebrafish143, but the concentration is sufficient enough to enable
reliable radioanalytical detection of the substance in different compartments of the fish. For all
sampling points, five fish were exposed, each in a beaker containing 500 mL artificial medium (OECD
305)281. CNT were dispersed in the media by use of an ultrasonication tip (Bandelin, Germany,
HD2070, 70W “pulse” 0.2 sec pulse 0.8 sec pause) for 10 min. Subsequently, 49,1 µL of a 14C-EE2
stock (targeted concentration 1 µL EE2/L; stock in MeOH) was spiked to the fresh dispersions. The
suspension was mixed using a glass stick and then incubated for one day at room temperature,
before adding the test organism. From the filtration experiments, EE2 and CNT were known to
significantly combine with each other within 24 h incubation. For the 168 h incubation treatment,
Chapter 4
85
one water renewal was performed after 72 h of exposure, where the fish were transferred to freshly
prepared suspensions. 24 h before sampling, the fish were transferred to small vessels with fresh
water containing dry fish food (TetraMin®, Tetra GmbH, Germany). Animals were allowed to eat for
10 minutes before they were relocated to the test vessels. This was performed as EE2 uptake in the
bile is stimulated by food digestion143. After exposure, fish were individually anesthetized by adding a
saturated solution of ethyl4-aminobenzoate (benzocaine, Sigma–Aldrich, Germany), exsanguinated
and weighted afterwards. Afterwards they dissected (according to OECD guideline 229282) to gut,
liver, and gallbladder as these are the main fish compartments to which EE2 is distributed143. The gills
were additionally dissected as they are known to be a target organ for CNT accumulation283. The
remaining fish was collected as well. Samples were placed on previously weighed aluminium foil
pieces and dried for 48 h in a compartment dryer (80°C). Afterwards, samples were weighed and
crushed in 2 mL MeOH before they were subjected to liquid scintillation counting. The remaining fish
was crushed by an ultra-turrax (T25, Janke & Kunkel, Germany) dispersion unit. The amount of EE2
per µg tissue dry weight was calculated for each compartment and the amount of EE2 in the whole
fish body was determined. These values could be obtained from the absolute amounts of
radioactivity measured in the samples.
4.2.5.3 Accumulation & elimination experiments
In order to determine the uptake and elimination behaviour of EE2 in and from different
compartments of the zebrafish, they were exposed to EE2 alone and in presence of CNT material in
the water. For this purpose, a test setup similar to the biodistribution experiments was chosen with
treatments containing of 1.0 mg CNT/L in combination with 1.0 µg 14C-EE2/L and treatments
containing only 1.0 µg 14C-EE2/L without CNT material. All suspensions were prepared, 24 h before
fish were added. An uptake phase of 48 h was followed by a 72 h elimination phase, because after
48 h, a maximum in the EE2 uptake by the organisms was reached (data from biodistribution
experiment). Moreover, preliminary tests showed that 24 h elimination was not sufficient to detect
any EE2 eliminated from the fish. Adult male fish were used in the test with 7 replicates for each
Chapter 4
86
treatment. 24 h before transferring the fish to fresh clear artificial water for elimination, fish were
caught and fed for 10 min with dry food. The same was performed 24 h before sampling after the
elimination time. After exposure, fish were anaesthetised, killed and dissected as described above
with two slight changes: a part of the skin of the fish was additionally dissected and the gallbladder
was transferred as a whole to a LSC vial directly after dissection to prevent bile loss.
4.2.5.4 Effect test (vitellogenin induction)
In the effect tests, the level of VTG per protein mass in the fish blood was determined. For the tests,
twenty-four mature male zebrafish of similar size were collected. Six different treatments were
scheduled: solvent control (80 µL EtOH), CNT control (800 µg Baytubes® C150P) with EtOH (80 µL),
2.2 ng EE2 and 6.6 ng EE2 without and with CNT (800 µg) (further referred to as EE2 2.2, EE2 6.6, CNT
2.2, and CNT 6.6). Treatments were prepared in glass beakers containing 800 mL OECD artificial
water. For the treatments containing CNT material, samples were dispersed as described above and
spiked with EE2 or EtOH afterwards. EE2 treatments were spiked from two prepared stock solutions
containing the necessary 14C-EE2 amount in 80 µL EtOH (0.01% solvent in the test vessels). All
suspensions were mixed 24 h before adding the fish to enable CNT and EE2 interaction. Fish were
exposed semi-statically for 14 days at 26±1°C and a 16:8 light/dark photoperiod with water renewal
every 2 days (3 days at the weekend). All fresh solutions or suspensions were again mixed 24 h
before adding the fish. Animals were fed during the water renewal by transferring them to small
vessels with fresh water with dry fish food and allowing them to eat for 10 minutes before relocating
them to the fresh mixtures.
During the test, analytical sampling was performed using one replicate per treatment containing EE2
(2.2 EE2; 6.6 EE2, 2.2 EE2 CNT and 6.6 EE2 CNT). These samples were transferred to a 2 L glass flask
(Shott, Duran, Germany) after the fish was caught for each water exchange and 500 mL ethyl acetate
(ROTISOLV® ≥99.8%, Roth, Germany) was added. The bottle was shaken by hand-power for 10 min to
allow the EE2 to settle in the EtOAc phase. Subsequently, water was separated from the EtOAc by a
separation funnel and EtOAc constrained in a rotary evaporator. When a volume of only several mL
Chapter 4
87
was left, the sample was subjected to LSC. Analytical samples were also prepared directly after
spiking the solutions with EE2 and after 24 h incubation of the mixtures to determine the start
amount for the fish incubation.
After exposure, fish were individually anesthetized in a vessel by adding a saturated solution of ethyl
4-aminobenzoate (benzocaine, Sigma–Aldrich) and weighted afterwards. Then, blood of the fish was
collected by cardiac puncture (OECD 229282) with a syringe (0.3 mL; MYJector, Terumo Europe N.V.,
Belgium). To avoid coagulation of blood and protein degradation, the syringe was prefilled with
100 µL of a phosphate buffered saline (PBS) (PBS tablet solved in 1 L Milli-Q Water, Sigma Aldrich,
Germany) solution containing heparin and aprotinin (>1000 units/mL and >2 trypsin inhibitory
units/mL, respectively; both Sigma Aldrich, Germany). Blood samples were immediately transferred
to 0.5 mL test tubes (Eppendorf Safe-Lock Tubes; VWR International GmbH, Germany), which were
stored on crushed ice, and then put in a container of liquid nitrogen to freeze them. After all samples
were taken, tubes were stored at -80°C until analysis. For further analysis, samples were cautiously
unfrozen on crushed ice. The fish bodies were dried in a compartment drier (80°C) for 48 h and then
stored in a desiccated environment until they were crushed in methanol (MeOH, ≥99% for synthesis,
Roth, Germany) by an ultra-turrax unit and then measured by LSC.
Since the amount of blood taken from each fish was variable (15-40 µL) and could not be very
accurately determined, the total protein amount of the blood samples was determined by use of a
BiCinchoninic Acid (BCA) test kit (Thermo Fisher Scientific Inc.; Sigma Aldrich, Germany) to be able to
express the blood VTG content in each fish as ng VTG per µg blood protein mass, afterwards. The
BCA test combines the reduction of Cu2+ to Cu1+ in the presence of protein in an alkaline medium (the
biuret reaction). Afterwards, it is possible to detect the cuprous cation (Cu1+) by colorimetric analysis.
The chelation of two BCA molecules with one reduced copper ion results in the formation of purple
BCA/copper complexes. The subsequent shift in colour is linearly proportional to the amount of
protein in the test medium and can be measured spectrophotometrically since the complexes exhibit
absorbance of light with a wavelength of 562 nm. The test was performed in a 96-well transparent
tissue culture test plate (flat transparent, TPP, Trasadingen, Switzerland). A standard containing
Chapter 4
88
bovine serum albumin (BSA) was mixed in PBS and following concentrations were prepared for
calibration: 600, 500, 400, 300, 200, 100, 50 µg BSA/mL. Standards were filled to the cell culture test
plates in duplicate with a volume of 25 µL per well. For the blood samples 5 µL unfrozen sample was
diluted with 20 µL PBS before adding them in triplicates to the test plates. Furthermore, six wells
were only filled with PBS to determine the non-specific binding (NSB) of the test. A control setup was
additionally made by adding 5 µL of the heparin-aprotinin solution for blood sampling to 20 µL of PBS
to determine the amount of proteins which were not from the fish blood. When all vials were filled,
the test kit reagents were mixed and 200 µL of the solution given to each well. The plates were
incubated in a compartment dryer at 60°C for 15 minutes. After cooling them to room temperature,
the absorbance at 562 nm (A562nm) was measured by means of a microplate reader (Infinite®200;
Tecan Group Ltd., Germany).
For data evaluation, the average A562nm obtained from the NSB wells was first subtracted from the
mean values of the standard replicates and the corresponding control average value from the blood
samples. To plot the data, a linear regression curve was fitted to the known protein content (x) vs.
corrected A562nm data (y). From its equation (y = a*x) and the A562nm values measured in plasma
samples, and taking the dilution factor into account, the plasma protein concentration of the test fish
could be calculated.
To determine the VTG content in the fish blood samples, a special Enzyme-Linked ImmunoSorbent
Assay (ELISA) kit was purchased from BiosenseTM Laboratories (Norway)284. The test kit contains 96-
well microtiter plates pre-coated with a specific capture antibody that binds to VTG present in
samples added to the wells. After loading the sample to the wells, several washing steps are
performed to remove unbound components. Then, another VTG-specific detecting antibody bound
to an enzyme-labelled secondary antibody is applied. Thus, VTG is “sandwiched” in the middle
between both specific antibodies. The enzyme conjugated to the secondary antibody is horseradish
peroxidase. This enzyme reacts with added o-phenylenediamine (OPD), leading to a shift in the
colour of the medium from transparent to yellow. The colour intensity can be measured
Chapter 4
89
spectrophotometrically (A492nm) in a microplate reader (Infinite®200; Tecan Group Ltd., Germany) and
is proportional to the amount of VTG present in the samples
For this assay, a 1:2 dilution series of a VTG standard was prepared by diluting pure solid VTG
(supported with the assay kit) to a concentration range from 125.00 - 0.12 ng VTG/mL with a volume
of 100 μL/well, in duplicate. Also, similar to the BCA assay, two wells were filled with 100 µL of the
dilution buffer (PBS - 1% BSA) for NSB correction of the data. Three different plates were filled to
measure all samples, and therefore, three different standard curves were gained from which the best
was chosen for data evaluation.
As control fish were expected to have no or only very low amounts of VTG in the blood samples,
these blood samples were diluted 1:150, 1:500 and 1:30000. For the fish exposed to EE2 dilutions of
1:500 and 1:30000 and 1:1800000 were prepared and given to the microtiter plate in triplicates of
100 µL each. The plate was sealed and incubated for 1 h (all incubation periods of the ELISA were
performed at room temperature, ~20°C). Then, samples were removed and the plate was washed
three times with 200 μL of a PBS - 0.05% Tween-20 buffer. 100 μL of the detecting antibody in
dilution buffer was given to each well, followed by a second incubation period of 1 h. Subsequently,
the plate was again washed three times with washing buffer to remove not bound detecting
antibody. After all wells were filled with 100 μL of diluted secondary antibody, the plate was
incubated a third time and then washed five times. In a final step, 100 μL of substrate solution was
applied to each well. The plate was incubated in the dark for 30 min. Afterwards, the reaction was
then interrupted by addition of 50 μL 2M H2SO4 to all wells. Five minutes later, the A492nm was
determined in a microplate reader (Infinite®200; Tecan Group Ltd., Germany). For data evaluation,
the standard and sample values were corrected by subtracting the mean of the NSB values. A
calibration curve was fitted to the NSB corrected data of the standard series with equation y = a*xb.
To determine the working range of the test, data points deviating from the linear log-log function
were omitted until the coefficient of determination (R2) was higher than 0.99. The unknown VTG
content (x) of diluted plasma samples was derived using the average A492nm measured (y). Afterwards,
dilution factor of the samples was corrected to gain the ng VTG/mL content of the blood samples. All
Chapter 4
90
values were taken from the dilution, where all values were in the working range, except for the
control samples where different dilutions were used.
To relate the VTG content in the blood samples to the protein content, samples of the BCA and the
ELISA were brought together to calculate the µg VTG/µg protein in the samples.
4.2.6 Statistics
For all tests, differences between treatments were analyzed by use of a Student t-test (one tailed,
p=0.05), after checking the data for normality with Shapiro-Wilk’s test and for variance homogeneity
with Levene’s test. Data analyses were performed by Excel (Microsoft Office, USA) and GraphPad
(GraphPad Prism6®, GraphPad Software, Inc, USA).
4.3 Results
4.3.1 Filtration Experiments
The recovery rate of the EE2 material ranged between 90 and 110 %. The results from the filtration
experiments were drawn in Figure 16. The radioactivity (14C-EE2) on the filter paper (associated with
the CNT) vs. incubation time is displayed. Data were already corrected by the amounts of CNT which
Table 3: Mean and standard derivation (SD) of the calculated weight percent of EE2 material associated with CNT over time, data from all filtration experiments
EE2 to CNT
ratio parameter 2.5 h 24 h 48 h 72 h
0.1 mg CNT+ 1:100 Mean - 0,0122 0,0524 0,0557
1 µg EE2
SD 0,0049 0,0129 0,0230
1 mg CNT+ 1:10000 Mean 0,0011 0,0020 0,0024 -
0.1 µg EE2
SD 0,0003 0,0004 0,0003
1 mg CNT+ 1:1000 Mean 0,0074 0,0229 0,0263 0,0360
1 µg EE2 SD 0,0036 0,0035 0,0099 0,0040
passed the filter papers, determined by 14C-CNT experiments (about 10%, data not shown) and the
amount of free EE2 adsorbed to the filter papers. For all treatments, an increasing amount of EE2
Chapter 4
91
material could be observed to associate with the CNT material over time. For the treatment with the
test concentration used for the biodistribution and accumulation & elimination fish experiments
(1 mg CNT/L + 1 µg EE2/L), the percentage of the radioactivity on the filter amounted to 6.9±3.1,
20.6±2.9, 25.7±9.1, and 38.7±5.0 after 2.5, 24, 48, and 72 h of incubation, respectively. For the
treatment with the lower EE2 amount, i.e. 1 mg CNT/L + 0.1 µg EE2/L, 10.7±2.7, 19.2±3.8, and
28.0±2.7 % of the radioactivity was detected on filter paper after 2.5, 24, and 48 h incubation time,
respectively (Figure 16 and Table 3). These values are similar to the values of the samples with higher
EE2 amounts, although higher amounts were detected because of the lower relative EE2
concentration. However, in treatments with lower CNT amounts (0.1 mg CNT/L + 1 µg EE2/L, the
percentages of radioactivity on the filter paper were 1.4±0.6, 4.7±1.4, and 4.8±1.8 after 2.5, 24, and
48 h of exposure, respectively.
24 h treatment pre-incubation for all dispersions and solutions in the fish experiments was chosen as
there was no significant difference in EE2 sorption to CNT after 24 and 48 h in the 1 mg CNT/L +
Figure 16: 14C-EE2 on filter paper (associated with the CNT) vs. incubation time (h), (A) 1 mg CNT/L + 0.1 µg EE2/L; (B) 1 mg CNT/L + 1 µg EE2/L; (C) 0.1 mg CNT/L + 1 µg EE2/L; columns represent the mean & whiskers represent the standard derivation on the mean of 5 replicates; RA: radioactivity
Chapter 4
92
1 µL EE2/L treatment. 72 h incubation would have been too difficult to manage during the tests.
Thus, for the fish experiments with 1 mg CNT/L and 1 µg EE2/L, 20.6±2.9% of the EE2 was aligned to
the CNT material when fish were applied to the system.
4.3.2 TGA
The TGA diagram of the pure EE2 material shows a full burning of the substance between 300 and
500°C, due to the oxygen atoms present in the substance (Figure 17A). Most of the substance (about
90%) vanished between 300 and 450°C. Therefore, the range between 300 and 450 °C was selected
as ‘EE2 burning zone’ to detect EE2 vanishing in the samples, in which also CNT material was present.
Figure 17: TGA results: (A) pure EE2 substance, temperature plotted against heat flow rate (red) and weight % (blue); (B) TGA results CNT (Baytubes® C150P) material, temperature plotted vs. weight % ; blue curve untreated material, green curve dispersed CNT material
Chapter 4
93
The pure Baytubes® C150P material only showed a small mass loss of about 4% during heating until
800°C (Figure 17B), whereas after dispersion, incubation and drying, the same material
showed a significant mass loss of about 50% until 800°C.
Table 4: Calculated weight ratios of EE2 material associated with the CNT material gained from the TGA results, and ratios of EE2 material to CNT material in the samples; n.d.: the value for 1000 µg EE2 and 10 mg CNT/L was negative and therefore not taken for evaluation as this might come from a poorly dispersed sample; *: value determined from a filtration experiment
1 mg CNT/L +
1 µg EE2/L *
10 mg CNT/L+
100 µg EE2/L
5 mg CNT/L +
100 µg EE2/L
10 mg CNT/L +
1000 µg EE2/L
5 mg CNT/L +
1000 µg EE2/L
weight % EE2 0.023 0.163 1.103 n.d. 4.876
ratio EE2/CNT 1:1000 1:100 1:50 1:10 1:5
For data evaluation, the percentage of mass loss between 300°C and 450°C in relation to the entire
mass loss during TGA was determined from the control samples containing only CNT material. From
the treatments containing EE2, the mass loss in this region was determined as well. The amount of
CNT was set to the weight remaining in the TGA after 150°C. The percentage determined for control
samples for the “EE2 burning region” was cleared with the amount of CNT in the sample. This value
was then subtracted from the corresponding weight loss in this region to clear out the amount of
weight loss in this region that could be explained by the CNT material. Afterwards, the percentage of
EE2 mass per CNT mass was determined.
Since the value for the treatment with 1000 µg EE2/L and 10 mg CNT/L was negative, this treatment
was excluded from data evaluation. This result might originate from a poorly dispersed sample. The
amounts of EE2 bound to the MWCNT increased with larger EE2 to CNT mass ratios (Table 4). Highest
amounts were detected in the sample with 5 mg CNT/L and 1000 µg EE2/L (ratio 5:1) with 4.88% of
the introduced material was EE2 material. For comparison, a treatment from the radioactive results
with 1 mg CNT/L and 1 µg EE2/L was added (Table 4). This shows that 0.02 % of the material weight
concerned EE2. Comparing the amounts with the weight ratios displayed in Table 4, EE2 weight
percentages fitted roughly to these values.
Chapter 4
94
Figure 18: Combined data of filtration experiments and TAG experiments showing ratio of EE2/CNT mass in the samples plotted against weight % EE2 at the CNT mass with a linear fit (formula displayed at the top) of the data, all vessels incubated for 24 h
Figure 18 shows combined weight
percent data of filtration experiments
(the ones with 1 mg CNT/L after 24 h
incubation) and the TGA results plotted
together against the EE2 to CNT ratio in
the experiment. A linear fit was plotted
with a quite good fit (R2=0.98). Sorption
capacities (SP0) of about 50 g EE2/kg CNT
material could be calculate from both,
filtration and TGA experiments, for 72 h
and 24 h incubation time, respectively.
4.3.3 Fish tests
During all test with zebrafish no mortality was observed. The recovery rate of the 14C-EE2 material
ranged between 90 and 110 % for the biodistribution and the accumulation & elimination
experiments.
4.3.3.1 Biodistribution
For the biodistribution experiment, fish were incubated with 1 mg CNT/L and 1 µg EE2/L for different
time periods. For all dissected organs, i.e. the gut, gills, gallbladder, and liver and for the whole fish
an accumulation of EE2 over 168 h could be observed (Figure 19). No significant differences in the
gut EE2 concentration was observed between 24 h and 168 h exposure (ranging between 0.005 to
0.008 µg EE2/mg tissue dry weight). Only after 6 h of exposure, the value was significantly lower
(about 0.002 µg EE2/mg tissue dry weight). The digestive tract of all organisms was filled with CNT
material as fish ingested them from the water phase, after which the nanomaterial accumulated in
the guts. This was already previously shown278. The amount of EE2 directed to the gills and the liver
did also not increase over time, i.e. no significant differences in the gill and liver EE2 concentrations
Chapter 4
95
were observed between at all time points. For the gallbladder, values increased over time. Only
between 72 and 168 h, the bile EE2 concentration slightly decreased. The bile concentrations were
the highest, i.e.(0.3 µg EE2 per mg tissue dry weight (Figure 19). Most gallbladders were destroyed
during the drying process and therefore for the next experiments the full bile was directly transferred
Figure 19: µg EE2 per whole mg tissue dry
weight (dw) or fish vs. exposure time
(hours), fish exposed to 1 µg EE2/L and
1 mg CNT/L, whiskers represent standard
deviation on the mean of 5 replicates
Chapter 4
96
to a LSC vial to not sophisticate the values by possible bile loss. EE2 accumulated in the complete fish
between 6 h to 24 h to an amount of about 0.18 µg EE2 per g fish ww. This amount was then stable
for the rest of the test duration. Because of these results, a 48 h accumulation phase was chosen for
the accumulation experiment since the amounts of EE2 in all fish compartments reached its
maximum during this exposure time. From the whole body data and the fish wet weight a
bioconcentration factor (BCF) of 280 L/kg fish wet weight for male zebrafish at 1 µg EE2/L and
1 mg CNT /L could be calculated since the uptake reached a plateau after 48 h exposure.
4.3.3.2 Accumulation and elimination
The experiment dealing with the accumulation and elimination of EE2 by fish showed that no
significant difference between the treatments with CNT presence and absence in the media could be
detected (Figure 20). Hence, between all organs and the whole fish after uptake and elimination no
significant difference in EE2 concentrations could be observed. After 48 h of uptake, the EE2
concentrations in the gallbladder, skin, and liver were slightly higher when fish were exposed to EE2
alone compared to when CNT was present in the medium. For the gut and the gill samples, the
picture was the other way around. The gut and gill concentrations in fish exposed to the combination
of EE2 and CNT were slightly higher than those of fish exposed to EE2 alone, after a 48 h uptake
phase. After the uptake phase, the highest amount of EE2 was detected in the gallbladder with
26±11 and 30±11 ng EE2 per organ in presence or absence of CNT in the media, respectively. After
the elimination phase, these values decreased to 5±4 and 6±2 ng EE2 per organ in presence or
absence of CNT. The lowest concentrations of EE2 were detected in the skin samples after the uptake
phase, with 0.34±0.07 and 0. 4±0. 2 ng EE2/mg tissue dw in presence and absence of CNT,
respectively. After elimination, these values decreased to a level of 0. 12±0.09 and 0. 15±0.
15 ng EE2/mg tissue dw with CNT present or absent.
The results for the liver showed that after the uptake phase the fish of the treatment with CNT all
showed nearly the same EE2 liver concentration (2.0±0.3 ng EE2/mg tissue dw) with only a narrow
Chapter 4
97
standard deviation, whereas the fish of the treatment without CNT showed a board range in the liver
EE2 content (2.7±1.4 ng EE2/mg tissue dw). Some fish had even quite high amounts of EE2 in this
organ (up to 4.5 ng EE2/mg tissue dw). However, the mean liver CNT concentrations were not
significant different between the treatments (Figure 20). Gut EE2 concentrations of 19±14 and
15±9 ng EE2/mg tissue dw were detected for the CNT and the EE2 only treatment after uptake,
respectively. After the elimination phase, the gut contents decreased to 3±1 ng EE2/mg tissue dw for
the CNT treatment and to 4±3 ng EE2/mg tissue dw for the EE2 only treatment. As was mentioned
above, guts were black due to CNT material after the uptake phase, but seemed to be cleared during
elimination. Only low EE2 concentrations in the gills were detected after uptake, i.e. 0. 8±0.
6 ng EE2/mg tissue dw for the CNT and 0.6±0.2 ng EE2/mg tissue dw for the EE2 only treatments.
After elimination, these concentrations decreased to levels of 0.2±0.2 and
0.4±0.4 ng EE2/mg tissue dw for CNT and EE2 only treatments, respectively. The mean amount of EE2
in the whole fish was a little higher for the treatments with CNT in the media than for the EE2 only
treatments after the accumulation phase. 0.41±0.11 µg EE2/g fish wet weight and
0.38±0.10 µg EE2/g were respectively detected (Figure 20). After elimination concentrations
amounted to 0.07±0.01 µg EE2/g fish ww and 0.14±0.12 µg EE2/g fish ww for the treatments with
and without CNT material. From this data, it can be calculated that the fish of the CNT EE2 treatment
eliminated about 83% of the internalized EE2 during 72 h in fresh media, whereas organisms exposed
to pure EE2 eliminated only about 63%.
Concluding, after transferring the fish to fresh media and letting them eliminate for 72 h with one
time feeding after 48 h, all tissue EE2 concentrations significantly decreased, except for the gills.
Some fish showed higher EE2 gill concentrations in the EE2 only treatment after elimination than
after the uptake phase, but it does not concern significant differences. Furthermore, no differences
were detected comparing the CNT and EE2 only treatments. For all compartments, the
concentrations were slightly lower after the elimination phase when CNT were present in the media.
Chapter 4
98
Figure 20: Amount of EE2 per mg dry weight (and µg EE2 per whole organ at bile) at different treatments: CNT (1 mg/L) and EE2 (1 µg/L) in combination (CNT) and EE2 only (EE2, 1 µg/L) with 48 h uptake (Up) followed by 72 h elimination (Eli), ww= wet weight, the whiskers represent the standard derivation on the mean of 7 replicates
Chapter 4
99
4.3.3.3 Effect tests (VTG)
The recovery rate for 14C-EE2 in the test system, i.e. the amount in water plus the amount in the fish,
was about 91% of the amount detected in samples taken at the beginning of the experiment, when
the fish were added to the systems after 24 h incubation of the EE2 and EE2+CNT suspensions. The
observed difference to 100% might be explained by the liquid:liquid extraction technique with EtOAc.
In particular, a low recovery rate was found when CNT material was present in the system. Probably,
EE2 bound to CNT material is not good extractable by ethyl acetate.
In Figure 21, the amount of EE2 per fish dry weight
was displayed calculated for the crushed dried fish
bodies after 14 d exposure. The EE2 concentration in
fish of the 2.2 ng EE2/L treatments with CNT present
in the media was significant higher than the one with
only EE2 in the media (6.64±3.46 and
3.56±0.35 ng EE2/g fish dw). Higher fish EE2
concentrations were obtained, when fish were
treated with the higher EE2 dose (6.6 ng/L). Here, there was no significant difference between the
treatments without and with CNT (14.86±7.14 and 17.86±4.50 ng EE2/g fish dw).
The amount of VTG in the fish blood samples could reliably be detected by the ELISA method and
related to the blood sample protein content via BCA (R2 for both calibration curves >0.99). Both
showed a good correlation of the standard sample data. Data evaluation results are shown in Figure
22. The data for the control samples and the ones for the 2.2 ng EE2 only treatment were near zero
as the amount of VTG in the blood samples was under the detection limit of the test. The fish of the
treatment with only 6.6 ng EE2/L in the water phase showed an induction of the VTG production with
blood VTG levels of 0.061±0.044 µg VTG/µg blood protein. When CNT were present during
incubation, the blood VTG contents amounted to 0.007±0.004 and
0.006±0.002 µg VTG/µg blood protein for the treatments with 2.2 and 6.6 ng EE2/L, respectively.
Figure 21: ng EE2 per g fish dw after 14 days exposure to 2.2 and 6.6 ng EE2, respectively; whiskers represent the standard deviation of the mean of 5 replikates
Chapter 4
100
Figure 22: µg VTG per µg fish blood protein after 14 days exposure; control treatments and 2.2 ng EE2 treatment were near zero as the VTG contents were under the detection limit of the ELISA; whiskers represent standard deviation, all samples are means of four replicates
There was no significant
difference between both CNT
treatments, but both were
significantly different from the
6.6 ng EE2 only treatment. This
treatment and the two CNT
treatments significantly differed
from the control and the
2.2 ng EE2 only treatment with
values near zero (Figure 22).
4.4 Discussion
Our first aim was to investigate the interactions of MWCNT and EE2. Filtration experiments showed
that EE2 associated with the CNT material over time (Figure 16). When a lower amount of CNT was
present in the medium (0.1 mg/L CNT) the relative amount of EE2 sorbed to the nanomaterial was
significantly lower than at higher CNT concentrations (1 mg/L). At 0.1 mg CNT/L, sorption equilibrium
was reached within 24 h of incubation. At concentrations of 0.1 µg EE2/L and 1 mg CNT/L, less
absolute amounts of EE2 were bound to CNT than at a 10fold higher concentration of EE2
(1 mg CNT/L and 1 µg EE2/L), but relative sorbed amounts were similar. It was concluded from these
experiments that an equilibrium was quickly reached and that at a certain CNT level sorption did not
increase with lower EE2 concentrations. Pan et al (2008)122 found that aggregated forms of CNT
exhibited a larger surface area than individual tubes. Agglomerates seem to have more adsorption
sites such as the surface, groove area, and interstitial pores where chemicals can bind. The authors
assumed that π- π electron donor-acceptor interactions were responsible for the EE2 adsorption to
Chapter 4
101
MWCNT. Furthermore, they proposed rearrangement of CNT bundles or aggregates to be
responsible for irreversible hysteresis of EE2. Hence, ethinylestradiol can partially be immobilized by
CNT material in the water phase.
It has to be mentioned that the study of Pan et al. (2008)122 was performed with undispersed
MWCNT material at lower concentrations (50 µg to 8 µg/L) combined with higher EE2 concentrations
(100 - 3000 µg/L). They also shook the material for one week to reach equilibrium. The authors did
not ultrasonicate their CNT material and therefore they estimated groove and interstitial areas at the
agglomerates to be of major importance for EE2 adsorption. In the present study, higher CNT and
lower EE2 concentrations were combined for only 72 h. The surface area can also be a possible site
for adsorption in our experiments due to dispersion of the CNT, but CNT material significantly
reagglomerated over time. Therefore, it is possible that the groove and interstitial area were of great
importance as EE2 sorption to CNT material increased over time in our experiments synchronic to
CNT reagglomeration. The reported amount of EE2 aligned to CNT material was about 10 weight % of
EE2. From this value, Pan et al.(2008)122 calculated an adsorption capacity (SP0) of about 100 g/kg for
EE2 to MWCNT. The sorption rates derived from our experiments were below this value (about
50 g/kg) since we used differed CNT material and less incubation time. EE2 concentrations.
Moreover, we obtained an equilibrium between sorbed and free CNT material. The fact that at a
lower CNT concentration, a significantly lower sorption rate was obtained in our experiment shows
that CNT mass and morphology, such as agglomeration state and surface modification, have the
biggest influence on EE2 adsorption.
Kah et al. (2011)127 investigated the adsorption of PAH to CNT (Baytubes® HP) over a wide
concentration range. They report that different studies deliver different results as there are
differences in CNT material and that environmentally relevant concentrations should be focused.
Therefore, they described the adsorption coefficients for PAH at CNT at lower concentrations to
differ from those at higher concentrations127. This means that estimations of the adsorption
behaviour from our adsorption experiments may not apply for the low concentrations used in the
VTG experiment. Studies dealing with more realistic conditions may give a better insight in the
Chapter 4
102
environmental fate and behaviour of CNT and pollutants. However, PAH were described to be able to
be fully desorbed from CNT in water285, whereas for EE2 an irreversible hysteresis was found122.
The thermogravimetrical measurements showed that dispersed CNT material had a differed
appearance compared to raw ones. Ultrasonication treatments seem to modify the CNT material as
the pristine CNT material did not show significant mass loss during TGA, whereas the mass loss of
dispersed CNT material was about 50%. About 7% of the mass loss from the dispersed CNT material
(Figure 17B) might be explained by residual water in the dispersed samples, which evaporates until a
temperature of 150°C. The remaining 43% mass loss during heating up to 800°C in N2 flux might be
explained by oxidation of the nanomaterial during ultrasonication treatment and drying afterwards,
as only oxygen at the CNT material could lead to pyrolysis in the N2 atmosphere. From the literature,
ultrasound treatment of carbon nanotubes is known to damage and oxidize nanotube material. For
example, formation of carboxylic groups at their surfaces was repeatedly reported105, 232, 234, 286.
Moreover, previous studies described that carboxyated MWCNT show a higher mass loss during TGA
compared to unfunctionalized material.
We found that thermogravimetrical analysis is not an adequate method for analysis of CNT-EE2
interactions at low (environmentally relevant) concentrations since high amounts of both materials
are needed for reliable analysis. Therefore, the concentrations used in the fish experiments were
several times lower than the ones prepared for TGA analysis. Nevertheless, when comparing the
values of the TGA to the values obtained from the filtration experiments (all after 24 h incubation), a
linear correlation could be found when EE2 to CNT concentration-ratio was plotted against the CNT
aligned EE2 content (Figure 18). Sorption of EE2 to CNT was linearly correlated to the concentration
ratio in a large concentration range of both materials.
The TGA results may explain the fact that the adsorption rates found in the present study did not fit
well to the study of Pan et al. (2008)122, as they used raw, agglomerated CNT material for their
experiments. Functionalized CNT material differed from unfunctionalized tubes as differences in the
adsorption behaviour for organic chemicals119, 238 and differences at the effect69 and biodistribution
level234 were described. Moreover, surface oxidation at the nanotubes might deliver possible
Chapter 4
103
degradation points as one study described oxidized CNT to be possibly biodegraded by
environmentally relevant bacteria68. Su et al. (2013)287 found long dispersed SWCNT with estimated
polar surfaces to be more stable in dispersions than short dispersed. This SWCNT material
transferred a bigger amount of phenanthrene inside the organisms. In general, for the transferability
from in vitro results to the situation in the environment, more complex adsorption and
agglomeration behaviour processes have to be considered as other substances as e.g. dissolved
organic matter (DOM) can affect adsorption behaviour of CNT288. Hence, there is a need for testing
the combinational effects in more complex systems since CNT might be applied in waste water
treatment in the future.
Zebrafish were previously reported to ingest CNT from the water phase278. Hence, during our
experiments, digestive tracts of all organisms were observed to be filled with CNT material. As EE2
was sorbed to the CNT material, as was proved by the filtration experiments, a higher accumulation
of EE2 in the digestive tract was expected. Nevertheless, the gut EE2 concentration was not
significant higher when CNT were present in the test system (Figure 20). The mean amount of EE2 in
the gut was only slightly higher when CNT were available in the medium. EE2 was nearly completely
purged from the gut when fish were transferred to fresh media for 72 h. Gut purging of CNT material
has already been reported and the major part of ingested CNT could be eliminated by the fish in clear
water278. Similar findings were obtained for the gills and the whole fish (Figure 20). The mean EE2
concentrations in fish tissues were slightly higher when CNT were present in the medium. The
filtration experiments outlined an amount of about 20% of the EE2 material in the water phase to be
bound to the CNT after 24 h incubation (time point when the fish were inserted in the treatments).
This fits roughly to the higher amount of EE2 accumulated in the fish in presence of CNT, which was
about 10% higher than for the EE2 only treatment (Figure 20). From the data, it can be calculated
that the organisms of the CNT EE2 treatment eliminated about 83% of the EE2 within 72 h, whereas
the zebrafish exposed to EE2 only eliminated about 63%. Hence, excretion was higher in the CNT
treatments, probably while CNT are easily purged from fish in fresh media278. Therefore, it seems
that CNT bound EE2 was largely excreted from the fish. It is also possible that CNT bound free EE2
Chapter 4
104
that was excreted by the organisms in the surrounding water phase and made it not accessible for
reuptake. For all fish compartments, the EE2 concentrations were slightly lower after the elimination
phase of 72 h in clear media and feeding, when CNT had been present in the media during uptake.
This may come from the fact that CNT purged from the gut in the elimination phase bind EE2 in the
media and hinder reuptake of EE2 by the fish. From the literature, it is known that SWCNT can
influence the body burden of phenantrene in fish, as the nanomaterial influences the liver and whole
body concentrations287. However, in our experiments only slight changes in the body burden could be
observed.
From the whole body EE2 amounts and the fish wet weight a bioconcentration factor (BCF) of
280 L/kg fish wet weight for male zebrafish at 1 µg EE2/L and 1 mg CNT /L could be calculated since a
steady state was reached after 48 h (Figure 19). Maes (2011)143 calculated a BCF of 960 L/kg ww for
male zebrafish exposed to 1 µg/L EE2 only. Due to the absence of CNT, the free EE2 concentration in
the water phase was higher at her investigations. Still both values are in the same range considering
that experiments with fish always deliver scatter in the data. Further research is needed to evaluate
these values and to verify whether CNT have a negative influence on EE2 accumulation.
The applied EE2 concentration for biodistribution and uptake & elimination experiments with Danio
rerio was quite high, i.e. 1 μg EE2/L, which is about 200 times higher than the previously reported
lowest observed effect concentration (LOEC) for plasma VTG induction in zebrafish during 14 days
(i.e. 5 ng/L)151. However, working with lower EE2 contents was not an option due to the detection
limits of the radioanalytical equipment. For the VTG experiments, lower EE2 concentrations were
chosen (2.2. & 6.6 ng/L), as concentrations of 5 to 10 ng ethinylestradiol per litre were repeatedly
reported to have an impact on the VTG induction in male zebrafish151, 289, 290. Fish effective
concentration of VTG induction in juvenile rainbow trout 14 d exposure for EE2 exposure were
already reported to range between 0.85 and 1.80 ng/L144. Furthermore, EC10 and EC50 values of 0.9
and 2.5 ng EE2/L, respectively, were previously determined for the induction of VTG in male zebrafish
after water exposure for eight days150. About 50% of the blood protein content were reported to be
consisting of VTG after exposure to more than 10 ng EE2143, 151. In the present study during 14 days of
Chapter 4
105
exposure of zebrafish to 2.2 ng EE2/L no VTG induction was found, but a significant induction for 6.6
ng EE2/L. Therefore, values were in the same effective concentration range as the concentrations
summarized above and alterations may be due to the fact that we used adult zebrafish for our
experiments or different test setups. The lower concentration (2.2 ng/L) did not cause any effect on
VTG induction but the higher concentration (6.6 ng/L) resulted in 6% of the total blood protein to be
consisting of VTG. In the studies cited above, blood plasma was used instead of the complete blood
and higher EE2 concentrations than 6.6 ng/L were tested (up to 10 ng/L).
The significant induction of VTG production at the lower EE2 concentration (2.2 ng/L) in presence of
CNT was an unexpected result, as we expected the CNT to hinder EE2 to be distributed to the
estrogenic receptor. In the EE2 only treatment at 2.2 ng/L, no VTG induction was observed (Figure
22). For both treatments with nanotube material, the amount of free EE2 able to bind to the
estrogenic acceptor in the fish seemed to be the same as the VTG induction was at the same level.
The detected whole body concentration of EE2 per fish showed higher values for the treatments with
CNT material than for the treatments without (Figure 21). This was also the case in the uptake
experiments, where the EE2 concentrations in the fish were slightly higher with CNT present in the
medium. Hence, the VTG induction results of both tested EE2 concentrations in presence of CNT
material might be explained by the results of the uptake experiment. There it was outlined that the
mean amount of EE2 found in the liver, i.e. the target organ for VTG induction, was slightly lower
when CNT were present in the sample. Furthermore, no high absolute amounts were detected in the
liver. We suppose that the mass flux from the gut to this organ may be hindered by CNT presence in
the gut.
CNT agglomerates were formed in the gut, and were described above to have a higher adsorption
potential for organic chemicals than dispersed CNT material122, 127. CNT adsorbed EE2 was
transported to the gut of the fish, where a desorption may lead to the presence of free EE2, which
can then be transported to the liver and induce VTG production. A transport of CNT-EE2 complexes
to the liver was not likely, as at the higher EE2 concentration a lower VTG induction compared to the
free EE2 treatment could be observed and CNT were not found to be transported to the liver of
Chapter 4
106
Danio rerio278. Therefore, we assume that at the 2.2 ng/L EE2 + 1 mg CNT/L treatment, a process
happened in the fish digestive tract comparable to the one described in a study of Schwab et al.
(2013)156. In their study, diuron sorbed to CNT agglomerates led to a locally higher contaminant
concentration at algae surfaces and thus induced a higher impact than the same concentration of
free diuron. We think that this also happened in the fish digestive tract where high CNT amounts
accumulated278. Total EE2 concentrations in the whole fish were higher for the treatments with CNT
in the media than for the ones exposed to ethinylestradiol only. Therefore, an accumulation of EE2 in
the digestive tract of the fish due to CNT adsorption could be supposed. Based on a probable
desorption of EE2 from CNT by digestive fluids and proteins239, 287, the local concentration of free EE2
in the fish was high enough to enable resorption of free EE2 to the estrogenic receptors in the liver
and then induce VTG production in the organisms. Su et al. (2013)287 reported phenantrene to be
removed from SWCNT surfaces in the gut during the digestion process287. Desorption of chemicals
from CNT influenced by gastrointeral digestive fluids and proteins was also described in another
study239. We suppose that the same happened in our experiments. For the 2.2 ng/L EE2 in
combination with CNT treatment, this process might explain the observed VTG induction as CNT
bound EE2 might have been concentrated in the fish gut from where it was transported to the
endocrine receptors in the liver.
It is possible that in the effect tests, all EE2 sorbed to the CNT material after 24 h pre incubation since
the EE2 concentrations were much lower than in the filtration experiments. Maybe no free EE2 was
present in the samples to induce the VTG production in the fish. An unpublished experiment
performed in our institute showed however that for higher concentrations of EE2 in combination
with the same amounts of CNT material, the inhibitory effect was negligible. This might be due to the
presence of higher amounts of unbound EE2 material in the gut of the organisms290.
To determine the interactions of EE2 with CNT material at the low ng/L concentration range of
ethinylestradiol, further experiments should be performed. A prediction of CNT-pollutant
interactions at low concentrations by observations made in experiments with higher concentrations
may not work as already found in a study dealing with CNT and PAHs127. Therefore, experiments with
Chapter 4
107
environmentally relevant concentrations and conditions are highly necessary for a risk assessment of
the combinational impact potential of carbon nanotubes. In our experiment, the induction of VTG
production in male zebrafish at a concentration of 2.2 ng EE2/L with a CNT concentration of 1 mg/L
was an unexpected result, which raises further questions.
Neither the solvent nor MWCNT caused VTG production in male zebrafish after 14 d exposure (Figure
22). All effects could therefore be aligned to EE2. When CNT was present in the medium, both EE2
levels resulted in the same low induction, which was significantly lower than the induction of the
6.6 ng EE2/L treatment. Hence, CNT seemed to lower the estrogenic potential, i.e. the bioavailability
of EE2. However, altered bioavailability and bioconcentration for organic pollutants in presence of
CNT was described in several studies. Reduced bioaccumulation of perfluorchemicals by Chironimus
plumosus larvae in sediment was found by Xia et al. (2012)129. Despite of this, Shen et al. (2011)130
reported polycyclic aromatic hydrocarbons (PAHs) to be more bioavailable for C. plumosus larvae in
sediment when a MWCNT content of 2% was present in the samples. They assumed that the uptake
and concentration of PAHs aligned to CNT was responsible for these results. This may explain what
happened in the lower EE2 concentration treatment of this study, where a VTG induction was
observed, whereas for free EE2 no induction was observed. However, results of sediment
experiments may not be comparable to those for pelagial living organisms.
For freshwater fish, lead toxicity was found to be affected by CNT presence. The LC50 for the heavy
metal was decreased as toxicity increased after combinational exposure of lead (Pb) and HNO3-
MWCNT (nitric acid functionalized MWCNT) to fish (synergism)291. HNO3-MWCNT itself was proven to
be not lethal in the tested concentrations and lead itself only at higher concentrations. The authors
proposed that the found combinational effect was caused by Pb2+ adsorption to the negatively
charged surface of the CNT292. They used HNO3 oxidized MWCNT, but this may fit to our results as
our CNT were shown by TGA to be oxidized due to the sonication treatment.
Park et al. (2010)155 reported fullerenes (nC60) to hinder the bioavailability of EE2 to zebrafish that
were dietary exposed with brine shrimps. No VTG gene expression in male zebrafish was found when
they were exposed to fullerenes in combination with EE2 in brine shrimps (gastrointestinal). When
Chapter 4
108
EE2 only exposed brine shrimps were fed to the fish, this led to a significant expression. However, a
problem in their study was that they could not assure that EE2 sorbed to nC60 in the brine shrimps
was sufficient enough for VTG induction. In the present study, we also may have free CNT in the
medium as in the filtration experiments, after 24 h incubation time, not all EE2 was adsorbed to the
CNT. The same may apply for the low EE2 concentrations in the effect test, but this has to be further
investigated in the future as different scenarios are possible for low pollutant concentrations127.
Another study dealing with fullerenes and ethinylestradiol in combination described a reduced
bioavailability of EE2 for fish when associated to fullerenes, while previous aging of the suspensions
increased this effect293. In the present study, an ageing of the nanotube EE2 dispersions was
performed as well, to enable a binding between both substances. However, nanotubes and
fullerenes surely differ from each other and therefore results have to be compared with caution.
In the present study, we used concentrations of 1 mg MWCNT/L for exposure of the fish. Similar
concentrations of MWCNT were reported to result in a distribution of 0.029 pg of the nanomaterial
to 100 g fish filet278. With a predicted environmental concentration (PEC) for surface water of only
4 pg/L5, and a BAF of 73, Maes et al. (2014)278 calculated an environmental body burden of
292 pg CNT/kg fish dry weight for D. rerio278. However, MWCNT concentrations used in the present
study were much higher as expected for the environment, and therefore, experiments with relevant
concentrations in combination with the used environmentally relevant EE2 concentrations should be
performed, to identify the potential impact of both substances in combination.
Concluding, an association of EE2 and CNT material in the water phase over time was observed. This
interaction might be influenced by CNT reagglomeration and possible surface modifications which
were found by TGA. Further tests should be performed, to draw an entire picture of the interactions,
as unexpected results were observed for low EE2 concentrations. The fish experiments showed a
bioaccumulation of EE2 with a BCF of 280 L/kg fish wet weight for male zebrafish exposed to 1
µg EE2/L and 1 mg CNT /L. The main target compartments for EE2 uptake were bile, gut and liver.
Only slight differences in the uptake and elimination behaviour of fish exposed to EE2 with and
Chapter 4
109
without CNT material in the media could be outlined. Slightly more mean uptake after 48 h was
detected for the treatments with CNT material but also better elimination within 72 h. The effect test
showed a surprising result, as for both treatments with CNT material in the samples (2.2 and
6.6 ng EE2/L) the VTG induction was low but at a similar intensity. For the treatments with free EE2
material, a significantly higher induction was found for the 6.6 ng EE2/L treatment and no induction
of EE2 was found at the lower concentration. To our knowledge, for the first time combinational
effects of CNT with relevant environmental EE2 concentrations were reported, which showed new
effects at low pollutant concentrations. These results raised the question what combinational
potency CNT material may exhibit in the environment. More complex studies should be performed to
assess possible risks of CNT at environmentally relevant concentrations of pollutants.
Chapter 4
110
Chapter 5
Overall discussion
Chapter 5
112
Chapter 5
113
In the following the general main conclusions of this work will be summarized followed by a short
discussion connecting the single parts of this thesis with each other in the frame of an end life cycle
assessment. During this open gaps in research and new questions raised by this work will be outlined.
In the present work, for the first time, we reported an end of life cycle assessment of CNT holding
polymeric composite, ranging from CNT material release towards ecotoxicological effects on primary
producers (green algae) and secondary effects of CNT presence on EE2 impact to zebrafish. This work
should deliver an overview of what impacts may cause CNT release from CNT composite products
and what hazardous impact to the environment released CNT material may exhibit.
For the first time release rates for CNT from CNT holding PC composites (1% CNT) were reported to
amount to 18 mg/m2 per anno (Chapter 2). Different degradation scenarios were studied in a series
and CNT release was quantified by 14C-labelled nanotubes, identifying sun-spectra irradiation to be
the main source of composite degradation. Hence, release of CNT material from nanocomposite
products could be estimated to occur in the environment. Release of radiation degraded surfaces
during different environmental relevant scenarios was significantly higher as the release from fresh
prepared samples. Furthermore, decelerated degradation rate of CNT holding polycarbonate
compared to pure PC and uncovering of CNT at the composite surface were observed. It could not be
investigated which morphological appearance the released CNT material had as it could have been
single nanotubes, agglomerated CNT, hybrid material containing polymer and CNT or a mixture of
different forms. Nevertheless, due to the fact that these products contain CNT material, all of them
may exhibit a hazardous impact for the environment. Their structure and impacts should be
investigated more detailed in the future.
A link has to be built between the global plastic debris problem (see chapter 1.4.2) and possible risks
of particle release from polymeric nanocomposite materials. Today, polymeric material could be
observed in several environmental compartments and organisms. Thus, CNT holding polymeric
Chapter 5
114
particles, because of their estimated productions scale up, may occur in similar compartments in the
future, too. Therefore, fate of CNT holding composite particles in the environment have to be
evaluated.
In the second part of the present work, CNT were shown to be bioavailable for the alga D.
subspicatus since a low amount of single nanotubes were observed inside the cells (Chapter 3). To
our knowledge, for the first time quantitative data for the association and dissociation of MWCNT
with algae cells were reported. A concentration factor of about 5000 L/kg algae dry weight was
calculated for algae exposed to 1 mg CNT/L for 72 h. Furthermore, an impact of CNT to the algae cells
over time was observed, manifested in a differed biochemical composition of the cells. ATR-FTIR
spectroscopy with subsequent PCA-LDA statistical analysis was identified to be a promising tool to
detect nanoparticle related alterations in different biomarker regions of algae. The used CNT
concentrations of 1 mg/L was non cytotoxic as no inhibitory effect on the growth of the algae was
observed at this CNT level. In the future, the underlying processes responsible for these alterations
have to be more profoundly investigated, as they might outline possible chronic impacts on the
organisms. Our data stress the importance of further studies dealing with food chain transfer of CNT,
since the material gets attached to or incorporated in algae cells and therefore can be transferred to
higher organisms in the food chain.
In the third research part of the present study, an adsorption of the environmental relevant pollutant
EE2 to CNT was observed in the water phase over time (Chapter 4). This interaction might have be
influenced by CNT reagglomeration and possible surface modifications detected in TGA
measurements, caused by the ultrasound dispersion of our samples. Further tests should be
performed, to draw an entire picture of the interactions, as non predictable results were observed
for low EE2 concentrations. The fish experiments showed an bioaccumulation of EE2 with a BCF of
280 L/kg fish ww for male zebrafish exposed to a 24 h aged combinational approach of 1 µg EE2 and
1 mg CNT per litre. Main target compartments for EE2 uptake were bile, gut and liver. However, only
slight differences in the uptake and elimination behaviour of fish exposed to EE2 with and without
Chapter 5
115
CNT material in the media could be outlined. Slightly increased uptake of EE2 was detected in the
presence of CNT material after 48 h, but also better elimination within 72 h within clear media. The
experiment, aiming at the VTG induction, showed a surprising results as for both EE2 treatments (2.2
and 6.6 ng EE2/L), the VTG induction was low but at a similar intensity, when CNT material was
present in the media. Despite to this, free EE2 material induced a significant higher level of VTG
production at 6.6 ng EE2/L, but no induction could be observed at the lower EE2 concentration.
To our knowledge, for the first time combinational effects of CNT with relevant environmental EE2
concentrations were reported, which showed novel effects at low pollutant concentrations. These
results raised the question what combinational potency CNT material may exhibit in the
environment. More complex studies should be performed to assess possible risks of CNT at
environmentally relevant concentrations of pollutants.
In this last section of this chapter, an assessment of these conclusions in the range of a end of life
cycle assessment will be drawn:
Release studies estimated CNT to be released from products primary to wastewater treatment
facilities and landfills during waste disposal16, 17, 43. However, these studies exhibit large uncertainties
especially for the end of the life cycle43. It is not clear what happens with CNT holding products
during disposal, especially for CNT embedded in composites as they may persist combusting
processes during incineration and could thus be deposited afterwards41, 44, 45. For carbon nanotubes,
first modelled predicted environmental concentrations for surface water amounted to
0.0005 µg CNT/L17 and sediment concentrations were estimated to be in µg/kg dry weight range17, 42.
In the present study, a release of 18 mg CNT/m2*year from 1% CNT holding PC composite material
was detected. About 1% of the CNT material embedded in our composite was released during
different degradative scenarios. These proofed that several impacts degrade a CNT holding polymer
product in the environment and thus could lead to CNT uncovering at its surface and also to release
Chapter 5
116
of nanomaterial. Degradation of multiwalled carbon nanotubes by enzymes observed for co-
metabolism, only for functionalized CNT or presence of defect structures as start points for the
degradation or transformation and over a quiet long period of time (about 80 years half time)63-65, 67,
68, 235. It should be investigated, whether CNT can be degraded at composite surfaces, too. However,
due to the fact that composite products with embedded CNT are durable products potential CNT
release may occur over a long period of time40.
Up to now, only released pure CNT material was considered to cause possible hazardous impact to
the environment. In the present work, a link between the global plastic debris problem (chapter
1.4.2) and the fate of CNT holding polymeric products in the environment was build. Plastic particles
can be detected in several environmental compartments and organisms and are known to cause
hazardous effects. Particles are observed to act as pollutant shuttles to organisms, to leech their
additives and to stuff the digestive tract of higher organisms178, 179, 181, 204, 211, 294. This presence of
polymeric particles all around the globe might be a possible hint to what would happen after a
production scale up of polymeric nanocomposite products. A possible release of these products to
the environment during waste disposal or failure recycling may surely lead to a global distribution.
Subsequently, the polymer matrix will be degraded by several environmental impacts and the
products are going to decompose to smaller pieces, which could be ingested by organisms.
Furthermore, nanotubes will be uncovered at the surface which could also cause additionally toxic
impacts (e.g. antibacterial78, 88) as a locally high concentration of nanomaterial is present on degraded
composite surfaces. Ingested `hedgehog` like particles with a polymeric core and protruding
uncovered CNT material at their surfaces can cause several hazardous impacts. Unicellular organisms
may attach to uncovered CNT material which may cause an immobilization of these organisms.
Furthermore, CNT can cause cytotoxic impacts as they could damage the cell walls 78, 88, 101, 104, shade
photosynthetic active organisms89, 91 and induce oxidative stress89, 91. In the present research
influences of CNT to the green algae Desmodesmus subspicatus were discovered (Chapter 3). The
CNT were found to associate with the algae over time and moreover a small amount was observed
Chapter 5
117
inside the cells. Exposure of algae to CNT induced several changes in different biomarker regions of
the cells which should be investigated more detailed in the future, as they could be hints for chronic
effects of CNT exposure towards algae. D. subspicatus can be influenced by the presence of
nanotubes released from CNT holding composite products or also possibly by CNT uncovered at the
composite surfaces when it may get in touch with them. The used concentration of 1 mg/L in the
algae tests were much higher than estimated values at the environment, but one study also observed
changes in the benthic community at environmentally relevant CNT concentrations118. However, the
need for more research at realistic environmental concentrations of CNT and polymeric CNT
composite fragments is high. Possible risks of these materials which cannot be extrapolated from in
vitro studies with higher CNT concentrations or optimized conditions should be evaluated.
Another possible toxic pathway for released CNT composite fragments is combinational toxicity. CNT
were known to act as good adsorbents for a board range of chemicals (Chapter 1.2.6) and similar
properties were also observed for plastic particles (Chapter 1.4.2). Hence, a combination of both
materials present at the surface of degraded CNT composite products may lead to an increased
adsorption of pollutants from the aqueous phase13, 14, 122, 295. After adsorption, a transport of the
loaded nanocomposite particles to organisms is possible, where the pollutants may become
bioavailable239, 287. The fact that CNT are able to influence the bioavailability of pollutants to
organisms is already described elsewhere128-131, 226, 280. In some studies a reduced, but in others a
increased bioavailability of pollutants in presence of CNT material e.g. in sediment systems was
described. Furthermore, a study of Schwab et al. (2013)156 showed diuron adsorbed to CNT
agglomerates to cause higher toxic impact to algae cells which stuck to these agglomerates, than in
treatments with duiron and algae only. Same might happen at degraded composite surfaces, where
protruded nanotubes loaded with pollutants may cause locally high concentrations of this pollutant
bioavailable for organisms which may also attach to the surfaces, as described above. Hence, the
interactions of CNT with several environmental relevant pollutants may outline a secondary effects
scenario which may play an important role in the environment and has not been taken into account
Chapter 5
118
for risk assessment up to now, also for released degraded nanotube composite fragments. Due to the
fact that CNT could possibly be modified (Chapter 2) and functionalized during the degradation
process of the composite, realistic test scenarios to evaluate the possible risk are highly necessary.
Additionally, leaching of chemicals from the polymer matrix itself and further attachment of these
chemicals to the nanocomposite surface may outline a possible process to cause hazardous impacts.
For example, bisphenol A (BPA) is known to known to leach from plastics204, 296, especially during
seawater exposure BPA was observed to leach from polycarbonate pellets297. Further, BPA is known
to be degraded by oxygen species297 which formation may be induced by CNT presence89, 298.
Leaching of additives from plastics to organisms is known203, 299, and CNT presence in polymeric
composite materials can possibly alter its behaviour and bioavailability. Bioconcentration factors for
BPA range from 5 to 68 L/kg300, 301, implying that BPA is not considered to be a bioaccumulative
compound. However, these values may be altered due to CNT adsorption122. This could lead to higher
body burdens or even more persistence of pollutants as CNT can alter mineralization rates of
phenanthrene due to material adsorption280. Research in this area is highly desired for the future to
deliver an entire picture of CNT holding composite impacts to the environment not only due to
possible CNT material release.
The observed bioconcentration of CNT to algae cells (Chapter 3) is a non often described observed
result, as bioaccumulation of CNT in organisms was not often described and when only low70, 71, 283.
However, algae could also eliminate the CNT material may influence bioavailability of other
chemicals and attachment may lead to a possible biomagnifications across the aquatic food chain
which should be investigated further. Also CNT loaded with pollutants attached to algae may outline
a possible toxic impact for primary consumers, as they may ingest a high amount of the pollutant in
combination with their food source which causes higher concentration factors as through the water
phase. As chronic exposure of CNT is not yet investigated in detail and ATR-FTIR used in the present
study showed changes in the biochemical composition of the algae cells, possible long termed effects
on aquatic populations have to be investigated in the future.
Chapter 5
119
The experiments with CNT and EE2 proofed, that interactions in aquatic systems could not reliably be
predicted from adsorption studies, as organisms and other factors may influence the pollutant CNT
interactions (Chapter 4). Effect tests with environmentally relevant concentrations of EE2 showed
interesting results for Danio rerio, as at a concentration of 2.2 ng EE2/l an induction of VTG could be
detected only in fish exposed to CNT and EE2 and not in zebrafish exposed to EE2 only. This shows
that experiments with environmentally relevant pollutant concentrations may show different, non
predictable behaviour of combinational exposure of nanoparticles and pollutants. However, the used
concentration of CNT of 1 mg/L was not predicted to be environmentally relevant. But as observed
for the composite degradation also small released fragments of CNT holding polymeric material with
uncovered nanotubes at their surface may leach EE2 from the water phase and provide it to the fish
when digested by them. Furthermore, BPA was described to induce VTG in zebrafish, which may link
the results to the additive leaching from plastic products and possible adsorption at degraded
composite particles described above302.
To conclude, the present work delivered new insights at the fate of CNT holding composite products
at the end of their life cycle: Moreover, new perspectives for CNT ecotoxicology were outlined and
secondary effects of CNT material and an environmentally relevant pollutant were investigated.
These insights may lead to a better understanding of the impact which nanotubes may have to the
aquatic organisms. However, more research is needed to assess all possibly risks of an estimated
release of CNT holding products or raw CNT material to the environment, as most studies till now do
not estimate a risk, but our results showed that there are big gaps in knowledge of CNT material
behaviour and its possible hazardous impacts.
Chapter 5
120
121
References
1. http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2011:275:0038:0040:EN:PDF, Comission recommondation on the definiton of nanomaterial. In EU recommondation 2011/696/EU: 20.10.2011. 2. Perez, S.; Farre, M.; Barcelo, D., Analysis, behavior and ecotoxicity of carbon-based nanomaterials in the aquatic environment. Trac-Trends Anal. Chem. 2009, 28, (6), 820-832. 3. Englert, B. C., Nanomaterials and the environment: uses, methods and measurement. Journal of Environmental Monitoring 2007, 9, 1154-1161. 4. Klaine, S. J.; Alvarez, P. J. J.; Batley, G. E.; Fernandes, T. F.; Handy, R. D.; Lyon, D. Y.; Mahendra, S.; McLaughlin, M. J.; Lead, J. R., Nanomaterials in the environment: Behavior, fate, bioavailability, and effects. Environmental Toxicology and Chemistry 2008, 27, (9), 1825-1851. 5. Gottschalk, F.; Sonderer, T.; Scholz, R. W.; Nowack, B., Modeled Environmental Concentrations of Engineered Nanomaterials (TiO2, ZnO, Ag, CNT, Fullerenes) for Different Regions. Environmental Science & Technology 2009, 43, (24), 9216-9222. 6. Ball, P., Roll up for the revolution. Nature 2001, 414, (6860), 142-144. 7. Masciangioli, T.; Zhang, W. X., Environmental technologies at the nanoscale. Environmental Science & Technology 2003, 37, (5), 102A-108A. 8. Lau, K. T.; Hui, D., The revolutionary creation of new advanced materials - carbon nanotube composites. Composites Part B-Engineering 2002, 33, (4), 263-277. 9. Mauter, M. S.; Elimelech, M., Environmental applications of carbon-based nanomaterials. Environmental Science & Technology 2008, 42, (16), 5843-5859. 10. Liu, Z.; Tabakman, S.; Welsher, K.; Dai, H. J., Carbon Nanotubes in Biology and Medicine: In vitro and in vivo Detection, Imaging and Drug Delivery. Nano Research 2009, 2, (2), 85-120. 11. De Volder, M. F. L.; Tawfick, S. H.; Baughman, R. H.; Hart, A. J., Carbon Nanotubes: Present and Future Commercial Applications. Science 2013, 339, (6119), 535-539. 12. Qu, X. L.; Alvarez, P. J. J.; Li, Q. L., Applications of nanotechnology in water and wastewater treatment. Water Research 2013, 47, (12), 3931-3946. 13. Joseph, L.; Zaib, Q.; Khan, I. A.; Berge, N. D.; Park, Y. G.; Saleh, N. B.; Yoon, Y., Removal of bisphenol A and 17 alpha-ethinyl estradiol from landfill leachate using single-walled carbon nanotubes. Water Research 2011, 45, (13), 4056-4068. 14. Purceno, A. D.; Teixeira, A. P. C.; de Souza, N. J.; Fernandez-Outon, L. E.; Ardisson, J. D.; Lago, R. M., Hybrid magnetic amphiphilic composites based on carbon nanotube/nanofibers and layered silicates fragments as efficient adsorbent for ethynilestradiol. Journal of Colloid and Interface Science 2012, 379, 84-88. 15. Nowack, B.; Bucheli, T. D., Occurrence, behavior and effects of nanoparticles in the environment. Environmental Pollution 2007, 150, 5-22. 16. Kohler, A. R.; Som, C.; Helland, A.; Gottschalk, F., Studying the potential release of carbon nanotubes throughout the application life cycle. Journal of Cleaner Production 2008, 16, (8-9), 927-937. 17. Mueller, N. C.; Nowack, B., Exposure modeling of engineered nanoparticles in the environment. Environmental Science & Technology 2008, 42, (12), 4447-4453. 18. Gottschalk, F.; Nowack, B., The release of engineered nanomaterials to the environment. Journal of Environmental Monitoring 2011, 13, (5), 1145-1155. 19. Gottschalk, F.; Sun, T.; Nowack, B., Environmental concentrations of engineered nanomaterials: review of modeling and analytical studies. Environ. Pollut. 2013, 181, 287-300. 20. Gottschalk, F.; Kost, E.; Nowack, B., Engineered nanomaterials in water and soils: A risk quantification based on probabilistic exposure and effect modeling. Environmental Toxicology and Chemistry 2013, 32, (6), 1278-1287.
122
21. Nowack, B.; Mueller, N. C.; Krug, H. F.; Wick, P., How to consider engineered nanomaterials in major accident regulations? Environmental Sciences Europe 2014, 26, (1), 2. 22. Oberdorster, G.; Stone, V.; Donaldson, K., Toxicology of nanoparticles: A historical perspective. Nanotoxicology 2007, 1, (1), 2-25. 23. Farre, M.; Gajda-Schrantz, K.; Kantiani, L.; Barcelo, D., Ecotoxicity and analysis of nanomaterials in the aquatic environment. Anal. Bioanal. Chem. 2009, 393, (1), 81-95. 24. Abbott, L. C.; Maynard, A. D., Exposure Assessment Approaches for Engineered Nanomaterials. Risk Analysis 2010, 30, (11), 1634-1644. 25. von der Kammer, F.; Ferguson, P. L.; Holden, P. A.; Masion, A.; Rogers, K. R.; Klaine, S. J.; Koelmans, A. A.; Horne, N.; Unrine, J. M., Analysis of engineered nanomaterials in complex matrices (environment and biota): General considerations and conceptual case studies. Environmental Toxicology and Chemistry 2012, 31, (1), 32-49. 26. Clark, K.; van Tongeren, M.; Christensen, F. M.; Brouwer, D.; Nowack, B.; Gottschalk, F.; Micheletti, C.; Schmid, K.; Gerritsen, R.; Aitken, R.; Vaquero, C.; Gkanis, V.; Housiadas, C.; de Ipina, J. M. L.; Riediker, M., Limitations and information needs for engineered nanomaterial-specific exposure estimation and scenarios: recommendations for improved reporting practices. Journal of Nanoparticle Research 2012, 14, (9). 27. Handy, R. D.; Owen, R.; Valsami-Jones, E., The ecotoxicology of nanoparticles and nanomaterials: current status, knowledge gaps, challenges, and future needs. Ecotoxicology 2008, 17, (5), 315-325. 28. Brooks, A. J.; Lim, H. N.; Kilduff, J. E., Adsorption uptake of synthetic organic chemicals by carbon nanotubes and activated carbons. Nanotechnology 2012, 23, (29), 13. 29. Iijima, S., HELICAL MICROTUBULES OF GRAPHITIC CARBON. Nature 1991, 354, (6348), 56-58. 30. Ajayan, P. M., Nanotubes from carbon. Chem. Rev. 1999, 99, (7), 1787-1799. 31. Newman, P.; Minett, A.; Ellis-Behnke, R.; Zreiqat, H., Carbon nanotubes: Their potential and pitfalls for bone tissue regeneration and engineering. Nanomedicine-Nanotechnology Biology and Medicine 2013, 9, (8), 1139-1158. 32. Murr, L. E.; Esquivel, E. V.; Bang, J. J.; de la Rosa, G.; Gardea-Torresdey, J. L., Chemistry and nanoparticulate compositions of a 10,000 year-old ice core melt water. Water Research 2004, 38, (19), 4282-4296. 33. See, C. H.; Harris, A. T., A review of carbon nanotube synthesis via fluidized-bed chemical vapor deposition. Ind. Eng. Chem. Res. 2007, 46, (4), 997-1012. 34. Grobert, N., Carbon nanotubes - becoming clean. Mater. Today 2007, 10, (1-2), 28-35. 35. Liu, X. M.; Spencer, J. L.; Kaiser, A. B.; Arnold, W. M., Selective purification of multiwalled carbon nanotubes by dielectrophoresis within a large array. Current Applied Physics 2006, 6, (3), 427-431. 36. Petersen, E. J.; Huang, Q. G.; Weber, W. J., Bioaccumulation of radio-labeled carbon nanotubes by Eisenia foetida. Environmental Science & Technology 2008, 42, (8), 3090-3095. 37. Worle-Knirsch, J. M.; Pulskamp, K.; Krug, H. F., Oops they did it again! Carbon nanotubes hoax scientists in viability assays. Nano Letters 2006, 6, (6), 1261-1268. 38. Buford, M. C.; Hamilton, R. F.; Holian, A., A comparison of dispersing media for various engineered carbon nanoparticles. Part Fibre Toxicol 2007, 4, 6. 39. Petersen, E. J.; Zhang, L.; Mattison, N. T.; O'Carroll, D. M.; Whelton, A. J.; Uddin, N.; Tinh, N.; Huang, Q.; Henry, T. B.; Holbrook, R. D.; Chen, K. L., Potential Release Pathways, Environmental Fate, And Ecological Risks of Carbon Nanotubes. Environmental Science & Technology 2011, 45, (23), 9837-9856. 40. Nowack, B.; Ranville, J. F.; Diamond, S.; Gallego-Urrea, J. A.; Metcalfe, C.; Rose, J.; Horne, N.; Koelmans, A. A.; Klaine, S. J., Potential scenarios for nanomaterial release and subsequent alteration in the environment. Environmental Toxicology and Chemistry 2012, 31, (1), 50-59. 41. Nowack, B.; David, R. M.; Fissan, H.; Morris, H.; Shatkin, J. A.; Stintz, M.; Zepp, R.; Brouwer, D., Potential release scenarios for carbon nanotubes used in composites. Environment international 2013, 59, 1-11.
123
42. Koelmans, A. A.; Nowack, B.; Wiesner, M. R., Comparison of manufactured and black carbon nanoparticle concentrations in aquatic sediments. Environmental Pollution 2009, 157, (4), 1110-1116. 43. Upadhyayula, V. K. K.; Meyer, D. E.; Curran, M. A.; Gonzalez, M. A., Life cycle assessment as a tool to enhance the environmental performance of carbon nanotube products: a review. Journal of Cleaner Production 2012, 26, 37-47. 44. Roes, L.; Patel, M. K.; Worrell, E.; Ludwig, C., Preliminary evaluation of risks related to waste incineration of polymer nanocomposites. Science of the Total Environment 2012, 417, 76-86. 45. Holder, A. L.; Vejerano, E. P.; Zhou, X. Z.; Marr, L. C., Nanomaterial disposal by incineration. Environmental Science-Processes & Impacts 2013, 15, (9), 1652-1664. 46. Oberdorster, G.; Oberdorster, E.; Oberdorster, J., Nanotoxicology: An emerging discipline evolving from studies of ultrafine particles. Environmental Health Perspectives 2005, 113, (7), 823-839. 47. Jaisi, D. P.; Elimelech, M., Single-Walled Carbon Nanotubes Exhibit Limited Transport in Soil Columns. Environmental Science & Technology 2009, 43, (24), 9161-9166. 48. Kasel, D.; Bradford, S. A.; Šimůnek, J.; Pütz, T.; Vereecken, H.; Klumpp, E., Limited transport of functionalized multi-walled carbon nanotubes in two natural soils. Environmental Pollution 2013, 180, (0), 152-158. 49. Kasel, D.; Bradford, S. A.; Šimůnek, J.; Heggen, M.; Vereecken, H.; Klumpp, E., Transport and retention of multi-walled carbon nanotubes in saturated porous media: Effects of input concentration and grain size. Water Research 2013, 47, (2), 933-944. 50. Ryman-Rasmussen, J. P.; Cesta, M. F.; Brody, A. R.; Shipley-Phillips, J. K.; Everitt, J. I.; Tewksbury, E. W.; Moss, O. R.; Wong, B. A.; Dodd, D. E.; Andersen, M. E.; Bonner, J. C., Inhaled carbon nanotubes reach the subpleural tissue in mice. Nature Nanotechnology 2009, 4, (11), 747-751. 51. Pacurari, M.; Castranova, V.; Vallyathan, V., SINGLE- AND MULTI-WALL CARBON NANOTUBES VERSUS ASBESTOS: ARE THE CARBON NANOTUBES A NEW HEALTH RISK TO HUMANS? Journal of Toxicology and Environmental Health-Part a-Current Issues 2010, 73, (5-6), 378-395. 52. Yamashita, K.; Yoshioka, Y.; Higashisaka, K.; Morishita, Y.; Yoshida, T.; Fujimura, M.; Kayamuro, H.; Nabeshi, H.; Yamashita, T.; Nagano, K.; Abe, Y.; Kamada, H.; Kawai, Y.; Mayumi, T.; Yoshikawa, T.; Itoh, N.; Tsunoda, S.; Tsutsumi, Y., Carbon Nanotubes Elicit DNA Damage and Inflammatory Response Relative to Their Size and Shape. Inflammation 2010, 33, (4), 276-280. 53. Kennedy, A. J.; Hull, M. S.; Steevens, J. A.; Dontsova, K. M.; Chappell, M. A.; Gunter, J. C.; Weiss, C. A., Factors influencing the partitioning and toxicity of nanotubes in the aquatic environment. Environmental Toxicology and Chemistry 2008, 27, (9), 1932-1941. 54. Chang, X. J.; Bouchard, D. C., Multiwalled Carbon Nanotube Deposition on Model Environmental Surfaces. Environmental Science & Technology 2013, 47, (18), 10372-10380. 55. Liu, X. Y.; Chen, G. X.; Keller, A. A.; Su, C. M., Effects of dominant material properties on the stability and transport of TiO2 nanoparticles and carbon nanotubes in aquatic environments: from synthesis to fate. Environmental Science-Processes & Impacts 2013, 15, (1), 169-189. 56. Smith, B.; Yang, J.; Bitter, J. L.; Ball, W. P.; Fairbrother, D. H., Influence of Surface Oxygen on the Interactions of Carbon Nanotubes with Natural Organic Matter. Environmental Science & Technology 2012, 46, (23), 12839-12847. 57. Su, F. S.; Lu, C. S., Adsorption kinetics, thermodynamics and desorption of natural dissolved organic matter by multiwalled carbon nanotubes. Journal of Environmental Science and Health Part a-Toxic/Hazardous Substances & Environmental Engineering 2007, 42, (11), 1543-1552. 58. Hyung, H.; Fortner, J. D.; Hughes, J. B.; Kim, J. H., Natural organic matter stabilizes carbon nanotubes in the aqueous phase. Environmental Science & Technology 2007, 41, 179-184. 59. Hyung, H.; Kim, J. H., Natural organic matter (NOM) adsorption to multi-walled carbon nanotubes: Effect of NOM characteristics and water quality parameters. Environmental Science & Technology 2008, 42, (12), 4416-4421. 60. Lin, D.; Li, T.; Yang, K.; Wu, F., The relationship between humic acid (HA) adsorption on and stabilizing multiwalled carbon nanotubes (MWNTs) in water: Effects of HA, MWNT and solution properties. Journal of Hazardous Materials 2012, 241, 404-410.
124
61. Zhang, L. W.; Pan, Z. W.; Huang, Q. G., Effect of primary particle size on colloidal stability of multiwall carbon nanotubes. Water Science and Technology 2013, 68, (10), 2249-2256. 62. Petersen, E. J.; Huang, Q. G.; Weber, W. J., Relevance of octanol-water distribution measurements to the potential ecological uptake of multi-walled carbon nanotubes. Environ. Toxicol. Chem. 2010, 29, (5), 1106-1112. 63. Allen, B. L.; Kichambare, P. D.; Gou, P.; Vlasova, I. I.; Kapralov, A. A.; Konduru, N.; Kagan, V. E.; Star, A., Biodegradation of Single-Walled Carbon Nanotubes through Enzymatic Catalysis. Nano Letters 2008, 8, (11), 3899-3903. 64. Allen, B. L.; Kotchey, G. P.; Chen, Y.; Yanamala, N. V. K.; Klein-Seetharaman, J.; Kagan, V. E.; Star, A., Mechanistic Investigations of Horseradish Peroxidase-Catalyzed Degradation of Single-Walled Carbon Nanotubes. Journal of the American Chemical Society 2009, 131, (47), 17194-17205. 65. Zhao, Y.; Allen, B. L.; Star, A., Enzymatic Degradation of Multiwalled Carbon Nanotubes. Journal of Physical Chemistry A 2011, 115, (34), 9536-9544. 66. Kotchey, G. P.; Gaugler, J. A.; Kapralov, A. A.; Kagan, V. E.; Star, A., Effect of antioxidants on enzyme-catalysed biodegradation of carbon nanotubes. Journal of Materials Chemistry B 2013, 1, (3), 302-309. 67. Flores-Cervantes, D.; Kohler, H.-P. E.; Maes, H. M.; Schaeffer, A.; Hollender, J., Slow biotransformation of carbon nanotubes by horseradish peroxidase. Environmental Science & Technology 2014. 68. Zhang, L. W.; Petersen, E. J.; Habteselassie, M. Y.; Mao, L.; Huang, Q. G., Degradation of multiwall carbon nanotubes by bacteria. Environmental Pollution 2013, 181, 335-339. 69. Fraczek-Szczypta, A.; Menaszek, E.; Syeda, T. B.; Misra, A.; Alavijeh, M.; Adu, J.; Blazewicz, S., Effect of MWCNT surface and chemical modification on in vitro cellular response. Journal of Nanoparticle Research 2012, 14, (10). 70. Hou, W. C.; Westerhoff, P.; Posner, J. D., Biological accumulation of engineered nanomaterials: a review of current knowledge. Environmental Science-Processes & Impacts 2013, 15, (1), 103-122. 71. Parks, A. N.; Portis, L. M.; Schierz, P. A.; Washburn, K. M.; Perron, M. M.; Burgess, R. M.; Ho, K. T.; Chandler, G. T.; Ferguson, P. L., Bioaccumulation and toxicity of single-walled carbon nanotubes to benthic organisms at the base of the marine food chain. Environmental Toxicology and Chemistry 2013, 32, (6), 1270-1277. 72. Zhu, Y.; Zhao, Q. F.; Li, Y. G.; Cai, X. Q.; Li, W., The interaction and toxicity of multi-walled carbon nanotubes with Stylonychia mytilus. Journal of Nanoscience and Nanotechnology 2006, 6, (5), 1357-1364. 73. Petersen, E. J.; Akkanen, J.; Kukkonen, J. V. K.; Weber, W. J., Biological Uptake and Depuration of Carbon Nano-tubes by Daphnia magna. Environmental Science & Technology 2009, 43, (8), 2969-2975. 74. Roberts, A. P.; Mount, A. S.; Seda, B.; Souther, J.; Qiao, R.; Lin, S. J.; Ke, P. C.; Rao, A. M.; Klaine, S. J., In vivo biomodification of lipid-coated carbon nanotubes by Daphnia magna. Environmental Science & Technology 2007, 41, (8), 3025-3029. 75. Kahru, A.; Dubourguier, H. C., From ecotoxicology to nanoecotoxicology. Toxicology 2010, 269, (2-3), 105-119. 76. Handy, R. D.; van den Brink, N.; Chappell, M.; Muhling, M.; Behra, R.; Dusinska, M.; Simpson, P.; Ahtiainen, J.; Jha, A. N.; Seiter, J.; Bednar, A.; Kennedy, A.; Fernandes, T. F.; Riediker, M., Practical considerations for conducting ecotoxicity test methods with manufactured nanomaterials: what have we learnt so far? Ecotoxicology 2012, 21, (4), 933-972. 77. Petersen, E. J.; Henry, T. B., Methodological considerations for testing the ecotoxicity of carbon nanotubes and fullerenes: Review. Environmental toxicology and chemistry / SETAC 2012, 31, (1), 60-72. 78. Hussain, M. A.; Kabir, M. A.; Sood, A. K., On the cytotoxicity of carbon nanotubes. Current Science 2009, 96, (5), 664-673. 79. Jackson, P.; Jacobsen, N. R.; Baun, A.; Birkedal, R.; Kuhnel, D.; Jensen, K. A.; Vogel, U.; Wallin, H., Bioaccumulation and ecotoxicity of carbon nanotubes. Chemistry Central Journal 2013, 7.
125
80. Edgington, A. J.; Roberts, A. P.; Taylor, L. M.; Alloy, M. M.; Reppert, J.; Rao, A. M.; Mao, J.; Klaine, S. J., The influence of natural organic matter on the toxicity of multiwalled carbon nanotubes. Environmental Toxicology and Chemistry 2010, 29, (11), 2511-2518. 81. Petersen, E. J.; Huang, Q. G.; Weber, W. J., Ecological uptake and depuration of carbon nanotubes by Lumbriculus variegatus. Environmental Health Perspectives 2008, 116, (4), 496-500. 82. Petersen, E. J.; Pinto, R. A.; Mai, D. J.; Landrum, P. F.; Weber, W. J., Influence of Polyethyleneimine Graftings of Multi-Walled Carbon Nanotubes on their Accumulation and Elimination by and Toxicity to Daphnia magna. Environmental Science & Technology 2011, 45, (3), 1133-1138. 83. Galloway, T.; Lewis, C.; Dolciotti, I.; Johnston, B. D.; Moger, J.; Regoli, F., Sublethal toxicity of nano-titanium dioxide and carbon nanotubes in a sediment dwelling marine polychaete. Environ Pollut 2010, 158, (5), 1748-55. 84. Ghafari, P.; St-Denis, C. H.; Power, M. E.; Jin, X.; Tsou, V.; Mandal, H. S.; Bols, N. C.; Tang, X., Impact of carbon nanotubes on the ingestion and digestion of bacteria by ciliated protozoa. Nat Nano 2008, 3, (6), 347-351. 85. Smith, C. J.; Shaw, B. J.; Handy, R. D., Toxicity of single walled carbon nanotubes to rainbow trout, (Oncorhynchus mykiss): Respiratory toxicity, organ pathologies, and other physiological effects. Aquatic Toxicology 2007, 82, (2), 94-109. 86. Thurnherr, T.; Brandenberger, C.; Fischer, K.; Diener, L.; Manser, P.; Maeder-Althaus, X.; Kaiser, J.-P.; Krug, H. F.; Rothen-Rutishauser, B.; Wick, P., A comparison of acute and long-term effects of industrial multiwalled carbon nanotubes on human lung and immune cells in vitro. Toxicology Letters 2011, 200, (3), 176-186. 87. Cheng, C.; Muller, K. H.; Koziol, K. K. K.; Skepper, J. N.; Midgley, P. A.; Welland, M. E.; Porter, A. E., Toxicity and imaging of multi-walled carbon nanotubes in human macrophage cells. Biomaterials 2009a, 30, (25), 4152-4160. 88. Kang, S.; Herzberg, M.; Rodrigues, D. F.; Elimelech, M., Antibacterial effects of carbon nanotubes: Size does matter. Langmuir 2008, 24, (13), 6409-6413. 89. Long, Z. F.; Ji, J.; Yang, K.; Lin, D. H.; Wu, F. C., Systematic and Quantitative Investigation of the Mechanism of Carbon Nanotubes' Toxicity toward Algae. Environmental Science & Technology 2012, 46, (15), 8458-8466. 90. Wei, L. P.; Thakkar, M.; Chen, Y. H.; Ntim, S. A.; Mitra, S.; Zhang, X. Y., Cytotoxicity effects of water dispersible oxidized multiwalled carbon nanotubes on marine alga, Dunaliella tertiolecta. Aquatic Toxicology 2010, 100, (2), 194-201. 91. Schwab, F.; Bucheli, T. D.; Lukhele, L. P.; Magrez, A.; Nowack, B.; Sigg, L.; Knauer, K., Are carbon nanotube effects on green algae caused by shading and agglomeration? Environmental Science & Technology 2011, 45, (14), 6136-44. 92. Bennett, S. W.; Adeleye, A.; Ji, Z.; Keller, A. A., Stability, metal leaching, photoactivity and toxicity in freshwater systems of commercial single wall carbon nanotubes. Water research 2013, 47, (12), 4074-85. 93. Schwab, F.; Bucheli, T. D.; Lukhele, L. P.; Magrez, A.; Nowack, B.; Sigg, L.; Knauer, K., Are Carbon Nanotube Effects on Green Algae Caused by Shading and Agglomeration? Environmental Science & Technology 2011, 45, (14), 6136-6144. 94. Neves, V.; Heister, E.; Costa, S.; Tilmaciu, C.; Borowiak-Palen, E.; Giusca, C. E.; Flahaut, E.; Soula, B.; Coley, H. M.; McFadden, J.; Silva, S. R. P., Uptake and Release of Double-Walled Carbon Nanotubes by Mammalian Cells. Advanced Functional Materials 2010, 20, (19), 3272-3279. 95. Blaise, C.; Gagne, F.; Ferard, J. F.; Eullaffroy, P., Ecotoxicity of selected nano-materials to aquatic organisms. Environ. Toxicol. 2008, 23, (5), 591-598. 96. Fengwei, M.; Ping, W.; Handong, L.; Zhixiang, Z., Growth inhibitions of four types of CNTs on Scenedesmus obliquus. Journal of Convergence Information Technology 2013, 8, (10), 176-182. 97. Wick, P.; Manser, P.; Limbach, L. K.; Dettlaff-Weglikowska, U.; Krumeich, F.; Roth, S.; Stark, W. J.; Bruinink, A., The degree and kind of agglomeration affect carbon nanotube cytotoxicity. Toxicology Letters 2007, 168, (2), 121-131.
126
98. Al-Jamal, K. T.; Nerl, H.; Muller, K. H.; Ali-Boucetta, H.; Li, S. P.; Haynes, P. D.; Jinschek, J. R.; Prato, M.; Bianco, A.; Kostarelos, K.; Porter, A. E., Cellular uptake mechanisms of functionalised multi-walled carbon nanotubes by 3D electron tomography imaging. Nanoscale 2011, 3, (6), 2627-2635. 99. Lacerda, L.; Raffa, S.; Prato, M.; Bianco, A.; Kostarelos, K., Cell-penetrating CNTs for delivery of therapeutics. Nano Today 2007, 2, (6), 38-43. 100. Cheng, J. P.; Flahaut, E.; Cheng, S. H., Effect of carbon nanotubes on developing zebrafish (Danio rerio) embryos. Environmental Toxicology and Chemistry 2007, 26, (4), 708-716. 101. Jia, G.; Wang, H. F.; Yan, L.; Wang, X.; Pei, R. J.; Yan, T.; Zhao, Y. L.; Guo, X. B., Cytotoxicity of carbon nanomaterials: Single-wall nanotube, multi-wall nanotube, and fullerene. Environmental Science & Technology 2005, 39, (5), 1378-1383. 102. Pantarotto, D.; Briand, J. P.; Prato, M.; Bianco, A., Translocation of bioactive peptides across cell membranes by carbon nanotubes. Chemical Communications 2004, (1), 16-17. 103. Chin, S. F.; Baughman, R. H.; Dalton, A. B.; Dieckmann, G. R.; Draper, R. K.; Mikoryak, C.; Musselman, I. H.; Poenitzsch, V. Z.; Xie, H.; Pantano, P., Amphiphilic helical peptide enhances the uptake of single-walled carbon nanotubes by living cells. Experimental Biology and Medicine 2007, 232, 1236-1244. 104. Cheng, C.; Porter, A. E.; Muller, K.; Koziol, K.; Skepper, J. N.; Midgley, P.; Welland, M., Imaging carbon nanoparticles and related cytotoxicity. Inhaled Particles X 2009, 151. 105. Toh, R. J.; Ambrosi, A.; Pumera, M., Bioavailability of Metallic Impurities in Carbon Nanotubes Is Greatly Enhanced by Ultrasonication. Chemistry-a European Journal 2012, 18, (37), 11593-11596. 106. Kam, N. W. S.; Jessop, T. C.; Wender, P. A.; Dai, H. J., Nanotube molecular transporters: Internalization of carbon nanotube-protein conjugates into mammalian cells. Journal of the American Chemical Society 2004, 126, (22), 6850-6851. 107. Kateb, B.; Van Handel, M.; Zhang, L. Y.; Bronikowski, M. J.; Manohara, H.; Badie, B., Internalization of MWCNTs by microglia: Possible application in immunotherapy of brain tumors. Neuroimage 2007, 37, S9-S17. 108. Liu, Q. L.; Chen, B.; Wang, Q. L.; Shi, X. L.; Xiao, Z. Y.; Lin, J. X.; Fang, X. H., Carbon Nanotubes as Molecular Transporters for Walled Plant Cells. Nano Letters 2009, 9, (3), 1007-1010. 109. Manna, S. K.; Sarkar, S.; Barr, J.; Wise, K.; Barrera, E. V.; Jejelowo, O.; Rice-Ficht, A. C.; Ramesh, G. T., Single-walled carbon nanotube induces oxidative stress and activates nuclear transcription factor-kappa B in human keratinocytes. Nano Lett. 2005, 5, (9), 1676-1684. 110. Pulskamp, K.; Diabate, S.; Krug, H. F., Carbon nanotubes show no sign of acute toxicity but induce intracellular reactive oxygen species in dependence on contaminants. Toxicology Letters 2007, 168, (1), 58-74. 111. Hsieh, S.-F.; Bello, D.; Schmidt, D. F.; Pal, A. K.; Rogers, E. J., Biological oxidative damage by carbon nanotubes: Fingerprint or footprint? Nanotoxicology 2012, 6, (1), 61-76. 112. Fent, K., Ökotoxikologie. 3. Edition ed.; Georg Thieme Verlag: Stuttgart, 2007. 113. Lindberg, H. K.; Falck, G. C. M.; Suhonen, S.; Vippola, M.; Vanhala, E.; Catalan, J.; Savolainen, K.; Norppa, H., Genotoxicity of nanomaterials: DNA damage and micronuclei induced by carbon nanotubes and graphite nanofibres in human bronchial epithelial cells in vitro. Toxicology Letters 2009, 186, (3), 166-173. 114. Tyurina, Y. Y.; Kisin, E. R.; Murray, A.; Tyurin, V. A.; Kapralova, V. I.; Sparvero, L. J.; Amoscato, A. A.; Samhan-Arias, A. K.; Swedin, L.; Lahesmaa, R.; Fadeel, B.; Shvedova, A. A.; Kagan, V. E., Global Phospholipidomics Analysis Reveals Selective Pulmonary Peroxidation Profiles upon Inhalation of Single-Walled Carbon Nanotubes. Acs Nano 2011, 5, (9), 7342-7353. 115. Guo, Y. Y.; Zhang, J.; Zheng, Y. F.; Yang, J.; Zhu, X. Q., Cytotoxic and genotoxic effects of multi-wall carbon nanotubes on human umbilical vein endothelial cells in vitro. Mutat. Res.-Genet. Toxicol. Environ. Mutag. 2011, 721, (2), 184-191. 116. Wild, E.; Jones, K. C., Novel Method for the Direct Visualization of in Vivo Nanomaterials and Chemical Interactions in Plants. Environmental Science & Technology 2009, 43, (14), 5290-5294. 117. Chen, X.; Schluesener, H. J., Multi-walled carbon nanotubes affect drug transport across cell membrane in rat astrocytes. Nanotechnology 2010, 21, (10), 10.
127
118. Velzeboer, I.; Peeters, E.; Koelmans, A., Multiwalled Carbon Nanotubes at Environmentally Relevant Concentrations Affect the Composition of Benthic Communities. Environmental Science & Technology 2013, 47, (13), 7475-7482. 119. Zhang, X.; Kah, M.; Jonker, M. T. O.; Hofmann, T., Dispersion State and Humic Acids Concentration-Dependent Sorption of Pyrene to Carbon Nanotubes. Environmental Science & Technology 2012, 46, (13), 7166-7173. 120. Cho, H. H.; Huang, H.; Schwab, K., Effects of Solution Chemistry on the Adsorption of Ibuprofen and Triclosan onto Carbon Nanotubes. Langmuir 2011, 27, (21), 12960-12967. 121. Cong, Q.; Yuan, X.; Qu, J., A review on the removal of antibiotics by carbon nanotubes. Water Science and Technology 2013, 68, (8), 1679-1687. 122. Pan, B.; Lin, D. H.; Mashayekhi, H.; Xing, B. S., Adsorption and hysteresis of bisphenol A and 17 alpha-ethinyl estradiol on carbon nanomaterials. Environmental Science & Technology 2008, 42, (15), 5480-5485. 123. Rao, G. P.; Lu, C.; Su, F., Sorption of divalent metal ions from aqueous solution by carbon nanotubes: a review. Sep Purif Technol 2007, 58, (1), 224-231. 124. Vukovic, G. D.; Marinkovic, A. D.; Colic, M.; Ristic, M. D.; Aleksic, R.; Peric-Grujic, A. A.; Uskokovic, P. S., Removal of cadmium from aqueous solutions by oxidized and ethylenediamine-functionalized multi-walled carbon nanotubes. Chemical Engineering Journal 2010, 157, (1), 238-248. 125. Cao, Q. M.; Yu, Q. M.; Connell, D. W.; Yu, G., Titania/carbon nanotube composite (TiO2/CNT) and its application for removal of organic pollutants. Clean Technologies and Environmental Policy 2013, 15, (6), 871-880. 126. Li, S.; Anderson, T. A.; Maul, J. D.; Shrestha, S.; Green, M. J.; Canas-Carrell, J. E., Comparative studies of multi-walled carbon nanotubes (MWNTs) and octadecyl (C-18) as sorbents in passive sampling devices for biomimetic uptake of polycyclic aromatic hydrocarbons (PAHs) from soils. Science of the Total Environment 2013, 461, 560-567. 127. Kah, M.; Zhang, X.; Jonker, M. T. O.; Hofmann, T., Measuring and Modelling Adsorption of PAHs to Carbon Nanotubes Over a Six Order of Magnitude Wide Concentration Range. Environmental Science & Technology 2011, 45, (14), 6011-6017. 128. Towell, M.; Browne, L.; Paton, G.; Semple, K., Impact of carbon nanomaterials on the behaviour of C-14-phenanthrene and C-14-benzo-[a] pyrene in soil. Environmental Pollution 2011, 159, (3), 706-715. 129. Xia, X. H.; Chen, X.; Zhao, X. L.; Chen, H. T.; Shen, M. H., Effects of Carbon Nanotubes, Chars, and Ash on Bioaccumulation of Perfluorochemicals by Chironomus plumosus Larvae in Sediment. Environmental Science & Technology 2012, 46, (22), 12467-12475. 130. Shen, M.; Xia, X.; Wang, F.; Zhang, P.; Zhao, X., Influences of multiwalled carbon nanotubes and plant residue chars on bioaccumulation of polycyclic aromatic hydrocarbons by Chironomus plumosus larvae in sediment. Environmental toxicology and chemistry / SETAC 2012, 31, (1), 202-9. 131. Li, S. B.; Turaga, U.; Shrestha, B.; Anderson, T. A.; Ramkumar, S. S.; Green, M. J.; Das, S.; Canas-Carrell, J. E., Mobility of polyaromatic hydrocarbons (PAHs) in soil in the presence of carbon nanotubes. Ecotoxicology and Environmental Safety 2013, 96, 168-174. 132. Wang, Z.; Zhao, J.; Song, L.; Mashayekhi, H.; Chefetz, B.; Xing, B., Adsorption and desorption of phenanthrene on carbon nanotubes in simulated gastrointestinal fluids. Environmental Science & Technology 2011, 45, (14), 6018-24. 133. Johnson, A. C.; Williams, R. J., A model to estimate influent and effluent concentrations of estradiol, estrone, and ethinylestradiol at sewage treatment works. Environmental Science & Technology 2004, 38, (13), 3649-3658. 134. Ying, G. G.; Kookana, R. S.; Ru, Y. J., Occurrence and fate of hormone steroids in the environment. Environment International 2002, 28, (6), 545-551. 135. Xu, J.; Wu, L.; Chen, W.; Jiang, P.; Chang, A. C.-S., Pharmaceuticals and Personal Care Products (PPCPs), and Endocrine Disrupting Compounds (EDCs) in Runoff from a Potato Field Irrigated with Treated Wastewater in Southern California. Journal of Health Science 2009, 55, (2), 306-310.
128
136. Desbrow, C.; Routledge, E. J.; Brighty, G. C.; Sumpter, J. P.; Waldock, M., Identification of estrogenic chemicals in STW effluent. 1. Chemical fractionation and in vitro biological screening. Environmental Science & Technology 1998, 32, (11), 1549-1558. 137. Williams, R. J.; Johnson, A. C.; Smith, J. J. L.; Kanda, R., Steroid estrogens profiles along river stretches arising from sewage treatment works discharges. Environmental Science & Technology 2003, 37, (9), 1744-1750. 138. Langston, W. J.; Burt, G. R.; Chesman, B. S.; Vane, C. H., Partitioning, bioavallability and effects of oestrogens and xeno-oestrogens in the aquatic environment. Journal of the Marine Biological Association of the United Kingdom 2005, 85, (1), 1-31. 139. Ternes, T. A.; Stumpf, M.; Mueller, J.; Haberer, K.; Wilken, R. D.; Servos, M., Behavior and occurrence of estrogens in municipal sewage treatment plants - I. Investigations in Germany, Canada and Brazil. Science of the Total Environment 1999, 225, (1-2), 81-90. 140. Jurgens, M. D.; Holthaus, K. I. E.; Johnson, A. C.; Smith, J. J. L.; Hetheridge, M.; Williams, R. J., The potential for estradiol and ethinylestradiol degradation in English rivers. Environmental Toxicology and Chemistry 2002, 21, (3), 480-488. 141. Ternes, T. A.; Kreckel, P.; Mueller, J., Behaviour and occurrence of estrogens in municipal sewage treatment plants - II. Aerobic batch experiments with activated sludge. Science of the Total Environment 1999, 225, (1-2), 91-99. 142. Robinson, B. J.; Hellou, J., Biodegradation of endocrine disrupting compounds in harbour seawater and sediments. Science of the Total Environment 2009, 407, (21), 5713-5718. 143. Maes, H. M. Fate of ethinylestradiol in the aquatic environment and the associated effects on organisms of different trophic levels. RWTH Aachen University, Aachen, Germany, 2011. 144. Thorpe, K. L.; Cummings, R. I.; Hutchinson, T. H.; Scholze, M.; Brighty, G.; Sumpter, J. P.; Tyler, C. R., Relative potencies and combination effects of steroidal estrogens in fish. Environmental Science & Technology 2003, 37, (6), 1142-1149. 145. Caldwell, D. J.; Mastrocco, F.; Anderson, P. D.; Laenge, R.; Sumpter, J. P., Predicted-no-effect concentrations for the steroid estrogens estrone, 17 beta-estradiol, estriol, and 17 alpha-ethinylestradiol. Environmental Toxicology and Chemistry 2012, 31, (6), 1396-1406. 146. Hense, B. A.; Severin, G. F.; Pfister, G.; Welzl, G.; Jaser, W.; Schramm, K. W., Effects of anthropogenic estrogens nonylphenol and 17 alpha-ethinylestradiol in aquatic model ecosystems. Acta Hydrochimica Et Hydrobiologica 2005, 33, (1), 27-37. 147. Orn, S.; Yamani, S.; Norrgren, L., Comparison of vitellogenin induction, sex ratio, and gonad morphology between zebrafish and Japanese medaka after exposure to 17 alpha-ethinylestradiol and 17 beta-trenbolone. Archives of Environmental Contamination and Toxicology 2006, 51, (2), 237-243. 148. Xu, H.; Yang, J.; Wang, Y. X.; Jiang, Q.; Chen, H.; Song, H. Y., Exposure to 17 alpha-ethynylestradiol impairs reproductive functions of both male and female zebrafish (Danio rerio). Aquatic Toxicology 2008, 88, (1), 1-8. 149. Van den Belt, K.; Verheyen, R.; Witters, H., Effects of 17 alpha-ethynylestradiol in a partial life-cycle test with zebrafish (Danio rerio): effects on growth, gonads and female reproductive success. Science of the Total Environment 2003, 309, (1-3), 127-137. 150. Rose, J.; Holbech, H.; Lindholst, C.; Norum, U.; Povlsen, A.; Korsgaard, B.; Bjerregaard, P., Vitellogenin induction by 17 beta-estradiol and 17 alpha-ethinylestradiol in male zebrafish (Danio rerio). Comparative Biochemistry and Physiology C-Toxicology & Pharmacology 2002, 131, (4), 531-539. 151. Versonnen, B. J.; Arijs, K.; Verslycke, T.; Lema, W.; Janssen, C. R., In vitro and in vivo estrogenicity and toxicity of o-, m-, and p-dichlorobenzene. Environmental Toxicology and Chemistry 2003, 22, (2), 329-335. 152. Lai, K. M.; Scrimshaw, M. D.; Lester, J. N., Prediction of the bioaccumulation factors and body burden of natural and synthetic estrogens in aquatic organisms in the river systems. Science of the Total Environment 2002, 289, (1-3), 159-168. 153. Lange, R.; Hutchinson, T. H.; Croudace, C. P.; Siegmund, F.; Schweinfurth, H.; Hampe, P.; Panter, G. H.; Sumpter, J. P., Effects of the synthetic estrogen 17 alpha-ethinylestradiol on the life-
129
cycle of the fathead minnow (Pimephales promelas). Environmental Toxicology and Chemistry 2001, 20, (6), 1216-1227. 154. Pan, B.; Xing, B. S., Adsorption Mechanisms of Organic Chemicals on Carbon Nanotubes. Environmental Science & Technology 2008, 42, (24), 9005-9013. 155. Park, J.-W.; Henry, T. B.; Menn, F.-M.; Compton, R. N.; Sayler, G., No bioavailability of 17 alpha-ethinylestradiol when associated with nC(60) aggregates during dietary exposure in adult male zebrafish (Danio rerio). Chemosphere 2010, 81, (10), 1227-1232. 156. Schwab, F.; Bucheli, T. D.; Camenzuli, L.; Magrez, A.; Knauer, K.; Sigg, L.; Nowack, B., Diuron Sorbed to Carbon Nanotubes Exhibits Enhanced Toxicity to Chlorella vulgaris. Environmental science & technology 2013, 47, (13), 7012-9. 157. Breuer, O.; Sundararaj, U., Big returns from small fibers: A review of polymer/carbon nanotube composites. Polymer Composites 2004, 25, (6), 630-645. 158. Andrews, R.; Jacques, D.; Minot, M.; Rantell, T., Fabrication of carbon multiwall nanotube/polymer composites by shear mixing. Macromolecular Materials and Engineering 2002, 287, (6), 395-403. 159. Safadi, B.; Andrews, R.; Grulke, E. A., Multiwalled carbon nanotube polymer composites: Synthesis and characterization of thin films. Journal of Applied Polymer Science 2002, 84, (14), 2660-2669. 160. Potschke, P.; Bhattacharyya, A. R.; Janke, A.; Goering, H., Melt mixing of polycarbonate/multi-wall carbon nanotube composites. Composite Interfaces 2003, 10, (4-5), 389-404. 161. Xie, X. L.; Mai, Y. W.; Zhou, X. P., Dispersion and alignment of carbon nanotubes in polymer matrix: A review. Materials Science & Engineering R-Reports 2005, 49, (4), 89-112. 162. Moniruzzaman, M.; Winey, K. I., Polymer nanocomposites containing carbon nanotubes. Macromolecules 2006, 39, (16), 5194-5205. 163. Spitalsky, Z.; Tasis, D.; Papagelis, K.; Galiotis, C., Carbon nanotube-polymer composites: Chemistry, processing, mechanical and electrical properties. Progress in Polymer Science 2010, 35, (3), 357-401. 164. Kingston, C.; Zepp, R.; Andrady, A.; Boverhof, D.; Fehir, R.; Hawkins, D.; Roberts, J.; Sayre, P.; Shelton, B.; Sultan, Y.; Vejins, V.; Wohlleben, W., Release characteristics of selected carbon nanotube polymer composites. Carbon 2014, 68, (0), 33-57. 165. Rahatekar, S. S.; Zammarano, M.; Matko, S.; Koziol, K. K.; Windle, A. H.; Nyden, M.; Kashiwagi, T.; Gilman, J. W., Effect of carbon nanotubes and montmorillonite on the flammability of epoxy nanocomposites. Polymer Degradation and Stability 2010, 95, (5), 870-879. 166. Schartel, B.; Braun, U.; Knoll, U.; Bartholmai, M.; Goering, H.; Neubert, D.; Potschke, P., Mechanical, thermal, and fire behavior of bisphenol a polycarbonate/multiwall carbon nanotube nanocomposites. Polymer Engineering and Science 2008, 48, (1), 149-158. 167. Kashiwagi, T.; Du, F. M.; Douglas, J. F.; Winey, K. I.; Harris, R. H.; Shields, J. R., Nanoparticle networks reduce the flammability of polymer nanocomposites. Nature Materials 2005, 4, (12), 928-933. 168. Ma, P. C.; Siddiqui, N. A.; Marom, G.; Kim, J. K., Dispersion and functionalization of carbon nanotubes for polymer-based nanocomposites: A review. Composites Part a-Applied Science and Manufacturing 2010, 41, (10), 1345-1367. 169. Potschke, P.; Bhattacharyya, A. R.; Janke, A., Melt mixing of polycarbonate with multiwalled carbon nanotubes: microscopic studies on the state of dispersion. European Polymer Journal 2004, 40, (1), 137-148. 170. Pegel, S.; Potschke, P.; Petzold, G.; Alig, I.; Dudkin, S. M.; Lellinger, D., Dispersion, agglomeration, and network formation of multiwalled carbon nanotubes in polycarbonate melts. Polymer 2008, 49, (4), 974-984. 171. Krause, B.; Bolcit, R.; Potschke, P., A method for determination of length distributions of multiwalled carbon nanotubes before and after melt processing. Carbon 2011, 49, (4), 1243-1247.
130
172. Lee, J.; Mahendra, S.; Alvarez, P. J. J., Nanomaterials in the Construction Industry: A Review of Their Applications and Environmental Health and Safety Considerations. Acs Nano 2010, 4, (7), 3580-3590. 173. Baur, J.; Silverman, E., Challenges and opportunities in multifunctional nanocomposite structures for aerospace applications. Mrs Bulletin 2007, 32, (4), 328-334. 174. Chen, L.; Hu, J.; Shen, X.; Tong, H., Synthesis and characterization of chitosan-multiwalled carbon nanotubes/hydroxyapatite nanocomposites for bone tissue engineering. Journal of Materials Science-Materials in Medicine 2013, 24, (8), 1843-1851. 175. Okamoto, M.; John, B., Synthetic biopolymer nanocomposites for tissue engineering scaffolds. Progress in Polymer Science 2013, 38, (10-11), 1487-1503. 176. Ghasemi, S.; Karami, H.; Khanezar, H., Hydrothermal synthesis of lead dioxide/multiwall carbon nanotube nanocomposite and its application in removal of some organic water pollutants. Journal of Materials Science 2014, 49, (3), 1014-1024. 177. Zoubos, H.; Koutsokeras, L. E.; Anagnostopoulos, D. F.; Lidorikis, E.; Kalogirou, S. A.; Wildes, A. R.; Kelires, P. C.; Patsalas, P., Broadband optical absorption of amorphous carbon/Ag nanocomposite films and its potential for solar harvesting applications. Solar Energy Materials and Solar Cells 2013, 117, 350-356. 178. Derraik, J., The pollution of the marine environment by plastic debris: a review. Marine Pollution Bulletin 2002, 44, (9), 842-852. 179. LAIST, D., OVERVIEW OF THE BIOLOGICAL EFFECTS OF LOST AND DISCARDED PLASTIC DEBRIS IN THE MARINE-ENVIRONMENT. Marine Pollution Bulletin 1987, 18, (6B), 319-326. 180. Yamashita, R.; Tanimura, A., Floating plastic in the Kuroshio Current area, western North Pacific Ocean. Marine Pollution Bulletin 2007, 54, (4), 485-488. 181. Gregory, M. R., Environmental implications of plastic debris in marine settings-entanglement, ingestion, smothering, hangers-on, hitch-hiking and alien invasions. Philosophical Transactions of the Royal Society B-Biological Sciences 2009, 364, (1526), 2013-2025. 182. Ryan, P. G.; Moore, C. J.; van Franeker, J. A.; Moloney, C. L., Monitoring the abundance of plastic debris in the marine environment. Philosophical Transactions of the Royal Society B-Biological Sciences 2009, 364, (1526), 1999-2012. 183. Costa, M. F.; Ivar do Sul, J. A.; Silva-Cavalcanti, J. S.; B. Araujo, M. C.; Spengler, A.; Tourinho, P. S., On the importance of size of plastic fragments and pellets on the strandline: a snapshot of a Brazilian beach. Environmental Monitoring and Assessment 2010, 168, (1-4), 299-304. 184. Andrady, A. L., Microplastics in the marine environment. Marine Pollution Bulletin 2011, 62, (8), 1596-1605. 185. Van Cauwenberghe, L.; Vanreusel, A.; Mees, J.; Janssen, C. R., Microplastic pollution in deep-sea sediments. Environmental Pollution 2013, 182, 495-499. 186. Desforges, J. P.; Galbraith, M.; Dangerfield, N.; Ross, P. S., Widespread distribution of microplastics in subsurface seawater in the NE Pacific Ocean. Mar Pollut Bull 2014, 79, (1-2), 94-9. 187. O'Brine, T.; Thompson, R. C., Degradation of plastic carrier bags in the marine environment. Marine Pollution Bulletin 2010, 60, (12), 2279-2283. 188. Cooper, D. A.; Corcoran, P. L., Effects of mechanical and chemical processes on the degradation of plastic beach debris on the island of Kauai, Hawaii. Marine Pollution Bulletin 2010, 60, (5), 650-654. 189. Zbyszewski, M.; Corcoran, P. L., Distribution and Degradation of Fresh Water Plastic Particles Along the Beaches of Lake Huron, Canada. Water Air and Soil Pollution 2011, 220, (1-4), 365-372. 190. Muller, C.; Townsend, K.; Matschullat, J., Experimental degradation of polymer shopping bags (standard and degradable plastic, and biodegradable) in the gastrointestinal fluids of sea turtles. Science of the Total Environment 2012, 416, 464-467. 191. Brandao, M. L.; Braga, K. M.; Luque, J. L., Marine debris ingestion by Magellanic penguins, Spheniscus magellanicus (Aves: Sphenisciformes), from the Brazilian coastal zone. Marine Pollution Bulletin 2011, 62, (10), 2246-2249. 192. Goldstein, M. C.; Rosenberg, M.; Cheng, L., Increased oceanic microplastic debris enhances oviposition in an endemic pelagic insect. Biology Letters 2012, 8, (5), 817-820.
131
193. Browne, M.; Niven, S.; Galloway, T.; Rowland, S.; Thompson, R., Microplastic Moves Pollutants and Additives to Worms, Reducing Functions Linked to Health and Biodiversity. Current Biology 2013, 23, (23), 2388-2392. 194. Choy, C. A.; Drazen, J. C., Plastic for dinner? Observations of frequent debris ingestion by pelagic predatory fishes from the central North Pacific. Marine Ecology Progress Series 2013, 485, 155-163. 195. LAIST, D., Overview of the biological effects of lost and discarded plastic debris in the marine- environment. Marine Pollution Bulletin 1987, 18, (6B), 319-326. 196. Possatto, F. E.; Barletta, M.; Costa, M. F.; do Sul, J. A.; Dantas, D. V., Plastic debris ingestion by marine catfish: an unexpected fisheries impact. Mar Pollut Bull 2011, 62, (5), 1098-102. 197. van Franeker, J. A.; Blaize, C.; Danielsen, J.; Fairclough, K.; Gollan, J.; Guse, N.; Hansen, P. L.; Heubeck, M.; Jensen, J. K.; Le Guillou, G.; Olsen, B.; Olsen, K. O.; Pedersen, J.; Stienen, E. W. M.; Turner, D. M., Monitoring plastic ingestion by the northern fulmar Fulmarus glacialis in the North Sea. Environmental Pollution 2011, 159, (10), 2609-2615. 198. Davison, P.; Asch, R. G., Plastic ingestion by mesopelagic fishes in the North Pacific Subtropical Gyre. Marine Ecology-Progress Series 2011, 432, 173-180. 199. Jantz, L. A.; Morishige, C. L.; Bruland, G. L.; Lepczyk, C. A., Ingestion of plastic marine debris by longnose lancetfish (Alepisaurus ferox) in the North Pacific Ocean. Marine Pollution Bulletin 2013, 69, (1-2), 97-104. 200. Rebolledo, E. L. B.; Van Franeker, J. A.; Jansen, O. E.; Brasseur, S. M. J. M., Plastic ingestion by harbour seals (Phoca vitulina) in The Netherlands. Marine Pollution Bulletin 2013, 67, (1-2), 200-202. 201. Browne, M. A.; Crump, P.; Niven, S. J.; Teuten, E.; Tonkin, A.; Galloway, T.; Thompson, R., Accumulation of Microplastic on Shorelines Woldwide: Sources and Sinks. Environmental Science & Technology 2011, 45, (21), 9175-9179. 202. Codina-Garcia, M.; Militao, T.; Moreno, J.; Gonzalez-Solis, J., Plastic debris in Mediterranean seabirds. Marine Pollution Bulletin 2013, 77, (1-2), 220-226. 203. Teuten, E. L.; Saquing, J. M.; Knappe, D. R. U.; Barlaz, M. A.; Jonsson, S.; Bjorn, A.; Rowland, S. J.; Thompson, R. C.; Galloway, T. S.; Yamashita, R.; Ochi, D.; Watanuki, Y.; Moore, C.; Pham, H. V.; Tana, T. S.; Prudente, M.; Boonyatumanond, R.; Zakaria, M. P.; Akkhavong, K.; Ogata, Y.; Hirai, H.; Iwasa, S.; Mizukawa, K.; Hagino, Y.; Imamura, A.; Saha, M.; Takada, H., Transport and release of chemicals from plastics to the environment and to wildlife. Philosophical Transactions of the Royal Society B-Biological Sciences 2009, 364, (1526), 2027-2045. 204. Koelmans, A. A.; Besseling, E.; Foekema, E. M., Leaching of plastic additives to marine organisms. Environ Pollut 2014, 187, 49-54. 205. Frias, J. P. G. L.; Sobral, P.; Ferreira, A. M., Organic pollutants in microplastics from two beaches of the Portuguese coast. Marine Pollution Bulletin 2010, 60, (11), 1988-1992. 206. Rochman, C.; Hoh, E.; Kurobe, T.; Teh, S., Ingested plastic transfers hazardous chemicals to fish and induces hepatic stress. Scientific Reports 2013, 3. 207. Holmes, L. A.; Turner, A.; Thompson, R. C., Adsorption of trace metals to plastic resin pellets in the marine environment. Environmental Pollution 2012, 160, 42-48. 208. Hirai, H.; Takada, H.; Ogata, Y.; Yamashita, R.; Mizukawa, K.; Saha, M.; Kwan, C.; Moore, C.; Gray, H.; Laursen, D.; Zettler, E. R.; Farrington, J. W.; Reddy, C. M.; Peacock, E. E.; Ward, M. W., Organic micropollutants in marine plastics debris from the open ocean and remote and urban beaches. Marine Pollution Bulletin 2011, 62, (8), 1683-1692. 209. Rochman, C. M.; Hoh, E.; Hentschel, B. T.; Kaye, S., Long-Term Field Measurement of Sorption of Organic Contaminants to Five Types of Plastic Pellets: Implications for Plastic Marine Debris. Environmental Science & Technology 2013, 47, (3), 1646-1654. 210. Engler, R. E., The Complex Interaction between Marine Debris and Toxic Chemicals in the Ocean. Environmental Science & Technology 2012, 46, (22), 12302-12315. 211. Teuten, E. L.; Rowland, S. J.; Galloway, T. S.; Thompson, R. C., Potential for plastics to transport hydrophobic contaminants. Environmental Science & Technology 2007, 41, (22), 7759-7764.
132
212. Rochman, C. M.; Manzano, C.; Hentschel, B. T.; Simonich, S. L. M.; Hoh, E., Polystyrene Plastic: A Source and Sink for Polycyclic Aromatic Hydrocarbons in the Marine Environment. Environmental Science & Technology 2013, 47, (24), 13976-13984. 213. Pandey, J. K.; Reddy, K. R.; Kumar, A. P.; Singh, R. P., An overview on the degradability of polymer nanocomposites. Polymer Degradation and Stability 2005, 88, (2), 234-250. 214. Wypych, G., Handbook of material weathering. 4th ed.; ChemTec Publishing: Toronto, Canada, 2008. 215. Nguyen, T.; Pellegrin, B.; Mermet, L.; Shapiro, A.; Gu, X.; Chin, J., Network Aggregation of CNT at the Surface of Epoxy/MWCNT Composite Exposed to UV Radiation. In Nanotechnology 2009: Fabrication, Particles, Characterization, MEMS, Electronics and Photonics, NIST, US: 2009; Vol. 1. 216. Hirth, S.; Cena, L.; Cox, G.; Tomovic, Z.; Peters, T.; Wohlleben, W., Scenarios and methods that induce protruding or released CNTs after degradation of nanocomposite materials. Journal of Nanoparticle Research 2013, 15, (4). 217. Petersen, E. J.; Lam, T.; Gorham, J. M.; Scott, K. C.; Long, C. J.; Stanley, D.; Sharma, R.; Alexander Liddle, J.; Pellegrin, B.; Nguyen, T., Methods to assess the impact of UV irradiation on the surface chemistry and structure of multiwall carbon nanotube epoxy nanocomposites. Carbon 2014, 69, (0), 194-205. 218. Wohlleben, W.; Meier, M. W.; Vogel, S.; Landsiedel, R.; Cox, G.; Hirth, S.; Tomovic, Z., Elastic CNT-polyurethane nanocomposite: synthesis, performance and assessment of fragments released during use. Nanoscale 2013, 5, (1), 369-380. 219. Golanski, L.; Guiot, A.; Pras, M.; Malarde, M.; Tardif, F., Release-ability of nano fillers from different nanomaterials (toward the acceptability of nanoproduct). Journal of Nanoparticle Research 2012, 14, (7). 220. Schlagenhauf, L.; Chu, B. T. T.; Buha, J.; Nuesch, F.; Wang, J., Release of Carbon Nanotubes from an Epoxy-Based Nanocomposite during an Abrasion Process. Environmental Science & Technology 2012, 46, (13), 7366-7372. 221. Som, C.; Berges, M.; Chaudhry, Q.; Dusinska, M.; Fernandes, T. F.; Olsen, S. I.; Nowack, B., The importance of life cycle concepts for the development of safe nanoproducts. Toxicology 2010, 269, (2-3), 160-169. 222. Eckelman, M. J.; Mauter, M. S.; Isaacs, J. A.; Elimelech, M., New Perspectives on Nanomaterial Aquatic Ecotoxicity: Production Impacts Exceed Direct Exposure Impacts for Carbon Nanotoubes. Environmental Science & Technology 2012, 46, (5). 223. Wohlleben, W.; Brill, S.; Meier, M. W.; Mertler, M.; Cox, G.; Hirth, S.; von Vacano, B.; Strauss, V.; Treumann, S.; Wiench, K.; Ma-Hock, L.; Landsiedel, R., On the Lifecycle of Nanocomposites: Comparing Released Fragments and their In-Vivo Hazards from Three Release Mechanisms and Four Nanocomposites. Small 2011, 7, (16), 2384-2395. 224. Lau, K. T.; Shi, S. Q., Failure mechanisms of carbon nanotube/epoxy composites pretreated in different temperature environments. Carbon 2002, 40, (15), 2965-2968. 225. Nguyen, T.; Pellegrin, B.; Bernard, C.; Gu, X.; Gorham, J. M.; Stutzman, P.; Stanley, D.; Shapiro, A.; Byrd, E.; Hettenhouser, R.; Chin, J., Fate of nanoparticles during life cycle of polymer nanocomposites. In Nanosafe2010: International Conference on Safe Production and Use of Nanomaterials, Journal ofPhysics:ConferenceSeries: 2011; Vol. 304. 226. Petersen, E.; Pinto, R.; Landrum, P.; Weber, W., Influence of Carbon Nanotubes on Pyrene Bioaccumulation from Contaminated Soils by Earthworms. Environmental Science & Technology 2009, 43, (11), 4181-4187. 227. Tessonnier, J.; Rosenthal, D.; Hansen, T.; Hess, C.; Schuster, M.; Blume, R.; Girgsdies, F.; Pfander, N.; Timpe, O.; Su, D.; Schlogl, R., Analysis of the structure and chemical properties of some commercial carbon nanostructures. Carbon 2009, 47, (7), 1779-1798. 228. Keller, A. A.; Wang, P.; Shi, Q. H.; Stucky, G., Enhanced environment mobility of carbon nanotubes in the presence of humic acid and removal from aqueous solution. Abstracts of Papers of the American Chemical Society 2009, 237. 229. ISO29664, Plastics - Artificial weathering including acidic deposition. In 2010b; p 16.
133
230. Kalbe, U.; Muller, W. W.; Berger, W.; Eckardt, J., Transport of organic contaminants within composite liner systems. Applied Clay Science 2002, 21, (1-2), 67-76. 231. Beamson, G.; Briggs, D., High resolution XPS of organic polymers. 1992. 232. Lu, K. L.; Lago, R. M.; Chen, Y. K.; Green, M. L. H.; Harris, P. J. F.; Tsang, S. C., Mechanical damage of carbon nanotubes by ultrasound. Carbon 1996, 34, (6), 814-816. 233. Nguyen, T.; Pellegrin, B.; Bernard, C.; Gu, X.; Gorham, J. M.; Stutzman, P.; Stanley, D.; Shapiro, A.; Byrd, E.; Hettenhouser, R.; Chin, J., Fate of nanoparticles during life cycle of polymer nanocomposites. In Nanosafe2010: InternationalConferenceonSafeProductionandUseofNanomaterials, Journal ofPhysics:ConferenceSeries: 2011; Vol. 304. 234. Nouara, A.; Wu, Q. L.; Li, Y. X.; Tang, M.; Wang, H. F.; Zhao, Y. L.; Wang, D. Y., Carboxylic acid functionalization prevents the translocation of multi-walled carbon nanotubes at predicted environmentally relevant concentrations into targeted organs of nematode Caenorhabditis elegans. Nanoscale 2013, 5, (13), 6088-6096. 235. Russier, J.; Menard-Moyon, C.; Venturelli, E.; Gravel, E.; Marcolongo, G.; Meneghetti, M.; Doris, E.; Bianco, A., Oxidative biodegradation of single- and multi-walled carbon nanotubes. Nanoscale 2011, 3, (3), 893-896. 236. Ging, J.; Tejerina-Anton, R.; Ramakrishnan, G.; Nielsen, M.; Murphy, K.; Gorham, J. M.; Nguyen, T.; Orlov, A., Development of a conceptual framework for evaluation of nanomaterials release from nanocomposites: Environmental and toxicological implications. Science of The Total Environment 2014, 473–474, (0), 9-19. 237. Shi, B.; Zhuang, X.; Yan, X.; Lu, J.; Tang, H., Adsorption of atrazine by natural organic matter and surfactant dispersed carbon nanotubes. Journal of Environmental Sciences-China 2010, 22, (8), 1195-1202. 238. Apul, O. G.; Shao, T.; Zhang, S. J.; Karanfil, T., Impact of carbon nanotube morphology on phenanthrene adsorption. Environmental Toxicology and Chemistry 2012, 31, (1), 73-78. 239. Wang, Z.; Zhao, J.; Song, L.; Mashayekhi, H.; Chefetz, B.; Xing, B., Adsorption and Desorption of Phenanthrene on Carbon Nanotubes in Simulated Gastrointestinal Fluids. Environmental Science & Technology 2011, 45, (14), 6018-6024. 240. Helland, A.; Wick, P.; Koehler, A.; Schmid, K.; Som, C., Reviewing the environmental and human health knowledge base of carbon nanotubes. Ciencia & Saude Coletiva 2008, 13, (2), 441-452. 241. Petersen, E. J.; Henry, T. B., Methodological considerations for testing the ecotoxicity of carbon nanotubes and fullerenes: Review. Environmental Toxicology and Chemistry 2012, 31, (1), 60-72. 242. Templeton, R. C.; Ferguson, P. L.; Washburn, K. M.; Scrivens, W. A.; Chandler, G. T., Life-cycle effects of single-walled carbon nanotubes (SWNTs) on an estuarine meiobenthic copepod. Environmental Science & Technology 2006, 40, (23), 7387-7393. 243. Cheng, C.; Muller, K. H.; Koziol, K. K. K.; Skepper, J. N.; Midgley, P. A.; Welland, M. E.; Porter, A. E., Toxicity and imaging of multi-walled carbon nanotubes in human macrophage cells. Biomaterials 2009, 30, (25), 4152-4160. 244. Llabjani, V.; Jones, K. C.; Thomas, G. O.; Walker, L. A.; Shore, R. F.; Martin, F. L., Polybrominated Diphenyl Ether-Associated Alterations in Cell Biochemistry as Determined by Attenuated Total Reflection Fourier-Transform Infrared Spectroscopy: a Comparison with DNA-Reactive and/or Endocrine-Disrupting Agents. Environmental Science & Technology 2009, 43, (9), 3356-3364. 245. Bi, X.; Walsh, M. J.; Wei, X.; Sheng, G.; Fu, J.; Martin-Hirsch, P. L.; Thomas, G. O.; Jones, K. C.; Martin, F. L., Infrared spectral analysis of MCF-7 cells treated with serum-lipid extracts segregates predominantly brominated flame retardant-exposed subjects from those with mainly organochlorine exposures. Environ Sci Technol 2007, 41, (16), 5915-22. 246. Riding, M. J.; Martin, F. L.; Trevisan, J.; Llabjani, V.; Patel, I. I.; Jones, K. C.; Semple, K. T., Concentration-dependent effects of carbon nanoparticles in gram-negative bacteria determined by infrared spectroscopy with multivariate analysis. Environmental pollution (Barking, Essex : 1987) 2012, 163, 226-34.
134
247. Bucholz, S.; Duff, D. G.; Michele, V.; Mleczko, L.; Muennich C.; RudolfR.; Wolf, A. Katalysator zur Herstellung von Kohlenstoffnanoröhrchen durch Zersetzung von gas-förmigen Kohlenverbindungen an einem heterogenen Katalysator, DE102004054959 A1. 2006. 248. OECD201, Freshwater Alga and Cyanobacteria, Growth Inhibition Test. In OECD Guideline: 2006. 249. Arnold, M. S.; Stupp, S. I.; Hersam, M. C., Enrichment of single-walled carbon nanotubes by diameter in density gradients. Nano Letters 2005, 5, (4), 713-718. 250. De Jonge, V. N., Quantitativeseparation of benthic diatoms from sedimentsusing density gradient centrifugation in the colloidal silica Ludox TM. Marine Biology (Berlin) 1979, 51, (3), 267-278. 251. Connel, D. W., Bioaccumulation of chemicals by aquatic organisms. In Ecotoxicology. Ecological Fundamentals, Chemical Exposure, and Biological Effects., Schüürmann G, M. B. e., Ed. Co-publication of John Wiley & Sons Inc., New York,
USA, and Spektrum Akademischer Verlag, Heidelberg, Germany: 1998; Vol. 14, pp 439-450. 252. Riding, M. J.; Martin, F. L.; Trevisan, J.; Llabjani, V.; Patel, I. I.; Jones, K. C.; Semple, K. T., Concentration-dependent effects of carbon nanoparticles in gram-negative bacteria determined by infrared spectroscopy with multivariate analysis. Environmental Pollution 2012, 163. 253. Llabjani, V.; Trevisan, J.; Jones, K. C.; Shore, R. F.; Martin, F. L., Binary Mixture Effects by PBDE Congeners (47, 153, 183, or 209) and PCB Congeners (126 or 153) in MCF-7 Cells: Biochemical Alterations Assessed by IR Spectroscopy and Multivariate Analysis. Environmental Science & Technology 2010, 44, (10), 3992-3998. 254. Martin, F. L.; German, M. J.; Wit, E.; Fearn, T.; Ragavan, N.; Pollock, H. M., Identifying variables responsible for clustering in discriminant analysis of data from infrared microspectroscopy of a biological sample. Journal of Computational Biology 2007, 14, (9). 255. Martin, F. L.; Kelly, J. G.; Llabjani, V.; Martin-Hirsch, P. L.; Patel, I. I.; Trevisan, J.; Fullwood, N. J.; Walsh, M. J., Distinguishing cell types or populations based on the computational analysis of their infrared spectra. Nature Protocols 2010, 5, (11), 1748-1760. 256. Kelly, J. G.; Trevisan, J.; Scott, A. D.; Carmichael, P. L.; Pollock, H. M.; Martin-Hirsch, P. L.; Martin, F. L., Biospectroscopy to metabolically profile biomolecular structure: a multistage approach linking computational analysis with biomarkers. Journal of Proteome Research 2011, 10, (4), 1437-1448. 257. Duygu, D.; Udoh, A. U.; Ozer, T.; Akbulut, A.; Erkaya, I.; Yildiz, K.; Guler, D., Fourier transform infrared (FTIR) spectroscopy for identification of Chlorella vulgaris Beijerinck 1890 and Scenedesmus obliquus (Turpin) Kützing 1833. African Journal of Biotechnology 2012, 11, (16), 3817-3824. 258. Sigee, D. C.; Bahram, F.; Estrada, B.; Webster, R. E.; Dean, A. P., The influence of phosphorus availability on carbon allocation and P quota in Scenedesmus subspicatus: A synchrotron-based FTIR analysis. Phycologia 2007, 46, (5), 583-592. 259. Dean, A. P.; Sigee, D. C.; Estrada, B.; Pittman, J. K., Using FTIR spectroscopy for rapid determination of lipid accumulation in response to nitrogen limitation in freshwater microalgae. Bioresource Technology 2010, 101, (12), 4499-4507. 260. Heraud, P.; Stojkovic, S.; Beardall, J.; McNaughton, D.; Wood, B. R., Intercolonial variability in macromolecular composition in p-starved and p-replete Scenedesmus populations revealed by infrared microspectroscopy. Journal of Phycology 2008, 44, (5), 1335-1339. 261. Sijm, D.; Broersen, K. W.; de Roode, D. F.; Mayer, P., Bioconcentration kinetics of hydrophobic chemicals in different densities of Chlorella pyrenoidosa. Environmental Toxicology and Chemistry 1998, 17, (9), 1695-1704. 262. Heraud, P.; Stojkovic, S.; Beardall, J.; McNaughton, D.; Wood, B. R., INTERCOLONIAL VARIABILITY IN MACROMOLECULAR COMPOSITION INP-STARVED AND P-REPLETE SCENEDESMUS POPULATIONS REVEALED BY INFRARED MICROSPECTROSCOPY. Journal of Phycology 2008, 44, (5), 1335-1339. 263. Zhang, S. J.; Jiang, Y. L.; Chen, C. S.; Creeley, D.; Schwehr, K. A.; Quigg, A.; Chin, W. C.; Santschi, P. H., Ameliorating effects of extracellular polymeric substances excreted by Thalassiosira pseudonana on algal toxicity of CdSe quantum dots. Aquatic Toxicology 2013, 126, 214-223.
135
264. Miao, A. J.; Schwehr, K. A.; Xu, C.; Zhang, S. J.; Luo, Z. P.; Quigg, A.; Santschi, P. H., The algal toxicity of silver engineered nanoparticles and detoxification by exopolymeric substances. Environmental Pollution 2009, 157, (11), 3034-3041. 265. Zhang, S. J.; Jiang, Y. L.; Chen, C. S.; Spurgin, J.; Schwehr, K. A.; Quigg, A.; Chin, W. C.; Santschi, P. H., Aggregation, Dissolution, and Stability of Quantum Dots in Marine Environments: Importance of Extracellular Polymeric Substances. Environmental Science & Technology 2012, 46, (16), 8764-8772. 266. Kostarelos, K.; Lacerda, L.; Pastorin, G.; Wu, W.; Wieckowski, S.; Luangsivilay, J.; Godefroy, S.; Pantarotto, D.; Briand, J. P.; Muller, S.; Prato, M.; Bianco, A., Cellular uptake of functionalized carbon nanotubes is independent of functional group and cell type. Nature Nanotechnology 2007, 2, (2), 108-113. 267. Kim, H. S.; Park, B. H.; Kang, M.; Yoon, J. S.; Jin, H. J., Characterization of polycarbonate/multiwalled carbon nanotube composites. In Experimental Mechanics in Nano and Biotechnology, Pts 1 and 2, Lee, S. B.; Kim, Y. J., Eds. Trans Tech Publications Ltd: Stafa-Zurich, 2006; Vol. 326-328, pp 1829-1832. 268. Tan, X.; Lin, C.; Fugetsu, B., Studies on toxicity of multi-walled carbon nanotubes on suspension rice cells. Carbon 2009, 47, (15), 3479-3487. 269. Porter, A. E.; Gass, M.; Bendall, J. S.; Muller, K.; Goode, A.; Skepper, J. N.; Midgley, P. A.; Welland, M., Uptake of Noncytotoxic Acid-Treated Single-Walled Carbon Nanotubes into the Cytoplasm of Human Macrophage Cells. Acs Nano 2009, 3, (6), 1485-1492. 270. Zhu, Y.; Ran, T. C.; Li, Y. G.; Guo, J. X.; Li, W. X., Dependence of the cytotoxicity of multi-walled carbon nanotubes on the culture medium. Nanotechnology 2006, 17, (18), 4668-4674. 271. Murdock, J. N.; Dodds, W. K.; Wetzel, D. L., Subcellular localized chemical imaging of benthic algal nutritional content via HgCdTe array FT-IR. Vibrational Spectroscopy 2008, 48, (2), 179-188. 272. Hirschmugl, C. J.; Bayarri, Z. E.; Bunta, M.; Holt, J. B.; Giordano, M., Analysis of the nutritional status of algae by Fourier transform infrared chemical imaging. Infrared Physics & Technology 2006, 49, (1-2), 57-63. 273. Murdock, J. N.; Wetzel, D. L., FT-IR Microspectroscopy Enhances Biological and Ecological Analysis of Algae. Applied Spectroscopy Reviews 2009, 44, (4), 335-361. 274. Domenighini, A.; Giordano, M., Fourier transform infrared spectroscopy of microalgae as a novel tool for biodiversity studies, species identification, and the assessment of water quality. Journal of Phycology 2009, 45, (2), 522-531. 275. Yang, K.; Zhu, L. Z.; Xing, B. S., Adsorption of polycyclic aromatic hydrocarbons by carbon nanomaterials. Environmental Science & Technology 2006, 40, (6), 1855-1861. 276. Chen, W.; Duan, L.; Zhu, D., Adsorption of polar and nonpolar organic chemicals to carbon nanotubes. Environmental Science & Technology 2007, 41, (24), 8295-8300. 277. Yang, K.; Xing, B. S., Adsorption of Organic Compounds by Carbon Nanomaterials in Aqueous Phase: Polanyi Theory and Its Application. Chemical Reviews 2010, 110, (10), 5989-6008. 278. Maes, H.; Stibany, F.; Giefers, S.; Daniels, B.; Deutschmann, B.; Baumgartner, W.; Schäffer, A., Accumulation and distribution of multiwalled carbon nanotubes in zebrafish (Danio rerio). In 2014. 279. Ternes, T. A.; Stumpf, M.; Mueller, J.; Haberer, K.; Wilken, R. D.; Servos, M., Behavior and occurrence of estrogens in municipal sewage treatment plants - I. Investigations in Germany, Canada and Brazil (vol 225, pg 81, 1999). Science of the Total Environment 1999, 228, (1), 87-87. 280. Xia, X. H.; Zhou, C. H.; Huang, J. H.; Wang, R.; Xia, N., Mineralization of phenanthrene sorbed on multiwalled carbon nanotubes. Environmental Toxicology and Chemistry 2013, 32, (4), 894-901. 281. OECD 305: Bioaccumulation in Fish: Aqueous and Dietary Exposure. In 1996. 282. OECD229, Fish Short Term Reproduction Assay. In OECD guideline for the testing of chemicals: 2012. 283. Maes, H.; Stibany, F.; Giefers, S.; Daniels, B.; Deutschmann, B.; Baumgartner, W.; Schäffer, A., Under construction: Accumulation and distribution of multiwalled carbon nanotubes in zebrafish (Danio rerio). In 2014. 284. Fenske, M.; van Aerle, R.; Brack, S.; Tyler, C. R.; Segner, H., Development and validation of a homologous zebrafish (Danio rerio Hamilton-Buchanan) vitellogenin enzyme-linked immunosorbent
136
assay (ELISA) and its application for studies on estrogenic chemicals. Comparative Biochemistry and Physiology C-Toxicology & Pharmacology 2001, 129, (3), 217-232. 285. Yang, K.; Xing, B. S., Desorption of polycyclic aromatic hydrocarbons from carbon nanomaterials in water. Environmental Pollution 2007, 145, (2), 529-537. 286. Titus, E.; Ali, N.; Cabral, G.; Gracio, J.; Babu, P. R.; Jackson, M. J., Chemically functionalized carbon nanotubes and their characterization using thermogravimetric analysis, Fourier transform infrared, and Raman spectroscopy. Journal of Materials Engineering and Performance 2006, 15, (2), 182-186. 287. Su, Y.; Yan, X. M.; Pu, Y. B.; Xiao, F.; Wang, D. S.; Yang, M., Risks of Single-Walled Carbon Nanotubes Acting as Contaminants-Carriers: Potential Release of Phenanthrene in Japanese Medaka (Oryzias latipes). Environmental Science & Technology 2013, 47, (9), 4704-4710. 288. Lerman, I.; Chen, Y. N.; Xing, B. S.; Chefetz, B., Adsorption of carbamazepine by carbon nanotubes: Effects of DOM introduction and competition with phenanthrene and bisphenol A. Environmental Pollution 2013, 182, 169-176. 289. Van den Belt, K.; Verheyen, R.; Witters, H., Reproductive effects of ethynylestradiol and 4t-octylphenol on the zebrafish (Danio rerio). Archives of Environmental Contamination and Toxicology 2001, 41, (4), 458-467. 290. Lögers, L. The influence of dissolved organic carbon and carbon nanotubes on the bioavailability of 17alpha-ethinylestradiol to the zebrafish (Danio rerio). RWTH Aachen University, Aachen, 2010. 291. Martinez, D. S. T.; Alves, O. L.; Barbieri, E.; Iop In Carbon nanotubes enhanced the lead toxicity on the freshwater fish, 3rd International Conference on Safe Production and Use of Nanomaterials (Nanosafe), Minatec, Grenoble, FRANCE, 2013
Nov 13-15, 2012; Minatec, Grenoble, FRANCE, 2012. 292. Yu, X.-Y.; Luo, T.; Zhang, Y.-X.; Jia, Y.; Zhu, B.-J.; Fu, X.-C.; Liu, J.-H.; Huang, X.-J., Adsorption of Lead(II) on O-2-Plasma-Oxidized Multiwalled Carbon Nanotubes: Thermodynamics, Kinetics, and Desorption. Acs Applied Materials & Interfaces 2011, 3, (7), 2585-2593. 293. Park, J.-W.; Henry, T. B.; Ard, S.; Menn, F.-M.; Compton, R. N.; Sayler, G. S., The association between nC(60) and 17 alpha-ethinylestradiol (EE2) decreases EE2 bioavailability in zebrafish and alters nanoaggregate characteristics. Nanotoxicology 2011, 5, (3), 406-416. 294. Barnes, D. K. A.; Galgani, F.; Thompson, R. C.; Barlaz, M., Accumulation and fragmentation of plastic debris in global environments. Philosophical Transactions of the Royal Society B-Biological Sciences 2009, 364, (1526), 1985-1998. 295. Boateng, L. K.; Heo, J.; Flora, J. R. V.; Park, Y. G.; Yoon, Y., Molecular level simulation of the adsorption of bisphenol A and 17 alpha-ethinyl estradiol onto carbon nanomaterials. Separation and Purification Technology 2013, 116, 471-478. 296. Hoekstra, E. J.; Simoneau, C., Release of Bisphenol A from Polycarbonate-A Review. Critical Reviews in Food Science and Nutrition 2013, 53, (4), 386-402. 297. Sajiki, J.; Yonekubo, J., Leaching of bisphenol A (BPA) to seawater from polycarbonate plastic and its degradation by reactive oxygen species. Chemosphere 2003, 51, (1), 55-62. 298. Shvedova, A. A.; Pietroiusti, A.; Fadeel, B.; Kagan, V. E., Mechanisms of carbon nanotube-induced toxicity: Focus on oxidative stress. Toxicology and Applied Pharmacology 2012, 261, (2), 121-133. 299. Koelmans, A.; Besseling, E., EM, Leaching of plastic additives to marine organisms. Environmental Pollution 2014, 187, 6. 300. Flint, S.; Markle, T.; Thompson, S.; Wallace, E., Bisphenol A exposure, effects, and policy: A wildlife perspective. Journal of Environmental Management 2012, 104, (0), 19-34. 301. Staples, C. A.; Dorn, P. B.; Klecka, G. M.; O'Block, S. T.; Harris, L. R., A review of the environmental fate, effects, and exposures of bisphenol A. Chemosphere 1998, 36, (10), 2149-2173. 302. Chow, W. S.; Chan, W. K. L.; Chan, K. M., Toxicity assessment and vitellogenin expression in zebrafish (Danio rerio) embryos and larvae acutely exposed to bisphenol A, endosulfan, heptachlor, methoxychlor and tetrabromobisphenol A. Journal of Applied Toxicology 2013, 33, (7), 670-678.
137
138