Acute toxicity to Daphnia magna in river water; Investigating … · 2017-10-18 ·...
Transcript of Acute toxicity to Daphnia magna in river water; Investigating … · 2017-10-18 ·...
Maria Andersson
Degree project for Master of Science 45 hec
Department of Biological and Environmental SciencesUniversity of Gothenburg 2012
Acute toxicity to Daphnia magna in river water; Investigating mitigation and bioavailability of pure cationic surfactants and mixtures with SPME
Summary Many surfactants are evaluated according to REACH and a key component in exposure assessment is
fate, which is influenced by various factors in the environment that can strongly reduce the toxicity
observed in the laboratory. Toxicity is e.g. mitigated by their tendency to interact with natural
organic matter (NOM) via hydrophobic interactions, but also electrostatically. Thus to determine the
toxic potential of a surfactant, a quantification of the freely dissolved concentration, i.e. the
bioavailable fraction, is necessary. AkzoNobel Surface Chemistry AB in Stenungsund, Sweden,
supported this study with the aim to investigate the bioavailability and thereby the true acute
toxicity of seven pure cationic surfactants and mixtures to Daphnia magna in river water using Solid-
Phase Micro Extraction (SPME). A method where polyacrylate-coated fibers are added to the acute
immobilization test (OECD 202) and the amount of sorbed surfactant on the fibers is directly
proportional to the freely dissolved concentration. The most toxic substances in this study were
hexadecylamine+2EO and didodecyldimethylammonium bromide, whereas the least toxic substances
were Ethomeen C/12 and dodecylamine+2EO. Toxicity is increasing for primary fatty amine
ethoxylates with the chain length increasing from 12 to 16 carbon atoms, caused by an increasing
hydrophobicity within the molecule. Sorption increases with increasing amount of NOM but the
mitigating effect is substance specific due to different sorption affinities and varies between 0.9 and
31.3 in this study. A general mitigation factor cannot be used, as the true toxicity will be either over-
or underestimated. Different sorption affinities of individual mixture components to NOM also
affects the composition of Ethomeen C/12, hence the mixture toxicity. The predicted mixture toxicity
is overestimated with Concentration Addition in all test media but the overestimation decreases with
increasing amount of NOM due to the altered composition.
Sammanfattning Många tensider utvärderas enligt REACH och en viktig del i exponeringsbedömningen är ämnets öde,
som påverkas av olika faktorer i miljön som till stor del kan minska observerad toxicitet i laboratoriet.
Toxiciteten kan t.ex. mildras genom deras benägenhet att interagera med naturligt organiskt material
(NOM) via hydrofoba interaktioner, men även elektrostatiska. Så för att bestämma den potentiella
toxiciteten hos en tensid krävs en kvantifiering av den fritt lösta koncentrationen, det vill säga den
biotillgängliga fraktionen. Denna studie stöddes av AkzoNobel Surface Chemistry AB i Stenungsund,
Sverige, i syfte att undersöka biotillgängligheten och därmed den sanna akuta toxiciteten av sju rena
katjoniska tensider och blandningar på Daphnia magna i flodvatten med Solid-Phase Micro
Extraction (SPME). En metod där polyakrylatbelagda fibrer tillsätts i det akuta immobiliseringstestet
(OECD 202) och mängden sorberad tensid på fibrerna är direkt proportionell mot den fritt lösta
koncentrationen. De giftigaste ämnena i denna studie var hexadecylamine+2EO och
didodecyldimethylammonium bromide, medan de minst giftiga ämnena var Ethomeen C/12 och
dodecylamine+2EO. Toxiciteten ökar för primära fettaminetoxylater då kedjelängden ökar från 12 till
16 kolatomer, som orsakas av ökad hydrofobicitet inom molekylen. Sorptionen ökar med ökande
mängd NOM men den mildrande effekten är ämnesspecifik på grund av olika sorptionsaffiniteter för
NOM och varierar mellan 0.9 och 31.3 i denna studie. En generell mildringsfaktor kan inte användas
eftersom den sanna toxiciteten kommer då antingen att över- eller underskattas. Olika affinitet för
sorption till NOM för enskilda blandningskomponenter påverkar även sammansättningen av
Ethomeen C/12, därmed blandningens toxicitet. Den predikterade blandningstoxiciteten överskattas
med Concentration Addition i alla testmedier men överskattningen minskar med ökad mängd NOM
till följd av den förändrade sammansättningen.
Table of contents
SUMMARY...........................................................................................................................................
SAMMANFATTNING ............................................................................................................................
TABLE OF CONTENTS ...........................................................................................................................
1. INTRODUCTION ......................................................................................................................... 1
1.1 SURFACTANTS.................................................................................................................................... 1
1.2 ENVIRONMENTAL FATE – WHAT IS BIOAVAILABLE? ................................................................................... 3
1.3 TESTED CATIONIC SURFACTANTS ........................................................................................................... 7
1.4 AIM ................................................................................................................................................. 8
1.5 HYPOTHESIS AND QUESTIONS ............................................................................................................... 8
2. MATERIALS AND METHODS........................................................................................................... 8
2.1 LITERATURE SEARCH ........................................................................................................................... 8
2.2 EXPERIMENTS .................................................................................................................................... 9
2.2.1 Chemicals................................................................................................................................ 9
2.2.2 Sampling and characterization of river water........................................................................ 9
2.2.3 Acute toxicity test with Daphnia magna .............................................................................. 10
2.2.4 Bioavailability test with Solid-Phase Micro Extraction (SPME) ............................................ 11
2.2.5 Statistical and mathematical calculations ........................................................................... 13
2.3 PROBLEMS ENCOUNTERED DURING THE EXPERIMENTS ............................................................................ 13
3. RESULTS & DISCUSSION................................................................................................................14
3.1 NOMINAL AND MEASURED CONCENTRATION......................................................................................... 14
3.1.1 Relationship between nominal and measured concentration.............................................. 14
3.1.2 Difference in nominal and measured concentration ............................................................ 15
3.1.3 Toxicity comparison between substances ............................................................................ 16
3.2 TEST MEDIA..................................................................................................................................... 20
3.2.1 Factor difference between different test media................................................................... 21
3.2.2 Changed conductivity in river water (HD to HD600) ............................................................ 26
3.3 DEGREE OF ETHOXYLATION ................................................................................................................ 27
3.4 TOXIC RESPONSE AS A FUNCTION OF THE ALKYL CHAIN LENGTH................................................................. 28
3.5 SINGLE SUBSTANCES AND MIXTURE TOXICITY......................................................................................... 32
3.6 FURTHER RECOMMENDATIONS ........................................................................................................... 37
4. CONCLUSIONS ..............................................................................................................................39
ACKNOWLEDGEMENTS ....................................................................................................................40
5. REFERENCES .................................................................................................................................41
APPENDIX A: CALCULATIONS OF TU FOR SINGLE SUBSTANCES AND MIXTURE ................................47
APPENDIX B: PHYSICO-CHEMICAL PROPERTIES AND TOXICITY DATA FOR TESTED SURFACTANTS...50
APPENDIX C: SPECIFICATIONS FOR NATURAL RIVER WATER AND DUTCH STANDARD WATER ........57
APPENDIX D: PREPARATIONS, RAW DATA AND RESULTS FROM TOXCALC. V5.0.23.........................59
APPENDIX E: PREPARATION OF CALIBRATION CURVES FOR SPME...................................................84
APPENDIX F: RELATIONSHIP BETWEEN NOMINAL AND MEASURED CONCENTRATION....................89
1
1. Introduction
1.1 Surfactants
A wide range of products and applications used by consumers and industry of today’s society contain
surface-active agents, or shortly surfactants. The use ranges from primary production processes to
enhancing the quality of finished products, hence surfactants appear in products such as motor oils,
pharmaceuticals, cosmetics, detergents, drilling muds and flotation agents and in recent decades also
in electronic printing, magnetic recording, biotechnology, microelectronics and viral research (Rosen
and Kunjappu, 2012). However, about 54% of the use is in different household products, including
detergents, fabric softeners, cosmetics and sanitizers (Banat et al., 2000; Rust and Wildes, 2008). In
1993, the annual world production of synthetic surfactants amounted to 7.2 million tons (Di Corcia,
1998) and in 2008, the annual production was 13 million tons (Reznik et al., 2010) and was expected
to increase by 2.8% annually until 2012 and 3.5-4% thereafter (Acmite, 2010).
Synthetic surfactants are economically important chemicals (Ying, 2006) and the main reason for this
is their ability to modify surface and interfacial properties between liquids, solids and gases. These
properties reside in their amphiphilic character, i.e. they generally contain a hydrophobic (nonpolar)
tail and a hydrophilic (polar, charged or uncharged) head (fig. 1). The chemical structure of
surfactants are not restricted to the simple schematic illustration shown in figure 1 but varies widely,
which gives them their different characteristics (Holmberg et al., 2003).
Figure 1. A schematic illustration of surfactant monomers and a micelle.
When a surfactant with an amphiphilic structure is dissolved in an aqueous solution, they prefer to
migrate to surfaces or interfacial regions. This is because the hydrophobic group is incapable of
hydrogen bonding and thus disrupts the normal water structure. As a consequence, the interfacial
tension or surface tension of the system is increased, which is defined as the interfacial free energy
per unit area of the boundary between two different phases (Holmberg et al., 2003; Rosen and
Kunjappu, 2012). By orientation of the hydrophilic group towards the aqueous phase and the
hydrophobic groups away from it, the interfacial tension is reduced and the normal water structure is
restored. Hence, surfactants are concentrating at the interfaces separating immiscible phases (Haigh,
1996) and by lowering the interfacial tension of the medium in which it is dissolved, two different
media or interfaces are able to mix or disperse readily as emulsions in water or other liquids
(Holmberg et al., 2003).
2
Surfactants are present as monomers when dissolved in an aqueous solution at low concentrations.
As the concentration of surfactant increases, the interface will eventually be saturated. At higher
concentrations micelles will be formed, i.e. aggregation of surfactants (fig. 1), when the hydrophobic
groups are oriented towards the center of the micelle and the hydrophilic groups towards the
aqueous phase. This aggregation occurs at a surfactant concentration called the critical micelle
concentration (CMC) (Holmberg et al., 2003) and varies with surfactant structure and solution
chemistry, e.g. temperature, presence of electrolytes and various organic compounds. In general, the
CMC decreases as the hydrophobic character of the surfactant increases and when electrolytes are
present (Haigh, 1996). Concentrations above the CMC enables surfactants to solubilise more of a
hydrophobic organic compound compared to what would dissolve in water alone (Haigh, 1996;
Roberts, 2000), thus reducing the interfacial tension that has increased due to the presence of
organic compounds (Holmberg et al., 2003).
Surfactants are represented in different forms but normally classified according to the presence of
formally charged groups on the hydrophilic moiety. These different types include cationic, anionic,
non-ionic and zwitterionic surfactants (Holmberg et al., 2003). Even though each surfactant have
unique properties and characteristics some common characteristics can be attributed to each class.
Anionic surfactants bears a negative charge, usually due to a sulphonate or sulphate group, and they
are used in detergents due to their detersive action and efficiency to remove particulate soils. This
benefit is possible due to the fact that anions are not prone to sorb to negatively charged substrates,
such as particulate soils, thereby hindering redeposition of undesirable soils on fabrics’ etcetera
(Rosen and Kunjappu, 2012). Anionic surfactants are also the largest surfactant class, with
approximately 60% of the world production, due to their ease and low cost of manufacture
(Holmberg et al., 2003). Non-ionic surfactants are the second largest surfactant class and contain no
ionic constituent and are thus compatible with charged molecules, e.g. ionic surfactants that result in
beneficial associations. They have also very low sensitivity to water hardness and pH, which makes
them very useful in liquid and powder detergents and to stabilize oil-in-water emulsions (Holmberg
et al., 2003; Rosen and Kunjappu, 2012).
Cationic surfactants bears a formal positive charge and thus adsorbs strongly onto most substrates in
the environment, e.g. metals, minerals, plastics, fibres, cell membranes etcetera, which are generally
negatively charged. This changes the surface properties and makes a hydrophilic surface behave as if
it was hydrophobic and vice versa, and thus impart special characteristics to the surface. Cationic
surfactants are the third largest surfactant class and are used as conditioning agents in fabric
softeners and hair care products, as corrosion inhibitors of metals in fuel and lubricating oils and as
anticaking agents in fertilizers. The smallest surfactant class is zwitterionic surfactants which may
have both positively and negatively charged moieties within the same molecule. They have their
optimal surface activity around neutral pH, hence they are used in personal care products (shower
gels, foam baths, shampoos, etc.) for their mildness and skin compatibility. They are often used
together with anionic or non-ionic surfactants to enhance properties such as foam or detergency
(Holmberg et al., 2003; Rosen and Kunjappu, 2012).
Differences in the nature of the hydrophobic group, which generally consists of long-chain
hydrocarbon residues, are also important for the properties and the characteristics of surfactants but
less pronounced than for the hydrophilic group. These structures includes differences in the length of
the alkyl group, branching and unsaturation, presence of an aromatic nucleus, polyoxypropylene or
3
polyoxyethylene and perfluoroalkyl or polysiloxane groups (Rosen and Kunjappu, 2012). Surfactants
are produced from petrochemical (synthetic) and/or oleochemical (renewable) feedstocks. The
petrochemical feedstocks are mainly derived from crude oil and converted to different surfactant
intermediates whereas oleochemical feedstocks are commonly derived from plant oil (palm and
coconut), plant carbohydrates (sorbitol, sucrose and glucose) and animal fat (tallow) (Holmberg et
al., 2003; Rust and Wildes, 2008).
1.2 Environmental fate – what is bioavailable?
Considering the widespread use and high consumption of surfactants and due to the fact that they
are mainly used in household products, such as laundry detergents, fabric softeners and hair care
products, they will be discharged to sewage treatment plants or directly to surface waters (Ying,
2006). Inevitably, aquatic organisms are exposed to different types of surfactants and their
degradation products at various concentrations in different environmental compartments. The total
surfactant concentration in wastewater may reach 10 mg/L in areas where it is extensively used,
although the aqueous concentration are below a few tens of µg/L (WHO, 1996). Some reported
concentrations for cationic surfactants, such as ditallow dimethylammonium chloride (DTDMAC), are
37 µg/L in river water, 334 µg/L in influent wastewater and 28 µg/L in effluents from sewage
treatment plants (Wee, 1984), 60 µg/L in surface waters (Versteeg et al., 1992) and up to 5870 mg/kg
in dry treated sewage sludge (Fernandez et al., 1996). Alkyltrimethylammonium compounds have
measured concentrations ranging from 361 to 6750 mg/kg in sediments, where the highest
concentration was observed in samples affected by effluents from wastewater treatment plants
(Lara-Martín et al., 2010). Dimethyldiesterarylammonium chloride have been measured in effluents
from wastewater treatment plants up to 503 µg/L (Barco et al., 2003).
Given a high enough concentration and a sufficient
length of time, a chemical and/or its metabolites
that come into contact with an organism and react
at an appropriate target site(s) will elicit an
adverse response or toxic effect. The effect is
concentration-dependent and this relationship
(fig. 2) varies with the chemical and species of
organism. To express and measure the toxicity of a
certain chemical to aquatic organisms, different
end points are used, e.g. the median effect
concentration (EC50). EC50 is the concentration
estimated to produce a certain effect, e.g.
immobility, in 50% of a test population over a
specific time period (Rand et al., 1995).
To improve the protection of human health and
the environment, all chemical substances that are produced within or imported to the European
market above 1 ton per year has to be assessed for its intrinsic properties. This is according to the
European legislation of chemicals, REACH (Registration, Evaluation, Authorisation and restriction of
Chemicals), which entered into force 1 June 2007 (Europa.eu, 2011). Ecotoxicological information is
gathered through exposure and effect assessments where tests are performed with standard test
organisms from at least three different trophic levels (algae, Daphnia and fish). The organisms are
Figure 2. A typical form of the concentration-response relationship.
EC50=0.1113 mg/L
4
exposed to the substance during a short (hours to a few days) or a longer (generally several days or
weeks) period of time to evaluate potential hazardous properties and possible acute or chronic
effects (ECHA, 2011). The freshwater micro crustacean Daphnia magna is included in the ecological
risk assessment and used in acute immobilization tests because they are a primary food source for
many fish species and convert phytoplankton and bacteria into animal protein, thus an ecologically
important species (Cooney, 1995). They have also been shown to be the most sensitive species to
some detergent chemicals according to Lewis and Suprenant (1983).
The mechanism of action of surfactants is widely believed to be narcotic, i.e. the toxicity is
dependent on the ability of the surfactant to partition from the aqueous environment into lipid
membranes of aquatic organisms (Rosen et al., 2001). Two different narcosis mechanisms have been
recognized and are based on log Pow (P=octanol/water partition coefficient) (Roberts and Castello,
2003) or log Kmw (membrane-water partition coefficient) (Robert and Castello, 2003:a). The first is
general narcosis developed by Könemann (1981) where the substance act by a non-specific
mechanism and is generally as toxic as their hydrophobicity indicates, i.e. a baseline toxicity. The
second is polar narcosis, developed by Saarikoski and Viluksela (1982), and accounts for polar
contributions to binding to membranes as the predicted baseline toxicity is generally lower than the
observed (Roberts and Costello, 2003:a). Toxicity is also related to bioavailability, which is the freely
available fraction of the surfactant that possibly can cross an organism’s cellular membrane from the
medium surrounding the organism (Semple et al., 2004). Cationic surfactants are found to be more
toxic than anionic surfactants, and anionic surfactants are more toxic than non-ionic surfactants. In
general, toxicity increases with an increase in the length of the hydrophobic group and decreases
with branching (Rosen and Kunjappu, 2012). EC50 values below 1 mg/L after a 48 h test with D.
magna and 96 h test with fish and algae are considered to be toxic (Holmberg et al., 2003).
Aquatic toxicity data are available for surfactants on different organisms, although the toxic effects
are more evaluated for anionic, e.g. linear alkylbenzene sulphonic acid (LAS), and non-ionic
surfactants, e.g. alcohol ethoxylate (AE), according to Ivankovic and Hrenovic (2010). For cationic
surfactants, aquatic toxicity data is available but less evaluated for their environmental fate and toxic
effects. Different quaternary ammonium compounds (QAC) exposed to several fish species have
reported EC50-48h values ranging between 0.49 and 8.24 mg/L (Singh et al., 2002).
Alkyltrimethylammonium compounds exposed to D. magna, such as cetyl trimethylammonium
chloride have LC50-48h (lethal concentration) ranging between 0.025-0.05 mg/L (Lewis and
Suprenant, 1983), whereas dodecyl-, tetradecyl- and hexadecyl trimethylammonium bromide have
reported EC50-24h of 0.37, 0.091 and 0.058 mg/L, respectively by Sandbacka et al. (2000) and 0.38,
0.14 and 0.13 mg/L, respectively by García et al. (2001). García et al. (2001) also showed that
substitution of a benzyl group for a methyl group appears to slightly increase the toxicity to D. magna
and reported EC50-24h values of 0.13, 0.13 and 0.22 mg/L for dodecyl benzyl dimethyl ammonium
bromide, tetradecyl benzyl dimethyl ammonium chloride and hexadecyl benzyl dimethyl ammonium
chloride, respectively. Arquad 2C-75 have reported LC50-96h for fish ranging between 0.26 and 0.787
mg/L, an LC50-48h of 0.295 mg/L for crustacean and EC50-72h for algae ranging between 0.06 and
0.386 mg/L (ECHA, 2012: CAS 68391-05-9). Clearly, the effect concentration is below 1 mg/L for the
most sensitive species D. magna and these values are all based on nominal concentrations, except
the highest mentioned toxicity data for Arquad 2C-75 on algae.
5
However, the risk assessment of the surfactant is to a large extent based on these laboratory studies
where the tested chemical is dissolved in a pure liquid media (OECD, 2004; van Wijk et al., 2009) and
the effect is then extrapolated to the real environment (TGD, 2003). A key component in exposure
assessment is fate, i.e. the concentration, transport, transformation and disposition of a surfactant
(Lyman, 1995), and that is influenced by various factors in the aquatic environment that can strongly
reduce the toxicity observed in the lab (Haigh, 1996; Alexander, 2000). Due to physical and chemical
properties of the surfactant, such as the molecular structure and the nature of structural groups
(amphiphilic structure), they have a tendency to form aggregates and a propensity to interact with
natural particles (Jones-Hughes and Turner, 2005). This will reduce their toxicity, i.e. mitigate their
effect.
Thus, sorption to natural organic matter (NOM) is an important property to consider regarding
surfactants as they can at low concentrations in natural water exists in either or both the dissolved
and the sorbed phase (Lyman, 1995). NOM is a complex mixture of compounds with different particle
sizes that can be separated into particulate, colloidal and dissolved fractions. Their functional groups
are diverse and have a broad range of interaction with surfactants, hence controls bioavailability and
toxicity. Humic acid is one of the most abundant components of the colloidal fraction of NOM
(Koopal et al., 2005), considered to be structured polyelectrolytes with an amphiphilic character
(Guetzloff and Rice, 1994) and soluble in aqueous solutions in a wide pH range and thus easily
transported in the aqueous environment (Koopal et al., 2004).
The impact of sorption is included in the environmental risk assessment for hydrophobic nonpolar
chemicals where the organic carbon/water partition coefficient (Koc) or the octanol/water partition
coefficient (Kow) can be used to describe the sorption to organic matter and subsequent reduced
bioavailability (TGD, 2003; van Wijk et al., 2009). However, the sorption of cationic surfactants to
natural organic matter is not only described by hydrophobic interaction and measured values are
therefore necessary (TGD, 2003). Depending on the aqueous properties, such as pH, salinity,
temperature and amount of suspended material (Rand et al., 1995), different sorption mechanisms
are potentially involved for ionic surfactants, such as ion exchange, ion pairing and hydrophobic
bonding (Jones-Hughes and Turner, 2005; Rosen and Kunjappu, 2012). The hydrophobic chains of
cationic surfactants binds to the organic fraction of suspended matter and of humic acid through van
der Waals forces, whereas the positively charged nitrogen group binds electrostatically to the
negatively charged binding sites of the sorbents, hence both hydrophobic and electrostatic attraction
are involved (Koopal et al., 2004; van Wijk et al., 2009). Surfactants differ in their hydrophobicity as
well as how much that is charged at a specific pH. The hydrophobic binding of surfactants to
substrates are assumed to concur with the equilibrium partition theory, whereas the electrostatic
interaction (ionic) is governed by other parameters not included in this theory (Thomas et al., 2009).
To account for sorption of cationic surfactants a quantification of the freely dissolved concentration
is necessary as this determines the toxic potential of a surfactant (Rufli et al., 1998) and not the
surfactants that are strongly sorbed to colloidal phases. It is the freely dissolved concentration that
controls evaporation, sorption, precipitation, biodegradation, bioconcentration and toxicity (Rico-
Rico et al., 2009) and a quantification of this provides information about the bioavailability and thus
the potential risk of cationic surfactants in the environment. The freely dissolved concentration is
measured with the method Solid-Phase Micro Extraction (SPME). It is a sampling technique with
polyacrylate-coated fibers that utilize the ion-exchange capacity of the fibers to sorb chemical
6
substances. The fibers are equilibrated for 24 hours in a test vessel and the concentration of
chemicals on the fibers is directly proportional to the freely dissolved concentration by applying a
compound specific fiber-water partitioning coefficient (Kfw). The SPME method began with
hydrophobic compounds and in recent years, the application of SPME has extended and includes also
more polar and ionized compounds. Difficulties with the calibration of SPME for ionic organics are
that the partitioning is influenced by the solution chemistry (pH, salinity, type of counter ions, etc.)
(Rico-Rico et al., 2009) and that they have an affinity to the test container. With optimized
experimental conditions, SPME calibration isotherms have been made for anionic and non-ionic
surfactants and all of them were linear at concentrations below their critical micelle concentration
(CMC) (Droge et al., 2007; Rico-Rico et al., 2009). At last, a few cationic surfactants have been tested
and evaluated with this technique (Chen et al., 2010).
Furthermore, the number of chemicals produced in today’s society is increasing and to perform
ecotoxicological tests on all of them are expensive, time consuming and raise questions about ethics.
Predictions of their environmental behaviour, effect and fate by a model is thus necessary.
Development of alternative hazard assessments are e.g. promoted by REACH (2006). Surfactants are
present as pure individual substances but also as mixtures of e.g. different carbon chain lengths and
structural groups, and these variations are numerous. Instead of testing every possible mixture
combination, the mixture toxicity can be predicted if the toxicity and the concentration of the
individual substances within the mixture are known.
One concept is Concentration Addition (CA) where the concentrations of the single substances are
added to yield the toxicity of the mixture. This predictive model is applied to substances believed to
have a similar mode of action described by Porsbring (2009). At first, each single substance in the
mixture is scaled to a common effect level, i.e. a toxic unit (TU) (see appendix A). The TU of a single
substance is the ratio between their concentration in the mixture and their effect concentration (e.g.
EC50) when tested individually. Addition of the single TUs gives the TU of the mixture and the
mixture conforms to CA when the TUs are equal, i.e. 1. However, if the addition of the single TUs will
be less than 1, their joint toxicity is greater than additive and a lower mixture concentration than
expected by CA is required to provoke an effect. Conversely, less than additive if the TU of the
mixture is higher than 1.
With all this in mind, the toxicity of surfactants are obvioulsy affected by several factors in the
aquatic environment. Therefore, a quantification of the bioavailable fraction with SPME is necessary
to describe their true toxicity and the focus is on cationic surfactants. The study was performed in
collaboration with AkzoNobel Surface Chemistry AB, in Stenungsund Sweden and Arnhem, the
Netherlands. AkzoNobel is a multinational chemical corporation headquartered in Amsterdam, the
Netherlands, which supplies industries and consumers worldwide with decorative paints,
performance coatings and specialty chemicals. They have operations in more than 80 countries and
employs around 55 000 people. Their approach is to find innovative solutions and sustainable
answers to customers (AkzoNobel, 2012). REACH put a greater responsibility on the companies to
evaluate substances and thus increase the competitiveness of the chemicals industry within
European Union (Europa.eu, 2011). Therefore, AkzoNobel’s main objective is to develop more
environmentally friendly surfactants.
7
1.3 Tested cationic surfactants
Fatty amines and their derivatives are examples of cationic surfactants, produced either from
synthetic or renewable feedstocks, where AkzoNobel Surface Chemistry is the world's leading
supplier. These cationic surfactants are based on alkyl groups ranging from carbon chain lengths C8 to
C22, with C12 to C18 chain lengths the most predominant (AkzoNobel, 2012). Available physico-
chemical properties for the seven tested surfactants are found in appendix B.
Primary, secondary, tertiary alkyl amines and their salts
(RNH3+X-) are uncharged and insoluble in water at a high pH
and therefore, not strictly cationic (Holmberg et al., 2003).
Dodecylamine (abbreviated C12) is a pure primary fatty
amine with a C12 carbon chain length (fig. 3). It has a pKa of
10.63 and is cationic at a pH below this value. Primary alkyl
amines sorb strongly to solid phases by van der Waals
forces and ionic interactions (e.g. ion pair formation and cation exchange). Dodecylamine is used for
manufacturing of primary alkyl amines, formulation of fuel additives, lubricants, coating agents for
fertilizer and products in textile industry, production of ethoxylates of primary alkyl amines, amine
derivatives, amides, as metal corrosion inhibitor, antistatic agents and rubber additive and flotation
agent in mining industry (ECHA, 2012:a).
The amine can be ethoxylated to yield an ethoxylated
amine. These surfactants can be cationic or non-ionic,
depending on the degree of ethoxylation and on the pH at
which they are used. They are considered as cationic
surfactants when the pH is low enough to provide the ionic
form. Ethoxylated amines are water-soluble over a large pH
range due to the fact that the ethoxylation degree mainly
governs the hydrophilic character of the fatty amine
(Holmberg et al., 2003). Dodecylamine +2EO,
hexadecylamine +2EO (fig. 4) and octadecylamine +2EO, abbreviated C12+2EO, C16+2EO and
C18+2EO respectively in this report, are pure primary fatty amine ethoxylates (PFAEO). They have
two ethoxylates attached to the amine and an alkyl chain length of C12, C16 and C18 carbon,
respectively. They have a pKa of 8.6 (Chen et al., 2012) and is therefore cationic under the test
conditions in this study. Ethomeen C/12 is a mixture of different fatty acid chain lengths, mainly C12
and C14 (appendix B), with two ethoxylates attached to the amine. The pKa is 8.8, hence it is cationic
under test conditions in this study. It is used in applications as pigment processing additives and as
thickening agents in polar solvents (AkzoNobel, 2011). It is also used in cosmetic products, cleaning
and care products, lubricants and greases, plastic articles and as corrosion protection (ECHA, 2012:b).
Quaternary ammonium compounds (QAC) contain a positively
charged nitrogen atom linked to four alkyl or aryl substituent’s
and the positive charge is permanent, regardless of pH (Rosen
and Kunjappu, 2012). Didodecyldimethylammonium bromide
(abbreviated DDAB) has two alkyl chains with C12 carbon
respectively, and two methyl groups attached to the amine. The
Figure 3. Chemical structure of dodecylamine.
Figure 4. Chemical structure of hexadecylamine +2EO.
Figure 5. Chemical structure of didodecyldimethylammonium bromide.
8
counter ion is bromide (fig. 5).
Arquad 2C-75 has two hydrophobic hydrocarbon chains, with
carbon chain lengths varying from C12 to C18 respectively, but
mainly C12 and C14 (AkzoNobel, 2012:a). The other two
substituents are methyl groups. They are all linked to a
positively charged nitrogen atom (fig. 6). It is used in industrial
settings and by professional workers for treatment of minerals,
application and manufacture of metal treatment products,
coatings (organic solvent-borne, water-borne, solvent-free
products and powder coatings), manufacturing of washing and cleaning products, cosmetic products
and application of agricultural and agro products. The use by consumers is mainly by application of
cosmetic products (ECHA, 2012:c).
1.4 Aim
The aim of this project was to investigate the bioavailability and thereby the true acute toxicity of
pure cationic surfactants and mixtures to Daphnia magna in river water using the SPME technique.
1.5 Hypothesis and questions
The hypothesis is that the toxicity of these surfactants is to a large extent determined by their
hydrophobicity due to a narcotic mechanism of action. Additionally, toxicity is also influenced by
their ability to also interact electrostatically with biological surfaces due to their cationic charge.
• Is there a difference between nominal and measured concentrations? If so, why?
• Which of the tested cationic surfactants is the most and least toxic ones, and why?
• What is the mitigation factor for these surfactants? Is it the same mitigation for all cationic
surfactants, i.e. is it possible to use a standard mitigation factor?
• How does carbon chain length affect toxicity? Is the response only a function of alkyl chain
length?
• How does the degree of ethoxylation affect toxicity?
• Is there a difference between single substances and mixtures regarding toxicity? Does the
cationic mixture conform to the predictive model Concentration Addition?
2. Materials and methods
2.1 Literature search
A literature search were performed for all the tested surfactants at the website of European
Chemicals Agency (ECHA, 2012) to gather physico-chemical and ecotoxicological information, by
searching on the individual CAS numbers. Aquatic toxicity data was also obtained at the ECOTOX
database (U.S. EPA, 2012). The scientific databases Web of Knowledge, Scopus, Sciencedirect and
Google Scholar were used to gather available scientific information about the tested surfactants and
related surfactants. The same databases were also used to find information about surfactants and
their environmental fate for the introduction.
Figure 6. Chemical structure of Arquad 2C-75.
9
2.2 Experiments
Experiments were performed during nine weeks from February to April 2012 at AkzoNobel
Ecotoxicology and Environmental Testing lab in Arnhem, the Netherlands. The focus was on
documenting and studying the bioavailability and the acute aquatic toxicity of pure cationic
surfactants and mixtures. The used methods were the Daphnia sp. Acute Immobilization test (OECD
202) and the SPME technique (Solid-Phase Micro Extraction).
2.2.1 Chemicals
Surfactants, fibers and NOM
Dodecylamine (C12), purity ≥99.5% and Didodecyldimethylammonium bromide, purity ≥98% from
Fluka Chemie GmbH, Switzerland. Arquad 2C-75, Dodecylamine (pure) +2EO, Hexadecylamine (pure)
+ 2EO, Octadecylamine + 2EO (pure) and Ethomeen C/12 from AkzoNobel Surface Chemistry AB,
Stenungsund, Sweden. Polyacrylate coated SPME fibers (30 µm: FSA110170 and 7 µm: FSA110124 5,
15) from Polymicro Technologies, Phoenix, Arizona US (www.polymicro.com). Humic acid (EC: 215-
809-6, CAS: 1415-93-6) from Sigma-Aldrich.
Test medium
The tests were performed in four different test media. First test medium was Dutch Standard Water
(DSW), having a pH of approximately 8.2, and conductivity between 550 and 650 µS/cm. It contains
per liter of de-ionized water: NaHCO3 [100 mg], CaCl2·2H2O [200 mg], MgSO4·7H2O [180 mg] and
KHCO3 [20 mg]. Second test medium was DSW with humic acid (HA) [20 mg/L] added, other
characteristics are the same as previous DSW. Third test medium was river water (HD) containing
suspended matter [2.4 mg/L] and humic acid with a conductivity of 283 µS/cm and a pH of 7.8.
Fourth test medium was HD water with DSW salts added to achieve a conductivity between 550 and
650 µS/cm. It contains per liter of HD water: NaHCO3 [50 mg], CaCl2·2H2O [100 mg], MgSO4·7H2O [90
mg] and KHCO3 [10 mg]. The dissolved oxygen and pH was measured and adjusted, if necessary, to
achieve an oxygen concentration >7 mg/L and a pH of 8.2 (±0.2).
Culture medium
Culturing media for D. magna were M4. It is based on concentrated stock mineral salt solutions
supplemented with vitamins. It was prepared by adding the stock solutions to de-ionized water
preferably one day before the animals were introduced. The vitamins were added to the culture
medium immediately before use. Following salts with final concentration in mg/L were used in M4:
CaCl2·2H2O [293.8], MgSO4·7H2O [123.3], NaHCO3 [64.8], KCl [5.8], MnCl2·4H2O [0.36], LiCl [0.31],
RbCl [0.071], SrCl2·6H2O [0.152], CuCl2·2H2O [0.017], ZnCl2 [0.013], CoCl2·6H2O [0.010], H3BO3 [2.86],
NaBr [0.016], KI [0.0033], Na2SeO3 [0.0022], FeSO4·7H2O [0.9955], Na2EDTA·2H2O [2.5],
Na2MoO4·2H2O [0.063], NH4VO3 [0.0006], NaSiO3·9H2O [10], NaNO3 [0.274], KH2PO4 [0.143] and
K2HPO4 [0.184]. Following vitamins with final concentration in mg/L are included in M4: thiamine
hydrochloride (B1) [0.075], cyanocobalamine (B12) [0.001] and biotin [0.00075].
D. magna were also cultured in HD water with the following vitamins and final concentration in mg/L
in HD water: thiamine hydrochloride (B1) [0.075], cyanocobalamine (B12) [0.001] and biotin [0.00075].
Culture medium, both M4 and HD, were renewed twice a week, every Tuesday and Friday.
2.2.2 Sampling and characterization of river water
The natural surface water used as test medium and culture medium is river water (abbreviated HD
from Heveadorp). It is sampled from a specific sample location in Heveadorp at Fonteinallee,
10
Doorwerth (Gelderland) with GPS coordinates: 51° 58’ 10.29” N, 5° 48’ 9.35” E (appendix C). The
sample point is situated in a ground water protection area under management from water company
Vallei Eem and the Gelderland province. There is no agriculture in the area and therefore no
concerns regarding pesticide use in the area. The water source has also been analysed for dissolved
heavy metals but there are no cause of concerns regarding this. The water is described of exceptional
quality and diverse in flora and fauna. The river water (HD) has a total suspended solids-particulate
matter (TSS) concentration of 2.4 mg/L. It was measured by filter 1 litre of river water through a 45
µm filter with known weight, then placed in the oven at 105°C for 24 hours and then weights the
filter again. The total organic carbon (TOC) concentration is 2.21 mg/L. The conductivity is 283 µS/cm
and has a pH of 7.8. The Ca2+ concentration is 34.3 mg/L (see appendix C).
2.2.3 Acute toxicity test with Daphnia magna
The toxicity tests for all tested substances were performed according to Organisation for Economic
Co-operation and Development Guideline 202 (OECD, 2004), a Daphnia sp. acute immobilisation test
with exposure duration of 48 hours. The acute toxicity to Daphnia magna is usually expressed as the
median effective concentration for immobilization. This is the concentration, which immobilizes 50 %
of the animals in a test batch within a period of continuous exposure (EC50). Furthermore, the
concentration causing no significant immobility (NOEC) and the lowest concentration causing
significant immobility in comparison to the control was determined (LOEC), if possible.
Test species
The test animals used in the acute toxicity tests were D. magna (water flea), taken from a stock
cultured in M4 and HD water. They were grown in 3 L beakers covered with glass plates and
contained about 2.5 L medium and the room temperature was between 18 to 23 °C. They were fed
with 2 ml of algae (Chlorella vulgaris, Pseudokircherinella subcapitata or Scenedesmus subspicatus)
six days per week and received feed equivalent to approximately 0.1 mg carbon per daphnia per day.
The animals used in the test were less than 24 hours old and obtained from parent animals aged
between 2 and 4 weeks. The day before the start of the test, the suitable group of test animals were
sieved in the afternoon (around 4.00 p.m.) to remove the juveniles. On the day of the test, the same
group were sieved in the morning (around 8.00 a.m.) again and the juveniles were collected in a dish
with dilution water. Animals cultured in M4 were collected in DSW and animals cultured in HD were
collected in HD water.
Test procedures
The test was performed as a static test for 48 hours with a light regime of 16 hours of ambient light
and 8 hours of darkness. A total of 20 animals divided into 4 batches of 5 animals in 200 ml of test
medium were tested at each concentration and in the control. Those animals that were not able to
swim within 15 seconds after gentle agitation of the test vessel were considered to be immobile and
were recorded. The number of animals being trapped at the surface was determined. These animals
were not regarded as immobile. The test vessels were not aerated during the test and the animals
were not fed. Glass beakers (test vessels) were covered with glass after introducing daphnia in them.
The test was inspected at 0, 24 and 48 hours.
Preparations of solutions/suspensions of the test substance
All test substances were soluble in water (specifications for each test substance, see appendix D). A
stock solution of approximately 100 mg/l was prepared by loading approximately 0.0100 gram of the
11
test substance, weighed out on an analytical balance and then filled up to the appropriate volume
(100 ml) with de-ionized water to achieve a 100 mg/L stock solution. The solution was then stirred or
sonicated whilst on ice (if required) for maximum two minutes until a homogenous solution was
formed. The pH was checked and adjusted with sodium hydroxide (1 M) or hydrochloric acid (1 M) if
required to approximately 8.2.
Test concentrations
To minimise contamination from previous tests, all glassware were rinsed in methanol and de-
ionized water prior to be used in the new test. Preliminary tests (range finding) were conducted for
all the tested substances with the following standard concentrations: 0.01, 0.1, 1.0 and 10 mg/L to
determine the range of concentrations for the definitive test. 100 ml of test solutions divided in two
test vessels and 200 ml of test medium (control) divided in four test vessels with 5 daphnids in each
were used.
For the definitive test, test solutions were prepared on the day of the test in 200 ml volumetric flasks
by diluting the stock solution in test media to achieve five test concentrations in a geometric series
with a separation factor not exceeding 2.2. The highest test concentration resulted in 100 per cent
immobilisation and the lowest test concentration resulted in no observable effect, compared to the
control. Controls containing only test medium was also included in the test.
Determination of dissolved oxygen, pH and temperature
The dissolved oxygen and pH were determined in the test vessels and adjusted, if necessary, before
the start (t=0h) of the test in the highest and lowest test concentrations and in the control. It was
also determined at the end of the test (t=48h). The temperature was also measured at the beginning
and at the end of the test.
2.2.4 Bioavailability test with Solid-Phase Micro Extraction (SPME)
Preparation of SPME fibers
The polyacrylate (PA) coated fibers used in the tests had a glass core of 110 µm diameter and a
thickness of either 7 (mainly ionic interaction) or 30 µm (mainly hydrophobic interaction), depending
on the test substance (table 1). Gloves were used to avoid contamination of the fibers while cutting
them in a length of 3.4 centimetres. Subsequently, they were activated by heating them up in GC
Oven 8000 series (Fisons instruments) with a helium flow of 30 ml/min and a temperature of 120°C
for at least 16 hours, then changed to a temperature of 60°C for at least two hours. After heating,
they were placed in a vial with de-ionized water for minimum 24 hours before they were used in the
test.
The CEC for the 7 µm PA fiber is much higher than for the 30 µm PA fiber (Chen et al., 2010), thus
used for the surfactants that are always positively charged (Arquad 2C-75) and where the pKa of the
substance is much higher than at the tested pH (dodecylamine). DDAB (QAC) were tested with the 30
µm due to a misunderstanding, but it is still possible to measure it based on hydrophobicity although
with less sensitivity. The 30 µm fiber were tested with the remaining substances and have a higher
affinity of the neutral species than for the ionized (cationic) species. Still, the calibrated fiber-water
isotherm at a certain pH reflects the freely dissolved concentration at that pH. Below the pKa of the
substance, which is the case for these substances, the 30 µm fiber extracts relatively much of the
neutral species from the solution and equilibrates with this low concentration, but this is instantly
12
replenished by the speciation constant of the compound at the solution pH and still represent the
total of freely dissolved cationic/neutral species at a given pH.
Test procedures
Approximately 24 hours after Daphnia immobility test start, four SPME fibers were added with
tweezers into two of the four replicates for each test concentration, including the control, and left to
equilibrate. After 24 hours, the SPME fibers were removed with tweezers, dried and placed in HPLC
vials. Subsequently, they were cut in three pieces and 1 ml of a mobile phase (table 1) was added to
each vial. Furthermore, 0.75 ml of the middle test concentration was transferred with a pipette to
vials already containing 0.75 ml of leaching solution. The vials were closed and analysed with Liquid
Chromatography/Mass Spectrometry (LC/MS).
Table 1. Mobile phases used for LC/MS and thickness of fibers used for each tested surfactant.
Mobile phase Tested substances Thickness of fiber (µm)
90:10 Methanol + 2% formic acid: H2O + 2% formic acid
Dodecylamine 7
Didodecyldimethylammonium bromide
30 50:50 Methanol:H2O + 0.65 ml TFA + 0.75 ml NH3 + 1.15 ml CH3COOH
Arquad 2C-75 7 Dodecylamine +2EO 30 Hexadecylamine +2EO 30 Octadecylamine +2EO 30
50:50 Methanol:2-propanol + 2 ml TFA + 2 ml NH3 + 2.5 ml CH3COOH
Ethomeen C/12 30
Preparations of solutions and calibration curves of the test substance
Calibration curve
A leaching solution was prepared with approximately 100 grams of MgCl2·6H2O weight into a 1 liter
Erlenmeyer flask and then added 500 ml of methanol and 2-propanol, respectively. The content was
shaken until the salt was dissolved completely. A stock solution of the test substance was prepared
with leaching solution as dilution media. Test solutions for the calibration curve were prepared by
diluting the stock solution in leaching solution to achieve eight concentrations in a range from 0 to
maximum 1500 µg/L. Control containing only leaching solution was also included in the calibration
curve (appendix E).
SPME calibration curve
A stock solution was prepared by loading an accurate amount of the test substance, weighed out on
an analytical balance and then filled up with the appropriate volume with de-ionized water. Test
solutions for the SPME calibration curve were prepared by diluting the stock solution in regular DSW
and modified DSW (45 % of the salts added, see appendix C) to achieve eight concentrations ranging
from 0 to maximum 1500 µg/L depending on the surfactant (appendix E). Two SPME fibers were
added with tweezers into each beaker for each test concentration, including the control, and left to
equilibrate. The fibers were removed after 24 hours using the same procedure as for the Daphnia
immobility (toxicity) test. Furthermore, 0.75 ml of the test solutions in the test vessels was
transferred with a pipette to vials already containing 0.75 ml of leaching solution. Subsequently, the
vials were closed and analyzed with Liquid Chromatography/Mass Spectrometry (LC/MS).
13
2.2.5 Statistical and mathematical calculations
The software ToxCalc v5.0.23 was used to calculate EC50, NOEC and LOEC. The method Trimmed
Spearman-Karber gave EC50 and the method Williams’ test gave NOEC and LOEC. The raw data from
the SPME analysis was treated in Microsoft Excel. TUs for the single substances and the mixture and
the predicted EC50 for the mixture are calculated with the equations given in appendix A.
2.3 Problems encountered during the experiments
A few problems were encountered during the experiments. One was floating daphnid’s due to a
different surface tension in the test vessel compared to the culture medium. An internal test,
following the OECD 202, was performed to see whether the temperature, test media or if the
beakers were contaminated or not (methanol wash) could have an influence. Four different test
media were used: M4, handmade DSW, tank DSW and HD water and the test was performed in three
different rooms: daphnia test room, daphnia culture room and lumbriculous room, with different
room temperatures. No significant results were obtained. Next hypothesis was if the test media was
aerated too much because the daphnid’s had air bubbles underneath the shell. Floating is not a huge
problem in an acute toxicity test since the animals are not fed during the test and therefore, they do
not have to swim around and eat algae to survive. In addition, the surfactant concentrates at
interfacial regions and the surface where the floating daphnid’s are and the acute toxicity is exerted
anyway. However, the floating seemed to be caused by the use of plastic pipettes as they most likely
released something into the water and thereby changed the surface tension. At higher test
concentrations the number of floating daphnid’s were less, probably because the surfactant lowered
the surface tension. Problem was solved by using glass pipettes. Another problem was that the
daphnid’s cultured in M4 had a lower survival rate and was much smaller and more pale compared to
the ones cultured in HD water. This resulted in high mortality in the DSW control during the tests and
tests had to be repeated.
Problem was also encountered with the SPME fibers. Normally, 1-2 % of the chemical in the test
vessel sorb to the fiber and therefore do not influence the toxicity to the test animals. The sorption
of Ethomeen O/12 was around 30% and for C16+2EO it was around 10%. Toxicity was thus altered
and the test was repeated with two additional beakers for each tested concentration (without test
animals) where the SPME fibers were placed. See also discussion “problems with determination of
truly dissolved concentration”.
14
3. Results & discussion
3.1 Nominal and measured concentration
3.1.1 Relationship between nominal and measured concentration
The ratio between nominal and measured concentration is presented in appendix F and is nearly
constant (linear) over the entire concentration range for all the substances where it was possible to
measure the freely dissolved concentration. Hence low residual variability and the model fits the
data very well. The lowest R2 value (0.884) has C16+2EO as a single substance in Ethomeen C/12
(measured with the LC/MS) in the test media DSW+HA. A linear relationship indicates that the
calibration curve (sorption isotherm) for the SPME fibers are effective over the concentration range,
hence the sorption of the surfactant to the fibers is directly proportional to the freely dissolved
concentration in the test vessels. A linear relationship also concludes that the concentration range is
below the CMC for the substances.
Problems with determination of truly dissolved concentration
Nevertheless, the truly dissolved concentration measured with SPME fibers couldn’t be determined
for dodecylamine (C12) and octadecylamine +2EO (C18+2EO) in this study. Other cationic
surfactants; Lilaflot D817M, hexadecylamine (C16), Armeen T, Ethomeen HT and Ethomeen O/12,
were aimed to be tested but due to low or no sorption to the SPME fibers in the calibration curve
they were not tested for their nominal toxicity either, although a range finding were performed on all
of them. These substances are more hydrophobic and less water-soluble than other cationic
surfactants tested in this study. Water solubility might be an explanation to why some substances
have difficulties with SPME, although not confirmed in this study. However, Lilaflot D817M should be
able to measure with SPME but the tested batch seemed to be old.
The sorption of truly dissolved dodecylamine to the SPME fibers in the calibration curve at an
aqueous concentration up to 25 µg/L is not significant, and at an aqueous concentration of 75 µg/L it
is slightly higher but still low. In the Daphnia immobility test, the sorption of truly dissolved
dodecylamine to the SPME fibers at the lowest nominal concentrations (0.03-0.06 mg/L in DSW and
0.1 mg/L in HD) are also not significant and it is first when the nominal concentration is 0.12 mg/L in
DSW and 0.2 mg/L in HD that the aqueous concentration corresponds to about 25 µg/L and thus
possible to sorb onto the fiber. Therefore, the true toxicity (EC50) couldn’t be calculated since the
true effect range of dodecylamine is within this low concentration range. Low sorption in this study
may be due to; 1) high sorption to the test vessels and/or daphnid´s instead of the SPME fibers, 2)
different type of fibers compared to the ones used by Utrecht university, i.e. another type of
activation might be necessary, 3) the test medium contains Ca2+ and other divalent cations which
strongly competes with the ion-exchange affinity of the cationic species to the fiber, 4) the
desorption volume, and 5) not applying a column in the LC/MS that separates the compound from
the “noise” eluting from the fibers. However, the Utrecht University in the Netherlands have
determined the aqueous detection limit for the application of SPME on dodecylamine to be 1.0 µg/L
(Chen et al., 2012). Further tests with dodecylamine and SPME is necessary to make the working
range of the SPME to cover the effect range of the substance, i.e. optimize the test conditions and
15
the analytics, so it is possible to fully evaluate the mitigating factor. The 30 µm fiber may be used
instead, as it extracts the small neutral fraction but it is less dependent on the electrolytes in the
solution, although dependent on the pH. One alternative to minimize loss of substance to other
surfaces is to precondition the glassware with the test substance (Rufli et al., 1998).
Determination of a SPME calibration isotherm for octadecylamine +2EO (C18+2EO) resulted in a non-
linear relationship. Above an aqueous concentration of approximately 65 µg/L for this substance, the
SPME fiber is saturated and the fiber concentration analyzed with LC/MS reached a maximum value
at approximately 30-40 µg/L. As a consequence, there will be an underestimation of the freely
dissolved concentration of C18+2EO at higher aqueous concentrations. Hence, the measured
concentrations and the true EC50 is going to be less reliable as the uncertainty around the truly
dissolved concentration of C18+2EO is increasing with increasing aqueous concentrations.
3.1.2 Difference in nominal and measured concentration
Figure 7. Nominal and measured concentration (EC50 in mg/L) in DSW for all tested surfactants. Measured EC50 is missing for dodecylamine (C12) and octadecylamine (C18+2EO) due to problems with the SPME (see section 3.1.1.).
The nominal and measured EC50 (mg/L) in DSW of all tested surfactants are presented in figure 7. A
comparison between nominal and measured concentration in DSW, where no suspended matter or
humic acid is present, clearly shows the “surface-acting” behavior of these substances. This
difference is mostly due to their strong tendency to adsorb to pipettes, glassware, Daphnia and other
surfaces during preparation of test concentrations and running of the test during 48 hours.
Table 2. Factor difference between nominal and measured EC50 in DSW for all tested surfactants.
Substance Factor
difference
Dodecylamine (C12) - Dodecylamine +2EO (C12+2EO) 1.856 Hexadecylamine +2EO (C16+2EO) 14.842 Octadecylamine +2EO (C18+2EO) - Ethomeen C/12 1.183 Didodecyldimethylammonium bromide (DDAB) 6.206 Arquad 2C-75 4.556
16
Furthermore, the difference is also distinctive between substances. The difference in nominal and
measured concentration is smallest for Ethomeen C/12 and C12+2EO, whereas for C16+2EO, DDAB
and Arquad 2C-75 the difference is larger (table 2). For Ethomeen C/12 the measured concentration
is even higher than the nominal. One hypothesis regarding this difference in sorption is related to
their hydrophobicity, i.e. increasing sorption with an increase in hydrophobicity of the molecule. An
increase of carbon atoms in the alkyl chain length is related to an increased hydrophobicity, hence
increasing sorption. It has for example been reported by Duman and Ayranci (2010) that tested
several cationic surfactants and found that hydrophobic interactions appeared to determine the
adsorption to activated carbon cloth (ACC), where an increase in carbon chain length were reflected
in an increased sorption to ACC.
The hydrophobicity in this study is based on log Kow (partitioning between octanol/water) modelled
with U.S. EPI Suite (2011). An increased value reflects a higher hydrophobicity, hence a higher
migration (or sorption) to surfaces or interfaces. The values for C12+2EO, C16+2EO and C18+2EO are
3.9, 5.86 and 6.85, respectively. The higher log Kow of C16+2EO explains the higher sorption
compared to C12+2EO in this study. Assuming the hypothesis is correct, the higher log Kow of
C18+2EO and the correlation should thus result in an even higher sorption compared to C16+2EO.
Unfortunately, the measured concentration is missing for C18+2EO and the hypothesis cannot be
confirmed in this study for these surfactants. The modelled log Kow for DDAB is 6.62 and that
surfactant also adsorb strongly to surfaces.
Ethomeen C/12 is a mixture of different carbon chain lengths, but mainly consists of C12+2EO
(≥50%), and has a log Pow value of 0.7 (AkzoNobel, 2011). It is a substance that show a weaker
tendency to adsorb to surfaces in the same way as C12+2EO, compared to Arquad 2C-75 which has a
log Pow of 4.8. More data on nominal and measured toxicity in standard water for a larger set of
cationic surfactants are necessary to be able to make any conclusions regarding hydrophobicity in
this case.
3.1.3 Toxicity comparison between substances
The toxicity (EC50 in mg/L) of the seven tested surfactants in four different test media are presented
in figure 8 based on nominal concentrations and in figure 9 for measured concentrations. The
nominal concentrations required to immobilise 50% of the D. magna population after 48h of
exposure (EC50) ranged from 0.026 to 1.61 mg/L whereas the required SPME derived aqueous
(measured) concentrations (EC50-48h) ranged from 0.0019 to 0.87 mg/L.
17
Nominal concentration
Figure 8. Nominal log EC50 (mg/L) for all tested substances in four different test media. The EC50 for C18+2EO in DSW+HA (orange) is an estimated value based on 60% mobile daphnid’s at 1.25 mg/L. All surfactants are tested in in DSW and HD, some of them in HD600 and DSW+HA (see “test media”).
Regarding nominal concentrations, the most toxic substance in DSW is C18+2EO (EC50=0.0264 mg/L),
whereas the least toxic substance is C12+2EO (EC50=0.681 mg/L). In the test media HD, the most
toxic substance is DDAB (EC50=0.107 mg/L) and the least toxic is once again C12+2EO (EC50=1.612
mg/L). Only two and five out of total seven surfactants were tested in HD600 and DSW+HA,
respectively (explanation see discussion “test media”). In DSW+HA, the most toxic was C16+2EO
(EC50=0.433 mg/L) and the least toxic was C12+2EO (EC50=1.131 mg/L), based on definitive results.
If the estimated EC50 for C18+2EO (1.3 mg/L) is taken into account, it is thus regarded as the least
toxic surfactant.
The nominal EC50 to D. magna in DSW is below 1 mg/L for all seven surfactants, in contrast to the
test media HD and DSW+HA where the highest EC50 value is around 1.6 mg/L (in HD). Based on 95%
confidence interval, the toxicity is not statistically different for C16+2EO, C18+2EO and DDAB in DSW
and C16+2EO, DDAB and Arquad 2C-75 in HD. C12 and Ethomeen C/12 are also not statistically
different in HD. In DSW+HA, the toxicity of C12+2EO, Ethomeen C/12 and DDAB is not statistically
different whereas Arquad 2C-75 is not statistically different from Ethomeen C/12 and DDAB.
In conclusion, the least toxic substance regarding nominal concentration is C12+2EO, regardless of
test media. Whereas the most toxic substance varies depending on test media, although C16+2EO
18
can be regarded as one the most toxic substance since the toxicity is not statistically different from
C18+2EO in DSW and DDAB in HD and still the most toxic substance in DSW+HA.
Measured concentration
Figure 9. Measured log EC50 (mg/L) for five tested substances in four different test media. Measured concentrations are missing for C12 and C18+2EO due to problems with the SPME. All surfactants are tested in DSW and HD, some of them in HD600 and DSW+HA (see “test media”).
Measured concentrations are missing for a few substances due to problems with the SPME (see
“problems with determination of truly dissolved concentration”). Regarding the SPME derived
aqueous (measured) concentrations (fig. 9), the most toxic substance in DSW is C16+2EO
(EC50=0.0019mg/L) whereas the least toxic substance is Ethomeen C/12 (EC50=0.389 mg/L). In HD,
the most toxic substance is DDAB (EC50=0.0034 mg/L) and the least toxic is C12+2EO (EC50=0.6327
mg/L). In DSW+HA, the most toxic is DDAB (EC50=0.0039 mg/L) and the least toxic is Ethomeen C/12
(EC50=0.874 mg/L). Only two surfactants were tested in HD600 and are thus the most and least toxic
substances.
EC50 based on measured concentrations are below 1 mg/L for all tested surfactants that was
possible to test with the SPME technique. However, the substances that had problems with the SPME
had nominal EC50 values below 1 mg/L and thus regarded as toxic. Based on 95% confidence
interval, the measured EC50 are not statistically different for Ethomeen C/12 and C12+2EO in DSW
and HD; DDAB and Arquad 2C-75 in DSW; C16+2EO, DDAB and Arquad 2C-75 in DSW+HA and DDAB
and C16+2EO in HD. Whereas the toxicity of all the other substances in the different test media are
statistically different.
In conclusion, the most toxic substance to D. magna in this study regarding measured concentrations
is C16+2EO, since it is not statistically different from DDAB in HD and DSW+HA. However, DDAB and
19
Arquad 2C-75 is also very toxic and the toxicity for Arquad 2C-75 seems to increase as the amount of
humic acid increases since the EC50 values are statistically different in HD and DSW+HA. The least
toxic substance is Ethomeen C/12, but C12+2EO is not very toxic either as it is not statistically
different from Ethomeen C/12 in DSW and HD. From the literature search, few or no toxicity studies
have been performed with these surfactants except for the tests that have been done for registration
according to REACH. Furthermore, a majority of the available reported values are based on nominal
concentrations.
The primary fatty amine, dodecylamine (C12), have a reported acute nominal EC50 to D. magna of
0.146 mg/L in freshwater (ref. 13 in appendix B), which is a factor 3.87 lower than the nominal value
in HD in this study. The freshwater that was used had a concentration of 17.6 mg/L of suspended
matter and a TOC of 5.9 mg C/L, i.e. a higher amount of NOM compared to the HD water used in this
study. A higher mitigation of the reported value would thus be expected, i.e. a higher EC50, because
when based on nominal concentrations, the variation in toxicity between substances in freshwater
will differ due to e.g. their tendency to sorb to the available amount of NOM. However, pH and
conductivity differ between these two test media and might explain the dissimilar result for these
two studies.
Reported toxicity values for Arquad 2C-75 are higher than the data in this study, both for nominal
and measured concentrations. The reported toxicity in HD water based on chronic nominal
concentrations to crustacean is 1.15 mg/L (EC10) (ref. 32 in appendix B) whereas the toxicity based
on acute measured concentrations to algae are varying from 0.148 to 0.386 mg/L (EC50-72h) (ref. 34
in appendix B). The reported chronic EC10 is a factor 8.5 higher than the acute EC50 in this study and
the lowest reported EC50 to algae is a factor 19.2 higher than the measured EC50 to D. magna in this
study (0.0077 mg/L) and the latter could be attributed to different sensitivity between species.
Previous studies have shown that sensitivity between different species of invertebrates towards the
same surfactant can differ up to 2300 times (Lewis and Suprenant, 1983) thus enhancing the
different sensitivity between the two different species algae and D. magna towards Arquad 2C-75.
For DDAB, a nominal LC50 (24h) of 1.2 mg/L (U.S. EPA, 2012 cas: 3282-73-3) to crustacean in
freshwater is also higher (a factor 11.2) than the nominal EC50 (48h) in HD in this study. The
exposure duration differ and the characteristics of the freshwater (amount of NOM, water hardness,
etcetera) is unknown and may explain the higher reported toxicity value. DDAB is the pure
compound and the toxicity is similar to Arquad 2C-75 (mixture) since the mixture is mainly composed
of this substance. The QAC are always positively charged and thus adsorbs rapidly and strongly to
negatively charged substrates (Ying, 2006) and their high toxicity may be explained by a high
electrostatical interaction with the membranes of aquatic organisms. Their relatively high
hydrophobicity also contributes to the toxicity as a baseline toxicity (Könemann, 1981).
The primary fatty amine ethoxylates are less toxic than the QAC, except C16+2EO, and this could be
attributed to a lower hydrophobicity. The reported data for Ethomeen C/12 are also higher than the
toxicity data in this study, although not as much as for the QAC. Reported acute nominal EC50 to D.
magna in standard water varies from 0.84 to 1.4 mg/L (AkzoNobel, 2012:b), a factor 2.6 to 4.3 higher
than this study. The nominal EC50 in HD water in this study is more consistent with the reported
chronic EC50 to D. magna of 0.405 mg/L (AkzoNobel, 2012:b). In comparison with algae, the reported
acute nominal EC50 of 0.107 mg/L in HD (AkzoNobel, 2012:b) is lower than in this study, in contrast
20
to the QAC where it was the other way around. Once again, sensitivity towards different surfactants
differ between species. The algae might be more sensitive towards PFAEO than QAC, which may be
due to their lower hydrophobicity.
The evaluation of cationic surfactants according to REACH are performed on mixtures, which either
consists of mainly short or longer alkyl chains. Thus, the toxicity of C12+2EO can partly be explained
by the toxicity of Ethomeen C/12 and the toxicity is similar regarding measured concentrations. The
higher toxicity of C16+2EO can be explained by another mixture, i.e. Ethomeen 18/16 (oleyl), which
mainly consists of chain lengths of 16 and 18 carbon atoms.
Reported chronic nominal EC50 on D. magna for Ethomeen C/12 and Ethomeen 18/16 (oleyl) are
0.405 and 0.0463 mg/L, respectively (AkzoNobel, 2012:b), that is a higher toxicity with higher alkyl
chain lengths and thus explaining the difference in toxicity between C12+2EO and C16+2EO.
However, it is based on nominal concentrations. Toxicity based on measured concentrations are few,
one reported acute EC50 to D. magna in standard water of Ethomeen 18/16 (oleyl) is 0.043 mg/L
(AkzoNobel, 2012:b), which is a factor 22.6 higher than the measured EC50 for C16+2EO and a factor
8.5 lower than the measured EC50 for C12+2EO in this study. Ethomeen 18/16 (oleyl) is a mixture
and a comparison with pure substances is not always straight-forward, which was seen for the
measured concentrations of the single substances in Ethomeen C/12 in the test media DSW+HA (see
“test media – Ethomeen C/12”). The higher reported measured EC50 of Ethomeen 18/16 (oleyl)
could be due to a lower solubility of the longer alkyl chain lenghts, especially since there was
problems with determination of measured concentrations of C18+2EO in this study. Thus, the true
toxicity is more exerted by the shorter alkyl chain lengths, which is less toxic.
The toxicity of C16+2EO is similar to Arquad 2C-75 and DDAB regarding measured concentrations
when the amount of NOM is increasing. However, the nominal concentrations indicate a higher
toxicity of C16+2EO. The lower nominal EC50 in DSW+HA for C16+2EO compared to the QAC is most
probably exerted by its hydrophobicity. The QAC are mitigated to a higher degree due to their
stronger tendency to adsorb to the humic acid via ion-exchange compared to the PFAEO. Thus, the
toxicity exerted by PFAEO is more based on hydrophobic interactions, whereas the toxicity of QAC is
based on electrostatic interaction. At last, it can be concluded that it is difficult to compare nominal
and measured concentrations since it obviously differs between substances due to sorption to
surfaces or interfaces. Further tests with SPME and more surfactants are necessary to actually be
able to compare them.
3.2 Test media
Two to four different test media are used in the tests and all tests are performed in DSW and HD
water since these two different media were supposed to be comparable when measuring the truly
dissolved concentration with SPME fibers. Due to problems with sorption and comparison of
measured EC50 values, the conductivity was changed in the HD water by adding salts (abbreviated
HD600) to be more similar to the conductivity in DSW. As a consequence, toxicity was altered and
further tests were done without this test medium. Instead, a second calibration curve for SPME in
DSW with a less amount of salts (abbreviated DSW modified) was used to get a better and more
equal comparison between DSW and HD. Many previous toxicity studies have been made in standard
water with purified humic acid (Chen et al., 2010; van Wijk et al., 2009; Ishiguro et al., 2007; Koopal
et al., 2004), thus a fourth test medium (DSW with 20 mg/L commercial humic acid) was used to get
21
another reference point. Commercial HA is used as a surrogate for natural aquatic humic substances
and accounts for almost 100% of the DOC (dissolved organic carbon) in those preparations, whereas
the HA in natural waters only account for approximately 50 to 75 % of the total DOC. The mitigation
factor are thus based on the detoxification in HD water, i.e. the real environment. The correspondent
effect of DOM (dissolved organic matter) in natural water will otherwise be overestimated (Haitzer et
al., 1998). As van Wijk et al. (2009) wrote “A good understanding of sorption in relation to toxicity is
needed to understand the relevant mitigating effects for chemicals” .
3.2.1 Factor difference between different test media
The nominal EC50 of a cationic surfactant varies widely depending on the test media. Due to sorption
to NOM naturally present in river water (HD) and added humic acid to DSW, the freely available
concentration will be the same in all test media since that is the bioavailable fraction. Therefore, the
measured EC50 will vary with a factor 2 maximum from measured EC50 in DSW for one substance. A
factor 2 is chosen as an acceptable difference when measuring the freely dissolved concentration
with SPME, based on experiments from previous investigations with SPME. The nominal and
measured EC50 of one surfactant is represented in each figure below, where all the EC50 are related
to the measured EC50 in DSW (which is set to 1). All EC50 values are presented in table 3 at the end
of section 3.2.1.
Figure 10. Nominal and measured EC50 for dodecylamine +2EO (C12+2EO), expressed as a factor different from measured EC50 in DSW.
For C12+2EO (fig. 10), the nominal EC50 varies with a factor of maximum 2.366 (DSW compared to
HD) and the measured EC50 between the different test media varies with a factor of maximum 1.724
(DSW compared to HD). This is the only substance where the nominal EC50 is higher in HD than in
DSW+HA, however the mitigation factor is still higher for DSW+HA (2.642) than for HD (2.548). Thus,
the mitigation, i.e. relating the laboratory conditions to the real environment, for this substance is a
factor of 2.5.
22
Figure 11. Nominal and measured EC50 for hexadecylamine +2EO (only measured EC50 in the graph to the right), expressed as a factor different from measured EC50 in DSW. The substance is not tested in HD600.
For C16+2EO, the variation in nominal EC50 between different test media is higher with a maximum
factor difference of 15.369 between DSW+HA and DSW (fig. 11). The difference between the
measured EC50 is also higher and varies with a factor of 2.263 for both HD and DSW+HA, compared
to measured DSW (see graph to the right in fig. 11). This value is slightly higher than the accepted
difference of a factor 2. Still, the effect of C16+2EO is mitigated with a factor of 100.791 in DSW+HA
and about 25.884 in HD, compared to a factor of about 2.6 for C12+2EO. The higher measured EC50
in the test media HD and DSW+HA compared to DSW might be due to the stronger sorptive
behaviour of C16+2EO to natural organic matter, because of the long hydrophobic alkyl chain. Thus,
the truly dissolved concentration that can adsorb to the SPME fibers and to the organisms is lower,
and subsequent toxicity is lower.
The detoxification increases as the amount of humic acid increases, as well as carbon chain length.
García et al. (2006) found that the sorption to activated sludge largely increased as the carbon chain
length of QAC increased from C12 to C16, van Wijk et al. (2009) reported a decrease in toxicity with an
increasing concentration of humic acid and Versteeg and Shorther (1992) reported that HA had a
concentration-dependent mitigating effect that was more prominent on the longer alkyl chain
lengths. Hence explaining the difference in mitigation factors between C12+2EO and C16+2EO. The
mitigation factor for C16+2EO is 25.9, a factor 10 higher than for C12+2EO and this is mostly due to
the longer alkyl chain.
Figure 12. Nominal and measured EC50 for ethomeen C/12 as a mixture, expressed as a factor difference from measured EC50 in DSW.
23
The variation in nominal EC50 between different test media of Ethomeen C/12 is also lower (fig. 12),
as it is for C12+2EO. The maximum variation is between DSW+HA and DSW with a factor 3.209.
Ethomeen C/12 as a mixture have a measured EC50 in DSW+HA that is 2.246 times higher than the
measured EC50 in DSW, compared to 1.362 in HD. Thus, in DSW+HA it varies more than the
acceptable factor 2. Here, the measured EC50 for the mixture is based on addition of the measured
concentration of the individual mixture components (see appendix B), since the analysis with LC/MS
extracts the different carbon chain lengths that Ethomeen C/12 consists of and not the mixture as a
whole. The mitigation factor is also low, the highest value of 1.207 is for DSW+HA. The corresponding
value for HD is 0.919.
Figure 13. Nominal EC50 for Ethomeen C/12 as a mixture and measured EC50 for single substances present in the mixture, expressed as a factor difference from measured EC50 for C12+2EO in DSW.
The extracted concentration of each carbon chain length that Ethomeen C/12 consists of,
recalculated to measured EC50, is presented in figure 13. The measured EC50 of the individual
mixture components clearly shows that it is C12+2EO that give rise to the higher factor difference
(3.264) between measured EC50 in DSW+HA and DSW. The fraction of each single substance in
Ethomeen C/12 measured with LC/MS is presented in appendix B. The measured EC50 in HD and
DSW+HA for C14+2EO are within a factor 2 different from the measured EC50 in DSW. Whereas for
C16+2EO, the measured EC50 is within a factor 2 different in HD but a factor of 3.542 different in
DSW+HA compared to DSW.
Ethomeen C/12 mainly consists of C12+2EO and thus explains why the measured EC50 for the
mixture (fig. 12) is higher in DSW+HA, as it is in fig. 13. However, the reason to why the EC50 is higher
is unclear. The measured EC50 for C12+2EO tested individually is within a factor 2 different in all test
media (see fig. 10), whereas it is higher than a factor two for C16+2EO when it is tested individually in
HD and DSW+HA (see fig. 11). Their presence in a mixture leads to an unexpected behavior. There
might be some sort of interaction of C12+2EO or competition between the single substances in the
mixture that causes this. One possible explanation is that C14+2EO and C16+2EO are stronger
competitors, i.e. have a stronger adsorption affinity for the sorption sites on humic acid, fibers and
the organisms. An increased adsorption affinity to clay and sediment with increasing alkyl chain
length has been demonstrated by Droge and Hermens (2010) with alcohol ethoxylate homologues.
This may thus confirm the indicated stronger sorption of C16+2EO to humic acid in DSW+HA since
24
that measured EC50 is 3.5 times lower than the measured EC50 in DSW, in contrast to the measured
EC50 for C12+2EO that is 3.3 times higher in DSW+HA than in DSW. The measured concentration for
Ethomeen C/12 is also based on addition of the concentrations of the single substances in the
mixture, the higher truly dissolved concentration of C12+2EO in the mixture will thus result in a
higher measured EC50 for the mixture in DSW+HA. If the concentration of NOM in HD were higher, a
similar result might have been observed there as well. Further discussed under “single substances
and mixture toxicity”.
According to these data, the mitigation factor increases with increasing amount of NOM and alkyl
chain length. The measured EC50 for C14+2EO is almost the same in all test media, whereas the
measured EC50 for C16+2EO is decreasing from DSW to DSW+HA, indicating a stronger sorption to
NOM. The mitigation factor for C16+2EO in DSW+HA is 87.967, compared to 9.695 for C14+2EO and
1.261 for C12+2EO. However, the mitigation is adjusted according to the entire mixture in the river
water and is set to 0.9.
Figure 14. Nominal and measured EC50 for didodecyldimethylammonium bromide (DDAB), expressed as a factor different from measured EC50 in DSW. DDAB is not tested in HD600.
The nominal EC50 of DDAB (fig. 14) in DSW+HA is 26.719 higher than in DSW. Corresponding value
for HD is 2.724. The measured EC50 is within a factor 2 different from DSW for both HD (1.853) and
DSW+HA (1.615). The mitigation factor is 31.324 for HD and 267.872 for DSW+HA, thus the
mitigation for this substance is 31.3.
25
Figure 15. Nominal and measured EC50 for Arquad 2C-75, expressed as a factor different from measured EC50 in DSW.
For Arquad 2C-75, the nominal EC50 is highest for DSW+HA with a factor of 16.370 higher than DSW
(fig. 15). For HD and HD600, the corresponding difference is a factor 2.752 and 2.874, respectively.
Regarding measured EC50, the acceptable factor of maximum 2 difference is exceeded in HD600
(2.472) and DSW+HA (2.400), compared to HD where the factor is only 1.403 different from DSW
(see graph to the right in fig. 15). The mitigation factor for the three different test media (HD, HD600
and DSW+HA) are 17.584, 5.296 and 178.978, respectively. A mitigation factor of 17.5 can be used
for this substance. Arquad 2C-75 is also a mixture of different carbon chain lengths and the measured
EC50 in the four different test media are here presented by the most dominant carbon chain length
detected with LC/MS. Apparently, the bioavailable fraction of a mixture may not be similar in
different test media due to changed fractions of individual mixture components as a result of their
different sorption affinities, which may explain why the measured EC50 in DSW+HA is a factor of 2.4
lower than in DSW.
QAC and other fatty amine derivatives have an amphiphilic structure, thus have the potential for
both hydrophilic and hydrophobic interactions with NOM. The most abundant component of NOM is
HA and due to their amphiphilic structure, they play a major role in controlling the bioavailability,
hence toxicity, of surfactants (Koopal et al., 2005). The type of sorption was not actually determined
for the tested surfactants in this study, only the differences in nominal and measured concentrations
in different test media. Two different binding mechanisms of cationic surfactants to organic matter
have been observed. One is through van der Waals forces (hydrophobic interaction) between the
apolar carbon chain of the surfactant and the organic fraction of suspended matter and humic acid,
and the other is through electrostatic interaction, i.e. ion-exchange, of the positively charged
nitrogen group to the negatively charged sites of humic acid (van Wijk et al., 2009). Cationic
surfactants thus binds electrostatically to humic acid, whereas nonionic surfactants don’t. Cationic
surfactants also binds hydrophobically to humic acid and this is demonstrated with an increase in
sorption with increasing alkyl chain length (Koopal et al., 2004), as can be seen for C12+2EO and
C16+2EO. This may support the stronger sorption of the QAC (DDAB and Arquad 2C-75) as they are
always positively charged compared to the PFAEO. Furthermore, according to van Wijk et al. (2009)
the CEC of the sorbent is more important than the organic matter content as the CEC results in an
additional electrostatic sorption, although not examined in this study.
26
According to this study, the mitigation of the cationic surfactants toxicity by sorption to NOM is
substance specific. Previous tests performed by AkzoNobel in river water have used a standard
mitigation factor of 10 for all substances when determining their true toxicity, i.e. the bioavailable
fraction. Consequently, both over- and underestimation of their true toxicity have been done. The
difference in mitigation factors varies from 0.9 to 31.3 in this study and is related to HD water with a
TSS of 2.4 mg/L, a TOC of 2.21 mg C/L and a water hardness of 5.56 °dH. For risk assessment
purposes, a standard mitigation factor for all surfactants may thus have serious implications.
Table 3. Nominal and measured EC50 with 95% CI in mg/L for all tested surfactants in different test media. A missing nominal or measured EC50 due to problems with SPME or not tested in that test medium at all, is denoted with (-). The nominal EC50 for C18+2EO in DSW+HA is an estimated value based on 60% mobile daphnid’s at 1.25 mg/L.
DSW (mg/L) HD (mg/L) HD600 (mg/L) DSW+HA (mg/L)
Substance Nom.
(95% CI)
Meas.
(95% CI)
Nom.
(95% CI)
Meas.
(95% CI)
Nom.
(95% CI)
Meas.
(95% CI)
Nom.
(95% CI)
Meas.
(95% CI)
C12 0.0849 (0.0712-0.101)
- 0.566 - - - - -
C12+2EO 0.681 (0.578-0.804)
0.367 (0.302-0.446)
1.612 (1.481-1.755)
0.633 (0.579-0.691)
1.086 (0.941-1.252)
0.526 (0.433-0.640)
1.131 0.428
C16+2EO 0.0282 (0.0253-0.0316)
0.0019 (0.0017-0.0020)
0.111 (0.0945-0.131)
0.0043 (0.0030-0.0062)
- - 0.433 (0.372-0.505)
0.0043 (0.0035-0.0054)
C18+2EO 0.0264 (0.0210-0.333)
- 0.217 (0.146-0.323)
- - - 1.3 -
Ethomeen
C/12
0.329 (0.309-0.351)
0.389 (0.358-0.424)
0.487 (0.411-0.578)
0.530 (0.420-0.669)
- - 1.056 (0.962-1.159)
0.874 (0.716-1.067)
Ethomeen
C/12
-C12+2EO
0.329 0.257 (0.236-0.279)
0.487 0.360 (0.289-0.450)
- - 1.056 0.837 (0.730-0.961)
Ethomeen
C/12
-C14+2EO
0.329 0.0906 (0.0833-0.0984)
0.487 0.131 (0.100-0.171)
- - 1.056 0.121 (0.104-0.142)
Ethomeen
C/12
-C16+2EO
0.329 0.0425 (0.0391-0.0462)
0.487 0.0379 (0.0299-0.0481)
- - 1.056 0.012 (0.0102-0.0141)
DDAB 0.0391 (0.0286-0.0535)
0.0063 (0.0037-0.0106)
0.107 (0.0825-0.137)
0.0034 (0.0020-0.0059)
- - 1.045 (0.906-1.204)
0.0039 (0.0030-0.0050)
Arquad
2C-75
0.0492 (0.0442-0.0548)
0.0108 (0.0091-0.0128)
0.135 (0.124-0.148)
0.0077 (0.0064-0.0093)
0.141 0.0267 0.805 (0.708-0.917)
0.0045 (0.0034-0.0058)
3.2.2 Changed conductivity in river water (HD to HD600)
The conductivity in river water (HD) was changed after the first test with Arquad 2C-75, because of a
higher measured EC50 value for HD than for DSW (not presented in this report). With an increased
salt concentration (Na2+, Ca2+ etc.) in the water, sorption to the fibers was expected to be lower due
to competition between positively charged salt ions and surfactants (Chen et al., 2010) and thus give
a better measurement on the freely dissolved concentration.
27
Nevertheless, the sorption to the fibers increased by adding salts to HD water, indicating a higher
amount of freely dissolved surfactants in the water (see appendix C), hence increasing toxicity. For
C12+2EO the nominal EC50 was increased with a factor of 1.485 by adding salts to HD water. The
difference in nominal EC50 between HD and HD600 is statistically significant with EC50 values of
1.612 (1.481-1.754) and 1.086 mg/L (0.941-1.252), respectively. However, the difference in measured
EC50 is not statistically significant, with EC50 values of 0.633 (0.579-0.691) and 0.526 mg/L (0.433-
0.640), respectively. Thus based on measured concentrations, the alkyl chain length and not the
cations seems to determine the toxicity of C12+2EO. In contrast, the difference in nominal EC50
between HD and HD600 is not statistically significant for Arquad 2C-75 with values of 0.141 and
0.135 mg/L (0.124-0.148), respectively. Whereas the difference in measured EC50 is, with values of
0.0077 (0.0064-0.0093) and 0.0267 mg/L, respectively. Here, the cation activity determines the
toxicity of Arquad 2C-75 and since there are more competitive inorganic cations available in HD600,
the toxicity of Arquad 2C-75 is thus lower (i.e. a higher EC50). A study by Hisano and Oya (2010) with
a mixture of an anionic and a cationic surfactant at different fractions resulted in a decreased toxicity
as the water hardness increased from 25 to 625 ppm. The mixture was assumed to be affected by the
existence of metal ions with the result of a decrease in toxicity. The decreasing toxicity was not seen
when the anionic surfactant was tested individually. This enhance the result of the decreased toxicity
of Arquad 2C-75 in HD600.
Instead, the salts seems to have an effect on the sorption to particles, i.e. negatively charged clay,
present in HD as they might be stronger competitors than the cationic surfactants. The bioavailability
and thereby the sorption of C12+2EO and Arquad 2C-75 to the fibers are thus increased, as opposed
to expectations. The factor difference in nominal and measured toxicity for C12+2EO is 2.548 in HD
and 2.064 in HD600, indicating a stronger sorption of C12+2EO to particles in HD and to the fibers in
HD600. For Arquad 2C-75, the corresponding values are 17.584 for HD and 5.296 for HD600. The
values for Arquad 2C-75 are higher as it is always positively charged and more suspectible to
competition of inorganic ions.
3.3 Degree of ethoxylation
Figure 16. Nominal EC50 for dodecylamine and dodecylamine +2EO in DSW and HD.
Dodecylamine (C12) with no ethoxylates is compared with dodecylamine +2EO (C12+2EO) that has
two ethoxylated groups attached to the amine (fig. 16). Based on nominal concentration, toxicity is
decreasing when two ethoxylated groups are attached to the amine. The toxicity between the two
substances is statistically different in both test media. C12 has an EC50 of 0.0849 mg/L (0.0712-
28
0.1011) in DSW, whereas C12+2EO has an EC50 of 0.6813 mg/L (0.5776-0.8037). For HD, the
corresponding toxicity values are 0.5657 mg/L and 1.6121 mg/L (1.4813-1.7544), respectively. There
is also a difference in sorption to suspended matter and humic acid between the substances. The
difference in EC50 between DSW and HD is higher for C12 (factor 6.663) than for C12+2EO (factor
2.366).
The ethoxylation mainly governs the hydrophilic character of the fatty amine (Holmberg et al., 2003),
thus makes C12+2EO more water soluble. A decreasing water solubility is reflected in an increasing
biophilic character and as a consequence the molecule is more likely to adsorb on lipid membranes
and disrupt different membrane functions (Singh et al., 2002) and this explains the higher toxicity of
the less water soluble C12. Decreasing water solubility may also increase the sorption affinity for
NOM present in HD and then explain the higher difference in sorption for C12. The pKa is also lower
for C12+2EO (8.6 compared to 10.63 for C12), which may affect how much of the substance that is
cationic under the actual test conditions. If the fraction cationic is lower, the sorption to negatively
charged substrates in river water may decreases and the sorption may then be mainly based on
hydrophobic interactions. Previous studies with alcohol ethoxylates (non-ionic surfactants) have
shown a decrease in toxicity with an increase in EO units based on nominal concentrations (Hisano
and Oya, 2010). Measured EC50 values are unfortunately missing for C12, hence a comparison
between the freely dissolved concentrations is not possible. Further tests with SPME and primary
fatty amines, as well as with ethoxylated groups attached, are necessary to determine how toxicity is
altered.
3.4 Toxic response as a function of the alkyl chain length
The toxic response of D. magna as a function of the alkyl chain length is based on PFAEO C12+2EO,
C16+2EO and C18+2EO. The nominal EC50 values are presented in figure 17 and the measured EC50
in figure 18. In general, the nominal EC50 is decreasing with increasing alkyl chain length, although
deviations occur at longer alkyl chain length. Regarding measured concentrations, the results
indicates an increasing toxicity with an increase in the number of carbon atoms in the alkyl chain.
29
Nominal concentration
Figure 17. Nominal log EC50 (mg/L) for three cationic surfactants with different alkyl chain lengths. The EC50 in DSW+HA for octadecylamine +2EO is estimated, based on 60% mobile daphnids at 1.25 mg/L. HD600 was only used as test medium for dodecylamine +2EO.
The toxicity is increasing with an increase in the carbon chain length from C12 to C16, based on both
nominal and measured concentrations (see fig. 17 and 18). For C16+2EO and C18+2EO, the nominal
EC50 is not statistically different in DSW with values of 0.0282 (0.0253-0.0316) and 0.0264 mg/L
(0.0209-0.0333), respectively. However, the toxicity is statistically different in HD and DSW+HA with
a decreasing toxicity from C16 to C18 with increasing amount of NOM. The EC50 values for C16+2EO
and C18+2EO in HD are 0.1113 (0.0945-0.1311) and 0.2171 mg/L (0.1457-0.3233), respectively. The
EC50 value for C16+2EO in DSW+HA is 0.433 mg/L (0.372-0.505) and for C18+2EO in DSW+HA it is
estimated to 1.3 mg/L, based on 60% mobile daphnid’s at 1.25 mg/L, but clearly shows that the
toxicity is decreasing with increasing carbon chain length and amount of NOM. The relationship
between C12 and C18 is not linear, it deviates after C16 and the reason for this is probably due to
solubility problems of C18+2EO. The relationship between C12 and C16 seems to be linear, however, a
conclusion about it cannot be made due to missing data for tetradecylamine (C14) +2EO.
Furthermore, sorption increases as the amount of NOM and the alkyl chain length increases. The
factor differences between DSW and HD are 2.37, 3.95 and 8.22 for C12, C16 and C18, respectively. For
DSW+HA the corresponding factor differences are 1.66, 15.4 and approximately 49.2. The factor
difference between DSW and DSW+HA for C18 is based on an estimated toxicity value. This increasing
sorption with increasing alkyl chain length is also seen for the factor difference between nominal and
measured EC50 for C12 and C16, with a factor 1.86 and 14.84, respectively, different for DSW and 2.55
and 25.88, respectively for HD. HD600 is not included since only C12+2EO is tested in this.
30
Measured concentration
Figure 18. Measured log EC50 (mg/L) for two cationic surfactants with different alkyl chain lengths. Measured EC50 is missing for C18+2EO (see “test media”) and only C12+2EO is tested in HD600.
Similarly to nominal concentrations, the measured EC50 is increasing with increasing carbon chain
length, from C12 to C16 (fig. 18). Measured concentration is, unfortunately, missing for C18+2EO. The
toxicity is statistically different in the three different test media between C12 and C16. For EC50 values,
see table 3, p. 26. Only C12+2EO is tested in HD600 with a result similar to HD. The linear relationship
between C12 and C16 is uncertain due to missing data for C14+2EO, same argument as for nominal
concentrations. In contrast to nominal concentrations, sorption is not increasing as the carbon chain
length and amount of NOM increases. The factor difference between DSW and HD is 1.72 and 2.26
for C12 and C16, respectively. The corresponding values between DSW and DSW+HA are 1.17 and 2.26,
respectively. This is because the EC50 here is based on the freely dissolved concentration of C12+2EO
and C16+2EO, hence the amount adsorbed to organic matter is excluded.
According to a QSAR model based on hydrophobic narcotic chemicals (general narcosis), an increased
carbon chain length gives the molecule a larger hydrophobic fraction and toxicity is thus expected to
increase (Könemann, 1981). A higher toxicity with an increase in the number of carbon atoms in the
alkyl chain have been reported for zwitterionic surfactants on D. magna and P. phosphoreum (García
et al. 2008), for cationic surfactants on D. magna and rainbow trout (Sandbacka et al., 2000), for
nonionic surfactants on D. magna and fathead minnow (Wong et al., 1997) and for cationic and
anionic surfactants on B. calyciflorus (Versteeg et al., 1997). The increasing toxicity was observed up
to a chain length of 14 carbon atoms based on nominal concentrations. The trend is the same for
measured concentrations as reported for anionic surfactants exposed to methanogenic
microorganisms, D. magna and P. promelas (García et al., 2006:a), and for cationic surfactants on P.
subcapitata (van Wijk et al. 2009) and P. promelas (Versteeg and shorter, 1992). Van Wijk et al.
31
(2009), Sandbacka et al. (2000) and Wong et al. (1997) reported a decreasing tendency after 14
carbon atoms, whereas Versteeg and Shorter (1992) reported an increase in toxicity up to a chain
length of 16 and 18 carbon atoms for monoalkyl QAC. The alkyl chain length vs. toxicity relationship
has also been reported for Ethomeen products with different carbon chain lengths. Reported
nominal LC50 (96 h) to fish in standard water are 0.5-0.6 mg/L for Ethomeen C/12 and 0.2 mg/L for
Ethomeen 14/12 (mainly C14+2EO) (AkzoNobel, 2012:b). The corresponding LC50 (96 h) on fish for
Ethomeen 18/16 (oleyl) is 0.1 mg/L (AkzoNobel, 2012:b), but this is based on measured
concentrations. Even if it is a measured concentration, it is only slightly lower than the nominal for
Ethomeen 14/12.
This enhance the results of this study, which shows an increasing toxicity with increasing carbon
atoms in the alkyl chain. However, the deviation occur at C16 in this study since C14+2EO is not
tested. Van Wijk et al. (2009) also reported a decreasing toxicity with increasing humic acid
concentrations, which is similar to this study when comparing the toxicity in HD and DSW+HA based
on nominal concentrations. In addition, the sorption to substrates in the study by van Wijk et al.
(2009) seemed to increase as the chain length increased from 10 to 18 carbon atoms, which is similar
to this study. That is, the effect is mitigated to a larger degree with a longer alkyl chain. Koopal et al.
(2004), Ishiguro et al. (2007) and van Wijk et al. (2009) reported that cationic surfactants binds to
humic substances via both electrostatic and hydrophobic interaction. A longer aliphatic chain gives
the molecule a stronger hydrophobic character (Ishiguro et al., 2007) and the stronger sorption of
C16+2EO compared to C12+2EO is thus the result of their longer aliphatic tail. The same for C18+2EO
compared to C16+2EO. In addition, the hydrophobicity of humic acid is increasing when long-chain
surfactants adsorbs to them and thus influence further adsorption of cationic surfactants as well as
other contaminants to humic acid (Koopal et al., 2004). Thus, the mitigation is increasing with
increasing humic acid concentrations, as well as an increase in carbon chain length due to its higher
hydrophobicity. The toxicity is also increasing with increasing carbon chain length but it seems to
have a tendency to diminish after 14 carbon atoms according to previous studies, both for nominal
and measured concentrations, and after 16 carbon atoms in this study based on nominal
concentrations.
Furthermore, cationic surfactants are very toxic compared to anionic and non-ionic surfactants (Singh
et al., 2002), and polar narcosis, i.e. polar contributions when binding to membranes (Saarikoski and
Viluksela, 1982) might be necessary to take into account as the predicted baseline toxicity is
generally lower than the observed for polar narcotics (Roberts and Costello, 2003:a). That is, the
toxicity is probably not only governed by the length of the alkyl chain. The pKa of the substance
together with the pH of the environment decides whether the substance is cationic or nonionic. The
studied PFAEO have a pKa of about 8.6 (Chen et al., 2012) and tested at a pH of 8.2, thus cationic
during lab conditions. However, the fraction of ionic species are supposed to be the same for
C12+2EO, C16+2EO and C18+2EO and the charge (polar moieties) thus governs the toxicity exerted
by the electrostatic interaction, whereas the alkyl chain length governs the toxicity exerted by
hydrophobic interactions. Although the cationic part contributes to the sorption and the toxicity, a
comparison of the sorption of primary fatty amine ethoxylates and subsequent toxicity, to aquatic
organisms in this study are mainly driven by hydrophobic interactions and also explains why C16+2EO
are more toxic than C12+2EO. In addition, the presence of NOM in the real environment reduces
toxicity as it competes with Daphnia as substrate for sorption. This is seen in HD and DSW+HA
compared to DSW. The additional electrostatic sorption of cationic surfactants to negatively charged
32
substrate is not considered in these QSAR calculations (van Wijk et al., 2009) and further enhance the
need for measurements of the truly dissolved concentrations to determine the true toxicity.
In contrast to previous studies, García et al. (2001) didn’t see any incremental increase in toxicity to
D. magna and P. phosphoreum with increasing carbon chain length for monoalkyl QACs. It was
attributed to a decreasing water solubility with increasing carbon chain length, with the result of
lower bioavailability, hence lower toxicity. This might enhance the results in this study as the
relationship between a chain length of C12 and C18 is not linear, with C18+2EO being less soluble than
C16+2EO, hence lower bioavailability and toxicity. However, a decreasing water solubility is also
related to an increased biophilic character of the molecule, and as a consequence it has a stronger
tendency to adsorb onto lipid membranes of aquatic organisms and disrupt different membrane
functions (Singh et al., 2002). Apparently, the water solubility of the molecule and subsequent
toxicity has a mutual limit. Since nominal and measured EC50 values are missing for C14+2EO, this
study can’t confirm if there is an increase in toxicity from C12 to C16 or if the tendency decreases after
C14. Nor can this study see the measured EC50 to D. magna for C18+2EO, due to problems with the
SPME, to fully evaluate the true toxicity and the relationship between alkyl chain length and toxicity.
3.5 Single substances and mixture toxicity
Chemicals are in these days tested for their intrinsic properties according to REACH which concerns
substances on their own, in preparations and in articles. Development of new alternative hazard
assessments are promoted (REACH, 2006) and since chemicals, e.g. surfactants, are not only present
as single substances in the environment, but rather as mixtures, predictive mixture toxicity models
can be used. Concentration Addition (CA) is a toxicity model for predicting mixture toxicity based on
substances with a similar mode of action. Three substances; dodecylamine +2EO, hexadecylamine
+2EO and octadecylamine +2EO, are tested individually to evaluate their nominal and measured
concentrations. A mixture of these substances, Ethomeen C/12, is also tested and the concept of CA
is applied to see whether it is possible to predict the toxicity of Ethomeen C/12 from the effect of the
single substances.
33
Nominal concentration
The nominal EC50 to D. magna of the single
substances and the mixture, both observed
and predicted, are presented in figure 19.
The nominal EC50 of C12+2EO, C16+2EO
and C18+2EO in the three different test
media represents their effect when tested
individually. Whereas the EC50 for C14+2EO
in the three different test media is
calculated from a linear relationship
between the other three logarithmic EC50
values (see appendix A). The toxicity is
increasing as the carbon chain length
increases, although it decreases after C16
and as the amount of humic acid increases
(see discussion “toxic response as a
function of the alkyl chain length”).
The nominal EC50 of Ethomeen C/12 is
increasing as the amount of NOM increases
as expected, both for observed and
predicted toxicity. In the three different test
media, the observed toxicity of the mixture
is between the highest and lowest toxicity
value of the single substances and this
result is also expected. The toxicity is well
predicted in HD and DSW+HA but is 3 times
higher than the observed in DSW. The joint
toxicity of the individual components in the
test media DSW is thus less than additive
and CA overestimate the mixture toxicity.
Figure 19. Nominal EC50 (mg/L) of individual mixture components and effect of the mixture, both observed and predicted. The EC50 for C14+2EO is calculated and for C18+2EO (orange) it is estimated based on 60% mobile daphnid’s at 1.25 mg/L.
34
Measured concentration
The measured concentrations, i.e. the
truly dissolved concentration that is
bioavailable and have the potential to
exert toxicity to D. magna, of the single
substances and the mixture are
presented in figure 20. The measured
EC50 is missing for C18+2EO and the
EC50 for C14+2EO is calculated from the
linear relationship between the
logarithmic EC50 values of C12+2EO and
C16+2EO (see appendix A). The
predicted toxicity of Ethomeen C/12 in
the three different test media (DSW, HD
and DSW+HA) are higher (23.8, 12.0 and
6.3 times, respectively) than the
observed. As a consequence, the toxicity
of Ethomeen C/12 is overestimated with
CA in all test media when measured
toxicity is considered. Meaning that the
joint toxicity of the individual mixture
components are less than additive and a
higher mixture concentration than
expected by CA is required to provoke
the same effect as the sum of the
individual mixture components.
However, the overestimation decreases
with increasing amount of humic acid.
Noteworthy is that the observed EC50
for Ethomeen C/12 in DSW+HA is higher
than the highest EC50 value for the
single substances (C12+2EO). This is
partly due to the higher factor difference
(2.246) from the measured EC50 in DSW
(see “factor difference between different
test media”), but could also be due to the
analytics as the measured concentrations
of Ethomeen C/12 is higher than the
nominal in DSW and HD. The difference
between the observed and predicted EC50 in DSW+HA would be smaller, if the factor were less than
2.
Figure 20. Measured EC50 (mg/L) of individual mixture components and effect of the mixture, both observed and predicted. The EC50 for C14+2EO is calculated. Measured EC50 is missing for C18+2EO.
35
The prediction of the EC50 for Ethomeen C/12 is dependent on the knowledge of the mixture
components and their individual fraction (Backhaus et al., 2003). According to the Certificate of
Analysis for Ethomeen C/12 (AkzoNobel, 2012:c), the mixture consists of alkyl chain lengths varying
from C8 to C18 with different fractions. The lower alkyl chain lengths (C8 to C10) are excluded in the
predicted mixture toxicity based on nominal concentrations, due to the unknown relationship
between C8 and C12. If a linear relationship is expected from C8 to C18, the difference between
observed and predicted mixture toxicity based on nominal concentrations are still the same. Their
TUs are very low and do not contribute substantially to the mixture toxicity. The predicted nominal
EC50 in all three test media is based on the weight fraction of each single substance in the mixture,
i.e. no considerations is taken regarding sorption to NOM in HD and DSW+HA since those fractions
are unknown for the longest alkyl chain (C18+2EO).
For mixture toxicity based on measured concentrations, only C12 to C16 are taken into account, both
for observed and predicted toxicity. The truly dissolved concentration of the three detectable single
substances within the mixture is measured with LC/MS in three different test media and the fraction
of each is calculated from the total concentration and presented in table 4. When no NOM is present,
i.e. in DSW, the measured fraction with LC/MS of C12+2EO, C14+2EO and C16+2EO are
approximately 0.7, 0.2 and 0.1, respectively. Which is similar, except for the higher fraction of
C12+2EO, to the determined mixture composition according to Certificate of Analysis (see appendix
B) of Ethomeen C/12 used in the prediction of the nominal EC50 for the mixture. However, when
NOM is present, i.e. in HD and DSW+HA, the measured fraction with LC/MS of C12+2EO, C14+2EO
and C16+2EO changes.
Table 4. Fraction of the individual mixture components measured with LC/MS in three different test media.
C12+2EO C14+2EO C16+2EO
DSW 0.70 0.20 0.10 HD 0.69 0.24 0.07 DSW+HA 0.87 0.12 0.010
Previous discussion about hydrophobicity is valid here, i.e. the longer alkyl chains have a higher
sorption affinity to NOM and other surfaces due to a higher hydrophobicity (García et al., 2006), with
the result that C14+2EO and C16+2EO are present at a lower fraction when there is a high amount of
NOM in the water. Conversely, C12+2EO is present at a higher fraction. The toxicity predicted with
CA based on measured concentrations is thus going to be largely exerted by the shorter alkyl chain
length (C12+2EO) in presence of NOM since the longer alkyl chains have a lower fraction in the
mixture. This is because the calculated TUs for the longer alkyl chains become smaller, compared to
when their fraction is higher as it is in DSW. Meaning that when no NOM is present, the predicted
mixture toxicity is more determined by the longer alkyl chain lengths as they will have a higher TU.
Apparently that is not the case regarding observed mixure toxicity in DSW since that measured EC50
is a factor of almost 24 higher than the predicted EC50.
If the substances are acting with a known similar or dissimilar mechanism of action, any increase or
reduction in the overall statistical uncertainty of the predicted mixture toxicity are thus, among
others, largely governed by the ratio of the individual substances within the mixture. Furthermore,
deviations from the prediction of the mixture toxicity may occur under environmental conditions due
36
to, e.g. physico/chemical interactions in or with the mixture. The predictive power of CA may also be
reduced due to synergistic or antagonistic effects because of interferences of the mixture
components (Backhaus et al., 2003). A limitation of the CA concept is thus that it is based on the
fraction of the substances related to their individual effect concentrations in the mixture.
Interactions with natural organic matter and different sorption affinities are not taken into account in
this equation.
The mixture toxicity is overestimated in DSW based on both nominal and measured concentrations,
but to a higher degree regarding measured concentrations. The predicted mixture toxicity in DSW
based on nominal concentrations is 3 times lower than the observed, whereas the corresponding
value based on measured concentrations is 23.8. This difference could be explained by lower
individual EC50 values regarding measured concentrations and an increasing difference between
nominal and measured concentrations with increasing alkyl chain length due to a stronger sorption
affinity, which results in higher TUs for measured concentrations.
The observed measured EC50 of Ethomeen C/12 in the three different test media should be and are
almost the same because the toxicity is based on the truly dissolved concentration, i.e. the
bioavailable fraction that is believed to exert the toxicity. The toxicity predicted with CA should
therefore also be the same as the observed in all test media. A factor 2 is an acceptable difference
with SPME between the measured EC50 in DSW with other test media and could therefore be
applied on the difference between the predicted measured EC50 as well. The difference are however
a factor 2.7 and 8.5 higher for HD and DSW+HA, respectively, than the predicted measured EC50 in
DSW. This could only be attributed to changed fractions of individual mixture compontents when
NOM are present due to different sorption affinities, which mitigate their effect differently, and
consequently, different predicted mixture toxicity.
CA is a concept based on the assumption that substances with a similar mode of action have an
additive mixture effect, thus exchangeable with other substances that have the same TU as they
have in a certain mixture. However, toxicity is not in general linearly related to molecular descriptors.
The ecotoxicity of surfactants are typically increasing logarithmically with a linear increase in the alkyl
chain length and applying the concept of CA is thus going to be largely governed by those mixture
components that are most toxic. As a consequence, the mixture toxicity will be overestimated (Boeije
et al., 2006) which is the case in this study when the predictive mixture toxicity is determined in DSW
based on both nominal and measured concentrations, but also in HD and DSW+HA based on
measured concentrations. The reason for this is that the presence of the most toxic substances is not
reflected in the calculated average structure of the mixture. That is, the nonlinearity of the most
toxic components impact is disproportionate to their molar abundance, whereas the calculation is
(Boeije et al., 2006). CA also interprets that it is the overall binding to the target site that determines
the effect and all organisms are susceptible to baseline toxicity since they all contain membranes
(Porsbring, 2009). The combined effects of the components are estimated well with CA if they belong
to this group of baseline toxicants (Könemann, 1981) or if they have an identical molecular
mechanism of action (Backhaus et al., 2003).
There are no available literature data on comparison between nominal and measured
concentrations, including toxicity, sorption and concentration addition on surfactants. Mixture
toxicity studies in general contains mixtures between anionic/non-ionic/cationic surfactants, not
37
cationic/cationic. A study by Hisano and Oya (2010) with a mixture of an anionic and a cationic
surfactant exposed to D. magna in standard water was less than additive as the TU were greater than
or equal to 1. This result was in agreement with another study with binary and ternary mixtures of
anionic, non-ionic and cationic surfactants, referred by Hisano and Oya (2010). It is also similar to this
study.
Boeije et al. (2006) reported a measured EC50 value for a mixture of non-ionic Alcohol Ethoxylates
(AE) that was 1.5 times higher than the EC50 predicted by CA, but due to variability in the analytical
recovery, the measured concentration could be overestimated with the result that the mixture
toxicity is actually more consistent with the CA than observed. Boeije et al. (2006) also referred to
other studies which states that the CA model is valid for AE but also for other baseline toxicants.
However, the toxicities of non-ionic surfactants are well predicted with the general narcosis equation
(Roberts and Castello, 2003) and thus enhance why the CA model is applicable for AE. Cationic
surfactants on the other hand, have been shown to act by a polar narcosis mechanism (Roberts and
Castello, 2003) and may explain why the mixture of PFAEO do not conform to CA. However, the
observed toxicity is lower than the predicted in this study and that should not be the case if they are
polar narcotics. Other factors might then influence and affect the mixture toxicity.
Although predictive toxicity models are very useful when considering economy, time-efficiency,
animal testing etcetera, in determining the toxicity of mixtures, the risk that they might over or
underestimate the mixture toxicity is still there. Regarding registration of surfactants according to
REACH, where the intrinsic properties of the surfactant are supposed to be evaluated, it is thus
better to test the substance, i.e. the mixture itself, to minimize this risk. From an environmental risk
assessment point of view, it is actually useless to test a specific mixture as the real environment
consists of an infinite amount of different mixtures.
Concluding summary
The SPME method used in this study measures the truly dissolved concentration of surfactants in the
water and that is the concentration believed to be bioavailable, thus have a potential to exert toxicity
to aquatic organisms. The method is very useful as the total concentration of surfactants in natural
water may be of less importance (Haitzer et al., 1998) when risk assessments are performed to
predict the potential effect and environmental concentrations, hence the risk posed by them to
aquatic organisms. However, the toxicity of these cationic surfactants to D. magna are probably
greater than their hydrophobicity imply as a consequence of their ability to also interact
electrostatically with biological surfaces. This study have only measured how much the effect is
mitigated and not how, that is hydrophobically or electrostatically. The aquatic toxicity of a pure
substance, e.g. one specific alkyl chain length attached to the amine, is assumed to be the same
regardless of test media when it is based on the truly dissolved concentration. Whereas the
composition of a mixture changes in different test media due to, e.g. different sorption affinities of
the individual mixture components, and this is reflected in the truly dissolved concentrations. As a
consequence, the toxicity is altered and more obvious in predictive toxicity models.
3.6 Further recommendations
This study examined the acute toxicity of one primary fatty amine, four primary fatty amine
ethoxylates and two quaternary ammonium compounds. Further tests with these and other related
cationic surfactants are necessary to fully evaluate the SPME method for cationic surfactants and to
38
be able to build a QSAR model for them regarding aquatic toxicity and bioavailability. One factor that
seems to affect the SPME method is the water solubility of the cationic surfactants. The less water
soluble it is, the more difficult it is. Water solubility of chemicals is a factor that matters for a QSAR
based on log P (Könemann, 1981). If the substance is infinitely soluble in water the toxicity is not
possible to predict with an equation baed on hydrophobicity, only slightly soluble substances is.
When further tests are done, a QSAR based on log P can be used to model the toxicity of primary
fatty amines, PFAEO and QAC, especially in mixture toxicity studies, to determine if the tested
cationics follow a general or polar narcotic mechanism of toxicity.
Long term toxicity test with D. magna and SPME should preferably be performed to se whether the
relationship from short term to long term is linear or not. In addition, test should preferably also be
performed on other organisms, e.g. algae and fish as the sensitivity differ between species. Daphnia
is believed to be the most sensitivie species towards cationic surfactants (Lewis and Suprenant, 1983)
based on nominal concentration. What would be the results if it is based on measured
concentrations?
What is the molecular mechanism of cationic surfactants towards different species of organisms, i.e.
how are the daphnia, algae and fish affected by cationic surfactants? In general, the toxicity of
surfactants are indicated to be determined by their affinity to adsorb onto the cell membrane, mainly
driven by nonspecific hydrophobic interactions, and their ability to penetrate the membrane bilayer
(Rosen et al., 2001). The plasma membrane consists of lipids and mostly phospholipids, which also
have an amphiphilic structure. Surfactants disrupts the hydrophobic interactions of the bilayer by
binding to the hydrophobic region of transmembrane proteins and the hydrophobic fatty acid tails,
thus forming protein-surfactant complexes and solubilizing the phospholipids (Alberts et al., 2004).
Fish may thus be affected as the water is constantly pumped through the gills, whereas algae has a
larger, negatively charged surface area and D. magna may be affected as they are filter feeders. If the
adsorption to cell membranes is mainly driven by nonspecific hydrophobic interactions, what is then
the difference in toxicity between e.g. non-ionic and cationic surfactants?
Limitations of this study
This study has focused on the nominal and measured EC50 values and all the comparisons within and
between substances are based on this. Further studies, e.g. on mixture toxicity, should preferably
look at the entire concentration range from EC1 to EC100 to better see any under- or overestimation
of mixture toxicity when the concept of CA is applied.
Sorption to humic acid may also enhance mobility of surfactants in the soil and hasn’t been
considered in this study.
At last, the pKa of the primary fatty amine ethoxylates are low and the pH in the test is high, with the
results that the molecule may not be entirely cationic. A determination of how much is cationic
under the test condition could be necessary, at least when comparison between other cationic
surfactants are made.
39
4. Conclusions SPME fibers extracts the freely dissolved concentration of the tested surfactants in various test
media and provides information about the bioavailability, thus the potential risk of cationic
surfactants in the environment. Based on the results from the acute immobility test (OECD 202) and
the SPME it can be concluded that sorption of the tested cationic surfactants to NOM in river water
(HD) clearly has a mitigating effect, although substance specific, on the acute toxicity to Daphnia
magna.
• The mitigation of each surfactant in different test media are determined and the difference
between nominal and measured concentrations of cationic surfactants are due to their
strong tendency to sorb to substrates via hydrophobic, as well as electrostatic interaction.
The toxicity is governed by both of these interactions, however this study haven’t examined
how much each of these interactions contribute and how it may differ between surfactants.
• The most toxic substance regarding measured concentrations is hexadecylamine +2EO,
although didodecyldimethylammonium bromide in HD and DSW+HA are not statistically
different from C16+2EO. Furthermore, Arquad 2C-75 is also very toxic and the toxicity seems
to increase as the amount of HA increases. The least toxic substance is Ethomeen C/12,
together with dodecylamine +2EO as it is not statistically different in DSW and HD.
• Mitigation factors for cationic surfactants are substance specific and varies from 0.9 to 31.3
in this study. A standard mitigation factor for all substances will inevitably lead to either
over- or underestimation of their true toxicity, depending on which surfactant it is.
• Toxicity is increasing with an increase from C12 to C16 in the alkyl chain for PFAEO, based on
both nominal and measured concentrations, and it is related to an increasing hydrophobicity
within the molecule. The tendency is decreasing from C16 to C18 regarding nominal
concentrations probably due to a lower water solubility.
• An addition of two ethoxylated groups to dodecylamine results in a higher nominal EC50,
both in DSW and HD, due to a higher water solubility of the molecule.
• The predictive toxicity model Concentration Addition overestimates the mixture toxicity of
Ethomeen C/12 in Dutch Standard Water based on nominal and measured concentrations,
the joint toxicity of the individual mixture components are thus less than additive. Regarding
measured concentrations, the overestimation decreasas as the amount of NOM increases
due to a changed composition of the mixture, i.e. the fraction of individual mixture
components, caused by different sorption affinities. This in turn affects the predicted toxicity.
An overall conclusion is that the SPME method is a good technique to quantify the truly dissolved
concentration of cationic surfactants, but further studies are necessary to properly evaluate the
method for these kind of substances to be able to find mitigation factors. These acute tests only gives
an explanation that the mitigation is substance specific for cationic surfactants and they may vary as
the amount of suspended matter, humic acid and other sorbents in the aquatic environment varies
from one place to another and over the year.
40
Acknowledgements This study was supported by AkzoNobel Surface Chemistry AB in Stenungsund, Sweden, so many
thanks for the financed trip and stay in Arnhem, the Netherlands. I would also like to gratefully
acknowledge my supervisor Bengt Fjällborg at AkzoNobel, Stenungsund Sweden, for the possibility to
perform my master thesis in ecotoxicology at a multinational chemical company and for all your help
during the experiments and writing.
I would also acknowledge my academic supervisor, Thomas Backhaus, from Gothenburg university
for academic support and input to my report.
Thanks to Mark Kean and Marc Geurts at AkzoNobel ecotoxicology lab in Arnhem the Netherlands,
for the possibility to perform my experiments and expert judgements during the experiments. Also
thanks for the assistance to find an apartment and later on cutleries to my small kitchenette.
Special thanks to Jacco van Dam, Bart Kluskens and Irmgard Garttener for all your help and assistance
during my experiments. Also thanks to everyone else I met in Arnhem, all of you made my visit to the
Netherlands wonderful and pleasant.
41
5. References Acmite (2010) Market Report: World Surfactant market, Acmite Market Intelligence. Available at:
<http://www.acmite.com/market-reports/chemicals/world-surfactant-market.html>
AkzoNobel (2012) Available at: <www.akzonobel.com> 2012-02-03.
AkzoNobel (2012:a) Available at: <http://sc.akzonobel.com/en/fabric-cleaning/Pages/product-
detail.aspx?prodID=8285> 2012-06-14.
AkzoNobel (2012:b) Aquatic toxicity data for REACH registration.
AkzoNobel (2012:c) Certificate of Analysis for Ethomeen C/12.
AkzoNobel (2011) Safety data sheet, according to Regulation (EC) No. 1907/2006 for Ethomeen C/12.
Available at: <http://sc.akzonobel.com/MSDS/12_GB_EN_000000000125.pdf> and <
http://sc.akzonobel.com/en/personalcare/Pages/product-detail.aspx?prodID=6671> 2012-06-14.
Alberts, B., Bray, D., Hopkin, K., Johnson, A., Lewis, J., Raff, M., Roberts, K. and Walter, P. (2004)
Essential cell biology, second edition. Garland Science, Taylor & Francis Group, New York.
Alexander, M. (2000) Aging, Bioavailability, and Overestimation of Risk from Environmental
Pollutants. Environmental Science and Technology, Vol. 34, pp. 4259-4265.
Backhaus, T., Altenburger, R., Arrhenius, Å., Blanck, H., Faust, M., Finizio, A., Gramatica, P., Grote, M.,
Junghans, M., Meyer, W., Pavan, M., Porsbring, T., Scholze, M., Todeschini, R., Vighi, M., Walter, H.
and Grimme, L.H. (2003) The BEAM-project: prediction and assessment of mixture toxicities in the
aquatic environment. Continental Shelf Research, Vol. 23, pp. 1757-1769.
Banat, I.M., Makkar, R.S. and Cameotra, S.S. (2000) Potential commercial applications of microbial
surfactants. Applied Microbial Biotechnology, Vol. 53, pp. 495-508.
Barco, M., Planas, C., Palacios, O., Ventura, F., Rivera, J. and Caixach, J. (2003) Simultaneous
quantitative analysis of anionic, cationic, and nonionic surfactants in water by electrospray ionization
mass spectrometry with flow injection analysis. Analytical Chemistry, Vol. 75, pp. 5129-5136.
Boeije, G.M., Cano, M.L., Marshall, S.J., Belanger, S.E., van Compernolle, R., Dorn, P.B., Gümbel, H.,
Toy, R. and Wind, T. (2006) Ecotoxicity quantitative structure-activity relationships for alcohol
ethoxylate mixtures based on substrance-specific toxicity predictions. Ecotoxicology and
Environmental Safety, Vol. 64, pp. 75-84.
Chen, Y., Droge, S.T.J. and Hermens, J.L. (2012) Power point presentation: Towards a better
understanding of the bioavailability and partition behaviour of cationic surfactants. Project update:
Jan 2012-jun2012. Utrecht University, Institute for Risk Assessment Sciences / Division Technology.
Chen, Y., Droge, S.T.J. and Hermens, J.L. (2010) Analyzing freely dissolved concentrations of cationic
surfactant utilizing ion-exchange capability of polyacrylate coated solid-phase microextraction fibers.
Journal of Chromatography A, doi:10.1016/j.chroma.2012.06.080.
42
Cooney, J.D. (1995) Ch 2: Freshwater tests, in Fundamentals of aquatic toxicology – effects,
environmental fate and risk assessment. Second edition, edited by Rand, G.M. Taylor & Francis,
Washington D.C.
Di Corcia, A. (1998) Characterisation of surfactants and their biointermediates by liquid
chromatography mass spectrometry. Journal of Chromatography A, Vol. 794, pp. 165-185.
Droge, S.T.J. and Hermens, J.L.M. (2010) Alcohol ethoxylate mixtures in marine sediment:
Competition for adsorption sites affects the sorption behaviour of individual homologues.
Environmental Pollution, Vol. 158, pp. 3116-3122.
Droge, S.T.J., Sinnige, T.L. and Hermens, J.L.M (2007) Analysis of freely dissolved alcohol ethoxylate
homologues in various seawater matrixes using solid-phase microextraction. Analytical Chemistry,
Vol. 79., pp. 2885-2891.
Duman, O. and Ayranci, E. (2010) Adsorptive removal of cationic surfactants from aqueous solutions
onto high-area activated carbon cloth monitored by in situ UV spectroscopy. Journal of Hazardous
Materials, Vol. 174, pp. 359-367.
ECHA (2012) European Chemicals Agency, Registered substances. Available at:
<http://echa.europa.eu/web/guest/information-on-chemicals/registered-substances> 2012-02-24.
ECHA (2012:a) Identified uses for dodecylamine. Available at:
<http://apps.echa.europa.eu/registered/data/dossiers/DISS-9d988c3c-1891-0cf3-e044-
00144f67d249/AGGR-dbada6ba-4de9-4b74-b88e-34132894b9ee_DISS-9d988c3c-1891-0cf3-e044-
00144f67d249.html#section_3_5> 2012-04-17.
ECHA (2012:b) Identified uses for Ethomeen C/12. Available at:
<http://apps.echa.europa.eu/registered/data/dossiers/DISS-98761353-424b-6f67-e044-
00144f67d031/AGGR-ea2ef8ed-1885-41b1-85b2-162ec1cbfb73_DISS-98761353-424b-6f67-e044-
00144f67d031.html#section_3_5> 2012-05-08.
ECHA (2012:c) Identified uses for Arquad 2C-75. Available at:
<http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-
00144f67d031/AGGR-22d56b0e-6254-4706-9b79-0b170d20772b_DISS-9fea2ec8-51a7-3ba2-e044-
00144f67d031.html#section_3_5> 2012-04-17.
ECHA (2011) Guidance on information requirements and chemical safety assessment, Part B: Hazard
assessment. European Chemicals Agency, Helsinki, Finland, Version 2.1. Available at:
<http://echa.europa.eu/documents/10162/13643/information_requirements_part_b_en.pdf>
Europa.eu (2011) Available at:
<http://ec.europa.eu/environment/chemicals/reach/reach_intro.htm> 2012-02-03.
Fernandez, P., Alder, A.C., Suter, M.J.F. and Giger, W. (1996) Determination of the quaternary
ammonium surfactant ditallowdimethylammonium in digested sludges and marine sediments by
supercritical fluid extraction and liquid chromatography with postcolumn ion-pair formation.
Analytical Chemistry, Vol. 68, pp. 921-929.
43
García, M.T., Campos, E., Marsal, A. and Ribosa, I. (2008) Fate and effects of amphoteric surfactants
in the aquatic environment. Environmental International, Vol. 34, pp. 1001-1005.
García, M.T., Campos, E., Sánchez-Leal, J. and Comelles, F. (2006) Sorption of alkyl benzyl dimethyl
ammonium compounds by activated sludge. Journal of Dispersion Science and Technology, Vol. 27,
pp. 739-744.
García, M.T., Campos, E., Sánchez-Leal, J. and Ribosa, I. (2006:a) Effect of linear alkylbenzene
sulphonates (LAS) on the anaerobic digestion of sewage sludge. Water research, Vol. 40, pp. 2958-
2964.
García, M.T., Ribosa, I., Guindulain, T., Sánchez-Leal, J. and Vives-Rego, J. (2001) Fate and effect of
monoalkyl quaternary ammonium surfactants in the aquatic environment. Environmental Pollution,
Vol. 111, pp. 169-175.
Guetzloff, T.F. and Rice, J.A. (1994) Does humic acid form a micelle? The Science of the Total
Environment, Vol. 152, pp. 31-35.
Haigh, S.D. (1996) A review of the interaction of surfactants with organic contaminants in soil. Science
of the total environment, Vol. 185, Issue: 1-3, pp. 161-170.
Haitzer, M., Höss, S., Traunspurger, W. and Steinberg, C. (1998) Effects of dissolved organic matter
(DOM) on the bioconcentration of organic chemicals in aquatic organisms – a review. Chemosphere,
Vol. 37, pp. 1335-1362.
Hisano, N. and Oya, M. (2010) Effects of surface activity on aquatic toxicity of binary surfactant
mixtures. Journal of Oleo Science, Vol. 59, pp. 589-599.
Holmberg, K., Jönsson, B., Kronberg, B. and Lindman, B. (2003) Surfactants and polymers in aqueous
solutions. Second edition, John Wiley & Sons Ltd, West Sussex, England.
Ishiguro, M., Tan, W. and Koopal, L.K. (2007) Binding of cationic surfactants to humic substances.
Colloids and Surfaces A: Physicochemical and Engineering aspects, Vol. 306, pp. 29-39.
Ivankovic, T. and Hrenovic, J. (2010) Surfactants in the environment. Arhiv za higijenu rada I
toksikologiju, Vol. 61, pp. 95-110.
Jones-Hughes, T. and Turner, A. (2005) Sorption of ionic surfactants to estuarine sediment and their
influence on the sequestration of phenanthrene. Environmental Science & Technology, Vol. 39, Issue:
6. Pp: 1688-1697.
Koopal, L.K., Saito, T., Pinheiro, J.P. and van Riemsdijk, W.H. (2005) Ion binding to natural organic
matter: General considerations and the NICA-Donnan model. Colloids and Surfaces A:
Physicochemical and engineering aspects, Vol. 265, pp. 40-54.
Koopal, L.K., Goloub, T.P. and Davis, T.A. (2004) Binding of ionic surfactants to purified humic acid.
Journal of colloid and interface science, Vol. 275, pp. 360-367.
Könemann, H. (1981) Quantitative structure-activity relationships in fish toxicity studies, part 1:
relationship for 50 industrial pollutants. Toxicology, Vol. 19, pp. 209-221.
44
Lara-Martín, P.A., Li, Xiaolin., Bopp, R.F. and Brownawell, B.J. (2010) Occurrence of
alkyltrimethylammonium compounds in urban estuarine sediments: Behentrimonium as a new
emerging contaminant. Environmental Science and Technology, Vol. 44, pp. 7569-7575.
Lewis, M.A. and Suprenant, D. (1983) Comparative acute toxicities of surfactants to aquatic
invertebrates. Ecotoxicology and environmental safety, Vol. 7, pp. 313-322.
Lyman, W.J. (1995) Chapter 15: Transport and transformation processes, in Fundamentals of aquatic
toxicology – effects, environmental fate and risk assessment. Second edition, edited by Rand, G.M.
Taylor & Francis, Washington D.C.
OECD (2004) OECD Guideline for testing of chemicals, Daphnia sp., Acute Immobilisation Test.
Organisation for Economic Co-operation and Development, adopted 13 April 2004. Available at:
<www.oecd.org> 2012-02-17.
Porsbring, T. (2009) On toxicant-induced succession in periphyton communities: effects of single
chemicals and chemical mixtures. Ph D-thesis at Faculty of science, department of plant and
environmental sciences, University of Gothenburg, Sweden.
Rand, G.M., Wells, P.G. and McCarty, L.S. (1995) Ch 1: Introduction to aquatic toxicology, in
Fundamentals of aquatic toxicology – effects, environmental fate and risk assessment. Second
edition, edited by Rand, G.M. Taylor & Francis, Washington D.C.
REACH (2006) Regulation (EC) No 1907/2006 of the European Parliament and of the Council, Available
at:
< http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2006:396:0001:0849:EN:PDF>
Reznik, G.O., Vishwanath, P., Pynn, M.A., Sitnik, J.M. Todd, J.J., Wu, J., Jiang, Y., Keenan, B.G., Castle,
A.B., Haskell, R.F., Smith, T.F., Somasundaran, P. And Jarrell, K.A. (2010) Use of sustainable chemistry
to produce an acyl amino acid surfactant. Applied microbiology and biotechnology, Vol. 86, pp. 1387-
1397.
Rico-Rico, A., Droge, S.T.J., Widmer, D. and Hermens, J.L.M. (2009) Freely dissolved concentrations of
anionic surfactants in seawater solutions: Optimization of the non-depletive solid-phase
microextraction method and application to linear alkylbenzene sulfonates. Journal of
Chromatography A, Vol. 1216, pp. 2996-3002.
Roberts, D.W. and Costello, J. (2003) QSAR and mechanism of action for aquatic toxicity of cationic
surfactants. QSAR and Combinatorial Science, Vol. 22, pp. 220-225.
Roberts, D.W. and Costello, J. (2003:a) Mechanisms of action for general and polar narcosis: A
difference in dimension. QSAR and Combinatorial Science, Vol. 22, pp. 226-233.
Roberts, D.W. (2000) Aquatic toxicity – Are surfactant properties relevant? Journal of Surfactants and
Detergents, Vol. 3, pp. 309-315.
Rosen, M.J. and Kunjappu, J.T. (2012) Surfactants and Interfacial Phenomena. 4 edition. John Wiley &
Sons, Inc., Hoboken, New Jersey.
45
Rosen, M., Li, F., Morrall, S.W. and Versteeg, D.J. (2001) The relationship between interfacial
properties of surfactants and their toxicity to aquatic organisms. Environmental Science and
Technology, Vol. 35, pp. 954-959.
Rufli, H., Fisk, P.R., Girling, A.E., King, J.M.H., Länge, R., Lejeune, X., Stelter, N., Stevens, C., Suteau, P.,
Tapp, J., Thus, J., Versteeg, D.J. and Niessen, H.J. (1998) Aquatic toxicity testing of sparingly soluble,
volatile and unstable substances and interpretation and use of data. Ecotoxicology and
Environmental safety, Vol. 39, pp. 72-77.
Rust, D. and Wildes, S. (2008) Surfactants: a market opportunity study update. OmniTech
International Ltd., Midland, MI.
Sandbacka, M., Christiansson, I. and Isomaa, B. (2000) The acute toxicity of surfactants on fish cells,
Daphnia magna and fish – A comparative study. Toxicology in Vitro, Vol. 14, pp. 61-68.
Saarikoski, J. and Viluksela, M. (1982) Relation between physicochemical properties of phenols and
their toxicity and accumulation in fish. Ecotoxicology and Environmental Safety, Vol. 6, pp. 501-512.
Semple, K.T., Doick, K.J., Burauel, P., Craven, A. And Harms, H. (2004) Defining bioavailability and
bioaccessibility of contaminated soil and sediment is complicated. Environmental Science and
Technology, pp. 228-231.
Singh, R.P., Gupta, N., Singh, S., Suman, R. And Annie, K. (2002) Toxicity of ionic and nonionic
surfactants to six macrobes found in Agra, India. Bulletin of environmental contamination and
toxicology, Vol. 69, pp. 265-270.
TGD (2003) Technical Guidance Document on Risk Assessment Part II. European Commission Joint
research Centre, European Chemicals Bureau. Available at:
<http://ihcp.jrc.ec.europa.eu/our_activities/public-
health/risk_assessment_of_Biocides/doc/tgd/tgdpart2_2ed.pdf>
Thomas, P.C., Velthoven, K., Geurts, M. and van Wijk, D. (2009) Bioavailability and detoxification of
cationics: II. Relationship between toxicity and CEC of cationic surfactants on Caenorhabditis elegans
(Nematoda) in artificial and natural substrates. Chemosphere, Vol. 75, pp. 310-318.
U.S. EPA (2012) ECOTOX Database, 4.0. provided by U.S. Environmental Protection Agency. Available
at: <http://cfpub.epa.gov/ecotox/> 2012-06-11.
U.S. EPI Suite (2011) Exposure Assessment Tools and Models: Estimation Program Interface (EPI)
Suite, provided by U.S. Environmental Protection Agency. Available at:
<http://www.epa.gov/oppt/exposure/pubs/episuite.htm> 2012-06-26.
Van Wijk, D., Gyimesi-van den Bos, M., Garttener-Arends, I., Geurts, M., Kamstra, J. and Thomas, P.
(2009) Bioavailability and detoxification of cationics: I. Algal toxicity of alkyltrimethyl ammonium salts
in the presence of suspended sediment and humic acid. Chemospere, Vol. 75, pp. 303-309.
Versteeg, D.J., Stanton, D.T., Pence, M.A. and Cowan, C. (1997) Effects of surfactants on the rotifer,
Brachionus calyciflorus, in a chronic toxicity test and in the development of QSARS. Environmental
Toxicology and Chemistry, Vol. 16, pp. 1051-1058.
46
Versteeg, D.J., Feijtel, T.C.J., Cowan, C.E., Ward, T.E. and Rapaport, R.A. (1992) An environmental risk
assessment for DTDMAC in the Netherlands. Chemosphere, Vol. 24, pp. 641-662.
Versteeg, D.J. and Shorter, S.J. (1992) Effect of organic carbon on the uptake and toxicity of
quaternary ammonium compounds to the fathead minnow, Pimephales promelas. Environmental
Toxicology and Chemistry, Vol. 11, pp. 571-580.
Wee, T.V. (1984) Determination of cationic surfactants in waste- and river waters. Water Research,
Vol. 18, pp. 223-225.
WHO (1996) Environmental Health Criteria 169: Linear alkylbenzene sulfonates and related
compounds. World Health Organization, Geneva.
Wong, D.C.L., Dorn, P.B. and Chai, E.Y. (1997) Acute toxicity and structure-activity relationships of
nine alcohol ethoxylate surfactants to fathead minnow and Daphnia magna. Environmental
Toxicology and Chemistry, Vol. 16, pp. 1970-1976.
Ying, G.G. (2006) Fate, behaviour and effects of surfactants and their degradation products in the
environment. Environment International, Vol. 32, pp. 417-431.
47
APPENDIX A: Calculations of TU for single substances and mixture TUs for the single substances are calculated from equation 1, where c is the concentration of a single
substance in the mixture divided by its individual effect concentration, e.g. EC50. Addition of the
single TUs gives the TU of the mixture (eq. 2 and 3) and the mixture conforms to CA when the TUs
are equal, i.e. 1 (eq. 4).
(1) (2) (3) (4)
To calculate the EC50 for the mixture, equation 3 is used and rearranged gives equation 5. The
concentrations of the single substances are expressed as a fraction of the mixture, where the mixture
is set to 1.
(5)
It is assumed that dodecylamine +2EO, tetradecylamine +2EO, hexadecylamine +2EO and
octadecylamine +2EO have a similar mode of action, hence the concept of Concentration Addition is
applied to predict the toxicity of the mixture, i.e. Ethomeen C/12.
Toxic Units (TU)
DSW
Single substances and mixture TU (Nominal) TU (Measured)
C12+2EO 0.734 1.924 C14+2EO 1.103 7.390 C16+2EO 3.546 51.842 C18+2EO 3.788 - Predicted (Ethomeen C/12) 9.171 61.156 Observed (Ethomeen C/12) 3.040 2.569
HD
Single substances and mixture TU (Nominal) TU (Measured)
C12+2EO 0.310 1.083 C14+2EO 0.337 4.635 C16+2EO 0.898 16.837 C18+2EO 0.461 - Predicted (Ethomeen C/12) 2.007 22.555 Observed (Ethomeen C/12) 2.052 1.886
DSW+HA
48
Single substances and mixture TU (Nominal) TU (Measured)
C12+2EO 0.442 2.027 C14+2EO 0.217 2.846 C16+2EO 0.231 2.293 C18+2EO 0.0770 - Predicted (Ethomeen C/12) 0.966 7.165 Observed (Ethomeen C/12) 0.947 1.144
Calculation of EC50 for tetradecylamine +2EO
The EC50 for tetradecylamine +2EO (C14+2EO) is calculated from the equation given by the linear
relationship between dodecylamine +2EO and hexadecylamine +2EO, and where it is possible also
with octadecylamine +2EO. The difference in EC50 values for C14+2EO calculated in DSW+HA with or
without the estimated EC50 for C18+2EO is negligible.
Figure 21. Calculation of nominal and measured EC50 (mg/L) for tetradecylamine +2EO in DSW.
Figure 22. Calculation of nominal and measured EC50 (mg/L) for tetradecylamine +2EO in HD.
49
Figure 23. Calculation of nominal and measured EC50 (mg/L) for tetradecylamine +2EO in DSW+HA. The nominal EC50 for octadecylamine +2EO (C18+2EO) is estimated (orange mark).
50
APPENDIX B: Physico-chemical properties and toxicity data for tested
surfactants
DODECYLAMINE
Physical and chemical properties
CAS number 124-22-1
EC number 204-690-6 IUPAC name Dodecan-1-amine
Molecular formula C12H27N
Molecular weight 185.35 g/mole Purity of the substance ≥99.5%
Physical state (20°C, 1013 hPa) Solid paste, colorless to slightly yellow1
Log Pow (20°C) 8.352 (calculated from solubility’s in pure solvents)
pKa (25°C) 10.633 pH (20°C) 12.44
Water solubility (25°C, pH 7) 3.5 g/L5
Surface tension (1.0 g/L, 20°C) 51.3 mN/m6
Vapour pressure (25°C) 0.41 Pa7
Biodegradation in water (Guideline 301 D)
>60% after 28 days8 (Readily biodegradable)
Biodegradation in water and sediment (OECD Guideline 303 A)
>99.98%9
Bioaccumulation (BCF) 173 L/kg10
(weight of evidence using log Kow)
1 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9d988c3c-1891-0cf3-e044-00144f67d249/AGGR-
9a2aab3f-7679-4761-8447-2b50d562cd75_DISS-9d988c3c-1891-0cf3-e044-00144f67d249.html#AGGR-9a2aab3f-7679-4761-8447-2b50d562cd75 2012-04-17. 2 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9d988c3c-1891-0cf3-e044-00144f67d249/AGGR-aa2a9c2f-a58f-4182-9b15-82cd553bc41d_DISS-9d988c3c-1891-0cf3-e044-00144f67d249.html#AGGR-aa2a9c2f-a58f-4182-9b15-82cd553bc41d 2012-04-17. 3 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9d988c3c-1891-0cf3-e044-00144f67d249/AGGR-
42c11e67-23fc-4b20-8f54-14ba9d1890d5_DISS-9d988c3c-1891-0cf3-e044-00144f67d249.html#AGGR-42c11e67-23fc-4b20-8f54-14ba9d1890d5 2012-04-17. 4 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9d988c3c-1891-0cf3-e044-00144f67d249/AGGR-
72d9a1a1-6ec2-4a22-8dbb-d9eda80d32fc_DISS-9d988c3c-1891-0cf3-e044-00144f67d249.html#AGGR-72d9a1a1-6ec2-4a22-8dbb-d9eda80d32fc 2012-05-08. 5 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9d988c3c-1891-0cf3-e044-00144f67d249/AGGR-34fd2527-564b-4e98-88d9-fe71bf598023_DISS-9d988c3c-1891-0cf3-e044-00144f67d249.html#AGGR-34fd2527-564b-4e98-88d9-fe71bf598023 2012-04-17. 6 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9d988c3c-1891-0cf3-e044-00144f67d249/AGGR-2a5643b9-54fc-44b4-8919-e1c9f3bc8806_DISS-9d988c3c-1891-0cf3-e044-00144f67d249.html#AGGR-2a5643b9-54fc-44b4-8919-e1c9f3bc8806 2012-04-17. 7 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9d988c3c-1891-0cf3-e044-00144f67d249/AGGR-
ee8516f2-7022-486c-9524-d3aab2a26566_DISS-9d988c3c-1891-0cf3-e044-00144f67d249.html#AGGR-ee8516f2-7022-486c-9524-d3aab2a26566 2012-04-17. 8 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9d988c3c-1891-0cf3-e044-00144f67d249/AGGR-01d56da3-463d-48b6-bad7-b6d9f6ff381c_DISS-9d988c3c-1891-0cf3-e044-00144f67d249.html#AGGR-01d56da3-463d-48b6-bad7-b6d9f6ff381c 2012-04-17. 9 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9d988c3c-1891-0cf3-e044-00144f67d249/AGGR-3448eb5e-b51a-47ad-bac2-3b7a45990ba4_DISS-9d988c3c-1891-0cf3-e044-00144f67d249.html#AGGR-3448eb5e-b51a-47ad-bac2-3b7a45990ba4 2012-04-17.
51
Ecotoxicological information
Species Effect
level
Test
conditions
Exposure
time
Nominal/
measured
Conc. in mg/L
(95% CI)
Reference
Danio rerio (fish) LC50 Freshwater 96 h Nominal 0.536 (0.486-0.589)
11
Danio rerio (fish) NOEC Freshwater 96 h Nominal 0.250 11
Danio rerio (fish) LC50 Standard
water 48 h Nominal 0.42 12
Phimephales
promelas (fish) LC50 Freshwater 96 h Measured 0.103 U.S. EPA,
2012 (Cas 124-22-1)
Daphnia magna
(crustacean) EC50 Freshwater 48 h Nominal 0.146 (95%CL:
0.136-0.157)
13
Pseudokirchnerella
subcapitata (algae) EC50 Freshwater 72 h Nominal 0.0516 (95%Cl:
0.484-0.552)
14
Pseudokirchnerella
subcapitata (algae) NOEC Freshwater 72 h Nominal 0.0125 14
Vibrio fisheri
(bacteria) EC50, resp.rate
Freshwater 3 h Nominal 28.2 15
DODECYLAMINE + 2EO
Physical and chemical properties
CAS number 1541-67-9
Molecular formula C16H35NO2
Molecular weight 273.52 g/mole Log Kow (modelled) 3.9 (U.S. EPI Suite, 2011)
pKa (determined with SPME) ~8.6 (Chen et al., 2012)
Ecotoxicological information
Species Effect
level
Test
conditions
Exposure
time
Nominal/
measured
Conc. in mg/L
(95% CI)
Reference
Fish LC50 Standard water
96 h Nominal 0.3 AkzoNobel, 2012:b
10
http://apps.echa.europa.eu/registered/data/dossiers/DISS-9d988c3c-1891-0cf3-e044-00144f67d249/AGGR-1c594938-0542-40b5-9dda-dc44f52b7267_DISS-9d988c3c-1891-0cf3-e044-00144f67d249.html#AGGR-1c594938-0542-40b5-9dda-dc44f52b7267 2012-04-14. 11 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9d988c3c-1891-0cf3-e044-00144f67d249/AGGR-3a43b46f-3a8b-4219-b768-b73746294db3_DISS-9d988c3c-1891-0cf3-e044-00144f67d249.html#AGGR-3a43b46f-3a8b-4219-b768-b73746294db3 2012-04-17. 12 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9d988c3c-1891-0cf3-e044-00144f67d249/AGGR-58d6540c-c04c-415e-acc5-96c13bd28f37_DISS-9d988c3c-1891-0cf3-e044-00144f67d249.html#AGGR-58d6540c-c04c-415e-acc5-96c13bd28f37 2012-04-17. 13
http://apps.echa.europa.eu/registered/data/dossiers/DISS-9d988c3c-1891-0cf3-e044-00144f67d249/AGGR-8a7da964-d387-40dd-bbd7-0b6ba9a08c11_DISS-9d988c3c-1891-0cf3-e044-00144f67d249.html#AGGR-8a7da964-d387-40dd-bbd7-0b6ba9a08c11 2012-05-21 14 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9d988c3c-1891-0cf3-e044-00144f67d249/AGGR-b55ae8ef-5706-4a73-b744-97ce42eeacfd_DISS-9d988c3c-1891-0cf3-e044-00144f67d249.html#AGGR-b55ae8ef-5706-4a73-b744-97ce42eeacfd 2012-04-17. 15 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9d988c3c-1891-0cf3-e044-00144f67d249/AGGR-07553001-1aa9-4ddf-abef-89d66892d58b_DISS-9d988c3c-1891-0cf3-e044-00144f67d249.html#AGGR-07553001-1aa9-4ddf-abef-89d66892d58b 2012-04-17.
52
HEXADECYLAMINE + 2EO
Physical and chemical properties
CAS number 18924-67-9 Molecular formula C20H43NO2
Molecular weight 329.56 g/mole Log Kow (modelled) 5.86 (U.S. EPI Suite, 2011)
OCTADECYLAMINE + 2EO
Physical and chemical properties
CAS number 10213-78-2
Molecular formula C22H47NO2 Molecular weight 357.61 g/mole
Log Kow (modelled) 6.85 (U.S. EPI Suite, 2011)
Figure 24. Chemical structure of octadecylamine +2EO.
ETHOMEEN C/12
Physical and chemical properties
CAS number 61791-31-9
EC number 263-163-9 Molecular formula R-N(CH2CH2O)Hm(CH2CH2O)Hn Physical state (20°C, 1013 hPa) Liquid, light yellowish*
Log Pow (25°C) 0.7* pKa (25°C) 8.816 (It is assumed that length of the hydrophobe has no
significant effect on the pKa values.)
pH 9-11* Water solubility (23°C, pH 7)
Surface tension (1.0 g/L, 23°C)
Vapour pressure (20°C) <0.1 hPa* Biodegradation in water (Guideline 301 D) > 60% (readily biodegradable)*
Bioaccumulation (BCF) No bioaccumulation is expected*
*=AkzoNobel, 2011.
Table 5. Distribution of single substances present in Ethomeen C/12 (*AkzoNobel, 2012:c)
Fatty acid chain
length
Distribution*
(%)
Distribution (%)
after SPME in DSW
Distribution (%) after
SPME in HD
Distribution (%) after
SPME in DSW+HA
C8 + 2EO 5 - - - C10 + 2EO 6 - - -
C12 + 2EO 50 70.6 68.5 86.8
C14 + 2EO 19 19.5 24.2 12.2 C16 + 2EO 10 9.85 7.24 0.986
C18 + 2EO 10 - - -
Total 100 100 100 100
16 http://apps.echa.europa.eu/registered/data/dossiers/DISS-98761353-424b-6f67-e044-00144f67d031/AGGR-0e2c3d82-593c-4057-aadf-23d0a5ea7d5a_DISS-98761353-424b-6f67-e044-00144f67d031.html#AGGR-0e2c3d82-593c-4057-aadf-23d0a5ea7d5a 2012-05-08.
53
From the SPME measurements on freely dissolved concentration of single substances present in
Ethomeen C/12, following substances were detectable with LC/MS; dodecylamine +2EO (C12+2EO),
tetradecylamine +2EO (C14+2EO) and hexadecylamine +2EO (C16+2EO). According to the table
above they constitute in total 79% of the mixture, but are regarded as 100%. Hence, the new
distribution in each test media is given by the measured concentration of a single substance divided
by the sum of the measured concentrations for the three detected single substances in that test
medium. The new distribution is used in calculations of the TU for the single substances regarding
measured toxicity.
Ecotoxicological information
Species Effect level Test
conditions
Exposure
time
Nominal/
measured
Conc.
(mg/L)
Reference
Brachydanio rerio
(fish) LC50 - 96 h - > 0.1 - 1 AkzoNobel,
2011
Fish (Ethomeen 14/12: pure C14+2EO)
LC50 Standard water
96 h Nominal 0.2 AkzoNobel, 2012:b
Fish LC50 Standard water
96 h Nominal 0.5 AkzoNobel, 2012:b
Fish LC50 Standard water
96 h Nominal 0.6 AkzoNobel, 2012:b
Daphnia magna
(crustacean) (Ethomeen 14/12: pure C14+2EO)
EC50 Standard water
48 h Nominal 0.17 AkzoNobel, 2012:b
Daphnia magna
(crustacean) EC50 Standard
water 48 h Nominal 1.4 AkzoNobel,
2012:b
Daphnia magna
(crustacean) EC50 Standard
water 48 h Nominal 0.84 AkzoNobel,
2012:b
Daphnia magna
(crustacean) EC50 - 48 h - > 0.1 – 1 AkzoNobel,
2011
Daphnia magna
(crustacean) EC50 Freshwater
(HD) 21 d Nominal 0.405 AkzoNobel,
2012:b Daphnia magna
(crustacean) EC10 Freshwater
(HD) 21 d Nominal 0.279 AkzoNobel,
2012:b Algae EC50 Freshwater
(HD) 72 h Nominal 0.107 AkzoNobel,
2012:b
Algae EC10 Freshwater (HD)
72 h Nominal 0.00916 AkzoNobel, 2012:b
DIDODECYLDIMETHYLAMMONIUM BROMIDE (DDAB)
Physical and chemical properties
CAS number 3282-73-3
EC number 221-923-7 Molecular formula C26H56BrN Molecular weight 462.63 g/mole
Log Kow (modelled) 6.62 (U.S. EPI Suite, 2011)
Ecotoxicological information
Species Effect level Test
conditions
Exposure
time
Nominal/
measured
Conc.
(mg/L)
Reference
Echinogammarus
tibaldii (crustacean) LC50 Freshwater 24 h Nominal 1.2 U.S. EPA,
2012 (Cas
54
3282-73-3)
ARQUAD 2C-75
Physical and chemical properties
CAS number 68391-05-9 EC number 269-924-1
IUPAC name N-C12-C18(even numbered)-alkyl-N,N-dimethyl-C12-C18(even numbered)-alkyl-1-aminium chloride
Molecular formula C30H46ClN
Molecular weight 474.3 g/mole
Physical state (20°C, 1013 hPa) Solid paste in the form of lumps17
Log Pow 4.8 18
pKa -
Water solubility (23°C, pH 7) 0.2 g/L19
Surface tension (0.2 g/L, 23°C) 28 mN/m20
Vapour pressure (25°C) 6.36E-13 hPa21 (calculated)
Dissociation constant (Kd) Di-coco C12-18 salts are at the whole pH range fully dissociated in water, consisting of ionised cationic surfactant and chloride as counter ion.
Biodegradation in water (Guideline 301 B) 61 % after 28 days22 (readily biodegradable)
Biodegradation in water (Guideline 301 D) 37 % after 28 days23
(not readily biodegradable) Closed bottle test inoculated with seawater
Biodegradation in water and sediment (OECD Guideline 303 A)
Biodegrades 96 %24 in properly operating conventional biological wastewater treatment plants.
Bioaccumulation (BCF) 70.79 L/kg25
wet weight Koc (adsorption/desorption) 7.32E+08 L/kg
Distribution in air, water, soil and sediment 0.019, 2.63, 37.46 and 59.8726 respectively
17
http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-c6018cf1-cd11-4e1e-a3b7-d58f51d9d8bf_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-c6018cf1-cd11-4e1e-a3b7-d58f51d9d8bf 2012-02-01. 18 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-51811570-9c20-4549-a945-7c1e40e5d905_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-51811570-9c20-4549-a945-7c1e40e5d905 2012-01-25. 19
http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-8534968c-5d8a-4194-81a1-6717f80afbd4_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-8534968c-5d8a-4194-81a1-6717f80afbd4 2012-01-25. 20
http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-0e02480a-8ba2-485b-88af-4b4023612dd9_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-0e02480a-8ba2-485b-88af-4b4023612dd9 2012-02-01. 21 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-80b9c19c-d608-491e-bf57-620759a5d49a_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-80b9c19c-d608-491e-bf57-620759a5d49a 2012-02-01. 22 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-2b150350-309a-4598-b9a7-aafb4498b83f_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-2b150350-309a-4598-b9a7-aafb4498b83f 2012-02-01. 23
http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-f5ea32ee-6d59-4d45-a4dc-993b2d5e2576_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-f5ea32ee-6d59-4d45-a4dc-993b2d5e2576 2012-02-01. 24 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-19c76d28-8d38-4f03-abe9-99b21cf636cb_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-19c76d28-8d38-4f03-abe9-99b21cf636cb 2012-02-01. 25 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-3905d869-ca88-4464-a52d-3c32508d690b_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-3905d869-ca88-4464-a52d-3c32508d690b 2012-02-01.
55
Ecotoxicological information
Species Effect level Test
conditions
Exposure
time
Nominal/
measured
Conc. in
mg/L (95%
CI)
Reference
Danio rerio (fish) LC50 Standard water
48 h Nominal 0.3 (0.27-0.32)
27
Danio rerio (fish) LC50 Standard water
96 h Nominal 0.26 (0.24-0.29)
27
Danio rerio (fish) NOEC Standard
water 96 h Nominal 0.23 27
Danio rerio (fish) LC50 Standard water
96 h Nominal 0.66 28
Cyprinodon
variegatus (fish) LC50 Saltwater 96 h Nominal 0.696 (98%
CL: 0.470-1.030)
29
Cyprinodon
variegatus (fish) NOEC Saltwater 96 h Nominal 0.47
29
Cyprinodon
variegatus (fish) LC50 Saltwater 96 h Nominal 0.787 (98%
CL: 0.530-1.170)
30
Cyprinodon
variegatus (fish) NOEC Saltwater 96 h Nominal 0.53
30
Acartia tonsa
(crustacean) LC50 Saltwater 48 h Nominal 0.295
(0.237-0.367)
31
Acartia tonsa
(crustacean) NOEC Saltwater 48 h Nominal 0.10
31
Daphnia magna
(crustacean) EC10 Freshwater
(HD) 21 days Measured
(initial) 1.15 32
Daphnia magna
(crustacean) NOEC, reprod.
Freshwater (HD)
21 days Measured (initial)
0.5 32
Phaeodactylum EC50, Saltwater 72 h Nominal 0.06 (0.06- 33
26
http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-ed88def5-8b6a-4ec9-b667-fabb0cbc50df_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-ed88def5-8b6a-4ec9-b667-fabb0cbc50df 2012-02-01. 27
http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-4c0d7e93-b696-4808-8b80-f938879bbbbf_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-4c0d7e93-b696-4808-8b80-f938879bbbbf 2012-02-01. 28 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-7f3efcf4-52a8-4ce0-8889-362202890283_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-7f3efcf4-52a8-4ce0-8889-362202890283 2012-02-01. 29 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-9cfe05fc-e5b2-41d5-bb5e-e5406131c3e8_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-9cfe05fc-e5b2-41d5-bb5e-e5406131c3e8 2012-02-01. 30
http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-66e67983-3971-47a8-a8bc-a03e334312e4_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-66e67983-3971-47a8-a8bc-a03e334312e4 2012-02-01. 31 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-ffc2ae8e-e29c-4e07-80c5-09d973428095_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-ffc2ae8e-e29c-4e07-80c5-09d973428095 2012-04-17. 32 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-8ad470f1-962e-4acf-a633-f3ed3b1be926_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-8ad470f1-962e-4acf-a633-f3ed3b1be926 2012-04-17.
56
tricornutum (algae) biomass 0-08) Phaeodactylum
tricornutum (algae) EC50, growth rate
Saltwater 72 h Nominal 0.13 (0.11-0.14)
33
Phaeodactylum
tricornutum (algae) NOEC Saltwater 72 h Nominal 0.009
33
Pseudokirchnerella
subcapitata (algae) EC50, growth rate
Freshwater (HD)
72 h Measured (initial)
0.386 (0.236-0.618)
34
Pseudokirchnerella
subcapitata (algae) EC10, growth rate
Freshwater (HD)
72 h Measured (initial)
0.13 (0.022-0.224)
34
Pseudokirchnerella
subcapitata (algae) NOEC, growth rate
Freshwater (HD)
72 h Measured (initial)
0.06 34
Pseudokirchnerella
subcapitata (algae) EC50, biomass
Freshwater (HD)
72 h Measured (initial)
0.148 (0.088-0.217)
34
Pseudokirchnerella
subcapitata (algae) EC10, biomass
Freshwater (HD)
72 h Measured (initial)
0.062 (0.006-0.0985)
34
activated sludge of a predominantly domestic sewage
EC50 Freshwater 3 h Nominal 68 (15 and 858)
35
Corophium sp. LC50 Sediment 10 days Nominal 850 36
Corophium sp. NOEC, mortality
Sediment 10 days Nominal 320 36
Abra alva (mollusc) EC50, fecal pellet production
Sediment, saltwater
120h Nominal 45 146 (±8110)
37
33 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-9f928cef-2dd7-4103-9553-4399ad9a140f_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-9f928cef-2dd7-4103-9553-4399ad9a140f 2012-04-17. 34 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-fafe5847-6867-4ff7-b43c-cb5e9bc36915_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-fafe5847-6867-4ff7-b43c-cb5e9bc36915 2012-02-01. 35
http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-303074d9-76a1-4b0b-9d8d-b67711418b79_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-303074d9-76a1-4b0b-9d8d-b67711418b79 2012-02-01. 36 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-1bc91ada-bfca-4990-8302-8ad27637dc44_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-1bc91ada-bfca-4990-8302-8ad27637dc44 2012-04-10. 37 http://apps.echa.europa.eu/registered/data/dossiers/DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031/AGGR-d45cb248-c01d-4961-bc85-27169b28e0ea_DISS-9fea2ec8-51a7-3ba2-e044-00144f67d031.html#AGGR-d45cb248-c01d-4961-bc85-27169b28e0ea 2012-04-10.
57
APPENDIX C: Specifications for natural river water and Dutch standard
water
Natural river water (HD)
Location Heveadorp (HD) at Fonteinallee, Doorwerth (Gelderland)
Sampling date 2012-02-20 Weather on the day of sampling Sunny, ca. 12 °C Colour Yellowish, clear pH 7.8
Conductivity [µS/cm] 283
Ca2+ [mg/L] 34.3
Mg2+ [mg/L] 3.27
Dissolved Oxygen [mg O2/L] 9.0
TOC [mg C/L] 2.21
Suspended matter [mg/L] 2.4
Hardness [°dH] 5.56
Figure 25. Sampling site for river water (marked with red balloon).
Dutch Standard Water (DSW)
Conductivity [µS/cm] 617
Ca2+ [mg/L] 61.5
Mg2+ [mg/L] 19.1
Hardness [°dH] 13.0
Modified Dutch Standard Water (0.45 ml/L salts added)
Conductivity [µS/cm] 291
Ca2+ [mg/L] 28.5
58
Mg2+ [mg/L] 11.0
Hardness [°dH] 6.53
59
APPENDIX D: Preparations, raw data and results from ToxCalc.
V5.0.23.
DODECYLAMINE
Preparation of stock solution
A stock solution of 100 mg/l was prepared by loading 0.0100 gram of dodecylamine, weighed out on
an analytical balance and then slightly heated in a water bath before filled up to the appropriate
volume (100 ml) with de-ionized water. The solution was then stirred and sonicated whilst on ice for
maximum two minutes until a homogenous solution was formed. The pH was adjusted with
hydrochloric acid (1 M) to approximately 5.0.
Table 6. Oxygen (mg/L), pH and temperature (°C) measurements of dodecylamine at the start of the test (T=0h) and at the end of the test (T=48h).
O2 (mg/L) pH Temp. (°C)
T=0h T=48h T=0h T=48h T=0h T=48h
DSW Control 9.2 8.5 8.2 8.0 20.5 20.5
0.03 mg/L 9.2 8.5 8.1 8.1 - -
0.48 mg/L 9.2 8.4 8.1 8.1 - - HD Control 8.8 8.9 8.0 8.0 20.5 20.5
0.10 mg/L 8.8 8.7 8.1 8.1 - - 1.6 mg/L 8.8 8.5 8.1 8.0 - -
Table 7. Nominal toxicity of dodecylamine to Daphnia magna in DSW.
Replicate
Nominal
test conc.
(mg/L) T=0h T=24h T=48h
1 0,000 5 5 5
2 0,000 5 5 5
3 0,000 5 5 5
4 0,000 5 5 5
1 0,030 5 5 5
2 0,030 5 5 5
3 0,030 5 5 5
4 0,030 5 5 5
1 0,060 5 5 5
2 0,060 5 5 4
3 0,060 5 5 3
4 0,060 5 5 4
1 0,120 5 4 2
2 0,120 5 5 0
3 0,120 5 5 2
4 0,120 5 5 0
1 0,240 5 3 0
2 0,240 5 5 0
3 0,240 5 5 0
4 0,240 5 5 0
1 0,480 5 0 0
2 0,480 5 0 0
Figure 26. Dose-response relationship (nominal) for dodecylamine to Daphnia magna in DSW.
60
3 0,480 5 0 0
4 0,480 5 0 0 Table 8. Nominal toxicity of dodecylamine to Daphnia magna in HD water.
Replicate
Nominal
test conc.
(mg/L) T=0h T=24h T=48h
1 0,000 5 5 5
2 0,000 5 5 5
3 0,000 5 5 5
4 0,000 5 5 5
1 0,100 5 5 5
2 0,100 5 5 5
3 0,100 5 5 5
4 0,100 5 5 5
1 0,200 5 5 5
2 0,200 5 5 5
3 0,200 5 5 5
4 0,200 5 5 5
1 0,400 5 5 5
2 0,400 5 5 5
3 0,400 5 5 5
4 0,400 5 5 5
1 0,800 5 2 0
2 0,800 5 3 0
3 0,800 5 3 0
4 0,800 5 3 0
1 1,600 5 0 0
2 1,600 5 0 0
3 1,600 5 0 0
4 1,600 5 0 0
DODECYLAMINE +2EO
Preparation of stock solution
A stock solution of 109 mg/l was prepared by loading 0.0109 gram of dodecylamine +2EO, weighed
out on an analytical balance and then filled up with approximately 80 ml of de-ionized water. The
solution was then stirred while pH was checked and adjusted to 4.6 with hydrochloric acid (1M). The
solution was sonicated whilst on ice for maximum two minutes until a homogenous solution was
formed. The pH was checked and adjusted again with sodium hydroxide (1 M) to approximately 7.5.
At last it was filled up to the appropriate volume (100 ml) with de-ionized water.
Table 9. Oxygen (mg/L), pH and temperature (°C) measurements of dodecylamine +2EO at the start of the test (T=0h) and at the end of the test (T=48h).
O2 (mg/L) pH Temp. (°C)
T=0h T=48h T=0h T=48h T=0h T=48h
DSW Control 8.7 8.8 8.1 7.9 21.4 22.3
0.08 mg/L 8.6 8.8 8.1 7.9 - - 1.28 mg/L 8.8 8.8 8.2 7.9 - -
HD Control 9.0 8.9 7.9 7.7 21.4 22.1
0.4 mg/L 9.1 8.9 7.9 7.7 - -
Figure 27. Dose-response relationship (nominal) for dodecylamine to Daphnia magna in HD water.
61
4.2 mg/L 9.2 8.9 7.9 7.6 - - DSW + HA
Control
8.8 8.9 8.0 7.9 21.5 22.1
0.2 mg/L 8.8 8.8 8.0 7.9 - - 3.2 mg/L 8.8 8.8 8.0 7.9 - -
HD 600
Control
8.7 8.9 8.0 8.0 21.4 22.0
0.4 mg/L 8.7 8.8 8.0 8.0 - - 4.2 mg/L 8.7 8.8 8.0 8.0 - -
Table 10. Toxicity of dodecylamine+2EO to Daphnia magna in DSW.
Replicate
Nominal test
conc. (mg/L)
Measured test
conc. (mg/L) T=0h T=24h T=48h
1 0,000 0,000 5 5 5
2 0,000 0,000 5 5 5
3 0,000 0,000 5 5 5
4 0,000 0,000 5 5 5
1 0,080 0,009 5 4 3
2 0,080 0,009 5 5 5
3 0,080 0,009 5 5 5
4 0,080 0,009 5 5 5
1 0,160 0,052 5 5 5
2 0,160 0,052 5 5 5
3 0,160 0,052 5 5 5
4 0,160 0,052 5 5 5
1 0,320 0,144 5 5 5
2 0,320 0,144 5 5 5
3 0,320 0,144 5 5 5
4 0,320 0,144 5 5 5
1 0,640 0,357 5 5 5
2 0,640 0,357 5 5 2
3 0,640 0,357 5 4 3
4 0,640 0,357 5 5 2
1 1,280 0,732 5 5 0
2 1,280 0,732 5 4 0
3 1,280 0,732 5 5 0
4 1,280 0,732 5 3 0
Figure 28. Dose-response relationship (left: nominal, right: measured) of dodecylamine +2EO to Daphnia magna in DSW.
62
Table 11. Toxicity of dodecylamine+2EO to Daphnia magna in HD water.
Replicate
Nominal test
conc. (mg/L)
Measured test
conc. (mg/L) T=0h T=24h T=48h
1 0,000 0,000 5 5 5
2 0,000 0,000 5 4 4
3 0,000 0,000 5 4 4
4 0,000 0,000 5 5 5
1 0,400 0,129 5 5 5
2 0,400 0,129 5 5 5
3 0,400 0,129 5 5 5
4 0,400 0,129 5 5 5
1 0,720 0,272 5 5 5
2 0,720 0,272 5 5 5
3 0,720 0,272 5 5 5
4 0,720 0,272 5 5 5
1 1,300 0,508 5 5 4
2 1,300 0,508 5 5 5
3 1,300 0,508 5 5 4
4 1,300 0,508 5 5 4
1 2,300 0,912 5 3 0
2 2,300 0,912 5 2 0
3 2,300 0,912 5 1 0
4 2,300 0,912 5 2 0
1 4,200 1,806 5 2 0
2 4,200 1,806 5 1 0
3 4,200 1,806 5 0 0
4 4,200 1,806 5 0 0
Figure 29. Dose-response relationship (left: nominal, right: measured) of dodecylamine +2EO to Daphnia magna in HD water.
Table 12. Toxicity of dodecylamine+2EO to Daphnia magna in HD water with a conductivity of 600 µS/cm.
Replicate
Nominal test
conc. (mg/L)
Measured test
conc. (mg/L) T=0h T=24h T=48h
1 0,000 0,000 5 5 5
2 0,000 0,000 5 5 5
3 0,000 0,000 5 5 5
4 0,000 0,000 5 5 5
1 0,400 0,118 5 5 5
2 0,400 0,118 5 5 5
63
3 0,400 0,118 5 5 5
4 0,400 0,118 5 5 5
1 0,720 0,307 5 5 5
2 0,720 0,307 5 5 4
3 0,720 0,307 5 5 4
4 0,720 0,307 5 5 5
1 1,300 0,679 5 5 2
2 1,300 0,679 5 4 3
3 1,300 0,679 5 2 0
4 1,300 0,679 5 4 1
1 2,300 1,414 5 1 0
2 2,300 1,414 5 0 0
3 2,300 1,414 5 0 0
4 2,300 1,414 5 0 0
1 4,200 2,443 5 0 0
2 4,200 2,443 5 0 0
3 4,200 2,443 5 0 0
4 4,200 2,443 5 0 0
Figure 30. Dose-response relationship (left: nominal, right: measured) of dodecylamine +2EO to Daphnia magna in HD water with a conductivity of 600 µS/cm.
Table 13. Toxicity of dodecylamine+2EO to Daphnia magna in DSW + 20 mg/l HA.
Replicate
Nominal test
conc. (mg/L)
Measured test
conc. (mg/L) T=0h T=24h T=48h
1 0,000 0,000 5 5 5
2 0,000 0,000 5 5 5
3 0,000 0,000 5 5 5
4 0,000 0,000 5 5 5
1 0,200 0,034 5 5 5
2 0,200 0,034 5 5 5
3 0,200 0,034 5 5 5
4 0,200 0,034 5 5 5
1 0,400 0,103 5 5 5
2 0,400 0,103 5 5 5
3 0,400 0,103 5 5 5
4 0,400 0,103 5 5 5
1 0,800 0,275 5 5 5
2 0,800 0,275 5 5 5
3 0,800 0,275 5 5 5
64
4 0,800 0,275 5 5 5
1 1,600 0,667 5 4 0
2 1,600 0,667 5 5 0
3 1,600 0,667 5 5 0
4 1,600 0,667 5 3 0
1 3,200 1,500 5 1 0
2 3,200 1,500 5 1 0
3 3,200 1,500 5 0 0
4 3,200 1,500 5 3 0
Figure 31. Dose-response relationship (left: nominal, right: measured) of dodecylamine +2EO to Daphnia magna in DSW + 20 mg/L HA.
HEXADECYLAMINE +2EO
Preparation of stock solution
A stock solution of 89.98 mg/l was prepared by loading 0.0098 gram of hexadecylamine +2EO,
weighed out on an analytical balance and then filled with approximately 80 ml of de-ionized water.
While stirring the solution, pH was checked to be 5.5 and adjusted to 3.1 with hydrochloric acid (1M).
The solution was then sonicated whilst on ice for maximum two minutes until a homogenous
solution was formed. The pH was adjusted with sodium hydroxide (1 M) to 7.9. A total volume of
108.91 ml of de-ionized water was added to achieve a 89.98 mg/L stock solution.
Table 14. Oxygen (mg/L), pH and temperature (°C) measurements of hexadecylamine +2EO at the start of the test (T=0h) and at the end of the test (T=48h).
O2 (mg/L) pH Temp. (°C)
T=0h T=48h T=0h T=48h T=0h T=48h
DSW Control 8.7 8.5 8.0 7.9 22.8 21.7
0.009 mg/L 8.8 8.5 8.0 8.0 - -
0.144 mg/L 8.8 8.5 8.0 8.0 - - DSW + HA
Control
8.6 8.5 8.4 8.0 22.7 21.4
0.08 mg/L 8.6 8.5 8.4 8.0 - -
1.28 mg/L 8.6 8.5 8.4 8.0 - -
New stock solution
A stock solution of 85 mg/l was prepared by loading 0.0085 gram of hexadecylamine +2EO, weighed
out on an analytical balance and then filled with approximately 80 ml of de-ionized water. While
stirring the solution, pH was checked to be 6.4 and adjusted to 3.1 with hydrochloric acid (1M). The
solution was then sonicated whilst on ice for maximum two minutes until a homogenous solution
65
was formed. The pH was adjusted with sodium hydroxide (1 M) to 7.2. At last it was filled up to the
appropriate volume (100 ml) with de-ionized water.
Table 15. Oxygen (mg/L), pH and temperature (°C) measurements of hexadecylamine +2EO at the start of the test (T=0h) and at the end of the test (T=48h).
O2 (mg/L) pH Temp. (°C)
T=0h T=48h T=0h T=48h T=0h T=48h
HD Control 9.3 8.8 8.1 7.8 21.2 20.2 0.03 mg/L 9.7 8.9 8.3 7.8 - -
0.31 mg/L 9.7 8.6 8.3 7.8 - -
Table 16. Toxicity of hexadecylamine+2EO to Daphnia magna in DSW.
Replicate
Nominal test
conc. (mg/L)
Measured test
conc, (mg/L) T=0h T=24h T=48h
1 0,000 0,000 5 5 5
2 0,000 0,000 5 5 5
3 0,000 0,000 5 5 5
4 0,000 0,000 5 5 5
1 0,009 0,001 5 5 5
2 0,009 0,001 5 5 5
3 0,009 0,001 5 5 5
4 0,009 0,001 5 5 5
1 0,018 0,001 5 5 5
2 0,018 0,001 5 5 5
3 0,018 0,001 5 5 5
4 0,018 0,001 5 5 5
1 0,036 0,003 5 5 1
2 0,036 0,003 5 5 1
3 0,036 0,003 5 5 0
4 0,036 0,003 5 5 1
1 0,072 0,003 5 3 0
2 0,072 0,003 5 2 0
3 0,072 0,003 5 3 0
4 0,072 0,003 5 4 0
1 0,144 0,014 5 3 0
2 0,144 0,014 5 4 0
3 0,144 0,014 5 0 0
4 0,144 0,014 5 0 0
66
Figure 32. Dose-response relationship (left: nominal, right: measured) of hexadecylamine +2EO to Daphnia magna in DSW.
Table 17. Toxicity of hexadecylamine+2EO to Daphnia magna in HD water.
Replicate
Nominal test
conc. (mg/L)
Measured test conc.
(normal DSW
cal.curve) (mg/L)
Measured test conc.
(modified DSW
cal.curve) (mg/L) T=0h T=24h T=48h
1 0,000 0,000 0,000 5 5 5
2 0,000 0,000 0,000 5 5 5
3 0,000 0,000 0,000 5 5 5
4 0,000 0,000 0,000 5 5 5
1 0,030 0,005 0,000 5 5 5
2 0,030 0,005 0,000 5 5 5
3 0,030 0,005 0,000 5 5 5
4 0,030 0,005 0,000 5 5 5
1 0,054 0,012 0,001 5 5 5
2 0,054 0,012 0,001 5 4 4
3 0,054 0,012 0,001 5 5 5
4 0,054 0,012 0,001 5 5 5
1 0,097 0,037 0,004 5 5 3
2 0,097 0,037 0,004 5 5 4
3 0,097 0,037 0,004 5 5 2
4 0,097 0,037 0,004 5 5 4
1 0,170 0,072 0,012 5 5 0
2 0,170 0,072 0,012 5 4 0
3 0,170 0,072 0,012 5 5 1
4 0,170 0,072 0,012 5 5 2
1 0,310 0,096 0,024 5 5 0
2 0,310 0,096 0,024 5 4 0
3 0,310 0,096 0,024 5 4 0
4 0,310 0,096 0,024 5 5 0
67
Figure 33. Dose-response relationship (nominal) of hexadecylamine +2EO to Daphnia magna in HD water.
Figure 34. Dose-response relationship (measured) of hexadecylamine +2EO to Daphnia magna in HD water (left: related to normal DSW calibration curve, right: related to modified DSW calibration curve).
Table 18. Toxicity of hexadecylamine+2EO to Daphnia magna in DSW + 20 mg/L HA.
Replicate
Nominal test
conc. (mg/L)
Measured test
conc. (mg/L) T=0h T=24h T=48h
1 0,000 0,000 5 5 5
2 0,000 0,000 5 5 5
3 0,000 0,000 5 5 5
4 0,000 0,000 5 5 5
1 0,080 0,001 5 5 5
2 0,080 0,001 5 5 4
3 0,080 0,001 5 5 5
4 0,080 0,001 5 5 5
1 0,160 0,001 5 5 5
2 0,160 0,001 5 5 5
3 0,160 0,001 5 5 5
4 0,160 0,001 5 5 5
1 0,320 0,003 5 5 5
2 0,320 0,003 5 5 3
3 0,320 0,003 5 4 4
4 0,320 0,003 5 5 5
1 0,640 0,008 5 5 1
2 0,640 0,008 5 5 1
3 0,640 0,008 5 5 0
68
4 0,640 0,008 5 5 0
1 1,280 0,022 5 5 0
2 1,280 0,022 5 4 0
3 1,280 0,022 5 5 0
4 1,280 0,022 5 3 0
Figure 35. Dose-response relationship (left: nominal, right: measured) of hexadecylamine +2EO to Daphnia magna in DSW + 20 mg/L HA.
OCTADECYLAMINE +2EO
Preparation of stock solution
A stock solution of 95.6 mg/l was prepared by loading 0.0096 gram of octadecylamine +2EO, weighed
out on an analytical balance and then filled with approximately 80 ml of with de-ionized water. The
solution was then stirred while adjusting the pH to 2.8 with hydrochloric acid (1M). The solution was
also sonicated whilst on ice for maximum two minutes until a homogenous solution was formed. The
pH was checked again and adjusted with sodium hydroxide (1 M) to 7.3. A total volume of 100.38 ml
of de-ionized water was added to achieve a 95.6 mg/L stock solution.
Table 19. Oxygen (mg/L), pH and temperature (°C) measurements of octadecylamine +2EO at the start of the test (T=0h) and at the end of the test (T=48h).
O2 (mg/L) pH Temp. (°C)
T=0h T=48h T=0h T=48h T=0h T=48h
DSW Control 8.9 8.8 8.3 8.1 21.0 20.0
0.008 mg/L 8.9 8.8 8.3 8.2 - - 0.128 mg/L 8.8 8.8 8.3 8.2 - -
HD Control 9.2 8.8 8.4 8.2 20.9 19.9
0.02 mg/L 9.7 8.8 8.4 8.1 - - 0.32 mg/L 9.7 8.9 8.4 8.0 - -
DSW + HA
Control
8.7 8.8 8.4 8.2 21.2 19.9
0.08 mg/L 8.7 8.8 8.4 8.2 - - 1.28 mg/L 8.7 8.7 8.4 8.2 - -
69
Table 20. Nominal toxicity of octadecylamine +2EO to Daphnia
magna in DSW.
Replicate
Nominal test
conc. (mg/L) T=0h T=24h T=48h
1 0,000 5 5 5
2 0,000 5 5 5
3 0,000 5 5 5
4 0,000 5 5 5
1 0,008 5 5 5
2 0,008 5 5 4
3 0,008 5 5 5
4 0,008 5 5 5
1 0,016 5 5 5
2 0,016 5 5 4
3 0,016 5 5 4
4 0,016 5 4 2
1 0,032 5 5 3
2 0,032 5 5 2
3 0,032 5 5 1
4 0,032 5 5 3
1 0,064 5 5 1
2 0,064 5 4 0
3 0,064 5 3 0
4 0,064 5 5 0
1 0,128 5 3 0
2 0,128 5 3 0
3 0,128 5 3 0
4 0,128 5 2 0 Table 21. Nominal toxicity of octadecylamine +2EO to Daphnia
magna in HD water.
Replicate
Nominal test
conc. (mg/L) T=0h T=24h T=48h
1 0,000 5 5 5
2 0,000 5 5 5
3 0,000 5 5 5
4 0,000 5 5 5
1 0,020 5 5 5
2 0,020 5 5 5
3 0,020 5 5 5
4 0,020 5 5 5
1 0,040 5 5 5
2 0,040 5 5 5
3 0,040 5 5 5
4 0,040 5 5 5
1 0,080 5 5 5
2 0,080 5 5 5
3 0,080 5 5 5
4 0,080 5 5 4
1 0,160 5 5 3
2 0,160 5 5 1
3 0,160 5 5 5
4 0,160 5 5 5
Figure 36. Dose-response relationship (nominal) of octadecylamine +2EO to Daphnia magna in DSW.
Figure 37. Dose-response relationship (nominal) of octadecylamine +2EO to Daphnia magna in HD water.
70
1 0,320 5 5 1
2 0,320 5 5 2
3 0,320 5 4 0
4 0,320 5 5 2
Table 22. Nominal toxicity of octadecylamine +2EO to Daphnia magna in DSW + 20 mg/L HA.
Replicate
Nominal test
conc. (mg/L) T=0h T=24h T=48h
1 0,000 5 5 5
2 0,000 5 5 5
3 0,000 5 5 5
4 0,000 5 5 5
1 0,080 5 5 5
2 0,080 5 5 5
3 0,080 5 5 5
4 0,080 5 5 5
1 0,160 5 5 5
2 0,160 5 5 5
3 0,160 5 5 5
4 0,160 5 5 5
1 0,320 5 5 5
2 0,320 5 5 5
3 0,320 5 5 5
4 0,320 5 5 5
1 0,640 5 5 5
2 0,640 5 5 4
3 0,640 5 5 5
4 0,640 5 5 4
1 1,280 5 5 4
2 1,280 5 4 0
3 1,280 5 5 4
4 1,280 5 5 4
ETHOMEEN C/12
Preparation of stock solution
A stock solution of 109.4 mg/l was prepared by loading 0.0124 gram of Ethomeen C/12, weighed out
on an analytical balance and then filled with approximately 80 ml of de-ionized water. The solution
was then stirred and pH was checked to be 8.0 and adjusted to 3.7 with hydrochloric acid (1M). The
solution was sonicated whilst on ice for maximum two minutes until a homogenous solution was
formed. The pH was checked and adjusted again with sodium hydroxide (1 M) to 7.6. A total volume
of 113.37 ml of de-ionized water was added to achieve a 109.4 mg/L stock solution.
Table 23. Oxygen (mg/L), pH and temperature (°C) measurements of Ethomeen C/12 at the start of the test (T=0h) and at the end of the test (T=48h).
O2 (mg/L) pH Temp. (°C)
T=0h T=48h T=0h T=48h T=0h T=48h
DSW Control 8.7 9.0 8.2 8.1 20.8 20.8 0.03 mg/L 8.8 9.1 8.2 8.1 - - 0.48 mg/L 8.8 9.1 8.3 8.1 - -
71
HD Control 10.4 9.1 8.0 7.9 21.0 20.9 0.15 mg/L 10.2 9.0 8.2 7.8 - -
2.4 mg/L 10.0 9.0 7.9 7.7 - -
DSW + HA
Control
8.6 9.2 8.3 8.1 20.7 20.8
0.2 mg/L 8.9 9.1 8.3 8.1 - -
3.2 mg/L 8.8 9.1 8.3 8.1 - -
Table 24. Toxicity of Ethomeen C/12 to Daphnia magna in DSW.
Replicate
Nominal
test
conc.
(mg/L)
Measured
test conc.
mixture
(mg/L)
Measured
test conc.
C12+2EO
(mg/L)
Measured
test conc.
C14+2EO
(mg/L)
Measured
test conc.
C16+2EO
(mg/L) T=0h T=24h T=48h
1 0,000 0,000 0,000 0,000 0,000 5 2 0
2 0,000 0,000 0,000 0,000 0,000 5 5 5
3 0,000 0,000 0,000 0,000 0,000 5 5 5
4 0,000 0,000 0,000 0,000 0,000 5 5 5
1 0,030 0,012 0,01 0,001 0,001 5 5 0
2 0,030 0,012 0,01 0,001 0,001 5 5 4
3 0,030 0,012 0,01 0,001 0,001 5 5 5
4 0,030 0,012 0,01 0,001 0,001 5 5 5
1 0,060 0,028 0,021 0,004 0,003 5 5 5
2 0,060 0,028 0,021 0,004 0,003 5 5 5
3 0,060 0,028 0,021 0,004 0,003 5 5 5
4 0,060 0,028 0,021 0,004 0,003 5 5 5
1 0,120 0,089 0,059 0,021 0,009 5 5 4
2 0,120 0,089 0,059 0,021 0,009 5 5 5
3 0,120 0,089 0,059 0,021 0,009 5 5 5
4 0,120 0,089 0,059 0,021 0,009 5 5 4
1 0,240 0,298 0,194 0,07 0,035 5 5 4
2 0,240 0,298 0,194 0,07 0,035 5 5 3
3 0,240 0,298 0,194 0,07 0,035 5 5 4
4 0,240 0,298 0,194 0,07 0,035 5 5 5
1 0,480 0,552 0,368 0,127 0,056 5 5 0
2 0,480 0,552 0,368 0,127 0,056 5 5 0
3 0,480 0,552 0,368 0,127 0,056 5 5 0
4 0,480 0,552 0,368 0,127 0,056 5 5 0
72
Figure 38. Dose-response relationship (nominal) of Ethomeen C/12 to Daphnia magna in DSW.
Figure 39. Dose-response relationship (measured) of Ethomeen C/12 (left: mixture, right: C12+2EO) to Daphnia magna in DSW.
Figure 40. Dose-response relationship (measured) of Ethomeen C/12 (left: C14+2EO, right: C16+2EO) to Daphnia magna in DSW.
73
Table 25. Toxicity of Ethomeen C/12 to Daphnia magna in HD water.
Replicate
Nominal test
conc. (mg/L)
Measured
test conc.
mixture
(mg/L)
Measured
test conc.
C12+2EO
(mg/L)
Measured
test conc.
C14+2EO
(mg/L)
Measured
test conc.
C16+2EO
(mg/L) T=0h T=24h T=48h
1 0,000 0,000 0,000 0,000 0,000 5 2 5
2 0,000 0,000 0,000 0,000 0,000 5 5 5
3 0,000 0,000 0,000 0,000 0,000 5 5 5
4 0,000 0,000 0,000 0,000 0,000 5 5 5
1 0,150 0,077 0,058 0,013 0,006 5 5 5
2 0,150 0,077 0,058 0,013 0,006 5 5 5
3 0,150 0,077 0,058 0,013 0,006 5 5 5
4 0,150 0,077 0,058 0,013 0,006 5 5 5
1 0,300 0,303 0,213 0,07 0,02 5 5 5
2 0,300 0,303 0,213 0,07 0,02 5 5 3
3 0,300 0,303 0,213 0,07 0,02 5 5 5
4 0,300 0,303 0,213 0,07 0,02 5 5 5
1 0,600 0,678 0,45 0,177 0,051 5 5 2
2 0,600 0,678 0,45 0,177 0,051 5 5 1
3 0,600 0,678 0,45 0,177 0,051 5 5 0
4 0,600 0,678 0,45 0,177 0,051 5 5 3
1 1,200 1,781 1,161 0,492 0,128 5 5 0
2 1,200 1,781 1,161 0,492 0,128 5 4 0
3 1,200 1,781 1,161 0,492 0,128 5 3 0
4 1,200 1,781 1,161 0,492 0,128 5 5 0
1 2,400 3,004 1,971 0,818 0,214 5 0 0
2 2,400 3,004 1,971 0,818 0,214 5 3 0
3 2,400 3,004 1,971 0,818 0,214 5 0 0
4 2,400 3,004 1,971 0,818 0,214 5 0 0
Figure 41. Dose-response relationship (nominal) of Ethomeen C/12 to Daphnia magna in HD water.
74
Figure 42. Dose-response relationship (measured) of Ethomeen C/12 (left: mixture, right: C12+2EO) to Daphnia magna in HD water.
Figure 43. Dose-response relationship (measured) of Ethomeen C/12 (left: C14+2EO, right: C16+2EO) to Daphnia magna in HD water.
Table 26. Toxicity of Ethomeen C/12 to Daphnia magna in DSW + 20 mg/L HA.
Replicate
Nominal
test conc.
(mg/L)
Measured
test conc.
mixture
(mg/L)
Measured
test conc.
C12+2EO
(mg/L)
Measured
test conc.
C14+2EO
(mg/L)
Measured
test conc.
C16+2EO
(mg/L) T=0h T=24h T=48h
1 0,000 0,000 0,000 0,000 0,000 5 5 5
2 0,000 0,000 0,000 0,000 0,000 5 5 5
3 0,000 0,000 0,000 0,000 0,000 5 5 5
4 0,000 0,000 0,000 0,000 0,000 5 5 5
1 0,200 0,076 0,069 0,006 0,000 5 5 5
2 0,200 0,076 0,069 0,006 0,000 5 5 5
3 0,200 0,076 0,069 0,006 0,000 5 5 5
4 0,200 0,076 0,069 0,006 0,000 5 5 5
1 0,400 0,230 0,204 0,024 0,002 5 5 5
2 0,400 0,230 0,204 0,024 0,002 5 5 3
3 0,400 0,230 0,204 0,024 0,002 5 5 5
4 0,400 0,230 0,204 0,024 0,002 5 5 5
1 0,800 0,624 0,543 0,073 0,008 5 5 5
2 0,800 0,624 0,543 0,073 0,008 5 5 5
3 0,800 0,624 0,543 0,073 0,008 5 5 3
4 0,800 0,624 0,543 0,073 0,008 5 5 5
1 1,600 1,864 1,585 0,256 0,023 5 4 0
75
2 1,600 1,864 1,585 0,256 0,023 5 5 0
3 1,600 1,864 1,585 0,256 0,023 5 5 0
4 1,600 1,864 1,585 0,256 0,023 5 5 0
1 3,200 3,457 2,818 0,587 0,053 5 1 0
2 3,200 3,457 2,818 0,587 0,053 5 0 0
3 3,200 3,457 2,818 0,587 0,053 5 1 0
4 3,200 3,457 2,818 0,587 0,053 5 0 0
Figure 44. Dose-response relationship (nominal) of Ethomeen C/12 to Daphnia magna in DSW + 20 mg/L HA.
Figure 45. Dose-response relationship (measured) of Ethomeen C/12 (left: mixture, right: C12+2EO) to Daphnia magna in DSW + 20 mg/L HA.
Figure 46. Dose-response relationship (measured) of Ethomeen C/12 (left: C14+2EO, right: C16+2EO) to Daphnia magna in DSW + 20 mg/L HA.
DIDODECYLDIMETHYLAMMONIUM BROMIDE
76
Preparation of stock solution
A stock solution of 130 mg/L was prepared by loading 0.0130 gram of didodecyldimethylammonium
bromide, weighed out on an analytical balance and then filled with approximately 80 ml of de-
ionized water. The solution was sonicated whilst on ice for maximum two minutes until a
homogenous solution was formed. The pH was checked to be 4.7 and adjusted with sodium
hydroxide (1 M) and hydrochloric acid (1M) to 8.5. A total volume of 100 ml of de-ionized water was
added to achieve a 130 mg/L stock solution.
Table 27. Oxygen (mg/L), pH and temperature (°C) measurements of didodecyldimethylammonium bromide at the start of the test (T=0h) and at the end of the test (T=48h).
O2 (mg/L) pH Temp. (°C)
T=0h T=48h T=0h T=48h T=0h T=48h
DSW Control 8.9 9.0 8.1 8.1 20.7 21.0
0.03 mg/L 8.9 8.9 8.2 8.1 - - 0.5 mg/L 8.9 8.9 8.3 8.1 - - HD Control 10.2 9.0 8.1 8.1 20.6 20.7 0.1 mg/L 10.2 9.0 8.0 7.9 - -
1.6 mg/L 10.2 9.0 8.0 7.9 - -
DSW + HA
Control
8.9 8.9 8.3 8.1 20.7 21.0
0.3 mg/L 9.0 8.9 8.3 8.1 - -
4.8 mg/L 8.9 8.9 8.4 8.1 - -
Table 28. Toxicity of Didodecyldimethylammonium bromide to Daphnia magna in DSW.
Replicate
Nominal test
conc. (mg/L)
Measured test
conc. (mg/L) T=0h T=24h T=48h
1 0,000 0,000 5 5 5
2 0,000 0,000 5 5 5
3 0,000 0,000 5 5 5
4 0,000 0,000 5 4 4
1 0,030 0,004 5 5 3
2 0,030 0,004 5 5 0
3 0,030 0,004 5 5 5
4 0,030 0,004 5 5 4
1 0,063 0,014 5 0 0
2 0,063 0,014 5 2 0
3 0,063 0,014 5 0 0
4 0,063 0,014 5 5 5
1 0,125 0,028 5 0 0
2 0,125 0,028 5 0 0
3 0,125 0,028 5 0 0
4 0,125 0,028 5 0 0
1 0,250 0,064 5 0 0
2 0,250 0,064 5 0 0
3 0,250 0,064 5 0 0
4 0,250 0,064 5 0 0
1 0,500 0,088 5 0 0
2 0,500 0,088 5 0 0
3 0,500 0,088 5 0 0
4 0,500 0,088 5 0 0
77
Figure 47. Dose-response relationship (left: nominal, right: measured) of didodecyldimethylammonium bromide to Daphnia
magna in DSW.
Table 29. Toxicity of Didodecyldimethylammonium bromide to Daphnia magna in HD water.
Replicate
Nominal test
conc. (mg/L)
Measured test
conc. (mg/L) T=0h T=24h T=48h
1 0,000 0,000 5 5 5
2 0,000 0,000 5 5 5
3 0,000 0,000 5 5 5
4 0,000 0,000 5 5 5
1 0,100 0,003 5 5 1
2 0,100 0,003 5 5 5
3 0,100 0,003 5 5 5
4 0,100 0,003 5 5 0
1 0,200 0,013 5 0 0
2 0,200 0,013 5 0 0
3 0,200 0,013 5 0 0
4 0,200 0,013 5 0 0
1 0,400 0,27 5 0 0
2 0,400 0,27 5 0 0
3 0,400 0,27 5 0 0
4 0,400 0,27 5 0 0
1 0,800 0,064 5 0 0
2 0,800 0,064 5 0 0
3 0,800 0,064 5 0 0
4 0,800 0,064 5 0 0
1 1,600 0,284 5 0 0
2 1,600 0,284 5 0 0
3 1,600 0,284 5 0 0
4 1,600 0,284 5 0 0
78
Figure 48. Dose-response relationship (left: nominal, right: measured) of didodecyldimethylammonium bromide to Daphnia
magna in HD water.
Table 30. Toxicity of Didodecyldimethylammonium bromide to Daphnia magna in DSW + 20 mg/L HA.
Replicate
Nominal test
conc. (mg/L)
Measured test
conc. (mg/L) T=0h T=24h T=48h
1 0,000 0,000 5 5 5
2 0,000 0,000 5 5 5
3 0,000 0,000 5 5 5
4 0,000 0,000 5 5 5
1 0,300 0,000 5 5 5
2 0,300 0,000 5 5 5
3 0,300 0,000 5 5 5
4 0,300 0,000 5 5 5
1 0,600 0,0013 5 5 5
2 0,600 0,0013 5 5 5
3 0,600 0,0013 5 5 5
4 0,600 0,0013 5 5 5
1 1,200 0,0056 5 5 2
2 1,200 0,0056 5 5 1
3 1,200 0,0056 5 5 0
4 1,200 0,0056 5 5 3
1 2,400 0,015 5 5 0
2 2,400 0,015 5 5 0
3 2,400 0,015 5 5 0
4 2,400 0,015 5 5 0
1 4,800 0,0351 5 2 0
2 4,800 0,0351 5 1 0
3 4,800 0,0351 5 1 0
4 4,800 0,0351 5 3 0
79
Figure 49. Dose-response relationship (left: nominal, right: measured) of didodecyldimethylammonium bromide to Daphnia
magna in DSW + 20 mg/L HA.
ARQUAD 2C-75
Preparation of stock solution
A stock solution of 106 mg/l was prepared by loading 0.0106 gram of Arquad 2C-75, weighed out on
an analytical balance and then filled up to the appropriate volume (100 ml) with de-ionized water.
The solution was then stirred until a homogenous solution was formed. The pH was measured to 5.4
and adjusted with sodium hydroxide (1 M) and hydrochloric acid (1 M) to 8.3.
Table 31. Oxygen (mg/L), pH and temperature (°C) measurements of Arquad 2C-75 at the start of the test (T=0h) and at the end of the test (T=48h).
O2 (mg/L) pH Temp. (°C)
T=0h T=48h T=0h T=48h T=0h T=48h
DSW Control 8.7 8.6 8.0 7.9 21.9 21.3 0.02 mg/L 8.8 8.8 8.0 7.9 - -
0.21 mg/L 8.8 8.8 8.0 8.0 - -
HD Control 8.7 8.7 8.2 7.9 21.8 21.1 0.1 mg/L 8.8 8.7 8.2 7.9 - -
1.6 mg/L 8.8 8.8 8.1 7.8 - -
DSW + HA
Control
8.7 8.7 8.2 8.0 21.8 21.3
0.08 mg/L 8.7 8.7 8.3 8.0 - -
1.3 mg/L 8.7 8.6 8.3 8.0 - -
HD 600
Control
8.8 8.8 8.3 8.1 21.7 20.9
0.1 mg/L 8.7 8.8 8.4 8.1 - -
1.6 mg/L 8.7 8.9 8.4 8.1 - -
Table 32. Toxicity of Arquad 2C-75 to Daphnia magna in DSW.
Replicate
Nominal test
conc. (mg/L) Measured test
conc. (mg/L) T=0h T=24h T=48h
1 0,000 0,000 5 5 5
2 0,000 0,000 5 5 4
3 0,000 0,000 5 5 5
4 0,000 0,000 5 5 5
1 0,020 0,003 5 5 5
2 0,020 0,003 5 5 5
3 0,020 0,003 5 5 5
80
4 0,020 0,003 5 5 5
1 0,036 0,006 5 5 5
2 0,036 0,006 5 5 3
3 0,036 0,006 5 5 5
4 0,036 0,006 5 5 5
1 0,065 0,018 5 1 0
2 0,065 0,018 5 5 2
3 0,065 0,018 5 3 0
4 0,065 0,018 5 2 0
1 0,120 0,040 5 0 0
2 0,120 0,040 5 0 0
3 0,120 0,040 5 0 0
4 0,120 0,040 5 0 0
1 0,210 0,111 5 0 0
2 0,210 0,111 5 0 0
3 0,210 0,111 5 0 0
4 0,210 0,111 5 0 0
Figure 50. Dose-response relationship (left: nominal, right: measured) of Arquad 2C-75 to Daphnia magna in DSW.
Table 33. Toxicity of Arquad 2C-75 to Daphnia magna in HD water.
Replicate Nominal test
conc. (mg/L) Measured test
conc. (mg/L) T=0h T=24h T=48h
1 0,000 0,000 5 5 5
2 0,000 0,000 5 5 5
3 0,000 0,000 5 5 5
4 0,000 0,000 5 5 5
1 0,100 0,004 5 5 3
2 0,100 0,004 5 4 4
3 0,100 0,004 5 5 5
4 0,100 0,004 5 5 5
1 0,200 0,018 5 4 0
2 0,200 0,018 5 5 1
3 0,200 0,018 5 5 0
4 0,200 0,018 5 5 0
1 0,400 0,047 5 0 0
2 0,400 0,047 5 0 0
3 0,400 0,047 5 0 0
4 0,400 0,047 5 0 0
1 0,800 0,101 5 0 0
2 0,800 0,101 5 0 0
81
3 0,800 0,101 5 0 0
4 0,800 0,101 5 0 0
1 1,600 0,161 5 0 0
2 1,600 0,161 5 0 0
3 1,600 0,161 5 0 0
4 1,600 0,161 5 0 0
Figure 51. Dose-response relationship (left: nominal, right: measured) of Arquad 2C-75 to Daphnia magna in HD water.
Table 34. Toxicity of Arquad 2C-75 to Daphnia magna in HD water with a conductivity of 600 µS/cm.
Replicate Nominal test
conc. (mg/L) Measured test
conc. (mg/L) T=0h T=24h T=48h
1 0,000 0,000 5 5 5
2 0,000 0,000 5 5 5
3 0,000 0,000 5 5 5
4 0,000 0,000 5 5 5
1 0,100 0,015 5 5 5
2 0,100 0,015 5 5 5
3 0,100 0,015 5 5 5
4 0,100 0,015 5 5 5
1 0,200 0,048 5 0 0
2 0,200 0,048 5 0 0
3 0,200 0,048 5 0 0
4 0,200 0,048 5 0 0
1 0,400 0,097 5 0 0
2 0,400 0,097 5 0 0
3 0,400 0,097 5 0 0
4 0,400 0,097 5 0 0
1 0,800 0,179 5 0 0
2 0,800 0,179 5 0 0
3 0,800 0,179 5 0 0
4 0,800 0,179 5 0 0
1 1,600 0,234 5 0 0
2 1,600 0,234 5 0 0
3 1,600 0,234 5 0 0
4 1,600 0,234 5 0 0
82
Figure 52. Dose-response relationship (left: nominal, right: measured) of Arquad 2C-75 to Daphnia magna in HD water with a conductivity of 600 µS/cm.
Table 35. Toxicity of Arquad 2C-75 to Daphnia magna in DSW + 20 mg/l HA.
Replicate Nominal test
conc. (mg/L) Measured test
conc. (mg/L) T=0h T=24h T=48h
1 0,000 0,000 5 5 4
2 0,000 0,000 5 5 5
3 0,000 0,000 5 5 5
4 0,000 0,000 5 5 5
1 0,080 -0,001 5 5 5
2 0,080 -0,001 5 5 5
3 0,080 -0,001 5 5 4
4 0,080 -0,001 5 5 5
1 0,160 0,001 5 5 5
2 0,160 0,001 5 5 5
3 0,160 0,001 5 5 5
4 0,160 0,001 5 5 5
1 0,325 0,000 5 4 3
2 0,325 0,000 5 5 4
3 0,325 0,000 5 5 5
4 0,325 0,000 5 5 4
1 0,650 0,003 5 5 5
2 0,650 0,003 5 5 4
3 0,650 0,003 5 5 5
4 0,650 0,003 5 5 5
1 1,300 0,011 5 3 0
2 1,300 0,011 5 4 0
3 1,300 0,011 5 3 0
4 1,300 0,011 5 4 0
83
Figure 53. Dose-response relationship (left: nominal, right: measured) of Arquad 2C-75 to Daphnia magna in DSW + 20 mg/L HA.
84
APPENDIX E: Preparation of calibration curves for SPME Preparation of calibration curve and SPME calibration curve for dodecylamine. Preparations for the
other surfactants are not presented here but the preparations are the same, the only thing that
differs between the substances is the concentration range in the calibration curve and the SPME
calibration curve. Calibration curves for each tested surfactant is presented below.
Preparation calibration curve for dodecylamine
A stock solution of 238 mg/L was prepared by loading 0.0119 gram of dodecylamine into a 50 ml
volumetric flask and brought up to volume using leaching solution. The solution was then
ultrasonicated until a homogenous solution was formed. From the stock solution [238 mg/L] a
dilution was made to the concentration of 1 mg/L by pipetting 0.210 ml into a new volumetric flask
and brought up to volume using leaching solution.
Calibration curve (µµµµg/L) Stock 1 mg/L
Volumetric flask (ml),
filled to the mark with leaching
solution
0 0 25 2.5 62.5 µl 25
5.0 125 µl 25
10 0.25 ml 25 25 0.625 ml 25 75 1.875 ml 25
150 3.75 ml 25 300 7.5 ml 25
Preparation SPME calibration curve for dodecylamine
A stock solution of 144 mg/L was prepared by loading 0.0072 gram of dodecylamine into a 50 ml
volumetric flask and brought up to volume with methanol. From the stock solution [144 mg/L] a
dilution was made to 10 mg/L by pipetting 3.472 ml into a new 50 ml volumetric flask and filled to
the mark with methanol. One SPME fiber (7 µm) was added to each beaker and left to equilibrate for
24 hours.
Concentration (µµµµg/L) Stock 10 mg/L Stock 144 mg/L Test vessel with DSW
(ml)
0 0 0 50 10 50 µl - 49.95
25 125 µl - 49.875
75 0.375 ml - 49.625 150 - 52.1 µl 49.948
300 - 104.2 µl 49.896
500 - 173.6 µl 49.826
85
Calibration curves
Fiber concentration plotted against the aqueous concentration of the tested surfactant. Fiber
concentration is unitless because it hasn’t been corrected for the fiber length and thickness of the
coating. However, the unit measured with LC/MS is µg/L.
Dodecylamine
Figure 54. Calibration curve for dodecylamine in DSW.
Dodecylamine +2EO
Figure 55. Calibration curve for dodecylamine +2EO in DSW.
Hexadecylamine +2EO
Figure 56. Calibration curve for hexadecylamine +2EO in DSW (normal DSW).
86
Figure 57. Calibration curve for hexadecylamine +2EO in modified DSW (less amount of salts, only 45 %).
Octadecylamine +2EO
Figure 58. Calibration curve for octadecylamine +2EO in DSW.
Ethomeen C/12
Figure 59. Calibration curves for dodecylamine +2EO as a single substance in the mixture Ethomeen C/12 in DSW and modified DSW (less amount of salts, only 45 %).
87
Figure 60. Calibration curves for tetradecylamine +2EO as a single substance in the mixture Ethomeen C/12 in DSW and modified DSW (less amount of salts, only 45 %).
Figure 61. Calibration curves for hexadecylamine +2EO as a single substance in the mixture Ethomeen C/12 in DSW and modified DSW (less amount of salts, only 45 %).
Didodecyldimethylammonium bromide
Figure 62. Calibration curves for didodecyldimethylammonium bromide in DSW and modified DSW (less amount of salts, only 45 %).
88
Arquad 2C-75
Figure 63. Calibration curves for the single substances in the mixture Arquad 2C-75 in DSW.
89
APPENDIX F: Relationship between nominal and measured
concentration Dodecylamine +2EO
Figure 64. Relationship between nominal and measured concentration (log mg/L) for dodecylamine +2EO in four different test media.
Hexadecylamine +2EO
90
Figure 65. Relationship between nominal and measured concentration (log mg/L) for hexadecylamine +2EO in three different test media.
Ethomeen C/12
91
Figure 66. Relationship between nominal and measured concentration (log mg/L) for Ethomeen C/12 in three different test media. The relationship is given for the mixture as well as for the individual substances measured with the LC/MS.
Didodecyldimethylammonium bromide
92
Figure 67. Relationship between nominal and measured concentration (log mg/L) for didodecyldimethylammonium bromide in three different test media. Based on the nominal and measured concentrations in mg/L (not logarithmic), the R2 value in DSW+HA is 0.9994.
Arquad 2C-75
Figure 68. Relationship between nominal and measured concentration (log mg/L) for Arquad 2C-75 in four different test media. Based on the nominal and measured concentrations in mg/L (not logarithmic), the R2 value in DSW+HA is 0.9726.