**3. Toxicity and environmental effects of triazines**

Triazine herbicides are generally of low acute toxicity for birds and mammals, although certain species show unexpected vulnerability for some of them, e.g. for sheep the fatal dose of simazine has been reported as 500-1400 mg/kg, while LD50 for rats is >5000 mg/kg (Stevens & Sumner, 1991). Acute toxicity data for some compounds are shown in Table 2.

Fate and Determination of Triazine Herbicides in Soil 45

However, the situation is less plausible when assessing the chronic toxicity of triazines. Significant scientific and public controversy has been increasing in the last decade especially regarding the effects of environmentally relevant concentrations of atrazine and its main transformation products desethylatrazine, desisopropylatrazine and hydroxyatrazine, resulting in the 2003 ban of atrazine products in European Union (Sass & Colangelo, 2006). Initial studies reported some carcinogenic, mutagenic and teratogenic effects of triazines only at the dose exceeding the maximal tolerable dose (Stevens & Sumner, 1991). However, environmentally relevant low concentrations of atrazine were later shown to adversely affect the normal male development in amphibians (Tavera-Mendoza et al., 2002), although the evidence is still not conclusive (Solomon et al., 2008). Adverse effects of atrazine were shown also for rats, both on male reproductive tract (Kniewald et al., 2000) and on oestrus in females (Eldridge et al., 1999). The latter is presumably due to the effect on hypothalamicpituitary-gonadal axis and not on intrinsic estrogenic effect of atrazine (Eldridge et al., 1999; Taketa et al., 2011). Similar effects have been observed for the main atrazine transformation products (Stanko et al., 2010). Besides these endocrine-disrupting properties, atrazine has been shown to affect immune function in mice and the effects persist some time after the exposure (Filipov et al., 2005). Other triazine herbicides are not that extensively covered regarding their chronic toxicity, presumably because they are less widely applied. However, USA Environmental Protection Agency (EPA) concludes that triazines and TPs with chlorine attached to the ring (see Fig. 1) have the same common mechanism of toxicity regarding their endocrine-related developmental, reproductive and carcinogenic effects

Atrazine 1900-3000 750 (rabbits) Simazine >5000 500-1400 (sheep) Cyanazine 180-380 NA Terbutylazine 1000-1590 NA Atraton 1465-2400 NA Terbumeton >650 NA Ametryn 110-1750 NA Prometryn 3150-5235 NA Terbutryn 2000-2980 3880 (mice) Table 2. Acute toxicity data for some triazines (IUPAC Agrochemical Information, 2011;

Stevens & Sumner, 1991). NA - not available.

(Environmental Protection Agency [EPA], 2011).

materials and various pollutants into the environment etc.

**4. Distribution of triazines in the environmental compartments** 

After the introduction in the environment, triazines are distributed between the three main environmental compartments, namely gaseous (air), aqueous (ground and surface waters) and solid (soil, sediments). The fourth important compartment interacting with the environment is biota: uptake of triazines into microorganisms and plants, which will be considered separately. Distribution is governed by the physicochemical properties of the compounds (Table 1) and is an ongoing process. There is a dynamic interchange of temporary equilibrium states and re-distribution, influenced by weather conditions, input of

Name oral LD50 /mg/kg (rats) oral LD50 /mg/kg (other species)


Fig. 1. Structures of some widely used triazines and more important transformation products.


Table 1. Some relevant physico-chemical parameters for the environmentally important triazines and their transformation products (Kaune et al., 1998; Noble, 1993; Shiu et al., 1990; Tomlin, 1994). NA - not available.

Chlorotriazines

Methoxytriazines

Methylthiotriazines

Degradation products

Fig. 1. Structures of some widely used triazines and more important transformation

Desisopropylatrazine 173.6 670 1.1-1.2 1.58 NA

Table 1. Some relevant physico-chemical parameters for the environmentally important triazines and their transformation products (Kaune et al., 1998; Noble, 1993; Shiu et al., 1990;

Hydroxyatrazine 197.3 5.9 1.4 5.2 1.110-3 (25 oC)

Name *M* / g/mol Water sol. / mg/L log*K*ow p*K*<sup>a</sup> *p* / Pa Atrazine 215.7 33 (20 oC) 2.2-2.7 1.7 4.010-5 (20 oC) Simazine 201.7 5 (20-22 oC) 2.2-2.3 1.65 8.110-7 (20 oC) Cyanazine 240.7 171 (25 oC) 1.8-2.0 1.85 2.110-7 (25 oC) Terbutylazine 229.8 8.5 (20 oC) 2.6-3.0 2.0 1.510-4 (25 oC) Atraton 211.3 1800 (20-22 oC) 2.3-2.7 4.2 NA Terbumeton 225.3 130 (20 oC) 2.7-3.1 4.7 2.510-5 (25 oC) Ametryn 227.1 185 (20 oC) 2.7-3.1 4.0 1.110-4 (20 oC) Prometryn 241.4 33-48 (20 oC) 3.3 4.1 1.3�10-4 (20 oC) Terbutryn 241.4 25 (20 oC) 3.1-3.7 4.3 2.210-4 (25 oC) Desethylatrazine 187.7 3200 1.5 1.65 1.210-2 (25 oC)

R1 R2 R3

atrazine Cl CH2CH3 CH(CH3)2 simazine Cl CH2CH3 CH2CH3 propazine Cl CH(CH3)2 CH(CH3)2

cyanazine Cl CH2CH3 C(CH3)2CN terbutylazine Cl CH2CH3 C(CH3)3

atratone OCH3 CH2CH3 CH(CH3)2 prometon OCH3 CH(CH3)2 CH(CH3)2 terbumeton OCH3 CH2CH3 C(CH3)3

ametryn SCH3 CH2CH3 CH(CH3)2 simetryn SCH3 CH2CH3 CH2CH3 prometryn SCH3 CH(CH3)2 CH(CH3)2 terbutryn SCH3 CH2CH3 C(CH3)3

desethylatrazine Cl H CH(CH3)2 desisopropylatrazine Cl CH2CH3 H desethyldesisopropylatrazine Cl H H hydroxyatrazine OH CH2CH3 CH(CH3)2

N

O

(CH <sup>N</sup> 3)2N

products.

(CH3)3C

N

N H N H

N

metribuzin (triazinone)

CH3

O

N N

O

Tomlin, 1994). NA - not available.

hexazinone (triazidinone)

NH2

N N

SCH3

R3 R2 General s-triazine structure (see table on the right)

R1

N


Table 2. Acute toxicity data for some triazines (IUPAC Agrochemical Information, 2011; Stevens & Sumner, 1991). NA - not available.

However, the situation is less plausible when assessing the chronic toxicity of triazines. Significant scientific and public controversy has been increasing in the last decade especially regarding the effects of environmentally relevant concentrations of atrazine and its main transformation products desethylatrazine, desisopropylatrazine and hydroxyatrazine, resulting in the 2003 ban of atrazine products in European Union (Sass & Colangelo, 2006). Initial studies reported some carcinogenic, mutagenic and teratogenic effects of triazines only at the dose exceeding the maximal tolerable dose (Stevens & Sumner, 1991). However, environmentally relevant low concentrations of atrazine were later shown to adversely affect the normal male development in amphibians (Tavera-Mendoza et al., 2002), although the evidence is still not conclusive (Solomon et al., 2008). Adverse effects of atrazine were shown also for rats, both on male reproductive tract (Kniewald et al., 2000) and on oestrus in females (Eldridge et al., 1999). The latter is presumably due to the effect on hypothalamicpituitary-gonadal axis and not on intrinsic estrogenic effect of atrazine (Eldridge et al., 1999; Taketa et al., 2011). Similar effects have been observed for the main atrazine transformation products (Stanko et al., 2010). Besides these endocrine-disrupting properties, atrazine has been shown to affect immune function in mice and the effects persist some time after the exposure (Filipov et al., 2005). Other triazine herbicides are not that extensively covered regarding their chronic toxicity, presumably because they are less widely applied. However, USA Environmental Protection Agency (EPA) concludes that triazines and TPs with chlorine attached to the ring (see Fig. 1) have the same common mechanism of toxicity regarding their endocrine-related developmental, reproductive and carcinogenic effects (Environmental Protection Agency [EPA], 2011).

#### **4. Distribution of triazines in the environmental compartments**

After the introduction in the environment, triazines are distributed between the three main environmental compartments, namely gaseous (air), aqueous (ground and surface waters) and solid (soil, sediments). The fourth important compartment interacting with the environment is biota: uptake of triazines into microorganisms and plants, which will be considered separately. Distribution is governed by the physicochemical properties of the compounds (Table 1) and is an ongoing process. There is a dynamic interchange of temporary equilibrium states and re-distribution, influenced by weather conditions, input of materials and various pollutants into the environment etc.

Fate and Determination of Triazine Herbicides in Soil 47

Leaching of atrazine into lower layers of the soil and eventually groundwater is generally affected by the same parameters as sorption. The mobility of compound in soil/sediment is expressed by retardation factor *R*f as determined by column lysimeters (Weber et al., 2007). For atrazine, *R*f has been shown to be inversely proportional to SOM content and related to pH and soil leaching potential (Weber et al., 2007). Presence of more polar SOM with higher ratio of polar functional groups, e.g. from the manure, has been postulated to result in stronger hydrogen bonding of atrazine and reduced desorption and mobility (Lima et al., 2010), although completely opposite results, i.e. stronger bonding to more hydrophobic humic matter, were reported elsewhere (Celano et al., 2008). Desorption and leaching is enhanced by the presence of surfactants, especially anionic (J.F. Lee et al., 2004; Ying et al., 2005), as well as dissolved organic matter (DOM) (Ling et al., 2006). However, great caution is needed when extrapolating results from these studies to predict the dissipation behaviour

of atrazine, as gross underestimations have been observed (Jablonowski et al., 2011).

extensively sorbed on mineral components of the soil/sediment (Stipičević et al., 2009).

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 **pKa**

ametryne

terbutryne

terbumeton

atratone

Fig. 2. Relation between p*K*a and % of sorbed compounds after 5-7 days of batch equilibrium

experiment on Florisil (SiO2, MgO). Adapted after Prosen et al. (2007).

atrazine

terbutylazine

60

70

80

90

**% of sorbed compound (5-7 days)**

100

110

Considerably less information about sorption and mobility in soil and sediments is available for other triazines or transformation products. Chlorotriazines are generally assumed to behave similarly to atrazine and this has been confirmed in some experiments for simazine (Mudhoo & Garg, 2011; Ying et al., 2005) or terbutylazine for humic organic matter (Celano et al., 2008). The latter is a less polar compared to atrazine and has been shown to exhibit greater extent of sorption on HS (Erny et al., 2011; Prosen & Zupančič-Kralj, 2000). In comparison of methylthio-, methoxy- and chlorotriazine sorption on sediments and mineral soil components, sorption intensity was related to the basicity (p*K*a) and water solubility of compounds, but not their log*K*ow (Prosen et al., 2007; Stipičević et al., 2009) - Fig. 2. Dealkylated triazine transformation products are weakly sorbed on humic substances compared to parent compounds (Erny et al., 2011), while hydroxyatrazine, a dechlorinated atrazine TP, is

Volatilization of triazines and their long-range atmospheric transport is a poorly researched process. It is supposed that, similar to other semivolatiles, triazines are transported by air masses absorbed on the particulate matter and deposit in cold atmospheric conditions (high mountains, higher geographical latitudes) mainly by wet deposition. Snow is an effective scavenger of particulate matter and associated pollutants from the atmosphere. Triazines have been detected both in snow and rainwater (Polkowska et al., 2000; Usenko et al., 2005).

Triazines are distributed mainly between aqueous and solids compartments. The main processes are partitioning and sorption on solid components, as well as solubilisation in the aqueous compartment followed by leaching into lower solid layers and eventually into groundwater. Living organisms present in both compartments contribute to the transport by uptaking the compounds and returning them mainly as transformation products. The majority of research has been done on atrazine in the last decade of 20th century and is encompassed in a recent review paper (Mudhoo & Garg, 2011). However, atrazine residues in soil have proven to be more persistent than previously expected (Jablonowski et al., 2011) and thus there is an ongoing need for further research on the soil behaviour of this compound (Barton & Karathanasis, 2003; Jablonowski et al., 2011; Kovaios et al., 2006; Ling et al., 2006). Atrazine is expected to be in its non-ionized form at the environmentally relevant pH values (see Table 1) and for uncharged compounds, it is generally accepted that they are sorbed on organic carbon fraction of the soils/sediments (Mudhoo & Garg, 2011). The main mechanism in operation is partitioning between aqueous and organic carbon phase, predominantly humic substances. Both overall partition coefficient *K*d and partition coefficient for organic carbon *K*oc are used to quantitatively express the extent of interaction. The reported values for the latter differ considerably from 25 to 600 L/kg OC (Mudhoo & Garg, 2011), which may reflect the differences in organic matter structure. Humic substances (HS) are heterogeneous and still poorly characterized macromolecules or supramolecular associations (Schaumann, 2006). A number of mechanisms have been proposed for the interaction of atrazine and HS: partitioning resulting from hydrophobic interactions (Lima et al., 2010; Prosen & Zupančič-Kralj, 2000), hydrogen bonding (Prosen & Zupančič-Kralj, 2000), electron transfer, charge transfer (Mudhoo & Garg, 2011). While atrazine is sorbed primarily onto soil organic matter (SOM), presence or addition of dissolved organic matter (DOM) may enhance the sorption at lower DOM concentration, but decrease it at higher DOM concentration (Ling et al., 2006; Mudhoo & Garg, 2011), which is a consequence of increased solubilisation of atrazine in the aqueous fraction with DOM.

Atrazine is sorbed on some mineral components of soils/sediments as well: aluminiumsaturated smectite (Mudhoo & Garg, 2011), silicagel (Kovaios et al., 2006) and Florisil (SiO2+MgO) (Prosen et al., 2007), but not calcite or alumina (Kovaios et al., 2006; Prosen et al., 2007). The proposed mechanism is electrostatic or electron-transfer interaction of atrazine with silanol groups (Kovaios et al., 2006; Prosen et al., 2007). Besides soil organic matter (SOM) content and presence of adsorbing minerals, other parameters govern the extent of atrazine sorption to environmental solids: pH, ionic strength, surface area, particle and pore size, presence of other compounds, especially surfactants (J.F. Lee et al., 2004), temperature (Mudhoo & Garg, 2011). Contact time is another important factor. Desorption hysteresis has been observed for longer contact times (Drori et al., 2005; Prosen & Zupančič-Kralj, 2000). The currently accepted model explaining the effect of contact time, nonlinear sorption kinetics, desorption hysteresis and conditioning effect of sorbate on sorbent affinity is the dual-mode sorption process of sorbate in the interchangeable rubbery and glassy state of polymerous SOM material (Schaumann, 2006).

Volatilization of triazines and their long-range atmospheric transport is a poorly researched process. It is supposed that, similar to other semivolatiles, triazines are transported by air masses absorbed on the particulate matter and deposit in cold atmospheric conditions (high mountains, higher geographical latitudes) mainly by wet deposition. Snow is an effective scavenger of particulate matter and associated pollutants from the atmosphere. Triazines have been detected both in snow and rainwater (Polkowska et al., 2000; Usenko et al., 2005). Triazines are distributed mainly between aqueous and solids compartments. The main processes are partitioning and sorption on solid components, as well as solubilisation in the aqueous compartment followed by leaching into lower solid layers and eventually into groundwater. Living organisms present in both compartments contribute to the transport by uptaking the compounds and returning them mainly as transformation products. The majority of research has been done on atrazine in the last decade of 20th century and is encompassed in a recent review paper (Mudhoo & Garg, 2011). However, atrazine residues in soil have proven to be more persistent than previously expected (Jablonowski et al., 2011) and thus there is an ongoing need for further research on the soil behaviour of this compound (Barton & Karathanasis, 2003; Jablonowski et al., 2011; Kovaios et al., 2006; Ling et al., 2006). Atrazine is expected to be in its non-ionized form at the environmentally relevant pH values (see Table 1) and for uncharged compounds, it is generally accepted that they are sorbed on organic carbon fraction of the soils/sediments (Mudhoo & Garg, 2011). The main mechanism in operation is partitioning between aqueous and organic carbon phase, predominantly humic substances. Both overall partition coefficient *K*d and partition coefficient for organic carbon *K*oc are used to quantitatively express the extent of interaction. The reported values for the latter differ considerably from 25 to 600 L/kg OC (Mudhoo & Garg, 2011), which may reflect the differences in organic matter structure. Humic substances (HS) are heterogeneous and still poorly characterized macromolecules or supramolecular associations (Schaumann, 2006). A number of mechanisms have been proposed for the interaction of atrazine and HS: partitioning resulting from hydrophobic interactions (Lima et al., 2010; Prosen & Zupančič-Kralj, 2000), hydrogen bonding (Prosen & Zupančič-Kralj, 2000), electron transfer, charge transfer (Mudhoo & Garg, 2011). While atrazine is sorbed primarily onto soil organic matter (SOM), presence or addition of dissolved organic matter (DOM) may enhance the sorption at lower DOM concentration, but decrease it at higher DOM concentration (Ling et al., 2006; Mudhoo & Garg, 2011), which is a consequence of

increased solubilisation of atrazine in the aqueous fraction with DOM.

of polymerous SOM material (Schaumann, 2006).

Atrazine is sorbed on some mineral components of soils/sediments as well: aluminiumsaturated smectite (Mudhoo & Garg, 2011), silicagel (Kovaios et al., 2006) and Florisil (SiO2+MgO) (Prosen et al., 2007), but not calcite or alumina (Kovaios et al., 2006; Prosen et al., 2007). The proposed mechanism is electrostatic or electron-transfer interaction of atrazine with silanol groups (Kovaios et al., 2006; Prosen et al., 2007). Besides soil organic matter (SOM) content and presence of adsorbing minerals, other parameters govern the extent of atrazine sorption to environmental solids: pH, ionic strength, surface area, particle and pore size, presence of other compounds, especially surfactants (J.F. Lee et al., 2004), temperature (Mudhoo & Garg, 2011). Contact time is another important factor. Desorption hysteresis has been observed for longer contact times (Drori et al., 2005; Prosen & Zupančič-Kralj, 2000). The currently accepted model explaining the effect of contact time, nonlinear sorption kinetics, desorption hysteresis and conditioning effect of sorbate on sorbent affinity is the dual-mode sorption process of sorbate in the interchangeable rubbery and glassy state Leaching of atrazine into lower layers of the soil and eventually groundwater is generally affected by the same parameters as sorption. The mobility of compound in soil/sediment is expressed by retardation factor *R*f as determined by column lysimeters (Weber et al., 2007). For atrazine, *R*f has been shown to be inversely proportional to SOM content and related to pH and soil leaching potential (Weber et al., 2007). Presence of more polar SOM with higher ratio of polar functional groups, e.g. from the manure, has been postulated to result in stronger hydrogen bonding of atrazine and reduced desorption and mobility (Lima et al., 2010), although completely opposite results, i.e. stronger bonding to more hydrophobic humic matter, were reported elsewhere (Celano et al., 2008). Desorption and leaching is enhanced by the presence of surfactants, especially anionic (J.F. Lee et al., 2004; Ying et al., 2005), as well as dissolved organic matter (DOM) (Ling et al., 2006). However, great caution is needed when extrapolating results from these studies to predict the dissipation behaviour of atrazine, as gross underestimations have been observed (Jablonowski et al., 2011).

Considerably less information about sorption and mobility in soil and sediments is available for other triazines or transformation products. Chlorotriazines are generally assumed to behave similarly to atrazine and this has been confirmed in some experiments for simazine (Mudhoo & Garg, 2011; Ying et al., 2005) or terbutylazine for humic organic matter (Celano et al., 2008). The latter is a less polar compared to atrazine and has been shown to exhibit greater extent of sorption on HS (Erny et al., 2011; Prosen & Zupančič-Kralj, 2000). In comparison of methylthio-, methoxy- and chlorotriazine sorption on sediments and mineral soil components, sorption intensity was related to the basicity (p*K*a) and water solubility of compounds, but not their log*K*ow (Prosen et al., 2007; Stipičević et al., 2009) - Fig. 2. Dealkylated triazine transformation products are weakly sorbed on humic substances compared to parent compounds (Erny et al., 2011), while hydroxyatrazine, a dechlorinated atrazine TP, is extensively sorbed on mineral components of the soil/sediment (Stipičević et al., 2009).

Fig. 2. Relation between p*K*a and % of sorbed compounds after 5-7 days of batch equilibrium experiment on Florisil (SiO2, MgO). Adapted after Prosen et al. (2007).

Fate and Determination of Triazine Herbicides in Soil 49

detection (selectivity, LODs)

Fig. 3. Schematic representation of the problems and solutions for triazine determination in

Technique Principle Advantages Disadvantages

selectivity identification

separation from the bulk matrix

> - recoveries not dependent on sample type - cheap

> - recoveries not dependent on sample type - cheap


environmentally acceptable



toxic,

(ta n d e m )

m a s s sp e ctro m e try

e x tra c tio n w ith p re c o n c e n tra tio n

separation m ethod w ith selected detection

> - time-consuming - high consumption of organic solvents - extracts have to be concentrated







amount

Determination of triazines in soil and sediment samples

solid environmental samples.

Ultrasonication Extraction (USE)

Supercritical Fluid Extraction (SFE)

Microwave-Assisted Solvent Extr. (MASE)

Pressurised Liquid Extraction (PLE) / Accelerated Solvent

Extr. (ASE)

Soxhlet Extraction (SE) continuous percolation

low concentration

many interfering compounds binding to organic and mineral components

of organic solvent

mixing, desorption of analytes from sample

supercritical fluid of low viscosity better penetrates the sample

microwave- assisted desorption of analytes

enhanced extraction efficiency of analytes due to solvents at high temperature and pressure (liquids above

(Andreu & Pico, 2004; Camel, 2000; Lesueur et al., 2008; Lopez-Avila, 1999).

Table 3. Common extraction techniques for triazines from the solid environmental samples

boiling point)

and sample components

components

Knowledge of sorption/desorption behaviour of triazines is frequently applied in bioremediation either to enhance their leaching or to stabilise the residues in the contaminated sites (Delgado-Moreno et al., 2010; Jones at al., 2011; J.F. Lee et al., 2004; Lima et al., 2010; Mudhoo & Garg, 2011; Ying et al., 2005).
