N-(3-aminopropyl)-N-dodecylpropane-1,3-diamine

Adsorption / desorption

Link to relevant study record(s)

Reference

Reference 1

Endpoint:

adsorption / desorption: screening

Type of information:

experimental study

Adequacy of study:

key study

Study period:

From May 08, 2003 to May 28, 2003

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

OECD Guideline 106 (Adsorption - Desorption Using a Batch Equilibrium Method)

GLP compliance:

yes (incl. QA statement)

Type of method:

batch equilibrium method

Media:

soil

Test temperature:

20±2°C

Details on study design: HPLC method:

Analyses were carried out via HPLC on cation-exchange column andfluorescence detection after post-column derivatisation with ortho-Phthaldialdehyde (OPA). The derivatisation reagent was added to the column effluent via T-connector where the solutions were mixed. The reaction took place in a reaction capillary of a length of about 1 m and afterwards the sample passed the fluorescence detector. Quantification was performed using external standards.
3 HPLC-pumps: Waters 510
HPLC-autosampler: Waters 712 WISP
Detection: Waters 470 fluorescence detector.

Analytical monitoring:

yes

Details on sampling:

Test concentrations: Based on the results of a preliminary test, the following five test concentrations were used for definitive testing:
- 100, 500, 1,000, 5,000 and 10,000 mg/L for soils no. 1, 2 and 4 and
- 50, 100, 500, 1,000 and 5,000 mg/L for soils no. 3 and 5.

Six stock solutions with the following concentrations were prepared in 0.01 M CaCl2:
I 0.5 g/L
II 1 g/L
III 5 g/L
IV 10 g/L
V 50 g/L
VI 100 g/L
Stock solutions 0.5 - 50 g/L were diluted from 100 g/L with 0.01 M CaCl2. From every stock solution 10 mL were spiked into the test vessel to obtain the final concentration in 100 mL 0.01 M CaCl2.

Details on matrix:

TEST SYSTEM:
Equilibrium adsorption coefficients on various soils were determined as a function of soil characteristics (e.g. organic carbon content, clay content, soil texture and pH-value). 5 soil types were used in order to cover as widely as possible the interactions with natural soils. Standard EURO soils no. 1, 2, 3, 4 and 5. These soils have varying adsorption capacities in relation to their content of organic matter, clay and metal oxides as well as pH and cation exchange capacity. All soils were air dried and sieved to a maximum particle size of 2 mm.

See below. Table 1 for details on test system or soils used.

Details on test conditions:

TEST SYSTEM AND CONDITIONS:
- Test vessel: 180 mL brown glass bottles with screw cap (covered with aluminium foil)
- Equilibration time/ Test duration: 24 h was the time set to reach adsorption equilibrium for all soils and the time to reach desorption equilibrium was set to 20 h for soils no. 1, 2, 3, 5 and 24 h for soil no. 4 (based on a preliminary test)
- Dispersion treatment : Agitation
- Replicates: Duplicates per soil
- Vehicle: None
- Test volume: 100 mL 0.01 M CaCl2
- Amount of soil: 1.0 g
- Soil/solution ratio: 1:100 (based on a preliminary test)
- Application: 10 mL of the stock solution were added to the test vessel using a pipette.
- pH stability of the test substance: pH 3.2 and 7.2 (with 102 and 107 % recovery after 48 h)

CONTROLS AND REFERENCE SUBSTANCE:
- Control: CaCl2 solution with lowest test substance concentration (without soil)
- Replicate: Single
- Reference substance: Not recommended according to the OECD Guideline 106

DEFINITIVE TEST METHODS:
- Adsorption step: Samples were shaken for 24 h to reach adsorption equilibrium after which the samples were centrifuged at 3,000 rpm for 3 minutes. The aqueous phase was then decanted.
- Desorption step: After the decanting of the aqueous phase from the adsorption step, it was replaced with an identical volume of fresh 0.01 M CaCl2
solution and samples were shaken for 20 h (soils no. 1, 2, 3 and 5) and 24 h (soil no. 4) to reach desorption equilibrium.
- Mass balance: After decanting of the aqueous phase from the desorption step, it was replaced with 20 mL 2-propanole and samples were shaken for 30 min to extract the test substance from the soil phase. Test temperature 20 ± 2 °C. As mass balance was lower than 90 % the test substance is considered to
be unstable during the time scale of the test. The definitive test was performed by analysing both the aqueous phase and the soil phase.

Sample No.:

#1

Type:

Kd

Remarks:

Not corrected

Value:

2 781 other: cm3/g

Temp.:

20 °C

Matrix:

Soil 1 (EURO soil 1: Clay)

% Org. carbon:

3.29

Remarks on result:

other: OC normalisation should not be applied to substances which sorb mainly based on ionic interaction

Key result

Sample No.:

#1

Type:

Kd

Remarks:

Danish EPA approach: For extrapolation to lower concentrations the data originating from the lowest aqueous measured concentrations are used letting the intercept equal zero.

Value:

1 957 other: cm3/g

Temp.:

20 °C

Matrix:

Soil 1 (EURO soil 1: Clay)

% Org. carbon:

3.29

Remarks on result:

other: OC normalisation should not be applied to substances which sorb mainly based on ionic interaction

Sample No.:

#2

Type:

Kd

Remarks:

Not corrected

Value:

598 other: cm3/g

Temp.:

20 °C

Matrix:

Soil 2: (Euro soil 2) Silt loam

% Org. carbon:

2.39

Remarks on result:

other: OC normalisation should not be applied to substances which sorb mainly based on ionic interaction

Key result

Sample No.:

#2

Type:

Kd

Remarks:

Danish EPA approach: For extrapolation to lower concentrations the data originating from the lowest aqueous measured concentrations are used letting the intercept equal zero.

Value:

561 other: cm3/g

Temp.:

20 °C

Matrix:

Soil 2: (Euro soil 2) Silt loam

% Org. carbon:

2.39

Remarks on result:

other: OC normalisation should not be applied to substances which sorb mainly based on ionic interaction

Sample No.:

#3

Type:

Kd

Remarks:

Not corrected

Value:

239 other: cm3/g

Temp.:

20 °C

Matrix:

Soil 3 (EURO soil 3) Loam

% Org. carbon:

3.32

Remarks on result:

other: OC normalisation should not be applied to substances which sorb mainly based on ionic interaction

Key result

Sample No.:

#3

Type:

Kd

Remarks:

Danish EPA approach: For extrapolation to lower concentrations the data originating from the lowest aqueous measured concentrations are used letting the intercept equal zero.

Value:

194 other: cm3/g

Temp.:

20 °C

Matrix:

Soil 3 (EURO soil 3) Loam

% Org. carbon:

3.32

Remarks on result:

other: OC normalisation should not be applied to substances which sorb mainly based on ionic interaction

Sample No.:

#4

Type:

Kd

Remarks:

!!!Not to be used!!

Value:

622 other: cm3/g

Temp.:

20 °C

Matrix:

Soil 4 (EURO soil 4) Silt

% Org. carbon:

1.36

Remarks on result:

other: Result of EURO soil 4 should not be used for extrapolation to lower concentrations

Sample No.:

#5

Type:

Kd

Remarks:

Not corrected

Value:

121 other: cm3/g

Temp.:

20 °C

Matrix:

Soil 5 (EURO soil 5) Loamy sand

% Org. carbon:

4.43

Remarks on result:

other: OC normalisation should not be applied to substances which sorb based ionic interaction

Key result

Sample No.:

#5

Type:

Kd

Remarks:

Danish EPA approach: For extrapolation to lower concentrations the data originating from the lowest aqueous measured concentrations are used letting the intercept equal zero.

Value:

145 other: cm3/g

Temp.:

20 °C

Matrix:

Soil 5 (EURO soil 5) Loamy sand

% Org. carbon:

4.43

Remarks on result:

other: OC normalisation should not be applied to substances which sorb mainly based on ionic interaction

Key result

Type:

Kd

Remarks:

Mean value for Soil 1,2,3 and 5

Value:

714 other: cm3/g

Temp.:

20 °C

Matrix:

Mean of Soil 1,2,3 and 5. Soil 4 was excluded because of the poor fit

Remarks on result:

other: OC normalisation should not be applied to substances which sorb mainly based on ionic interaction

Details on results (HPLC method):

The method was validated and tested in regard to the:
- linearity
- precision
- limit of quantification (LOQ) and limit of detection (LOD)

Adsorption and desorption constants:

Determination of percentage adsorption and distribution coefficient (Kd):
For determination of Kd a concentration of 100 mg/L was prepared in duplicates for soil no. 3 and 5 and 500 mg/L for soils no. 1, 2 and 4.
- Distribution coefficient (Kd; cm3/g): Soil 1: 2781; Soil 2: 598; soil 3: 239; soil 4: 622; soil 5: 121
- Koc (cm3/g): Soil 1: 84,529; Soil 2: 25,021; soil 3: 7,199; soil 4: 45,735; soil 5: 2,731

- Results of adsorption isotherms (FREUNDLICH adsorption coefficients):
- KadsF: Soil 1: 11,466; Soil 2: 2,784; Soil 3: 449; Soil 4: 17,254; Soil 5: 88
- log KadsF : Soil 1: 4.05; Soil 2: 3.44; Soil 3: 2.65; Soil 4: 4.23; Soil 5: 1.94
- KOCF: Soil 1: 34,8511; Soil 116,485; Soil 3: 13,524; Soil 4: 1268,676; Soil 5: 1,986

Determination of Kdes values and % desorption: For determination of the desorption coefficient (Kdes) and percentage desorption, soil sample which had achieved equilibrium in the adsorption kinetics step were used:
- Results of desorption isotherm (FREUNDLICH desorption coefficients):
- Desorption coefficient (Kdes; cm3/g): Soil 1: 2,685; Soil 2: 428; soil 3: 259; soil 4: 284; soil 5: 524
- K desF: Soil 1: 8251; Soil 2: 3,209; Soil 3: 656; Soil 4: 3,562; Soil 5: 815.
- log KdesF: Soil 1: 3.91; Soil 2: 3.50; Soil 3: 2.81; Soil 4: 3.55; Soil 5: 2.91.

Recovery of test material:

Results of mass balance:
Soil 1:
at 100 ug/cm3 mean RR(%): 5
at 500 ug/cm3 mean RR(%): 93
at 1,000ug/cm3 mean RR(%): 103
at 5,000 ug/cm3 mean RR(%): 126
at 10,000 ug/cm3 mean RR(%): 106

Soil 2:
at 100 ug/cm3 mean RR(%): 79
at 500 ug/cm3 mean RR(%): 132
at 1,000ug/cm3 mean RR(%): 146
at 5,000ug/cm3 mean RR(%): 103
at 10,000 ug/cm3 mean RR(%): 83

Soil 3:
at 100 ug/cm3 mean RR(%): 85
at 500 ug/cm3 mean RR(%): 94
at 1,000ug/cm3 mean RR(%): 96
at 5,000ug/cm3 mean RR(%): 123
at 10,000 ug/cm3 mean RR(%): 102

Soil 4:
at 100 ug/cm3 mean RR(%): 43
at 500 ug/cm3 mean RR(%): 130
at 1,000ug/cm3 mean RR(%): 121
at 5,000ug/cm3 mean RR(%): 106
at 10,000 ug/cm3 mean RR(%): 95

Soil 5:
at 100 ug/cm3 mean RR(%): 59
at 500 ug/cm3 mean RR(%): 67
at 1,000ug/cm3 mean RR(%): 92
at 5,000ug/cm3 mean RR(%): 56
at 10,000 ug/cm3 mean RR(%): 19

Concentration of test substance at end of adsorption equilibration period:

Results of screening test - adsorption at 20°C:
- Quantity adsorbed [ug] per gram of soil: soil 1: 43611.5; soil 2: 37605.0; soil 3: 6112.6; soil 4: 37648.2; soil 5: 4711.8
- Test material adsorbed [%]: soil 1: 96; soil 2: 85; soil 3: 70; soil 4: 86; soil 5: 54

Concentration of test substance at end of desorption equilibration period:

Results of screening test - desorption:
-Quantity desorbed [ug] : soil 1: 1562.26; soil2: 7072.53; soil 3: 1699.56; soil 4:9720.74; soil 5: 752.46
-[%] of adsorbed test material, which is desorbed: soil 1: 3.9; soil 2: 19.5; soil 3: 28.3; soil 4: 26.6; soil 5: 16.3

Transformation products:

no

Details on results (Batch equilibrium method):

Due to the non linearity of the curves which is caused by typical cationic surfactant behaviour at higher test concentrations, the Kfads values as indicated below were recalculated using the linear part of the curves. This was done to allow extrapolation to lower test concentrations which is needed for risk assessment purposes. The test concentrations excluded from the Kfads calculation are indicated below.

C0 Cadsaq Cadss Kd Clay soil (Kfads = 4083.2 r2 = 0.99 excluding 5000 and 10000)
[µg/cm3] [µg/cm3] [µg/g] [g/cm3] SAND total [%] 3.30
Soil 1 100 2.87 8750.5 3048.97 SILT total [%] 21.90
500 27.24 42384.3 1555.96 CLAY total [%] 75.00
1000 66.10 83535.6 1263.78 pH in CaCl2 5.7
5000 3388.85 93009.6 27.45 total carbon [%] 3.5
10000 6923.82 170507.4 24.63 organic carbon [%] 3.29
C.E.C. [mval/100g] 32.4

Silt loam (Kfads = 36.6 r2 = 0.99 excluding 5000 and 10000)
Soil 2 100 23.26 6415.7 275.83 SAND total [%] 13.40
500 56.65 38247.7 675.16 SILT total [%] 64.10
1000 105.62 77289.0 731.77 CLAY total [%] 22.60
5000 3641.47 64481.1 17.71 pH in CaCl2 7.2
10000 6072.51 254135.5 41.85 total carbon [%] 10.9
organic carbon [%] 2.39
C.E.C. [mval/100g] 28.9

Loam (Kfads = 448.7 r2 = 0.99 excluding 5000)
Soil 3 50 12.39 3097.6 250.01 SAND total [%] 46.40
100 31.44 5513.2 175.36 SILT total [%] 36.80
500 169.49 26306.5 155.21 CLAY total [%] 17.00
1000 450.72 41175.1 91.35 pH in CaCl2 5.9
5000 3774.97 50156.8 13.29 total carbon [%] 3.6
organic carbon [%] 3.32
C.E.C. [mval/100g] 16.6


Silt (Kfads = 25293 r2 = 0.88 excluding 10000)
Soil 4 100 SAND total [%] 4.1
500 60.65 37639.1 620.60 SILT total [%] 75.70
1000 287.26 58204.0 202.62 CLAY total [%] 20.30
5000 3643.85 63904.3 17.54 pH in CaCl2 6.8
10000 7325.01 123970.2 16.92 total carbon [%] 1.7
organic carbon [%] 1.36
C.E.C. [mval/100g] 17.3
Due to the loss of the 100 mg/L test concentration and the low r2 for soil 4 it is not possible to reliably extrapolate to lower concentrations. The results of soil 4 should therefore not be used for extrapolation i.e. risk assessment.

Loamy sand (Kfads = 88.5 r2 = 0.99)
Soil 5 50 16.61 2651.4 159.63 SAND total [%] 71.60
100 39.22 4692.3 119.64 SILT total [%] 12.70
500 168.08 26315.4 156.56 CLAY total [%] 6.00
1000 250.01 61404.1 245.61 pH in CaCl2 3.2
5000 total carbon [%] 3.7
organic carbon [%] 4.43
C.E.C. [mval/100g] 24.1

Based on the adsorption isotherm results, the test substance demonstrated a high tendency to adsorb to the five test soils. There was no correlation between the degree of adsorption and the organic carbon content of the soil. This suggests that other mechanisms were involved in the adsorption of the test substance to soil (e.g. interactions with the inorganic matter of the soil). A correlation between Kd and the content of clay could be assumed.Therefore, on the basis of these results the test substance can be considered immobile in soil. Based on the desorption isotherm results,the test substance demonstrated a low desorption, which underlines the tendency to adsorb on soils.

Validity criteria fulfilled:

yes

Conclusions:

Under the study conditions, the test substance demonstrated a high tendency to adsorb to the five soils. There was no correlation between degree of adsorption and organic carbon content of the soil. This suggested that other mechanisms were involved in the adsorption to soil (e.g. interactions with inorganic matter). A correlation between Kd and the content of the clay could be assumed. The deviation between Koc and KFOC is lower for soils with low content of clay and silt. Desorption was also very low (4-28%) after 48 h for all soils, indicating that the test substance adsorbed on the soils.

Executive summary:

A study was conducted to determine the adsorption and desorption potential of the test substance (not radiolabelled) to soil according to OECD Guideline 106, in compliance with GLP. The amount of substance adsorbed to soil was calculated from its depletion in the solution (indirect method). Equilibrium adsorption coefficients on various soils were determined as a function of soil characteristics (e.g. organic carbon content, clay content, soil texture and pH). Five soil types were used in order to cover as widely as possible the interactions with natural soils. The soil/solution ratio used was 1:100 (1 g soil and 100 mL solution). Based on preliminary tests, adsorption time was set to 24 h. Desorption time was set to 20 h for Soils no. 1, 2, 3, 5 and 24 h for Soil no. 4. Preliminary studies confirmed that there was no adsorption to the apparatus or filter material and stability of the test substance in the pH range of selected soils. A preliminary mass balance showed recoveries between 68 and 143%. Therefore, both aqueous and soil phases were analysed in the definitive study. Based on the results of the preliminary tests, 100, 500, 1000, 5000 and 10000 mg/L were selected as nominal test substance concentrations for adsorption/desorption isotherms for Soils no. 1, 2 and 4. For Soils 3 and 5, the nominal test substance concentrations were 50, 100, 500, 1000 and 5000 mg/L. Test substance concentrations were determined in CaCl2 solution via HPLC using post column derivatisation. The concentration adsorbed to soil was then determined by calculation.

The distribution coefficient (Kd, in cm3/g), Kd corrected for % organic carbon (Koc, in cm3/g); desorption coefficients (Kdes, in cm3/g); Freundlich adsorption/desorption coefficients (including KFOC; KadsF; log KadsF and KdesF; log KdesF) of the test substance in 5 soils were:

• Soil 1 (clay): Kd = 2781; Koc = 84,529; Kdes= 2685; KFOC= 348511; KadsF= 11466; log KadsF= 4.05; KdesF= 8251; log KdesF= 3.91
• Soil 2 (silt loam): Kd = 598; Koc = 25021; Kdes= 428; KFOC= 116485; KadsF= 2784; log KadsF= 3.44; KdesF= 3,209; log KdesF= 3.5
• Soil 3 (loam): Kd = 239; Koc = 7199; Kdes= 259; KFOC= 13524; KadsF= 449; log KadsF= 2.65; KdesF= 656; log KdesF= 2.82
• Soil 4 (silt): Kd = 622; Koc = 45,735; Kdes= 284; KFOC= 1268676; KadsF= 17254; log KadsF= 4.23; KdesF= 3562; log KdesF= 3.55
• Soil 5 (loamy sand): Kd = 121; Koc = 2731; Kdes= 524; KFOC= 1986; KadsF= 88; log KadsF= 1.94; KdesF= 815; log KdesF= 2.91
• Kd values originating from the lowest aqueous measured concentrations letting the intercept equal zero: 1957 (Soil 1), 561 (Soil 2), 194 (Soil 3) and 145 (Soil 5). Overall mean Kd = 714 cm3/g (Soil 4 was excluded from the calculation of the mean as results for this soil were inconsistent)

Under the study conditions, the test substance demonstrated a high tendency to adsorb to the five soils. There was no correlation between degree of adsorption and organic carbon content of the soil. This suggests that other mechanisms are involved in the adsorption to soil (e.g. interactions with inorganic matter), as expected for a cationic surfactant. In this case, Kd is a more relevant descriptor for the substance than Koc. Furthermore, non-linear sorption is normally observed in standard sorption desorption tests with cationic surfactants. Extrapolation based on these non-linear Freundlich isotherms to environmental representative concentration gives frequently unrealistic predictions. A Danish EPA report (Styczen and Petersen, 2004; RSS included) comes to the conclusion that, for these substances, the most reliable method of extrapolation is to use the data originating from the lowest aqueous measured concentrations letting the intercept equal zero. Using this approach, the observed Kd values for loamy sand, silty sand and clay loam are 1957 (Soil 1), 561 (Soil 2), 194 (Soil 3) and 145 (Soil 5) cm3/g, respectively, with an overall mean of 714 cm3/g. Soil 4 was excluded from the calculation of the mean as results for this soil were inconsistent (Geffke, 2003).

Reference 2

Endpoint:

adsorption / desorption: screening

Type of information:

experimental study

Adequacy of study:

supporting study

Study period:

2007

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

OECD Guideline 106 (Adsorption - Desorption Using a Batch Equilibrium Method)

GLP compliance:

yes

Type of method:

batch equilibrium method

Media:

soil

Radiolabelling:

yes

Analytical monitoring:

yes

Duration:

0.5 h

Initial conc. measured:

5 other: mg/L

Temp.:

25 °C

Key result

Type:

Kd

Value:

>= 2 970 - <= 10 500 other: cm3/g

Type:

Koc

Value:

> 26 000 - < 551 000 other: cm3/g

Details on results (Batch equilibrium method):

The test substance showed a very high affinity to all five soil types, which was confirmed by the high overall Kd values of 2970 to at least 10500 cm3/g.

Validity criteria fulfilled:

yes

Conclusions:

Under the study conditions, the test substance had little or no potential for mobility in soil.

Executive summary:

A study was conducted to determine adsorption/desorption of the test substance (radiolabelled) in soil according to OECD Guideline 106 in compliance with GLP. The main test was performed using a reduced adsorption equilibration period of 30 min. The initial test substance concentration used was 5.0 mg/L (measured). The preliminary test showed that the test substance had a very high affinity for soil, therefore the main test was performed at the maximum soil to solution ratio recommended by the guideline (1:100). A high percentage adsorption was detected for all soil types. The overall Kd range was 2970 to at least 10500 g/cm3 (average of 6322 g/cm3), the respective Koc range was from 26000 to at least 551000 g/cm3. Under the study conditions, the test substance had little or no potential for mobility in soil (O’Connor, 2007).

Description of key information

Adsorption-desorption study 1 (Geffke, 2003)

Under the study conditions, the test substance demonstrated a high tendency to adsorb to the five soils. There was no correlation between degree of adsorption and organic carbon content of the soil. This suggests that other mechanisms are involved in the adsorption to soil (e.g. interactions with inorganic matter), as expected for a cationic surfactant. In this case, Kd is a more relevant descriptor for the substance than Koc.

Furthermore, non-linear sorption is normally observed in standard sorption desorption tests with cationic surfactants. Extrapolation based on these non-linear Freundlich isotherms to environmental representative concentration gives frequently unrealistic predictions. A Danish EPA report (Styczen and Petersen, 2004; RSS included) comes to the conclusion that, for these substances, the most reliable method of extrapolation is to use the data originating from the lowest aqueous measured concentrations letting the intercept equal zero. Using this approach, the observed Kd values for loamy sand, silty sand and clay loam are 1957 (Soil 1), 561 (Soil 2), 194 (Soil 3) and 145 (Soil 5) cm3/g, respectively, with an overall mean of 714 cm3/g. Soil 4 was excluded from the calculation of the mean as results for this soil were inconsistent (Geffke, 2003).

Adsorption-desorption study 2 (O'Conner, 2007)

A study was conducted to determine adsorption/desorption of the test substance (radiolabelled) in soil according to OECD Guideline 106 in compliance with GLP. The main test was performed using a reduced adsorption equilibration period of 30 min. The initial test substance concentration used was 5.0 mg/L (measured). The preliminary test showed that the test substance had a very high affinity for soil, therefore the main test was performed at the maximum soil to solution ratio recommended by the guideline (1:100). A high percentage adsorption was detected for all soil types. The overall Kd range was 2970 to at least 10500 g/cm3 (average of 6322 g/cm3), the respective Koc range was from 26000 to at least 551000 g/cm3. Under the study conditions, the test substance had little or no potential for mobility in soil (O’Connor, 2007).

Discussion

Cationic substances such as the test substance do not behave in the environment such as standard (narcotic) substances; partitioning in soil and sediment is mainly based on ionic interactions and not linked to organic carbon (Droge and Goss, 2013). Therefore, for the present assessment, it is relevant to use the Kd rather than the Koc as key parameter.

Two adsorption-desorption studies were conducted, covering a wide range of concentrations. The study by Geffke (2003), was run with non-radiolabeled substance, so that starting concentrations were higher than in the study by O’Connor (2007) which was performed with labelled substance. Kd values for this study were recalculated from the lowest aqueous measured concentration(s) letting the intercept equal to zero, as per Styczen and Petersen (2004).

For risk assessment purposes, the overall mean Kd from the two studies will be used, i.e. 3518 cm3/g.

References

• Styczen M and Petersen S, 2004. Scenarios and model describing fate and transport of pesticides in surface water. http://www2.mst.dk/udgiv/publications/2004/87-7614-251-5/pdf/87-7614-252-3.pdf. Report no: 87-7614-252-3.
• Droge S and Goss KU, 2013. Development and Evaluation of a New Sorption Model for Organic Cations in Soil: Contributions from Organic Matter and Clay Minerals (publication), Environ. Sci. Technol. Testing laboratory: Department of Analytical Environmental Chemistry, Helmholtz Centre for Environmental Research−UFZ,, Report no: doi: 10.1021/es4031886. Owner company; APAG representing AKZO Nobel Surface Chemistry AB, Ceca SA, Kao Chemicals Europe and Clariant Produkten (Deutch land) GmbH, Report date: Dec 17, 2013.

Key value for chemical safety assessment

Other adsorption coefficients

Type:

log Kp (solids-water in soil)

Value in L/kg:

3.546

at the temperature of:

20 °C

Additional information

Adsorption-desorption study 1 (Geffke, 2003)

A study was conducted to determine the adsorption and desorption potential of the test substance (not radiolabelled) to soil according to OECD Guideline 106, in compliance with GLP. The amount of substance adsorbed to soil was calculated from its depletion in the solution (indirect method). Equilibrium adsorption coefficients on various soils were determined as a function of soil characteristics (e.g. organic carbon content, clay content, soil texture and pH). Five soil types were used in order to cover as widely as possible the interactions with natural soils. The soil/solution ratio used was 1:100 (1 g soil and 100 mL solution). Based on preliminary tests, adsorption time was set to 24 h. Desorption time was set to 20 h for Soils no. 1, 2, 3, 5 and 24 h for Soil no. 4. Preliminary studies confirmed that there was no adsorption to the apparatus or filter material and stability of the test substance in the pH range of selected soils. A preliminary mass balance showed recoveries between 68 and 143%. Therefore, both aqueous and soil phases were analysed in the definitive study. Based on the results of the preliminary tests, 100, 500, 1000, 5000 and 10000 mg/L were selected as nominal test substance concentrations for adsorption/desorption isotherms for Soils no. 1, 2 and 4. For Soils 3 and 5, the nominal test substance concentrations were 50, 100, 500, 1000 and 5000 mg/L. Test substance concentrations were determined in CaCl2 solution via HPLC using post column derivatisation. The concentration adsorbed to soil was then determined by calculation.

The distribution coefficient (Kd, in cm3/g), Kd corrected for % organic carbon (Koc, in cm3/g); desorption coefficients (Kdes, in cm3/g); Freundlich adsorption/desorption coefficients (including KFOC; KadsF; log KadsF and KdesF; log KdesF) of the test substance in 5 soils were:

• Soil 1 (clay): Kd = 2781; Koc = 84,529; Kdes= 2685; KFOC= 348511; KadsF= 11466; log KadsF= 4.05; KdesF= 8251; log KdesF= 3.91
• Soil 2 (silt loam): Kd = 598; Koc = 25021; Kdes= 428; KFOC= 116485; KadsF= 2784; log KadsF= 3.44; KdesF= 3,209; log KdesF= 3.5
• Soil 3 (loam): Kd = 239; Koc = 7199; Kdes= 259; KFOC= 13524; KadsF= 449; log KadsF= 2.65; KdesF= 656; log KdesF= 2.82
• Soil 4 (silt): Kd = 622; Koc = 45,735; Kdes= 284; KFOC= 1268676; KadsF= 17254; log KadsF= 4.23; KdesF= 3562; log KdesF= 3.55
• Soil 5 (loamy sand): Kd = 121; Koc = 2731; Kdes= 524; KFOC= 1986; KadsF= 88; log KadsF= 1.94; KdesF= 815; log KdesF= 2.91
• Kd valuesoriginating from the lowest aqueous measured concentrations letting the intercept equal zero: 1957 (Soil 1), 561 (Soil 2), 194 (Soil 3) and 145 (Soil 5). Overall mean Kd = 714 cm3/g (Soil 4 was excluded from the calculation of the mean as results for this soil were inconsistent)

Under the study conditions, the test substance demonstrated a high tendency to adsorb to the five soils. There was no correlation between degree of adsorption and organic carbon content of the soil. This suggests that other mechanisms are involved in the adsorption to soil (e.g. interactions with inorganic matter), as expected for a cationic surfactant. In this case, Kd is a more relevant descriptor for the substance than Koc.

Furthermore, non-linear sorption is normally observed in standard sorption desorption tests with cationic surfactants. Extrapolation based on these non-linear Freundlich isotherms to environmental representative concentration gives frequently unrealistic predictions. A Danish EPA report (Styczen and Petersen, 2004; RSS included) comes to the conclusion that, for these substances, the most reliable method of extrapolation is to use the data originating from the lowest aqueous measured concentrations letting the intercept equal zero. Using this approach, the observed Kd values for loamy sand, silty sand and clay loam are 1957 (Soil 1), 561 (Soil 2), 194 (Soil 3) and 145 (Soil 5) cm3/g, respectively, with an overall mean of 714 cm3/g. Soil 4 was excluded from the calculation of the mean as results for this soil were inconsistent (Geffke, 2003).

Adsorption-desorption study 2 (O'Conner, 2007)

A study was conducted to determine adsorption/desorption of the test substance (radiolabelled) in soil according to OECD Guideline 106 in compliance with GLP. The main test was performed using a reduced adsorption equilibration period of 30 min. The initial test substance concentration used was 5.0 mg/L (measured). The preliminary test showed that the test substance had a very high affinity for soil, therefore the main test was performed at the maximum soil to solution ratio recommended by the guideline (1:100). A high percentage adsorption was detected for all soil types. The overall Kd range was 2970 to at least 10500 g/cm3 (average of 6322 g/cm3), the respective Koc range was from 26000 to at least 551000 g/cm3. Under the study conditions, the test substance had little or no potential for mobility in soil (O’Connor, 2007).

Discussion

Cationic substance such as the test substance do not behave in the environment such as standard (narcotic) substances; partitioning in soil and sediment is mainly based on ionic interactions and not linked to organic carbon (Droge and Goss, 2014). Therefore, for the present assessment, it is relevant to use the Kd rather than the Koc as key parameter.

Two adsorption-desorption studies were conducted, covering a wide range of concentrations. The study by Geffke (2003), was run with non-radiolabeled substance, so that starting concentrations were higher than in the study by O’Connor (2007) with labelled substance. Kd values for this study were recalculated from the lowest aqueous measured concentration(s) letting the intercept equal to zero, as per Styczen and Petersen (2004). For risk assessment purposes, the overall mean Kd from the two studies will be used, i.e. 3518 cm3/g.

References

Styczen M and Petersen S, 2004. Scenarios and model describing fate and transport of pesticides in surface water.http://www2.mst.dk/udgiv/publications/2004/87-7614-251-5/pdf/87-7614-252-3.pdf. Report no: 87-7614-252-3.

Droge S and Goss KU, 2013. Development and Evaluation of a New Sorption Model for Organic Cations in Soil: Contributions from Organic Matter and Clay Minerals (publication), Environ. Sci. Technol. Testing laboratory: Department of Analytical Environmental Chemistry, Helmholtz Centre for Environmental Research−UFZ,, Report no: doi: 10.1021/es4031886. Owner company; APAG representing AKZO Nobel Surface Chemistry AB, Ceca SA, Kao Chemicals Europe and Clariant Produkten (Deutch land) GmbH, Report date: Dec 17, 2013.