Registration Dossier

Diss Factsheets

Environmental fate & pathways

Biodegradation in water and sediment: simulation tests

Currently viewing:

Administrative data

Link to relevant study record(s)

Referenceopen allclose all

Endpoint:
biodegradation in water: sediment simulation testing
Type of information:
experimental study
Adequacy of study:
key study
Study period:
26 June 2014 to 02 March 2015
Reliability:
1 (reliable without restriction)
Rationale for reliability incl. deficiencies:
guideline study
Qualifier:
according to guideline
Guideline:
OECD Guideline 308 (Aerobic and Anaerobic Transformation in Aquatic Sediment Systems)
Version / remarks:
2002
Deviations:
no
GLP compliance:
yes (incl. QA statement)
Radiolabelling:
yes
Oxygen conditions:
aerobic
Inoculum or test system:
natural water / sediment: freshwater
Details on source and properties of surface water:
TEST SYSTEM
Two natural aquatic sediment systems from sites in the UK were used for the study. The test systems were selected to provide contrasting properties with respect to particle size distribution, organic carbon content and pH.
The sediments and their overlying waters were supplied by Land Research Associates, Derby, UK and certificated by Agvise, Northwood, North Dakota, US. The water samples were characterised with respect to pH, hardness, dissolved and total organic carbon, total nitrogen and total phosphorus. On arrival at the testing facility, the sediments and associated waters were stored in accordance with ISO standard 10381-6 to maintain viability.

CALWICH ABBEY
- Details on collection: The sediment and associated water were collected from Calwich Abbey Lake, Ashbourne, Staffordshire on 26 August 2014. Taken from the lake by bucket and passed through a 212 μm sieve into two 20 litre containers. Pesticide history - only one agricultural field abuts the lake. All other shores are wooded. The tenant of the agricultural land stated that no pesticide has been used on it in the past 5 years.
- Storage conditions: ca. 5 °C in the dark
- Storage length: 9 days
- Temperature (°C) at time of collection: 16.4
- pH at time of collection: 8.75
- Electrical conductivity: 532 µS/cm
- Oxygen concentration: 125 %
- Hardness (CaCO3): 257 mg equivalent CaCO3/L
- Dissolved organic carbon: 1.6 ppm
- Water filtered: Yes
- Type and size of filter used, if any: The water samples had been passed through a 212 μm sieve.

SWISS LAKE
- Details on collection: The sediment and associated water were collected from Swiss Lake, Chatsworth, Derbyshire on 27 August 2014. Scooped from lake then passed through a 212 μm sieve into two 20 litre containers. Pesticide history - only one agricultural field abuts the lake. All other shores are wooded. The tenant of the agricultural land stated that no pesticide has been used on it in the past 5 years.
- Storage conditions: ca. 5 °C in the dark
- Storage length: 9 days
- Temperature (°C) at time of collection: 14.7
- pH at time of collection: 8.72
- Electrical conductivity: 128 µS/cm
- Oxygen concentration: 127.1 %
- Hardness (CaCO3): 35 mg equivalent CaCO3/L
- Dissolved organic carbon: 8.7 ppm
- Water filtered: Yes
- Type and size of filter used, if any: The water samples had been passed through a 212 μm sieve.
Details on source and properties of sediment:
TEST SYSTEM
The sediments and their overlying waters were supplied by Land Research Associates, Derby, UK and certificated by Agvise, Northwood, North Dakota, US. The sediments were fully characterised, with respect to particle size distribution, organic carbon, pH, cation exchange capacity, total nitrogen and total phosphorous. The sediments were classified under the USDA system. On arrival at the testing facility, the sediments and associated waters were stored in accordance with ISO standard 10381-6 to maintain viability.
Prior to use the moisture content of the sediments was determined. The microbial biomass of the sediments was determined by the fumigation-extraction method. Samples of sediment were sent for biomass determination at the beginning (post-acclimatisation) and the end of the study (after the last sampling point).

CALWICH ABBEY
- Details on collection: The sediment and associated water were collected from Calwich Abbey Lake, Ashbourne, Staffordshire on 26 August 2014. Scooped from the top 5 cm of sediment onto the bank to drain slightly and then passed through a 2 mm sieve into two 10.4 litre kegs. Pesticide history - only one agricultural field abuts the lake. All other shores are wooded. The tenant of the agricultural land stated that no pesticide has been used on it in the past 5 years.
- Storage conditions: ca. 5 °C in the dark
- Storage length: 9 days
- Textural classification: USDA Textural Class Silt loam (39 % sand, 52 % silt and 9 % clay); A.D.A.S. Textural Class Sandy silt loam (37 % sand, 54 % silt and 9 % clay)
- pH: pH in 1:1 soil:water ratio: 7.2; pH in 1 N KCl: 7.1; pH in 0.01 M CaC12 (1: 2): 7.1
- Organic carbon (%): 5.0
- CEC (meq/100 g): 10.7
- Biomass: Initial: 752.1 mg C/kg sediment; Final: 891.0 mg C/kg sediment
- % Moisture content: 158.69
- Sediment samples sieved: Yes, the sediments had been passed through a 2 mm sieve

SWISS LAKE
- Details on collection: The sediment and associated water were collected from Swiss Lake, Chatsworth, Derbyshire on 27 August 2014. Scooped from lake top 5 cm of sediment onto the bank to drain slightly in a bucket then passed through a 2 mm sieve into one 15.4 litre keg. Pesticide history - only one agricultural field abuts the lake. All other shores are wooded. The tenant of the agricultural land stated that no pesticide has been used on it in the past 5 years.
- Storage conditions: ca. 5 °C in the dark
- Storage length: 9 days
- Textural classification: USDA Textural Class Loamy sand (87 % sand, 8 % silt and 5 % clay); A.D.A.S. Textural Class Loamy sand (83 % sand, 12 % silt and 5 % clay)
- pH: pH in 1:1 soil:water ratio: 6.6; pH in 1 N KCl: 6.2; pH in 0.01 M CaC12 (1: 2): 6.1
- Organic carbon (%): 0.71
- CEC (meq/100 g): 3.1
- Biomass: Initial: 108.7 mg C/kg sediment; Final: 129.5 mg C/kg sediment
- % Moisture content: 32.64
- Sediment samples sieved: Yes, the sediments had been passed through a 2 mm sieve
Duration of test (contact time):
98 d
Initial conc.:
138 mg/L
Based on:
test mat.
Parameter followed for biodegradation estimation:
radiochem. meas.
Details on study design:
EXPERIMENTAL SETUP
- Duration of test: 98 days
- Water/sediment condition: Freshly sampled, sediment sieved (≤ 2 mm), entire water/sediment systems pre-incubated under test conditions for 7 days prior to treatment.
- Target application rate 46.8 μg per flask
- Concentration in test system: Nominal: 46.8 μg per flask (equivalent to initial concentration of 0.138 mg/L in water column); Measured: 46.5 μg per flask (equivalent to 0.136 mg/L based on 340 mL overlying water); % Target: 99.4 %.
- Number of replications: Two
- Test apparatus: Water sediment flasks containing sediment and natural water (approximate volume ratio of 1:3 for sediment: water).
- Weight of sediment per vessel (to give ca. 3 - 4 cm depth): ca. 150 g for Calwich Abbey (58 g oven dried equivalent (ode)); ca. 140 g for Swiss Lake (105 g ode).
- Volume of natural water per vessel (to give ca. 12 cm total depth): ca. 340 mL for Calwich Abbey; ca. 340 mL for Swiss Lake.
- Test material application: Identity of solvent: Water: Acetonitrile (90:10); Volume of application solution: 320 μL; Application method: Positive displacement pipette.
- Traps for CO2 and organic volatiles: One ethylene glycol trap followed by two potassium hydroxide traps.
- Is there any indication of the test material absorbing to the walls of the test apparatus: No
- Temperature 20 ± 2 °C
- Lighting: Dark
- Aeration: Aerobic conditions in the water phase were maintained by constant passage of moist air through the sample flasks and out through the trap solutions.

The sediment and associated water were added to specially adapted individual glass incubation flasks with a screw top and straight sides of approximately 600 mL capacity (ca 6.0 cm diameter). Each had an associated air-tight flask head with side-arm fittings to permit the passage of air through the flask. The flasks were connected to a series of trap vessels.
For both water sediment systems, eighteen flasks were prepared for treatment with [14C]-test material, allowing duplicate samples to be taken at each of six specified sampling timepoints whilst leaving six flasks which could be used as treated spares. Additionally, four flasks were prepared and remained untreated.
Approximately 58 g oven-dried equivalent of Calwich Abbey Lake sediment and 105 g oven-dried equivalent of Swiss Lake sediment (each sieved to 2 mm), along with ca. 340 mL of the associated water, was dispensed into the glass flasks. The samples were allowed to acclimatise under study conditions for 7 days prior to application of the test material. A ratio of approximately 1:3 v/v (sediment: water) were obtained for the Calwich Abbey and Swiss Lake systems, (ca. 3 cm depth of sediment with overlying water ca. 12 cm). All flasks were attached to an incubation system through which moistened air was bubbled, at a rate that allowed aeration of the water without disturbance of the sediment water interface. The passage of air was controlled by the use of flow restrictors consisting of lengths of glass rod with a capillary along their entire length. These ensured a uniform flow rate into each flask and allowed individual flasks to be disconnected without disrupting the flow through those remaining. Each flask was connected to a series of three traps, the first containing ethylene glycol, and the second and third containing 2M potassium hydroxide.
The water/sediment systems were incubated at 20 ± 2 °C in the dark until there was complete phase separation and the oxygen levels, pH and redox potentials had been established. Following treatment, each flask was returned to the incubation room where the temperature was maintained at 20 ± 2 °C throughout the course of the study.

CONTROL FLASKS
Eight flasks were prepared for each sediment type to be used for determination of sediment biomass (four each for initial and final biomass samples). The four flasks for the final biomass determination were treated with 320 μL of water: acetonitrile 90:10 (to mimic the test material application in the study flasks in case of adverse effects on the microbial activity) remained on the system throughout the study. These were also used to measure the water and sediment conditions (pH, oxygen and redox potential) throughout the duration of the study.

PREPARATION AND APPLICATION OF TEST MATERIAL
- Preparation of Treatment Solution
The treatment solution was prepared by transferring 2 mL of the supplied stock solution (1.36 mg/mL) to a 20 mL volumetric flask and making to volume with acetonitrile/water 10/90. Aliquots (100 μL) of this solution were diluted to 25 mL with acetonitrile and triplicate 100 μL aliquots of this solution were counted by LSC to determine the exact concentration of the solution. The concentration of the treatment solution was determined to be 0.145 mg/mL.
- Application Procedures
The water-sediment systems were each treated with the [14C]-treatment solution (320 μL) using a positive displacement pipette, adding the solution dropwise onto the water surface. All of the sample flasks were treated on 11 September 2014. Treatment checks were made before, during and after treatment by dispensing an equal volume of treatment solution directly into a 20 mL volumetric flask. The treatment checks were diluted to volume with acetonitrile and the radioactivity measured by LSC to determine the treatment rate achieved.
- Justification of Study Application Rate
The amount to be applied was based upon considerations of the water phase concentration that would result from direct overspray from a field application, at a rate of 1 380 g/ha, to a water body of 100 cm depth. The water/sediment systems were treated therefore with ca. 46.8μg of [14C]-test material per flask. This application equated to an initial water phase concentration of 0.138 mg/L and thus significantly less than the aqueous solubility of the test material which is >860 mg/L in water.

MEASUREMENT OF REDOX POTENTIAL, DISSOLVED OXYGEN AND PH OF SEDIMENT AND WATER
The redox potential of the sediment and water was measured using an InLab® RedoxMicro ™ Redox electrode, 3M KCl, AgCl saturated reference electrolyte which was calibrated using a commercially available Redox calibration solution. For the conversion to the standard redox potential +200 mV should be added to the redox values obtained.
To measure the redox potential in water, the RedoxMicro ™ electrode was suspended in the water through the neck of the flask, ensuring it did not come in contact with the sediment surface. For measurement of the sediment redox potential, the electrode was carefully inserted into the sediment by passing through the water into the sediment via the neck of the flask.
The pH electrode was calibrated using previously prepared buffer solutions. The pH of the water was measured by inserting the electrode through the neck of the flask, ensuring that it was not in contact with the sediment surface, and recording the value obtained.
Prior to use, the oxygen meter was calibrated to zero in a commercially available zero oxygen solution, and to 100 % in air. The oxygen electrode was then inserted into the water phase of the flasks and the reading taken.

TRAPPING OF VOLATILE PRODUCTS
The trapping solutions, consisting of three 50 mL tubes, the first containing ethylene glycol followed by two containing 2M potassium hydroxide, were connected in series to the incubation flasks. Moist air was bubbled constantly, at a consistent rate, through the flasks and trap vessels during the course of the study, with interruption to the flow only occurring during flask maintenance. The traps were removed and replaced at appropriate intervals to ensure efficient trapping of evolved volatile metabolites. The trap solutions were removed and analysed.

SAMPLING INTERVALS
Duplicate flasks, and their associated traps, were removed at each sampling interval. Samples were taken at zero time and following 7, 14, 29, 56, 81 and 98 days of incubation.

CALCULATIONS
- Determination of Degradation Kinetics
DT50 and DT90 values for the degradation of the test material in the water phase and in the total water/sediment systems were determined following the recommendations of the FOCUS work group. The degradation kinetics were estimated according to FOCUS recommended procedures for determination of persistence endpoints using the software CAKE. A model input data set was derived from the individual data for each time-point, with figures taken from the water phase and the total system. The following kinetic models were tested in order to determine the best-fit kinetic model:

- Simple first order model (SFO):
Mp(t) = M0exp^(-kt)
Where:
Mp(t) = Total amount of chemical present at time t
M0 = Total amount of chemical present at time t = 0
k = Rate constant [d^-1]

- First order multi compartment model (FOMC):
Mp(t) = M0[(t/b) + 1]^-a
Where:
MP(t) = Total amount of chemical present at time t
M0 = Total amount of chemical present at time t = 0
a = Shape parameter determined by CV of k values
b = Location parameter

- Bi-exponential model (double first order in parallel, DFOP):
M(t) = M0(ge^-k1t + (1 - g)e^-k2t)
Where:
M(t) = Total amount of chemical present at time t
M0 = Total amount of chemical present at time t = 0
g = The fraction of M0 applied to compartment 1
k1 = Rate constant in compartment 1 [d^-1]
k2 = Rate constant in compartment 2 [d^-1]

- Hockey-Stick model (HS)
For t < tb: Mp(t) = M0exp(-k1t)
For t ≥ tb: Mp(t) = M0exp(-k1tb - k2(t - tb))
Where:
MP(t) = Total amount of chemical present at time t
M0 = Total amount of chemical present at time t = 0
k1 = Rate constant until break time (tb) [d^-1]
k2 = Rate constant after break time (tb) [d^-1]
tb = Break-time [d]

The best-fit kinetic model was selected on the basis of a visual assessment of the goodness of fit (diagrams of measured and calculated values versus time, diagrams of residuals versus time) and on the basis of the chi^2 scaled-error criterion. The significance of the estimated parameters was also confirmed by a single-sided t-test. A t-test probability of <0.05 (>95 % parameter significance) is usually considered sufficiently small.
The dissipation times DT50 and DT90 (time until 50 or 90 % of disappearance) were calculated by the software from the optimised kinetic parameters for the best-fit kinetic model.

STATISTICAL METHODS
Where a mean or average value is quoted it is the arithmetical mean that is used:
Mean = Σx / n

The standard deviation (std dev) from the mean, σn-1 is calculated from:
σn-1 = √((Σ(mean - x)^2) / n - 1)
Where n is the number of values of x.
This calculation assumes that only a sample of the population of values of x is known and was performed in Excel using the STDEV function.

The coefficient of variation, CV, is calculated from:
CV = (σn-1 x 100) / mean
Test performance:
The actual application rate of [14C]-test material to individual test vessels was determined as 23 003 443 dpm (46.5 μg, 99.4 % of target). The treatment rate checks demonstrated good homogeneity of the treatment solutions during application.

- Properties of Test Systems
The temperature was maintained at 20 ± 2 °C throughout the duration of the study.
The pH of the waters remained at approximately 7 - 8 for both the Calwich Abbey and the Swiss Lake systems during the incubation period (average pH 7.7 and 7.4, respectively). The redox potentials showed that the water phases remained aerobic throughout the study whilst the sediments were anaerobic and remained in this state for the duration of the study. The oxygen content of the water phases for both systems fluctuated around mean values of ca. 8.7 mg/L throughout the study (8.8 mg/L for Calwich Abbey and 8.6 mg/L for Swiss Lake).

- Verification of Chromatographic Procedures
HPLC column recoveries were found to be quantitative with 98.2 - 110.5 % (mean 104.1 %) of injected radioactivity collected post-column for the selected samples.

- Storage Stability
Water samples and sample extracts were stored refrigeraged (ca. 5 °C) whilst waiting for analysis otherwise were stored frozen (ca. < -15 °C). Water and sediment samples were profiled by HPLC on the same day of sampling in all cases with only one exception, the 56 day samples which were analysed within 24 h of generation.
Compartment:
natural water / sediment: freshwater
% Recovery:
96.9
Remarks on result:
other: Calwich Abbey system
Compartment:
natural water / sediment: freshwater
% Recovery:
99.7
Remarks on result:
other: Swiss Lake system
Key result
Compartment:
natural water / sediment: freshwater
DT50:
83.2 d
Type:
other: Hockey stick model
Temp.:
20 °C
Remarks on result:
other: Calwich Abbey system
Key result
Compartment:
natural water / sediment
DT50:
244 d
Type:
other: Simpe first order
Temp.:
20 °C
Remarks on result:
other: Swiss Lake system
Transformation products:
not specified
Remarks:
No intermediate metabolites were detected at, or above, an amount representing 5 % of the applied test material
Details on results:
MATERIAL BALANCE
The overall recovery of radioactivity was good throughout the study with mean values of 96.9 % of applied radioactivity (AR) for Calwich Abbey and 99.7 % AR for Swiss Lake. Recoveries for individual flasks ranged from 93.1 to 102.0 % AR for the Calwich Abbey system and 95.3 to 101.7 % AR for the Swiss Lake system.

DISTRIBUTION OF RADIOACTIVITY
In the Calwich Abbey system the levels of radioactivity in the water declined steadily from 101.3 AR to 48.2 % AR at day 81. After this initial lag phase, however, the rate of decline from the water accelerated rapidly, reaching 4.1 % AR at day 98. The total extractable radioactivity in the sediment rose from 0.0 % AR at time zero to a maximum of 23.0 % at day 56 before declining to 6.6 % at the end of the study at day 98. The unextracted radioactivity rose from 0.0 % at time zero to 32.3 % at day 98. The presence of an initial metabolic lag phase was also evident in the evolution of CO2. Over the first 56 days only 4.9 % CO2 was released, however, this had increased to 14.8 % AR by day 81 and reached 50.1 % AR by day 98.
In the Swiss Lake system there was a slower transfer of the applied radioactivity from the water to the sediment, when compared with Calwich Abbey system, and thus resulting in a much greater percentage of applied radioactivity remaining in the water at the end of the incubation period. The radioactivity in the water declined from 101.1 % AR at time zero to 62.7 % AR by day 98. The total radioactivity in the sediment increased from 0.0 % at time zero to 21.8 % by day 98, with the extractable portion reaching a maximum of 12.8 % AR at day 56 and the unextractable reaching a maximum of 10.4 % AR at day 98. The degree of mineralisation to CO2 was less than in the Calwich Abbey system accounting for 13.5 % AR by day 98.

DISSIPATION OF THE TEST MATERIAL AND FORMATION AND DECLINE OF METABOLITES
- Water
HPLC analysis showed that the test material dissipated from the water phase of both systems. In the Calwich Abbey system, the test material declined steadily from 101.3 % AR at time zero to 47.5 % AR at day 81 but degradation then accelerated rapidly with only 3.7 % AR remaining as test material at day 98. In the Swiss Lake system, the test material declined at a steadier rate from 101.1 % at time zero to 62.1 % by day 98. In both systems, several minor metabolites were observed, none of which accounted for more than 0.8 % AR.
- Sediment
In the Calwich Abbey system, HPLC analysis showed that the test material reached a maximum level of 22.1 % AR at day 56 and subsequently declined to 5.0 % AR by day 98. Two minor metabolites were detected, neither accounting for >1.6 % AR.
In the Swiss Lake system, HPLC analysis showed that the test material reached a maximum level of 13.0 % AR at day 14 and declined to 11.4 % AR by the end of the study. A single minor metabolite was observed at 0.1 % AR at day 98.
- Total System
In the total Calwich Abbey system, comprising sediment and overlying water, the test material declined from 101.3 % AR at time zero to 66.8 % AR at day 81. After day 81 degradation accelerated rapidly such that only 8.7 % AR remained as test material by day 98.
In the total Swiss Lake system, the applied test material declined more steadily, from 101.1 % AR at time zero to 73.4 % AR at day 98.
Several minor metabolites were detected throughout the course of the study but none exceeded 5 % AR at any time point in the total system (combined water and sediment) of either system.

CHARACTERISATION OF NON-EXTRACTABLE RADIOACTIVITY
The level of unextracted radioactivity following extraction reached a maximum level of 32.3 % AR in the Calwich Abbey system at day 98. In the Swiss Lake system bound residues reached a maximum of 10.4 % AR at day 98. Further characterisation was therefore carried out on both sediments.
The unextracted radioactivity was principally associated with the insoluble humin fraction, in both systems (65.2 % of total non-extractable for Calwich Abbey, 52.9 % Swiss Lake). The remainder was largely associated with the humic acid fraction (34.4 % of total non-extractable for Calwich Abbey, 46.5 % for Swiss Lake). The fulvic acid fraction represented the remaining residues (0.4 % of total non-extractable for Calwich Abbey, 0.6 % Swiss Lake).

CONFIRMATION OF TEST MATERIAL AND DEGRADATES WITH MASS SPECTROMETRY
Full scan accurate mass data was collected for the standard and samples analysed using a Q-Extactive Mass Spectrometer. A mass error window of ± 5 ppm, calculated from the theoretical exact mass of the appropriate ions, was applied to full scan data and the selected ions were extracted.
The test material was confirmed in the following samples:
Day 29: Two samples of water confirmed to contain the test material.
Day 81: Two samples of sediment confirmed to contain the test material.

DEGRADATION KINETICS
In the Calwich Abbey system, degradation of the test material exhibited a lag-phase of moderate degradation followed by rapid decline, the Hockey-stick (HS) model was selected as best-fit to the data for the water phase and the total system. The DT50 and DT90 values were thus calculated as 72.5 and 91.9 days in the water and 83.2 and 96.7 days in the total system, respectively.
In the Swiss Lake system the SFO model was selected as best-fit for the degradation in the water and for the total system. The DT50 and DT90 values were thus calculated as 171 and 567 days in the water and 244 and 810 days in the total system, respectively.
The degradation kinetics for the test material were also estimated according to FOCUS recommended procedures for determination of modelling endpoints.

CONCLUSIONS
In natural water/sediment systems, under aerobic conditions in the dark at 20 °C, the test material was shown to degrade to carbon dioxide and unextractable sediment bound residues. No intermediate metabolites were detected at, or above, an amount representing 5 % of the applied test material.
In one of the systems tested, Calwich Abbey, degradation of the test material accelerated rapidly after an initial lag phase, with best-fit overall DT50 values of 72.5 days in the water phase and 83.2 days in the total water/sediment system. Degradation in the second system, Swiss Lake, was steadier, with best-fit DT50 values of 171 days in the water phase and 244 days in the total water/sediment system.

Distribution and Recovery of Radioactivity in Calwich Abbey System (as % of applied radioactivity)

Incubation Time (Days)

Sample ID

Water

Sediment Extracts (1-3)

Unextracted

Total in Sediment (extracted and unextracted)

Total Volatiles*

TOTAL

0

CA01

101.98

0.01

0.00

0.01

-

101.98

CA02

100.66

0.00

0.00

0.00

-

100.66

Mean

 

101.32

0.00

0.00

0.00

-

101.32

7

CA03

92.36

8.94

0.52

9.46

0.10

101.92

CA04

97.88

2.53

0.10

2.63

0.04

100.54

Mean

 

95.12

5.74

0.31

6.05

0.07

101.23

14

CA05

81.04

14.06

1.78

15.85

0.32

97.21

CA06

84.98

11.58

1.34

12.92

0.73

98.63

Mean

 

83.01

12.82

1.56

14.38

0.53

97.92

29

CA07

75.10

19316

1.61

20.77

0.38

96.25

CA08

72.17

20.21

2.19

22.40

0.03

94.59

Mean

 

73.63

19.69

1.90

21.59

0.20

95.42

56

CA10

60.19

22.48

7.37

29.85

3.80

93.84

CA11

57.12

23.59

7.16

30.76

6.04

93.91

Mean

 

58.65

23.04

7.27

30.30

4.92

93.87

81

CA14

46.36

20.13

13.65

33.78

16.00

96.13

CA15

50.08

21.30

10.02

31.32

13.62

95.01

Mean

 

48.22

20.71

11.83

32.55

14.81

95.57

98

CA09

4.15

6.36

33.19

39.56

49.34

93.05

CA12

4.06

6.89

31.44

38.34

50.76

93.16

Mean

 

4.11

6.63

32.32

38.95

50.05

93.10

 

Average

96.92

* The volatile radioactivity was trapped in the KOH traps and was confirmed as carbon dioxide by precipitation as barium carbonate for gravimetric analyses.

 

Distribution and Recovery of Radioactivity in Swiss Lake System (as % of applied radioactivity)

Incubation Time (Days)

Sample ID

Water

Sediment Extracts (1-3)

Unextracted

Total in Sediment (extracted and unextracted)

Total Volatiles*

TOTAL

0

SL50

100.58

0.00

0.00

0.00

-

100.58

SL51

101.67

0.00

0.00

0.00

-

101.67

Mean

 

101.12

0.00

0.00

0.00

-

101.12

7

SL53

88.42

10.78

0.71

11.49

0.25

100.16

SL54

94.65

6.26

0.43

6.69

0.16

101.49

Mean

 

91.53

8.52

0.57

9.09

0.20

100.83

14

SL55

81.91

14.91

2.27

17.18

0.79

99.88

SL56

82.21

11.12

2.42

13.54

2.31

98.06

Mean

 

82.06

13.02

2.35

15.36

1.55

98.97

29

SL57

86.59

11.88

0.75

12.63

0.24

99.46

SL58

86.71

12.45

1.00

13.45

0.15

100.31

Mean

 

86.65

12.17

0.88

13.04

0.19

99.88

56

SL60

79.06

12.19

3.82

16.01

4.58

99.65

SL61

74.51

13.43

7.78

21.21

5.51

101.24

Mean

 

76.79

12.81

5.80

18.61

5.04

100.44

81

SL52

71.68

10.32

4.53

14.85

10.88

97.42

SL67

73.08

12.64

4.22

16.86

9.70

99.65

Mean

 

72.38

11.48

4.37

15.86

10.29

98.53

98

SL62

63.38

11.21

7.06

18.27

13.68

95.33

SL63

61.96

11.73

13.64

25.37

13.21

100.54

Mean

 

62.67

11.47

10.35

21.82

13.45

97.93

 

Average

99.67

* The volatile radioactivity was trapped in the KOH traps and was confirmed as carbon dioxide by precipitation as barium carbonate for gravimetric analyses

 

Composition of Radioactivity in Total Water/Sediment System, Calwich Abbey (as % of applied radioactivity by HPLC)

Incubation Time (Days)

Sample ID

% AR

Unknown RRT 0.35 - 0.64

Unknown RRT 0.94 - 0.96

Test Material RRT 1.00

TOTAL

0

CA01

101.98

-

-

101.98

101.98

CA02

100.66

-

-

100.66

100.66

Mean

 

101.32

-

-

101.32

101.32

7

CA03

101.30

0.14

-

101.16

101.30

CA04

100.41

0.45

-

99.96

100.41

Mean

 

100.85

0.29

-

100.56

100.85

14

CA05

95.10

-

0.40

94.70

95.10

CA06

96.56

-

0.25

96.32

96.56

Mean

 

95.83

-

0.32

95.51

95.83

29

CA07

94.27

-

0.56

93.71

94.27

CA08

92.38

-

0.61

91.76

92.38

Mean

 

93.32

-

0.59

92.73

93.32

56

CA10

82.67

0.22

0.96

81.49

82.67

CA11

80.71

0.16

0.86

79.69

80.71

Mean

 

81.69

0.19

0.91

80.59

81.69

81

CA14

66.49

1.01

1.03

64.45

6.49

CA15

71.37

0.48

1.81

69.08

71.37

Mean

 

68.93

0.75

1.42

66.76

68.93

98

CA09

10.52

0.00

1.53

8.99

10.52

CA12

10.96

0.79

1.72

8.45

10.96

Mean

 

10.74

0.39

1.63

8.72

10.74

 

Composition of Radioactivity in Total Water/Sediment System, Swiss Lake (as % of applied radioactivity by HPLC)

Incubation Time (Days)

Sample ID

% AR

Unknown RRT 0.64 - 0.82

Unknown RRT 0.94 - 0.96

Test Material RRT 1.00

Unknown RRT 1.06

TOTAL

0

SL50

100.58

-

-

100.58

-

100.58

SL51

101.67

-

-

101.67

-

101.67

Mean

 

101.12

-

-

101.12

-

101.12

7

SL53

99.20

-

-

99.20

-

99.20

SL54

100.91

-

-

100.91

-

100.91

Mean

 

100.05

-

-

100.05

-

100.05

14

SL55

96.82

-

-

96.82

-

96.82

SL56

93.33

-

-

93.33

-

93.33

Mean

 

95.07

-

-

95.07

-

95.07

29

SL57

98.47

-

-

98.47

-

98.47

SL58

99.16

-

-

99.16

-

99.16

Mean

 

98.82

-

-

98.82

-

98.82

56

SL60

91.25

-

-

91.25

-

91.25

SL61

87.94

-

-

87.94

-

87.94

Mean

 

89.60

-

-

89.60

-

89.60

81

SL52

82.01

1.06

0.23

80.51

0.20

82.01

SL67

85.72

0.43

0.00

85.29

0.00

85.72

Mean

 

83.87

0.75

0.11

82.90

0.10

83.87

98

SL62

74.60

0.51

0.16

73.92

0.00

74.60

SL63

73.68

0.72

0.00

72.96

0.00

73.68

Mean

 

74.14

0.62

0.08

73.44

0.00

74.14

 

Characterisation of Non-Extractable Residues by Organic Matter Fractionation

Calwich Abbey System

Sampling Interval (Days)

Flask Number

As % of applied radioactivity

Fulvic Acid

Humic Acid

Humin

Total

98

CA09

0.10

8.02

15.22

23.34

As % of total non-extractable

0.43

34.35

65.22

100

Swiss Lake System

Sampling Interval (Days)

Flask Number

As % of applied radioactivity

Fulvic Acid

Humic Acid

Humin

Total

98

SL63

0.05

3.97

4.51

8.53

As % of total non-extractable

0.61

46.53

52.87

100

 

Summarised Best-Fit Kinetic Parameters

System

Phase

Kinetic Model

DT50 (days)

DT90 (days)

 

Chi^2 (%)

Confidence interval

Visual Fit

Calwich Abbey

Water

HS

72.5

91.9

2.9

k1: 1.50E-09

k2: 0.4714

Good

Total sytem

HS

83.2

96.7

1.3

k1: 2.38E-07

k2: 3.37E-08

Good

Swiss Lake

Water

SFO

171

567

4.0

8.58E-07

Good

Total sytem

SFO

244

810

2.5

5.93E-07

Good

 

Validity criteria fulfilled:
yes
Conclusions:
Under the conditions of this study, [14C]-test material in natural water/sediment systems was shown to degrade ultimately to carbon dioxide and unextractable sediment bound residues.
Executive summary:

The route and rate of degradation of the test material in two water/sediment systems was investigated in accordance with the standardised guideline OECD 308 under GLP conditions.

The route and rate of degradation of [14C]-test material has been investigated under aerobic conditions at 20 ± 2 °C in two contrasting water sediment systems in the dark. The water sediment systems were Calwich Abbey [Staffordshire, UK, silt loam sediment, pH 7.2, 5.0 % OC] and Swiss Lake [Derbyshire, UK, loamy sand sediment, pH 6.6, 0.7 % OC].

The systems were incubated in glass flasks containing sediment and associated water at a ratio of approximately 1:3 (v/v). Throughout the experiment the flasks were maintained in the dark at 20 ± 2 °C whilst attached to an incubation system allowing air to be bubbled through the surface water and then through a system for trapping volatile degradates. The water/sediment systems were incubated for 7 days prior to application of [14C]-test material to allow the systems to equilibrate. The redox potential of the sediment and water in control flasks was measured at regular intervals during the incubation. The pH and dissolved oxygen content of the water were also measured.

[14C]-test material was applied to the water surface of individual water sediment systems at a target application rate equivalent to an initial concentration of 0.138 mg/L in the water phase, equivalent to the direct overspray from a field application at a rate of 1 380 g/ha to a water body of 100 cm depth. The treated sediment systems were incubated for up to 98 days.

At zero time (immediately after treatment) and at intervals of 7, 14, 29, 56, 81 and 98 days after treatment duplicate flasks and their corresponding traps were removed from the incubation system. The water was carefully decanted, and the sediment was extracted once with acetonitrile, then two more times with acetonitrile: water (80:20 v/v) at ambient temperature. Extracted sediment samples were air-dried (for the early time points), ground to a fine powder and the residual radioactivity quantified by combustion. For the later time points (day 56 onwards), in order to avoid possible loss of any volatile material during drying, the sediments were not dried before combustion. The radioactivity in the water, the sediment extracts and the volatile traps was quantified by liquid scintillation counting (LSC).

Following extraction, the water phase and the sediment extracts were analysed directly by reverse phase high performance liquid chromatography (HPLC).

The overall recovery of radioactivity was good throughout the study with mean values of 96.9 % of applied radioactivity (AR) for the Calwich Abbey system and 99.7 % AR for the Swiss Lake system. Recoveries for individual flasks ranged from 93.1 to 102.0 % AR for the Calwich Abbey system and from 95.3 to 101.7 % AR for the Swiss Lake system.

In the Calwich Abbey system the levels of radioactivity in the water declined steadily from 101.3 % AR to 48.2 % AR at day 81. After this initial lag phase, however, the rate of decline from the water accelerated rapidly, reaching 4.1 % AR at day 98. The total extractable radioactivity in the sediment rose from 0.0 % AR at time zero to a maximum of 23.0 % at day 56 before declining to 6.6 % at the end of the study at day 98. The unextracted radioactivity rose from 0.0 % at time zero to 32.3 % at day 98. The presence of an initial metabolic lag phase was also evident in the evolution of CO2. Over the first 56 days only 4.9 % CO2 was released, however, this had increased to 14.8 % AR by day 81 and reached 50.1 % AR by day 98.

In the Swiss Lake system there was a slower transfer of the applied radioactivity from the water to the sediment, when compared with Calwich Abbey system, and thus resulting in a much greater percentage of applied radioactivity remaining in the water at the end of the incubation period. The radioactivity in the water declined from 101.1 % AR at time zero to 62.7 % AR by day 98. The total radioactivity in the sediment increased from 0.0 % at time zero to 21.8 % by day 98, with the extractable portion reaching a maximum of 12.8 % AR at day 56 and the unextractable reaching a maximum of 10.4 % AR at day 98. The degree of mineralisation to CO2 was less than in the Calwich Abbey system accounting for 13.5 % AR by day 98.

The chromatographic results showed that in both the Calwich Abbey and Swiss Lake systems, the applied test material partitioned into the sediment and was degraded to form minor metabolites, none exceeding >5.0 % AR.

In the total Calwich Abbey system, the applied test material declined from 101.3 % AR at time zero to 66.8 % AR at day 81. After day 81 degradation accelerated rapidly such that only 8.7 % AR remained as test material by day 98. No significant metabolites (i.e. >10 % AR, >5 % AR at consecutive sampling intervals or >5 % AR without decline at end of study) were detected. In the water phase, the applied test material declined from 101.3 % AR at time zero to 47.5 % AR at day 81 and to 3.7 % AR at day 98. In the sediment extracts, the test material reached a maximum level of 22.1 % AR at day 56 and subsequently declined to 5.0 % AR by day 98.

In the total Swiss Lake system, the applied test material declined from 101.1 % AR at time zero to 73.4 % AR at day 98. No significant metabolites (i.e. >10 % AR, >5 % AR at consecutive sampling intervals or >5 % AR without decline at end of study) were detected. In the water phase, the applied test material declined from 101.1 % AR at time zero to 62.1 % AR at day 98. In the sediment extracts, the test material reached a maximum level of 13.0 % AR at day 14 and subsequently declined slightly to 11.4 % AR by the end of the study at day 98.

The dissipation of the test material from the water phase and degradation in the total system was evaluated according to the FOCUS guidance document on degradation kinetics using the most appropriate model for the best fit to the data set. In the Calwich Abbey system, with the data exhibiting an initial lag-phase of moderate degradation followed by rapid decline, the Hockey-stick (HS) model was selected as best-fit to the data for the water phase and for the total system. In the Swiss Lake system the SFO model was selected as best-fit for the degradation in the water and for the total system.

Under aerobic conditions in the Calwich Abbey system, [14C]-test material dissipated rapidly from the water phase after an initial lag phase with a best-fit overall DT50 value of 72.5 days.

Dissipation from the water phase was slower in the Swiss Lake system with a DT50 of 171 days. The degradation in the total water/sediment systems again showed differences in the two systems, with best fit DT50 values of 83.2 and 244 days for the Calwich Abbey and Swiss Lake systems, respectively.

In conclusion, under the conditions of this study, [14C]-test material in natural water/sediment systems was shown to degrade ultimately to carbon dioxide and unextractable sediment bound residues.

Endpoint:
biodegradation in water and sediment: simulation testing, other
Remarks:
biodegradation in water and sediment
Type of information:
experimental study
Adequacy of study:
supporting study
Study period:
June 1990 to August 1991
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
comparable to guideline study with acceptable restrictions
Qualifier:
according to guideline
Guideline:
EPA Subdivision N Pesticide Guideline 162-4 (Aerobic Aquatic Metabolism)
Deviations:
no
GLP compliance:
yes
Radiolabelling:
yes
Oxygen conditions:
aerobic
Inoculum or test system:
natural water / sediment: freshwater
Details on source and properties of surface water:
- Details on collection: Two different water-sediment systems (soil and water each) were collected from the area north-west and north of Hamburg on June 27, 1990.
stream "Grosser Graben", near Krempe/Elmshorn in a marsh area surrounded by arable land and pastures; in this report called "Krempe".
stream "Ohlau", near Lentforden surrounded by pastures; in this report called "Ohlau".
- Storage length: Water parameters were determined within 23 h after collecting the samples, after the sediment had settled.
- pH at time of collection: Krempe: 7.6; Ohlau 7.7
- Oxygen concentration (mg/l) initial/final: Krempe: 7.5 mg/L initial; Ohlau 5.3mg/L initial.
Details on source and properties of sediment:
- Details on collection: Two different water-sediment systems (soil and water each) were collected from the area north-west and north of Hamburg on June 27, 1990.
stream "Grosser Graben", near Krempe/Elmshorn in a marsh area surrounded by arable land and pastures; in this report called "Krempe".
stream "Ohlau", near Lentforden surrounded by pastures; in this report called "Ohlau".
- Textural classification: Krempe: loamy-sandy; Ohlau: Sandy
- pH at time of collection: Krempe: 6.2; Ohlau 6.9; determined in 0.01 m CaCl 2 solution.
- Organic carbon (%): Krempe: 1.85; Ohlau: 0.25
- CEC (meq/100 g): Krempe: 8; Ohlau: 18
- Particle size Krempe:
< 0.002 mm: 1.2
0.002 - 0.020 mm: 1.0
0.006 - 0.060 mm: 1.0
0.020 - 0.060 mm: 6.1
0.060 - 0.200 mm: 41.0
0.200 - 0.600 mm: 46.9
0.600 - 2.0 mm: 2.8
- Particle size Ohlau:
< 0.002 mm: 3.0
0.002 - 0.020 mm: 1.0
0.006 - 0.060 mm: 2.8
0.020 - 0.060 mm: 6.9
0.060 - 0.200 mm: 35.4
0.200 - 0.600 mm: 49.1
0.600 - 2.0 mm: 1.8
Duration of test (contact time):
3 mo
Initial conc.:
1.88 other: mg/kg
Based on:
test mat.
Parameter followed for biodegradation estimation:
radiochem. meas.
Details on study design:
Radioactivity measurements
Determination of radioactivity in liquid samples
Aliquots of the liquid samples (solutions) in amounts up to 1 mL were weighed into glass scintillation vials and scintillation cocktail (Hionic Fluor (R) of Packard Instrument GmbH, Frankfurt, FRG) was added. In general, threefold samples were prepared from each solution.
The radioactivity was then determined in a liquid scintillation spectrometer, model Tri-Carb 4530 of Packard Instrument GmbH according to a standard operation procedure.
From the dpm values obtained the specific and total radioactivities were calculated under consideration of the aliquot used for the measurements

Calculation of specific radioactivities and equivalents of the test material
The basis for conversion of dpm’s into kBq or Bq was: 1 kBq ~ 60 000 dpm or 1 Bq = 60 dpm respectively.

Specific radioactivities in liquids and solids were calculated from the following expression:

Specific radioactivity (Bq/g) = dpm / (60 x m)

where m is the weight (in g) of the aliquot of liquid or solid used for the radioactivity determination.
From the individual values the mean value and the standard deviation were calculated.
The total radioactivity (At) contained in solution or solid was calculated from the specific radioactivities by multiplication with the respective weights. This was related to the radioactivity originally applied (Aa) to the test system resulting in the percentage of radioactivity contained in the respective solution or solid.

Percentage = (At [kBq] / Aa [kBq]) · 100

Combustion of dried sediment samples
Aliquots of the sediment samples were combusted after extraction and drying in a Tri-Carb sample oxidiser model 306 of Packard Instrument GmbH, Frankfurt, FRG according to a standard operation procedure. In general, five parallel samples were combusted per sample. The radioactivity in the resulting samples was determined by liquid scintillation counting.

Microbiological investigations
The total germ count was determined according to Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung, mikrobiologische Verfahren (Gruppe K) DIN 38411, Teil 5) method K 5, omitting the membrane filtration.

Degradation of the test material
Preparation of the solution of [14C]-test material to be applied to the test system
Unlabelled test material and [14C]-test material were dissolved in methanol, resulting in a concentration of 2.2 mg (14 C-labelled)-test material per mL.

Test procedure
For investigation of the degradation of the test material the general procedure was followed as outlined in our drafted standard operation procedure (SOP).

Test device
The incubation flask (volume ca. 1.3 L) is equipped with a device for slow stirring of the water only, a trap for volatile components and a tube for introducing air. The trap, containing quarz wool (being impregnated with paraffin oil) and soda lime, is permeable to air. Slow stirring of the water is accomplished by placing the flask on a shaking cupboard with revolving movement. Thus, aerobic conditions are achieved in the water. Dissolved oxygen, temperature and pH are measured at each sampling date.

Preparation of the flasks
A total of 12 flasks was prepared per system, each containing 55 g (dry weight) soil and 550 mL water. They were placed on the shaking cupboard and allowed to equilibrate for 2 weeks before the 14C-labelled test material was applied.

Application of the test material
0.5 mL each of the methanol solution of [14C]-test material were applied to 10 of the 12 flasks prepared. For determination of the amount of test material applied to the flasks, three replicate applications were done in the same way into methanol. The radioactivity in these solutions was determined via LSC. Thus 1.14 mg or 905.6 kBq of the material were present in the flask, corresponding to a concentration of 1.88 mg per kg system. No test material was applied to both flasks allocated to monitoring of the biological system.

Sampling and extraction procedure
At each sampling 50 mL water were withdrawn from the system for determination of O2 content, pH and temperature. Then volatile compounds were driven from the headspace of the flask into the trap by a stream of air (10 min, 10 mL/sec); subsequently the aqueous layer in the flask was set to approximately pH 10 with aqueous 1 N KOH and the contents of the flasks was transferred into centrifuge vials; the portion of water previously withdrawn for determination of the water parameters was used for rinsing and thus added to the system again.

After separation of the layers by centrifugation an aliquot of the aqueous phase was acidified (pH 1) with aqueous 1N HCl and extracted with dichloromethane (5 times, 30 mL each). After determination of the radioactivity in the combined solution it was investigated by means of radio-TLC.
Another aliquot of the aqueous layer was acidified in the same way and in a closed flask, heated to 70 °C
for 30 min during which time a slight stream of nitrogen was passed through the solution and trapped in an absorption solution for CO2. The radioactivity in this solution was determined.
The sediment was extracted with methanol/water (1:1) followed by methanol. After determination of the radioacitivity the solutions were combined and investigated via radio-TLC.
After drying, the radioactivity retained in the sediment was determined by means of combustion/LSC.

Carbonate in extracted sediment samples
For determination of the amounts of 14C-labelled carbonate, an aliquot of the sediment was suspended in aqueous HCl (pH 1) in a closed system and during stirring and heating (70 °C) a slight stream of nitrogen was driven over this solution and passed through an absorption liquid for carbon dioxide. The radioactivity in this solution was determined.

Investigation of extracted sediment samples
An aliquot (ca. 5 g) of the dry sediment was extracted three times with 20 mL 0.1 N NaOH each with ultrasonication and washed with water. After determination of the radioactivity in the alkaline solution this was acidified with 1 N HCl to pH 1.
The precipitate formed was separated by centrifugation and washed with 0.1 N HCl. The radioactivity was determined in the acidic solution, in the precipitate (after dissolution in 0.1 N NaOH) and in the dried sediment.
Reference substance:
other: 2-Methyl-4-chlorophenol
Compartment:
sediment
% Recovery:
11
Remarks on result:
other: Krempe
Compartment:
sediment
% Recovery:
8
Remarks on result:
other: Ohlau
Compartment:
entire system
% Recovery:
0.4
Remarks on result:
other: Parent compound
Compartment:
other: CO2
% Recovery:
80
Compartment:
natural water / sediment: freshwater
DT50:
21 d
Type:
other: Degradation curve
Other kinetic parameters:
other: DT90: 28 days
Transformation products:
not specified
Evaporation of parent compound:
yes
Volatile metabolites:
yes
Residues:
not specified
Details on results:
The degradation of the test material in two different water/ sediment systems was investigated under aerobic conditions using radioanalytical methods. The sediments employed, named "Krempe" and “Ohlau", were a loamy-sandy and a sandy in character, respectively.
1.14 mg [ring-u-14C)-test material dissolved in 0.5 mL methanol were applied to each flask containing 55 g (dry weight) sediment and 550 mL water, being equivalent to an application rate of 1.88 mg per kg system. According to TLC-investigation in two systems the radiochemical purity of the test material applied was 87 to 88 %. This figure was got by a real worst-case estimation, when every count was considered. By this the impurities were split into 15 separate peaks, the most important of which accounted for no more than 2 - 3 % each.
The study was conducted at 22 ± 2 °c in the dark over 92 days with 7 samplings at 0, 3, 7, 14, 30, 60 and 91/92 days after application of the test material (the last sampling for the system Krempe was done 91 dafter application, whereas this was done for the system Ohlau after 92 d).

Water parameters
Temperature, pH and O2-concentration were determined in the flask at each sampling date. For the system Ohlau, a slight decrease in pH was observed from day 30 to day 92. The o2-concentration started between 6 and 7 mg/L, after a slight increase at day 3 it dropped to 3.8 mg/L in the system Krempe at day 7 and subsequently recovered to more than 8 mg/L. This decrease in oxygen concentration was paralleled by some turbidity in the systems which was observed starting after ca. day 7 and reaching the maximum at ca. day 10; subsequently it slowly disappeared. This turbidity was not observed in the sample for the control of the biological system, to which no test material had been applied.
However, the data showed that there were always aerobic conditions in the water phases.
The total germ number in the water was determined within 20 h after collecting the system from the respective source and during the experiment on a water/sediment sample which was contained in a degradation flask but which contained no test material. The total germ counts after collecting the system were approximately 7 x 10^3 in the Krempe water and approximately 3 x 10^5 in the Ohlau water. The counts decreased to ca. 20 00 (Krempe) or ca. 3 600 (Ohlau) respectively within ca. two weeks until application and further decreased slightly until the end of the experiment.
During development of the turbidity in the water the total germ count was determined in a Krempe sample to which 14C-labelled test material had been applied. On day 10 after application the germ count was 5.9 x 10^6, on day 18 it was 5.8 x 10^4 . This finding indicates that the microorganisms involved in the breakdown of the herbicide used the solvent (methanol) or the active ingredient as a carbon source.

Radioactivity in the different parts of the system
The data show that in both systems Krempe and Ohlau between day 3 and day 14 after application of the substance the distribution of radioactivity in the different parts was found to be rather constant: 81 to 85 % of the radioactivity applied were found in the water, 14 to 18 % in the sediment-and at most 4 % in the volatiles. Between day 14 and day 91/92 (Krempe/Ohlau) the percentage in the volatiles, being carbon dioxide increased rapidly to 83 % (Krempe) or to 80 % (Ohlau) whereas the amount in the water decreased to 3 % or less. The amount in the sediment increased to 19 % (Krempe) or to 29 % (Ohlau). But whereas in the beginning most of the radioactivity was extractable from the sediment into methanol/water, at the end most of it was retained.

Identity of the radioactivity
As far as significant amounts of radioactivity were found in the water (until day 14 after application) most of this was extractable into dichloromethane. For TLC investigation of these latter extract solutions different TLC systems were examined. Radio-TLC investigation of the dichloromethane extracts mentioned revealed that almost all of the radioactive residue consisted of parent [14C]-test material. Some more parent compound could be extracted from the sediment and identified via radio-TLC as well. Thus, the total percentage of parent compound found in both systems until day 14 was more than 83 %, related to the amount applied. After day 14 the amount of parent compound decreased rapidly to 0.4 % or less.
No significant amounts of 2-methyl-4-chloro-[14C]-phenol as potential metabolite could be identified.
After day 14 up to 83 % of the radioactivity applied were found directly in volatile compounds which could be trapped with soda lime and thus were assigned as carbon dioxide. Minor amounts of carbon dioxide or carbonate were additionally found in the water or in the sediment respectively.
From the data it is concluded that at day 30, when significant amounts of carbon dioxide had been formed, slight amounts of volatile radioactivity - probably carbon dioxide - were lost during manipulation of water until determination of the CO2. For this reason, from day 60 the aqueous layer was set slightly alkaline after sweeping the volatile compounds into the trap and before opening the degradation flask. This resulted in an improved recovery of radioactivity at least for the system Krempe.

Radioactivity retained in the sediment
After extraction of the sediment with methanol in the samples of day 30 to day 91/92 radioactivity was retained therein in amounts of 11 to 26 %. For characterising this radioactivity the sediment samples were extracted with aqueous sodium hydroxide. This extract was subsequently acidified. It could be demonstrated that roughly half of the radioactivity from the sediment was extractable into aqueous sodium hydroxide being equivalent to 7 to 12 % (Krempe) or to 7 to 15 % (Ohlau) related to the amount applied. For the samples "day 91/92" approximately 4 or 8 % respectively of the radioactivity, to be considered as humic acids, were precipitated by hydrochloric acid (pH 1) whereas 3 to 4 % (fulvic acids) were soluble in the acidic aqueous solution.

TLC Systems and Rf-Values

No.

Solvent*

Rf Value

Test Material

2-methyl-4-chlorophenol

A

Chloroform cyclohexane acetic-acid= 80 20 10

0.7

0.7

B

Benzene dioxan acetic acid= 90 25 4

0.5

0.8

F

Acetonitrile water aq. ammonia (25 %) = 80 18 2

0.5

0.9

The adsorbent was silica gel Merck No. 5715 in all cases.

* All values are v/v without chamber saturation.

 

Data on the Water During the Experiment

Period after

application

[d]

Krempe

Ohlau

pH

O2 Conc.**

[mg/L]

Temp

[°C]

pH

O2 Conc.

[mg/L]

Temp

[°C]

0

7.4

6.6

21.5

6.6

7.0

21.5

3

7.7

7.7

21.7

6.6

7.1

21.7

7

7.3

3.8

21.8

6.9

7.3

21.8

14

7.9

7.

21.9

7.0

8.1

21.9

30

7.7

8.0

21.2

6.4

8.7

21.2

60

7.3

8.3

22.1

5.8

8.5

21.1

91/ 92*

7.6

8.5

23.1

6.1

8.2

23.5

*The last sampling was on day 91 for the systems Krempe and on day 92 for the system Ohlau.

**At 22 °Cthe saturation is equivalent to 9 mg o2 per L.

 

Microbiological Data on the Water in the Control Systems

Day After Application

Total Germ [count/mL]**

Krempe

Ohlau

-19*

7300

240 · 10^3

7500

360 · 10^3

0

920

5100

440

2200

540

 

30

240

1140

1510

910

2400

 

60

780

2190

87

 

91

183

158

204

208

380

 

* Day of collecting the water-sediment samples from the respective sources

** The individual results of replicate sample are shown.

 

Radioactivity Distribution and Balance During the Degradation of the Test Material

Test Period

[d]

Amount of Radioactivity [%]*

Sediment

Water

Volatiles

(Balance)

CO2

Others

System: Krempe

0

6.2

99.2

ND

ND

105.4

3

13.6

85.1

0.3

0.4

99.3

7

16.3

83.9

1.0

0.5

101.7

14

16.5

83.0

1.5

0.4

101.4

30

26.5

16.2

47.9

0.2

90.8

60

19.9

3.3

76.1

< 0.1

99.3

91

18.7

1.6

82.9

< 0.1

13.2

System: Ohlau

0

5.7

98.5

ND

ND

104.2

3

14.0

85.3

1.0

0.4

100.7

7

14.5

84.7

1.8

0.3

101.3

14

17.5

81.3

3.1

0.4

102.3

30

28.0

6.6

58.0

0.3

92.9

60

16.4

6.2

67.9

0.1

90.6

92

28.9

3.0

79.7

< 0.1

111.6

*Percentage of radioactivity related to the amount originally applied to the system

ND:Not determined

 

Amounts of Parent Compound During the Degradation of the Test Material

Test Period

[d]

Amount of Parent Compound [%]* in Extracts

Sediment

Water

System: Krempe

0

5.2

92.3

97.5

3

9.0

75.1

84.1

7

11.0

81.0

92.0

14

10.5

75.7

86.2

30

0.6

0.6

1.2

60

0.2

0.4

0.6

91

< 0.1

< 0.4

< 0.4

System: Ohlau

0

5.1

94.5

99.9

3

9.1

71.7

80.8

7

7.8

79.7

87.5

14

8.6

74.3

82.9

30

1.1

0.7

1.8

60

0.2

< 0.3

< 0.5

92

< 0.1

< 0.3

< 0.5

*Percentage of radioactivity found at the Rf value for parent compound, related to the amount originally applied to the system

 

Comprehensive Table for the Degradation of Test Material

Test Period

[d]

Amount of Radioactivity [%]* Found

Parent Compound

Degradates**

Unextract.

[Sediment]

Polar

[Water]

CO2

[Balance]

System: Krempe

0

97.5

5.4

0.7

2.0

ND

105.6

3

84.1

2.9

3.2

2.0

0.6

92.8

7

92.0

3.2

4.0

2.1

1.1

102.4

14

86.2

2.3

4.8

2.9

2.3

98.4

30

1.2

4.7

22.2

2.0

61.0

91.1

60

0.6

2.9

17.4

1.2

77.2

99.3

91

< 0.4

< 2.6

16.9

0.6

84.2

101.7†

System: Ohlau

0

99.9

2.9

0.4

1.8

ND

105.0

3

80.8

7.6

2.9

2.3

1.0

94.6

7

87.5

6.5

4.8

2.2

1.8

102.8

14

82.9

2.5

3.6

2.6

3.2

98.8

30

1.8

5.0

23.4

3.1

58.2

91.5

60

< 0.5‡

< 5.4‡

10.7

4.7

68.2

89.5

92

< 0.3

< 3.2

25.9

2.3

79.9

108.1⁑

*Percentage of radioactivity related to the amount originally applied to the system = 100%

** Consists of 1 to 5 peaks in radio-TLC

ND: Not determined

This figure of 101.7%is the sum of unextractable, polar and CO2 only.

The extract from the sediment contained 0.2% parent compound and 5.4% degradates, whereas the total amount of radioactivity in the extract from the water was 0.3 only which was not investigated via radio-TLC.

⁑ This figure of 108.1% is the sum of unextractable, polar and co2 only.

Validity criteria fulfilled:
not specified
Conclusions:
Under the conditions of the study the degradation of the test material proceeded rather rapidly in both systems. The main degradation product was carbon dioxide, which after three months comprised more than 80 % of the radioactivity originally applied to the system. Within this time the amount of parent compound decreased to less than 0.4 %. At the end of the test period three months after application approximately 4 to 8 % of the radioactivity were found in humic acids in the systems, another 3 to 4 % in fulvic acids. Ca. 11 % were retained in one sediment, ca. 8 % only in the other.
A disappearance time (DT) 50 of 20 to 21 days was found, the DT 90 was 28 days. This low DT 90 value may be explained by the exponential growth of the microbial degraders after a lack phase in the beginning of the experiment.
Executive summary:

The degradation of the test material in sediment/water was assessed according to Pesticide Assessment Guidelines Subdivision N, § 162-4 and in compliance with GLP.

The degradation of the test substance [14 C]-test material, being labelled in the aromatic ring was investigated in two different water/sediment systems under aerobic conditions over a period of three months at 22 °C. 1.88 mg test material was applied per kg of the system.

After an incubation time of approximately two weeks, the degradation of the test material proceeded rather rapidly in both systems. The main degradation product was carbon dioxide, which after three months comprised more than 80 % of the radioactivity originally applied to the system. Within this time the amount of parent compound decreased to less than 0.4 %.

At the end of the test period three months after application approximately 4 to 8 % of the radioactivity were found in humic acids in the systems, another 3 to 4 % in fulvic acids. Ca. 11 % were retained in one sediment, ca. 8 % only in the other.

For both systems a disappearance time (DT) 50 of 20 to 21 days was found, the DT 90 was 28 days. This low DT 90 value may be explained by the exponential growth of the microbial degraders after a lack phase in the beginning of the experiment.

Endpoint:
biodegradation in water: sediment simulation testing
Type of information:
experimental study
Adequacy of study:
supporting study
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
guideline study with acceptable restrictions
Qualifier:
according to guideline
Guideline:
other: BBA, Part IV, Section 5.1
Version / remarks:
1990
Deviations:
no
GLP compliance:
yes
Radiolabelling:
yes
Oxygen conditions:
aerobic
Inoculum or test system:
natural water / sediment: freshwater
Details on source and properties of surface water:
- Details on collection: Samples of sediment and water were collected from two sites, a stream located on Aldhams Farm, Manningtree, Essex U.K. (system 96/03 (Manningtree)) and from the River Roding, Boarded Barns Farm, Ongar, Essex U.K. (system 96/04 (Ongar)). At collection the water was transferred to 60 L plastic drums.
- Temperature (°C) at time of collection: Immediately prior to sample collection the water temperature (measured just below the water surface):
Manningtree: 9.2 °C
Ongar: 7.6 °C
- pH at time of collection: Immediately prior to sample collection the water temperature (measured just below the water surface):
Manningtree: 5.57
Ongar: 6.94
- Redox potential (mv) initial/final: Immediately prior to sample collection the water temperature (measured just below the water surface):
Manningtree: 248 mV
Ongar: 225 mV
- Oxygen concentration:
Oxygen content sediment level:
Manningtree: 61 %
Ongar: 80 %
Oxygen content 5 cm above sediment:
Manningtree: 63 %
Ongar: 82 %
- Hardness (CaCO3):
Manningtree: 319 ppm
Ongar: 413 ppm
- Dissolved organic carbon:
Manningtree: 18.59 ppm
Ongar: 59.46 ppm
- Total Nitrogen:
Manningtree: 24.9 mg/L
Ongar: 20.2 mgL
- Total Phosphorous:
Manningtree: 0.68 mg/L
Ongar: 0.24 mg/L
Details on source and properties of sediment:
- Details on collection: Samples of sediment and water were collected from two sites, a stream located on Aldhams Farm, Manningtree, Essex U.K. (system 96/03 (Manningtree)) and from the River Roding, Boarded Barns Farm, Ongar, Essex U.K. (system 96/04 (Ongar)). The sediment was collected using a spade.
- Storage conditions: Transferred to large plastic bags and maintained in the dark at 4 °C prior to use.
- Textural classification (i.e. %sand/silt/clay):
Manningtree: Sandy silt loam (BBA); Loam (USDA)
Ongar: Clay loam (BBA and USDA)
- pH at time of collection:
Manningtree: 6.7 (H2O); 6.3 (KCl)
Ongar: 8.6 (H2O); 7.9 (KCl)
- Organic carbon (%):
Manningtree: 5.3 %
Ongar: 3.1 %
- CEC (meq/100 g):
Manningtree: 11.9
Ongar: 63.2
- Biomass:
Biomass at the start of the study:
Manningtree: 349 μg C g^-1
Ongar: 219 μg C g^-1
Biomass at the end of the study:
Manningtree: 390 μg C g^-1
Ongar: 222 μg C g^-1
- Sediment samples sieved: Yes. The sediments were sieved to 5 mm at the collection site.
- Total nitrogen:
Manningtree: 3836.4 mg/kg
Ongar: 1988.1 mg/kg
- Total phosphorous:
Manningtree: 984.3 mg/kg
Ongar: 1043.1 mg/kg
Duration of test (contact time):
100 d
Initial conc.:
992 mg/L
Based on:
test mat.
Parameter followed for biodegradation estimation:
radiochem. meas.
Details on study design:
TEST CONDITIONS
- Volume of test solution/treatment: An acetonitrile solution containing 0.992 mg mL-1 of [14C]-test material, equivalent to 1.123 kg ai ha^-1 was dispensed in aliquots of 100 μL (0.099 mg) dropwise, using a Gilson microman positive displacement pipette
(Anachem, Luton, U.K.), on a single occasion, into the water (approx. 300 mL) of each flask. The quantity of test material per flask was 99.2 μg compared to the theoretical value of 132.54 μg.
- Test temperature: At 20 °C ± 2 °C.
- Aeration of dilution water: Throughout the acclimatisation and experimental period a continuous flow of moistened, carbon dioxide-tree air was passed through the water, at a rate sufficient to allow aeration and gentle movement and at the same time avoiding mixing of the two phases. The passage of air was controlled by the use of flow restrictors, these consisted of lengths of glass with a capillary along their whole length. This narrow hole allows sufficient air to pass into each flask and for flask to be disconnected without disrupting the flow through others.
- Continuous darkness: Yes. The aquatic incubation units were maintained in the dark.

TEST SYSTEM
- Culturing apparatus: The sediments were sieved to 2 mm and added to individual glass flasks (ca. 7.5 cm internal diameter) to a depth of 2 - 2.5 cm; 0.2 mm filtered water was added to an approximate depth of 5 cm above the sediment. This level was maintained throughout the study by addition of de-ionised water, as necessary. The experiment was initiated after pre-incubation of the incubation flasks for approximately 6 weeks in the dark at 20 °C ± 2 °C, to enable acclimatisation of the systems with respect to oxygen content, pH, redox potential and complete phase separation.
- Details of trap for CO2 and volatile organics if used: Moistened carbon dioxide-free air was passed through the water in each unit and the effluent air current evolved passed through an ethylene glycol and two 2M potassium hydroxide traps to trap volatiles and 14CO2 respectively.

SAMPLING
- Sampling frequency: Duplicate units of each sediment type were removed for analysis at intervals, zero hours (immediately after dose application), 24 and 48 hours, 7, 14, 22, 30, 61 and 100 days after application. An extra sampling point at 80 days was added for the Manningtree system (due to poor replication of samples) The redox potential of the sediment and water, and the oxygen content and pH of the water, were measured in equivalent units prior to analysis.
- Sampling method used per analysis type:
Water phase: As much as possible of the water phase was decanted from the sediment and the sediment quantitatively transferred to a 250 mL plastic bottle using HPLC grade water. This was then centrifuged at 3 000 rpm (1 860g) for 10 mins and the aqueous fraction decanted and added to the initial water. The total volume was measured and the radioactivity it contained was determined from aliquots (250 μL - 1.0 mL) taken for liquid scintillation counting (LSC). The samples were analysed by HPLC (direct injection) and TLC.
Sediment: Once transferred to the 250 mL plastic bottle.
Step 1: Allow sediment to air dry at room temperature and combust (after combustion if > 5 % dose remaining, then rehydrate with water overnight and follow the procedure below).
Supplementary Extraction (after step 1 above) (Later Sampling Times)
Step 1: Centrifuge and decant off water.
Step 2: Add acetonitrile (150 mL), shake for 30 mins and centrifuge at 3 000rpm for 10 mins.
Step 3: Repeat step 2, until procedure has been repeated twice.
Step 4: Combine and count extracts.
Step 5: If > 5 % dose remains unextracted repeat procedure using an ultrasonic probe (High intensity ultrasonic processor, 375 watt model, Jencons, U.K.) in place of shaking.
Step 6: If > 5 % dose remains repeat procedure using soxhlet extraction (perform only once) in place of shaking.
Step 7: If > 5 % dose still remains repeat steps 1 - 4 using acetonitrile: water:f ormic:acid (80: 20: 0.1 M) (v/v/w) (perform once only).
Step 8: If > 5 % dose still remains repeat step 7 using 0.5 M aqueous formic acid in water (perform once only).
Not all sediments were extracted to steps 7 & 8.
Procedural recoveries were monitored at all stages of analysis and were generally greater than 85 %. Selected concentrated extracts were analysed for the test material and degradation products by high performance liquid chromatography and for confirmation of parent, in a second contrasting system, TLC
- Characterisation of Unextracted Residues: For selected samples, following exhaustive extraction with organic and organic/aqueous solvents, the distribution of the radioactivity between the humin, fulvic and humic acid fractions of the sediment was determined for two samples from each aquatic sediment system.
Two, ~20 g portions of air dried sediment for each system were shaken on a wrist action shaker for 24 hours with 2 5mL of 0.01 M calcium chloride. Following centrifugation at 3 000rpm (1 860g) for 10 minutes, the supernatant was removed, the volume
recorded, and aliquots removed for LSC. The pellet was resuspended in 45 mL of 0.5 M sodium hydroxide solution and shaken for a further 24 hours. Following centrifugation at 3 000rpm (1 860g) for 10 minutes, the supernatant was removed, and the sediment rinsed with two 20 mL portions of 0.5 M sodium hydroxide. The sediment residue was then rinsed with three 10 mL portions of deionised water. The extracts and washings were combined, the volume measured and aliquots removed for assay by LSC.
The solution was acidified to about pH 1 with 6 M hydrochloric acid and re-centrifuged as above. The supernatant was removed and aliquots (3 x 250 μL) taken for LSC. Approximately 25 mL of 0.1 M sodium hydroxide was added to the residue. The volume of the supernatant was measured and (3 x 250 μL) aliquots removed for LSC.

ANALYSIS OF EVOLVED RADIOLABELLED PRODUCTS:
- Radioactivity in the trapping solutions was quantified by LSC at each sampling time and at other times as required. A day 30 and a day 61 trap solution (from system 96/04 (Ongar)) were used to establish that the [14C] in the volatile traps was carbon dioxide. The addition of excess 1 M BaCl2 and 1 M Na2CO3 (present to help 'seed' the reaction) resulted in the precipitation of insoluble Ba 14CO3, thus removing the 14CO2 from solution. Therefore the difference between the radioactivity content of the supernatant before and after the addition of the BaCI2 was equivalent to the amount of 14CO2 that was present in solution.
- combustion of sediment residues: Following extraction the sediment residues were air dried, weighed and ground in a Labtechnics LM1-P pulverising mill (Glen Creston Ltd., Stanmore, Middlesex, U.K.). Triplicate (occasionally quadruplicates) aliquots approximately 0.2 g, accurately weighed, were thoroughly mixed with approximately equal volumes of cellulose then combusted using a Packard Model 387 Oxidiser. The combustion products were absorbed in Carbosorb-E, mixed with Permafluor E+ and the radioactivity measured by LSC.
- Liquid Scintillation Counting: All samples were given unique sequential vial numbers. The samples were counted using a Kontron Betamatic V or Beckman LS6500 liquid scintillation counter, the degree of quench being determined by an external standard channels ratio technique. Calibration curves, which were checked at regular intervals, were stored in the computer as second order polynomials and used to determine counting efficiency. Thus, unknown samples were counted both in the presence and absence of an external standard and, using the stored curve coefficients, the counting efficiency was computed. Following the determination of counting efficiency and subtraction of the background value, the radiocarbon content (DPM) of the sample was automatically calculated.

CALCULATIONS
- Calculation of Total Activity in Liquid Extracts: The total activity in any liquid sample was calculated as follows.

Total DPM = (DPM in aliquot / volu,e counted (mL)) x volume of sample (mL)

- Calculation of Total Unextracted Activity: The total unextracted activity was calculated from the combustion results as follows:

Total DPM = (DPM in aliquot / weight combusted (g)) x air dry weight of extracted soil (g)

- Calculation of % dose in sample: The % dose present in any sample was calculated as follows:

% dose = (DPM in sample extract / DPM in total dose) x 100 %
Compartment:
natural sediment
% Recovery:
110
Remarks on result:
other: Aquatic sediment 96/03
Compartment:
natural sediment
% Recovery:
106.2
Remarks on result:
other: Aquatic sediment 96/04
Compartment:
natural water
DT50:
49.23 d
Type:
not specified
Temp.:
20 °C
Remarks on result:
other: System 96/03
Compartment:
entire system
DT50:
60.66 d
Type:
not specified
Temp.:
20 °C
Remarks on result:
other: System 96/03
Compartment:
natural water
DT50:
24.25 d
Type:
not specified
Temp.:
20 °C
Remarks on result:
other: System 96/04
Compartment:
entire system
DT50:
22.97 d
Type:
not specified
Temp.:
20 °C
Remarks on result:
other: System 96/04
Transformation products:
not measured
Evaporation of parent compound:
not measured
Volatile metabolites:
yes
Residues:
not measured
Details on results:
Good recoveries were achieved for both aquatic sediment systems, with an overall mean of 110.0 % (range 101.5 - 115.4 %) for the Manningtree system and 106.2 % (range 99.1 % - 114.7 %), for the Ongar system, except for one replicate at 66.9 % which was due to poor 14CO2 trapping efficiency. The aquatic sediment systems described in this report, consisted of oxygenated water and a highly reductive environment, (mV readings -354 mv to 36.4 mV for the sediments) at and below the sediment surface. The biomass figures for both systems were measured before and after incubation and showed the sediments were both active biologically.

DISTRIBUTION OF RADIOACTIVITY BETWEEN WATER AND SEDIMENT
There was a steady transfer of radioactivity from the water to the sediment of both systems. The rate of mineralisation to carbon dioxide was at a similar rate in both systems commencing at 24 h and reaching -3 % (Manningtree) and -2 % (Ongar) after 30 and 22 days respectively, after this time the rate of mineralisation increased in both systems, and reached a maximum, at 100 days, of ~56 % (range 42.63 - 73.43 %).
This lag phase, followed by a rapid decline of parent material is typical of the phenoxy type herbicides and has been shown in previous studies. The initial lag phase where degradation is relatively slow shows an 'adaption' of the system to the presence of the active ingredient. Once the system has adapted then degradation is swift and almost complete mineralisation to 14CO2 occurs. The lack of major intermediates underlines the speed of this metabolism. This pattern was less marked in the Manningtree (96/03) system but was still evident.
At day 100 the two sediments showed slight differences. 15 % of the applied radioactivity was in the water for system Manningtree (96/03) and 2 % for the Ongar system (96/04). After 100 days extractable residues were 5 % (96/03, Manningtree) and 3 % (96/04, Ongar). The increase of unextractable residues in the sediments was gradual over time and at a similar rate in both systems reaching maxima of 28 % (96/03, Manningtree) and 24 % (96/04, Ongar) at 100 days. As the amount of 14CO2 increases the bound residue levels also increase which may indicate reincorporation of 14CO2 into organisms within the sediment and the radioactivity seen in combustions would be derived from these carbon fragments. Therefore, their isolation and identification is likely to be very difficult.
Fractionation into fulvic acid, humic acid and humin showed that the radioactivity was mainly associated with the humin for both systems (up to 38 % of applied radioactivity) with some associated with the humic acid (up to 3 % of applied radioactivity) and some associated with the fulvic acid (up to 5.4 % of applied).

COMPOSITION OF RADIOACTIVITY IN THE AQUATIC SYSTEMS
The test material remained the main component of all chromatographic samples from both water and sediment. The test material was identified as the main extractable component of the sediment. The amounts of the test material in the extracts reached a maximum at 14 days for both sediment systems.
The radioactivity in the 2M KOH traps was characterised as carbon dioxide by barium carbonate precipitation and was the only metabolite to exceed 10 %, reaching means of 55 % (range 78.91 % - 31.61 % 96/03 (Manningtree)) and 57 % (range 73.43 - 42.63 % 96/04 (Ongar)) at 100 days. Other metabolites were individually less than 5 % in the extracts. One exception was Ongar samples at 30 days, in which metabolite 1 totalled 10 % in the complete system. This metabolite seemed transient and had disappeared completely at 100 days. Attempts were made to identify this metabolite by mass spectroscopy but these were unsuccessful.

DEGRADATION OF THE TEST MATERIAL WITHIN THE AQUATIC SEDINEMNT SYSTEMS
Using data derived from HPLC of the sample extracts the DT50 and DT90 for the water phase and for the complete aquatic system were calculated using different mathematical models.
The Timme-Frehse (V. 2.0) (Bayer AG) program can fit the data to a number of first and second order decay curves and display the 'best-fit' results. When the data was analysed using this programme the best fit was found to be first order.
Kim V.1.0 (Schering AG.) was also used as a model. In this case the best model was not first order, but a three compartment decay model with linear slow release.
The ideal set of results would have a fit closest to 1.0 (-1.0 in KIM) in any of the programmes, although they are statistically different measurements. Using this criterion KIM gave the results that are quoted in the summary of this report, since this software produces the graphs with the best fit criteria. The DT50 for the water of both systems is different, approximately 49 days for 96/03 (Manningtree) and 23 days for 96/04 (Ongar) The main differences occur in the DT50 of the total system, where system 96/03 (Manningtree) shows a less rapid DT50 than system 96/04 (Ongar). The onset of decline of the test material in the extracts of sediment from system 96/03 (Manningtree) is at day 61 rather than at day 30 for the sediment of system 96/04 (Ongar) The DT90 values are figures extrapolated by the software used and in reality their actual values may be different to these figures.

Mean Recovery

 

Mean Recovery (% of Applied Radioactivity)

Time

0 h

24 h

48 h

7 Days

14 Days

22 Days

30 Days

61 Days

80 Days

100 Days

Overall Mean

Aquatic sediment 96/03

112.1

114.0

112.3

114.0

112.6

108.5

109.9

108.8

103.7

87.4

110.0

Aquatic sediment 96/04

112.0

112.6

109.9

10.4

110.1

105.8

101.6

106.7

n/a

107.9

106.2

 

Partition of Radioactivity Between Water and Sediment

 

Partition of Radioactivity Between Water and Sediment as % Applied Radioactivity

Time

 

0 h

24 h

48 h

7 Days

14 Days

22 Days

30 Days

61 Days

80 Days

100 Days

Aquatic sediment 96/03

Water

111.29

110.03

106.38

105.74

92.40

94.65

88.32

42.73

42.98

15.43

Sediment

0.77

3.93

5.74

7.45

18.31

11.59

18.61

32.40

37.74

33.47

Aquatic sediment 96/04

Water

109.16

108.14

105.29

101.73

97.39

92.14

11.37

7.04

n/a

1.80

Sediment

2.8

4.37

4.46

7.57

10.33

10.69

32.43

44.41

n/a

27.68

Degradation of the Test Material Within the Aquatic Sediment Systems

Water

96/03 Manningtree

96/04 Ongar

Model

DT50

(Days)

DT90

(Days)

Fit

Model

DT50

(Days)

DT90

(Days)

Fit

Excel

First order fit

39.6

131.63

0.916

First order fit

12.19

40.49

0.760

Timme Frehse

(best fit)

First order fit

39.63

NC

0.915

First order fit

NC

NC

0.210

Kim

Three compartment decay with linear slow release

49.23

155.47

-0.989

Three compartment decay with linear slow release

24.25

33.27

-0.994

Total

96/03 Manningtree

96/04 Ongar

Model

DT50

(Days)

DT90

(Days)

Fit

Model

DT50

(Days)

DT90

(Days)

Fit

Excel

First order fit

49.21

163.47

0.881

First order fit

15.55

51.66

0.777

Timme Frehse

(best fit)

First order fit

47.76

NC

0.880

First order fit

15.58

51.77

0.714

Kim

Three compartment decay with linear slow release

69.90

175.32

-0.993

Three compartment decay with linear slow release

22.97

26.42

-0.999

 

Validity criteria fulfilled:
not specified
Conclusions:
Under the conditions of the study the test material is readily degraded in the aerobic aquatic systems and is therefore unlikely to persist in the aquatic environment.
Executive summary:

The rate of test material degradation in two aquatic sediment systems was assessed according to BBA Guidelines, Part IV, Section 5-1 and in compliance with GLP.

The degradation of [14C]-test material, applied at a rate equivalent to 1.123 kg ai ha^-1, has been studied in two different aquatic sediment systems over a period of 100 days at 20 °C. The two sediments differed in that the Manningtree sediment had higher organic carbon and nitrogen content, whilst the Ongar system was more alkaline with a much higher cation exchange capacity.

The incubation was performed in glass flasks (approximately 7.5 cm internal diameter), containing sediment to an approximate depth of 2.5 cm covered with associated water to an approximate depth of 5 cm above the sediment. The units were maintained in the dark at 20 °C ± 2 °C. The water/sediment systems were incubated for approximately 6 weeks to enable acclimatisation, prior to [14C]-test material application to the surface water.

Moistened carbon dioxide-free air was supplied under positive pressure through the water in each unit and the effluent air passed through an ethylene glycol and two 2M potassium hydroxide traps, to trap any volatile products and liberated carbon dioxide (14CO2) respectively.

A good recovery of radioactivity was obtained for both systems with an overall mean recovery of 110.0 % (range 114.0 - 103.7 %) for system 96/03 (Manningtree). And 105. 5% (range 112.6 - 87.4 %) for system 96/04 (Ongar).

There was a steady transfer of the radioactivity from the water to the sediment in both systems, commencing at 111.29 % in the water phase at zero time and declining to 15.43 % (Manningtree system) and 1.80 % (Ongar system) after 100 days.

Extractable residues rose from 0 % at the first time-point (no extraction was made) to a maximum of 13.48 % at 14 days (Manningtree system). The Ongar system reached a maximum extracted residue level of 6.64 % after 14 days. Unextactable residues in both systems fluctuated with time from 0.77 % (Manningtree system) and 2.80 % (Ongar system) at time zero to 27.98 % (Manningtree system) at 100 days and 39.67 % (Ongar system) at 61 days.

Volatiles characterised as carbon dioxide accounted for 55 % (range 79 - 31 %) (Manningtree system) and 58 % (range 42 - 73 %) (Ongar system) of the applied dose after 100 days. Volatile materials, trapped in ethylene glycol, accounted for a negligible fraction (< 0.1 %) of the applied dose after 100 days. The test material was found to be the main component present in both the water and sediment phases. In addition significant mineralisation to carbon dioxide occurred. Some minor metabolites were observed in both the sediment and water phases, none of which exceeded 10 % of applied radioactivity in the total system.

Manningtree sediment: Water phase

DT50 = 49.23 days

DT90 = 155.47 days

Manningtree sediment: Complete system

DT50 = 60.66 days

DT90 = 175.23 days

Ongar sediment: Water phase

DT50 = 24.25 days

DT90 = 33.27 days

Ongar sediment: Complete phase

DT50 = 22.97 days

DT90 = 26.42 days

Under the conditions of the study the test material is readily degraded in the aerobic aquatic systems and is therefore unlikely to persist in the aquatic environment.

Endpoint:
biodegradation in water: simulation testing on ultimate degradation in surface water
Type of information:
experimental study
Adequacy of study:
key study
Study period:
11 September 2013 to 20 November 2013
Reliability:
1 (reliable without restriction)
Rationale for reliability incl. deficiencies:
guideline study
Qualifier:
according to guideline
Guideline:
OECD Guideline 309 (Aerobic Mineralisation in Surface Water - Simulation Biodegradation Test)
Version / remarks:
2004
Deviations:
no
GLP compliance:
yes (incl. QA statement)
Radiolabelling:
yes
Oxygen conditions:
aerobic
Inoculum or test system:
natural water: freshwater
Details on source and properties of surface water:
- Details on collection: Natural aerobic surface water was sourced from the Rhineland-Palatinate (67374 Hanhofen, Germany, 49°31’N, 08°32’O, where the use of phenoxy herbicide in surrounding agricultural areas is unlikely). Water was collected from the top 6 cm of the natural resource on 11 September 2013.
- Storage conditions: The water was transported in polyethylene containers to the laboratory and kept for 7 days at about 4 °C in the dark under aeration. Prior to use particles were removed by sedimentation.
- Temperature at time of collection: 17.7 °C
- pH at time of collection: 8.28
- Oxygen concentration: Below the water surface: 8.82 mg/L; Water/sediment interface: 6.38 mg/L
- Dissolved organic carbon: 8.6 mg/L
- Biochemical Oxygen Demand (BOD5): < 3 mg/L
Duration of test (contact time):
58 d
Initial conc.:
10 µg/L
Based on:
test mat.
Remarks:
radiolabelled (spiked radioactivity of ca. 0.04 MBq/ vessel)
Initial conc.:
100 µg/L
Based on:
test mat.
Remarks:
radiolabelled (spiked radioactivity of ca. 0.41 MBq/ vessel)
Parameter followed for biodegradation estimation:
CO2 evolution
radiochem. meas.
Details on study design:
TEST CONDITIONS
- Volume of test solution/treatment: The test vessels were each filled with 500 mL surface water
- Test temperature: 20 ± 2 °C
- pH adjusted: no
- Aeration of dilution water: Each system was aerated by a slight orbital movement of the test vessel on an orbital shaker
- Suspended solids concentration: < 10
- Continuous darkness: yes

TEST SYSTEM
The study was performed under continuous air flow using 1000 mL all-glass metabolism flasks (inner diameter: ca. 10.1 cm; surface: ca. 80 cm²). Any carbon dioxide generated in the flasks was trapped by two sodium hydroxide reservoirs. Any organic volatiles generated in the flasks were trapped by Tenax as an adsorbent. During the test period vessels were moistened by an all-glass metabolism flask which was connected in front of the test flasks and which was filled with pure water. For sterile test systems a 0.45 μm filter was used in front of this water flask. The pH-value and the dissolved oxygen concentration value were measured throughout the study from the blank controls once a week.
- Application: The application volume was 58 μL (5 μg test material/58 μL application solution) and 582 μL (50 μg test material/582 μL application solution). 58 and 582 μL application solution were applied in small drops directly onto the surface of the water of each test flask. The concentration of the organic solvent was below 0.2 % of the amount of water present.
- Reference material application: The application volume was 350 μL (5 μg reference material/350 μL reference material solution).
Number of culture flasks/concentration:
- Test samples (at each concentration): 14 (plus 4 spare)
- Sterilised test samples (100 µg/L): 2
- Reference samples: 6
- Solvent reference samples: 6 (2 samples spiked with 582 μL acetonitrile to monitor the influence of the solvent to the biodegradability)
- Blank controls: 2
- Solvent blank controls: 2 (spiked with 582 μL acetonitrile to monitor the influence of the solvent to the biodegradability)

SAMPLING OF TRAPS
- Details of trap for CO2 and volatile organics if used: The organic volatile traps (Tenax) were extracted with acetone and the amount of radioactivity was determined by liquid scintillation counting (LSC). The radioactive carbon dioxide evolved in the test system was trapped by sodium hydroxide (2 Molar) solution in two separate reservoirs, which were connected to the flask (each 60 mL). They were monitored for radioactivity (by LSC) at the sampling date of the corresponding flask.

SAMPLING
- Sampling frequency: Duplicate samples (from two separate flasks) of the test systems were taken for analysis on the following study days after treatment:
Test samples (10 and 100 µg/L): 0, 2, 7, 13, 19, 29, 58
Sterilised test samples (100 µg/L): 58
Reference and solvent reference samples: 13
Blank and solvent controls: 0
- Sample storage before analysis: Samples were analysed immediately by LSC. Thereafter they were stored in a freezer at approx. -18 °C. Stability of the stored samples was demonstrated by re-measuring samples stored frozen for 68 days.

EXTRACTION
- Organic Volatiles: The organic volatiles were extracted from the Tenax trap with 15 mL acetone for 1 hour on a flatbed shaker. The radioactivity in the extract was determined by liquid scintillation counting (LSC) of an aliquot (2 x 0.5 mL).
- Carbon Dioxide: The sodium hydroxide trapped carbon dioxide was determined by LSC (2 x 1 mL).
- Water Phase: The water was submitted to the following scheme: The radioactivity in the water was determined directly by LSC of an aliquot (2 x 1 mL) taken from the water phase before it was poured out of the incubation flask. To two aliquots a further 100 μL acetic acid was added and an hour later the radioactivity remaining in the aliquots was measured by LSC to determine the amount of carbon dioxide, which was dissolved in the water phase.
The remaining water phase was transferred to a HDPE flask. An aliquot was concentrated by rotary evaporation and characterised by NP-TLC.

CALCULATIONS
The disappearance time (DT) of the test material was calculated after analysis of the last sampling, including information about the dissipation/degradation kinetics according to the recommendations of EC document 9188/VI/97 rev. 8 (2000).
The calculation of the rate constant and the initial concentration was performed using the software KinGUII (2011). The initial concentration at 0 d was included in the parameter optimisation procedure, but for an optimal fit, the value was allowed to be estimated by the model.
Single first order kinetics (SFO) was assumed for the determination of the rate constants. If an acceptable determination coefficient (R > 0.7, EC document 9188/VI/97) was determined, the first order kinetic model was concluded to be an appropriate model for the degradation of the test material.
The equation describing first order kinetics is: Ct = C0 x e^(-kt)
Where:
C0 = initial concentration
t = time after application of the test material
Ct = concentration at the time t
k = rate constant

DT50 and DT90 values (time taken for 50 and 90 % of the test material to dissipate) were calculated from the kinetic parameters as follows:
DTx = [ln {100 / (100 - x)}] / k

First order multi compartment kinetic model (FOMC) was assumed for the determination of the rate constants in cases where 1st order kinetic did not provide an adequate fit to the data. The equation describing first order multi compartment kinetic is:
Ct = C0 x [(t / b) + 1]^-a
Where:
C0 = initial concentration
t = time after application of the test material
Ct = concentration at the time t
a,b = rate constants

Double first order in parallel kinetic model (DFOP) was assumed for determination of the rate constants in cases where it provided a better fit to the data than first order multi compartments kinetic. The equation describing double first order in parallel kinetic is:
Ct = C0 x [ g x e^(-kt) + (1 - g) x e^(-k2t)
Where:
C0 = initial concentration
t = time after application of the test material
Ct = concentration at the time t
k1,k2 = rate constants
g = fraction of degradation occurring under rate constant k1
Reference substance:
benzoic acid, sodium salt
Remarks:
(radiolabelled: phenyl ring)
Compartment:
natural water: freshwater
% Recovery:
97
Remarks on result:
other:
Remarks:
10 μg/L test material
Compartment:
natural water: freshwater
% Recovery:
100
Remarks on result:
other:
Remarks:
100 μg/L test material
Remarks on result:
not measured/tested
Remarks:
The mineralisation rate was negligible. Only amounts <5 % AR were detected as CO2.
Key result
Compartment:
natural water: freshwater
DT50:
2 501 d
Type:
(pseudo-)first order (= half-life)
Temp.:
20 °C
Remarks on result:
other: 10 µg/L test material
Key result
Compartment:
natural water: freshwater
DT50:
9 621 d
Type:
(pseudo-)first order (= half-life)
Temp.:
20 °C
Remarks on result:
other: 100 µg/L test material
Transformation products:
no
Remarks:
For both concentrations no metabolite was formed during the incubation period in the water system.
Volatile metabolites:
no
Remarks:
At both concentrations <1 % AR of organic volatiles were trapped
Details on results:
PERCENT OF THE RADIOACTIVITY
Each sampling during the test was done in duplicate, i.e. from two separate test flasks. The mean recoveries from the water system during the 58 days of incubation were 97 - 102 % AR for the applied amount of 10 μg/L (Table 1) and 100 - 101 % for the applied amount of 100 μg/L (Table 3).

DISTRIBUTION AND CHARACTERISATION OF THE RADIOACTIVITY (10 μg/L)
The radioactivity in the water phase showed mean recoveries of 97 - 102 %. The mineralisation rate was negligible. Only amounts <5 % AR were detected as CO2. The amount of solved CO2 in the sodium hydroxide traps was negligible and is therefore not mentioned separately. Organic volatiles were detected <1 % AR.
The radioactivity in water phase was analysed by liquid scintillation counting (LSC) and normal phase TLC. The results are presented in Table 2.
[14Ctest material could be determined with 96 - 101 % AR during the incubation period. No metabolite could be observed.
For the evaluation of the data three different kinetic models were tested in order to find the most suitable approach based on the chi² error criterion and visual assessment. The rate of disappearance of [14C]test material in the water phase was calculated by using single first order (SFO). The DT50 and DT90 values for [14C]test material (10 µg/L) were determined to be 2501 days and 8308 days in the water phases, respectively.

DISTRIBUTION AND CHARACTERISATION OF THE RADIOACTIVITY (100 μg/L)
The radioactivity in the water phase showed mean recoveries of 100 - 101 %. The mineralisation rate was negligible. Only amounts <5 % AR were detected as CO2. The amount of solved CO2 in the sodium hydroxide traps was negligible and is therefore not mentioned separately. Organic volatiles were detected <1 % AR.
The radioactivity in water phase was analysed by liquid scintillation counting (LSC) and normal phase TLC. The results are presented in Table 4.
[14C]test material could be determined with 99 - 101 % AR during the incubation period. No metabolites were observed.
For the evaluation of the data three different kinetic models were tested in order to find the most suitable approach based on the chi² error criterion and visual assessment. The rate of disappearance of [14C]test material in the water phase was calculated by using single first order (SFO). The DT50 and DT90 values for [14C]test material (100 µg/L) were determined to be 9621 days and 31 960 days in the water phases, respectively.

INVESTIGATION OF STERILISED TEST SAMPLES
The sterilised test samples were analysed 58 days after treatment. The radioactivity in the water phase of sterilised test samples showed mean recovery of 100 %. The mineralisation rate was negligible. Only amounts <5 % AR were detected as CO2. The amount of solved CO2 in the sodium hydroxide traps was negligible. Organic volatiles were detected <1 % AR. No transformation products could be observed; therefore, hydrolysis rate was negligible. The results are presented in Table 5.
Results with reference substance:
INVESTIGATION OF REFERENCE TEST SAMPLES
The reference samples and the solvent containing reference samples were analysed 13 days after treatment. All samples showed mean recoveries of mass of 93 - 97 %. The microbiological activity of test water was sufficient, because the mineralisation rate was 82 - 87 %. Therefore, the test was valid.
The system was biologically active and no influence of solvent of the activity of the system could be observed.

Table 1: Distribution of radioactivity between water phase, total carbon dioxide and organic volatiles (% of the applied radioactivity, mean of duplicate values) for the applied amount of 10 μg/L

Sampling interval (days)

0

2

7

13

19

29

58

Radioactivity in water phase

98

115*

97

102

99

98

n.a.

100

79*

97

100

93

96

97

Mean

99

97*

97

101

96

97

97

Total carbon dioxide

2

<1

1

<1

<1

<1

n.a.

n.d.

<1

2

<1

<1

<1

2

Mean

2

<1

2

<1

<1

<1

2

Organic volatiles

n.a.

<1

<1

<1

<1

<1

n.a.

n.a.

<1

<1

<1

<1

<1

<1

Mean

n.a.

<1

<1

<1

<1

<1

<1

Mean recovery

100

99*

99

102

98

98

97**

n.d.: not detected

n.a.: not analysed

*Samples of DAT 2 flowed together during the experiment. Replicate 1 of DAT 58 was not used for metabolite characterisation, because water was lost during treatment of sample.

**Replicate 1 of DAT 58 was not used for evaluation for experimental reasons.

Table 2: Distribution of the radioactivity of the water phase (% of the applied radioactivity, mean of duplicate values) for applied amount of 10 μg/L test material

Sampling interval (days)

0

2

7

13

19

29

58

Parent

99

97

97

101

96

97

97*

*Only replicate two was used for evaluation

Table 3: Distribution of radioactivity between water phase, total carbon dioxide and organic volatiles (% of the applied radioactivity, mean of duplicate values) for the applied amount of 100 μg/L

Sampling interval (days)

0

2

7

13

19

29

58

Radioactivity in water phase

101

100

100

99

99

99

101

99

101

97

101

98

100

98

Mean

100

101

99

100

99

99

99

Total carbon dioxide

n.d.

< 1

< 1

< 1

< 1

< 1

< 1

n.d.

< 1

< 1

2

1

< 1

< 1

Mean

n.d.

< 1

< 1

2

1

< 1

< 1

Organic volatiles

n.a.

< 1

< 1

< 1

< 1

< 1

< 1

n.a.

< 1

< 1

< 1

< 1

< 1

< 1

Mean

n.a.

< 1

< 1

< 1

< 1

< 1

< 1

Mean recovery

100

101

100

101

100

100

100

n.d.: not detected

n.a.: not analysed

Table 4: Distribution of the radioactivity of the water phase (% of the applied radioactivity, mean of duplicate values) for applied amount of 100 μg/L test material

Sampling interval (days)

0

2

7

13

19

29

58

Parent

100

101

99

100

99

99

99

Table 5: Distribution of the radioactivity of the water phase (% of the applied radioactivity, mean of duplicate values) for applied amount of 100 μg/L test material (sterilised test samples)

Sampling interval (days)

58

Radioactivity in water phase

100

99

Mean

99

Total carbon dioxide

1

< 1

Mean

1

Organic volatiles

< 1

< 1

Mean

< 1

Mean recovery

100

Validity criteria fulfilled:
yes
Conclusions:
The DT50 values for [14C]test material were determined to be 2501 days (10 μg/L) and 9621 days (100 μg/L) in the water phases.
Executive summary:

The rate of degradation of the test material, and the number and quantity of formed metabolites, were investigated in a study which was conducted in accordance with the standardised guideline OECD 309, under GLP conditions.

The study was performed under aerobic conditions in the dark using natural aerobic surface water taken from a large water body. To be able to determine the degradation rate and to follow the transformation products two different application rates (10 and 100 μg/L) of radio labelled test material were applied. Assuming a specific activity of 8.24 MBq/mg this corresponds to a spiked radioactivity of about 0.04 and 0.41 MBq per vessel, respectively.

Water taken from Rhineland-Palatinate (67374 Hanhofen, Germany, 49°31’N, 08°32’O) was used as the test system. This water had a dissolved organic carbon content of 8.6 mg/L and a BOD5 of < 3 mg/L.

The water system was incubated in the dark at 20 ± 2°C under constant bubbling of air through the water. The incubation period after treatment was 58 days. Organic volatiles and carbon dioxide were trapped.

Duplicate samples were taken for analysis at specified intervals up to 58 days after application. The radioactivity was quantified by liquid scintillation counting and characterised by normal phase thin layer chromatography. Reversed phase thin layer chromatography was used for confirmation of metabolites in selected samples.

The mean recoveries of both test concentrations were within the range 97 to 101 % of the applied radioactivity (AR).

The mineralisation rate was negligible for both tested concentrations. Only amounts < 5 % AR were detected as CO2. The amount of solved CO2 in the sodium hydroxide traps was negligible. Organic volatiles were detected < 1 % AR. For both concentrations no metabolites were formed during the incubation period in the water system.

The test system was validated by the help of the reference material sodium benzoate. After 13 days 82 - 87 % of the reference material was mineralised.

The DT50 values for [14C]test material were determined to be 2501 days (10 μg/L) and 9621 days (100 μg/L) in the water phases. The DT90 values for [14C]test material were determined to be 8308 days (10 μg/L) and 31 960 days (100 μg/L) in the water phases. Therefore, according to the results of the study, the test material is persistent.

Endpoint:
biodegradation in water: sediment simulation testing
Type of information:
other: Re-analysis of study data according to the latest guidelines.
Adequacy of study:
supporting study
Study period:
not applicable
Reliability:
4 (not assignable)
Rationale for reliability incl. deficiencies:
secondary literature
Reason / purpose for cross-reference:
reference to same study
Qualifier:
according to guideline
Guideline:
other: FOCUS (2006) guidance on kinetics
Deviations:
not specified
Principles of method if other than guideline:
The report covers the re-analysis of the data from a previous study, according to the latest guidelines, in order to provide the most appropriate data for triggers for further studies and for inputs into environmental models of pesticide fate, specifically for use in FOCUS STEPS 1-4 for surface water modelling (FOCUS 2001, 2012).
FOCUS (2006) guidance on kinetics was used to evaluate two water-sediment datasets, Manningtree and Ongar. Two ‘Levels’ of analysis were used: Level P1, looking at a single compartment model and Level P2, looking at a two compartment model with transfer between the water and sediment and vice versa.
GLP compliance:
no
Remarks:
Summary of data
Details on study design:
In an experimental study into the behaviour of the test material in water-sediment systems conducted previously, kinetic parameters to describe the degradation of the test material in the system were derived. However, this study was performed prior to the development of FOCUS (2006) guidance on the interpretation of degradation kinetics. The data have therefore been re-assessed, according to the latest guidelines, in order to provide the most appropriate data for triggers for further studies and for inputs into environmental models of pesticide fate, specifically for use in FOCUS STEPS 1-4 for surface water modelling (FOCUS 2001, 2012).

STATISTICAL METHODS
- Input data (Manningtree)
The data on the residues of the materials were handled as % equivalent of the parent material (test material). The reported data were processed before fitting, following the guidance from FOCUS (2006). Specifically, the guidance states that: ‘All samples < LOD are set to ½ LOD. All samples after the first non-detect (< LOD) should be omitted unless positive detections above LOQ are made later in the experiment’.
In addition, for analysis of degradation kinetics at P1 for sediment, the curve-fitting used only the decline phase and the data-points prior to this were omitted. The initial fit of kinetics to the Manningtree data-set was acceptable, both visually and statistically. Therefore, the need to further identify outliers or weight the data was unnecessary.

- Input data (Ongar)
The process of data preparation for the Ongar system also used the FOCUS (2006) guidance. To deal with data-points with no detection and the data-points from the sediment prior to the peak in the sediment, the same approach as used for Manningtree was used as described previously. However, initial curve fitting with all the remaining data-points resulted in a poor fit, both visually and statistically, for SFO and FOMC kinetics. It is clear from the data that there are some unusual kinetics operating in the Ongar system, with a slow degradation rate (or lag phase) at the start followed by a very rapid degradation phase. This was evident in the water column data, but not the sediment data.
With an unacceptable fit, the next part of the decision tree in FOCUS (2006) required a modification of the fitting routine by excluding outliers. In this case there were two options: -
1. Remove the earliest data-points and fit the kinetics to the rapid degradation phase. The authors of the original report suggest that ‘this lag phase, followed by a rapid decline of parent material is typical of the phenoxy type herbicides….The initial lag phase where degradation is relatively slow shows an ‘adaption’ of the system to the presence of the active ingredient’.
2. Remove the data-points at 30 days and 61 days. The Manningtree data did not show such an unusual shape and therefore calculating degradation on the rapid phase may not reflect the true nature of degradation of test material. In addition, this second option would result in a conservative DT50 from SFO kinetics, which could then be used in FOCUS surface water modelling.
In this case, option 2 was selected. The same process was used to remove outliers in the water column data.

Kinetic modelling procedure
- Conceptual model of degradation scheme
According to FOCUS (2006), there are 2 possible options for conducting this analysis, termed ‘Levels’. These Levels are:
Level P1: Analysis of the parent material in a single compartment to generate DissT50 (the time taken for a 50 % decline in mass or concentration of the substance to occur by dissipation from the environment or an environmental compartment after it has been applied to, formed in, or transferred to, an environmental compartment).
Level P2: Analysis of the parent material with distribution between 2 compartments to generate DegT50 (the time taken for a 50 % decline in mass or concentration of a substance to occur by degradation from the environment or an environmental compartment after it has been applied to, formed in, or transferred to, an environmental compartment).
The data for the test material were of sufficient quality to conduct both the P1 and P2 analyses.
The P1 analyses were performed separately for the aqueous phase, the sediment phase and the whole system.

- Kinetics models used
A range of kinetic models were applied:
Level P1 and Level P2: Simple first order (SFO)
Level P1 only: First order multi-compartment (FOMC), Dual first order in parallel (DFOP), Hockey Stick (HS)
The differential forms of the equations associated with each of these models are as shown below:
SFO: dM / dt = -k.M
FOMC: dM / dt = (-α / β) M [(t / β) + 1]^-1
DFOP: dM / dt = ([k1ge^-k1t + k2(1-g)e^-k2t] / [ ge^-k1t + (1-g)e^-k2t])M
HS: dM / dt = -k1.M ← t ≤ tb
HS: dM / dt = -k2.M ← t > tb
Where:
M = Total amount of chemical present at time t (molar % of parent)
t = Time since the beginning of the experiment (d)
tb = Time of the breakpoint between ‘early’ and ‘late’ times (d)
k = Rate constant (d^-1)
k1 = Rate constant in compartment 1 (DFOP) or at early times (HS) (d^-1)
k2 = Rate constant in compartment 2 (DFOP) or at late times (HS) (d^-1)
α = Shape parameter
β = Location parameter
g = Fraction of parent compound applied into compartment 1
In the integrated form of the above equations, an additional variable is needed: M0 is the amount of chemical present at time 0.

For Level P1, SFO and FOMC were tested initially. If the fit was unacceptable, DFOP and HS were tested.
For Level P2, the SFO model was applied.

- Software tool for curve fitting
The kinetic analyses were conducted using the fitting software Kingui2 version 2.2012.320.1629. The parameters were optimised by using the Iteratively Reweighted Least Squares (IRLS) algorithm.
The set of differential equations which define the kinetic model were solved with numerical methods, implemented in R (R language for statistical computing and graphics). The initial guess for the parameters was specified manually. Data were not weighted. The initial concentration of test material in the water and total system was not constrained, whilst the initial concentration of test material in the sediment was constrained to zero (for analysis at P2).

- Fitting procedure and statistical analysis
Level P1:
The first stage was to fit the decline of the parent material (test material) to the models SFO and FOMC. These models were compared, by evaluating the following elements of the results:
- the visual fit (the curve should go through the data points);
- the distribution of residuals (the residuals should be scattered randomly around zero);
- the confidence interval around the kinetic parameters (the confidence interval for a rate constant should not include zero); and
- the χ²-error test (the minimum χ-error % to pass significance test at 5 % should not exceed approximately 15 %).
The Level P1 analysis was conducted separately for the aqueous phase, the sediment phase and the whole system.
In order to derive kinetic parameters for use in other models of environmental fate, the SFO model was favoured as long as it provided a good-enough description of the data. If the fit was not adequate, then SFO-equivalent parameters were derived from one of the other kinetic models. In order to derive kinetic parameters for use as triggers for other studies, endpoints calculated from the ‘best-fit’ model were used.

Level P2:
The model SFO was used to simultaneously fit the data from the aqueous and sediment phases. The model was checked by evaluating the following elements of the results:
- the visual fit (the curve should go through the data points);
- the distribution of residuals (the residuals should be scattered randomly around zero);
- the confidence interval around the kinetic parameters (the confidence interval for a rate constant should not include zero);
- the χ²-error test (the minimum χ-error % to pass significance test at 5 % should not exceed approximately 15 %);
- the back-transfer rate should be greater than zero; and
- the Fsed test should be passed, i.e. the transfer rate constant from water to sediment divided by the sum of the two transfer rate constants should fall within a specific range. This range is dependent on the sorption coefficient KD. Test material sorption is pH dependent, with a pH of 5.3-7.7 (in water solution) corresponding to a KOC ranging from 167-12 L/kg. Using the KOC associated with higher pH (pH more than 5.5, to act conservatively for run-off in FOCUS surface water scenarios) and accounting for the organic carbon content of the sediments tested, the test material has a sorption coefficient in the range of approximately 0.7 to 1.2 L/kg, meaning that the appropriate range for Fsed is approximately 0.27 to 0.54 (FOCUS, 2006).
Compartment:
water
DT50:
38 d
Type:
other: P1 DissT50 (SFO)
Compartment:
sediment
DT50:
39 d
Type:
other: P1 DissT50 (FOMC)
Compartment:
entire system
DT50:
46 d
Type:
other: P1 DegT50 (SFO)

Level P1

- Aqueous phase

The results of the assessment criteria for acceptance of fitted kinetics at Level P1 for the test material in the aqueous phase for both Manningtree and Ongar are as follows:

Assessment and Selection of Kinetic Models for Data on the Test Material from Manningtree Systems at Level P1 in the Aqueous Phase

Assessment Criteria

SFO

FOMC

Comments

Visual assessment of residuals

Pass

Pass

χ² error (%)

8.212

10.44

t-test on rate constants

Pass

Fail

Best fit kinetics for trigger values

SFO

Best fit kinetics for modelling

SFO

Assessment and Selection of Kinetic Models for Data on the Test Material from Ongar Systems at Level P1 in the Aqueous Phase

Assessment Criteria

SFO

FOMC

Comments

Visual assessment of residuals

Pass

Pass

χ² error (%)

13.69

16.99

t-test on rate constants

Pass

Fail

Best fit kinetics for trigger values

SFO

Best fit kinetics for modelling

SFO

- Sediment phase

The results of the assessment criteria for acceptance of fitted kinetics at Level P1 for the test material in the sediment phase for both Manningtree and Ongar are as follows:

Assessment and Selection of Kinetic Models for Data on the Test Material from Manningtree Systems at Level P1 in the Sediment Phase

Assessment Criteria

SFO

FOMC

DFOP

HS

Comments

Visual assessment of residuals

Pass

Pass

Pass

Pass

χ² error (%)

21.63

24.28

28.41

24.7

t-test on rate constants

Fail

Fail

Fail

Fail

Best fit kinetics for trigger values

SFO*

Best fit kinetics for modelling

SFO*

* No kinetics models fitted to this data acquires the statistical support necessary to be accepted. The identification of the best fit is determined in a case-by-case basis. In this case, SFO kinetics have a slightly lower χ² error with an acceptable visual fit and so it has been selected as the best fit.

Assessment and Selection of Kinetic Models for Data on the Test Material from Ongar Systems at Level P1 in the Sediment Phase

Assessment Criteria

SFO

FOMC

DFOP

HS

Comments

Visual assessment of residuals

Pass

Pass

Pass

Pass

χ² error (%)

62.32

60.26

67.03

88.9

t-test on rate constants

Fail

Fail

Fail

Fail

Best fit kinetics for trigger values

FOMC*

Best fit kinetics for modelling

Back calculated
from FOMC**

* No kinetics model acquired full statistical support, however FOMC kinetics had the lowest χ² error with an acceptable visual fit and so it has been selected as the best fit.

** Modelling endpoints require the use of SFO kinetics for DT50 which can be back calculated from FOMC using DT90 / 3.32.

- Whole System

The results of the assessment criteria for acceptance of fitted kinetics at Level P1 for the test material in the whole system for both Manningtree and Ongar are as follows:

Assessment and Selection of Kinetic Models for Data on the Test Material from Manningtree Systems at Level P1 in the Whole System

Assessment Criteria

SFO

FOMC

Comments

Visual assessment of residuals

Pass

Pass

χ² error (%)

8.8

10.53

t-test on rate constants

Pass

Fail

Best fit kinetics for trigger values

SFO

Best fit kinetics for modelling

SFO

Assessment and Selection of Kinetic Models for Data on the Test Material from Ongar Systems at Level P1 in the Whole System

Assessment Criteria

SFO

FOMC

Comments

Visual assessment of residuals

Pass

Pass

χ² error (%)

12.01

16.05

t-test on rate constants

Pass

Fail

Best fit kinetics for trigger values

SFO

Best fit kinetics for modelling

SFO

Summary for P1

The fitted values for the DissT50 and DissT90 are shown below. The values for Manningtree for both water and the whole system are similar to those previously reported in the original study. However, the values for Ongar for both water and the whole system are higher than those originally reported. The cause of the differences is likely due to the preparation of the data for analysis (exclusion of outliers) and the use of SFO kinetics rather than the three compartment decay model used in the original analysis. This results in a more conservative DT50.

Summary of DissT50 and DissT90 for Use as Values in Modelling and to Trigger Further Studies Resulting from the Analysis of Water-Sediment Studies of the Test Material at Level P1

System

Endpoint

Water

Sediment

Whole System*

Manningtree

DissT50 trigger [d]

49

130

59

DissT90 trigger [d]

161

432

196

DissT50 modelling [d]

49

130

59

Best fit kinetics

SFO

SFO

SFO

Ongar

DissT50 trigger [d]

30

12

35

DissT90 trigger [d]

100

131

117

DissT50 modelling [d]

30

40**

35

Best fit kinetics

SFO

FOMC

SFO

Geometric Mean

DissT50 trigger [d]

38

39

46

DissT90 trigger [d]

127

238

152

DissT50 modelling [d]

38

72

46

*Note that whole system values at Level P1 are actually DegT50 values rather than DissT50 values as the analysis can be sure that removal of the test material from the system is ultimately via degradation rather than dissipation to another compartment.

**Back calculated from FOMC DT90 / 3.32

Level P2

The results of the assessment criteria for SFO kinetics fitted to both Manningtree and Ongar data-sets are as follows.

Results of the Assessment Criteria for SFO Kinetics Fitted to Both Manningtree and Ongar Data-Sets at Level P2

Assessment Criteria

Manningtree

Ongar

Visual Fit

For water

Pass

Pass

For sediment

Pass

Pass

Distribution of the residuals

For water

Pass

Pass

For sediment

Pass

Pass

Confidence interval around the kinetic parameters

Degradation in aqueous phase

Pass

Pass

Degradation in sediment phase

Fail

Fail

Transfer co-efficient from water to sediment

Pass

Pass

Transfer co-efficient from sediment to water

Fail

Fail

χ2 error test

For water

8.622

14.79

For sediment

27.92

70.92

For whole system

11.381

21.17

Back-transfer rate (sediment to water)

Is the value positive?

Yes, pass

Yes, pass

Fsed test

0.99, Fail

0.99, Fail

These results at Level P2 show that the kinetic fit for two compartment analysis is not acceptable for either Manningtree or Ongar data-sets.

Choice of DT50 for surface water modelling of the test material

The inability to acquire a statistically supported fit at Level P2 requires the use of results from Level P1 in FOCUS modelling. At FOCUS STEPS 1 & 2 the use of the whole system DegT50 calculated at Level P1 can be used for the appropriate compartment according to set guidance.

With respect to FOCUS STEP 3, FOCUS (2001, 2012) states:

‘Experience of following this FOCUS kinetics guidance has shown that in the vast majority of cases first order whole system DT50 are selected for calculating the geometric mean (in accordance with the procedures defined for P-I, as the statistical criteria for accepting a P-II approach are rarely satisfied). In this situation (only P-I assessment accepted) the usual evaluation practice has been to ascribe the whole system DT50 to the water phase for compounds with a Koc< ca. 100 mL/g or to the sediment phase for compounds with a Koc > ca. 2000 mL/g and use a default of 1000 days for the other compartment. This is considered by Member State regulators to be a reasonable "rule of thumb". For compounds with Koc between 100 and 2000 mL/g, the FOCUS kinetics advice regarding running simulations with both combinations for ascribing the whole system DT50 and default and selecting the results that give the highest concentrations for the risk assessment should be followed. It shouldn’t be forgotten that often the highest concentrations in sediment and water originate from the contrary simulation approaches.’

The sorption of the test material is known to be pH dependent, KOC decreasing with increasing pH. With tests conducted on sandy soils with pH between 5.2 - 5.3 (in water solution), the KOC of the test material ranged between 135 - 167 L/kg. With tests conducted on sandy soils with a pH of 5.7 - 7.7 (in water solution), the KOC of the test material ranged between 12 - 42.9 L/kg. With respect to surface water and the run-off exposure route, the lower KOC associated with a higher pH will result in less sorption and therefore greater run-off to surface waters in the aqueous phase, thereby providing a conservative assessment for the range of KOC values available. Therefore, using a KOC in the 12 - 42.9 L/kg range means that the whole system DegT50 should be allocated to the water phase and the default DT50 of 1000 days should be allocated to the sediment phase.

Choice of DT50 and DT90 for Use to Trigger Further Studies

The DT50 values for use as trigger values from the water phase and whole system can be taken directly from kinetics analysis at P1. The values are similar to those calculated from the original report. The use of P1 values for sediment may also be appropriate, but the user should be aware that the fit for sediment at Level P1 was quite weak, with large χ² errors in all models used.

Recommended Endpoints for Use as Trigger Values for Further Testing as the Geometric Mean Calculated from Fitted Kinetic Analysis

Purpose

Parameter

Water

Sediment

Whole system

Trigger

Kinetic Level

P1 DissT50

P1 DissT50

P1 DegT50

Type of Kinetics

SFO

FOMC

SFO

DT50 [d]

38

39

46

DT90 [d]

127

238

152

 

Validity criteria fulfilled:
not applicable
Conclusions:
The data did not allow a firm relationship to be built-up between the partitioning of test material to the sediment and degradation in the water phase in a two-compartment analysis. This was primarily a function of limited back-transfer from the sediment back into the water column.
Executive summary:

The report covers the re-analysis of data from a previous study of the behaviour of the test material in water-sediment systems, according to the latest guidelines, in order to provide the most appropriate data for triggers for further studies and for inputs into environmental models of pesticide fate, specifically for use in FOCUS STEPS 1-4 for surface water modelling (FOCUS 2001, 2012).

FOCUS (2006) guidance on kinetics was used to evaluate two water-sediment datasets, Manningtree and Ongar. Two ‘Levels’ of analysis were used: Level P1, looking at a single compartment model and Level P2, looking at a two-compartment model with transfer between the water and sediment and vice versa.

The kinetic analysis conducted shows some differences to that originally performed. The values for Manningtree for both water and the whole system are similar to those previously reported. However, the values for Ongar for both water and the whole system are higher than those originally reported. The cause of the differences is likely due to the preparation of the data for analysis (exclusion of outliers) and the use of an SFO kinetics model rather than the three-compartment decay model used in the original analysis. This results in conservative degradation kinetic values for use in environmental fate modelling.

The data did not allow a firm relationship to be built-up between the partitioning of test material to the sediment and degradation in the water phase in a two-compartment analysis. This was primarily a function of limited back-transfer from the sediment back into the water column. In practice, this means that whole system DegT50 must be used in FOCUS STEP 3 modelling for one compartment (water) and the default DT50 of 1000 days used for the other compartment (sediment). This allocation is driven by the KOC of the test material as recommended by FOCUS (2012) guidance.

Endpoint:
biodegradation in water: sediment simulation testing
Type of information:
experimental study
Adequacy of study:
supporting study
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
study well documented, meets generally accepted scientific principles, acceptable for assessment
Qualifier:
no guideline followed
Principles of method if other than guideline:
The natural attenuation of the test material was studied by determining changes in enantiomeric fraction in different redox environments down gradient from a landfill in the Lincolnshire limestone. In order to confirm the processes occurring in the field, microcosm experiments were undertaken using limestone acclimatised in different redox zones.
GLP compliance:
not specified
Specific details on test material used for the study:
Racemic mixture, deposited in a landfill in the Lincolnshire limestone aquifer.
Radiolabelling:
no
Oxygen conditions:
aerobic/anaerobic
Inoculum or test system:
other: natural groundwater / sediment
Details on source and properties of surface water:
Groundwaters for use in the laboratory microcosms were collected from six boreholes. Groundwater samples were collected using a Grundfos SP1 submersible pump, in 3 L polyethylene terephthalate (PET) bottles with no headspace. The PET bottles have an integral barrier layer polymer to preclude gas permeation.
Details on source and properties of sediment:
Microbially colonised sediments for use in the microcosms were prepared by suspending polypropylene mesh cylinders (Plastock Polymers, Birkenhead, UK) containing crushed Lincolnshire limestone (1 mm dia.) down the boreholes for three months. The limestone was obtained from a quarry in the vicinity of the landfill in the Lincolnshire limestone and correlated with the lithological horizon of the limestone in the aquifer.
Duration of test (contact time):
224 d
Parameter followed for biodegradation estimation:
test mat. analysis
Details on study design:
MICROCOSM PREPARATION
Twenty four microcosms were constructed using standard, 250 mL amber-glass, reagent bottles. Twelve of these were fitted with screw-cap Mininert® valves to provide a set of anaerobic microcosms (i.e. six sterile and six non-sterile). The remaining 12 were reserved for a set of aerobic microcosms, and these were fitted with ordinary screw-caps containing PTFE-faced liners.
Anaerobic microcosms were assembled in a glovebox under an atmosphere of nitrogen. Moist sediment (ca. 50 g dry weight) was added to each bottle followed by 200 mL of the respective groundwater. Microcosms that were to be sterile controls also contained 0.4 g of sodium azide (i.e. 0.2 % w/v). The anaerobic microcosms were placed in stainless steel vessels, through which nitrogen could be flushed, and agitated gently on shakers installed in an incubator maintained at 10 ± 1 °C. Periodically the microcosms were taken from their nitrogen-filled containers and sampled by syringe. Septa on the Mininert® valves were intentionally cored so that nitrogen flowing across the face of the valve would be preferentially drawn into the microcosms at this point by the partial vacuum created on withdrawal of a sample.
Similar aerobic microcosms fitted with screw caps were prepared. Periodic sampling involved removing the screw-cap and taking a sample by syringe. After sampling, and before return to the incubator, headspace air in the aerobic microcosms was exchanged for fresh air to maintain aerobic conditions.

Transformation products:
yes
Remarks:
(R-isomer degradation - anaerobic)
No.:
#1
Details on results:
Results from the individual microcosm experiments are discussed below.

1. Methanogenic/sulphate reducing microcosms (Borehole 28)
No degradation was observed in the methanogenic microcosms from Borehole 28. The EF remained at 0.50 ± 0.01 in both sterile and non-sterile microcosms throughout their 224 day operation, although there was some decline in test material concentration from 3 100 to 2 800 µg/L. Since this occurred in both sterile and non­ sterile microcosms, to the same extent, it was attributed to abiotic processes, possibly chemical reaction. The walls of the non-sterile microcosm developed a coating of black iron sulphide during the incubation and sulphate concentration declined from 105 to 12 mg/L in 119 days, indicating active sulphate reduction. This did not occur in the sterile microcosm.
It appears that in this microbially active environment, micro-organisms for mecoprop degradation are either not present, or are being inhibited, or are preferentially degrading other organic compounds.
After 119 days a nitrate amendment (543 mg/L) was made to assess the effect of this terminal electron acceptor on test material degradation. Although no degradation was observed the nitrate supplement did result in a steady increase in sulphate in the non-sterile microcosm from 12 mg/L at 119 days to 623 mg/L at 224 days. This was attributed to microbially mediated oxidation of both the black iron sulphide, observed on the surfaces of the sediment after acclimatisation, and that formed on the walls of the microcosm. The stoichiometry of this oxidation (3 moles of nitrate produce 2 moles of sulphate) corresponds to the observed decrease in nitrate concentration from 543 to 1 mg/L over the same period (i.e. the consumption of 542 mg/L nitrate led to the production of 611 mg/L sulphate). It is possible that no test material degraded because all the available nitrate was consumed in sulphide oxidation.

2. Nitrate-reducing microcosms (Borehole 32)
Initially, these microcosms contained 4 µg/L of R-isome compared with 680 µg/L of S-isomer (EF = 0.01). Although the concentration of R-isomer was close to the detection limit there were indications that after six days the R-isomer had disappeared from the non-sterile microcosm but not from the sterile one. This implied either enantioselective anaerobic biodegradation of R-isomer or inversion of R-isomer to S-isomer. To confirm which, the microcosms were spiked with 500 µg/L racemic test material after 18 days. This resulted in an EF of 0.29 in the non-sterile microcosm and 0.31 in the sterile microcosm. Thereafter, in the non-sterile microcosm the EF fell to 0.28 after 3 days then to 0.13 after 13 days and was 0.00 after 17 days. No evidence of increasing concentration of S-isomer signifying (R)- to (S)-inversion was observed. EF in the sterile microcosm, which received no further test material amendments, remained at 0.31 ± 0.02 for the ensuing 161 days of the experiment. A further spike of R-isomer to the non-sterile microcosm, producing a concentration of 10 471 µg/L, was made 46 days after the microcosm was first started. After this amendment S-isomer remained constant at 1 263 ± 18 µg/L throughout, but that R-isomer was transformed with the concomitant growth of the metabolite 4-CMP. Constancy of S-isomer concentration, during the profound decrease in R-isomer concentration, confirmed that biodegradation and not chiral inversion was the process involved.
The maximum concentration of 4-CMP achieved was 6 856 µg/L which would have required stoichiometric conversion of 10 322 µg/L of test material which corresponds well with the initial concentration of 10 471 µg/L for R-isomer. Only when R-isomer disappeared did biodegradation of 4-CMP commence. This suggested that 4-CMP was a transient metabolite formed by cleavage of the ether linkage between the aromatic and propionic acid moieties of R-isomer.
A plot of R-isomer concentration against time produced a graph that gave a coefficient of regression (R^2), of 0.9733 when a linear fit was employed, indicating that degradation could be described by zero-order kinetics with a degradation rate of 0.65 mg/L/day. Whilst zero-order kinetics provide a reasonable fit to the data they do not imply any understanding of the complex kinetics of the underlying microbial processes and the extraction of a zero-order anaerobic degradation rate is intended merely to give some approximate idea of relative disappearance rates when comparing anaerobic and aerobic degradation in the microcosms.
Nitrate reduction was identified as the terminal electron accepting process for the following reasons. This groundwater (Borehole 32) had the highest nitrate concentration of all microcosm groundwaters studied at 29.9 mg/L. Nitrate concentration declined progressively throughout the experiment to below detection limit (< 0.2 mg/L) some 73 days after the final large R-isomer spike, (i.e. 119 days from initial construction of the microcosm), by which time all R-isomer and all 4-CMP had degraded. This then preceded the 119 day nitrate addition, 500 µg/L, that had been made to all anaerobic microcosms. Stoichiometric considerations reveal that 27.7 mg/L of nitrate would be required to completely mineralise the 10 898 µg/L of R-isomer that had been present in the microcosm since its construction and subsequent spikings. Nitrite, which is a transient species generated during microbial nitrate reduction, was detected only in the non-sterile microcosm during the experiment. It was not detected in the sterile microcosm.
Iron reduction may be ruled out because there was very little iron(II) in Borehole 32 groundwater at the start of the experiment (0.05 mg/L) and this concentration remained constant throughout. Had iron reduction occurred steadily increasing concentrations of aqueous iron(II) would have been expected due to reductive dissolution of insoluble iron(III) minerals. Likewise, there was no evidence of sulphate reduction since sulphate remained steady at 150 mg/L in both sterile and non-sterile microcosms.

3. Induced nitrate-reducing microcosms (Boreholes 4, 25 and 9)
Similar behaviour to that in Borehole 32 microcosms was found in some of the other anaerobic non-sterile microcosms (Boreholes 4, 25 and 9) after the 119 day nitrate addition. Thus, for Borehole 4 microcosm, degradation of R-isomer commenced promptly following addition. Microcosms for Boreholes 25 and 9 behaved analogously to the Borehole 4 microcosm, but with lag times of around 20 and 35 days, respectively, following nitrate amendment before degradation commenced and with maximum 4-CMP formation at 200 and 220 days from initial construction, respectively.
In common with Borehole 32, groundwaters from Boreholes 4, 25 and 9 had EF < 0.5, but, they also had lower nitrate concentrations; 12.0, 3.85 and < 0.6 mg/L, respectively. In each non-sterile microcosm assembled using these groundwaters nitrate content had been completely consumed after 53 days with no observable impact upon test material concentrations. It is interesting to note that lag periods associated with R-isomer biodegradation, following nitrate amendment, correlate with proximity of the boreholes to Borehole 32. Thus, Borehole 4, only 90 m down-flow from Borehole 32 displays, in common, no lag preceding test material biodegradation. For Borehole 25 which is about 500 m up-flow from Borehole 32 the lag period is about 20 days and for Borehole 9 situated about a kilometre down-flow the lag is approximately 35 days. Differences in lag times may indicate the relative degree of acclimatisation of micro-organisms responsible for enantioselective anaerobic R-isomer biodegradation at the various locations, i.e. reflect the length of time since they were actively engaged in this process.

4. Sulphate reducing microcosms (Borehole 35)
No perceptible changes in enantiomeric fraction throughout the 224 day duration of investigation were observed in the microcosm pair constructed with groundwater from the borehole situated furthest from the landfill (Borehole 35), 4.5 km down-gradient, in the confined sulphate-reducing region of the aquifer. Because this groundwater contained about 0.7 µg/L total test material and only produced measurable concentrations of test material after pre-concentration of a large volume by solid-phase extraction, it was amended with racemic mecoprop to produce microcosm concentrations of each enantiomer of 129 µg/L (non-sterile) and 138 µg/L (sterile), well above the HPLC limit of detection. For both sterile and non-sterile microcosms enantiomeric fraction remained at 0.50 ± 0.01 throughout, regardless of the 119 day nitrate amendment. NPOC was low at < 1.00 mg/L, less than at Borehole 9, situated 1.5 km from the landfill, where NPOC was 3.29 mg/L. Both these boreholes occur in the confined region of the aquifer, and have the lowest NPOC values of the six groundwaters examined, yet both show differing anaerobic test material-degrading behaviour. Thus, Borehole 9, following a lag period, yielded rapid anaerobic biodegradation of test material when sufficient nitrate was present, whereas nitrate would not induce Borehole 35 to react in this manner. However, Borehole 9 is much closer to the unconfined zone of the aquifer and its groundwater does not possess the sulphate-reducing characteristics of Borehole 35 groundwater which result from prolonged confinement. Lack of test material-degrading activity in Borehole 35, in contrast to Borehole 9, may be due to an oligotrophic environment in which no acclimatised test material-degraders are supported because of low test material and NPOC concentrations. Alternatively, it could be that the sulphate-reducing conditions found to prevail at Borehole 35 inhibit anaerobic test material degradation.

5. Aerobic microcosms (Boreholes 32, 4 and 9)
Microcosms, that had exhibited enantioselective biodegradation of only R-isomer when kept under anaerobic conditions, were able aerobically to biodegrade test material rapidly and enantioselectively but in a very different fashion. Non-sterile aerobic microcosms from Boreholes 32, 4 and 9 exhibited fast degradation of both test material enantiomers with no perceptible lag phase. To quantify biodegradation rates these three non­sterile microcosms were spiked with racemic mecoprop to give initial test material concentrations of about 15 000 µg/L. Zero-order rate constants and associated linear regression coefficients reveal faster biodegradation of S-isomer, with ultimate biodegradation of both enantiomers.
Rates obtained for aerobic degradation are significantly greater than those for anaerobic degradation, but it should be borne in mind that a constant plentiful supply of atmospheric oxygen was provided to all aerobic microcosms. Lack of a lag-phase suggests that the microorganisms responsible for aerobic transformation were acclimatised and had therefore at some recent stage been performing this function. It seems probable that in the vicinity of Boreholes 32, 4 and 9, where water level fluctuations show large seasonal variations redox conditions can shift from anaerobic to aerobic and vice versa, but the microbial community is capable of degrading test material under either condition.

6. Aerobic microcosms (Boreholes 25, 28 and 35)
The remaining non-sterile microcosms constructed with groundwaters from Boreholes 25, 35 and 28 also displayed aerobic test material biodegradation but only after significant and differing lag times of 20, 60 and 100 days respectively. Aerobic behaviour of the microcosm pair prepared from Borehole 25 groundwater was typical of degradation encountered with these three groundwaters. The lag-phase suggests that microorganisms responsible for aerobic transformation were present but were not acclimatised. No 4-CMP was detected in any aerobic microcosm chromatograms, probably because its transformation rate was equal to or greater than that of test material under the conditions prevailing in our experiments.
Validity criteria fulfilled:
not applicable
Conclusions:
The laboratory microcosms confirmed the processes contributing to variations in enantiomeric fraction of test material down gradient from a landfill in a limestone aquifer.
They revealed that the test material does not degrade in methanogenic/sulphate-reducing and iron-reducing conditions. In nitrate-reducing conditions R-isomer degrades but S-isomer does not. In aerobic conditions S-isomer degrades at a faster rate than R-isomer. There was no indication of enantiomeric inversion.
It was also possible to show that anaerobic transformation proceeded by initial cleavage of the ether linkage leading to formation of the transient metabolite 4-CMP. Only when all R-isomer had been converted to 4-CMP did the metabolite itself begin to biodegrade anaerobically.
Executive summary:

The natural attenuation of test material was studied by determining changes in enantiomeric fraction in different redox environments down gradient from a landfill in the Lincolnshire limestone. Such changes could be due to differential metabolism of the enantiomers, or enantiomeric inversion. In order to confirm the processes occurring in the field, microcosm experiments were undertaken using limestone acclimatised in different redox zones.

No biodegradation was observed in the methanogenic, sulphate-reducing or iron-reducing microcosms. In the nitrate-reducing microcosm S-isomer did not degrade but R-isomer degraded with zero order kinetics at 0.65 mg/L/day to produce a stoichiometric equivalent amount of 4-chloro-2-methylphenol. This metabolite only degraded when the R-isomer disappeared. In aerobic conditions S- and R-isomer degraded with zero order kinetics at rates of 1.90 and 1.32 mg/L/day, respectively.

The addition of nitrate to dormant iron-reducing microcosms devoid of nitrate stimulated anaerobic degradation of R-isomer after a lag period of about 20 days and was associated with the production of 4-chloro-2-methylphenol. Nitrate addition to sulphate-reducing/methanogenic microcosms did not stimulate  test material degradation. However, the added nitrate was completely utilised in oxidising sulphide to sulphate. There was no evidence for enantiomeric inversion.

The study reveals new evidence for fast enantioselective degradation of R-isomer under nitrate-reducing conditions.

Endpoint:
biodegradation in water: sediment simulation testing
Type of information:
read-across from supporting substance (structural analogue or surrogate)
Adequacy of study:
supporting study
Justification for type of information:
Please see the read-across justification report in IUCLID Section 13.
Reason / purpose for cross-reference:
read-across source
Transformation products:
yes
No.:
#1
Endpoint:
biodegradation in water and sediment: simulation testing, other
Remarks:
Biodegradation in groundwater and sediment.
Type of information:
experimental study
Adequacy of study:
supporting study
Study period:
not reported
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
study well documented, meets generally accepted scientific principles, acceptable for assessment
Qualifier:
no guideline followed
Principles of method if other than guideline:
The potential of a riparian fen to mineralise herbicides that could leach from an adjacent catchment area was investigated. Slurries were prepared from sediment and ground water collected from different parts of a wetland representing different redox conditions. The slurries were amended with O2, NO3^-, SO4^2-, and CO2, or CO2 alone as electron acceptors to simulate the in situ conditions and their ability to mineralise the herbicide test material. In addition, the abundance of bacteria able to utilise O2, NO3^-, SO4^2- + CO2 and CO2 as electron acceptors was investigated along with the O2–reducing and methanogenic potential of the sediment.
GLP compliance:
not specified
Radiolabelling:
yes
Remarks:
Radiolabelled and unlabelled test material.
Oxygen conditions:
aerobic/anaerobic
Inoculum or test system:
natural water / sediment
Details on source and properties of surface water:
- Details on collection: Ground water was collected from polyethylene piezometers nested at the stations using an immersion pump and silicone tubing.
- Storage conditions: Samples were stored at 4 °C until use.
- Analysis of ground water: Oxygen, pH, and conductivity were measured in the field in a flow cell mounted with electrodes. Phosphate was measured by a modified ascorbic acid method, nitrite was measured spectrophotometrically. Other anions were analysed by ion chromatography. Ammonium was measured as indophenol. Ferrous iron was measured spectrophotometrically. Methane was analysed by GC-FID. Alkalinity was determined by titration. The total organic carbon content of the sediment was determined by combustion.

Details on source and properties of sediment:
The sediment type was sand at Stations 0 and 5, sandy peat at S9-I, and peat at S9-II. Sediment cores were collected using a stainless steel borer and the sediment immediately transferred to sterile air-tight containers, completely filling them.

Duration of test (contact time):
473 d
Initial conc.:
ca. 0.25 µg/L
Based on:
test mat.
Parameter followed for biodegradation estimation:
O2 consumption
Details on study design:
RESPIRATION RATES
Aerobic respirationwas determined as oxygen consumption in 116-mL serum bottles sealed with butyl rubber stoppers. Approximately 10 g of wet sediment was added to five replicate bottles and incubated at 10 °C. A test tube containing 2 mL 1 M NaOH was inserted in each bottle to trap evolved CO2. The pressure in the bottles was measured daily for 1 wk using a pressure transducer with a digital readout voltmeter. The pressure transducer had a range of 0 to 15 psi (0 – 103 kPa) with an accuracy of 0.1 psi ± 2 %. Five empty bottles served as controls to compensate for variation in atmospheric pressure. Standard curves were produced by withdrawal of known amounts of air from the bottles.
Potential methanogenesis was determined in similar bottles amended with sediment corresponding to 15 g dry weight and 30 mL ground water. Hydrogen (12.5 mL) and Na-acetate (3.4 mM) were added as electron donors. The flasks were incubated at 10 °C and evolution of methane followed on a gas chromatograph for 45 d.

MINERALISATION STUDIES
For studies of herbicide mineralisation, 100 mL of a stock solution containing the radiolabelled and unlabelled test material in methanol was added to sterile flasks. This corresponded to a total of 0.75 µg herbicide per flask or ca. 25 µg/L ground water. The flasks contained 1667 Bq of test material. The methanol was allowed to evaporate before addition of the sediment. Flasks (100 mL) with ground glass joints were used for the aerobic experiments, while 116-mL butyl rubber stoppered serum bottles with aluminum crimp seals were used for the anaerobic experiments.
The slurries were prepared in triplicate using wet sediment corresponding to 10 g dry weight and 30 mL ground water sampled from the same depth as the sediment. The sediment–ground water slurries from site S0 were set up aerobically while those from site S5 were set up in an atmosphere of N2 with the addition of 10 mM KNO3. The slurries from sites S9-II and S9-I were set up anaerobically in an atmosphere of 30 % CO2 and 70 % N2. The flasks simulating sulfate-reducing conditions were amended with 0.3 mL stock solution of Na2SO4 to supply 10 mM SO4^2- in the slurries (not taking into account the water added with the wet sediment). Slurries simulating methanogenic conditions were not amended with additional electron acceptors.
To trap the 14CO2 generated by mineralisation of the radiolabelled test material, a test tube containing 1 mL 1 M NaOH was mounted vertically in the culture flasks held in place by a metal needle inserted into the butyl rubber stopper. The NaOH was changed at each sampling occasion. The NaOH solution in the aerobic flasks was sampled using a pipette in a flow chamber while that in the anaerobic flasks was sampled anaerobically through the stopper using a N2–flushed disposable syringe mounted with a valve and a 0.9- x 120-mm stainless steel needle. 10 mL OptiPhase HiSafe scintillation fluid was added to each sample and the radioactivity determined using a Wallac 1409 liquid scintillation counter. Calculations on cumulative mineralisation were corrected for quenching and background decay. The data were reported as the average 14CO2 evolution for the triplicate slurries.
Radiolabelled 14CH4 in the headspace at the end of the experiment was combusted and measured as 14CO2. A 5-mL sample of headspace was passed through the flame ionization detector of a gas chromatograph and trapped in ethanolamine and 2-methoxyethanol (1:7 v/v).
To check for 14CO2 in the headspace of the slurries, the samples were also injected while the flame ionisation detector was switched off.
% Degr.:
36
Parameter:
O2 consumption
Sampling time:
473 d
Remarks on result:
other: S0
% Degr.:
8
Parameter:
O2 consumption
Sampling time:
473 d
Remarks on result:
other: S5
% Degr.:
10
Parameter:
O2 consumption
Sampling time:
473 d
Remarks on result:
other: S9-I
% Degr.:
13
Parameter:
O2 consumption
Sampling time:
473 d
Remarks on result:
other: S9-II
Transformation products:
not measured
Details on results:
GEOCHEMISTRY AND MICROBIOLOGY
Analysis of the water chemistry at the sample sites verified that the ground water became increasingly reduced as it flowed toward the recipient stream and that sediment and ground water samples in fact represented different redox zones. Oxygen at aerobic levels was only detected at site S0, located on the hillslope. Sites S0 and S5 closest to the ground water recharge were rich in nitrate (> 1 mM), whereas sites S9-I and S9-II further along the transect were anaerobic and depleted of nitrate. The sulfate concentration was 927 µM at S9-I but only 94 M at S9-II, where methane was observed at a concentration of 24 mM. The presence of ammonium and ferrous iron at S9-I and S9-II further indicated the reduced state of these sites.

BACTERIAL ABUNDANCE AND RESPIRATION RATES
Aerobic bacterial abundance was one order of magnitude greater in the aerobic sediment (S0: 5.8 x 10^6 CFU) than in the denitrifying sediment (S5: 6.4 x 10^5 CFU). The abundance of denitrifying bacteria was lower than that of aerobic bacteria under both aerobic (S0) and denitrifying (S5) conditions. The number of bacteria using the prevalent electron acceptors was lower in the sulfate-reducing (S9-I) and methanogenic (S9-II) sediment than in the aerobic (S0) and denitrifying (S5) sediment. Both sulfate-reducing and methanogenic bacteria were more abundant in the methanogenic sediment (S9-II) than in the sulfate-reducing sediment (S9-I). However, the potential rates of methanogenesis did not differ significantly between these two sediments.

MINERALISATION
The test material was readily mineralised and was mineralised under both aerobic and anaerobic conditions. Under aerobic conditions (S0 slurries) mineralisation started immediately, and 36 % of the test material had mineralised by the end of the experiment, 473 d later. Under anaerobic conditions (S5 slurries), in contrast, only 8 % of the test material mineralised during the course of the experiment. Similarly low rates of mineralisation were obtained under both sulfate-reducing (S9-I: 10 %) and methanogenic conditions (S9-II; 13 %). Moreover, mineralisation of the test material in both S9-I and S9-II slurries was independent of the presence of added sulfate since the omission of sulfate from S9-I slurries and the addition of sulfate to S9-II slurries did not significantly affect the mineralisation of the test material. Also, there was no significant difference between the amounts of accumulated 14CO2 from test material in anaerobic slurries amended with nitrate and anaerobic slurries with sulfate or without additional electron acceptor. The presence of anaerobic conditions in the S9-I and S9-II slurries was confirmed in two ways.
(i) To ensure that oxygen had not entered the slurries during sampling, methane was measured in the culture flasks that had been sampled during the experiment and compared with unsampled controls. There was no significant difference in methane concentration in the sampled and unsampled slurries except for the S9-II slurries containing atrazine. Hence, methanogenesis was not inhibited due to oxygen entering the slurries during sampling.
(ii) Radiolabelled methane in the headspace of slurries was transformed to 14CO2 and measured. Approximately 0.1 to 1.5 % of the 14C initially added as [ring-U-14C]test material was recovered as 14CH4 after 460 d, corresponding dealky to 1.0 to 12.3 % of the accumulated 14CO2.

Bacterial Density and Activity at the Aerobic and Denitrifying Sites

Site

Aerobic Bacteria

(CFU/g dry soil*)

O2 Consumption

(µmol/g dry soil/day)

Denitrifying Bacteria

(cells/g dry soil)

0

5.8 x 10^6

0.36

< 7.0 x 10^5

5

6.4 x 10^5

nd

4.2 x 10^5

* CFU: Colony forming units.

nd: Not determined

Bacterial Density and Activity at the Sulfate-Reducing and Methanogenic Sites

Site

Sulfate-Reducing Bacteria

(cells/g dry soil)

Methanogenic Bacteria

(cells/g dry soil)

Methanogenesis

(µmol CH4/g dry soil/day)

9-I

3 x 10^1

3 x 10^1

1.8

9-II

1.5 x 10^5

10^3

1.5

Validity criteria fulfilled:
not applicable
Conclusions:
The test material was readily mineralised and was mineralised under both aerobic and anaerobic conditions. Under aerobic conditions (S0 slurries) mineralisation started immediately, and 36 % of the test material had mineralised by the end of the experiment, 473 d later. Under anaerobic conditions (S5 slurries), in contrast, only 8 % of the test material mineralised during the course of the experiment. Similarly low rates of mineralisation were obtained under both sulfate-reducing (S9-I: 10 %) and methanogenic conditions (S9-II; 13 %).
Executive summary:

The potential of a riparian fen to mineralise herbicides that could leach from an adjacent catchment area was investigated. Slurries were prepared from sediment and ground water collected from different parts of a wetland representing different redox conditions. The slurries were amended with O2, NO3^-, SO4^2-, and CO2, or CO2 alone as electron acceptors to simulate the in situ conditions and their ability to mineralise the herbicide test material. In addition, the abundance of bacteria able to utilise O2, NO3^-, SO4^2- + CO2 and CO2 as electron acceptors was investigated along with the O2–reducing and methanogenic potential of the sediment.

Study findings revealed that aerobic bacterial abundance was one order of magnitude greater in the aerobic sediment (S0: 5.8 x 10^6 CFU) than in the denitrifying sediment (S5: 6.4 x 10^5 CFU). The abundance of denitrifying bacteria was lower than that of aerobic bacteria under both aerobic (S0) and denitrifying (S5) conditions. The number of bacteria using the prevalent electron acceptors was lower in the sulfate-reducing (S9-I) and methanogenic (S9-II) sediment han in the aerobic (S0) and denitrifying (S5) sediment. Both sulfate-reducing and methanogenic bacteria were more abundant in the methanogenic sediment (S9-II) than in the sulfate-reducing sediment (S9-I). However, the potential rates of methanogenesis did not differ significantly between these two sediments.

Under the conditions of the study, the test material was readily mineralised and was mineralised under both aerobic and anaerobic conditions. Under aerobic conditions (S0 slurries) mineralisation started immediately, and 36 % of the test material had mineralised by the end of the experiment, 473 d later. Under anaerobic conditions (S5 slurries), in contrast, only 8 % of the test material mineralised during the course of the experiment. Similarly low rates of mineralisation were obtained under both sulfate-reducing (S9-I: 10 %) and methanogenic conditions (S9-II; 13 %). Moreover, mineralisation of the test material in both S9-I and S9-II slurries was independent of the presence of added sulfate since the omission of sulfate from S9-I slurries and the addition of sulfate to S9-II slurries did not significantly affect the mineralisation of the test material. Also, there was no significant difference between the amounts of accumulated 14CO2 from test material in anaerobic slurries amended with nitrate and anaerobic slurries with sulfate or without additional electron acceptor.

Endpoint:
biodegradation in water and sediment: simulation testing, other
Remarks:
Biodegradation in groundwater and sediment.
Type of information:
read-across from supporting substance (structural analogue or surrogate)
Adequacy of study:
supporting study
Justification for type of information:
Please see the read-across justification report in IUCLID Section 13.
Reason / purpose for cross-reference:
read-across source
% Degr.:
36
Parameter:
O2 consumption
Sampling time:
473 d
Remarks on result:
other: S0
% Degr.:
8
Parameter:
O2 consumption
Sampling time:
473 d
Remarks on result:
other: S5
% Degr.:
10
Parameter:
O2 consumption
Sampling time:
473 d
Remarks on result:
other: S9-I
% Degr.:
13
Parameter:
O2 consumption
Sampling time:
473 d
Remarks on result:
other: S9-II
Endpoint:
biodegradation in water: sediment simulation testing
Type of information:
experimental study
Adequacy of study:
supporting study
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
study well documented, meets generally accepted scientific principles, acceptable for assessment
Qualifier:
no guideline followed
Principles of method if other than guideline:
Test material degradation in different redox environments was tested within a Lincolnshire Limestone aquifer. The objective of the study was to assess whether changes in the enantiomeric fraction can be used to indicate natural attenuation in the field. The study involved sampling groundwater and relating systematic changes in enantiomeric fraction to different redox environments down gradient of the landfill. Laboratory microcosms were set up with groundwater and limestone acclimatised with the in situ microbial consortium to confirm the behaviour of the test material enantiomers under controlled conditions.
GLP compliance:
not specified
Specific details on test material used for the study:
The study concerned test material (racemic mixture of R- and S-isomers), originally deposited in a landfill.
Radiolabelling:
no
Oxygen conditions:
aerobic/anaerobic
Inoculum or test system:
other: in situ microbial consortium
Details on source and properties of surface water:
Groundwater samples were obtained from boreholes down gradient of the Ailsworth Road Landfill. It was not possible to obtain samples from within this landfill but heavily contaminated samples were collected from boreholes 28 and 29 adjacent to the Ben Johnson's Landfill site. All boreholes were already constructed with relatively long screen intervals, and were sampled after purging three times the borehole volume using a stainless steel Grundfos SPI submersible pump with PTFE tubing. Field measurements included pH, temperature, redox potential (Pt electrode) and dissolved oxygen (DO2 electrode) using a flow cell. Samples for organic analysis were collected by completely filling a 2.5 L amber glass Winchester sample bottle containing 2.5 g sodium azide (NaN3) as a preservative (biocide). The samples were analysed for major and trace inorganic species, for total organic carbon (TOC) and for test material. Further analytical details are found in Williams et al. (2001).

On flowing eastwards, the aerobic groundwater evolves naturally to become sulphate reducing and saline. Samples were categorised into redox zones.

Methanogenic/sulphate­reducing conditions (boreholes 28 and 29) are identified by high TOC and the presence of methane that was actively degassing from these boreholes. Fe-reducing conditions are represented by the presence of reduced iron and low nitrate (borehole 25); nitrate reducing by low reduced iron and the presence of nitrate (boreholes 24, 4 and 32); aerobic/sub-oxic by low TOC without nitrate or reduced iron (boreholes 9, 11 and 18). Boreholes 16 and 17 and other boreholes further east in the confined zone represent the transition to sulphate-reducing/saline conditions. Since redox is known to control the fate of the test material, these redox environments formed the basis against which to assess changes in the enantiomeric fraction.

The heavily contaminated boreholes 28 and 29 down gradient from Ben Johnson's Landfill were degassing methane and contained racemic test material.
Details on source and properties of sediment:
Crushed Lincolnshire Limestone
Details on inoculum:
The Lincolnshire Limestone had been suspended in boreholes for several months to allow microbial colonisation.
Parameter followed for biodegradation estimation:
test mat. analysis
Details on study design:
BACKGROUND
During the 1980s, an estimated 40 tonnes of the test material contained in tank washings was discharged into two of the sites giving concentrations of up to 39 mg/L in the waste. Subsequently, up to 8 µg/L test material was detected in a public supply borehole at Etton 2.5 km east of the landfills. Groundwater is known to flow downdip to the east of the fault where the aquifer is confined by relatively impermeable strata and where boreholes are artesian and gradient of the landfill. Laboratory microcosms were set up with groundwater and limestone acclimatised with the in situ microbial consortium to confirm the behaviour of the test material enantiomers.

MICROCOSM DESIGN
Laboratory microcosms were set up using groundwater and crushed Lincolnshire Limestone, which had been suspended in boreholes for several months to allow microbial colonisation. Each microcosm was composed of 200 mL groundwater and 50 g limestone in a 250-mL amber glass bottle. Where necessary, the groundwater was amended with racemic test material to ensure sufficiency of each enantiomer at the start of the experiment. The microcosms were incubated at 10 °C aerobically (periodically exposed to the atmosphere) and anaerobically (under a nitrogen atmosphere) and compared with identical controls sterilised with NaN3.
Four microcosms were set up with groundwater from different redox environments along the flow path. These represent methanogenic/sulphate-reducing, iron-reducing, nitrate-reducing and aerobic conditions.
Transformation products:
yes
Remarks:
R-isomer degradation product in Fe/nitrate-reducing zones
No.:
#1
Details on results:
MICROCOSMS
Groundwater from Borehole 32 which lies on the boundary between aerobic and nitrate­reducing conditions was used in two sets of microcosms which were incubated aerobically and anaerobically.

- Borehole 32 (aerobic)
Both enantiomers disappeared over a period of 6 days in the non­sterile microcosm, while there was no change in the sterile control. The S-isomer degraded with zero-order kinetics at a rate of 1.90 mg/L/day while the R-isomer degraded more slowly at 1.32 mg/L/day.

- Borehole 32 (nitrate-reducing)
In contrast, under anaerobic conditions and with exactly the same starting material, the S-isomer did not degrade, but the R-isomer degraded with first-order kinetic rate of 0.65 mg/L/day. At the same time, there was a stoichiometric build-up of 4-chloro-2-methylphenol, which only began to degrade after the R-isomer had disappeared. Brief detection of NO2 during the experiment suggested that nitrate may have been acting as the terminal electron acceptor, but nitrate was not routinely measured. The sterile control did not change.

- Borehole 25 (Fe-reducing)
No degradation was observed in this microcosm, which initially contained only 3.85 mg/L nitrate but 15 mg/L of reduced iron. After 119 days, nitrate was added to produce a concentration of 491 mg/L nitrate, and after a lag period of 21 days the R-isomer degraded with the build-up of 4-chloro-2-methylphenol as in the anaerobic microcosm borehole 32. It is not known whether degradation would have eventually taken place in this microcosm without the addition of nitrate, but the relatively quick response suggests that nitrate was probably acting as an electron acceptor.

- Borehole 28 (methanogenic/sulphate reducing)
No change was observed in either the sterile or non-sterile microcosms using water from the heavily contaminated aquifer where the presence of sulphide and methane was indicative of methanogenic/sulphate-reducing conditions.


Validity criteria fulfilled:
not applicable
Conclusions:
This study has shown that systematic change in the enantiomeric fraction of racemic test material deposited in the landfills at Helpston is a useful indicator of in situ processes and provides insight into the relative behaviour of the enantiomers under different redox conditions. Test material behaviour inferred from field data is supported by the results of microcosm experiments conducted under aerobic and anaerobic conditions. No degradation occurred in methanogenic, sulphate-reducing or iron-reducing microcosms. In nitrate­reducing microcosms, S-isomer did not degrade, but R-isomer degraded with zero-order kinetics at 0.65 mg/L/day. The associated buildup of 4-chloro-2-methyl­phenol indicated cleavage of the propionic moiety at the ether linkage, and this metabolite only degraded after the R-isomer had disappeared. In aerobic microcosms, the S-isomer degraded faster than the R-isomer both following zero-order kinetics 1.32 and 1.9 mg/L/day. The nature of the microbial community responsible for test material degradation is currently under study.
Executive summary:

Test material degradation in different redox environments was tested within a Lincolnshire Limestone aquifer. The objective of the study was to assess whether changes in the enantiomeric fraction can be used to indicate natural attenuation in the field. The study involved sampling groundwater and relating systematic changes in enantiomeric fraction to different redox environments down gradient of the landfill. Laboratory microcosms were set up with groundwater and limestone acclimatised with the in situ microbial consortium to confirm the behaviour of the test material enantiomers under controlled conditions.

This study has shown that systematic change in the enantiomeric fraction of racemic test material deposited in the landfills at Helpston is a useful indicator of in situ processes and provides insight into the relative behaviour of the enantiomers under different redox conditions. Test material behaviour inferred from field data is supported by the results of microcosm experiments conducted under aerobic and anaerobic conditions. No degradation occurred in methanogenic, sulphate-reducing or iron-reducing microcosms. In nitrate­reducing microcosms, S-isomer did not degrade, but R-isomer degraded with zero-order kinetics at 0.65 mg/L/day. The associated buildup of 4-chloro-2-methyl­phenol indicated cleavage of the propionic moiety at the ether linkage, and this metabolite only degraded after the R-isomer had disappeared. In aerobic microcosms, the S-isomer degraded faster than the R-isomer both following zero-order kinetics 1.32 and 1.9 mg/L/day. The nature of the microbial community responsible for test material degradation is currently under study.

Endpoint:
biodegradation in water: sediment simulation testing
Type of information:
read-across from supporting substance (structural analogue or surrogate)
Adequacy of study:
supporting study
Justification for type of information:
Please see the read-across justification report in IUCLID Section 13.
Reason / purpose for cross-reference:
read-across source
Transformation products:
yes
Remarks:
R-isomer degradation product in Fe/nitrate-reducing zones
No.:
#1
Endpoint:
biodegradation in water: simulation testing on ultimate degradation in surface water
Type of information:
experimental study
Adequacy of study:
supporting study
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
study well documented, meets generally accepted scientific principles, acceptable for assessment
Qualifier:
no guideline followed
Principles of method if other than guideline:
The degradation of the test material in water under laboratory conditions was studied using high-resolution gas chromatography/mass spectrometry.
GLP compliance:
no
Radiolabelling:
no
Oxygen conditions:
aerobic
Inoculum or test system:
natural water: freshwater
Details on source and properties of surface water:
- Details on collection (e.g. location, sampling depth, contamination history, procedure): Surface water was collected from five lakes and two rivers situated in the central and eastern region of Switzerland.
The lakes investigated were the Zürichsee (Lake Zurich) and Walensee (sampling near the centers) and the Greifensee, Sempachersee, and Hallwilersee (sampling at outlets).
The rivers studied were the Aabach, one of the tributaries to Greifensee (sampling at the inlet to this lake), and the river Reuss (sampling at Ottenbach, ≈34 km from Luzern).
Zürichsee (406 m above sea level (msl); two basins, total surface area 88.4 km^2; mean water residence time, 1.2 yr) is situated in a more populated area, although there are still some agricultural activities nearby (catchment area, 1760 km^2; population, 315 000). It receives most of its water from Walensee (419 msl; surface area, 24.1 km^2; mean water residence time, 1.25 yr), which is situated in a more mountainous region. Greifensee is a smaller lake (435 msl; surface area, 8.4 km^2; mean water residence time, 1.1 yr) situated 10 km east of Zürich. There are agricultural activities in its catchment area (160 km^2), but the relatively high population (≈100 000 inhabitants) in this area cause an additional input of anthropogenic compounds. Sempachersee (504 msl; surface area, 14.4 km^2) is situated 40 km southwest of Zürich. This lake has a relatively long mean water residence time of 17 yr. In its relatively small catchment area (61 km^2) are intense agricultural activities, and there is a relatively small population of about 12 000 inhabitants. Finally, Hallwilersee (449 msl; surface area, 10.0 km^2; mean water residence time, 3.8 yr) is situated 25 km southwest of Zürich and has, as does Sempachersee, a predominant input from agricultural activities in its catchment area (138 km^2) (population, 23 000).
Typically, these lakes are stratified during the warmer season (April-November) with development of an epilimnion (thermocline at 8 - 15 m) and a hypolimnion. In late fall and winter, the waters are mixed down to significant depths. The residence times for water in the epilimnia during stratification are shorter (5 - 10x) than the mean water residence times indicated above. Lateral mixing in these lakes is fast (within days) as compared to other processes that influence the behavior of the compounds studied. In Sempachersee and Hallwilersee aeration programs are in progress, resulting in total vertical mixing in winter and increased hypolimnion mixing in summer. Additionally, a small mountain lake (Jörisee; 2519 msl; surface area, ≈0.05 km^2) with inputs only from snow, ice, and rain was analysed.
Lakewater for incubation was taken from Sempachersee on August 2, 1996, and the river Aabach on August 29, 1996. No fortifications were made to these waters.
Additionally, incubation experiments were carried out with fortified water, using water from Sempachersee and the river Aabach, collected on November 4, 1996.
Surface water (depth, 1 m) from the lakes was collected with standard equipment and filled on-site into methanol-rinsed 1-L glass bottles. A fossil groundwater (zero-contaminant water) was also analysed for control purposes.
- Water filtered: No. The water samples were not filtered; however, course particles were removed by sedimentation. Therefore, the concentrations reported include the amounts dissolved and those adsorbed on fine, suspended particles.
Duration of test (contact time):
>= 50 - <= 129 d
Parameter followed for biodegradation estimation:
test mat. analysis
Details on study design:
TEST CONDITIONS
- Test temperature: The water from Sempachersee and from the river Aabach were incubated at room temperature, 20 - 23 °C.
- Aeration of dilution water: From the amounts of oxygen dissolved and the relatively small amounts of dissolved organic matter, it can be assumed that the incubation conditions in both experiments were aerobic.

TEST SYSTEM
- Culturing apparatus:
The water from Sempachersee was incubated in 1 L green glass bottles.
The water from the river Aabach was incubated in one batch in a 2.5 L clear glass bottle.

- Other: There was some algal growth observed in the clear glass bottle for the water from the river Aabach.

SAMPLING
- Sampling frequency:
Water from Sempachersee - Periodically, one of the bottles was removed from storage, and the water was analysed as described.
Water from the river Aabach - Periodically, 0.25 L subsamples were removed and analysed as described.

- Other: Extraction from the acidified (pH ≈ 2) water was effected with a reusable small column containing a macroporous polystyrene adsorbent, and the analytes were eluted and recovered. After methylation with diazomethane, the extracts were concentrated (≈0.5 mL in diethyl ether) and passed through a small silica column (0.7 g of silica gel 60, deactivated with 5 % water; 5 mm i.d. Pasteur pipet) topped with 10 mm of sodium sulfate. The analytes (as methyl esters, ME) were eluted with 10 mL of n-hexane-methylene chloride (1:1). After careful concentration and dilution to 100 - 200 μL with n-hexane, 2 μL aliquots were used for analysis by high-resolution gas chromatography/mass spectrometry (HRGC/MS).


CONTROL AND BLANK SYSTEM
- Other: One 2.5-L batch of each were fortified at 50 - 100 ng/L with the test material. The waters were incubated at room temperature in amber bottles for up to 89 d. Periodically, 0.25 L subsamples were removed and analysed as described. The fortifications were made by adding 1.0 mL of a 250 ng/mL solution of the compounds in distilled water, prepared from stock solutions of the compounds in methanol.
Remarks on result:
not measured/tested
Transformation products:
not measured
Evaporation of parent compound:
not measured
Volatile metabolites:
not measured
Residues:
not measured
Details on results:
The concentrations of the test material were reduced during the 50 - 129 d of incubation in both experiments. The data fitted approximate first-order kinetics without a lag phase.
The data indicate the test material is relatively stable in the aquatic environment whereas it degrades relatively quickly in soil (half-life, 7 - 22 d). The data for the test material show an approximately constant concentration (55 ng/L; lag phase) up to 21 d, followed by a gradual decrease to ≈ 11 ng/L at 89 d, the largest decrease occurring between 21 and 63 d.
Validity criteria fulfilled:
not applicable
Conclusions:
Under the conditions of the study the concentrations of the test material were reduced during the 50 - 129 d of incubation in both experiments. The data fitted approximate first-order kinetics without a lag phase.
The data indicate the test material is relatively stable in the aquatic environment whereas it degrades relatively quickly in soil (half-life, 7 - 22 d). The data for the test material show an approximately constant concentration (55 ng/L; lag phase) up to 21 d, followed by a gradual decrease to ≈ 11 ng/L at 89 d, the largest decrease occurring between 21 and 63 d.
Executive summary:

The degradation of the test material in surface water was determined in surface water collected from five lakes and two rivers situated in the central and eastern region of Switzerland.

The water from Sempachersee was incubated in 1 L green glass bottles and incubated for up to 129 d at room temperature (20 - 23 °C). Periodically, one of the bottles was removed from storage, and the water was analysed.

The water from the river Aabach was incubated in one batch in a 2.5-L clear glass bottle for up to 50 d at room temperature while being stirred with a Teflon bar. Periodically, 0.25 L subsamples were removed and analysed. From the amounts of oxygen dissolved and the relatively small amounts of dissolved organic matter, it can be assumed that the incubation conditions in both experiments were aerobic.

There was some algal growth observed in the clear glass bottle.

Additionally, incubation experiments were carried out with fortified water, using water from Sempachersee and the river Aabach. One 2.5-L batch of each were fortified at 50 - 100 ng/L with the test material.

The waters were incubated at room temperature in amber bottles for up to 89 d. Periodically, 0.25 L subsamples were removed and analysed. The fortifications were made by adding 1.0 mL of a 250 ng/ mL solution of the compounds in distilled water, prepared from stock solutions of the compounds in methanol.

Extraction from the acidified (pH ≈ 2) water was effected with a reusable small column containing a macroporous polystyrene adsorbent, and the analytes were eluted and recovered. After methylation with diazomethane, the extracts were concentrated (≈0.5 mL in diethyl ether) and passed through a small silica column (0.7 g of silica gel 60, deactivated with 5 % water; 5 mm i.d. Pasteur pipet) topped with 10 mm of sodium sulfate. The analytes (as methyl esters) were eluted with 10 mL ofn-hexane-methylene chloride (1:1). After careful concentration and dilution to 100 - 200μL withn-hexane, 2 μL aliquots were used for analysis by high-resolution gas chromatography/mass spectrometry (HRGC/MS).

Under the conditions of the study the concentrations of the test material were reduced during the 50 - 129 d of incubation in both experiments. The data fitted approximate first-order kinetics without a lag phase.

The data indicate the test material is relatively stable in the aquatic environment whereas it degrades relatively quickly in soil (half-life, 7 - 22 d). The data for the test material show an approximately constant concentration (55 ng/L; lag phase) up to 21 d, followed by a gradual decrease to ≈ 11 ng/L at 89 d, the largest decrease occurring between 21 and 63 d.

Description of key information

Roohi & Perry (2015)

Under the conditions of this study, [14C]-test material in natural water/sediment systems was shown to degrade ultimately to carbon dioxide and unextractable sediment bound residues.

Traub (2014)

The DT50 values for [14C]test material were determined to be 2 501 days (10 μg/L) and 9 621 days (100 μg/L) in the water phases. The DT90 values for [14C]test material were determined to 8 308 days (10 μg/L) and 31 960 days (100 μg/L) in the water phases. Therefore, according to the results of the study, the test material is persistent.

Supporting Study: Bieber & Krӧhn (1991)

Under the conditions of the study the degradation of the test material proceeded rather rapidly in both systems. The main degradation product was carbon dioxide, which after three months comprised more than 80 % of the radioactivity originally applied to the system. Within this time the amount of parent compound decreased to less than 0.4 %. At the end of the test period three months after application approximately 4 to 8 % of the radioactivity were found in humic acids in the systems, another 3 to 4 % in fulvic acids. Ca. 11 % were retained in one sediment, ca. 8 % only in the other.

A disappearance time (DT) 50 of 20 to 21 days was found, the DT 90 was 28 days. This low DT 90 value may be explained by the exponential growth of the microbial degraders after a lack phase in the beginning of the experiment.

Supporting Study: Cooper & Unsworth (1996)

Under the conditions of the study the test material is readily degraded in the aerobic aquatic systems and is therefore unlikely to persist in the aquatic environment.

Supporting Study: Buser & Müller (1998)

Under the conditions of the study the concentrations of the test material were reduced during the 50 - 129 d of incubation in both experiments. The data fitted approximate first-order kinetics without a lag phase.

The data indicate the test material is relatively stable in the aquatic environment whereas it degrades relatively quickly in soil (half-life, 7 - 22 d). The data for the test material show an approximately constant concentration (55 ng/L; lag phase) up to 21 d, followed by a gradual decrease to ≈ 11 ng/L at 89 d, the largest decrease occurring between 21 and 63 d.

Supporting Study: Hazlerigg & Garrett (2014)

The data did not allow a firm relationship to be built-up between the partitioning of test material to the sediment and degradation in the water phase in a two-compartment analysis. This was primarily a function of limited back-transfer from the sediment back into the water column.

Read-Across Substance: Harrison et al. (2003)

The laboratory microcosms confirmed the processes contributing to variations in enantiomeric fraction of test material down gradient from a landfill in a limestone aquifer.

They revealed that the test material does not degrade in methanogenic/sulphate-reducing and iron-reducing conditions. In nitrate-reducing conditions R-isomer degrades but S-isomer does not. In aerobic conditions S-isomer degrades at a faster rate than R-isomer. There was no indication of enantiomeric inversion.

It was also possible to show that anaerobic transformation proceeded by initial cleavage of the ether linkage leading to formation of the transient metabolite 4-CMP. Only when all R-isomer had been converted to 4-CMP did the metabolite itself begin to biodegrade anaerobically.

Read-Across Substance: Williams et al. (2003)

This study has shown that systematic change in the enantiomeric fraction of racemic test material deposited in the landfills at Helpston is a useful indicator of in situ processes and provides insight into the relative behaviour of the enantiomers under different redox conditions. Test material behaviour inferred from field data is supported by the results of microcosm experiments conducted under aerobic and anaerobic conditions. No degradation occurred in methanogenic, sulphate-reducing or iron-reducing microcosms. In nitrate­reducing microcosms, S-isomer did not degrade, but R-isomer degraded with zero-order kinetics at 0.65 mg/L/day. The associated buildup of 4-chloro-2-methyl­phenol indicated cleavage of the propionic moiety at the ether linkage, and this metabolite only degraded after the R-isomer had disappeared. In aerobic microcosms, the S-isomer degraded faster than the R-isomer both following zero-order kinetics 1.32 and 1.9 mg/L/day. The nature of the microbial community responsible for test material degradation is currently under study.

Read-Across Substance: Larsen et al. (2001)

The test material was readily mineralised and was mineralised under both aerobic and anaerobic conditions. Under aerobic conditions (S0 slurries) mineralisation started immediately, and 36 % of the test material had mineralised by the end of the experiment, 473 d later. Under anaerobic conditions (S5 slurries), in contrast, only 8 % of the test material mineralised during the course of the experiment. Similarly low rates of mineralisation were obtained under both sulfate-reducing (S9-I: 10 %) and methanogenic conditions (S9-II; 13 %).

Key value for chemical safety assessment

Additional information

Roohi & Perry (2015)

The route and rate of degradation of the test material in two water/sediment systems was investigated in accordance with the standardised guideline OECD 308 under GLP conditions. The study was awarded a reliability score of 1 in accordance with the criteria set forth by Klimisch et al. (1997).

The route and rate of degradation of [14C]-test material has been investigated under aerobic conditions at 20 ± 2 °C in two contrasting water sediment systems in the dark. The water sediment systems were Calwich Abbey [Staffordshire, UK, silt loam sediment, pH 7.2, 5.0 % OC] and Swiss Lake [Derbyshire, UK, loamy sand sediment, pH 6.6, 0.7 % OC]. The systems were incubated in glass flasks containing sediment and associated water at a ratio of approximately 1:3 (v/v). Throughout the experiment the flasks were maintained in the dark at 20 ± 2 °C whilst attached to an incubation system allowing air to be bubbled through the surface water and then through a system for trapping volatile degradates. The water/sediment systems were incubated for 7 days prior to application of [14C]-test material to allow the systems to equilibrate. The redox potential of the sediment and water in control flasks was measured at regular intervals during the incubation. The pH and dissolved oxygen content of the water were also measured.

[14C]-test material was applied to the water surface of individual water sediment systems at a target application rate equivalent to an initial concentration of 0.138 mg/L in the water phase, equivalent to the direct overspray from a field application at a rate of 1 380 g/ha to a water body of 100 cm depth. The treated sediment systems were incubated for up to 98 days.

At zero time (immediately after treatment) and at intervals of 7, 14, 29, 56, 81 and 98 days after treatment duplicate flasks and their corresponding traps were removed from the incubation system. The water was carefully decanted, and the sediment was extracted once with acetonitrile, then two more times with acetonitrile: water (80:20 v/v) at ambient temperature. Extracted sediment samples were air-dried (for the early time points), ground to a fine powder and the residual radioactivity quantified by combustion. For the later time points (day 56 onwards), in order to avoid possible loss of any volatile material during drying, the sediments were not dried before combustion. The radioactivity in the water, the sediment extracts and the volatile traps was quantified by liquid scintillation counting (LSC).

Following extraction, the water phase and the sediment extracts were analysed directly by reverse phase high performance liquid chromatography (HPLC).

The overall recovery of radioactivity was good throughout the study with mean values of 96.9 % of applied radioactivity (AR) for the Calwich Abbey system and 99.7 % AR for the Swiss Lake system. Recoveries for individual flasks ranged from 93.1 to 102.0 % AR for the Calwich Abbey system and from 95.3 to 101.7 % AR for the Swiss Lake system.

In the Calwich Abbey system the levels of radioactivity in the water declined steadily from 101.3 % AR to 48.2 % AR at day 81. After this initial lag phase, however, the rate of decline from the water accelerated rapidly, reaching 4.1 % AR at day 98. The total extractable radioactivity in the sediment rose from 0.0 % AR at time zero to a maximum of 23.0 % at day 56 before declining to 6.6 % at the end of the study at day 98. The unextracted radioactivity rose from 0.0 % at time zero to 32.3 % at day 98. The presence of an initial metabolic lag phase was also evident in the evolution of CO2. Over the first 56 days only 4.9 % CO2 was released, however, this had increased to 14.8 % AR by day 81 and reached 50.1 % AR by day

98.

In the Swiss Lake system there was a slower transfer of the applied radioactivity from the water to the sediment, when compared with Calwich Abbey system, and thus resulting in a much greater percentage of applied radioactivity remaining in the water at the end of the incubation period. The radioactivity in the water declined from 101.1 % AR at time zero to 62.7 % AR by day 98. The total radioactivity in the sediment increased from 0.0 % at time zero to 21.8 % by day 98, with the extractable portion reaching a maximum of 12.8 % AR at day 56 and the unextractable reaching a maximum of 10.4 % AR at day 98. The degree of mineralisation to CO2 was less than in the Calwich Abbey system accounting for 13.5 % AR by day 98.

The chromatographic results showed that in both the Calwich Abbey and Swiss Lake systems, the applied test material partitioned into the sediment and was degraded to form minor metabolites, none exceeding >5.0 % AR.

In the total Calwich Abbey system, the applied test material declined from 101.3 % AR at time zero to 66.8 % AR at day 81. After day 81 degradation accelerated rapidly such that only 8.7 % AR remained as test material by day 98. No significant metabolites (i.e. >10 % AR, >5 % AR at consecutive sampling intervals or >5 % AR without decline at end of study) were detected. In the water phase, the applied test material declined from 101.3 % AR at time zero to 47.5 % AR at day 81 and to 3.7 % AR at day 98. In the sediment extracts, the test material reached a maximum level of 22.1 % AR at day 56 and subsequently declined to 5.0 % AR by day 98.

In the total Swiss Lake system, the applied test material declined from 101.1 % AR at time zero to 73.4 % AR at day 98. No significant metabolites (i.e. >10 % AR, >5 % AR at consecutive sampling intervals or >5 % AR without decline at end of study) were detected. In the water phase, the applied test material declined from 101.1 % AR at time zero to 62.1 % AR at day 98. In the sediment extracts, the test material reached a maximum level of 13.0 % AR at day 14 and subsequently declined slightly to 11.4 % AR by the end of the study at day 98.

The dissipation of the test material from the water phase and degradation in the total system was evaluated according to the FOCUS guidance document on degradation kinetics using the most appropriate model for the best fit to the data set. In the Calwich Abbey system, with the data exhibiting an initial lag-phase of moderate degradation followed by rapid decline, the Hockey-stick (HS) model was selected as best-fit to the data for the water phase and for the total system. In the Swiss Lake system the SFO model was selected as best-fit for the degradation in the water and for the total system.

Under aerobic conditions in the Calwich Abbey system, [14C]-test material dissipated rapidly from the water phase after an initial lag phase with a best-fit overall DT50 value of 72.5 days.

Dissipation from the water phase was slower in the Swiss Lake system with a DT50 of 171 days. The degradation in the total water/sediment systems again showed differences in the two systems, with best fit DT50 values of 83.2 and 244 days for the Calwich Abbey and Swiss Lake systems, respectively.

In conclusion, under the conditions of this study, [14C]-test material in natural water/sediment systems was shown to degrade ultimately to carbon dioxide and unextractable sediment bound residues.

Traub (2014)

The rate of degradation of the test material, and the number and quantity of formed metabolites, were investigated in a study which was conducted in accordance with the standardised guideline OECD 309, under GLP conditions.The study was awarded a reliability score of 1 in accordance with the criteria set forth by Klimisch et al. (1997).

The study was performed under aerobic conditions in the dark using natural aerobic surface water taken from a large water body. To be able to determine the degradation rate and to follow the transformation products two different application rates (10 and 100 μg/L) of radio labelled test material were applied. Assuming a specific activity of 8.24 MBq/mg this corresponds to a spiked radioactivity of about 0.04 and 0.41 MBq per vessel, respectively.

Water taken from Rhineland-Palatinate (67 374 Hanhofen, Germany, 49°31’N, 08°32’O) was used as the test system. This water had a dissolved organic carbon content of 8.6 mg/L and a BOD5 of < 3 mg/L.

The water system was incubated in the dark at 20 ± 2°C under constant bubbling of air through the water. The incubation period after treatment was 58 days. Organic volatiles and carbon dioxide were trapped.

Duplicate samples were taken for analysis at specified intervals up to 58 days after application. The radioactivity was quantified by liquid scintillation counting and characterised by normal phase thin layer chromatography. Reversed phase thin layer chromatography was used for confirmation of metabolites in selected samples.

The mean recoveries of both test concentrations were within the range 97 to 101 % of the applied radioactivity (AR).

The mineralisation rate was negligible for both tested concentrations. Only amounts < 5 % AR were detected as CO2. The amount of solved CO2 in the sodium hydroxide traps was negligible. Organic volatiles were detected < 1 % AR. For both concentrations no metabolites were formed during the incubation period in the water system.

The test system was validated by the help of the reference material sodium benzoate. After 13 days 82 - 87 % of the reference material was mineralised.

Under the conditions of this study, the DT50 values for [14C]test material were determined to be 2501 days (10 μg/L) and 9621 days (100 μg/L) in the water phases. The DT90 values for [14C]test material were determined to be 8308 days (10 μg/L) and 31 960 days (100 μg/L) in the water phases. Therefore, according to the results of the study, the test material is persistent.

Supporting Study: Bieber & Krӧhn (1991)

The degradation of the test material in sediment/water was assessed according to Pesticide Assessment Guidelines Subdivision N, § 162-4 and in compliance with GLP. The study was awarded a reliability score of 2 in accordance with the criteria set forth by Klimisch et al. (1997).

The degradation of the test substance [14 C]-test material, being labelled in the aromatic ring was investigated in two different water/sediment systems under aerobic conditions over a period of three months at 22 °C. 1.88 mg test material was applied per kg of the system.

After an incubation time of approximately two weeks, the degradation of the test material proceeded rather rapidly in both systems. The main degradation product was carbon dioxide, which after three months comprised more than 80 % of the radioactivity originally applied to the system. Within this time the amount of parent compound decreased to less than 0.4 %.

At the end of the test period three months after application approximately 4 to 8 % of the radioactivity were found in humic acids in the systems, another 3 to 4 % in fulvic acids. Ca. 11 % were retained in one sediment, ca. 8 % only in the other.

For both systems a disappearance time (DT) 50 of 20 to 21 days was found, the DT 90 was 28 days. This low DT 90 value may be explained by the exponential growth of the microbial degraders after a lack phase in the beginning of the experiment.

Supporting Study: Cooper & Unsworth (1996)

The rate of test material degradation in two aquatic sediment systems was assessed according to BBA) Guidelines, Part IV, Section 5-1 and in compliance with GLP. The study was awarded a reliability score of 2 in accordance with the criteria set forth by Klimisch et al. (1997).

The degradation of [14C]-test material, applied at a rate equivalent to 1.123 kg ai ha^-1, has been studied in two different aquatic sediment systems over a period of 100 days at 20 °C. The two sediments differed in that the Manningtree sediment had higher organic carbon and nitrogen content, whilst the Ongar system was more alkaline with a much higher cation exchange capacity.

The incubation was performed in glass flasks (approximately 7.5 cm internal diameter), containing sediment to an approximate depth of 2.5 cm covered with associated water to an approximate depth of 5 cm above the sediment. The units were maintained in the dark at 20 °C ± 2 °C. The water/sediment systems were incubated for approximately 6 weeks to enable acclimatisation, prior to [14C]-test material application to the surface water.

Moistened carbon dioxide-free air was supplied under positive pressure through the water in each unit and the effluent air passed through an ethylene glycol and two 2M potassium hydroxide traps, to trap any volatile products and liberated carbon dioxide (14CO2) respectively.

A good recovery of radioactivity was obtained for both systems with an overall mean recovery of 110.0 % (range 114.0 - 103.7 %) for system 96/03 (Manningtree). And 105. 5% (range 112.6 - 87.4 %) for system 96/04 (Ongar).

There was a steady transfer of the radioactivity from the water to the sediment in both systems, commencing at 111.29 % in the water phase at zero time and declining to 15.43 % (Manningtree system) and 1.80 % (Ongar system) after 100 days.

Extractable residues rose from 0 % at the first time-point (no extraction was made) to a maximum of 13.48 % at 14 days (Manningtree system). The Ongar system reached a maximum extracted residue level of 6.64 % after 14 days. Unextactable residues in both systems fluctuated with time from 0.77 % (Manningtree system) and 2.80 % (Ongar system) at time zero to 27.98 % (Manningtree system) at 100 days and 39.67 % (Ongar system) at 61 days.

Volatiles characterised as carbon dioxide accounted for 55 % (range 79 - 31 %) (Manningtree system) and 58 % (range 42 - 73 %) (Ongar system) of the applied dose after 100 days. Volatile materials, trapped in ethylene glycol, accounted for a negligible fraction (< 0.1 %) of the applied dose after 100 days. The test material was found to be the main component present in both the water and sediment phases. In addition significant mineralisation to carbon dioxide occurred. Some minor metabolites were observed in both the sediment and water phases, none of which exceeded 10 % of applied radioactivity in the total system.

Manningtree sediment: Water phase

DT50 = 49.23 days

DT90 = 155.47 days

Manningtree sediment: Complete system

DT50 = 60.66 days

DT90 = 175.23 days

Ongar sediment: Water phase

DT50 = 24.25 days

DT90 = 33.27 days

Ongar sediment: Complete phase

DT50 = 22.97 days

DT90 = 26.42 days

Under the conditions of the study the test material is readily degraded in the aerobic aquatic systems and is therefore unlikely to persist in the aquatic environment.

Supporting Study: Buser & Müller (1998)

The degradation of the test material in surface water was determined in surface water collected from five lakes and two rivers situated in the central and eastern region of Switzerland. The study was awarded a reliability score of 2 in accordance with the criteria set forth by Klimisch et al. (1997).

The water from Sempachersee was incubated in 1 L green glass bottles and incubated for up to 129 d at room temperature (20 - 23 °C). Periodically, one of the bottles was removed from storage, and the water was analysed.

The water from the river Aabach was incubated in one batch in a 2.5-L clear glass bottle for up to 50 d at room temperature while being stirred with a Teflon bar. Periodically, 0.25 L subsamples were removed and analysed. From the amounts of oxygen dissolved and the relatively small amounts of dissolved organic matter, it can be assumed that the incubation conditions in both experiments were aerobic.

There was some algal growth observed in the clear glass bottle.

Additionally, incubation experiments were carried out with fortified water, using water from Sempachersee and the river Aabach. One 2.5-L batch of each were fortified at 50 - 100 ng/L with the test material.

The waters were incubated at room temperature in amber bottles for up to 89 d. Periodically, 0.25 L subsamples were removed and analysed. The fortifications were made by adding 1.0 mL of a 250 ng/ mL solution of the compounds in distilled water, prepared from stock solutions of the compounds in methanol.

Extraction from the acidified (pH ≈ 2) water was effected with a reusable small column containing a macroporous polystyrene adsorbent, and the analytes were eluted and recovered. After methylation with diazomethane, the extracts were concentrated (≈0.5 mL in diethyl ether) and passed through a small silica column (0.7 g of silica gel 60, deactivated with 5 % water; 5 mm i.d. Pasteur pipet) topped with 10 mm of sodium sulfate. The analytes (as methyl esters) were eluted with 10 mL ofn-hexane-methylene chloride (1:1). After careful concentration and dilution to 100 - 200μL withn-hexane, 2 μL aliquots were used for analysis by high-resolution gas chromatography/mass spectrometry (HRGC/MS).

Under the conditions of the study the concentrations of the test material were reduced during the 50 - 129 d of incubation in both experiments. The data fitted approximate first-order kinetics without a lag phase.

The data indicate the test material is relatively stable in the aquatic environment whereas it degrades relatively quickly in soil (half-life, 7 - 22 d). The data for the test material show an approximately constant concentration (55 ng/L; lag phase) up to 21 d, followed by a gradual decrease to ≈ 11 ng/L at 89 d, the largest decrease occurring between 21 and 63 d.

Supporting Study: Hazlerigg & Garrett (2014)

The report covers the re-analysis of data from a previous study (Cooper & Unsworth, 1996) of the behaviour of the test material in water-sediment systems, according to the latest guidelines, in order to provide the most appropriate data for triggers for further studies and for inputs into environmental models of pesticide fate, specifically for use in FOCUS STEPS 1-4 for surface water modelling (FOCUS 2001, 2012). The study was awarded a reliability score of 4 in accordance with the criteria set forth by Klimisch et al. (1997).

FOCUS (2006) guidance on kinetics was used to evaluate two water-sediment datasets, Manningtree and Ongar. Two ‘Levels’ of analysis were used: Level P1, looking at a single compartment model and Level P2, looking at a two-compartment model with transfer between the water and sediment and vice versa.

The kinetic analysis conducted shows some differences to that originally performed. The values for Manningtree for both water and the whole system are similar to those previously reported. However, the values for Ongar for both water and the whole system are higher than those originally reported. The cause of the differences is likely due to the preparation of the data for analysis (exclusion of outliers) and the use of an SFO kinetics model rather than the three-compartment decay model used in the original analysis. This results in conservative degradation kinetic values for use in environmental fate modelling.

The data did not allow a firm relationship to be built-up between the partitioning of test material to the sediment and degradation in the water phase in a two-compartment analysis. This was primarily a function of limited back-transfer from the sediment back into the water column. In practice, this means that whole system DegT50 must be used in FOCUS STEP 3 modelling for one compartment (water) and the default DT50 of 1000 days used for the other compartment (sediment). This allocation is driven by the KOC of the test material as recommended by FOCUS (2012) guidance.

Read-Across Substance: Harrison et al. (2003)

The natural attenuation of test material was studied by determining changes in enantiomeric fraction in different redox environments down gradient from a landfill in the Lincolnshire limestone. Such changes could be due to differential metabolism of the enantiomers, or enantiomeric inversion. In order to confirm the processes occurring in the field, microcosm experiments were undertaken using limestone acclimatised in different redox zones. The study was awarded a reliability score of 2 in accordance with the criteria set forth by Klimisch et al. (1997).

No biodegradation was observed in the methanogenic, sulphate-reducing or iron-reducing microcosms. In the nitrate-reducing microcosm S-isomer did not degrade but R-isomer degraded with zero order kinetics at 0.65 mg/L/day to produce a stoichiometric equivalent amount of 4-chloro-2-methylphenol. This metabolite only degraded when the R-isomer disappeared. In aerobic conditions S- and R-isomer degraded with zero order kinetics at rates of 1.90 and 1.32 mg/L/day, respectively.

The addition of nitrate to dormant iron-reducing microcosms devoid of nitrate stimulated anaerobic degradation of R-isomer after a lag period of about 20 days and was associated with the production of 4-chloro-2-methylphenol. Nitrate addition to sulphate-reducing/methanogenic microcosms did not stimulate  test material degradation. However, the added nitrate was completely utilised in oxidising sulphide to sulphate. There was no evidence for enantiomeric inversion.

The study reveals new evidence for fast enantioselective degradation of R-isomer under nitrate-reducing conditions.

Read-Across Substance: Williams et al. (2003)

Test material degradation in different redox environments was tested within a Lincolnshire Limestone aquifer. The objective of the study was to assess whether changes in the enantiomeric fraction can be used to indicate natural attenuation in the field. The study was awarded a reliability score of 2 in accordance with the criteria set forth by Klimisch et al. (1997). The study involved sampling groundwater and relating systematic changes in enantiomeric fraction to different redox environments down gradient of the landfill. Laboratory microcosms were set up with groundwater and limestone acclimatised with the in situ microbial consortium to confirm the behaviour of the test material enantiomers under controlled conditions.

This study has shown that systematic change in the enantiomeric fraction of racemic test material deposited in the landfills at Helpston is a useful indicator of in situ processes and provides insight into the relative behaviour of the enantiomers under different redox conditions. Test material behaviour inferred from field data is supported by the results of microcosm experiments conducted under aerobic and anaerobic conditions. No degradation occurred in methanogenic, sulphate-reducing or iron-reducing microcosms. In nitrate­reducing microcosms, S-isomer did not degrade, but R-isomer degraded with zero-order kinetics at 0.65 mg/L/day. The associated buildup of 4-chloro-2-methyl­phenol indicated cleavage of the propionic moiety at the ether linkage, and this metabolite only degraded after the R-isomer had disappeared. In aerobic microcosms, the S-isomer degraded faster than the R-isomer both following zero-order kinetics 1.32 and 1.9 mg/L/day. The nature of the microbial community responsible for test material degradation is currently under study.

Read-Across Substance: Larsen et al. (2001)

The potential of a riparian fen to mineralise herbicides that could leach from an adjacent catchment area was investigated. The study was awarded a reliability score of 2 in accordance with the criteria set forth by Klimisch et al. (1997). Slurries were prepared from sediment and ground water collected from different parts of a wetland representing different redox conditions. The slurries were amended with O2, NO3^-, SO4^2-, and CO2, or CO2 alone as electron acceptors to simulate the in situ conditions and their ability to mineralise the herbicide test material. In addition, the abundance of bacteria able to utilise O2, NO3^-, SO4^2- + CO2 and CO2 as electron acceptors was investigated along with the O2–reducing and methanogenic potential of the sediment. The study was awarded a reliability score of 2 in accordance with the criteria set forth by Klimisch et al. (1997).

Study findings revealed that aerobic bacterial abundance was one order of magnitude greater in the aerobic sediment (S0: 5.8 x 10^6 CFU) than in the denitrifying sediment (S5: 6.4 x 10^5 CFU). The abundance of denitrifying bacteria was lower than that of aerobic bacteria under both aerobic (S0) and denitrifying (S5) conditions. The number of bacteria using the prevalent electron acceptors was lower in the sulfate-reducing (S9-I) and methanogenic (S9-II) sediment han in the aerobic (S0) and denitrifying (S5) sediment. Both sulfate-reducing and methanogenic bacteria were more abundant in the methanogenic sediment (S9-II) than in the sulfate-reducing sediment (S9-I). However, the potential rates of methanogenesis did not differ significantly between these two sediments.

Under the conditions of the study, the test material was readily mineralised and was mineralised under both aerobic and anaerobic conditions. Under aerobic conditions (S0 slurries) mineralisation started immediately, and 36 % of the test material had mineralised by the end of the experiment, 473 d later. Under anaerobic conditions (S5 slurries), in contrast, only 8 % of the test material mineralised during the course of the experiment. Similarly low rates of mineralisation were obtained under both sulfate-reducing (S9-I: 10 %) and methanogenic conditions (S9-II; 13 %). Moreover, mineralisation of the test material in both S9-I and S9-II slurries was independent of the presence of added sulfate since the omission of sulfate from S9-I slurries and the addition of sulfate to S9-II slurries did not significantly affect the mineralisation of the test material. Also, there was no significant difference between the amounts of accumulated 14CO2 from test material in anaerobic slurries amended with nitrate and anaerobic slurries with sulfate or without additional electron acceptor.