4-(1,1,3,3-tetramethylbutyl)phenol

Adsorption / desorption

Link to relevant study record(s)

Reference

Reference 1

Endpoint:

adsorption / desorption: screening

Type of information:

experimental study

Adequacy of study:

key study

Study period:

1990

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

guideline study with acceptable restrictions

Qualifier:

according to guideline

Guideline:

EPA OTS 796.2750 (Sediment and Soil Adsorption Isotherm)

Deviations:

yes

Remarks:

deviations (soil processing, use of carrier solvent) had no impact on the study results

GLP compliance:

yes

Type of method:

batch equilibrium method

Media:

soil

Specific details on test material used for the study:

Details on properties of test surrogate or analogue material (migrated information):
No details reported

Radiolabelling:

no

Test temperature:

21.8 - 21.9 °C

Details on study design: HPLC method:

not applicable

Analytical monitoring:

yes

Details on sampling:

WATER SAMPLES:
- 1 ml (equilibrium experiment) or 2 ml (isotherm experiment) samples of supernatant were filtered through 0.2 µm teflon filters, diluted 1:1 with acetonitrile, diluted (if necessary), and analyzed by HPLC
- no extraction step was used due to the small sample size

SOIL SAMPLES:
- Soil samples (2 g) were extracted with 4 ml of acetonitrile in glass vials utilizing manual agitation for two minutes per sample.
- The resulting extract was filtered through a 0.2 µm teflon filter, diluted (if necessary), and analyzed by HPLC using the instrument conditions noted below

Details on matrix:

COLLECTION AND STORAGE
- Geographic location: soil #1: collected from lawn of the Testing Laboratory in Lionville, PA
soil #2: collected from residential lawn in Downingtown, PA
soil #3: collected from the flood plain of French Creek in Phoenixville, PA
- Collection procedures: shovel
- Storage conditions: at field moisture content in glass jars with teflon-lined lids at approx. 4 °C
- Soil preparation: soil aliquots were air dried overnight, ground with a motar and pestle, and sieved through a 100 mesh brass soil sieve; all soil were autoclaved in the test vessel before experiments were conducted


PROPERTIES
- Soil texture
soil #1: - % sand: 36
- % gravel: 3
soil #2: - % sand: 25
- % gravel: 4
soil #3: - % sand: 19
- Soil taxonomic classification: soil #1 + #2 + #3: ML or CL
- Soil classification system: USCS
- pH: soil #1: 7.1
soil #2: 7.3
soil #3: 6.4
- Organic carbon (%): soil #1: 0.82
soil #2: 10.2
soil #3: 8.6
- CEC (meq/100 g): soil #1: 28.4
soil #2: 46.2
soil #3: 24.6
- exchangeable bases (meq/100 g): soil #1: 27.8
soil #2: 45.8
soil #3: 17.2
- exchangeable acids (meq/100 g): soil #1: 0.60
soil #2: 0.40
soil #3: 7.4
- Natural moisture content (%), dry basis: soil #1: 16.8
soil #2: 44.6
soil #3: 58.7

Details on test conditions:

TEST CONDITIONS
1) Determination of Equilibrium Time:
- a stock solution (1 µg/ml) was prepared in autoclaved 0.01 M Ca(NO3)2 from a 4-nonylphenol stock solution of 34.49 mg/ml
- 10 ml of stock solution was added to 2 g of the designated soil type in the test vessel
- vessels were agitated on wrist-action shakers for 8 h and then centrifuged
- centrate samples were analyzed by HPLC
- the concentration of the test compound was to be below the minimum detectable level, and as a result, the experiment was terminated
- a second stock solution 10 µg/ml was prepared in autoclaved 0.01 M Ca(NO3)2 from a 4-nonylphenol stock solution of 34.49 mg/ml
- 10 ml of stock solution was added to 2 g of the designated soil type in the tets vessel
- vessels were agitated on wrist-action shakers, vessels were sampled on days 1, 2, 3, 4, 5, and 6.
- vessels were centrifuged and a known volume of centrate (1.0 ml) was sampled from each vessel and analyzed by HPLC
- an equilibration time of 3 days was determined to be adequate for all three soil types

2) First Adsorption Isotherm Experiment
- six test solutions (10, 20, 40, 60, 80, and 100 mg/ml) were prepared as dilutions of the 34.49 mg/ml stock solution in autoclaved 0.01 M Ca(NO3)2
- triplicate test vessels were filled with 2 g of the designated soil type and 10 ml of test solution
- triplicate controls (no soil) and blanks (no test substance) were also prepared for each soil type and test concentration
- all test vessels were placed on wrist-action shakers at a speed setting of 3.5
- test vessels were sampled on day 3 and analyzed by HPLC
- analytical data from this experiment were highly variable and did not correlate to nominal test concentrations; the problem was traced in the spiking procedure; it was surmised that variable amounts of the test substance were transferred to the test vessels because of the limited solubility of 4-nonylphenol in aqueous solutions

3) Second Adsorption Isotherm Experiment
- six stock solutions of the test substance were prepared in methanol at nominal concentrations of 25, 50, 100, 150, 200, and 250 mg/ml (stock solutions, measured concentrations: 22.1, 46.9 95.5, 134, 183, and 233 mg/ml)
- triplicate test vessels were filled with 2 g of the designated soil type and 10 ml of autoclaved test solution
- vessels were spiked with 4 µl of the appropriate stock solution to yield nominal test concentrations of 10, 20, 40, 60, 80, and 100 mg/ml, respectively
- single controls (no soil) and blanks (no test substance) were also prepared for each soil type and test concentration
- all test vessels were placed on wrist-action shakers at approx. 110 rpm for 3 days and analyzed by HPLC afterwards

Sample No.:

#1

Duration:

3 d

Initial conc. measured:

10 other: µg/ml

pH:

7.1

Sample No.:

#2

Duration:

3 d

Initial conc. measured:

10 other: µg/ml

pH:

7.3

Sample No.:

#3

Duration:

3 d

Initial conc. measured:

10 other: µg/ml

pH:

6.4

Computational methods:

- Equilibrium time experiment: linear plots of aqueous 4-nonylphenol concentration as function of time were prepared
- Freundlich adsorption coefficients (K): * soil #1: 4009
* soil #2: 2301
* soil #3: 5164
- Soil partitioning coefficient: * soil #1: 3.603
* soil #2: 3.362
* soil #3: 1.348
- Slope of Freundlich adsorption isotherms: * soil #1: 0.738
* soil #2: 1.004
* soil #3: 0.742
- Linear correlation coefficient (R²) of Freundlich isotherm: * soil #1: 0.972
* soil #2: 0.920
* soil #3: 0.973
- Freundlich isotherm plots were produced by plotting log(mass of test substance/mass of soil in test vessel) as a function of log(equilibrium concentration of the test substance in the aqueous phase). The slope and the y-intercept yielded the constant 1/n and K (Freundlich adsorption coefficient), respectively. When the constant 1/n is numerically equal to 1, the Freundlich adsorption coefficient, K, becomes equal to the soil partition coeffiecient, Kd.

Type:

Koc

Value:

>= 23 000 - <= 489 000

Temp.:

22 °C

% Org. carbon:

>= 0.82 - <= 10.2

Type:

log Koc

Value:

>= 4.36 - <= 5.69

Temp.:

22 °C

% Org. carbon:

>= 0.82 - <= 10.2

Details on results (HPLC method):

not applicable

Adsorption and desorption constants:

Freundlich adsorption constant: * soil #1: 4009
* soil #2: 2301
* soil #3: 5164

Recovery of test material:

- Uncorrected for extraction efficiency: 55 to 77 %
- Corrected for extraction efficiency: 61 to 97 %

Concentration of test substance at end of adsorption equilibration period:

Concentration in liquid phase after 3 d (initial concentration):
- soil #1: 0.0696 µg/ml
- soil #2: 0.4879 µg/ml
- soil #3: 0.0019 µg/ml

Concentration of test substance at end of desorption equilibration period:

not applicable

Sample no.:

#1

Duration:

3 d

% Adsorption:

ca. 99

Sample no.:

#2

Duration:

3 d

% Adsorption:

ca. 99

Sample no.:

#3

Duration:

3 d

% Adsorption:

ca. 99

Transformation products:

not measured

Details on results (Batch equilibrium method):

ANALOGUE APPROACH JUSTIFICATION:
NP and PTOP are not multi-functional compounds. Both substances have only one functional group which is in both substances the same (-OH). Therefore, the reliability of a read-across is not negatively influenced.

Evaluation of the purity and impurity profiles:
1) Target chemical (PTOP):
The target chemical PTOP is analytical pure (99.2 %). Therefore, no impurities have to be considered which might have influence on the overall adsorption/desorption behaviour of PTOP.
2) Source chemical (NP):
Relevant impurities are 2-Nonylphenol (5 % w/w), and 2,4-Dinonylphenol (5 % w/w). As 2-Nonylphenol has the same molecular formula and molecular weight and a similar structure to 4-Nonylphenol, physico-chemical properties, and therefore also an adsorption/desorption behaviour similar to that of 4-Nonylphenol can be predicted. Thus, the contamination of the test substance with 2-nonylphenol can be neglected with regards to adsorption/desorption. 2,4-Dinonylphenol has a higher water solubility (20 mg/L at 20 °C; source: IFA GESTIS) than 4-Nonylphenol but can still be classified as slightly soluble. Because of the chemical structure of 2,4-Dinonylphenol a potential for adsorption can be assumed. Therefore, 2,4-Dinonylphenol has probably a negligible impact on the adsorption/desorption behaviour of 4-Nonylphenol.

Comparison of the physico-chemical properties of the target (PTOP) and source chemical (NP):
Apart from the physical form and melting point, NP and PTOP have similar properties with regards to molecular formula and weight, water solubility (both have very low water solubility), boiling point, relative density, vapour pressure, and n-octanol/water partition coefficient (see Table 1). In conclusion PTOP and NP are in very good agreement with regards to their physico-chemical properties and should therefore show similar behaviour in the environment.
Both substances have a log Kow > 3 which indicate a high potential for adsorption to soil and sediment. For NP this assumption is confirmed by the Weston study which concludes that “4-nonylphenol may be expected to adsorb strongly to soils and sediments in the environment”. PTOP has a only slightly lower log Kow of 4.12 than NP (log Kow = 4.48). Thus, it can be assumed that PTOP will also strongly tend to adsorb to soils and sediments.



MAIN TEST: PERFORMANCE
- Test material stability during adsorption phase: The stability of the test substance has been determined by Roy F. Weston Inc. in study report no. 90-095
- Experimental conditions maintained throughout the study: Yes
- Buffer/test substance interactions affecting sorption: no
- Further chemical interactions: no


TRANSFORMATION PRODUCTS: not measured


JUSTIFICATION FOR ANALOGUE APPROACH (Read-across from 4-nonylphenol (NP), branched, to 4-tert-octylphenol (PTOP)):
- The data point of the source chemical is relevant and reliable for the purpose of the read-across.
- The source and target chemical is not a multi-functional compound. Both substances have only one functional group which is in both substances the same (-OH). Therefore, the reliability of a read-across is not negatively influenced.
- Comparison of the physico-chemical properties of the target and source chemicals, particularly the physical form, molecular weight, water solubility, particle size and structure, partition coefficient and vapour pressure, provides information to their adequate similarity. Apart from the physical form and melting point, NP and PTOP have similar properties with regards to molecular formula and weight, water solubility (both have very low water solubility), boiling point, relative density, vapour pressure, and n-octanol/water partition coefficient. In conclusion PTOP and NP are in very good agreement with regards to their physico-chemical properties and should therefore show similar behaviour in the environment.
Both substances have a log Kow > 3 which indicate a high potential for adsorption to soil and sediment. For NP this assumption is confirmed by the Weston study which concludes that “4-nonylphenol may be expected to adsorb strongly to soils and sediments in the environment”. PTOP has a only slightly lower log Kow of 4.12 than NP (log Kow = 4.48). Thus, it can be assumed that PTOP will also strongly tend to adsorb to soils and sediments.

Statistics:

No details reported

The Freundlich adsorption coefficients, K, were 4009, 2301, and 5164 in soil #1, soil #2, and soil #3, respectively.  The Freundlich constant 1/n was numerically equal to 1 for soil #2. In this case, the Freundlich adsorption coefficient, K, becomes equal to the soil partition coeffiecient, Kd. The adsorption Koc value (calculated from Kd) is approx. 23000.

Validity criteria fulfilled:

yes

Conclusions:

The results of this study indicate that 4-nonylphenol may be expected to adsorb strongly to soils and sediments in the environment. Because of the similarity with regards to physico-chemical properties between 4-nonylphenol and 4-tert-octylphenol it can be concluded that 4-tert-octylphenol may also be expected to adsorb strongly to soils and sediments.

Executive summary:

The adsorption characteristics of 4-Nonylphenol, branched, was studied in three soils (USA) of light brown/brown gravelly sandy silt or clay at a pH of 6.4 to 7.1, and organic carbon contents of 0.82 %, 10.2 %, and 8.6 %, respectively, in a batch equilibrium experiment.  The experiment was conducted in accordance with the EPA OTS 796.2750 (Sediment and Soil Adsorption Isotherm), and in compliance with the USEPA TSCA GLP standard. 

 

The adsorption phase of the study was carried out by equilibrating 2 g air-dried soil with 10 ml of autoclaved test solution (4-Nonylphenol in methanol (solvent vehicle)) at nominal concentrations of 10, 20, 40, 60, 80, and 100 mg/ml 4 -nonylphenol at approx. 22 ºC for 3 days.  The equilibrating solution used was 0.01 M Ca(NO₂)₂, with a soil/solution ratio of 2 g/10 ml.  

 

The supernatant solution after adsorption was separated by centrifugation. The test substance was subsequently quantified in samples of supernatant and soil. Soil samples were extracted with 4 ml of acetonitrile. Extract and supernatant samples were analyzed by HPLC. The adsorption parameters were calculated using the Freundlich adsorption isotherm.

 

After 3 days of equilibration, 99 % of the applied 4-nonylphenol, branched, was adsorbed in all soil types. The Freundlich adsorption coefficients, K, were 4009, 2301, and 5164 in soil #1, soil #2, and soil #3, respectively.  The calculated Kd values were 489,000, 23,000, and 60,000 for soil #1, #2, and #3, respectively. The adsorption Koc value (calculated from Kd) were in the range form 4.35 to 5.69.

 

The results of this study indicate that 4-nonylphenol may be expected to adsorb strongly to soils and sediments in the environment.

 

This study is classified acceptable and satisfies the guideline requirement for an adsorption study in soil.

 

 

Rational for read-across:

4 -nonylphenol and 4 -tert-octylphenol have, apart from the physical form and melting point, similar properties with regards to molecular formula and weight, water solubility (both have very low water solubility), boiling point, relative density, vapour pressure, and n-octanol/water partition coefficient. In conclusion 4 -nonylphenol and 4 -tert-octylphenol are in very good agreement with regards to their physico-chemical properties and should therefore show similar behaviour in the environment.

Both substances have a log Kow > 3 which indicate a high potential for adsorption to soil and sediment. For 4 -nonylphenol this assumption is confirmed by the results of this study which concludes that “4- nonylphenol may be expected to adsorb strongly to soils and sediments in the environment”. 4 -tert-octylphenol has a only slightly lower log Kow of 4.12 than 4 -nonylphenol (log Kow = 4.48). Thus, it can be assumed that 4 -tert-octylphenol will also strongly tend to adsorb to soils and sediments.

Reference 2

Endpoint:

adsorption / desorption: screening

Type of information:

read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

weight of evidence

Justification for type of information:

In accordance with Regulation (EC) 1907/2006 Annex XI (1.5) and the relevant ECHA guidance documents, the substances detailed in the table below are grouped for the purposes of read across to reduce the need for unnecessary repeat testing on the basis that the substances are similar on the basis of a common functional groups.

Reason / purpose for cross-reference:

read-across source

Reason / purpose for cross-reference:

read-across source

Type:

log Koc

Value:

4.36 - <= 5.69

Temp.:

22 °C

% Org. carbon:

>= 0.82 - <= 10.2

Type:

Koc

Value:

>= 23 000 - <= 489 000

Temp.:

22 °C

% Org. carbon:

>= 0.82 - <= 10.2

Type:

log Koc

Value:

>= 3.54 - <= 4.26

Type:

Koc

Value:

>= 3 500 - <= 18 000

Conclusions:

The read across for 4-tert-octylphenol (CAS: 140-66-9); is based upon the analogous substances to which basic form, degree of substitution of functional groups is not considered to effect the proposed read across for the endpoint of adsorption/desorption. Based on the information available for the read across substances, the substance is expected to adsorb strongly onto organic matter.

Reference 3

Endpoint:

adsorption / desorption

Remarks:

other: literature review

Type of information:

other: literature review

Adequacy of study:

supporting study

Reliability:

4 (not assignable)

Rationale for reliability incl. deficiencies:

other: Secondary literature

Qualifier:

no guideline followed

Principles of method if other than guideline:

literature review

GLP compliance:

no

Media:

sewage sludge

Type:

Koc

Value:

>= 3 500 - <= 18 000

Type:

log Koc

Value:

>= 3.54 - <= 4.26

Validity criteria fulfilled:

not applicable

Conclusions:

Octylphenol strongly adsorbs to soil and sediments.

Executive summary:

"The alkylphenols (AP) and lower mole alkylphenol ethoxylates (APE) have low water solubilities, and thus tend to adsorb strongly onto organic matter. Adsorption is likely to play an important role in the removal of AP and to some extent APE in wastewater treatment plants and as a sequestration process in soil, sediment and sludge. Adsorption to solids is generally expressed as an organic-carbon (OC) normalized partition coeffcient, Koc, which is the ratio of the concentration in organic fraction of solids divided by concentration in water. For octylphenol (OP), Koc values reported from laboratory studies vary widely from about 3,500 to 18,000 L/kg (UKEA 2005)."

Reference 4

Endpoint:

adsorption / desorption: screening

Type of information:

experimental study

Adequacy of study:

key study

Study period:

1996-1997

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: Meets generally accepted scientific standards, well documented and acceptable for assessment.

Qualifier:

equivalent or similar to guideline

Guideline:

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

GLP compliance:

not specified

Type of method:

batch equilibrium method

Media:

sediment

Specific details on test material used for the study:

Details on properties of test surrogate or analogue material (migrated information):
PHYSICO-CHEMICAL PROPERTIES
[*) Properties that should be provided for the HPLC method]
- Water solubility*: 12 mg/l
- log Pow*: 4.12

Radiolabelling:

yes

Test temperature:

Room temperature (20-22 °C)

Analytical monitoring:

yes

Details on sampling:

1) Measuring sorption kinetics to bed-sediment and suspended sediment in a shaking batch system
- Samples were shaken at 120 rev./min for 2 min, 20 min, 4.5 h, 24 h and 48 h at room temperature (20-22 °C).
- At each sampling time, three replicate tubes were removed for sacrificial sampling.
- The procedure involved each replicate tube being sampled three times by removing 700 µl into 700 µl methanol with a syringe.
- This was then added to a further syringe connected to a 0.45-pm PTFE filter to remove the solids.
- The sample was filtered directly into a scintillation vial and mixed with 4.5 ml of Ultima Gold scintillant (Canberra Packard).
- After desorption the tubes were centrifuged again, and the aqueous phase sampled again in the presence of methanol and passed through a 0.45 µm filter into a scintillation vial and counted as described before.



2) Measuring sorption kinetics to bed-sediments in a static batch system
- The flasks were then gently agitated at 60 rev./min for 10 min on an orbital shaker to help disperse the [14C]octylphenol throughout the suspension.
- The flasks were then incubated without further agitation at 20-22 °C and sampled at 10 min, 30 min, 2 h, 5.5 h, 18 h, and 24 h by removing triplicate samples of 700 µl into syringes containing 700 µl of methanol and filtering via 0.45-µm PTFE filter into scintillation vials prior to counting.


3) Measuring equilibrium distribution coefficients for bed-sediments and suspended sediments
- Samples were shaken at 120 rev /min for 24 h at room temperature (20-22 °C).
- The tubes were then centrifuged for 30 min at 4749 x g. Each replicate tube was then sampled using the technique described previously.
- A second equilibrium distribution coefficient (desorption) was measured for the R. Aire (6 September 1996), R. Caliler (6 September 1996), R. Thames, Wallingford bridge (6 September 1996), and R. Thames, Wallingford slipway (8 September 1996) samples.
- In this case, after centrifugation the supernatant was discarded by pipette and replaced by fresh filtered river water.
- The samples were then shaken again for 24 h and sampled as described above.
- The sampling method was also as described previously.

Details on matrix:

Details on collection (e.g. location, depth, contamination history, procedure:

- River water and sediment material (both in suspension and from the bed) were collected from two rivers: the Aire and Calder, which run through
urban industrial landscapes. High concentrations of alkylphenols have been previously reported in these rivers including nonylphenol in river waters and bed/suspended sediments. It could be demonstrated that a long stretch of the R. Aire in 1994 was very oestrogenic to fish. River water and sediment was also collected from the River Thames, 54 km downstream of Oxford. This reach of the Thames does not have the same history of contamination by micro-organic xenobiotic molecules as the Aire and Calder.
- River water was collected 2-5 cm below the surface in 1-1 polypropylene flasks at a distance of about 2 m from the river banks.
- Collections were made from the R. Aire (6 September 1996 and 18 December 1996), NGK SE379288, 8 km downstream of Leeds, and H . Calder (6 September 1996 and 16 January 1997) NGR SE405207, 16 km downstream of Wakefield.
- River samples were returned to the laboratory and pH, dissolved organic carbon and sediment load measured prior to storage in the dark at 4°C for up to 1 month prior to use in experiments.
- Bed-sediment samples were also collected from the same locations, on the same dates, by skimming-off the top 2-5 cm of bed-material with wide neck 1-1 flasks.
- The bed-sediment samples were allowed to air dry before being sieved (2 mm) and mixed.
- The samples were stored at room temperature prior to use.
- Bed-sediments were collected and stored in the same way from the R. Thames near Wallingford bridge, NGR SU610895 (6 September 1996) and from a nearby slipway, NGR SU614903 (8 September 1996), from the river bottom midchannel (using a weighted metal collecting tube, 5 December 1996) and from inside a nearby boathouse on 15 April 1997.
- River water was collected from the same location on 4 June 1997.

Details on test conditions:

1) Measuring sorption kinetics to bed-sediment and suspended sediment in a shaking batch system
- 5 g of bed-sediments (R. Calder, 6 September 1996 and River Thames, 8 September 1996), prepared as described above, were added to 45-ml PTFE screw-top centrifuge tubes (Nalgene).
- River water from the same origin as the bed-sediments was prepared by 0.2 pm filtration (thus removing all suspended sediments).
- Fifteen ml were added to the bed-sediments.
- [14C]Octylphenol (3.75 µl) was added from a methanol stock solution to triplicate samples to give a final concentration of 50 µg/I.
- The [14C]octylphenol was ring-labelled, with a specific activity of 3.23 MBq/mg and a purity of 98% (Amersham International).
- The same quantity of [14C]octylphenol was added to centrifuge tubes containing ultra-pure water only, to act as a standard.
- The centrifuge tubes were laid on their sides on an orbital shaker and shaken at 120 rev./min for 2 min, 20 min, 4.5 h, 24 h and 48 h at room temperature (20-22°C).
- At each sampling time, three replicate tubes were removed for sacrificial sampling.
- The procedure involved each replicate tube being sampled three times by removing 700 µl into 700 µl methanol with a syringe.
- This was then added to a further syringe connected to a 0.45-pm PTFE filter to remove the solids.
- The sample was filtered directly into a scintillation vial and mixed with 4.5 ml of Ultima Gold scintillant (Canberra Packard).
- Each sample was then counted for 5 min with a Beckman LS 6500 instrument. The amount of radioactivity still present in the aqueous phase was compared with the sediment-free pure water controls and the amount sorbed calculated by difference.
- The tubes were then centrifuged for 30 min at 4749 x g and the supernatant discarded.
- Fifteen ml of fresh 0.2 µm filtered river water were then added to replace the discarded supernatant.
- The bed-sediments and fresh water were then shaken for a further 20 min to allow the [14C]octylphenol still sorbed to the bed-sediments to re-partition into the new octylphenol-free aqueous phase.
- The tubes were then centrifuged again, and the aqueous phase sampled again in the presence of methanol and passed through a 0.45 µm filter into a scintillation vial and counted as described before.
- To measure the sorption kinetics to suspended sediment, river water was collected from the R. Aire on 18 December 1996.
- The sample was placed on a stirrer with magnetic flea and stirred to ensure the suspended sediments remained suspended.
- Fifteen ml were then removed by pipette and added to three 45-ml PTFE centrifuge tubes.
- [14C]Octylphenol was then added to give a final concentration of 50 µg/I to triplicate tubes as described above.
- A standard was also made up to 50 µg/I in pure water.
- The tubes were placed on an orbital shaker at 120 rev./min and incubated at room temperature and samples were taken after 1 h and 18 h, using the same method as with the bed-sediments.


2) Measuring sorption kinetics to bed-sediments in a static batch system
- Using triplicate 150-ml PTFE conical flasks for each bed-sediment, 10 g R. Calder (6 September 1996) bed-sediment was overlain with 50 ml of 0.2 µm filtered river water (from the same river as the sediment).
- To each flask, together with a standard containing 50 ml ultra-pure water, 250 µl of [14C]octylphenol were added from a 10000 µg/l aqueous stock solution to give a final concentration of 50 µg/I.
- The flasks were then gently agitated at 60 rev. /min for 10 min on an orbital shaker to help disperse the [14C]octylphenol throughout the suspension.
- Earlier trials, adding 250 µl of the tie acid red 1 (Sigma Chemicals) to 50 ml, indicated a reasonable dispersion of a solute through the suspension under these conditions.
- The flasks were then incubated without further agitation at 20-22°C and sampled at 10 min, 30 min, 2 h, 5.5 h, 18 h, and 24 h by removing triplicate samples of 700 µl into syringes containing 700 µl of methanol and filtering via 0.45-µm PTFE filter into scintillation vials prior to counting.


3) Measuring equilibrium distribution coefficients for bed-sediments and suspended sediments
- 5 g of bed-sediments, prepared as described above, were added to 45-ml PTFE centrifuge tubes.
- Fifteen ml of 0.2-pm filtered river water, taken from the same location, were added to the bed-sediments.
- [14C]octylphenol (1.87, 3.75, and 7.5 µl) was added from the methanol stock solution to triplicate samples to give concentrations of 25, 50, and 100 µg/I, respectively.
- The same quantities of [14C]octylphenol were added to centrifuge tubes containing ultra-pure water only to act as standards.
- The centrifuge tubes were laid on their sides on an orbital shaker and shaken at 120 rev /min for 24 h at room temperature (20-22°C).
- The tubes were then centrifuged for 30 min at 4749 x g. Each replicate tube was then sampled using the technique described previously.
- The amount of radioactivity still present in the aqueous phase was compared with the sediment-free pure water controls and the amount sorbed calculated by difference.

- A second equilibrium distribution coefficient (desorption) was measured for the R. Aire (6 September 1996), R. Caliler (6 September 1996), R. Thames, Wallingford bridge (6 September 1996), and R. Thames, Wallingford slipway (8 September 1996) samples.
- In this case, after centrifugation the supernatant was discarded by pipette and replaced by fresh filtered river water.
- The samples were then shaken again for 24 h and sampled as described above.
- To measure equilibrium distribution coefficients for suspended sediments, the sample bottle containing the river water was placed on a stirrer with magnetic flea and stirred to ensure the suspended sediments remained suspended.
- Fifteen ml were then removed by pipette and added to 45-ml PTFE centrifuge tubes.
- [14C]Octylphenol was then added to give concentrations of 25, 50, and 100 µg/l to triplicate tubes as described previously.
- The sampling method was also as described previously. This experiment was followed for river water collected from the R. Aire on 6 September 1996 and on 18 December 1996, and the R. Calder (16 January 1097). This method was also used for R. Thames (4 June 1997) suspended sediments but when no OP sorption could be detected, it was decided to repeat the experiment using a higher concentration of suspended sediments. For 5 h, a high capacity peristaltic pump was used to feed water direct from the R. Thames (5 cm below the surface) into a continuous-flow centrifuge. The collected sediment was then scraped from the walls of the centrifuge and used after 4 days storage at 4 °C. In this case the experiment was repeated with the suspended sediment concentration increased to 40 times (from 20 to 800 mg/l') over the unconcentrated sample.
- The weight of suspended sediments was measured by filtering 300-400 ml through a dry pre-weighed 0.2-µm filter.
- The filter was then allowed to air dry and re-weighed.

Sample No.:

#1

Duration:

2 min

Initial conc. measured:

50 other: µg/l

Computational methods:

- Adsorption and desorption coefficients (Kd): details on calculation not reported

Type:

Kd

Value:

>= 6 - <= 700

Temp.:

22 °C

% Org. carbon:

>= 0.08 - <= 7.02

Remarks on result:

other: A wide variety of Kd values were obtained. Regression analysis showed a relationship between the quantity of OP sorbed and the amount of TOC in the sediments, as well as the quantity of different particles (size) present.

Type:

Koc

Value:

>= 3 500 - <= 18 500

Temp.:

22 °C

% Org. carbon:

>= 0.08 - <= 7.02

Adsorption and desorption constants:

For details see table 1

Recovery of test material:

Not applicable; the amount of radioactivity in the aqueous phase was compared with the sediment free pure water controls and the amount sorbed was calculated by difference.

Concentration of test substance at end of adsorption equilibration period:

Not applicable; concentration of test substance was not measured in solid and liquid phase; the amount of radioactivity in the aqueous phase was compared with the sediment free pure water controls and the amount sorbed was calculated by difference.

Concentration of test substance at end of desorption equilibration period:

Not applicable; concentration of test substance was not measured in solid and liquid phase; the amount of radioactivity in the aqueous phase was compared with the sediment free pure water controls and the amount sorbed was calculated by difference.

Transformation products:

not measured

Details on results (Batch equilibrium method):

1) Sorption kinetics to bed-sediments in the shaking batch system
- With the shaking batch system an equilibrium between OP in the liquid and solid phases is achieved within 2 min, for example, for the R. Calder (6 September 1996) and R. Thames (Wallingford slipway, 8 September 1996) bed-sediments. Therefore, 24h would be more than adequate to allow the establishment of equilibrium and distribution coefficient measurement.
- The experimental system did give wide variations between replicates on occasions. This difference was believed to be due to an increase in suspended (flocculated) clay particles in the solution over time.
- A variable quantity of these particles was sucked into the sampling syringe where bound OP could have become dissociated from the particles in the presence of methanol. This would give a higher apparent aqueous OP concentration. This problem was eliminated in subsequent experiments by centrifugation prior to filtration.
- With the R. Calder samples, the ratio of OP in the aqueous phase before and after the desorption step did appear to be influenced, to a small degree, with time of initial exposure to the sorbent, no such hysteresis effect was observed with the R. Thames samples.


2) Sorption kinetics to bed-sediment in a static batch system
- With the static batch system, much of the OP remained in the aqueous phase for the first 5.5 h. This feature was probably due to poor dispersion throughout the system.
- By 24 h the solid phase has adsorbed up to 50% of the OP (Kd of 10). The same distribution coefficient, 30-45 l/kg, which was obtained within 2 min within the shaking system was not reached with the static system after 24 h; it may take months for an organic compound to penetrate all the micropores and to diffuse throughout the organic matrices of soil aggregates.


3) Equilibrium distribution coefficients to, and their relationship with, bed-sediment characteristics
- The curves drawn to calculate K, for OP between the river water (suspended sediment-free) and bed-sediments were derived from the assumption that a linear isotherm exists at these low OP concentrations.
- Environmental concentrations would be unlikely to exceed 100 µg/l in UK rivers
- In some cases the regression was poor, and in these instances the result may have been influenced by experimental errors in adding the [14C]octylphenol to the samples, and also by sorbent heterogeneity. Variation due to sampling errors was reduced by taking the average of three for each replicate.

- A second distribution coefficient was calculated for some sediments in which the aqueous phase was replaced with fresh OP-free water (2nd equilibrium).
- Just as with the kinetic experiment, there was no evidence of hysteresis with the sample of R. Thames Wallingford slipway bed-sediments. The other bed-sediments did show evidence of hysteresis, most notably R. Are (6 September 1996) sample. This sample had a higher TOC content (2.4%) than
the others (0.08-0.1 2%). The hysteresis effect has been noted before with hydrophobic organic chemicals and bed-sediments. There is still no general agreement to explain the phenomena. Factors such as a much slower return to equilibrium due to retarded diffusion and tortuosity within the sorbent, low diffusion within the organic matter, or the sorbate becoming irreversibly bound to the sorbent have all been suggested to play a part.


4) Sorption kinetics to suspended sediment
- As the K, value did not increase with time, it would appear that equilibrium is achieved rapidly with the batch shaking system for OP and the suspended sediments.
- Compared to the 1-h reading an apparent slight reduction In proportion of OP sorbed after 18 h can be seen.
- The batch shaking method keeps the sediments in suspension and an equilibrium is quickly achieved.
- Whilst the experimental system cannot be considered a perfect representation of the natural one, it is more representative than the bed-sediment batch experiments, and may be considered adequate for comparative studies between different suspended sediment samples.


5) Equilibrium partition coefficients suspended sediments
- For the rivers Aire and Calder much higher Kd values of between 48000 and 226000 l/kg were obtained for the suspended sediments.
- For the amount of suspended sediment present (around 3 mg/l) this translates to 30-40% OP sorbed to the solid phase.
- Examination of suspended sediments from these rivers under the light microscope revealed the particles to resemble organic aggregates 100-600 µm in diameter. These were often associated with attached protozoa (20-50 µm) and bacteria. Little or no algae or inorganic minerals could be seen.
- Assuming that the suspended sediments were 100% particulate organic matter, and the organic matter comprises 58% organic carbon, then the Koc values are between 87 000 and 390000 l/kg.
- Therefore, the organic carbon of the Aire and Calder suspended sediments was between 8 and 38 times more effective at sorbing OP than that of their respective bed-sediments.
- As many of the suspended particles were quite large, up to 600 µm, the explanation for this may be more than simply a case of a large surface area/volume ratio.
- As can be seen for the two results for R. Aire suspended sediments, where the sediment collected on 18 December 1996 gave a K, four times larger than that collected on 6 September 1996, the quality of the suspended sediment sorbent can change with time for the same reach. Similar variability has been found for NP in the R. Aire.
- The suspended sediment collected from the R. Thames on 4 June 1997 appeared to consist largely of algae, when viewed through the microscope. These particles gave a Kd of 1900 l/kg, which was one or two orders of magnitude less than for the suspended particulate organic matter from the Aire and Calder.
- For the 20 mg/l suspended sediments, present in the Thames when the river was sampled, this translates to a potential to sorb only 1.5% of OP in the aqueous phase. Assuming the algae were 58% carbon, this would give a Koc, of 3281, much lower than that for the Aire and Calder suspended sediments and also lower than the value for most bed-sediments.

Statistics:

No details reported.

Table 1: Particle, caly mineralogy and total organic carbon (TOC) compositions by mass of the collected bed-sediments and their calculated Kd and Koc values

 

Characteristics	Aire (a)	Aire (b)	Calder (a)	Calder (b)	Thames W’ford Bridge	Thames W’ford slipway	Thames mid-channel	Thames W’ford boat house
% Clay (< 2 µm)	9.2	7.4	3.3	6.6	1.0	2.8	5.5	4.8
% Silt (2-63 µm)	72.9	62.8	19.3	55	8.3	12.1	38	43
% Sand (63-900 µm)	17.9	29.8	77	38.4	90.7	85.1	56	51.8
Kaolinite^a	50	20	50	25	10	10	10	10
Illite^a	30	65	40	65	10	10	10	10
Smectite^a	20	15	10	10	80	80	80	80
% TOC	2.42	7.02	0.12	5.71	0.09	0.08	1.76	2.88
Kd(l/kg)	259	707	11	582	6	15	61	112
Koc(l/kg)	10373	10071	10833	10193	6722	18500	3466	3958

  ^aSemi-quantitative assessment of < 2 µm fraction

Table 2: Regression analysis of log K_dagainst particle size and TOC

Characteristics	R²	Slope (x coefficient)	S.E. of slope (x coefficient)
Sand content	-0.87	-0.028	0.004
Silt content	0.87	0.031	0.005
Clay content	0.80	0.26	0.05
TOC content	0.86	0.28	0.05

 

The greater the proportion of clay (<2 µm), or silt (2 -63 µm) in the sediments, the greater the amount of sorption; similarly the- greater the proportion of sand particles (63 -900 µm)the less sorption. The R² value for the log Kd to clay regression (0.80) was poorer than the other (0.86 -0.87).

Table 3: The DOC, pH and quantity of suspended sediments of the collected river water samples compared with K_dand K_ocvalues calculated for their suspended sediments

Sample	pH	DOC (mg/l)	Sediment concentration (mg/l)	Kd(l/kg)	Koc(l/kg)
Aire (a)	8.0	ND^a	7.0	47777	82374
Aire (b)	7.3	5.6	3.0	206440	355931
Calder (b)	7.5	6.8	2.5	226384	390317
Thames	8.8	3.0	20	1903	3281

 

Validity criteria fulfilled:

not applicable

Conclusions:

4- tert-octylphenol strongly adsorbs to sediments.

Executive summary:

Laboratory batch techniques were used to study the sorptive behaviour of 4-t-octylphenol (OP) to sediments from three English rivers of contrasting water quality. Samples were taken from industrially polluted lower reaches of the Aire and Calder rivers, in the catchment as well as a rural reach of the River Thames, in Oxfordshire in the South of England.

The results showed that given either sufficient time or mixing, a large proportion of OP in solution will sorb to the bed-sediments, with distribution coefficients (Kd) of 6-700 l/kg and organic carbon normalised partition coefficients (Koc) 3500 -18000 1/kg. The sediments which sorbed the highest quantities of OP had higher total organic carbon and a greater proportion of clay and silt particles. There was evidence in some bed-sediments of a sorption-desorption hysteresis effect between OP and sediment.

The suspended sediments, on a carbon for carbon basis, adsorbed 5-35 times more OP than their respective bed-sediments in the Aire and Calder rivers: microscopic examination suggested that the suspended sediments were predominantly organic aggregates. The suspended sediments of the Thames adsorbed far less OP than those of the Aire and Calder: microscopic examination revealed these suspended sediments to he largely algae.

The work predicts that suspended sediments may play a key role in the fate of OP in the industrial reaches of English Rivers. In the comparatively rural reach of the Thames, a higher proportion of OP might be predicted to remain free in solution.

Reference 5

Endpoint:

adsorption / desorption: screening

Type of information:

experimental study

Adequacy of study:

supporting study

Study period:

2006

Reliability:

3 (not reliable)

Rationale for reliability incl. deficiencies:

other: Documentation insufficient for assessment.

Qualifier:

equivalent or similar to guideline

Guideline:

OECD Guideline 121 (Estimation of the Adsorption Coefficient (Koc) on Soil and on Sewage Sludge using High Performance Liquid Chromatography (HPLC))

Principles of method if other than guideline:

Samples with a high content of particulate matter were filtered by a 0.45 mm glass fibre filter. For analysis of the selected alkylphenols and alkylphenolethoxylates, the water samples were acidified with sulphuric acid to pH After addition of 25 ml surrogate standard (4-NP-d6), the filter cake was extracted in a Soxhlet apparatus using dichloromethane as solvent. The extract was then evaporated and changed to acetonitrile and analysed employing the above-mentioned methodology.

GLP compliance:

not specified

Type of method:

HPLC estimation method

Media:

sewage sludge

Radiolabelling:

no

- Sorbed concentrations were obtained by measuring the bound fraction.

- The total concentration values are obtained by summing up the dissolved and the sorbed fractions.

- Dissolved and sorbed fractions of the investigated compounds were determined separately, thus allowing the calculation of solid–liquid distribution coefficients (Kd) for those substances.

- Octylphenol (OP) is of minor relevance than nonylphenol and notably lower concentrations were found in the influents of the nine WWTPs.

- The total concentration of OP concentrations in the influents averaged 363 ng/l (16 -857 ng/l).

- The Kd values for OP averaged 975 l/kg (515 - 1559 l/kg).

- As the organic content of solids in the influent was not measured, no normalisation to this fraction is possible (determination of Koc).

Validity criteria fulfilled:

not applicable

Conclusions:

The calculated Kd values for octylphenol indicate that this substance strongly adsorbs to soil and sediments.

Executive summary:

Several surfactants were monitored in treated and untreated sewage in nine municipal wastewater treatment plants(WWTPs) in western Austria. The nine sampled WWTPs cover a wide variety referring to size and applied treatment technology.

The investigation focused amongst others on octylphenol (OP). OP was analysed separately in the liquid phase and in the solid phase.

OP was of minor importance with total influents concentrations below 1 µg/l. Kd values for OP varied between 515 and 1559 l/kg.

As the organic content of solids in the influent was not measured, no normalisation to this fraction is possible (determination of Koc).

The calulated Kd values for octylphenol indicate that this substance strongly adsorbs to soil and sediments.

Description of key information

Laboratory batch techniques were used to study the sorption/adsorption of 4-t-octylphenol (OP) to sediments from three English rivers of contrasting water quality (Johnson et al., 1998). Samples were taken from industrially polluted lower reaches of the Aire and Calder rivers, as well as from a rural reach of the River Thames, in Oxfordshire in the South of England. 
In a second report (Clara et al., 2007), several surfactants were monitored in treated and untreated sewage in nine municipal wastewater treatment plants (WWTPs) in western Austria. The nine sampled WWTPs were representative of a wide variety of facilities with reference to size and applied treatment technology. The investigation focused amongst others on octylphenol (OP). OP was analysed separately in the liquid phase and in the solid phase. 

Additional information

The results from Johnson et al. (1998) showed that given either sufficient time or mixing, a large proportion of OP in solution will sorb to the bed-sediments, with distribution coefficients (Kd) of 6-700 l/kg and organic carbon normalised partition coefficients (Koc) of 3500 -18000 1/kg. The sediments which sorbed the highest quantities of OP had higher total organic carbon and a greater proportion of clay and silt particles.

In the study conducted by Clara et al., 2007, OP was a minor constituent of total influents, with concentrations below 1 µg/l. Kd values determined for OP varied between 515 and 1559 l/kg, and overlap the range of Kd values determined by Johnson et al. (1998).

The calculated Kd values for octylphenol indicate that this substance strongly adsorbs to soil and sediments.

These results are supported by the results of the literature review conducted by Melcer at al. (2007). For OP, Koc values reported from different laboratory studies vary widely from about 3,500 to 18,000 l/kg.