Hydroxyacetone

Adsorption / desorption

Link to relevant study record(s)

Reference

Endpoint:

adsorption / desorption: screening

Type of information:

experimental study

Adequacy of study:

key study

Study period:

2018/03/26-2018/03/27

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

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

Deviations:

no

Qualifier:

according to guideline

Guideline:

EU Method C.19 (Estimation of the Adsorption Coefficient (KOC) on Soil and Sewage Sludge Using High Performance Liquid Chromatography (HPLC))

Deviations:

no

GLP compliance:

yes (incl. QA statement)

Remarks:

Landesamt für Umwelt, 55116 Mainz (15.05.2018)

Type of method:

HPLC estimation method

Media:

soil/sewage sludge

Specific details on test material used for the study:

The test item was stored in a closed vessel at room temperature (20 ± 5°C).

Radiolabelling:

no

Test temperature:

25± 0.5 °C

Details on study design: HPLC method:

EQUIPMENT
- Apparatus: HPLC
- Type: HPLC_7 (Thermo Ultimate)
Componenents: Solvent Rack, Pump LPG-3400SD, Autosampler WPS-3000TSL, Column Compartment TCC-3200 with 6-fold column switch valve
Manufacturer: Thermo Fisher Scientific
Software: CHROMELEON 6.80 SR15b Build 4981
HPLC column Reprosil Pur 120 CN 3 μm 125 x 4.6 mm with pre-column
Phenomenex SecurityGuard CN, 4x3 mm
Eluent isocratic methanol/water 55/45 (% v/v), degassed
Flow rate 0.8 mL/min
Injection volume 10 μL
Column temperature 25 ± 0.5 °C.
Wavelength for evaluation 200 nm (reference mix)
268 nm (test item)
- Detection system: UV-Detector DAD-3000

MOBILE PHASES
A mixture of 55% methanol and 45% water, isocratic, was used. The mix was produced by the mixture of 550 mL methanol and 450 mL water. pH of the eluent was approx. 5 - 6 (measured with pH indicator paper). The temperature was held constant at 25 ± 0.5 °C by use of a HPLC column oven.

DETERMINATION OF DEAD TIME
- Method: (by inert substances) The dead time t0 was measured by using an unretained compound (sodium nitrate) which was included in the mix of the reference items.

REFERENCE SUBSTANCES
- Identity: Sodium nitrate (2mg/L), Acetanilide (4 mg/L), Methylbenzoate (4 µL/L), Naphthalene (2 mg/L), 1,2,3-Trichlorobenzene (4 mg/L), Phenanthrene (4 mg/L), 4,4’-DDT (10 mg/L)

DETERMINATION OF RETENTION TIMES
Pre-Tests
A solution of the test item in eluent was prepared and injected in order to check detectability and to find an approximate range for log KOC.
In the preliminary test, a solution of the test item with a concentration of 1 mL/L was injected in order to check retention time and the signal height. The solution gave a large peak at 2.1 min. A test item solution with nominal concentration 200 μL/L was injected and showed a peak with sufficient height and retention time 2.1 min. Based on this result, a suitable mix of reference items was chosen (with Acetanilide as substance with the lowest log KOC and a retention time of approx. 2.8 minutes). The test item solution with nominal concentration 200 μL/L was used for the determination of the log KOC of the test item.

Determination of the Log KOC of the Test Item
One vial was filled with the reference item mix and one vial with the test item solution. The vials were analysed using HPLC with the program described above. Three injections were measured from the reference item mix, three injections from the test item solution and again three injections from the reference item mix.
The data was saved, the retention times of the seven peaks of the reference item mix and the retention time of the peak of the test item were used for further calculations.

REPETITIONS
- Number of determinations: Three injections were measured from the reference item mix, three injections from the test item solution and again three injections from the reference item mix.

EVALUATION
- Calculation of capacity factors k': The capacity factor k´ for every peak was calculated as follows:k' = tR - t0/ t0, with tr= retention time, t0= dead time (retention time of sodium nitrate)
- Calculation of retention times: The data was saved, the retention times of the seven peaks of the reference item mix and the retention time of the peak of the test item were used for further calculations.
- Determination of the log Koc value: The calibration function was drawn up from the six measurements of the capacity factors of the reference items. By plotting the (known) log KOC values of the reference items against the mean log capacity factors and performing linear regression, a calibration function was obtained. With the calculated capacity factors for the test item, each log KOC from the three determinations
was calculated. Finally, mean and standard deviation of the log KOC were calculated.

Analytical monitoring:

not specified

Details on sampling:

A solvent blank followed by three injections of a solution of the reference items were analysed with HPLC, then three injections of a solution of the test item; and finally three injections of the solution with the reference items, in total ten runs

Details on matrix:

HPLC with a cyanopropyl chemical bound resin on a silica base column

Details on test conditions:

no details given

Computational methods:

not specified

Key result

Type:

log Koc

Value:

0.668 dimensionless

pH:

5.5

Temp.:

25 °C

Remarks on result:

other: estimated by extrapolation

Remarks:

below lowest reference substance

Type:

log Koc

Value:

< 1.25 dimensionless

pH:

5.5

Temp.:

25 °C

Details on results (HPLC method):

Retention times of reference substances used for calibration: Sodium nitrate (1.050 min); Acetanilid (2.808 min); Methylbenzoate (3.308 min); Napthalene (4.893 min); 1,2,3-Trichlorobenzene (5.614 min); Phenanthrene (8.540 min); 4,4'-DDT (24.061 min)
Details of fitted regression line (log k' vs. log Koc): log k' = 0.2519 * log KOC - 0.1256, with a coefficient of determination r2 = 0.991
Average retention data for test substance: 2.208 min.

Table 1a: Retention Times (RT) Test Item

Peak 1 Test Item	RT
	min.
Measurement 1	2.208
Measurement 2	2.208
Measurement 3	2.208
Mean	2.208
Standard deviation	0.000

The retention time of the test item was shorter than the one of the reference substance with the lowest log K_OC. However, as the retention times were highly reproducible and well above the dead time for an unretained substance extrapolation of the log K_OCof the test item is possible.

Table 1b Capacity Factor, log Capacity Factor, log K_OC Test Item

Peak 1 Test Item	k'	log k'	log Koc
Measurement 1	1.1032	0.0426	0.668*
Measurement 2	1.1032	0.0426	0.668*
Measurement 3	1.1032	0.0426	0.668*
Mean			0.668*
Standard deviation			0.000
Confidence limit			0.428

*extrapolated values

log K_OC was calculated from the capacity factor as follows: log K_OC = (log k' + 0.1256) / 0.2519

Validity criteria fulfilled:

no

Remarks:

As the calculated log KOC of the peak lay below the lowest log KOC of the reference items (Acetanilide with Log KOC 1.25), the log KOC of this peak is estimated to be 0.668 ± 0.428 (< 1.25, based on lowest value for the concurrent reference substances).

Conclusions:

The correlation log k’/log KOC is sufficiently accurate, as calibration gave a coefficient of determination r2 = 0.991. Using the correlation log k’ / log KOC, the log KOC of the peak of the test item Hydroxyacetone was extrapolated as 0.668 ± 0.428 (mean and confidence interval). In consequence, the log KOC for Hydroxyacetone is estimated to be 0.668 ± 0.428 (< 1.25 based on lowest value for the concurrent reference substances): No observations were made which might give doubts concerning the validity of the study outcome.

Executive summary:

The study was performed following OECD guideline 121 and EU method C.1, in compliance with GLP. Using a HPLC with a cyanopropyl chemical bound resin on a silica base column, seven reference items with different retention times and sodium nitrate for the determination of the dead time were used to produce a calibration curve, since retention time on cyanopropyl columns and K_OCare correlated. The reference items were chosen on behalf of the results of the pre-test.

A solvent blank followed by three injections of a solution of the reference items were analysed with HPLC, then three injections of a solution of the test item; and finally three injections of the solution with the reference items, in total ten runs. For each reference item, the capacity factor k’ was calculated from the retention time of sodium nitrate and the retention time of the respective reference item. A calibration function was set up using the literature values for K_OC of the reference items and the mean capacity factor of the six determinations. In the graph log k’ versus log K_OC, linear regression was performed, giving r² = 0.991 and r = 0.996.

The chromatogram of the test item showed one peak (mean RT = 2.208 min), eluting earlier as the reference substance with the lowest log K_OC value. However, as the retention times were highly reproducible and well above the dead time for an unretained substance extrapolation of the log K_OC of the test item is possible. With the calibration function log k’ versus log K_OC, the corresponding log K_OC was calculated as 0.668 ± 0.428 (extrapolated log K_OC and confidence interval). In consequence, the log K_OC for Hydroxyacetone is estimated to be 0.668 ± 0.428 (< 1.25 based on lowest value for the concurrent reference substances).

Description of key information

OECD 121, GLP, HPLC method, log K_OC = 0.668 ± 0.428 (< 1.25 based on lowest value for the concurrent reference substances).

Key value for chemical safety assessment

Koc at 20 °C:

4.656

Additional information

The study was performed following OECD guideline 121 and EU method C.1, in compliance with GLP. Using a HPLC with a cyanopropyl chemical bound resin on a silica base column, seven reference items with different retention times and sodium nitrate for the determination of the dead time were used to produce a calibration curve, since retention time on cyanopropyl columns and K_OCare correlated. The reference items were chosen on behalf of the results of the pre-test.

A solvent blank followed by three injections of a solution of the reference items were analysed with HPLC, then three injections of a solution of the test item; and finally three injections of the solution with the reference items, in total ten runs. For each reference item, the capacity factor k’ was calculated from the retention time of sodium nitrate and the retention time of the respective reference item. A calibration function was set up using the literature values for K_OCof the reference items and the mean capacity factor of the six determinations. In the graph log k’ versus log K_OC, linear regression was performed, giving r2 = 0.991 and r = 0.996.

The chromatogram of the test item showed one peak (mean RT = 2.208 min), eluting earlier as the reference substance with the lowest log K_OCvalue. However, as the retention times were highly reproducible and well above the dead time for an unretained substance extrapolation of the log K_OCof the test item is possible. With the calibration function log k’ versus log K_OC, the corresponding log K_OCwas calculated as 0.668 ± 0.428 (extrapolated log K_OCand confidence interval). In consequence, the log K_OCfor Hydroxyacetone is estimated to be 0.668 ± 0.428 (< 1.25 based on lowest value for the concurrent reference substances).