Benzoic acid, 2-([1,1'-biphenyl]-4-ylcarbonyl)-

Adsorption / desorption

Link to relevant study record(s)

Reference

Endpoint:

adsorption / desorption: screening

Type of information:

calculation (if not (Q)SAR)

Adequacy of study:

key study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

accepted calculation method

Justification for type of information:

Log Koc of the test substance was calculated using regression model equations for general, organic acid class and ionisable compounds. See below under 'methods' for applicability domain.

Principles of method if other than guideline:

The soil adsorption coefficient (Koc) value for the test substance was estimated using the log Pow (Partition coefficient) correlation approach of the Log Koc regression models equations for general, organic acid class and ionisable compounds.

Key result

Type:

log Koc

Value:

ca. 1.34 - ca. 2.35 dimensionless

Remarks on result:

other: Koc: 21.7 to 221.5 L/kg; calculated 'general' log Kow based regression equations

Key result

Type:

log Koc

Value:

ca. 1.39 dimensionless

Remarks on result:

other: Koc: 24.4 L/kg; calculated using 'organic acid' specific log Kow based regression equation

Key result

Type:

log Koc

Value:

ca. 1.74 dimensionless

Remarks on result:

other: Koc: 54.72 L/kg; calculated using log Kow based regression equations for ionisable compounds

Key result

Type:

log Koc

Value:

ca. 0.65 - ca. 3 dimensionless

Remarks on result:

other: Koc: 4.5 to 1043 L/kg; calculated 'general' log S based regression equations

Key result

Type:

log Koc

Value:

ca. 1.57 dimensionless

Remarks on result:

other: Koc: 37.18 L/kg; calculated using 'aromatics' specific log S based regression equation

Results

Koc was calculated using range of regression equations, i.e., general wide variety, class specific and ionisable coumpound specific. The test substance is a aromatic organic acid, therefore the equations related to organi acid or aromatics class have been selected as one the criteria to generate the Koc value for the test substance. Apart from chemical class specific equations, general equations also have selected due to their well documented development and large data sets of Koc values. As the test substance is ionisable substance, the regression equation for ionisable compound also used to calculate the Koc. Prediction of log Koc can be improved by treating neutral and ionic fractions separately and therefore probably is superior to methods that merge both fractions without considering the differences between neutral compounds and ions. Therefore, considering that the test substance is an acid with an predicted pKa of 3.47, the neutral and ionic fractions at pH 5.8 was determined to be 0.0047 and 0.9953 respectively using Chemicalize (please refer to the attached PDF for details on results).

Table 1: Calculations of Koc based on regression models equations (general and Ionisable compound)

Regression Models Used to Estimate Log Koc from Log Kow
	Ǿneutral fraction	0.0047
	Ǿionic fraction	0.9953
Equation Number	Log Kow
	1.78
(I)	EPISuite (Doucette, 2000)	Log Koc = 0.8679 Log Kow - 0.0004
	Log Koc	1.54
(II)	Variety, mostly pesticides (Kenaga and Goring, 1980)	log Koc = = 1.377 + 0.544 log Kow
	Log Koc	2.35
(III)	Alcohols and organic acids	log Koc = 0.47 log Kow + 0.5
	Log Koc	1.34
(IV)	Wide variety (Gerstl, 1990)	log Koc = 0.679 log Kow + 0.663
	Log Koc	1.87
(V)	Hydrophobics (Sabljic et al 1995)	log Koc = 0.81 log Kow + 0.10
	Log Koc	1.54
(VI)	Wide variety (Baker et al 1997)	log Koc = 0.903 log Kow + 0.094
	Log Koc	1.70
(VII)	Franco and Trapp (2008) - acids (pH 5.8)	Log Koc = Log (Ǿn*10^(0.54 log Kow + 1.11) + Ǿion*10^(0.11 log Kow + 1.54))
	Log Koc	1.74
(VIII)	Organic acids (EC, 2003)	Log Koc = 0.60 log Kow + 0.32
	Log Koc	1.39
Franco and Trapp 2008	Equa. (VII)	1.74
Average of all log Koc values	Equa. (I) + (II) + (III) + (IV) + (V) + (VI) + (VIII)	1.68
Selected Log Koc value		1.74
	KOC	54.72 L/kg

Table 2: Calculations of Koc based on regression models equations based on log S:

Regression Models Used to Estimate Log Koc from Log S
Solubility (mg/L	13.5	MP
Log S	1.130333768	231
Equa. (I)	Variety, mostly pesticides (Kenaga and Goring, 1980)	log Koc = – 0.55 log S + 3.64 (S in mg/L)
	Log Koc	3.02
Equa. (II)	Variety, mostly pesticides (Briggs, 1981)	log Koc = 0.51 [log S + (0.01 MP – 0.25)] + 0.8
	(0.01 MP – 0.25)	2.06
	log S + (0.01 MP – 0.25)	3.190333768
	Log Koc	2.43
Equa. (III)	Wide variety (Gerstl, 1990)	log Koc = (–0.508 x log S + 0.953) – Fc*
	Log Koc	0.65
Equa. (IV)	PAHs, aromatics (Karickhoff et al., 1979)	log Koc = logS + 0.44
	Log Koc	1.57
Average	Equa. (I) + (II) + (III) + (IV)	1.92
Log Koc in equation III is estimated using FC correction factor for non-halogenated aromatic hydrocarbons. Gerstl 1990 expressions were developed with well documented and/or large data sets of Koc values.

Based on the above calculations, the log Koc values were found to range as follows:

- Log Koc from Method 1 (using non-ionic log Kow based regression based equations): ranged from 1.34 to 2.35 using general class equations and was 1.39 using 'organic acid' class equation (mean = 1.68; Koc = 47.38 L/kg)

- Log Koc from Method 2 (using non-ionic log S based regression based equations): 0.65 to 3.02 using general class equations and was 1.57 using 'aromatics' class equation (mean = 1.92; Koc = 82.44 L/kg)

- Log Koc from Method 3 (using ionic log Kow based regression based equations): 1.74 (Koc = 54.72 L/kg)

Given that the test substance is ionic, the prediction of log Koc by treating neutral and ionic fractions separately is considered superior to methods that merge both fractions without considering the differences between neutral compounds and ions (Franco and Trap, 2008). Therefore, the log Koc of 1.74 (i.e., equivalent to Koc of 54.72 L/kg) calculated from Franco and Trapp (2008) equation has been selected as key value for this endpoint.

Validity criteria fulfilled:

not applicable

Conclusions:

Using the regression model equation for ionisable compounds based on partition coefficient value, the Koc value of the test substance was determined to be 54.72 L/kg (log Koc = 1.74).

Executive summary:

The soil adsorption coefficient (Koc) value for the test substance, PBBA, was determined using the well-known log Kow or log S based log Koc regression models equations. To calculate a more reliable value and to reduce the overall uncertainty, multiple equations, which could be categorised as general, class-specific (i.e., organic acid or aromatics) (Doucette WJ, 2000) and ionisable compound based (Franco and Trapp, 2008), were used for the calculations. The Kow-based log Koc values were calculated using an experimentally determined log Kow value of 1.78 and a predicted фn of 0.0047 and a minimum фion of 0.9953 (using Chemicalise), for the Franco et al., equation. The solubility-based log Koc values were calculated using experimentally determined water solubility of 13.5 mg/L and melting point of 231°C.

The Kow-based log Koc values were calculated to range from 1.34 to 2.35, using general equations and was 1.39 using ‘organic acid class’ specific equation and 1.74 using the ionisable compound-based equation. The solubility-based log Koc values were calculated to range from 0.65 to 3, using general equations and was 1.57 using ‘aromatics class’ specific equation. This range of Koc indicates low to moderate sorption to soil / sediment and moderate to slow migration potential to ground water (US EPA, 2012). Given that the test substance is ionic, the prediction of log Koc by treating neutral and ionic fractions separately is considered superior to methods that merge both fractions without considering the differences between neutral compounds and ions (Franco and Trap, 2008). Therefore, the log Koc of 1.74 (i.e., equivalent to Koc of 54.72 L/kg) calculated from Franco and Trapp (2008) equation has been selected as key value for this endpoint.

Description of key information

Using the log Kow-based regression equation for ionisable and acidic compounds, the Koc of the test substance was calculated to be 54.72 L/kg (log Koc: 1.74).

Key value for chemical safety assessment

Koc at 20 °C:

54.72

Additional information

Given the limitation of the publicly available QSAR models for log Koc estimation of ionic compounds, the log Koc has been calculated using log Kow or solubility (log S) based regression equations proposed for general and specific classes, as a comparison.

The soil adsorption coefficient (Koc) value for the test substance, PBBA, was determined using the well-known log Kow or log S based log Koc regression models equations. To calculate a more reliable value and to reduce the overall uncertainty, multiple equations, which could be categorised as general, class-specific (i.e., organic acid or aromatics) (Doucette WJ, 2000) and ionisable compound based (Franco and Trapp, 2008), were used for the calculations. The Kow-based log Koc values were calculated using an experimentally determined log Kow value of 1.78 and a predicted фn of 0.0047 and a minimum фion of 0.9953 (using Chemicalise), for the Franco et al., equation. The solubility-based log Koc values were calculated using experimentally determined water solubility of 13.5 mg/L and melting point of 231°C.

The Kow-based log Koc values were calculated to range from 1.34 to 2.35, using general equations and was 1.39 using ‘organic acid class’ specific equation and 1.74 using the ionisable compound-based equation. The solubility-based log Koc values were calculated to range from 0.65 to 3, using general equations and was 1.57 using ‘aromatics class’ specific equation. This range of Koc indicates low to moderate sorption to soil / sediment and moderate to slow migration potential to ground water (US EPA, 2012). Given that the test substance is ionic, the prediction of log Koc by treating neutral and ionic fractions separately is considered superior to methods that merge both fractions without considering the differences between neutral compounds and ions (Franco and Trap, 2008). Therefore, the log Koc of 1.74 (i.e., equivalent to Koc of 54.72 L/kg) calculated from Franco and Trapp (2008) equation has been selected as key value for this endpoint.

Possible processes behind the sorption of organic chemicals to soil and sediment are ion bonding or ligand exchange, chemiosorption (formation of a bond, usually covalent, with the soil molecular structure), ion–dipole and dipole–dipole interactions, charge transfer, hydrogen bonding, and hydrophobic bonding (Van der Waals forces). The most chemically active component of the soil is the colloidal fraction, which consists of organic matter and inorganic clay minerals. Both components display a negative electrical charge at the surface. The effect of this charge can be measured by the cationic exchange capacity, which on average is 50 meq/100 g for clays and 290 meq/100 g for humic acids. Electrical forces involving charge transfer (40 kJ/mol) are stronger than hydrophobic bonding (4 kJ/mol) so that they dominate when present. Thus, a different degree of sorption of anions, cations, and neutral molecules can be expected, with cations showing the highest potential for sorption, due to electrical attraction (Franco and Trapp, 2008).

Therefore, considering that the test substance is an anionic, its sorption potential can be expected to be much lesser than other known cationic compounds, which is in line with the calculated log Koc derived based on Franco and Trapp, 2008 proposed equation for ionisable compounds.