Registration Dossier

Environmental fate & pathways

Adsorption / desorption

Currently viewing:

Administrative data

Link to relevant study record(s)

Referenceopen allclose all

Endpoint:
adsorption / desorption
Remarks:
adsorption
Type of information:
(Q)SAR
Adequacy of study:
weight of evidence
Study period:
2011
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
other: Use of a valid QSAR, the substance falls within the applicability domain of the (Q)SAR model, the prediction fits for the regulatory purpose, and the information is enough documented.
Justification for type of information:
. Relevance of the model
The model KOCWIN v2.00 as part of EPISUITE 4.0 software package (Syracuse Research Corporation, SRC) is reported as a usable valid model by ECHA (R6: QSARs and grouping of chemicals, May 2008). Calculation is performed according to the logKow value methodology.

2. Validation of the model

- Endpoint (OECD Principle 1)
The Endpoint concerned by the KOCWIN v2.00 prediction is the IUCLID endpoint 5.4.1: adsorption/desorption. The predicted variable is Koc, which is the soil/water partition coefficient.

- Algorithm (OECD Principle 2)
Koc calculated from logKow (from Doucette, 2000):

For non-polar compound (no correction factor):
log Koc = 0.8679 Log Kow - 0.0004
(n = 68, r2 = 0.877, std dev = 0.478, avg dev = 0.371)

With correction factors:
log Koc = 0.55313 Log Kow + 0.9251 + ΣPfN
where ΣPfN is the summation of the products of all applicable correction factor coefficients multiplied by the number of times (N) that factor is counted for the structure.

- Applicability domain (OECD Principle 3)

Training set molecular weights: 32.04 – 665.02 g/mol (average 224.4 g/mol)
Validation molecular weights: 73.14 – 504.12 g/mol (average 277.8 g/mol)

l-limonene have a molecular weight in the applicability domain.

- Uncertainty of the prediction (OECD Principle 4)
The following table gives statistical information for the log Kow-based regression: training and validation sets. The statistics pertain to the experimental log Koc and the Log Kow estimated log Koc:

Log Kow Methodology

Training Training Validation
No Corrections with Corrections Data set

number 68 447 150**
r2 corr coef 0.877 0.855 0.778
std deviation 0.478 0.396 0.679
avg deviation 0.371 0.307 0.494

** eight ammonium and metal salt compounds were removed from the Validation dataset

3. Adequacy of result for classification & labelling and/or risk assessment
KOCwin v2.00 estimates are values of Koc (partition coefficient between soil and water), they are therefore adequate for derivation of Koc (endpoint value) and completion of IUCLID endpoint 5.4.1 of adsorption/desorption.
Qualifier:
according to
Guideline:
other: Reach guidance on QSAR - R.6
Deviations:
no
Principles of method if other than guideline:
Koc was calculated from experimental Log Kow
GLP compliance:
no
Remarks:
Calculated results
Type of method:
other: QSAR
Media:
soil
Radiolabelling:
no
Computational methods:
1. Relevance of the model
The model KOCWIN v2.00 as part of EPISUITE 4.0 software package (Syracuse Research Corporation, SRC) is reported as a usable valid model by ECHA (R6: QSARs and grouping of chemicals, May 2008). Calculation is performed according to the logKow value methodology.

2. Validation of the model

- Endpoint (OECD Principle 1)
The Endpoint concerned by the KOCWIN v2.00 prediction is the IUCLID endpoint 5.4.1: adsorption/desorption. The predicted variable is Koc, which is the soil/water partition coefficient.

- Algorithm (OECD Principle 2)
Koc calculated from logKow (from Doucette, 2000):

For non-polar compound (no correction factor):
log Koc = 0.8679 Log Kow - 0.0004
(n = 68, r2 = 0.877, std dev = 0.478, avg dev = 0.371)

With correction factors:
log Koc = 0.55313 Log Kow + 0.9251 + ΣPfN
where ΣPfN is the summation of the products of all applicable correction factor coefficients multiplied by the number of times (N) that factor is counted for the structure.

- Applicability domain (OECD Principle 3)

Training set molecular weights: 32.04 – 665.02 g/mol (average 224.4 g/mol)
Validation molecular weights: 73.14 – 504.12 g/mol (average 277.8 g/mol)

l-limonene have a molecular weight in the applicability domain.

- Uncertainty of the prediction (OECD Principle 4)
The following table gives statistical information for the log Kow-based regression: training and validation sets. The statistics pertain to the experimental log Koc and the Log Kow estimated log Koc:

Log Kow Methodology

Training Training Validation
No Corrections with Corrections Data set

number 68 447 150**
r2 corr coef 0.877 0.855 0.778
std deviation 0.478 0.396 0.679
avg deviation 0.371 0.307 0.494

** eight ammonium and metal salt compounds were removed from the Validation dataset

3. Adequacy of result for classification & labelling and/or risk assessment
KOCwin v2.00 estimates are values of Koc (partition coefficient between soil and water), they are therefore adequate for derivation of Koc (endpoint value) and completion of IUCLID endpoint 5.4.1 of adsorption/desorption.

Key result
Type:
Koc
Value:
6 324
Remarks on result:
other: L/kg
Transformation products:
no
Validity criteria fulfilled:
yes
Conclusions:
The Koc of l-limonene predicted from Log Kow is 6324 L/kg.
Executive summary:

Koc has been predicted by KOCWIN v2.00 from EPISUITE 4.0 software reported as a usable valid model by ECHA (R6: QSARs and grouping of chemicals, May 2008).

KOCwin v2.00 estimates are values of Koc (partition coefficient between soil and water), they are therefore adequate for derivation of Koc (IUCLID endpoint 5.4.1 of adsorption/desorption) and for the purpose of risk assessment.

Calculation was performed according to the MCI, Molecular Connectivity Index methodology and l-limonene falls in the applicability domain of this QSAR model.

The Koc of l-limonene predicted from the Log Kow of 4.38 is 6324 L/kg.

Endpoint:
adsorption / desorption
Remarks:
adsorption
Type of information:
(Q)SAR
Adequacy of study:
weight of evidence
Study period:
2011
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
other: Use of a valid QSAR, the substance falls within the applicability domain of the (Q)SAR model, the prediction fits for the regulatory purpose, and the information is enough documented.
Justification for type of information:
1. Relevance of the model

The model KOCWIN v2.00 as part of EPISUITE 4.0 software package (Syracuse Research Corporation, SRC) is reported as a usable valid model by ECHA (R6: QSARs and grouping of chemicals, May 2008). Calculations are performed according to the methodology MCI, Molecular Connectivity Index.

2. Validation of the model

- Endpoint (OECD Principle 1)
The Endpoint concerned by the KOCWIN v2.00 prediction is the IUCLID endpoint 5.4.1: adsorption/desorption. The predicted variable is Koc, which is the soil/water partition coefficient.

- Algorithm (OECD Principle 2)
The methodology, Molecular Connectivity Index (Meylan et al., 1992), is used in KOCWIN v2.00 model.
For non-polar compound (no correction factor):
log Koc = 0.5213 MCI + 0.60
(n = 69, r2 = 0.967, std dev = 0.247, avg dev = 0.199)
With correction factors:
log Koc = 0.5213 MCI + 0.60 + ΣPfN
where ΣPfN is the summation of the products of all applicable correction factor coefficients multiplied by the number of times (N) that factor is counted for the structure.

- Applicability domain (OECD Principle 3)
Training set molecular weights: 32.04 – 665.02 g/mol (average 224.4 g/mol)
Validation molecular weights: 73.14 – 504.12 g/mol (average 277.8 g/mol)
l-limonene have a molecular weight in the applicability domain.

- Uncertainty of the prediction (OECD Principle 4)
The following table gives statistical information for the MCI (Molecular Connectivity Index, first-order) training and validation Datasets. The statistics pertain to the experimental log Koc and the MCI estimated log Koc:

MCI Methodology

Training Training Validation
No Corrections with Corrections Data set

number 69 447 158
r2 corr coef 0.967 0.900 0.850
std deviation 0.247 0.340 0.583
avg deviation 0.199 0.273 0.459

3. Adequacy of result for classification & labelling and/or risk assessment
KOCwin v2.00 estimates are values of Koc (partition coefficient between soil and water), they are therefore adequate for derivation of Koc (endpoint value) and completion of IUCLID endpoint 5.4.1 of adsorption/desorption.
Qualifier:
according to
Guideline:
other: Reach guidance on QSAR - R.6
Deviations:
no
Principles of method if other than guideline:
Calculations by Molecular Connectivity Index (MCI)
GLP compliance:
no
Remarks:
Calculated results
Type of method:
other: QSAR
Media:
soil
Radiolabelling:
no
Computational methods:
1. Relevance of the model

The model KOCWIN v2.00 as part of EPISUITE 4.0 software package (Syracuse Research Corporation, SRC) is reported as a usable valid model by ECHA (R6: QSARs and grouping of chemicals, May 2008). Calculations are performed according to the methodology MCI, Molecular Connectivity Index.

2. Validation of the model

- Endpoint (OECD Principle 1)
The Endpoint concerned by the KOCWIN v2.00 prediction is the IUCLID endpoint 5.4.1: adsorption/desorption. The predicted variable is Koc, which is the soil/water partition coefficient.

- Algorithm (OECD Principle 2)
The methodology, Molecular Connectivity Index (Meylan et al., 1992), is used in KOCWIN v2.00 model.
For non-polar compound (no correction factor):
log Koc = 0.5213 MCI + 0.60
(n = 69, r2 = 0.967, std dev = 0.247, avg dev = 0.199)
With correction factors:
log Koc = 0.5213 MCI + 0.60 + ΣPfN
where ΣPfN is the summation of the products of all applicable correction factor coefficients multiplied by the number of times (N) that factor is counted for the structure.

- Applicability domain (OECD Principle 3)
Training set molecular weights: 32.04 – 665.02 g/mol (average 224.4 g/mol)
Validation molecular weights: 73.14 – 504.12 g/mol (average 277.8 g/mol)
l-limonene have a molecular weight in the applicability domain.

- Uncertainty of the prediction (OECD Principle 4)
The following table gives statistical information for the MCI (Molecular Connectivity Index, first-order) training and validation Datasets. The statistics pertain to the experimental log Koc and the MCI estimated log Koc:

MCI Methodology

Training Training Validation
No Corrections with Corrections Data set

number 69 447 158
r2 corr coef 0.967 0.900 0.850
std deviation 0.247 0.340 0.583
avg deviation 0.199 0.273 0.459

3. Adequacy of result for classification & labelling and/or risk assessment
KOCwin v2.00 estimates are values of Koc (partition coefficient between soil and water), they are therefore adequate for derivation of Koc (endpoint value) and completion of IUCLID endpoint 5.4.1 of adsorption/desorption.
Key result
Type:
Koc
Value:
1 120
Remarks on result:
other: L/kg
Transformation products:
no
Validity criteria fulfilled:
yes
Conclusions:
Based on MCI calculation, the predicted Koc for l-limonene is 1120 L/kg.
Executive summary:

Koc has been predicted by KOCWIN v2.00 from EPISUITE 4.0 software reported as a usable valid model by ECHA (R6: QSARs and grouping of chemicals, May 2008).

KOCwin v2.00 estimates are values of Koc (partition coefficient between soil and water), they are therefore adequate for derivation of Koc (endpoint value) and completion of IUCLID endpoint 5.4.1 of adsorption/desorption.

Calculation was performed according to the MCI, Molecular Connectivity Index methodology and l-limonene falls in the applicability domain of this QSAR model.

The Molecular Connectivity Index method predicted a Koc for l-limonene of 1120 L/kg.

Endpoint:
adsorption / desorption
Remarks:
adsorption
Type of information:
(Q)SAR
Adequacy of study:
weight of evidence
Study period:
2011
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
other: Use of a valid QSAR, the substance falls within the applicability domain of the (Q)SAR model, the prediction fits for the regulatory purpose, and the information is enough documented.
Justification for type of information:
1. Relevance of the model
The derivation of Koc from Kow values as described in the TGD (2003) and in EUSES is reported as a usable valid model by ECHA (R6 : QSARs and grouping of chemicals, May 2008).

2. Validation of the model

- Endpoint (OECD Principle 1)
The Endpoint concerned by the Koc calculations with TGD (2003) equations is the IUCLID endpoint 5.4.1: adsorption/desorption. The predicted variable is Koc, which is the soil/water partition coefficient.

- Algorithm (OECD Principle 2)
The equations defined by Sabljic and Güsten (1995) for 19 different chemical classes of compounds. A full list of the equations for the 19 chemical classes can be found in the EU Technical Guidance Document on Risk Assessment Part III, Chapter 4, Section 4.3 (pp 24 – 27).

- Applicability domain (OECD Principle 3)
The applicability domain of the sorption models developed by Sabljic and Güsten (1995) depends on the considered chemical class, and is based on logKow range. The applicability domain of TGD (2003) equations for Koc calculations from Kow, for the various defined chemical classes can be found in the EU Technical Guidance Document on Risk Assessment Part III, Chapter 4, Section 4.3 (pp 24 – 27).

The chemical class “predominantly hydrophobics” is defined as compounds containing only carbon, hydrogen and halogen (F, Cl, Br, I) atoms, “nonhydrophobic chemicals” being all chemicals that cannot be classified as predominantly hydrophobics. Fourteen class-specific QSAR models were produced for all nonhydrophobic classes of chemicals, in order to give more appropriate equations.
l-limonene is not inorganic, nor a salt, nor a surfactant, nor an organometallic compound, or nor an ionized chemical. l-limonene is not hydrophobic. As non-hydrophobic, its chemical class should be clearly defined if possible. l-limonene can be defined as “Nonhydrophobics”, as it do not fit in any other chemical domain.
It has a logKow (4.38) in the validity domain of the model, allowing the use of the following equation: logKoc = 0.52 logKow + 1.02.
Thus the applicability of the equations defined by Sabljic and Güsten (1995) for l-limonene is confirmed.

- Uncertainty of the prediction (OECD Principle 4)
Uncertainty of the predictions is given in the EU Technical Guidance Document on Risk Assessment Part III, Chapter 4, Section 4.3,Table 4, p.26 as:
n=390,r2=0.63,s.e.=0.56.

3. Adequacy of result for classification & labelling and/or risk assessment
The described equations aim at calculating Koc values (partition coefficient between soil and water), they are therefore adequate for derivation of Koc (endpoint value) and completion of endpoint 5.4.1 of IUCLID dossier.
Qualifier:
according to
Guideline:
other: Reach guidance on QSAR - R.6
Deviations:
no
Principles of method if other than guideline:
Koc was calculated from experimental Log Kow according to equations from TGD 2003.
GLP compliance:
no
Remarks:
Calculated results
Type of method:
other: QSAR
Media:
soil
Radiolabelling:
no
Computational methods:
1. Relevance of the model
The derivation of Koc from Kow values as described in the TGD (2003) and in EUSES is reported as a usable valid model by ECHA (R6 : QSARs and grouping of chemicals, May 2008).

2. Validation of the model

- Endpoint (OECD Principle 1)
The Endpoint concerned by the Koc calculations with TGD (2003) equations is the IUCLID endpoint 5.4.1: adsorption/desorption. The predicted variable is Koc, which is the soil/water partition coefficient.

- Algorithm (OECD Principle 2)
The equations defined by Sabljic and Güsten (1995) for 19 different chemical classes of compounds. A full list of the equations for the 19 chemical classes can be found in the EU Technical Guidance Document on Risk Assessment Part III, Chapter 4, Section 4.3 (pp 24 – 27).

- Applicability domain (OECD Principle 3)
The applicability domain of the sorption models developed by Sabljic and Güsten (1995) depends on the considered chemical class, and is based on logKow range. The applicability domain of TGD (2003) equations for Koc calculations from Kow, for the various defined chemical classes can be found in the EU Technical Guidance Document on Risk Assessment Part III, Chapter 4, Section 4.3 (pp 24 – 27).

The chemical class “predominantly hydrophobics” is defined as compounds containing only carbon, hydrogen and halogen (F, Cl, Br, I) atoms, “nonhydrophobic chemicals” being all chemicals that cannot be classified as predominantly hydrophobics. Fourteen class-specific QSAR models were produced for all nonhydrophobic classes of chemicals, in order to give more appropriate equations.
l-limonene is not inorganic, nor a salt, nor a surfactant, nor an organometallic compound, or nor an ionized chemical. l-limonene is not hydrophobic. As non-hydrophobic, its chemical class should be clearly defined if possible. l-limonene can be defined as “Nonhydrophobics”, as it do not fit in any other chemical domain.
It has a logKow (4.38) in the validity domain of the model, allowing the use of the following equation: logKoc = 0.52 logKow + 1.02.
Thus the applicability of the equations defined by Sabljic and Güsten (1995) for l-limonene is confirmed.

- Uncertainty of the prediction (OECD Principle 4)
Uncertainty of the predictions is given in the EU Technical Guidance Document on Risk Assessment Part III, Chapter 4, Section 4.3,Table 4, p.26 as:
n=390,r2=0.63,s.e.=0.56.

3. Adequacy of result for classification & labelling and/or risk assessment
The described equations aim at calculating Koc values (partition coefficient between soil and water), they are therefore adequate for derivation of Koc (endpoint value) and completion of endpoint 5.4.1 of IUCLID dossier.

Key result
Type:
Koc
Value:
1 984
Remarks on result:
other: L/kg
Transformation products:
no
Validity criteria fulfilled:
yes
Conclusions:
The Koc of l-limonene predicted from experimental Log Kow of 4.38 and following equations of Sabljic and Güsten (1995) for non hydrophobic substances is 1984 L/kg.
Executive summary:

Koc has been predicted by calculations from experimental Log Kow and using equations of Sabljic and Güsten (1995) reported as a usable valid model by ECHA (R6: QSARs and grouping of chemicals, May 2008).

The described equations aim at calculating Koc values (partition coefficient between soil and water), they are therefore adequate for derivation of Koc (endpoint value) and completion of endpoint 5.4.1 of IUCLID dossier.

Calculation was performed according to the equation for non hydrophobic substances and l-limonene falls in the applicability domain of this QSAR model.

The Koc of l-limonene predicted from the Log Kow of 4.38 is 1984 L/kg.

Description of key information

Weight of evidence: The geometric mean of the Koc predicted values by valid QSARs is considered for chemical safety assessment = 2413 L/kg.

Key value for chemical safety assessment

Koc at 20 °C:
2 413

Additional information

ECHA Guidance (chapter R.7.a, paragraph R.7.1.15) recommends calculating the Koc endpoint value as the geometric mean of results obtained by means of QSARs. Koc has been estimated by 3 valid QSAR methods: 1. with KOCWIN from EPISUITE 4.0 using CAS number as input and the Molecular Connectivity Method (MCI) = 1120 L/kg 2. with KOCWIN from EPISUITE 4.0 using experimental Log Kow and CAS number as input= 6324 L/kg 3. from experimental Log Kow with equation of Sabljic and Güsten (1995) for non hydrophobic susbtances as reported in TGD (2003) = 1984 L/kg

The geometric mean value 2413 L/kg is considered as the endpoint defining the substance for adsorption and is used as input parameter for Environmental Risk Assessment.