Valeryl chloride

Partition coefficient

Administrative data

Endpoint:

partition coefficient

Type of information:

(Q)SAR

Adequacy of study:

weight of evidence

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

results derived from a valid (Q)SAR model and falling into its applicability domain, with limited documentation / justification

Justification for type of information:

1. SOFTWARE
The Estimation Programs Interface (EPI) SuiteTM was developed by the US Environmental Protection Agency's Office of Pollution Prevention
and Toxics and Syracuse Research Corporation (SRC). It is a screening-level tool, intended for use in applications such as to quickly screen
chemicals for release potential and "bin" chemicals by priority for future work. Estimated values should not be used when experimental
(measured) values are available.

2. MODEL (incl. version number)
EPI SUMMARY (v4.11), Log Kow(version 1.68 estimate)

3. SMILES OR OTHER IDENTIFIERS USED AS INPUT FOR THE MODEL
O=C(CCCC)CL

4. SCIENTIFIC VALIDITY OF THE (Q)SAR MODEL
  QMRF: Estimation of Octanol-Water Partition Coefficient using KOWWIN v1.68 (EPI Suite v4.11)
1.0 QSAR identifier
1.1 QSAR identifier (title)
Estimation of Octanol-Water Partition Coefficient using KOWWIN v1.68 (EPI Suite v4.11)
1.2 Other related models
-
1.3 Software coding the model
KOWWIN v1.68 (EPI Suite v4.11)

2.0 General information
2.1 Date of QMRF
01 Nov. 2013
2.2 QMRF author and contact details
BASF SE, Department of Product Safety, Ludwigshafen, Germany
2.3 Date of QMRF update(s)
-
2.4 QMRF update(s)
-
2.5 Model developer(s) and contact details
WM Meylan and PH Howard, US EPA
2.6 Date of model development and/or publication
Publication of first model: 1995
2.7 References to main scientific papers and/or software package
Meylan, W.M. and P.H. Howard. 1995. Atom/fragment contribution method for estimating octanol-water partition coefficients. J. Pharm. Sci. 84: 83-92
2.8 Availability of information about the model
The model is non-proprietary and can be downloaded freely from US EPA.
2.9 Availability of another QMRF for exactly the same model
No (http://qsardb.jrc.it/qmrf/).

3.0 Defining the endpoint
3.1 Species
3.2 Endpoint
Log Kow at 25 °C
3.3 Comment on the endpoint
Regulation (EC) No 1907/2006 [REACH], Annex VII, 7.8 Partition coefficient octanol/water
3.4 Endpoint units
mg/L
3.5 Dependent variable
log S (mol/L)
3.6 Experimental protocol
The log Kow of a substance can be determined according to OECD guideline 107 or OECD guideline 117.
3.7 Endpoint data quality

4.0 Defining the algorithm
4.1 Type of model
QSAR (QSPR)
4.2 Explicit algorithm
KOWWIN uses a "fragment constant" methodology to predict log Kow. In a "fragment constant" method, a structure is divided into fragments (atom or larger functional groups) and coefficient values of each fragment or group are summed together to yield the log Kow estimate. KOWWIN’s methodology is known as an Atom/Fragment Contribution (AFC) method. Coefficients for individual fragments and groups were derived by multiple regression of 2447 reliably measured log Kow values. Correction factors were developed to be able to consider larger or more complex substructures than the fragments (“atoms”).

log Kow =Σ(fini)+Σ(cjnj)+0.229
(num = 2447, r2= 0.982, std dev = 0.217, mean error = 0.159)
Where:
1)Σ(fini) is the summation offi(the coefficient for each atom/fragment) timesni(the number of times the atom/fragment occurs in the structure)
2)Σ(cjnj) is the summation ofcj(the coefficient for each correction factor) timesnj(the number of times the correction factor occurs (or is applied) in the molecule).
4.3 Descriptors in the model
SMILES: structure of the compound as SMILES notation; The chemical structure (as SMILES) is the only required descriptor for estimating the water solubility. All other descriptors can be derived from the chemical structure.
Fragments: The structure is divided into fragments (atom or larger functional groups).
Correction factors: The correction factors were derived from a multiple linear regression that correlated differences between the experimental log Kow and the log Kow estimated by the initial equation [log Kow = Σ(fini) + b; where b is the linear equation constant] with the correction factor descriptors.
4.4 Descriptor selection
Coefficients for individual fragments and groups were derived by multiple regression of 2447 reliably measured log Kow values. Correction factors were derived from differences between initially estimated log Kow values and the experimental data.
4.5 Algorithm and descriptor generation
SMILES: user entered (required for estimation)
Fragments and correction factors:taken from database
4.6 Software name and version for descriptor generation
KOWWIN v1.68 (EPI Suite v4.11)
4.7 Descriptor/Chemicals ratio
Training dataset: 2447 compounds with reliably measured data for log Kow
Descriptors:2 (Fragments and correction factors derived from chemical structure)

5.0 Defining the applicability domain
5.1 Description of the applicability domain of the model
The applicability domain is described by the range of the molecular weight of the training set as well as the identification and number of instances of a given fragment. The number of instances of a given fragment is compared to the maximum for all training set compounds. If this number exceeds the maximum number or a fragment is not identified, the estimate of a substance is regarded to be less accurate.
5.2 Method used to assess the applicability domain
The applicability is assessed by comparing the molecular weight and the given fragments of a compound with the corresponding values of the training set.
5.3 Software name and version for applicability domain assessment
-
5.4 Limits of applicability
Range of Molecular Weights in the Training set:
- Minimum = 18.02 g/mol
- Maximum = 719.92 g/mol
- Average = 199.98 g/mol
Structural features:
- See 6.1

6.0 Defining goodness-of-fit and robustness
6.1 Availability of the training set
The KOWWIN training and validation datasets can be downloaded from the Internet at:http://esc.syrres.com/interkow/KowwinData.htm.
Substructure searchable formats of the data can be downloaded at:http://esc.syrres.com/interkow/EpiSuiteData_ISIS_SDF.htm.
6.2 Available information for the training set
- CAS number
- Chemical name
- Molecular formula
- Data set (training set, validation set, diverse subsets)
- Measured log Kow
- Estimated log Kow (KOWWIN)
- Residual
- Reference for experimental data
6.3 Data for each descriptor variable for the training set
Descriptors derived from chemical structure, therefore not listed in database
6.4 Data for the dependent variable (response) for the training set
-
6.5 Other information about the training set
-
6.6 Pre-processing of data before modelling
-
6.7 Statistics for goodness-of-fit
Total Training Set Statistics:
- number in dataset = 2447
- correlation coefficient (r2) = 0.982
- standard deviation = 0.217
- absolute deviation = 0.159
- average molecular weight = 199.98

7.0 Defining predictivity
7.1 Availability of the external validation set
See 6.1
7.2 Available information for the external validation set
See 6.2
7.3 Data for each descriptor variable for external validation set
See 6.3
7.4 Data for the dependent variable for the external validation set
See 6.4
7.5 Other information about the external validation set
-
7.6 Experimental design of test set
References to data given
7.7
Total Validation Set Statistics:
- number in dataset = 10946
- correlation coefficient (r2) = 0.943
- standard deviation = 0.479
- absolute deviation = 0.356
- average Molecular Weight = 258.98

The external validation set of 10946 compounds contains 372 compounds that exceed the domain of instances of a given fragment or correction factor maximum for all training set compounds. The estimation accuracy for these compounds is:

Exceed Fragment Instance Domain - Accuracy Statistics:
- number in dataset = 372
- correlation coefficient (r2) = 0.939
- standard deviation = 0.731
- absolute deviation = 0.564
- average Molecular Weight = 460.0

Exceed Molecular Weight Domain - Accuracy Statistics:
- number in dataset = 103
- correlation coefficient (r2) = 0.879
- standard deviation = 0.815
- absolute deviation = 0.619
- average Molecular Weight = 802.16

Exceed BOTH Fragment & MW Domain - Accuracy Statistics:
- number in dataset = 75
- correlation coefficient (r2) = 0.879
- standard deviation = 0.905
- absolute deviation = 0.706
- average Molecular Weight = 812.70

Appendix F of the User’s Guide lists the 75 compounds that exceed both the fragment instance domain and molecular limit domain.
7.8 Predictivity - Assessment of the external validation set
-

8.0 Providing a mechanistic interpretation
8.1 Mechanistic basis of the model
The octanol-water partition coefficient is a physical property used extensively to describe a chemical's lipophilic or hydrophobic properties.KOWWIN uses a "fragment constant" methodology to predict log P. In a "fragment constant" method, a structure is divided into fragments (atom or larger functional groups) and coefficient values of each fragment or group are summed together to yield the log P estimate. KOWWIN’s methodology is known as an Atom/Fragment Contribution (AFC) method.

9.0 Miscellaneous information
9.1 Comments
-
9.2 Bibliography
- Meylan, W.M. and P.H. Howard. 1995. Atom/fragment contribution method for estimating octanol-water partition coefficients. J. Pharm. Sci. 84: 83-92
- US EPA (2012). On-Line KOWWIN User’s Guide.



5. APPLICABILITY DOMAIN
1. Substance
See “Test material identity”

2. General information
2.1 Date of QPRF
See “Data Source (Reference)”

2.2 QPRF author and contact details
See “Data Source (Reference)”

3. Prediction
3.1 Endpoint (OECD Principle 1)
Endpoint: Octanol-water partition coefficient (log Kow)
Dependent variable. Octanol-water partition coefficient (log Kow)

3.2 Algorithm (OECD Principle 2)
Model or submodel name. KOWWIN
Model version: v. 1.68
Reference to QMRF: QMRF: Estimation of Octanol-Water Partition Coefficient
using KOWWIN v1.68 (EPI Suite v4.11)
Predicted value (model result): See “Results and discussion”
Input for prediction: Chemical structure via CAS number or SMILES
Descriptor values: - Chemical structure
- Fragments
- Correction factors

3.3 Applicability domain (OECD principle 3)
Domains:
1) Molecular weight (range of test data set: 18.02 to 719.92 g/mol; On-Line KOWWIN User’s Guide, Ch. 6.2.3 Estimation Domain)
Substance (not) within range (120.58 g/mol)
2) Fragments: Number of instances of the identified fragments does not exceed the maximum number as listed in Appendix D (On-Line KOWWIN User’s Guide)
fulfilled.
3) Fragments: Substance has a functional group(s) or other structural features not represented in the training set, and for which no fragment coefficient was developed (Appendix D, On-Line KOWWIN User’s Guide)
applicable.
4) Correction factors: Number of instances of the identified correction factor does not exceed the maximum number as listed in Appendix D (On-Line KOWWIN User’s Guide)
fulfilled.

3.4 The uncertainty of the prediction (OECD principle 4)
According to REACH Guidance Document R.7a, (Nov. 2012), solubility in water is difficult to model accurately. For this reason, as well as the fact that the experimental error on solubility measurements can be quite high (generally reckoned to be about 0.5 log unit), the prediction of aqueous solubility is not as accurate as is the prediction of octanol/water partitioning.

3.5 The chemical mechanisms according to the model underpinning the predicted result (OECD principle 5)
The water solubility of a substance depends on its affinity for water as well as its affinity for its own crystal structure. In general, substances with high melting points have poor solubility in any solvent.

References
- US EPA (2012). On-Line KOWWIN User’s Guide.

Data source

Reference

Reference 1

Reference Type:

other company data

Title:

Unnamed

Year:

2020

Report date:

2020

Reference 2

Reference Type:

other: Estimation software

Title:

Estimation Programs Interface Suite for Microsoft Windows, v4.11

Author:

US-EPA

Year:

2012

Bibliographic source:

United States Environmental Protection Agency, Washington, DC, USA

Materials and methods

Test guideline

Qualifier:

no guideline required

Principles of method if other than guideline:

Estimation of log Kow using KOWWIN v1.68

GLP compliance:

no

Type of method:

calculation method (fragments)

Partition coefficient type:

octanol-water

Test material

Test material information

Constituent 1

Reference substance name:

Valeryl chloride

EC Number:

211-330-1

EC Name:

Valeryl chloride

Cas Number:

638-29-9

Molecular formula:

C5H9ClO

IUPAC Name:

pentanoyl chloride

Test material form:

liquid

Results and discussion

Partition coefficient

Type:

log Pow

Partition coefficient:

1

Temp.:

25 °C

Remarks on result:

other: The substance is within the applicability domain of the model.

Any other information on results incl. tables

SMILES : O=C(CCCC)CL

CHEM : Pentanoyl chloride

MOL FOR: C5 H9 CL1 O1

MOL WT : 120.58

-------+-----+--------------------------------------------+---------+--------

TYPE | NUM | LOGKOW FRAGMENT DESCRIPTION | COEFF | VALUE

-------+-----+--------------------------------------------+---------+--------

Frag | 1 | -CH3 [aliphatic carbon] | 0.5473 | 0.5473

Frag | 3 | -CH2- [aliphatic carbon] | 0.4911 | 1.4733

Frag | 1 | -CL [chlorine, aliphatic attach] | 0.3102 | 0.3102

Frag | 1 | -C(=O)- [carbonyl, aliphatic attach] |-1.5586 | -1.5586

Const | | Equation Constant | | 0.2290

-------+-----+--------------------------------------------+---------+--------

NOTE | | Acetyl halides hydrolyze....estimate questionable! |

-------+-----+--------------------------------------------+---------+--------

Log Kow = 1.0012

Applicant's summary and conclusion

Executive summary:

QPRF: KOWWIN v1.68 (01 Nov. 2013)

 

1.	Substance	See “Test material identity”
2.	General information
2.1	Date of QPRF	See “Data Source (Reference)”
2.2	QPRF author and contact details	See “Data Source (Reference)”
3.	Prediction
3.1	Endpoint (OECD Principle 1)	Endpoint	Octanol-water partition coefficient (log Kow)
		Dependent variable	Octanol-water partition coefficient (log Kow)
3.2	Algorithm (OECD Principle 2)	Model or submodel name	KOWWIN
		Model version	v. 1.68
		Reference to QMRF	QMRF: Estimation of Octanol-Water Partition Coefficient using KOWWIN v1.68 (EPI Suite v4.11)
		Predicted value (model result)	See “Results and discussion”
		Input for prediction	Chemical structure via CAS number or SMILES
		Descriptor values	- Chemical structure - Fragments - Correction factors
3.3	Applicability domain (OECD principle 3)	Domains:
		1) Molecular weight (range of test data set: 18.02 to 719.92 g/mol; On-Line KOWWIN User’s Guide, Ch. 6.2.3 Estimation Domain)	Substance (not) within range (120.58 g/mol)
		2) Fragments: Number of instances of the identified fragments does not exceed the maximum number as listed in Appendix D (On-Line KOWWIN User’s Guide)	fulfilled.
		3) Fragments: Substance has a functional group(s) or other structural features not represented in the training set, and for which no fragment coefficient was developed (Appendix D, On-Line KOWWIN User’s Guide)	applicable.
		4) Correction factors: Number of instances of the identified correction factor does not exceed the maximum number as listed in Appendix D (On-Line KOWWIN User’s Guide)	fulfilled.
3.4	The uncertainty of the prediction (OECD principle 4)	According to REACH Guidance Document R.7a, (Nov. 2012), solubility in water is difficult to model accurately. For this reason, as well as the fact that the experimental error on solubility measurements can be quite high (generally reckoned to be about 0.5 log unit), the prediction of aqueous solubility is not as accurate as is the prediction of octanol/water partitioning.
3.5	The chemical mechanisms according to the model underpinning the predicted result (OECD principle 5)	The water solubility of a substance depends on its affinity for water as well as its affinity for its own crystal structure. In general, substances with high melting points have poor solubility in any solvent.

 

References

- US EPA (2012). On-Line KOWWIN User’s Guide.

 

 

Assessment of estimation domain (molecular weight, fragments, correction factors):

 

SMILES : O=C(CCCC)CL

CHEM : Pentanoyl chloride

MOL FOR: C5 H9 CL1 O1

MOL WT : 120.58

-------+-----+--------------------------------------------+---------+--------

TYPE | NUM | LOGKOW FRAGMENT DESCRIPTION | COEFF | VALUE

-------+-----+--------------------------------------------+---------+--------

Frag | 1 | -CH3 [aliphatic carbon] | 0.5473 | 0.5473

Frag | 3 | -CH2- [aliphatic carbon] | 0.4911 | 1.4733

Frag | 1 | -CL [chlorine, aliphatic attach] | 0.3102 | 0.3102

Frag | 1 | -C(=O)- [carbonyl, aliphatic attach] |-1.5586 | -1.5586

Const | | Equation Constant | | 0.2290

-------+-----+--------------------------------------------+---------+--------

NOTE | | Acetyl halides hydrolyze....estimate questionable! |

-------+-----+--------------------------------------------+---------+--------

Log Kow = 1.0012