Registration Dossier

Physical & Chemical properties

Partition coefficient

Currently viewing:

Administrative data

Link to relevant study record(s)

partition coefficient
Type of information:
Adequacy of study:
key study
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
results derived from a valid (Q)SAR model and falling into its applicability domain, with adequate and reliable documentation / justification
Justification for type of information:
EPISuite (v4.11)

2. MODEL (incl. version number)
KOWWIN v1.68

1,2-Benzenedicarboxylic acid, di-C8-10-alkyl esters
1,2-Benzenedicarboxylic acid, di-C8-10-alkyl esters

A reliable QSAR model was used to calculate the partition coefficient of dihexyl ether. The logKow value was calculated using the KOWWIN v1.68 module embedded within the EPISuite computer model. KOWWIN™ estimates the log octanol-water partition coefficient, log Kow, of chemicals using an atom/fragment contribution method. EPISuite and its modules (including KOWWIN) have been utilized by the scientific community for prediction of phys/chem properties and environmental fate and effect properties since the 1990’s. The program underwent a comprehensive review by a panel of the US EPA’s independent Science Advisory Board (SAB) in 2007. The SAB summarized that the EPA used sound science to develop and refine EPISuite. The SAB also stated that the property estimation routines (PERs) satisfy the Organization for Economic Cooperation and Development (OECD) principles established for quantitative structure-activity relationship ((Q)SAR) validation.
The EPISuite modules (including KOWWIN) have been incorporated into the OECD Toolbox. Inclusion in the OECD toolbox requires specific documentation, validation and acceptability criteria and subjects EPISuite to international use, review, providing a means for receiving additional and ongoing input for improvements. KOWWIN is listed as one of the QSARs for use in predicting partition coefficient (Kow) of organic compounds values in the literature referenced in Guidance on information requirements and chemical safety assessment Chapter R.7a: Endpoint specific guidance. In summary, the EPISuite modules (including KOWWIN) have had their scientific validity established repeatedly.
- Defined endpoint and unambiguous algorithm:
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 h fragment or group are summed together to yield the log P 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 P values. KOWWIN’s "reductionist" fragment constant methodology (i.e. derivation via multiple regression) differs from the "constructionist" fragment constant methodology of Hansch and Leo (1979) that is available in the CLOGP Program (Daylight, 1995). See the Meylan and Howard (1995) journal article for a more complete description of KOWWIN’s methodology.
- Defined domain of applicability:
According to the KOWWIN documentation, there is currently no universally accepted definition of model domain. However, the documentation does provide information for reliability of the calculations. Estimates will possibly be less accurate for compounds that 1) have a MW outside the ranges of the training set compounds and 2) and/or that have more instances of a given fragment than the maximum for all training set compounds. It is also possible that a compound may have a functiona
l group(s) or other structural features not represented in the training set, and for which no fragment coefficient was developed.
- Appropriate measures of goodness-of-fit and robustness and predictivity:
Total Training Set Statistics: Number of substances in dataset = 2447, Correlation coef (r2) = 0.982, Standard deviation = 0.217, Absolute deviation = 0.159, Avg. Molecular Weight = 199.98.
Training Set Estimation Error: within ≤ 0.10 - 45.0%, within ≤0.20 - 72.5%, within ≤ 0.40 - 92.4%, within ≤ 0.50 - 96.4%, within ≤ 0.60 - 98.2%.

As described above, according to the KOWWIN documentation, there is currently no universally accepted definition of model domain. In general, the intended application domain for all models embedded in EPISuite is organic chemicals. Specific compound classes, besides organic chemicals, require additional correction factors. Indicators for the general applicability of the KOWWIN model are the molecular weight of the target substance and the identified number of individual fragments in comparison to the training set. The training set molecular weights are within the range of 18.02 - 719.92 with an average molecular weight of 199.98 (Validation set molecular weights: 27.03 - 991.15 and average of 258.98).
The molecular weight range of the 1,2-Benzenedicarboxylic acid, di-C8-10- alkyl ester isomers is 390 - 447, which falls within the range of both, the training set and the validation set.

The maximum number of instances of that fragment in any of the 2447 training set compounds and 10946 validation set compounds (the minimum number of instances is of course zero, since not all compounds had every fragment) are available in the model documentation. The following (max.) numbers of fragments were found in the target chemical, the respective maximum training set numbers for each fragment are given in the last column:
TYPE | NUM | Fragment | COEFF | VALUE | Max. NUM training set (validation set)
Frag | 2 | -CH3 [aliphatic carbon] | 0.5473 | 1.0946 | 13 (20)
Frag | 14, 16, 18 | -CH2- [aliphatic carbon] | 0.4911 | 7.8576 | 18 (28)
Frag | 6 | Aromatic Carbon | 0.2940 | 1.7640 | 24 (30)
Frag | 2 | -C(=O)O [ester, aromatic attach] |-0.7121 | -1.4242 | 2 (2)
Const | | Equation Constant | | 0.2290

MOL FOR: C26 H42 O4
MOL WT : 418.62
Log Kow(version 1.69 estimate): 9.52

MOL FOR: C28 H46 O4
MOL WT : 446.68
Log Kow(version 1.69 estimate): 10.50
Experimental Database Structure Match:
CAS Num : 000084-77-5
Exp Log P: 9.05
Exp Ref : ELLINGTON,JJ (1999)

MOL FOR: C24 H38 O4
MOL WT : 390.57
Log Kow(version 1.69 estimate): 8.54
Experimental Database Structure Match:
CAS Num : 000117-84-0
Exp Log P: 8.10
Exp Ref : ELLINGTON,JT & FLOYD,TL (1996)

Out of the identified fragments, none did exceed the maximum number found in the training set substances, as well as in the validation set. In addition to the fragment identification, no correction factors had to be applied. The results are further supported by the experimental database structure matches for dioctyl phtalate and didecyl phtalate. As a result 1,2-Benzenedicarboxylic acid, di-C8-10-alkyl esters would not be considered outside the estimation domain.

Based on the experimental difficulties for certain compound classes, the KOWWIN calculations are fit for the purpose of identifying a certain partition coefficient range. In case of the underlying target substance, the calculated logKow values are in a similar range (8.54 - 10.5) as the obtained experimental results and, hence, adequate for the purpose of classification and labelling and/or risk assessment. The estimated logKow (calculation based on fragment contribution) was ≥ 8.54.
The representative SMILES notations used for the predictions were: CCCCCCCCCCOC(=O)c1ccccc1C(=O)OCCCCCCCC, CCCCCCCCCCOC(=O)c1ccccc1C(=O)OCCCCCCCCCC, and CCCCCCCCOC(=O)c1ccccc1C(=O)OCCCCCCCC.
The KOWWIN predicted partition coefficient value is considered valid and fit for purpose.

Documentation of the KOWWIN model is provided in the following references:
Akamatsu, M., Y. Yoshida, H. Nakamura, M. Asao, H. Iwamura and T. Fujita. 1989. Hydrophobicity of di- and tripeptides having un-ionizable side-chains and correlation with substituent and structural parameters. Quant. Struct.-Act. Relat. 8: 195-203.
Akamatsu, M., S.I. Okutani, K. Nakao, N.J. Hong and T. Fujita. 1990. Hydrophobicity of N-acetyl, di- and tripeptide amides having unionizable side chains and correlation with substituent and structural parameters. Quant. Struct.Act. Relat., 9: 189-194.
Akamatsu, M. and T. Fujita. 1992. Quantitative analyses of hydrophobicity of di- to pentapeptides having nonionizable side chains with substituent and structural parameters, J. Pharm. Sci. 81: 164- 174.
Barbato, F., G. Caliendo, M.I. LaRotonda, P. Morrica, C. Silipo and A. Vittoria. 1990. Relationships be tween octanol-water partition data, chromatographic indexes and their dependence on pH in a set of beta-adrenoceptor blocking agents. Farmaco, 45:, 647-663.
Daylight. 1995. CLOGP Program. Daylight Chemical Information Systems. Von Karman Ave.,Irvine, C A 92715. (web-site as of March 2008:
Hansch, C and Leo, A.J. 1979. Substituent Constants for Correlation Analysis in Chemistry and Biolo gy; Wiley: New York, 1979.
Hansch. C., A. Leo and D. Hoekman. 1995. Exploring QSAR. Hydrophobic, Electronic, and Steric Constants. ACS Professional Reference Book. Washington, DC: American Chemical Society.
Howard, P.H. and M. Neal. 1992. Dictionary of Chemical Names and Synonyms. Lewis Publishers, Chelsea, MI (ISBN 0-87371-396-6)
Meylan, W.M. and P.H. Howard. 1995. Atom/fragment contribution method for estimating octanol- water partition coefficients. J. Pharm. Sci. 84: 83-92.
Meylan, W.M. P.H. Howard and R.S. Boethling. 1996. Improved method for estimating water solubility from octanol/water partition coefficient. Environ. Toxicol. Chem. 15: 100-106.
Meylan, WM, Howard, PH, Boethling, RS et al. 1999. Improved Method for Estimating Bioconcentrati on / Bioaccumulation Factor from Octanol/Water Partition Coefficient, Environ. Toxicol. Chem. 18(4): 664-672.
Meylan, W.M. and P.H. Howard. 2005. Estimating octanol-air partition coefficients with octanol-water partition coefficients and Henry's law constants. Chemosphere 61:640-644.
Morrison, R.T. and R.N. Boyd. 1973. Organic Chemistry. Third Ed., Boston, MA: Allyn and Bacon, Inc. p. 1133-38.
Nieder, M., W. Stroesser and J. Kappler. 1987. Octanol/buffer partition coefficients of different betablockers", Arzneim.- Forsch., 37: 549-550.
Ribo, J.M. 1988. The octanol/water partition coefficient of the herbicide chlorsulfuron as a function of p
H. Chemosphere 17: 709-15.
Sangster, J. 1994. LOGKOW Databank. A databank of evaluated octanol-water partition coefficients (Log P) on microcomputer diskette. Montreal, Quebec, Canada: Sangster Research Laboratories. Current Internet access available at:
Streitwieser, A. Jr. and C.H. Heathcock. 1985. Introduction to Organic Chemistry. Third Ed., NY: Macmillan Publ. Co. p. 926-32.
Takacs-Novak, K., M. Jozan and G. Szasz. 1995. Lipophilicity of amphoteric molecules expressed by the true partition coefficient. International J. Pharm. 113: 47-55.
US EPA. 1992. Dermal Exposure Assessment: Principles and Applications. EPA/600/8-91-011B, January 1992, Interim Report. U.S. Environmental Protection Agency, Exposure Assessment Group, Office of Health and Environmental Assessment, Washington, DC.
US EPA. [2012]. Estimation Programs Interface Suite™ for Microsoft® Windows, v 4.11. United States Environmental Protection Agency, Washington, DC, USA.
Yamana, T., A. Tsuji, E. Mikyamoto and O. Kubo. 1977. Novel method for determination of partition coefficients fo penicillins and cephalosporins by high-pressure liquid chromatography. J. Pharm. Sci. 66: 747-9.
other: REACH Guidance on QSARs R.6
Principles of method if other than guideline:
Meylan WM and Howard PH (1995) Atom/fragment contribution method for estimating octanol-water partition coefficients. J. Pharm. Sci. 84: 83-92.
Type of method:
calculation method (fragments)
Partition coefficient type:
Key result
log Pow
Partition coefficient:
> 8.5 - <= 10.5
Remarks on result:
other: QSAR predicted value

KOWWIN predicted that that 1,2-Benzenedicarboxylic acid, di-C8-10-alkyl esters have a logKow in the range of 8.5 -10.5.

Description of key information

The calculated log Kow range for 1,2-Benzenedicarboxylic acid, di-C8-10-alkyl esters is 8.5 - 10.5.

Key value for chemical safety assessment

Log Kow (Log Pow):
at the temperature of:
25 °C

Additional information

The logKow has been determined experimentally as ~10.4. Based on the structural composition, the calculated values for the various isomers are considered more reliable opposed to a single value.

The experimental logKow of didecyl phtalate is reported in the literature as 9.05 (Ellington, 1999) and the experimental logKow of dioctyl phtalate is reported as 8.10 (Ellington and Floyd, 1996). The calculations are in the same range as the available experimental information for the two isomers with 10.50 and 8.54, respectively.