Benzyl benzoate

Henry's Law constant

Link to relevant study record(s)

Reference

Endpoint:

Henry's law constant

Type of information:

(Q)SAR

Adequacy of study:

key study

Study period:

2021-02-11

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 adequate and reliable documentation / justification

Justification for type of information:

1. SOFTWARE
Individual model HENRYWIN included in the Estimation Programs Interface (EPI) Suite v4.1. There is also a standalone version HENRYWIN 3.21 available.

2. MODEL (incl. version number)
HENRYWIN v3.20 included in EPI-Suite v4.11, 2000-2012.

3. SMILES OR OTHER IDENTIFIERS USED AS INPUT FOR THE MODEL
A SMILES NOTATION/ CAS NUMBER was entered in the initial data entry screen. In the structure window, the molecular weight, structural formula and the structure of the input SMILES notation is shown.

4. SCIENTIFIC VALIDITY OF THE (Q)SAR MODEL
a. Defined endpoint: The Henry´s Law Constant (HLC) of organic compounds at 25°C is estimated. The standalone version is also able to calculate the Henry´s Law Constant over a temperature range (0 to 50°C).

b. Explicit algorithm: HENRYWIN requires only a chemical structure to make these predictions. There are two different methods to estimate the HLC, the bond contribution methodology and the group contribution methodology.
The bond contribution methodology used to derive the bond contribution values is identical to that of Hine and Mookerjee (1991). Each compound is split into a summation of the individual bonds which comprise the compound. The summation is set equal to the compound's LWAPC (log water-to-air partition coefficients).
The final bond contribution method SAR equation for HENRYWIN is:
LWAPC = ∑ [ (Bi) (Nj) + (Ci) (Mj) ] [this equals the log reciprocal value of the HLC]
LWAPC is the summation of the bond contribution value of each bond (Bi) times the number of instances of each bond (Nj) plus the correction factor value of each factor (Ci) times the number of instances of each correction factor (Mj).
The group contribution methodology and original group contribution values in HENRYWIN are taken directly from Hine and Mookerjee (1975). Each compound is split into a summation of the individual groups which comprise the compound. For deriving values, the summation is set equal to the compound's LWAPC. In most cases a group is taken to contain one polyvalent atom and the monovalent atoms bonded to it, but the group is characterized by the nature of the atoms to which it is attached as well as those it contains.

c. Applicability domain: Appendix D and E of the help menu of the EPI Suite data entry page provide for each bond contribution and correction factor used in regressing the bond methodology the maximum number of instances of that bond or correction factor in any of the training set compounds.
Ranges for the 442 compound dataset (Appendix G) used for regressing the bond method coefficients (via least-square analysis):
Molecular weight:
Minimum: 26.04 g/mol
Maximum: 451.47 g/mol
Average: 144.64 g/mol

Henry's law constant (atm m3/mol):
Minimum: 5.65*10-14 atm m3/mol (57.25 Pa m3/mol)
Maximum: 2.03*10+1 atm m3/mol (2.05*106 Pa m3/mol)

d. Limits of applicability: Users may wish to consider the possibility that estimates are less accurate for compounds outside the MW range of the training set compounds, and/or that have more instances of a given bond or correction factor than the maximum for all training set compounds (see list in Appendix D for more details).

HENRYWIN estimates two separate HLC values (one using the group method and one using the bond method). The bond contribution methodology splits a compound into smaller units (one atom to another atom only). The bond method includes individual hydrogen bond values; the group method does not. Both the group method and bond method are susceptible to estimates resulting in "Missing Fragments". When a compound is split into groups or bonds, one or more of the resulting groups or bonds may not have a value in the library of available values. The group method is much more likely to have a "Missing Fragment" occurrence (meaning an HLC estimate is not possible).

Experience with HENRYWIN has shown that the difference between bond and group methods can vary by as much as 2 orders of magnitude for some compounds with many functional groups. The estimation from the group method is sometimes preferred unless the bond method uses a correction factor. In general, the bond method is preferred when a correction factor is applied. An independent evaluation (Altschuh et al., 1999) for a diverse set of organic chemicals found the bond method more accurate than the group method. The group method generates inaccurate estimates for certain types of structures, such as hexachlorocyclohexanes.

Using the bond method, a major part of the statistical inaccuracy of the 1376 compound dataset occurs when the LWAPC exceeds 5 (this equals: unitless HLC of 1*10-5, 2.45*10-7 atm m3/mol (25°C), 0.0248 Pa m3/mol).
Any organic compound with a HLC less than 3.0*10-7 atm m3/mol is considered essentially non-volatile from water (Thomas, 1990). The exposure evaluation branch of the U.S. Environmental Protection Agency (OPPT) uses a cut-off of 1.0*10-8 atm m3/mol for HLC estimates when evaluating Pre-Manufacture Notice chemicals (PMNs); any estimate less than the cut-off is considered 1.0*10-8 atm m3/mol.

e. Staticstics for goodness-of-fit:
Bond method:
number in dataset: 442
correlation coeff (r2): 0.977
standard deviation: 0.4
average deviation: 0.249
[all statistical data is related to the LWAPC value]

Group method:
number in dataset: 318
correlation coeff (r2): 0.956
standard deviation: 0.397
average deviation: 0.223
[all statistical data is related to the LWAPC value]

f. Predictivity – Statistics obtained by external validation:
Bond method:
number in dataset: 1376
correlation coeff (r2): 0.79
standard deviation: 1.54
average deviation: 1.00
[all statistical data is related to the LWAPC value]

Group method:
number in dataset: 1031
correlation coeff (r2): 0.85
standard deviation: 1.05
average deviation: 0.85
[all statistical data is related to the LWAPC value]

g. Mechanistic interpretation: The Henry’s law constant is an important factor in determining the environmental fate of chemicals. Indeed, this constant is a fundamental input for fugacity models that estimate the multimedia partitioning of chemicals.
This model is based on the calculation of respective descriptor values, with the help of experimentally derived HLCs, for defined groups that comprise a compound. For modelling each compound is than split into its subgroups and the respective values are summed up to yield the HLC.

h. Uncertainty of the prediction: SUBSTANCE is not highly complex and the rules applied for the substance appears appropriate. An individual uncertainty for the investigated substance is not available. (ANPASSEN, FALLS NICHT ZUTREFFEND)

i. Chemical and biological mechanisms according to the model underpinning the predicted result: No information available.

5. APPLICABILITY DOMAIN
A. Domains:
a. Descriptor domains:
i. Molecular weight: With a molecular weight of XXX g/mol the substance is within / out of the range of the training set (26.04 – 451.47 g/mol).
ii. Structural fragment domain: Regarding the structure, the fragment descriptors used by the program for the estimation are complete and listed in Appendix D, E of the HENRYWIN help file. (ÜBERPRÜFEN UND GGF ANPASSEN)
iii. Mechanism domain: No information available.
iv. Metabolic domain: Not relevant.
b. Structural analogues: No information available.
i. Considerations on structural analogues: No information available.

6. ADEQUACY OF THE RESULT
a. Regulatory purpose: The data may be used under any regulatory purpose.
b. Approach for regulatory interpretation of the model result: If no experimental data are available, the estimated value may be used to fill data gaps needed for hazard and risk assessment. Further the value is used for other calculations.
c. Outcome: The prediction of the Henry´s Law Constant yields a useful result for further evaluation.
d. Conclusion: The result is considered as useful for regulatory purposes.

Principles of method if other than guideline:

Estimation Program Interface EPI-Suite version 4.11: HENRYWIN (v3.20). The bond contribution method is used to calculate the volatility of the substance from the aqueous phase.
The Estimation Program Interface was developed by the US Environmental Agency's Office of Pollution Prevention and Toxics, and Syracuse Research Corporation (SRC). © 2000 - 2015 U.S. Environmental Protection Agency for EPI SuiteTM (Published online in November 2012).

GLP compliance:

no

Key result

H:

0.284 Pa m³/mol

Temp.:

25 °C

Remarks on result:

other: Bond estimation

Validity of the model:

1. Defined Endpoint: Henry's Law Constant

2. Unambinguous algorithm: For benzyl benzoate the following descriptors were applied:

bond estimation method

class	bond contribution description	value
Hydrogen	2  Hydrogen to Carbon (aliphatic) Bonds	-0.2394
Hydrogen	10  Hydrogen to Carbon (aromatic) Bonds	-1.5429
Fragment	1  C-Car	0.1619
Fragment	1  C-O	1.0855
Fragment	12  Car-Car	3.1657
Fragment	1  Car-CO	1.2387
Fragment	1  CO-O	0.0714

3. Applicability domain:

With a molecular weight of 212.25 g/mole the substance is within the range of the training set (26.04 - 451.47 g/mol). Regarding the structure, the fragment descriptors used by the program for the estimation are complete.

4a. Statistical characteristics (bond method):

number in dataset: 442

correlation coeff. (r²): 0.977

standard deviation: 0.4

average deviation: 0.249 [all statistical data related to the LWAPC value]

4b. Statistical characteristics (group method):

number in dataset: 318

correlation coeff. (r²): 0.956

standard deviation: 0.397

average deviation: 0.223 [all statistical data related to the LWAPC value]

5. Mechanistic interpretation:

The Henry's law constant is an important factor in determining the environmental fate of chemicals. Indeed, this constant is a fundamental input for fugacity models that estimate the multimedia partitioning of chemicals.

This model is based on the calculation of respective descriptor values, with the help of experimentally derived HLC's for defined groups that comprise a compound. For modelling each compound is then split into its subgroups and the respective values are summed up to yield the HLC.

Adequacy of prediction:

The result for benzyl benzoate falls within the applicability domain described above and the estimation rules applied for the substance appears appropriate. Therefore the predicted value can be considered reliable yielding a useful result for further assessment.

Conclusions:

The QSAR determination of the Henry's Law Constant for benzyl benzoate using the model HENRYWIN included in the Estimation Program Interface (EPI) Suite v4.11 revealed a value of 0.284 Pa*m³/mol at 25°C for the unaffected molecule of the substance as any decomposition (e.g. hydrolysis) of the substance is not taken into account by the program.

Executive summary:

The QSAR determination of the Henry's Law Constant for benzyl benzoate using the model HENRYWIN included in the Estimation Program Interface (EPI) Suite v4.11 revealed a value of 0.284 Pa*m³/mol at 25°C for the unaffected molecule of the substance as any decomposition (e.g. hydrolysis) of the substance is not taken into account by the program.

Description of key information

The QSAR determination of the Henry's Law Constant for benzyl benzoate using the model HENRYWIN included in the Estimation Program Interface (EPI) Suite v4.11 revealed a value of 0.284 Pa*m³/mol at 25°C for the unaffected molecule of the substance as any decomposition (e.g. hydrolysis) of the substance is not taken into account by the program.

Key value for chemical safety assessment

Henry's law constant (H) (in Pa m³/mol):

0.284

at the temperature of:

25 °C