Vinasses, residue of fermentation containing biomass of Corynebacterium glutamicum

Partition coefficient

Link to relevant study record(s)

Reference

Reference 1

Endpoint:

partition coefficient

Type of information:

other: expert statement

Adequacy of study:

supporting study

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

other: Expert statement The general term Vinasses is used as a synonym/surrogate for the substance to be registered.

Principles of method if other than guideline:

Expert statement

GLP compliance:

no

1. STATEMENT

 

In result of a theoretical assessment of the testing options to determine the n-octanol/water partition coefficient of Vinasses it was concluded that it is technically not possible to determine a reliable experimental log Pow value to be used for hazard and risk assessment purposes. The main reasons leading to this conclusion can be found in the complex and unknown nature of Vinasses (UVCB substance) in combination with technical limitations of analytical methods and procedures.

 

Also, due to the variability, diversity and complexity of the constituents of Vinasses, and because full details on composition and chemical structures are not known, it was concluded that calculation methods can neither be applied to derive estimates of log Pow values for Vinasses.

 

 

2. DETAILED EXAMINATION OF THE ANALYTICAL TASK:

 

2.1. SCOPE

To evaluate the possibility of determining the n-octanol/water partition coefficient of Vinasses by aid of any of the conventional test methods, i.e. shake flask method (slow-stirring method), HPLC-method and estimation method (via the solubility in the pure solvents) as well as calculation methods (QSARs).

 

 

2.2. BACKGROUND INFORMATION ON THE SUBSTANCES OF INTEREST

 

Vinasses are by-products obtained after the fermentation of molasses/sugar/other feedstock, using different microbial strains, in the production of alcohol, yeast and other organic substances. Due to the complex nature of the biological raw materials in combination with the use of different micro-organisms during the fermentation process, leading to an inherent variability of the composition of Vinasses, these substances have been categorized as typical substances of Unknown or Variable composition, Complex reaction products or Biological materials (UVCB).

 

In a thorough search for the most appropriate description of Vinasses, the Vinasses Consortium has identified five different types, or sub-groups, of Vinasses, primarily based on the production process (potential biomass removal and salt-enrichment) and on the macro-composition.

 

The typical macro-composition of Vinasses can best be described by:

Ash residue

Total organic carbon

Total nitrogen

P

K

S

Na

Another parameter used for identification is the pH-value.

 

 

2.3. THEORETICAL CONSIDERATIONS

 

As a consequence of the complex nature of Vinasses, i.e. the presence of a wide variety of both organic and inorganic constituents, it can be stated that none of the conventional guideline testing methods will be suited to determine the n-octanol/water partitioning coefficient of the substances as a whole.

In more detail, testing according to the recommended OECD and EU guidelines is in principle not possible for the following reasons:

 

 

- General considerations:

Following the definition of the n-octanol/water partition coefficient, all test methods require that measurements on ionizable substances should be made only in their nonionized form (free acid or free base). Due to the nature of Vinasses it can be assumed that some constituents, e.g. the N-containing compounds or amino acids, contain ionizable groups, whereas others don’t. This assumption can be supported by the pH ranges indicated for the various Vinasses, i.e. min. pH 3 and max. pH 9. Therefore it can be considered practically impossible to assure that all ionizable constituents are present in their non-ionized form during the course of the test at one single pH. The performance of multiple tests at different pH-values is however not an option as the experimental test designs are either developed for pure substances or for mixtures containing comparable substances (details, see below).

Also, it is general knowledge that the presence of certain substances, e.g. inorganic salts, can influence the solubility of other substances (constituents of substances) in a solvent.

 

 

- Shake Flask Method (EU A.8, OECD 107; log Pow range -2 to 4):

This method is applicable for relatively pure substances only in order to fulfill the applicability requirement of the Nernst partition law, which constitutes the underlying principle of the shake flask method. The presence of various (soluble) constituents in one or both phases may affect the results due to deviations from the partition law (e.g. caused by dissociation or association of the dissolved molecules). As stated above, Vinasses are very complex substances, a.o. with a high degree of inorganic content, which may alter the composition/purity of the test phases. Therefore, an optimal partitioning of the various constituents in the two phases according to Nernst’s law cannot be guaranteed.

 

Additionally, this test requires substance-specific analytical methods that are sensitive enough to determine very low concentrations of the various constituents in either phase. For complex substances, a.o. UVCB’s such as Vinasses, it is very often not possible to develop such sensitive specific analytical methods, partly because the great number of constituents cannot be adequately resolved and also because the detector sensitivity is not the same towards all of the various constituents possibly present.

 

Alternatively, non-specific analytical methods, such as total organic carbon (TOC) analysis or gravimetric analysis (mass difference of loading rate and undissolved fraction after drying) could be considered as tools to determine the concentration of dissolved substance in the test phases. However, both approaches have downsides:

- TOC analysis is quite suitable to determine concentrations in the aqueous phase, but for analysis of the dissolved Vinasses fraction in the n-octanol phase it is not considered a suitable technique due to a high background value from the n-octanol phase.

- With respect to gravimetric analysis, it is considered technically challenging to remove all residual solvent from the undissolved fractions, especially n-octanol. In case of water, filtration before drying could be a useful aid to remove as much as possible of the solvent and thus obtain a sensible value for the mass loss of in the water fraction. However for n-octanol filtration seems not a suitable procedure to decrease the residual content, mainly due to its higher viscosity as compared to water, which will hinder optimal flow through filters, if not even clog them. A consequence of incomplete removal of residual solvent is that the weight of the undissolved fraction may be overestimated, and thus indicate a false mass loss in the respective phase.

- For either non-specific analysis method it is questionable if the approach is in line with the principle of the test, i.e. partitioning of 1 constituent in one or two phases. Even if TOC or gravimetric analysis would be suitable for the determination of substance concentrations in the n-octanol phase, the dissolved fraction in water is very likely going to contain different species than the n-octanol phase. As a consequence, the derived log Pow may be based on a wide variety of constituents which not necessarily are dissolved in both phases (pure substances: the solubilities in both phases are related to each other to derive log Pow).

 

A similar reasoning for non-applicability based on partition law and analytical demands is true for an alternative flask method, i.e. the

slow-stirring method (OECD123, log Pow range up to 8.2).

Furthermore, this method was developed for the determination of log Pow values of highly hydrophobic substances. Vinasses contain a variety of sugars which commonly are soluble in water, so log Pow values of constituents are expected below the range for (highly) hydrophobic substances (log Pow > 4).

 

In conclusion, the shake flask method and the slow stirring method are not considered suitable methods to generate reliable results for the n-octanol/water partitioning behavior of Vinasses that could be useful for hazard assessment purposes.

 

 

- HPLC method (EU A.8, OECD 117; log Powrange 0.3 to 6.5):

This method is preferably applied to sufficiently pure substances in order to allow for assignment of peaks in the chromatograms to the respective constituents. For complex substances or mixtures, meaningful results can be obtained, provided that the various constituents present can be adequately resolved and that the analytical detector used has the same sensitivity towards all the constituents present. Also, in case of a great number of peaks in the chromatogram identification of individual peaks may be needed to determine the upper and lower ends of a log Powrange, or to allow for e.g. grouping of similar constituents. In view of the uncertainties of the composition of the Vinasses, such efforts are not considered useful.

 

Furthermore, the experimental design of the test demands that the substance is completely dissolved in mobile phase or a solvent compatible with the mobile (and stationary) phase. In the case of Vinasses, the inorganic constituents, and maybe even some organic constituents, are not going to fulfill this criterion, thus only the soluble portion of the substances would be available for determination of log Powby the HPLC-method. Also, remaining undissolved substance in test solutions can harm the HPLC system and/or alter the stationary phase.

 

Another argument for the non-applicability of the HPLC-method is the fact that it is not known which constituents are potentially ionizable (see also general considerations), thus defining the correct test conditions, i.e. at which pH the test should be performed, is not possible.

 

For technical and practical reasons, the HPLC-method is not considered a suitable method to generate reliable results for the n-octanol/water partitioning behavior of Vinasses that could be useful for hazard assessment purposes.

 

 

- Estimation method (as mentioned in OECD 117, preliminary estimate):

This method is often used to derive a preliminary estimate of the log Pow of a substance from its solubility in the individual, pure solvents. As is the case for the shake flask method, substance-specific analytical methods are required that are sensitive enough to determine also very low concentrations in either phase. For complex substances, a.o. UVCB’s such as Vinasses, it is very often not possible to develop such sensitive specific analytical methods, partly because the great number of constituents cannot be adequately resolved and also because the detector sensitivity is not the same towards all of the various constituents possibly present.

 

Considerations on the use of non-specific analytical methods can be found in the shake flask method section (see above).

 

In conclusion, the estimation method (from solubility in pure solvents) is not considered a suitable method to generate reliable results for the n-octanol/water partitioning behavior of Vinasses that could be useful for hazard assessment purposes.

 

 

- Calculations methods (as described in the Annex to OECD 117):

Most calculation methods are based on the theoretical fragmentation of the molecule of interest into suitable substructures for which reliable log Pow increments are known. After summation of all fragment values and applying the correction terms for intramolecular interactions (such as H-bonds) the log Pow estimate is derived. For simple molecules of low molecular weight and with one or two functional groups, the reliability of calculation models is considered to be good, whereas it decreases with increasing complexity of the substance of interest.

 

In order to perform calculations, details on the chemical structures of the constituents of substances are required, e.g. to derive SMILES notations as input values for QSAR’s. Additionally, the composition of the substance should be well described to be able to derive a weight-averaged log Powvalue for hazard assessment purposes. In the case of Vinasses, due to the variability, diversity and complexity of the constituents present, neither criterion is fulfilled, because full details on composition and chemical structures are not known.

 

Therefore, the calculation methods (QSAR’s) are not considered suitable methods to generate reliable results for the n-octanol/water partitioning behavior of Vinasses that could be useful for hazard assessment purposes.

 

Reference 2

Endpoint:

partition coefficient

Data waiving:

study technically not feasible

Justification for data waiving:

other:

Description of key information

not applicable

Additional information

Due to the complex inorganic and organic composition of the substance, the partition coefficient of the substance as such is not possible to be determined experimentally nor estimated using QSAR. Detailed explanation is given in the waiving statement and supporting expert statement.

In the result of the theoretical assessment it is concluded that determination of the reliable experimental n-octanol/water partition coefficient of the substance is not technically feasible.

Currently available testing methods (and guidelines): Shake Flask Method (EU A.8, OECD TG 107), HPLC Method (EU A.8, OECD TG 117), Slow-Stirring Method (OECD TG 123) and estimation of the partition coefficient based on individual solubilities in water and n-octanol (EU A.8, Appendix 1) are non-applicable for the substance. The main reasons leading to this conclusion can be found in the complex and unknown nature of Vinasses (UVCB substance) in combination with technical limitations of analytical methods and procedures.

QSAR predictions for each component individually are also not applicable either due to the complex composition and the unknown or inorganic structure of the constituents. Furthermore a rough estimated single log K_ow value or a broad range of log K_ow values for the substance is not meaningful and would not contribute to hazard assessment or classification and labelling of the substance.

The testing possibilities were considered and evaluated also by the testing laboratory WIL Research Europe which confirmed in the expert statement that the determination of the reliable log K_ow value to be used for the hazard and risk assessment purposes for Vinasses is technically not feasible.

The general term Vinasses is used as a synonym/surrogate for the substance to be registered.