PhD student Spardha Virendra Jhamb (now postdoc) and Associate Professor Xiaodong Liang. Photo by Christian Ove Carlsson.

Virtual prediction of bio-degradability

Thursday 08 Oct 20
by Morten Andersen


Xiaodong Liang
Associate Professor
DTU Chemical Engineering
+45 45 25 28 77
Companies striving to improve the sustainability of their products are helped by software developed in the KT Consortium. The project was initiated by industry member Syngenta.

Some companies manufacture products that will unavoidably end up in the environment and should thus preferably be readily biodegradable. Prompted by a member company, Syngenta, researchers in the KT Consortium have developed a new tool able to predict the biodegradability of most new components that a company might consider to use.

“As long as the molecular structure of the compound under consideration is known and fairly simple, we are able to calculate its biodegradability with high accuracy. The uncertainty in the method increases, the more irregular the structure of the new substance is, and the less it resembles known structures,” says Associate Professor at DTU Chemical Engineering Xiaodong Liang. He has supervised the development which has been carried out by PhD student Spardha Virendra Jhamb and visiting student Irene Hospital, Universite de Lorraine, France.

The tool produces a biodegradability value with two or three decimals, usually in the range from 0 to 1.

Xiaodong Liang further notes that for most practical purposes a company would normally be happy to just know whether the given substance is readily biodegradable (values above 0.5), non-biodegradable (values below 0.2) or somewhere in-between.

May replace costly lab experiments

Owned by ChemChina, Syngenta is a leading provider of agricultural chemicals and seeds, supplying farmers worldwide. The company is committed to improving the sustainability of its products in accordance with the United Nation’s Sustainable Development Goals. These efforts involve improving and documenting biodegradability. Here, the idea of building a virtual tool for prediction of the biodegradability properties came up—supplementing or possibly even replacing costly and time-consuming lab experiments.

As Syngenta has been a KT Consortium member for a decade, it was a natural step for the company to approach the group at DTU Chemical Engineering.

“The first communication on the idea took place in January 2018, so development has been fairly rapid,” says Xiaodong Liang.

The first challenge was obviously to establish what is meant by the term biodegradable. It was decided to measure the biodegradability of a chemical using a combination of two oxidation parameters.

The first parameter is BOD (Biochemical Oxygen Demand), which is the amount of oxygen demanded by aerobic bacteria when metabolising a test compound. Common practise of labs in many countries have established measuring BOD after five days of incubation at 20°C as a global standard.

COD (Chemical Oxygen Demand) is the amount of oxygen consumed during oxidation of a test compound with hot, acidic dichromate, providing a measure of the total amount of oxidizable matter present in the sample.

High rate of success

The BOD/COD ratio can be seen as a measure for biodegradability. A 1.00 value would indicate that the compound is fully biodegradable.

With this in place, the next step was to develop a model for biodegradability estimation. The chosen model is based on the socalled group-contribution methods, which are simple, easy to use and widely used in chemical process system engineering.

Finally, results from the model were compared with experimental data derived partly from literature, and partly from Syngenta’s extensive database.

“The results are promising, with a regression coefficient of 0.92 and an average absolute error of 0.05, and a 96 per cent success of classifying a compound as ‘readily biodegradable” or not,’ says Xiaodong Liang.

He further notes that the methodology can be applied not only by Syngenta:

“The group-contribution methods are quite general. We will be able to use them for different properties and object functions in other DTU Chemical Engineering projects. Also, we are about to integrate them in our ICAS software (Integrated Computer Aided Systems, the main deliverable from the KT Consortium), thereby allowing other consortium member companies to benefit.”

Related videos  

video thumbnail image

Computerome - CFIM

Photo: Mikal Schlosser

video thumbnail image

Show more

News and filters

Get updated on news that match your filter.