Finechem is a software tool to estimate the resource use and environmental impacts of petrochemical production based on the molecular structure, circumventing the need for a process analysis. Due to the limited amount of input information, results cannot replace a thorough process analysis. Nevertheless, Finechem can be of use if there is a lack of process data, e.g.
1. Download R from http://www.r-project.org/. R is available free of charge.
2. Download Finechem.zip. Create a subfolder named "Finechem" in the R folder (note that R is case-sensitive, so an uppercase F is important). Usually, this is a directory named something like "R-2.8.1" inside a directory named "R" (it does not matter if your version number is different). Unpack Finechem.zip and move all files into the Finechem folder.
3. Start R. When first starting, select "File/Change dir..." to set the standard directory of R to be the directory in which the Finechem folder is located. Again, this will be the directory named something like "R-2.8.1" inside a directory named "R".
4. In the menu, select "Packages/Load Package...". Load the "nnet" package.
5. In the menu, select "File/Load Workspace...". Load the "Finechem_models.RData" file from the Finechem folder.
6. Open "Finechem_input.csv", a comma-separated values file. Any text editor will do, although software such as Excel may be more convenient. Insert the input for the chemicals to be modeled. The tool will automatically determine the number of rows and model each chemical. All 10 columns have to be filled. In order, the values to be entered are:
Important Note: This is just a shortened overview of how to apply the tool. The supporting information of the Green Chemistry article linked at the bottom of the page contains a complete description of how to determine the descriptors for a given molecule. This information can be freely accessed even without access to the Green Chemistry journal.
The first row of the file contains the labels for these parameters, do not remove this row. When first opened, "Finechem_input.csv" contains 3 rows of chemical data, for ethanol, tetrahydrofurane and benzaldehyde. These can be overwritten with your data.
7. Copy and paste the following text into the command line in R:
8. After the calculations are finished, open "Finechem.prediction.csv", a comma-separated values file. The output is organized in 6 columns. First, the prediction and uncertainty for the CED, then the same for the GWP and finally the Eco-Indicator 99.
Notes: As the models are black-box models, some results may be obviously wrong, especially if random numbers are entered and not the description of a real chemical. In addition, results are less likely to be reliable for very basic chemicals where very little information is given to the models. Some chemical classes were not covered well by the training dataset, therefore uncertainties will be higher and results may be unreliable. Among these classes where the use of Finechem is not recommended are:
These classes of chemicals may be supported in the future as data become available.
These models only work for petrochemical synthesis. If there is cause to believe that a chemical is synthesized enzymatically, results may not reflect the real situation.
Please read the articles and especially the supplementary material of the "Bridging data gaps" article for more information.
News and Updates:
20.1.2010 Important News: A bug was discovered which sometimes leads to errors if the input file has less than three lines. This can be avoided by replacing the examples in the input file instead of deleting them.
You may send a message to Stavros Papadokonstantakis to indicate your interest in being informed about future releases and updates of the Finechem model. You will receive an email notification if an important update is relased (and only then).
Bugs and Problems:
Please do not hesitate to contact Stavros Papadokonstantakis to report bugs or problems.
G. Wernet, S. Hellweg, U. Fischer, S. Papadokonstantakis and K. Hungerbuhler. Molecular Structure-Based Models of Chemical Inventories using Neural Networks. Env. Sci. Technol., 2008, 42, 6717-6722.
G. Wernet, S. Papadokonstantakis, S. Hellweg and K. Hungerbuhler. Bridging data gaps in environmental assessments: Modeling impacts of fine and basic chemical production. Published in Green Chemistry 2009. DOI: 10.1039/B905558D
Diese Website wird in älteren Versionen von Netscape ohne graphische Elemente dargestellt. Die Funktionalität der Website ist aber trotzdem gewährleistet. Wenn Sie diese Website regelmässig benutzen, empfehlen wir Ihnen, auf Ihrem Computer einen aktuellen Browser zu installieren. Weitere Informationen finden Sie auf
The content in this site is accessible to any browser or Internet device, however, some graphics will display correctly only in the newer versions of Netscape. To get the most out of our site we suggest you upgrade to a newer browser.