Thermodynamics of Task Specific Materials on Facebook

Software and Computational Skills

COSMOTherm software version 3.0, release 15.0 and TURBOMOLE

Software packages by COSMOlogic. In the COSMO-RS methodology, ab initio modelling of the screening charge density on the molecule surfaces is combined with dielectric continuum method with infinite permittivity to model the chemical potential.1 A large number of equilibrium thermodynamic properties are derived from the chemical potential both for pure compounds and for mixtures. TURBOMOLE and COSMOTherm software packages that we acquired recently will be used by our team to predict thermodynamic properties of ionic liquids and to design new molecules. Also, COSMO-RS modelling will be used in predictions the vapour pressure of terpenes and secondary aerosols, both pure and of their mixtures with water, and their air-water partition coefficients. In addition, the adsorption of the studied molecules at the air-water interface may be studied.

1Klamt, A. Conductor-like Screening Model for Real Solvents: A New Approach to the Quantitative Calculation of Solvation Phenomena J. Phys. Chem. 1995, 99, 2224-2235.

Mathematical Gnostics Toolbox

This is a set of functions written in Octave. Special care is taken to make these functions compatible across a wide range of Octave versions. The functions can be used both interactively and as a part of user-written programs. Several tools are ready to use.

Marginal analysis allows for testing homogeneity of data samples. If a sample is homogeneous, tolerance intervals and intervals of typical data may be estimated. These intervals are useful for testing agreement between data sets as well as for critical evaluation.

Estimation of distribution function by robust nonparametric method. It is a kernel estimation algorithm where the kernel is derived from theory and the smoothness of fit is based on the equality of entropy of the data sample and the distribution. It is suitable both for unimodal and multimodal distributions. It is routinely used for estimation of particle size distribution in atmospheric aerosols.

Robust linear regression is an extension of a well-known technique of regression along an influence curve where the influence curve is derived from mathematical gnostics and the minimum penalty estimate may also be used for improving the results. Polynomial regression and regression of excess quantities by Redlich-Kister polynomials are directly available.

Nonlinear robust regression is an extension of the above mentioned method to the nonlinear cases. An example of a directly available algorithm is regression by the Volker-Tamam-Fulcher equation.

Near Real Time Analysis of SMPS Measurement

This software was developed in cooperation with the Laboratory of Aerosols Chemistry and Physics. It analyses the data measured by the scaning mobility particle sizer (SMPS) by methods of mathematical gnostics. The graphical output is automatically presented on a web page. Graphs are generated by gnuplot, scripts for controlling the whole job are written in bash and perl, the web pages are made in PHP. The system is highly configurable and may also be used for off-line analysis of a set of SMPS data files even on a cluster of computers. File locking is implemented in such a way that it works reliably even on NFS and SSHFS file systems. The project was awarded a gold medal in the International Invention Show INOVA 2011 in Zagreb, Croatia, November 2011.

Reporting Tool for the EBAS and GAW-WDCA Databases

This software was developed in cooperation with the Laboratory of Aerosols Chemistry and Physics for the EUSAAR and ACTRIS projects. It serves for submission of aerosol data to the international databases. The tool is developed as an object oriented in perl with some parts written in Octave. It uses internal files in the JSON format and hence is highly configurable. It follows the new GAW-WDCA format. Currently supported instrument is SMPS, support for other instruments will be added within year 2015.

© 2015,  ttsm.icpf.cas.cz, e-mail: wagner AT icpf.cas.cz