Abstract

The objective of the study was to point out the influence of the molecular properties of volatile organic compounds (VOCs) on the intensity of the energetic interactions with activated carbon. The integral adsorption enthalpies of 44 organic species were first measured onto one type of granular activated carbon. Experimental data showed that depending on the nature of the VOC, the adsorption enthalpy may vary from 40 to 80 kJ.moF'. To account for the influence of the molecular properties on the variations observed, quantitative structure propem relationships (QSPRs) were investigated. QSPRs were set up through different statistical approaches which enabled to discriminate the molecular characteristics which have a significant influence on the adsorption energy. Physical data representative of both dimensional and electronic properties of the organic molecules were retained to form the input variable set. As a simplest tool, a multiple linear regression was first investigated. The best linear regression obtained involved 3 explicative variables : the ionization potential, the polarisability and the molar mass. The linear model permits to compute the adsorption energies by less than 15% in error. In a second approach, a non linear model was also attempted, using neural networks. According to the size of the experimental database, the number of neurons in the input layer was restricted to 3. The best neural network was selected after training was achieved with 56 input variable triplets. It is noticeable that the same molecular properties previously involved in the linear regression, also merge in the best neural network. The predictive ability of the neural network is then proved to be comparable to that of the linear regression.

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.