MULTI-SENSOR INTELLIGENT SYSTEM FOR THE DETERMINATION OF VOLATILE ORGANIC COMPOUNDS AND SAFETY ASSESSMENT OF POLYMER MATERIALS
Rubrics: 2. CHEMICAL TECHNOLOGY
Authors:
2.
Kazan National Research Technological University
Russian Federation
Russian Federation
Type:
Article
Pages:
from 59
to 64
Status:
In work
Language:
Russian
Keywords:
POLYMER, MONOMER, SENSORS, THERMOS-OXIDATIVE DESTRUCTION
Abstract (English):
A multisensory intelligent system is proposed to determine volatile organic compounds and assess the safety of polymer materials. The obtained results of the analysis of the surface morphology of the created sensor films, gas-sensitive and operational characteristics allow us to recommend arrays of 6-8 sensors for the analysis of various groups of polymer materials under static and dynamic conditions. Arrays of piezo quartz sensors have been created to detect volatile products of thermal oxidative degradation of polymers. It is shown that the selected polymer and specific sensor films are characterized by high sorption capacity and mass sensitivity with respect to the main volatile markers and classes of volatile food compounds (aldehydes, ketones, alcohols, phenols, etc.), the cross-selectivity of the sensors to a number of analytes under study makes it possible to identify their identification features in a multicomponent sample. A methodological scheme of gas-phase analysis of surfaces of various groups of polymer materials has been developed, taking into account the characteristics of the object of study and the tasks of qualitative and quantitative analysis. It is shown that algorithms for extracting and reducing the dimensionality of data containing information about the studied object allow forming matrices of parameters of the necessary informativeness for the overall assessment of the object and/or certain characteristics (quantitative determination of various odor fractions), demonstrate a significant increase in the efficiency of further use of multidimensional data processing methods. A method of intelligent analysis of chemical images of a multisensory system is proposed for simultaneous solving problems of qualitative (classification and identification, detection of sample falsification) and quantitative analysis (prediction of quantitative indicators) of polymer materials, including ranking parameters of various informativeness by chemical image and using optimized machine learning algorithms to build classification and regression models.
A multisensory intelligent system is proposed to determine volatile organic compounds and assess the safety of polymer materials. The obtained results of the analysis of the surface morphology of the created sensor films, gas-sensitive and operational characteristics allow us to recommend arrays of 6-8 sensors for the analysis of various groups of polymer materials under static and dynamic conditions. Arrays of piezo quartz sensors have been created to detect volatile products of thermal oxidative degradation of polymers. It is shown that the selected polymer and specific sensor films are characterized by high sorption capacity and mass sensitivity with respect to the main volatile markers and classes of volatile food compounds (aldehydes, ketones, alcohols, phenols, etc.), the cross-selectivity of the sensors to a number of analytes under study makes it possible to identify their identification features in a multicomponent sample. A methodological scheme of gas-phase analysis of surfaces of various groups of polymer materials has been developed, taking into account the characteristics of the object of study and the tasks of qualitative and quantitative analysis. It is shown that algorithms for extracting and reducing the dimensionality of data containing information about the studied object allow forming matrices of parameters of the necessary informativeness for the overall assessment of the object and/or certain characteristics (quantitative determination of various odor fractions), demonstrate a significant increase in the efficiency of further use of multidimensional data processing methods. A method of intelligent analysis of chemical images of a multisensory system is proposed for simultaneous solving problems of qualitative (classification and identification, detection of sample falsification) and quantitative analysis (prediction of quantitative indicators) of polymer materials, including ranking parameters of various informativeness by chemical image and using optimized machine learning algorithms to build classification and regression models.
Keywords:
POLYMER, MONOMER, SENSORS, THERMOS-OXIDATIVE DESTRUCTION
POLYMER, MONOMER, SENSORS, THERMOS-OXIDATIVE DESTRUCTION
