The natural rubber foam was prepared by a one-step foaming process, and the effects of density, expansion ratio and closed-cell properties were studied by microscopy. The purpose of this study was to investigate novel semi-vulcanized and fully vulcanized processes for the manufacture of natural rubber and carbon black foams by compression molding using three different blowing agents. The closed cells of natural rubber used as shoe soles are formed by azodicarbonamide (ADC) and organic blowing agents. The experimental design uses azodicarbonamide (ADC) A, B, C and control three blowing agents. Data analysis was performed on images in ImageJ software to obtain more quantitative results. Create experimental data for full vulcanization (AF, BF, and CF) and semi-vulcanization (AH, BH, and CH) processes. Different azodicarbonamide structural models based on digital microscope images are shown. AH, BH and CH with semi-vulcanization process showed higher expansion ratio and cell value. Microscopic observation shows that the rubber foam structure with the highest expansion ratio and cell number is more uniform, with more AH cell connections and a lower density value of 0.6480. Furthermore, AH and CH using ADC exhibited higher values than BH using organic blowing agents, respectively.
Azodicarbonamide is widely used as a gluten fortifier in the flour industry in many countries, but some studies have proved that it is dangerous or unhealthy to people, and it is not suitable to be added to flour. It has become an urgent need to apply fast, convenient and non-invasive technology for safety detection in the process of food analysis. In this paper, the Vis/NIR reflection spectrum analysis technique was used to predict the concentration of azodicarbonamide in flour based on the analysis of physical properties. Spectral data were obtained from 400 to 2498 nm by scanning 101 samples prepared by the stepwise dilution method. Further, the combination of leave-one-out cross-validation and Mahalanobis distance method was used to eliminate abnormal spectral data, and the correlation coefficient method was used to select characteristic bands. Partial least squares method, backpropagation neural network and radial basis function were used to establish prediction models. By comparing the prediction results of the three models, the radial basis function model has the best prediction results, and its correlation coefficient (R), root mean square error of prediction (RMSEP) and performance deviation ratio (RPD) reach 0.99996 and 0.5467, respectively. 116.5858 and 116.5858.

practical application

Azodicarbonamide has been banned or restricted in many countries. This paper presents a method for predicting the concentration of azodicarbonamide in wheat flour, which will be used in a rapid, convenient and non-invasive detection device.