application of image processing in the analysis of ph-sensitive optical sensors

Abstract: ph is an important factor in many fields of chemical, biological, and environmental sciences. importantly, the ph level is useful as a qualitative, diagnostic, or surrogate parameter in industrial, physiological, or clinical practice [1]. many methods have been developed for determination of ph but optical sensors (optodes) have several advantages such as simplicity, rapidity, low cost, capability for incorporation into easy-to-use kits for remote and continuous monitoring and the selectivity and sensitivity for trace analysis [2]. chemometric methods have been widely employed for simultaneous determination of ph. optodes and chemometric methods are well-established research areas in analytical chemistry for metal ion and ph determination. however, there are little efforts to employ both tools together [3-5].in this study a novel optical sensor has been proposed for sensitive determination of ph based on immobilization of alizarin red s on a triacetylcellulose membrance. a change in ph of solution caused a color change in the sensor containing the ligand, which was also visible in spectroscopic absorption and digital images that obtains by a smartphone. ph values of solutions were studied by calibration modeling and using application of pixel selection to predict the ph and comparison with experimental data based on absorption spectra and digital images of 26 sensors in solutions with different ph . in other words, pixel of images was employed as input data for the construction of chemometric models, exploiting multivariate tools, and then validated using independent measurements. the studied model showed valuable result and satisfactory output validation parameters, which made the model acceptable that can help researchers to design of new sensors for determination of various parameters such as ph, drugs and metals in aqueous solutions.