Research PaperA chemosensor array for the colorimetric identification of some carboxylic acids in human urine samples

- A simple colorimetric sensor array was developed to identify similar carboxylic acids.
- Eleven carboxylic acids were classified in water with 100% model accuracy.
- The array capability to discriminate urine based complex analytes was examined.
- The urines with abnormal content of carboxylic acids were recognized by chemosensor.
- Pattern recognition methods were used to evaluate performance of the sensor array.

A simple colorimetric sensor array containing eight chemosensors from commercial dyes and metal salts was designed to detect a number of important carboxylic acids in human urine. Some common chemometric methods, including principal component analysis (PCA), linear discriminant analysis (LDA) and k-nearest neighbor (kNN) based hierarchical cluster analysis (HCA) were used to test the discriminatory power of the array. The eight-member sensor array perfectly identified eleven carboxylic acids in water with 100% classification accuracy. In order evaluate the potential of sensor array used in biological environments (biofluids), carboxylic acids in human urine were analyzed and 100% classification accuracy was achieved. In addition, the array's performance in the semi-quantitative identification of carboxylic acids was investigated, and the results showed that the sensor array can discriminate seven typical carboxylic acids at concentrations ranging from 100 to 1000 μmol L−1. These results illustrate the potential use of the sensor array for disease diagnosis and other biomedical monitoring applications.

138