ABSTRACT
Colorimetric assays typically offer a rapid and convenient method to assess analytes that span healthcare monitoring to water quality testing. However, such tests can only provide qualitative results when employed in resource-limited settings or require bulky and expensive equipment such as lab spectrophotometers to allow quantitative measurements. In this paper, we report on the use of a handheld colorimeter to quantitatively determine the concentration of analytes in a manner that is independent of ambient lighting or initial sample color. The method combines the response of the sensor with first-principles modeling that better describes the nature of the assay compared to linear-in-parameters regression modeling that is typically performed in other studies. This method was successfully demonstrated using a number of colorimetric assays: (1) determination of solution pH using a universal indicator, (2) quantification of the DNase presence using a DNA-gold nanoparticle assay, and (3) quantification of the concentration of the antibiotic tetracycline using a cell-based assay.
ABSTRACT
As a proof of concept, a rapid assay consisting of a cell-based biosensor (CBB) panel of pure bacterial strains, a fluorescent dye, and partial least squares (PLS) modeling was developed to assess the nitrification inhibition potential of industrial wastewater (WW) samples. The current standard method used to assess the nitrification inhibition potential is the specific nitrification rate (SNR) batch test, which requires approximately 4 h to complete under the watch of an experienced operator. In this study, we exposed the CBB panel of seven bacterial strains (nitrifying and non-nitrifying) to 28 different industrial WW samples and then probed both the membrane integrity and cellular activity using a commercially available "live/dead" fluorescent dye. The CBB panel response acts as a surrogate measurement for the performance of nitrification. Of the seven strains, four (Nitrospira, Escherichia coli, Bacillus subtilis, Bacillus cereus) were identified via the modeling technique to be the most significant contributors for predicting the nitrification inhibition potential. The key outcome from this work is that the CBB panel fluorescence data (collected in approximately 10 min) can accurately predict the outcome of an SNR batch test (that takes 4 h) when performed with the same WW samples and has a strong potential to approximate the chemical composition of these WW samples using PLS modeling. Overall, this is a powerful technique that can be used for point-of-use detection of nitrification inhibition.
