Year: 2025
Type: article
Abstract: Though widely used for the accurate determination of host–guest binding constants, traditional fluorescence titrations are often too laborious for high‐throughput (HT) or multivariable screening. Herein, it is demonstrated that standard microplate readers can approximate binding affinities across multiple supramolecular systems, different analytes, and stoichiometries, offering a scalable and time‐efficient alternative to fluorimeters. Using diverse model systems, including a pH‐responsive dye, a Zn 2+ ‐binding fluorophore (8‐hydroxyquinoline‐5‐sulfonic acid) (1:2), a ratiometric anion‐responsive terpyridine complex (ZnCl 2 (BPh‐tpy)) which binds PPi (3:1), and a neutral guest‐binding system (cucurbit[7]uril with proflavine, 1:1), it is shown that plate‐reader measurements yield binding constants that closely mirror those obtained from the fluorimeter. Moreover, the HT multi‐well plate format enables the simultaneous acquisition of large datasets that support statistically robust chemometric analysis. In each case, Support Vector Machine regression models are trained to predict analyte concentration with high accuracy (prediction errors of 2.7–4.3%). Additionally, the methodology provides practical estimations of limits of detection and limits of quantification using the same plate data, further enhancing its utility. This platform preserves analytical rigor and introduces a path toward integrating statistical robustness and machine learning in sensor development, using instrumentation readily available in most research settings at a fraction of the cost.
Source: Analysis & Sensing
Authors: Austin R. Sartori, Anjusha Prakash, Elnaz Safari, Mikhail Zamkov, Pavel Anzenbacher
Institution: Bowling Green State University
Language: English
Subfield: Organic Chemistry
Field: Chemistry
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $3,840
OpenAlex ID: W4413239087