Development of Low-Cost Electrochemical Sensors Coupled with Machine Learning for In-Situ Detection of Heavy Metals in Groundwater
Year: 2026
Type: article
Abstract: Groundwater is the primary source of drinking water for billions of people, yet contamination by arsenic, lead, cadmium, mercury, chromium and copper remains widespread, particularly in South Asia.Conventional laboratory methods such as ICP-MS and atomic absorption spectrometry are accurate but expensive, centralised and slow, which limits the spatial and temporal resolution of monitoring.Electrochemical sensors, especially those based on stripping voltammetry at screen-printed and nanomaterialmodified electrodes, offer a portable, inexpensive and sensitive alternative.Their field performance, however, is degraded by overlapping peaks, intermetallic interference, matrix effects, electrode fouling and drift.Machine learning (ML) provides a data-driven route to resolve these problems by extracting concentration information from complex voltammetric signals and by correcting for drift and interference.This article reviews the health and regulatory background, the sensing principles, low-cost electrode and instrument design, the integration of ML algorithms, and the architecture of in-situ deployments.Key challenges, including data scarcity, model transferability, sensor longevity and validation, are discussed and future directions are outlined.
Institution: Government of Andhra Pradesh
Language: English
Subfield: Electrochemistry
Field: Chemistry
Domain: Physical Sciences
Keywords: heavy metal detection, groundwater contamination, electrochemical sensor, stripping voltammetry, machine learning, in situ monitoring, arsenic, lead, cadmium, mercury, chromium, copper, screen-printed electrodes, nanomaterial-modified electrodes, sensor drift correction, matrix effects, electrode fouling, drinking water safety, low-cost instrumentation
Open Access status: diamond
OpenAlex ID: W7234117239
