Electrospraying Zwitterionic Copolymers as an Effective Biofouling Control for Accurate and Continuous Monitoring of Wastewater Dynamics in a Real-Time and Long-Term Manner

Abstract

Long-term continuous monitoring (LTCM) of water quality can provide high-fidelity datasets essential for executing swift control and enhancing system efficiency. One roadblock for LTCM using solid-state ion-selective electrode (S-ISE) sensors is biofouling on the sensor surface, which perturbs analyte mass transfer and deteriorates the sensor reading accuracy. This study advanced the anti-biofouling property of S-ISE sensors through precisely coating a self-assembled channel-type zwitterionic copolymer poly(trifluoroethyl methacrylate-random-sulfobetaine methacrylate) (PTFEMA-r-SBMA) on the sensor surface using electrospray. The PTFEMA-r-SBMA membrane exhibits exceptional permeability and selectivity to primary ions in water solutions. NH 4 + S-ISE sensors with this anti-fouling zwitterionic layer were examined in real wastewater for 55 days consecutively, exhibiting sensitivity close to the theoretical value (59.18 mV/dec) and long-term stability (error <4 mg/L). Furthermore, a denoising data processing algorithm (DDPA) was developed to further improve the sensor accuracy, reducing the S-ISE sensor error to only 1.2 mg/L after 50 days of real wastewater analysis. Based on the dynamic energy cost function and carbon footprint models, LTCM is expected to save 44.9% NH 4 + discharge, 12.8% energy consumption, and 26.7% greenhouse emission under normal operational conditions. This study unveils an innovative LTCM methodology by integrating advanced materials (anti-fouling layer coating) with sensor data processing (DDPA).

Machine Learning Algorithm Integrated with Real-Time In Situ Sensors and Physiochemical Principle-Driven Soft Sensors toward an Anaerobic Digestion-Data Fusion Framework

“…41,42 Before clustering, all the MEA sensor data were processed with denoising data processing algorithm (DDPA) previously developed to alleviate the data drifting caused by electronic hardware problems (e.g., unstable sensor connections) and/or sensor material deterioration in the AD system (e.g., suspended particle attachment, S-ISM polymer matrix leaching). 25,38 The details of the MLA data set are demonstrated in Table S1. To fully exploit the MAPS and ADSS data, four MLAs (logistic regression (LR), Support vector machines (SVM), Artificial neural network (ANN), and ensembled random forest (RF)) regularly used in wastewater and AD studies were selected given their good prediction accuracy while being trained on water parameters (e.g., pH, conductivity, temperature).…”

“…The Nernstian response maintains the best performance between 0 and 1500 mg/L (∼1200 mg/ L of AD NH the MAPS data errors. 25,38 The electrochemical response had severe noises (Figure S2) with the signal-noise-ration (SNR) of only 1.33 (vs. good signals: >25 25 ). Through the DDPA preprocessing, the discrepancy of Nernstian-based open-circuit potential sensor readings (NH 4 + , pH, and ORP) from the labbased validation results was reduced to 0.11%, 1.33%, and 4.72%, respectively.…”

“…It should be noted that the prediction of the AD processes is still limited due to two major obstacles. First, the harsh environment of AD systems impedes the lifespan of the physical-based MEA sensors (∼1 week) because of the extreme environment (>10000 mg/L COD, >20000 mg/L biomass content) compared to regular wastewater (∼200 mg/ L COD) 38 and, thus, hinders the integrity of the long-term sensor readings (>6 months or even years) in the real-world AD systems. Second, current MLAs still heavily rely on statistics with little consideration of theories and principles associated with physiochemical and biochemical aspects of wastewater and biosystems.…”

Mass Fabrication and Smart Deployment of “Calibration-free" Miniature Solid-State Potentiometric Sensors towards Digital Water Infrastructure

“…It should be underscored that CF-MSP-ISM sensors only had a single calibration before being deployed in wastewater for 14 days and functioned accuracy without additional calibration. Sensors in other studies were calibrated on a daily basis to minimize data drifting effects and maintain sensor reading accuracy 9,17 . This exceptional long-term wastewater monitoring performance was mainly ascribed to the enhancement of anti-fouling capability (Fig.…”

Electro Fluid Dynamics: A Route to Design Polymers and Composites for Biomedical and Bio-Sustainable Applications

“…The copolymer and non-solvent solution were co-electrosprayed in order to induce copolymer precipitation. The coated membrane utilizes the microphase separation of the zwitterionic copolymer to form hydrophilic nanochannels acting as pores to mitigate biofouling without compromising the diffusion of primary ions (e.g., NH4+ for NH4+ S-ISE sensors) throughout the ISM matrix and augment the long-term accuracy and stability of the S-ISE sensors in wastewater [ 132 ].…”