Operando investigation of the synergistic effect of electric field treatment and copper for bacteria inactivation

The study addresses the need for alternatives to chemical disinfectants (e.g., chlorine and antibiotics) due to drawbacks such as disinfection by-product (DBP) formation and antimicrobial resistance (AMR). Electric field treatment (EFT), an electrophysical method that inactivates pathogens via electroporation, offers promise for food and water disinfection but faces challenges including high voltage and energy requirements. Locally enhanced electric field treatment (LEEFT) reduces voltage/energy demand using nanowire-modified electrodes, yet long-term robustness of nanowires limits scalability. Prior work shows EFT can synergize with chemical disinfectants (e.g., chlorine, ozone) to reduce dosages and DBPs by targeting cell membranes. Copper (Cu) is a known biocide that disrupts membranes and intracellular components, but its use in drinking water is constrained by health limits (1.3 mg/L). Previous EFT configured within a copper ionization cell achieved ~6 log E. coli removal at ~0.2 mg/L Cu, highlighting potential for point-of-use treatment. However, the cellular-level mechanism of EFT–Cu synergy remains unclear. This work employs a lab-on-a-chip (LOAC) platform to observe and quantify the EFT–Cu mechanism in situ at microscale, testing varied pulse widths, field strengths, treatment times, and Cu concentrations to define synergistic effects, performance, temporal dynamics, and single-cell responses.

The paper situates EFT among disinfection strategies, noting chlorine’s efficacy but DBP formation and antibiotics’ success but rising AMR. EFT inactivates via electroporation (distinct from electrochemical oxidation), with challenges in voltage/energy and chamber design. LEEFT with nanowires has demonstrated low-voltage, high-log reductions in bacteria, though electrode robustness for long-term operation remains a hurdle. Synergistic applications have shown that EFT with chlorine or ozone can achieve high log inactivation at markedly reduced disinfectant doses, mitigating DBP (e.g., chlorine by-products, bromate) formation. Copper’s antimicrobial action involves membrane damage, lipid peroxidation, and ion uptake rather than primary DNA degradation; it is used in specific contexts (e.g., pools, plumbing) but limited in drinking water by toxicity thresholds, despite being essential at low doses. Prior EFT–Cu systems achieved significant microbial reductions at very low Cu concentrations, motivating mechanistic elucidation at the cell level using microfluidic platforms that enable high-throughput, precise electric field control and imaging.

A lab-on-a-chip (LOAC) device was fabricated by standard photolithography and lift-off on 4-inch Borofloat glass: negative photoresist (Futurrex NR9-1500py) patterning via maskless aligner (Heidelberg MLA 150), development (RD-6), e-beam evaporation of 20 nm Ti and 200 nm Au (Denton Explorer 14), and lift-off in acetone. The chip features symmetric curved-edge gold electrodes generating a near-linear electric field gradient along the x-axis; 30 repeating channels enable parallel replicates and 120 data points per channel due to symmetry. COMSOL simulations informed field gradients. Model organism: Staphylococcus epidermidis (ATCC 12228) grown in nutrient broth at 35 °C ~15 h, washed 3× in 10 mM phosphate buffer (pH 8.5). Chips were pre-coated with poly-L-lysine for bacterial immobilization; ~50–100 µL of cell suspension was deposited for 50 min, then gently rinsed to remove unattached cells. Copper solutions (0–2 mg/L as Cu) were prepared from CuSO4 in DI water; sodium sulfate was added to match conductivity (~10 µS/cm) across concentrations. EFT employed square-wave pulses (50–80 V) with three regimes, all with 20 ms effective treatment time and 0.1% duty cycle: (1) 500 ns width, 500 µs period, 40k pulses; (2) 1 µs width, 1 ms period, 20k pulses; (3) 2 µs width, 2 ms period, 10k pulses. Pulses were applied using an Avtech AV-1010-B pulse generator triggered by a Keysight 33509B. Imaging and analysis: After EFT/Cu application, chips rested typically 2 h before single staining with 2 µM propidium iodide (PI). PI-positive cells (membrane-compromised) were counted as inactivated, accounting for recovery from reversible electroporation during the rest period. Imaging used a Zeiss Axio Observer 7 with DIC and fluorescence channels, CCD camera. MATLAB (R2023a) processed images and computed inactivation percentages versus electric field regions and time. For most experiments, up to 30 channels were analyzed with 95% confidence intervals; time-series experiments (180 min, images every 30 min) had 5 channel replicates due to processing constraints. Statistics included 95% confidence intervals; no a priori power analysis; some channels excluded for defects; experiments were not randomized or blinded.

- Baseline Cu-only on LOAC: 1 mg/L Cu yielded ~15% inactivation; 2 mg/L Cu ~30% (Supplementary Fig. 1). - Example EFT–Cu condition (500 ns pulse width, 500 µs period, 80 V, 20 ms effective time, 0.5 mg/L Cu): in a high-field region (25–37 kV/cm), inactivation averaged ~53% at 37 kV/cm and ~8% at 25 kV/cm, showing strong field dependence. - Lethal electroporation threshold (LET; 50% inactivation) shifts with Cu concentration at 500 ns pulse width: ~38 kV/cm (0 mg/L Cu), ~35 kV/cm (1 mg/L Cu), ~31 kV/cm (2 mg/L Cu), indicating copper lowers the required electric field for lethal electroporation. - Pulse width effect: To avoid ROS/bubble formation, max fields were 39 kV/cm (500 ns), 32 kV/cm (1 µs), 25 kV/cm (2 µs). Despite lower applicable fields at longer pulse widths, overall inactivation increased with longer pulses. With 2 mg/L Cu: at 25 kV/cm, 2 µs achieved ~90% inactivation vs ~35% at 1 µs and ~35% at 500 ns; at 30 kV/cm, 1 µs achieved ~84% vs ~43% at 500 ns. - Synergistic Cu dose-response slopes (percent inactivated per mg/L Cu): without EFT (0 kV/cm), ~16%/mg·L. With EFT: 500 ns—~27%/mg·L at 32 kV/cm and ~38%/mg·L at 35 kV/cm; 1 µs—~29%/mg·L at 30 kV/cm; 2 µs—~29%/mg·L at 25 kV/cm. Increasing pulse width enables similar synergistic slopes at lower field strengths. - Time-resolved synergy (500 ns pulses; Cu-only 2 mg/L; EFT-only 0 mg/L; EFT–Cu 2 mg/L): At 35 and 37 kV/cm, the theoretical additive (Cu-only + EFT-only) reached 50% inactivation at ~165 and ~100 min, whereas measured EFT–Cu reached ~50% at ~35 and ~15 min (up to ~5× faster). At 33 kV/cm and 90 min, theoretical additive was ~23% vs measured EFT–Cu ~77% (synergistic enhancement up to ~300%). EFT-only showed rapid, localized inactivation at high-field regions with minimal residual effect; Cu-only increased slowly and uniformly over time; EFT–Cu combined immediate EFT-induced deaths with ongoing Cu-induced inactivation across the channel. - Single-cell operando analysis: PI saturation times indicated different permeabilization kinetics—Cu-only reached full staining at ~44 s; EFT-only at high field at ~3.3 s; EFT–Cu reached complete saturation at ~21 s after pulses ceased, showing an initial slow uptake followed by a rapid increase, consistent with EFT-enhanced membrane permeability facilitating faster Cu penetration.

The findings demonstrate that EFT increases bacterial membrane permeability, lowering the electric field threshold for lethal electroporation and accelerating copper ion uptake, thereby producing a synergistic inactivation effect. Quantitatively, copper dosing shifts LET to lower field strengths and increases the Cu dose-response slope when combined with EFT. Longer pulse widths likely extend pore open times, permitting achievement of similar synergistic effects at lower field strengths, which can reduce energy input and mitigate ROS/bubble formation. Time-course experiments show that combining EFT with Cu transforms the inherently slow, residual Cu killing into a faster, more efficient process, substantially outperforming the theoretical additive of independent treatments. Single-cell measurements corroborate operando that EFT primes cells for accelerated Cu permeation, distinguishing immediate EFT deaths from delayed Cu-mediated inactivation. These mechanistic insights support customizable operation (balancing Cu dose, field strength, pulse width, and treatment time) for different application constraints, and underscore EFT’s unique compatibility with membrane-active chemical disinfectants compared to other physical methods that target different cellular components.

This study provides operando, microscale visualization and quantification of the synergistic inactivation mechanism of EFT combined with copper using a lab-on-a-chip platform with a controlled electric field gradient. It establishes that copper lowers the lethal electroporation threshold, quantifies enhanced Cu dose-response slopes under EFT, and shows that longer pulse widths enable similar synergistic efficacy at lower fields. Time-series and single-cell analyses reveal that EFT primes bacterial membranes for accelerated Cu uptake, enabling faster and greater overall inactivation than the additive expectation of separate treatments. These results offer actionable guidance for optimizing EFT–Cu disinfection toward lower energy and lower Cu dosages, informing point-of-use and scalable systems. Future research should investigate operation in regimes allowing controlled ROS generation, explore broader pulse parameters and effective treatment times, evaluate water quality factors (conductivity, pH, temperature, turbidity), test additional pathogens and disinfectant combinations, and translate microscale insights to robust, long-term macroscale implementations.

- Operating windows were constrained to avoid ROS generation and bubble formation at higher field strengths, limiting the parameter space explored. - The LOAC platform provides microscale, immobilized-cell observations that may not fully capture bulk/hydrodynamic effects or long-term electrode behavior; robustness and scale-up remain to be validated. - Time-series experiments had reduced replicates (n=5 channels) due to processing/time constraints (versus up to 30 channels typically); some channels were excluded for defects. - Experiments were not randomized and investigators were not blinded; no a priori power analysis was conducted. - Only one model organism (S. epidermidis) and Cu concentrations up to 2 mg/L were tested; broader organismal and water-matrix diversity was not evaluated. - Reactive oxygen species effects and EFT combinations with ROS-generating conditions were not examined in depth in this study.