Real-time environmental surveillance of SARS-CoV-2 aerosols

The COVID-19 pandemic continues to cause global morbidity and mortality, with airborne transmission recognized as a predominant infection pathway. Rapid, affordable, and real-time environmental monitoring of SARS-CoV-2 in air has been lacking; existing offline air sampling approaches (e.g., PILS, cyclone samplers, and filters followed by RT-qPCR) are labor-intensive, require skilled personnel, and entail long turnaround times (1–24 hours), limiting timely public health responses. The research question addressed is whether a high-flow aerosol sampler integrated with an ultrasensitive biosensor can enable rapid (minutes) and accurate detection of airborne SARS-CoV-2 at environmentally relevant low concentrations. The study aims to design, build, and validate a pathogen Air Quality (pAQ) monitor capable of near-real-time detection of SARS-CoV-2 aerosols, bridging key technological gaps in high-flow sampling and rapid, specific biosensing for low-abundance airborne viruses. This capability is important for enabling point-of-need surveillance and immediate mitigation actions in indoor environments.

Prior studies employed offline air sampling techniques such as condensation growth-based particle-into-liquid samplers (PILS), wet-wall cyclones, and filter sampling, with detection by RT-qPCR, to identify SARS-CoV-2 RNA in diverse environments including hospitals, public transport, residences, and outdoor air. Although effective for surveillance, these methods provide no real-time feedback and have long processing times. Evidence indicates that high-flow samplers can improve virus recovery by consolidating larger air volumes, increasing detection probability at low concentrations; e.g., higher detection rates were observed at 150 lpm versus 50 lpm in patient rooms. High-flow PILS can collect pathogen-laden aerosols directly into liquid for downstream quantification, but few systems integrate such samplers with real-time sensors. Biosensors have emerged as rapid, specific, and low-cost alternatives to RT-qPCR, showing strong performance in saliva, nasal swabs, and exhaled breath condensate; however, peer-reviewed demonstrations of biosensors for detecting airborne SARS-CoV-2 had been lacking. These gaps motivated integrating a high-flow PILS with a real-time immunobiosensor for airborne virus detection.

System design: The pAQ monitor integrates a batch-type wet-wall glass cyclone (wet cyclone) with a micro-immunoelectrode (MIE) biosensing unit that includes an automated liquid handling system, reagent reservoirs (hypochlorous acid, PBS, and 1% BSA in PBS for calibration), a potentiostat, and a microcomputer. The wet cyclone is connected to a high-flow vacuum pump to sample air at approximately 1,000 lpm. Before sampling, ~15 mL of PBS is added to the cyclone. Ambient air enters tangentially, forming a vortex and a rotating PBS film that captures incoming aerosols on the wetted wall; uncaptured aerosols exit to a HEPA filter. Typical sampling duration is 5 minutes, after which the collected liquid is transferred to the detection unit. Biosensing: Screen-printed carbon electrodes (SPCEs) are functionalized by covalent attachment of a llama-derived nanobody against the SARS-CoV-2 spike protein. Electrochemical detection uses square-wave voltammetry (SWV) from 0 to 1 V at 15 Hz, measuring the oxidation of tyrosine residues in the spike protein at approximately −0.65 V. Electrodes are pre-treated in PBS and blocked with 1% BSA to reduce nonspecific binding. The magnitude of the oxidation peak current at −0.65 V correlates with viral concentration. Sensor baseline is established with 1% BSA in PBS; 2 mL of the cyclone sample is then analyzed, and results are recorded within ~30 s. Cyclone performance modeling: CFD simulations estimated size-dependent collection efficiency, yielding >95% efficiency for particles >1 µm and a 50% cutoff diameter of ~0.4 µm. Chamber experiments: Inactivated SARS-CoV-2 Washington strain (WA-1) was aerosolized in a well-mixed, sealed 21 m³ stainless steel chamber. The wet cyclone was compared with two commercial PILS devices: BioSampler (SKC Inc.) and Liquid Spot Sampler (LSS; Aerosol Devices). Devices sampled simultaneously for 10 minutes to ensure adequate recovery for RT-qPCR. Three virus loading regimes were tested: low (<500 copies/m³), medium (500–10,000 copies/m³), and high (>10,000 copies/m³), with duplicate or triplicate runs. Post-sampling, viral RNA in collection liquids was quantified by RT-qPCR. Air-volume-normalized concentrations and collection media concentrations were compared across samplers. Field sampling in infected households: The pAQ monitor was shipped to two apartments occupied by SARS-CoV-2-positive individuals. A total of seven air samples were collected using the wet cyclone in the apartments, alongside three control samples in a virus-free room. RT-qPCR measured Ct values to assess presence of SARS-CoV-2 RNA. Limit of detection (LoD) and sensitivity: Variant-specific LoD for 5-minute sampling was determined for WA-1, Delta, Beta, and BA.1 strains. Sensitivity was evaluated by comparing MIE detection outcomes against RT-qPCR across replicated aerosolization tests for WA-1 and BA-1. Operational workflow: Prior to each run, baseline SWV is acquired with calibrant (1% BSA in PBS). The sample is introduced to the MIE vial via peristaltic pumps, and SWV is repeated to quantify the oxidation peak height. Automated liquid handling manages calibration, sample introduction, rinsing, and waste disposal.

• The wet cyclone exhibited >95% collection efficiency for particles >1 µm with a cutoff diameter of ~0.4 µm (CFD modeling).
• In 10-minute chamber tests with aerosolized WA-1, the wet cyclone produced on average ~10× and ~50× higher viral RNA concentration in collection media than the BioSampler and LSS, respectively. Under low concentration conditions (<500 copies/m³), only the wet cyclone yielded quantifiable RNA by RT-qPCR; BioSampler and LSS samples were below quantification limits.
• Air-volume-normalized RNA concentrations from the wet cyclone were lower than BioSampler under medium/high conditions but similar to or higher than LSS, consistent with known trade-offs for high-flow samplers (potential underestimation per unit volume due to sampling losses).
• Real-world apartment sampling: All seven air samples from two SARS-CoV-2-positive households were RT-qPCR positive with Ct values 32.7–34.9, whereas three control-room samples were negative, indicating detectable but low airborne RNA in occupied spaces.
• Variant-specific LoD for 5-minute sampling: WA-1 = 35 copies/m³; Delta = 7 copies/m³; Beta = 9 copies/m³; BA.1 = 23 copies/m³. These LoDs align with environmentally relevant indoor concentrations.
• Sensitivity: 77% for WA-1 and 83.3% for BA-1 in aerosol tests.
• The integrated system provides near-real-time operation: 5-minute sampling coupled with ~30-second biosensor detection, using ~1000 lpm high-flow sampling to enhance capture in low-concentration settings.

The study addresses the critical gap in real-time airborne SARS-CoV-2 surveillance by integrating a high-flow wet-wall cyclone sampler with an ultrasensitive MIE biosensor. High-flow sampling enabled concentration of large air volumes, improving detection at low ambient virus concentrations and allowing minute-scale time resolution. Laboratory intercomparisons demonstrated superior or comparable recovery to commercial samplers, particularly under low concentration conditions where high-flow sampling is most advantageous. Field testing in infected households confirmed detection of low-level airborne RNA in occupied spaces while control locations remained negative, reinforcing the system’s real-world applicability for point-of-need surveillance. Although air-volume-normalized concentrations from the wet cyclone may be underestimated relative to low-flow samplers due to potential sampling losses, the system’s goal of timely detection is met, enabling rapid risk assessment and intervention. Variant-dependent LoDs and sensitivities reflect epitope-specific binding differences but remain well within environmentally relevant ranges, supporting broader surveillance utility. Overall, the pAQ monitor provides a practical path toward real-time monitoring in schools, residences, healthcare, and public venues, with potential to inform immediate mitigation strategies.

This work demonstrates a proof-of-concept pathogen Air Quality (pAQ) monitor capable of near-real-time detection of airborne SARS-CoV-2 by combining a ~1000 lpm wet-wall cyclone PILS with a nanobody-functionalized MIE biosensor. The system achieves 5-minute temporal resolution, rapid electrochemical readout, variant-specific LoDs of 7–35 RNA copies/m³, and sensitivities of 77–83%. It outperforms or matches commercial samplers in recovery, especially at low concentrations, and detects SARS-CoV-2 RNA in real indoor environments. The platform is suitable for continuous or grab-sample surveillance and can be extended to multiplex detection of other respiratory pathogens by employing additional target-specific nanobodies. Future work should focus on noise reduction, rigorous performance validation across diverse environments and aerosol matrices, characterization of potential interferents, and optimization for accurate air-volume-normalized quantification.

• High operational noise (75–80 dB) may affect occupant comfort; efforts are underway to reduce noise to <65 dB via low-noise motors and acoustic liners.
• Potential underestimation of air-volume-normalized virus concentrations with high-flow wet cyclone sampling due to evaporative losses, particle losses to chamber or sampler walls, re-entrainment, and particle bounce.
• Variant-dependent biosensor response due to spike protein mutations can alter binding efficiency and signal strength.
• Reported LoDs apply primarily to aerosols >1 µm (near 100% collection efficiency); LoDs for submicron particles will vary with size-dependent recovery.
• The proof-of-concept system requires further testing for robustness across varied indoor environments and aerosol compositions; potential interfering agents in air need comprehensive assessment.