Electromagnetic deactivation spectroscopy of human coronavirus 229E

Waveguide fabrication and assembly

The air cooled waveguide was designed to be 3D printed from PLA and metallized using aluminum foil tape. This design methodology has been previously proven in ^12,13,18 and allows for rapid prototyping of waveguide components. Each waveguide was printed in four sections, which allowed the aluminum tape to be applied to a flat surface to reduce wrinkling. The sections of the waveguide were slotted and screwed together. The waveguides each were fed by standard off-the-shelf waveguide launchers and were tapered to accommodate the size of the viral sample tube. To cool the samples, holes were cut into the sides of the waveguide and filled with a grating. This grating allowed air to pass through the waveguide, but contain the RF power inside the guide. For each waveguide these were printed as a separate piece and were metallized on the waveguide side with aluminum tape and on the outside with a conductive copper based paint (MG-Chemicals 843WB) to increase the isolate. Each waveguide had four of these gratings; two had axial fans to force air into the waveguide, and the other two acted as exit ports with ducts to direct the warm air away from the fan intake. Finally, each waveguide included a location for the sample tube. This was a hole in the side of the waveguide that fit snugly to the test-tube and was metallized and had a metallized shielding cap that together contained the RF energy in the waveguide.

Microwave generation and amplification

A signal generator (Anritsu MG3694A) is used to generate microwave tones at desired frequencies. Power amplification stages are used to increase the power of the signal generator tones such that 2W of power is delivered to the experimental waveguide input. Multiple power application configurations were required to cover the large spectrum studied. In the 0.8–8.2 GHz range, five package amplifiers (Analog Devices HMC659LC5) are used: one in series and four in parallel, which are power combined to reach the target power level. In the 8.2–19.5 GHz and 20–40 GHz ranges, a single packaged amplifier (Mini-Circuits ZVE-3W-183+ and Qorvo QPA2640D, respectively), is used for power amplification. The power response of all amplification stages is characterized by sweeping the incident power from the signal generator and measuring the output power on a spectrum analyzer (Anritsu E4446A). This procedure is repeated at every frequency used in the experimental virus deactivation sweep plan. Thereafter, the power amplification characterization data is saved to memory of a digital-assist embedded system which interfaces with the signal generator. This system corrects intrinsic frequency-variations within each power amplification stage by adjusting the incident signal generator power such that the output power will be precisely 2W (33 dBm) at all frequencies.

Virus deactivation experiments

The 0.8–40 GHz spectrum investigated is discretized into 10 sub-bands based on the supported frequency range of each waveguide designation used, which are summarized in Table 2. The experimental range of 0.8– 40 GHz was selected to include and surround those in^2,3 where resonances for comparable viruses were observed. Multiple waveguides were used to cover this band so that each test could be carried out at the waveguide’s fundamental frequency. It was crucial to operate in the waveguide’s fundamental frequency so that the field max was in the center of each viral sample. Each sub-band uses an identical microwave sweep plan, consisting of 10 equally-spaced discrete tones within its respective band. The microwave generation and amplification stage produces each tone with incident power of 2W to the waveguide for 45 seconds, in ascending order, for a total sweep time of 7.5 minutes. The total sweep time was selected to be comparable to that in² which achieved significant viral reduction was observed for. Equal-concentration live virus samples are prepared and split into experimental and control groups. Both groups, control and experimental, each contain three samples so that every experiment (sub-band) can be repeated three times to analyze repeatability and average virus reduction. All samples are stored in an ice bath for the duration of the experiments. Experimental samples are temporarily removed from the ice bath and inserted into the waveguide, which thereafter are exposed to propagating microwave fields according to the sweep plan described. Control samples receive no exposure to microwaves. For every sub-band, the experiment is repeated for three trials. Plaque assay analysis is used to determine the average reduction in active virus of experimental samples relative to the control samples.

Reduced-serum medium heating characterization

Reduced-serum medium (OptiMEM) heating is characterized to determine the amount of virus heating during virus deactivation experiments. A sample is prepared containing equal volume of medium used in experimental trials. Firstly, the temperature of the sample is measured to determine ambient room temperature. Thereafter, the sample is inserted into the experimental set-up where the virus deactivation sweep plan is executed, exposing the sample to microwaves within the waveguides. Upon completion of the sweep plan, the sample temperature is immediately measured to characterize any heating due to microwave exposure. This procedure is repeated for every sub-band included in this study in order to verify that sample heating is sufficiently low and would not contribute to virus deactivation in all cases.

Reduced-serum medium complex permittivity characterization

The open-ended coaxial probe method is used to measure the complex permittivity of the reduced-serum medium. A vector network analyzer (Anritsu MS4644B) is used to measure reflections from the probe tips. Dielectric probes are calibrated using open, short, and deionized water standard measurements. Reduced-serum medium is transferred into a clean 50 mm diameter beaker to prepare a sufficiently large and uniform sample of the liquid. The probe tips are submerged a depth of 10 mm into the medium and measured using the vector network analyzer. Complex permittivity information (dielectric constant, loss tangent) is then computed over frequency using the calibration measurements. An empirical model is generated using the permittivity measurements with a previously developed methodology¹⁴. This model accounts for the frequency-variation of the EM properties of the medium, which significantly improves simulation accuracy.

Virus and cells

HCoV-229E was obtained from BEI Resources (NR-52726) and propagated as previously described¹⁹. HCoV-229E stocks were titrated by standard plaquing assay on Huh7 cells¹⁹. Huh7 cells (JCRB0403) were obtained from the Japanese Collection of Research Bioresources Cell Bank. Cells were cultured in Dulbecco’s minimal essential medium (DMEM) with 10% FBS, 50 U/mL penicillin and 50 \(\mu\)g/mL streptomycin at 37 \(^\circ\)C in 5% CO₂.

HCoV-229E inactivation assays

HCoV-229E was diluted to 1 x \(10^6\) plaque-forming units (PFU)/mL in OptiMEM (ThermoFisher Scientific, 31985062). Aliquots (1 mL) of the diluted virus were distributed into 1.5 mL screw cap tubes (Fisher Scientific 02-681-372) and subjected to the various microwave treatments described. Subsequently, viral infectivity was assessed by plaquing assay. Huh7 cells plated the day before in 12-well plates at a density of 3.5 x \(10^5\) cells/well were infected with serially diluted HCoV-229E samples for 2 hours at 37 \(^\circ\)C. After removing the inoculum, the cell monolayers were overlaid with 1.2% carboxymethylcellulose in DMEM containing 2% FBS and incubated at 33 \(^\circ\)C in 5% CO₂ until 4 days post-infection. Cells were fixed and stained with a crystal violet staining solution (1% crystal violet in 17% methanol in H₂0) to enable visualization of plaques. Plaques were counted to determine the viral titer.