Development of a photocatalytic filter to control indoor air quality

Filter prototyping

Photocatalytic filters have been developed and characterized, able not only to break down solid particles, but also, thanks to the use of photocatalysis (9-10-11-12-13-14-15), to oxidize VOCs and kill the bacteria that were mechanically locked onto the filter.

The photocatalytic filter was developed starting from an activated carbon filter with the following characteristics: substrate polyester, active carbon 24%, active carbon surface area 40,000 m²/m², flame retardancy FH2 (EN 60335-2-21), weight 150 ± 15 g/m² (EN 29073-1), thickness 5.5 ± 1.5 mm (EN ISO 9073-2), and air permeability (pressure drop 25 Pa) 1,050 ± 263 L/m²/s (EN ISO 9237).

The filters were treated using sol-gel technology to induce the precipitation of a titanium dioxide film (16-17-18). Sol 121-AB (NextMaterials Ltd.) with the following composition was used: water 95%, titanium dioxide 2%, functionalized silica 1%-2%, isopropanol 0-3%, acids ≈1%, dopant ≈1%, methanol <0.5%, additives (silver salt) 1% and semitransparent.

LEDs used for the experiments were SMD 5050, wavelength 395 nm (purple), voltage 12 V, wattage 240 mW/each.

Antimicrobial activity

To assess in vitro the time-dependent reduction of a microbial population after exposure to different antimicrobial devices, we first checked the antibacterial effectiveness of the filters per se. Treated filters were grouped into 2 lots (Lot A and Lot B photocatalytic filters) and compared to a commercially sourced filter (Commercial filter) (Fig. 1).

Briefly, disc-shaped samples (Ø = 15 mm, surface area = 1.75 cm²) were cut from each kind of filter and transferred into a sterile 24-well plate (Fig. 2). An aliquot of gram-negative Escherichia coli (strain JM109, cat. 53323; American Type Culture Collection [ATCC], Manassas, VA, USA) was transferred onto a sterile lysogeny broth agar plate and incubated at 37°C for 60 minutes to allow the growth of bacterial colonies. After incubation, using a sterile inoculating loop, a single colony was transferred into a 50 mL polypropylene tube (Corning, Fisher Scientific, Rodano, Italy) filled with 5 mL of sterile LB liquid growth medium and incubated overnight (ON) at 37°C under shaking, to produce sufficient microbial suspension. The number of bacteria within the microbial suspension was roughly estimated by measuring the optical density (OD) of the suspension at λ = 600 nm (OD_600 nm) by means of a Nanodrop 2000 spectrophotometer (Thermo Fisher Scientific, Segrate, Italy). Afterwards, the inoculum was prepared by diluting the microbial suspension in sterile LB to an OD_600 nm = 0.02. Bacteria were next allowed to grow until they reached an OD_600 nm ≈0.6 (exponential growth phase). The inoculum was finally centrifuged for 10 minutes at 4,000 g , washed once in sterile deionized water (dH₂O), pelleted again, and then the cells were resuspended in dH₂O to give a bacterial concentration of ≈5 × 10⁸ bacteria/mL.

Following a preincubation at 4°C, samples were brought into contact with the inoculum. The experiments were carried out according to the Assessment of Antimicrobial Activity Using a Time-Kill Procedure (ASTM E2315) (19) and Measurement of Antibacterial Activity on Plastics and Other Non-porous Surfaces (ISO 22196) (20) standards, slightly modified in terms of the bacterial inoculum (10³ times higher bacterial load as compared with standards) and the temperature at which the experiments were performed (4°C instead of 35°C ± 1°C).

Briefly, each test sample (commercial filter, n = 3; photocatalytic filters, n = 2) was inoculated with 1 mL of the bacterial inoculum and incubated at 4°C for different durations (30 and 60 minutes). During the experiments, the photocatalytic filters underwent UV irradiation (Fig. 3). Bacteria inoculated in wells without any filter were used as positive controls (untreated bacteria).

At different time intervals, a 30-µL aliquot of bacterial suspension kept in contact with each sample was removed from the well, serially diluted in sterile dH₂O (1:10, 1:100, 1:1,000), plated onto P60 LBA plates and incubated ON at 37°C to allow the growth of bacterial colonies. After incubation, the number of viable bacteria was evaluated by counting the colony forming units (CFUs) grown on agar plates. The number of viable cells was standardized to the plate area (cm²) and the dilution factor. Finally, data were normalized with respect to the number of viable cells in control samples at each time step, considering the number of untreated bacteria as 100%. Results were expressed as means ± standard error of the mean (SEM).

Evaluation of TSP and VOC abatement

To evaluate TSP and VOC abatement, a square device closed on the sides was developed ad hoc. It consisted of:

Tests were performed in a Glove Box (100 L volume) polluted by one half medium nicotine content cigarette, able to rapidly generate a large amount of TSPs and VOCs.

The following testing procedure was used:

The results obtained were then averaged (n = 2) and calculated as percentage decrease compared with the value at the time of the start of the fan (16 minutes).

Antibacterial effectiveness

A low pressure drop filter was developed and characterized in this work. Due to its photocatalytic behavior (3-4-5-6), such a photocatalytic filter was able to break down solid particles, oxidize VOCs and kill the bacteria that were mechanically blocked onto its surface.

The antibacterial effectiveness of the filters was evaluated following the Time-Kill Procedure (ASTM E2315), assessed by measuring the reduction of the microbial population over time. From visual inspection, the highest antimicrobial activity was observed for both Lot A and B photocatalytic filters. Results are reported in Table I. Figure 4 displays the results of the antibacterial tests carried out on the commercial filter and Lot A photocatalytic filter. It is worthy of note that already after 30 minutes of incubation, any viable bacteria was detected. The maximum efficiency was found already at the shortest duration and was maintained thereafter.

A time-killing effect was observed also for the commercial filter. Indeed, when in contact with this kind of filter, the number of viable cells decreased by up to 41% after 30 minutes of incubation, and it dropped to 36% after 60 minutes, before decreasing to very low levels only after 120 minutes.

VOC and TSP depolluting effectiveness

As regards the results of the experimentation of depolluting with the tested device, the results are shown in Table II and Figure 5 as regards the ability to decrease TSPs, and in Table III and Figure 6 with regard the ability to decrease VOC.

In Figure 5 it is apparent that, in the absence of ventilation (i.e., in static conditions), TSPs were roughly stable over the entire duration of the experiment, whereas a device equipped with Lot A photocatalytic filter induced a dramatic decrease in TSP load over time, until approaching 100% removal after 120 minutes.

Figure 6 shows the time course for VOCs. In the absence of ventilation (i.e., in static conditions), VOCs displayed a slow but significant decrease over time. In contrast, a ventilation system equipped with Lot A photocatalytic filter in the presence of UV irradiation (air purifier) induced an abrupt decrease in VOC content during the first 30 minutes, while the their content was roughly stable thereafter, likely because not all volatile chemicals had been photocatalytically oxidized.