Figure 1 shows a schematic of the apparatus and time-dependent observation of ozone density during the build- up phase. The ozone density gradually accumulates to 22,500 ppm in 80 minutes. For shorter operation times, the final density will not reach the maximum value. Final density varied depending on the wall temperature and humidity. Ozone density measurement was based on a UV absorption method (RokiTechno Co., Type ZM 20/200). The ozone generator consists of two co-planar-type discharge panels, 50 × 100 mm in width and length covered with a grid of a metallic pattern with square sections 5 × 5 mm, 17 columns by 8 rows in a face-to-face set-up with a 17 mm gap, excited by 20 kHz, 28 kV peak-to-peak sinusoidal voltages. Microscopic flashover was generated from the metallic grid to a ground electrode pattern on the opposite side through the dielectric base plate. A high-density ozone mixture of 32,000 ppm was generated in oxygen (99.5%) and supplied to the discharge region. The ozone-oxygen mixture was conducted through a Teflon-lined tube and injected through two apertures, 7 mm in diameter, located along the center line of the side wall.

The wall temperature was maintained at constant value with PID controlled flexible heaters to maintain the temperature in the internal space at 50˚C. Biological indicators were located at the center of the reactor, at 140 mm from the bottom and each corner inside. The reactor chamber dimensions were 31 L, 360 mm wide, 270

mm deep, and 320 mm high, and made from aluminum with no artificially anodized coating. Steam from distilled water, 0.5 mL∙min⁻¹ (liquid phase), was generated in a perforated brass block, heated to 120˚C and supplied into the reactor chamber. The relative humidity in the reactor chamber was maintained at 72%. The value was determined by the condensation at the lowest temperature from the chamber wall. Ozone mixture working gas was decomposed by using a manganese peroxide-coated carbon honeycomb catalysis, and the outflow to the exterior environment was controlled to be lower than 0.05 ppm. This emission level meets the operational exposure limits in the United Kingdom, other parts of Europe, the United States, Canada, and Japan: 0.1 ppm for regular exposure. The ozone density in the exhaust gas was measured with a GASTEC absorption tube. Deterioration of plastic material was measured with IR absorption using a microscopic Fourier-Transform Infrared (FTIR) spectrometer (Jansen Microscopic-FTIR) and processed with a workstation (WS/IR-8300, Jasco Co.). Attenuated total-reflectance (ATR) data are not available for these samples.

In the first experiment, a biological indicator, made of Schleicher & Schuell filter paper #470, 6.4 × 38.1 mm in width and length, respectively, was used as a carrier for Bacillus atrophaeus ATCC 9372 spores of 2.0 × 10⁶ CFU. ATCC stands for the American Type Culture Collection. CFU stands for colony-forming units, the density of active end-spores to be detected as individual colonies after incubation. Samples of microorganisms were selected from the library of the Raven Biological Laboratories Company.

Although maximum disinfection grade can be obtained with direct exposure, radicals are supplied through a small perforation of a porous sheet covering a plastic sterile casing. Thus, the compatibility study with such sterile packaging is important for use in transportation and for storage. Compatibility involves two conditions: #1. The antibacterial effect is sufficient in the region separated by the porous sheet; #2. The disinfection process will not deteriorate the sheet or the sterile casing. The selected package materials were polyethylene, poly(propylene), glassine paper, and a Tyvek sheet. Glassine paper is commercially available wax-applied cellophane, with small holes approximately of 1 μm in size at a density of 200 holes per 1 mm². Tyvek is a high-density polyethylene non-woven fabric of 0.15 mm in thickness. The density of microscopic apertures in this non-woven fabric is not identified. The selected microorganisms were spore-forming, gram-positive bacteria: Geobacillus stearothermophilus ATCC 7953, Bacillus atrophaeus ATCC 9372, and Bacillus subtilis ATCC 6653. G. stearothermophilus is approved as a standard indicator for autoclaving, which is pressurized, high-temperature steam sterilization, and B. atrophaeus is an approved indicator for gas sterilization using ethylene oxide. From a clinical view, opportunistic pathogens: Aspergillus niger ATCC 16404, Pseudomonas aeruginosa ATCC 9027, Escherichia coli JCM 1649 Staphylococcus aureus ATCC 6538, and a yeast-like fungus, Candida albicans ATCC 10231, were prepared. JCM stands for the Japan Collection of Microorganisms, a library of available microorganisms. The density of the microorganisms was adjusted in the following way. Selected species were incubated in a soybean casein digest (SCD)—fluid culture medium. A small amount of culture fluid was transferred to an SCD-agar culture medium and incubated. A small amount of the cultured sample was diluted in distilled water, and density was adjusted to approximately 10⁶ CFU per 0.1 mL. An amount of 0.1 mL of the suspension was applied to sterilized carriers and formed circular region of 8 to 10 mm in diameter after drying. The carrier materials were a stainless steel plate of 15 × 15 mm² and 1.5 mm in thickness, a slide glass, a filter paper carrier, and a dental impression made of VPS. The hydrophilic surface characteristics and flexibility of VPS allow stronger attachment of microorganisms, compared with a smooth surface, such as a slide glass. The selected material is commercially available Exafine from GC Medical Corp., Japan. This material is equivalent to Ewaflex, marketed by GC America in the United States, Exafine samples were cut out into 15 × 15 mm² samples, including one or two impressions of teeth, Candida albicans ATCC 10231 (2.2 ×10⁶ CFU) was applied onto the carriers. Reference biological indicators were prepared with Candida albicans applied on a stainless plate, and the sterilization time was compared. After the disinfection process, the biological indicator was incubated in 100 mL of soybean casein digest (SCD)—culture fluid medium. In this process, we followed protocols of Japanese Pharmacopeias [14] . Incubation temperature was 55˚C to 60˚C for G. stearothermophilus, 30˚C to 35˚C for B. atrophaeus, B. subtilis, P. aeruginosa, E. coli, and S. aureus, and 25˚C for the mold bacteria: Aspergillus niger, and the yeast-like fungus: Candida albicans. After 7 days of incubation, result of the disinfection was assessed based on the turbidity of the culture media. The culture medium remains clear if 6 × log₁₀ grade sterilization was successful. Conversely, the culture media appears cloudy or muddy, if some part of the microorganisms survived exposure to the ozone. The experimental result is shown in Tables 1-5.

The final experiment involved the disinfection of dental adhesive resin contaminated with Aspergillus niger and Candida albicans. The selected material was Superbond C & B from SunMedical Co. in Japan. This compound is equivalent to C & B-Metabondfrom Porkell Inc., marketed in North America. Super bond C & B is a self- curing dental adhesive system containing granular poly (methyl methacrylate) (PMMA) resin and a liquid: 5%

Biological indicator: B. atrophaeus ATCC 9372 (2.0 × 10⁶ CFU).

^bBiological indicator: B. atrophaeus ATCC 9372 (8.48 × 10⁵ CFU).

Table 4. Sterilization in ozone mixture at 180 g/Nm³. (a) Sterilization of commercial biological indicators; (b) Effect on packaging material, humidity 100%; (c) Effect on packaging material, temperature, humidity 100%.

(a)

(b)

^cDamage of bonding observed.

(c)

^dDamage of non-woven sheet observed.

4-META/95% MMA-TBB. MMA is methyl methacrylate monomer, 4-META 4-methacryloxyethyl trimellitate anhydride, and TBB: tributyl-borane hardening catalyst. This material is applied in oral surgery as a coating and teeth adhesive (enamel and dentin), metallic parts and dental resin.

The experimental result is summarized as following tables. In the tables “−” minus indicates the successful disinfection and “+” plus indicates a denial of sterility. Table 1 show the results of spore-forming bacteria, Bacillus atrophaeus, on a filter paper carrier and wrapped in sterile packages. Glassine paper required 120 min exposure, while the Tyvek package required 60 min for disinfection, showing the present scheme to be compatible with Tyvek packages. This experiment was conducted at different temperatures of 30˚C, 50˚C and 70˚C. In the case of Glassine paper, sterilization time was shorter at higher temperatures. In Tyvek packages, no significant influence from temperature was observed. Table 2 shows a comparison of disinfection times for spore-forming bacteria: Geobacillus stearothermophilus, Bacillus atrophaeus, and Bacillus subtilis. Sterilization of these spore- forming bacteria required rather long period of 90 to 120 minutes. Table 3 shows the dependence of sterilization time on carrier material. Although Bacillus atrophaeus was sterilized on glass plate or filter paper carrier, the biological indicator on stainless steel carrier was positive. Probably, the iron site on the carrier acts as quencher of ozone.

In the next experiment, higher density was realized using a small reactor made of acrylic-Lucite with 40 mm internal diameter. Table 4 shows the result of disinfection experiment using high density ozone at 180 g/Nm³, i.e., 84,000 (ppm). In Table 4(a), wall temperature was raised to 50˚C and humidity 100%. Two biological indicators of 3 M Attest^TM 1262 and Attest^TM 1264, containing Bacillus atrophaeus 3.9 × 10⁶ CFU with approved EOG kill time 31.77 min and Geobacillus stearothermophilus 5.7 × 10⁵ CFU were disinfected in the original plastic sterile capsules within 30 minutes. Table 4(b) shows the result of experiment using biological indicators enclosed in Glassine and Tyvek sterile packages. Although the disinfection was completed in 30 minutes, glassine package was broken at the corner. In this case, glassine sheet is not appropriate for the sterility package for use as a sterile package for ozone/vapor disinfection. Table 4(c) shows the result of experiment at 65˚C. In this experiment, some deterioration of non-woven sheet was observed in 10 minutes, whereas such deterioration was not observed at 50˚C. Thus, 50˚C is the appropriate working temperature for the practical use of ozone disinfection system. Although higher disinfection efficiency can be realized at higher wall temperatures, at 65˚C, deterioration of sterile package may cause possible denial of sterility assurance.

Table 5 shows the result of disinfection for opportunistic pathogens, Aspergillus niger, Pseudomonas aeruginosa, Escherichia coli, Staphylococcus aureus and Candida albicans. This result indicates that these pathogenic species can be disinfected in sterile packages. Figure 2 shows control showing turbidity and clear samples indicating the disinfection, from left to right: control, exposed samples for 10, 20, and 30 minutes.

Figure 3 shows a VPS dental impression and acryl-resin for filler for rebuilding the tooth surf. Table 6 shows a result of the disinfection experiment of Candida albicans applied to smooth stainless steel and shaped Exafine tooth impression. Figure 4, for example, shows the typical result of incubation (from left to right): an untreated control, the second and the third containers show positive results, while the fourth and fifth containers are clear indicating negative result (−). Candida albicans samples applied to impression material located at the position in the reactor and locations indicated in Figure 5. Table 7 shows the results of this experiment. 15 minutes of exposure and short aeration time are sufficient for complete disinfection.

Biological indicator: Candida albicans ATCC 10231.

^aBiological indicator: Candida albicans ATCC 10231 (1.4 × 10⁶ CFU).

Next, we show the disinfection of Aspergillus niger in granular acrylic resin. A PMMA resin of 0.16 g was measured with a small-sized measuring spoon and placed in a small petridish, and a distilled water suspension of Aspergillus niger was applied. The final population was adjusted to be approximately to 3.4 × 10⁵ CFU. Similar experiment was repeated for Candida albicans (1.7 × 10⁶ CFU) sample. The top of acryl resin shows more like solid shell than granular resin. After the disinfection cycle, whole resin samples were incubated in an SCD fluid culture medium. The incubation result shows that disinfection was completed in 10 and 15 minutes, without significant deterioration in solidity when mixed PMMA resin was formed. The results are summarized in Table 8. FTIR spectra were measured for a sliced sample of Exafine.