AIHA Journal 64:533–537 (2003) Ms.

#559

Surface Germicidal Effects of

AUTHORS
Chih-Shan Li
Ozone for Microorganisms
Yu-Chun Wang
Graduate Institute of
Environmental Health, College of
Public Health, National Taiwan
University, Room 1449, No. 1,
Jen Ai Road, 1st Section, 100,
Taipei, Taiwan, R.O.C.; e-mail:
csli@ccms.ntu.edu.tw
In this study the influences of microorganism species, relative humidity, and ozone dosage on
ozone surface disinfection were evaluated. Bacterial and fungal cultures were spread on agar
plates and exposed to ozone. The selected microorganisms included Escherichia coli, Bacillus
subtilis, Candida famata, and Penicillium citrinum. Results showed that microorganism survival
fraction and ozone dosage (ozone concentration times exposure time) have an exponential
relationship. Results also indicated that E. coli was the most sensitive organism to ozone
exposure. E. coli required only very low ozone doses of 2–2.5 and 3.5–4 mg to obtain 50 and
80% inactivation, respectively. In addition, P. citrinum was more resistant than E. coli and
required ozone doses of 40–60 and 60–120 mg to obtain 50 and 80% inactivation. In addition,
spores of B. subtilis were observed to be the most resistant organism, requiring ozone doses
of 40–75 and 145–150 mg to obtain 50 and 80% inactivation. Yeast was less resistant than P.
citrinum and B. subtilis, requiring ozone doses of 10 and 15–19 mg to obtain 50 and 80%
inactivation. It was clearly indicated that the ozone dose differences for 80% microorganism
inactivation could be as high as 40 times between B. subtilis and E. coli. Ozone surface
germicidal efficiency increased as relative humidity increased, which could be related to more
radicals generated from ozone reaction with more water vapor at higher relative humidity. It
was concluded that ozone should be highly effective and provide a reliable safety factor in
treating contaminated surface. In addition, workers might need to wear suitable respiratory
protection at high ozone level operation.
Keywords: bacteria, fungi, germicidal effects, microorganism, ozone, surface
decontamination

This work was supported
by grant NSC 90-2320-
B-002-180 from National
Science Council, Taiwan,
Republic of China.
M
icroorganisms are almost ubiquitous
on the earth, but some places, such as
in hospital cleanrooms and food indus-
try packing processes, need aseptic en-
vironments. Moreover, excessive reproduction of
microorganisms is hazardous to human health.
Currently, numerous methods are used to miti-
gate the biological contamination, including
physical (heating, filtration, ultraviolet germicidal
irradiation, freezing) and chemical (ozone and
ethylene oxide) methods.
Ozone is well known to have antibacterial ac-
tivity, is less expensive to generate, and although
toxic, rapidly dissociates to oxygen. In general, re-
actions of ozone with various compounds occur
in two different and coexisting modes, one in-
volving direct reactions of molecular ozone and
the other being a free radical-mediated destruc-
tion mode.(1) Moreover, ozone might inactivate
microorganisms by causing damage to their ge-
netic materials. Currently, ozone usage for micro-
organism inactivation in solution is widely used.
It was demonstrated that survival efficiencies of
Escherichia coli were observed to 0.003 at 1.3
SHORT COMMUNICATIONS
ppm and 10 sec,(2) 0.00003 at 0.81 ppm and 30
min,(3) 0.00015 at 12 ppm and 62 sec,(4) and zero
at 2 ppm and 15 sec.(5) In addition, survival effi-
ciency of Bacillus subtilis was found to be 0.01 at
2.2 ppm and 90 sec.(6) The results show that
ozone had certain germicidal effects on microor-
ganisms. However, there was no consistent rela-
tionship found between microorganism survival
efficiency and ozone exposure dosages. Moreover,
it was indicated that ozone concentrations, pH
value, water temperature, residence time, mixing
degree, and organic compounds could also influ-
ence microorganism survival efficiency.
Regarding surface germicidal effects of ozone,

Copyright 2003, American Industrial Hygiene Association AIHA Journal (64) July/August 2003 533
 it was found that survival efficiencies of E. coli and Staphylococcus
aureus were less than 0.0001 at 300–1500 ppm with an exposure
time of 10–480 sec.(7) Moreover, it was found that hardy spores
have more resistance to ozone exposure. It was demonstrated that
microorganism survival efficiency was higher at higher relative hu-
midity (RH). In addition, it was also observed that microorganism
survival efficiency was lower when the microorganism-contaminated
surface could interact with ozone.(8) However, some data indicated
that there were no significant increases of ozone germicidal effects
for some microorganism species when ozone levels increased.(9) It
was concluded that ozone germicidal effects depended strongly on
microorganism characteristics, ozone concentration, exposure time,
and surface materials. Currently, the proposed ozone regulation
standard of the U.S. National Ambient Air Quality Standard(10) and
Taiwan Environmental Protection Agency is 120 ppb (1-hour av-
erage) and 100 ppb (8-hour average) for the U.S. Occupational
Safety and Health Administration.(11)
Until now, there has been no systematic evaluation regarding
microorganism survival efficiency in relation to ozone exposure.
In this investigation the influences of RH, microorganism species,
ozone concentration, and exposure time on ozone surface disin-
fection were evaluated in a laboratory chamber. The ozone dose
versus surface microorganism survival profiles for different micro-
organisms and RH were investigated.
and finally resuspended in PBS at a concentration of ca. 105/mL
for subsequent studies. Strains of P. citrinum were cultured on
malt extract agar (MEA, pH 5.6, Merck) and incubated at room
temperature (ca. 258C) for 7 days. Spores were then harvested into
sterile distilled water and centrifuged at 2000 rpm for 5 min re-
peatedly (three times) and finally resuspended in sterile distilled
water at a concentration of ca. 105 CFU/mL for subsequent
studies.
Experimental Apparatus
Solid Media on Agar Plates
The suitable diluted culture of 0.2 mL was spread on the surface
of dry TSA/MEA agar plates. In addition, the culture plates were
dried for 20 min in laminar flow. The microorganism concentra-
tions in each plate were below 100 colonies. For triplicate ozone-
exposed and ozone-unexposed sample, TSA and MEA agar plates
were incubated for 24 and 48 hours at 37 and 258C before count-
ing, respectively. The survival fraction was calculated from the
number of colonies forming on the ozone-exposed plates in com-
parison with those of the ozone-unexposed control plates. At least
triplicate tests were performed for each set. In the presented fig-
ures each error bar represents one standard deviation on the mean
of triplicate experiments. The test system was located in a chemical
hood so that the exhausted gas was vented outside.

RH Regulating Unit
MATERIALS AND METHODS A previous investigation(14) demonstrated that the survival fraction
of airborne microorganisms might be highly influenced by RH.
Tested Microorganisms Therefore, two RHs (55–60 and 85–90%) were chosen to repre-
sent medial and humid conditions. The experimental apparatus
Cultures used in this evaluation included E. coli from the Graduate
used in this study consisted of a compressed air system, RH con-
Institute of Microbiology, College of Medicine, National Taiwan
ditioner, and an ozone exposure chamber. The humidified gas
University, and B. subtilis (CCRC 12145). The recovery of gram-
stream was generated by passing pure compressed air through a
negative and non-spore-forming E. coli, representative of a sensi-
humidity saturator. The humidity/temp sensor (Hygromer-A1,
tive bacterial strain, could be compared with that of gram-positive,
Rotronic) was mounted inside the chamber to monitor air tem-
endospore-forming, and hardy B. subtilis.(12) An active E. coli cul-
perature and RH throughout the trials.
ture was inoculated into TSA (trypticase soy agar; Difco Labora-
tories, Detroit, Mich.) plates and incubated for 24 hours at Ozone Exposure Unit
378C.(12) The colonies were later aseptically washed using sterile
The ozone exposure chamber was approximately 23 L in volume
phosphate buffered saline (PBS) in a 15-mL sterile conical cen-
(ID 14 cm and height 38 cm). The ozone was generated from an
trifuge tube, which was capped and centrifuged at 2500 rpm
ozone generator (OZ1PCS-V/SW, Ozotech, Yreka, Calif.).
(model 2010, Kubota, Japan) for 5 min. Colonies were resus-
Ozone was monitored by an ozone analyzer (model 401, Ad-
pended in PBS at a concentration of ca. 105 CFU/mL for sub-
vanced Pollution Instruments, San Diego, Calif.) with a detection
sequent studies. For B. subtilis, the cells were initially inoculated
limit of 1.0 ppb. Preliminary results indicated that the following
on TSA plates for sporulation at 378C for 7 days. Bacterial growth
conditions should apply for observing varying effectiveness of sur-
was harvested into sterile distilled water, agitated at 45 rpm for
face microorganism germicidal effects of ozone. It was found that
more than 24 hours at room temperature, and heated for 10 min
exposure times were in the range of 30 to 150 min (30 min in-
at 808C to inactivate vegetative cells. The resulting spore suspen-
ternal) at ozone concentrations of 8, 12, and 16 ppm for B. subtilis
SHORT COMMUNICATIONS

sion was centrifuged at 2500 rpm for 5 min and finally resus-
and P. citrinum; 600 ppb, 900 ppb, and 1.2 ppm for E. coli; and
pended in sterile distilled water at a concentration of ca. 105
1.2, 1.8, and 2.4 ppm for yeast. Although the assessed ozone
CFU/mL for subsequent studies. Microscopic evaluation showed
concentrations were higher than the current ozone regulation
that the centrifuged B. subtilis suspensions contained only pure
standards of the U.S. National Ambient Air Quality Standard and
spores.
Taiwan Environmental Protection Agency (120 ppb, 1-hour av-
Two fungal strains frequently found in Taiwan were used as
erage), as well as that of the U.S. Occupational Safety and Health
the challenge cultures in this investigation. Spores of Penicillium
Administration (100 ppb, 8-hour average), previous investigations
citrinum Thom (CCRC 33168) and cells of Candida famata
had shown that ozone concentration could decline rapidly to 4%
(CCRC 22304) were used to represent mold and yeast, respec-
of the original levels in 100 cm distance because of its high oxi-
tively. To produce cultures, 1 mL of the stored yeast suspension,
dation ability.
in the concentration range of 105–106 cells/mL, was used as an
inoculum for 100 mL of YM broth (Difco Laboratories) in a 250
Microorganism Survival Fraction Versus Ozone Exposure
mL shake flask.(13) This suspension was incubated at 100 rpm in
a shaker for a period of up to 24 hours at room temperature. The The total dosage amount to which airborne microorganisms were
cells were then washed three times in a centrifuge at 2500 rpm exposed is the product of the ozone exposure on the microbe.

534 AIHA Journal (64) July/August 2003

 FIGURE 2. Survival fraction of yeast exposed to ozone on agar
FIGURE 1. Survival fraction of E. coli exposed to ozone on agar
surface
surface

The following equation describes the results of microorganism

survival fraction versus ozone dosage: in Figure 1) the survival fractions at 600 ppb and 120 min of
exposure, 900 ppb and 90 min of exposure, and 1200 ppb and
Ns 90 min of exposure were 0.29, 0.03, and 0.004, respectively, at
3 e2KD
No 55% RH. At 85% RH the survival fractions at 600 ppb and 120
where min of exposure, and 900 ppb and 60 min of exposure were 0.27
No 5 CFU per plate of microorganism unexposed to ozone and 0.07, respectively. Previously, it was reported that the inacti-
(CFU/plate) vation ozone concentration for E. coli was approximately 1
Ns 5 CFU per plate of microorganism survival after ozone ppm,(15) which agreed with the current results. For yeast (as shown
exposure (CFU/plate) in Figure 2) the survival fractions at 1.2 ppm and 150 min of
D 5 ozone dosage, mg 5 1.962 3 ozone conc. (ppm) 3 exposure, 1.8 ppm and 120 min of exposure, and 2.4 ppm and
Q3t 90 min of exposure were 0.33, 0.21, and 0.19, respectively, at
Q 5 ozone gaseous flow rate (m3/min) 55% RH. At 85% RH the survival fractions at 1.2 ppm and 150
min of exposure, and 1.8 ppm and 120 min of exposure were SHORT COMMUNICATIONS
t 5 time of ozone exposure (min)
K 5 microorganism susceptibility factor (1/mg) 0.07 and 0.09, respectively.
For B. subtilis (as shown in Figure 3), the survival fractions at
Statistical Calculation 8 ppm and 150 min of exposure, 12 ppm and 150 min of expo-
sure, and 16 ppm and 150 min of exposure were 0.42, 0.27, and
The comparisons of ozone dosage, two RHs, and four microor-
0.23, respectively, at 55% RH. At 85% RH the survival fractions
ganism species for survival fraction were performed using multiple
at 8 ppm and 150 min of exposure, and at 12 ppm and 150 min
regression analysis.
of exposure were 0.27 and 0.24, respectively. For P. citrinum (as
shown in Figure 4), the survival fractions at 8 ppm and 150 min
RESULTS AND DISCUSSION of exposure, 12 ppm and 150 min of exposure, and 16 ppm and
120 min of exposure were 0.41, 0.21, and 0.14, respectively, at
he relationships between microorganism survival fraction and 55% RH. At 85% RH the survival fractions at 8 ppm and 120 min
T ozone surface exposure time at different ozone concentrations
and RHs were evaluated in this investigation. For E. coli (as shown
of exposure, and at 12 ppm and 60 min of exposure were 0.01
and 0.15, respectively.

AIHA Journal (64) July/August 2003 535

 FIGURE 3. Survival fraction of B. subtilis spore exposed to
ozone on agar surface
Data revealed that for all four types of the evaluated microor-
ganisms, the survival fraction declined exponentially with ozone
dosage increase. It was demonstrated that amount of ozone sur-
face disinfection, being dependent on ozone dose (ozone exposure
duration times ozone concentration), was not individually influ-
enced by ozone level and exposure time. Using the simple expo-
nential regression analyses, the K values (microorganism suscep-
tibility factor, 1/mg), commonly used as the indicator of
sensitivity of the test microorganism, were demonstrated to vary
across a wide range depending on microorganism species. Signif-
icantly, the microorganism susceptibilities of E. coli were the high-
est (0.32–0.43), and those of B. subtilis were the lowest (0.0078–
SHORT COMMUNICATIONS
0.0109). In comparison with some previous data,(16–19) it was also
clearly indicated that the microorganism susceptibility factor de-
pends strongly on microorganism species, and varies in a wide
range. However, there were no exponential relationships found
between ozone dosage and survival efficiency in previous investi-
gations.(7,9,20) These findings might be related to microorganism
masking effects.
The current results indicated that E. coli was the most sensitive
organism to ozone exposure in this study. E. coli required only
very low ozone doses of 2–2.5 and 3.5–4 mg to obtain 50 and
80% inactivation, respectively. In addition, P. citrinum was more
resistant than E. coli and required ozone doses of 40–60 and 60–
120 mg to obtain 50 and 80% inactivation, respectively. Spores of
B. subtilis were observed to be the most resistant organism and
required ozone doses of 45–70 and 145–150 mg to obtain 50 and
536 AIHA Journal (64) July/August 2003
FIGURE 4. Survival fraction of P. citrinum exposed to ozone on
agar surface
80% inactivation, respectively. Also, yeast was less resistant than P.
citrinum and B. subtilis and required ozone doses of 10 mg and
15–19 mg to obtain 50 and 80% inactivation, respectively. In ad-
dition, it was clearly indicated that the ozone dose differences for
80% microorganism inactivation could be as high as 40 times be-
tween B. subtilis and E. coli. These microorganism susceptibility
order results well agreed with those found in some previous
studies.(8,9,21)
In regard to RH effects on ozone surface-disinfection, it was
observed that survival fractions at 85% RH were lower than those
found at 55% RH for E. coli, B. subtilis, and P. citrinum, but not
for Candida famata. It was clearly demonstrated that microor-
ganism inactivation rates from ozone exposure were higher at
higher RH. This might be related to generation of more radicals
from ozone reacted with more water vapor at higher RH. A pre-
vious report(8) also indicated that there was significant survival dif-
ference for Rhodotorula glutinis at different RH conditions.
In this investigation, the evaluated ozone concentrations were
higher than ozone regulation standards (120 ppb, 1-hour average)
of the U.S. National Ambient Air Quality Standard and the Tai-
wan Environmental Protection Agency, as well as (100 ppb, 8-
hour average) those of the U.S. Occupational Safety and Health
Administration. As seen in a previous investigation, ozone con-
centration can decrease rapidly to 4% of the original concentration
in 100 cm distance because of its high oxidation ability. On the
other hand, it might be necessary to disinfect the surface in the
chamber, rather than in open space, to prevent high-level ozone
exposure for surface microorganism decontamination by workers.
Moreover, it was found that the microorganism survival fraction
declined exponentially with ozone dosage increase and was not
individually influenced by ozone level and exposure time. There-
fore, the authors suggest that surface microorganism disinfection
should be performed for longer exposure times and at lower ozone
levels. In summary, ozone disinfection should be a promising tech-
nique for surface microorganism contamination.
CONCLUSION
he influences of microorganism species, RH, and ozone dosage
T on ozone surface disinfection were evaluated. It was demon-
strated that the microorganism survival fraction and ozone dosage
have an exponential relationship. The results indicated that micro-
organism susceptibility to ozone exposure is in the order of E. coli,
yeast, P. citrinum, and B. subtilis. Ozone dosage differences for
80% microorganism inactivation were as high as 40 times between
B. subtilis and E. coli. Ozone surface microorganism germicidal
efficiencies increased as RH increased, which could be related to
the higher number of radicals from ozone reaction with more wa-
ter vapor at higher RH. It was concluded that ozone should be
highly effective and provide a reliable safety factor in treating mi-
croorganism contaminated surfaces.
REFERENCES
1. Kim, J.-G., A.E. Yousef, and S. Dave: Application of ozone for
enhancing the microbiological safety and quality of foods: A review.
J. Food Prot. 62:1071–1087 (1999).
2. Katzenelson, K., and H.I. Shuval: Studies on the disinfection of
water by ozone: Viruses and bacteria. In R.G. Rice and M.E. Brown-
ing, editors, First International Symposium on Ozone for Water &
Wastewater Treatment, vol. 1. Washington, D.C.: Hampson Press,
1973.
3. Finch, G.R., D. Smith, and M.E. Stiles: Dose-response of Esche-
richia coli in ozone demand-free phosphate buffer. Wat. Res. Tech. 22:
1563–1570 (1988).
4. Bunning, G., and D.C. Hempel: Vital-fluorochromization of micro-
organisms using 39,69-diacetylfluorescein to determine damages of cell
membranes and loss of metabolic activity by ozonation. Ozone Sci.
Eng. 18:173–181 (1996).
5. Burleson, G.R, T.M. Murray, and M. Pollard: Inactivation of vi-
ruses and bacteria by ozone, with and without sonication. Appl.
Microbiol. 29:340–344 (1975).
6. Botzenhart, K., G. Tarcson, and M. Ostruschka: Inactivation of
bacteria and coliphages by ozone and chlorine dioxide in a continuous
flow reactor. Wat. Sci. Tech. 27:363–370 (1993).
7. Kowalski, W.J., W.P. Bahnfleth, and T.S. Whittam: Bacterial effects
of high airborne ozone concentrations on Escherichia coli and Staph-
ylococcus aureus. Ozone Sci. Eng. 20:205–221 (1998).
8. Foarde, K.K., D.W. VanOsdell, and R.S. Steiber: Investigation of
gas-phase ozone as a potential biocide. Appl. Occup. Environ. Hyg. 12:
535–542 (1997).
9. Moore, G., C. Griffith, and A. Peters: Bactericidal properties of
ozone and its potential application as a terminal disinfectant. J. Food
Prot. 63:1100–1106 (2000).
10. U.S. Environmental Protection Agency (EPA): National Air Qual-
ity and Emissions Trend Report. Washington, D.C.: EPA, 1999.
11. ‘‘List of Highly Hazardous Chemicals, Toxics and Reactives (Man-
datory).’’ Code of Federal Regulations Title 29, Part 1926.64, Appen-
dix A. March 4, 1992.
12. Jensen, P.J., W.F. Todd, G.N. David, and P.V. Scarpino: Evalua-
tion of eight bioaerosol samplers challenged with aerosols of free bac-
teria. Am. Ind. Hyg. Assoc. J. 53:660–667 (1992).
13. Griffiths, W.D., I.W. Stewart, A.R. Reading, and S.J. Futter: Ef-
fect of aerosolisation, growth phase and residence time in spray and
collection fluids on the culturability of cells and spores. J. Aerosol Sci.
27:803–820 (1996).
14. Lin, C.Y., and C.S. Li: Control effectiveness of ultraviolet germicidal
irradiation on bioaerosols. Aerosol Sci. Tech. 36:474–478 (2002).
15. Hunt, N.K., and B.J. Marinas: Inactivation of Escherichia coli with
ozone chemical and inactivation kinetics. Water Res. 33:2633–2641
(1999).
16. Kundsin, R.B.: Characterization of Mycoplasma aerosols as to viabil-
ity, particle size, and lethality of ultraviolet radiation. J. Bacteriol. 91:
942–944 (1966).
17. David, H.L.: Response of mycobacteria to ultraviolet radiation. Am.
Rev. Resp. Dis. 108:1175–1184 (1973).
18. Keller, L.C., T.L. Thompson, et al.: UV light-induced survival re-
sponse in a highly radiation-resistant isolate of the Moraxella acine-
tobacter group. Appl. Environ. Microbiol. 43:424–429 (1982).
19. Mongold, J.: DNA repair and the evolution of transformation in
Haemophilus influenzae. Genetics 132:893–898 (1992).
20. Berrington, A.W., and S.J. Pedler: Investigation of gaseous ozone
for MRSA decontamination of hospital side-rooms. J. Hosp. Infect. 40:
61–65 (1998).
21. Masaoka, T., T. Kubota, S. Namiuchi, et al.: Ozone decontami-
nation of bioclean rooms. Appl. Environ. Microbiol. 43:509–513
(1982).
SHORT COMMUNICATIONS
AIHA Journal (64) July/August 2003 537