Ozone Antimicrobial Efficacy
EPA/600/R-08-137
December 2007
Ozone Antimicrobial Efficacy
By
Karin Foarde and Gary Eaton, Ph.D.
RTI International
3040 Cornwallis Road
P.O. Box 12194
Research Triangle Park, NC 27709-2194
GSA Contract No. GS-10F-0283K Order EP05C000057
RTI Project No. 09432.001
Project Officer
Marc Menetrez, Ph.D.
U.S. Environmental Protection Agency
National Risk Management Research Laboratory
Research Triangle Park, NC 27711
-------
Ozone Antimicrobial Efficacy
Notice
The U.S. Environmental Protection Agency (EPA) through its Office of Research and Develop-
ment (ORD) managed the research described in this report. The research was performed by RTI.
The report has been subjected to the Agency's peer and administrative review and has been
approved for publication as an EPA document. Any opinions expressed in this report are those of
the author and do not, necessarily, reflect the official positions and policies of the EPA. Any
mention of products or trade names does not constitute recommendation for use by the EPA.
Foreword
The U.S. Environmental Protection Agency (EPA) is charged by Congress with protecting the
Nation's land, air, and water resources. Under a mandate of national environmental laws, the
Agency strives to formulate and implement actions leading to a compatible balance between
human activities and the ability of natural systems to support and nurture life. To meet this
mandate, EPA's research program is providing data and technical support for solving environ-
mental problems today and building a science knowledge base necessary to manage our
ecological resources wisely, understand how pollutants affect our health, and prevent or reduce
environmental risks in the future.
The National Risk Management Research Laboratory (NRMRL) is the Agency's center for
investigation of technological and management approaches for preventing and reducing risks
from pollution that threaten human health and the environment. The focus of the Laboratory's
research program is on methods and their cost-effectiveness for prevention and control of
pollution to air, land, water, and subsurface resources; protection of water quality in public water
systems; remediation of contaminated sites, sediments and ground water; prevention and control
of indoor air pollution; and restoration of ecosystems. NRMRL collaborates with both public and
private sector partners to foster technologies that reduce the cost of compliance and to anticipate
emerging problems. NRMRL's research provides solutions to environmental problems by:
developing and promoting technologies that protect and improve the environment; advancing
scientific and engineering information to support regulatory and policy decisions; and providing
the technical support and information transfer to ensure implementation of environmental
regulations and strategies at the national, state, and community levels.
This publication has been produced as part of the Laboratory's strategic long-term research plan.
It is published and made available by EPA's Office of Research and Development to assist the
user community and to link researchers with their clients.
Sally C. Gutierrez, Director
National Risk Management Research Laboratory
in
-------
Ozone Antimicrobial Efficacy
Abstract
Ozone is a potent germicide that has been used extensively for water purification. The germicidal activity
of ozone in water has been reported by many authors (see for example U.S. EPA, 1999); however, there is
limited information on the biocidal activity of ozonated air as a treatment for contaminated surfaces.
Understanding of the biocidal capability of ozone against the microorganisms primarily responsible for
indoor air quality biocontaminant problems is still relatively limited.
In a previous research project, we evaluated the biocidal efficacy of three to 10 ppm ozone on selected
microorganisms in laboratory chamber studies under controlled conditions (Foarde et al., 1997). The
organism challenge consisted of vegetative organisms or spores dried on surfaces. The study was
conducted in two phases. First, an extensive series of tests employing glass slides as the test surface were
performed under ideal conditions of ozone exposure in which intensive efforts were made to minimize (or
eliminate) ozone losses in the chambers. Second, a short series of tests was performed using building
materials as the test surfaces. We found that ozone concentrations of 6 to 10 ppm were required to obtain
3-log reductions in colony-forming units (CPUs). The results from the second phase of the study, where
spores of Penicillium spp. were deposited on actual building material surfaces, showed no reduction in
CPUs after a 23-hr exposure to 9 ppm of ozone. For the denser materials (ceiling tiles), test levels of
ozone were not attained, probably due to the reaction with the substrate.
The objective of this project was to expand on work from the earlier study by testing the effect of ozone
at much higher levels (up to 1000 ppm for 24 hours) on a variety of microorganisms. The goal of these
experiments was to ascertain the biocidal efficacy of ozone against four organisms - two bacteria (one
spore and one vegetative organism) and two fungi (one spore and one vegetative organism). A series of
experiments was performed using either glass slides or gypsum wallboard as the test surface. This series
of experiments confirmed the results of the earlier experiments that the organisms on glass slides were
more readily killed than organisms on building materials for higher levels of ozone. It would be
reasonable to assume that the difference in ozone efficacy between the two test surfaces was due at least
in part to the ability of the gypsum wallboard to inactivate the ozone and thus to protect the spores
deposited on its surface. Also as in the earlier experiments, increasing RH increases the biocidal
capability of ozone.
Because adverse health effects differ by organism and susceptibility of the exposure population, no
standard acceptable level of contamination exists, nor does any required level of efficacy for decontami-
nating building materials in the field, therefore, a key issue in evaluating the efficacy of any biocide,
including ozone, is to determine the acceptable number of CPUs remaining after treatment. For example,
in these experiments the inoculum was usually at least 1 x 106 CPU. A 1 log reduction (90% inactivation
efficiency) would mean that 1 x 105 CPU remained after exposure. If a 4 log reduction (99.99%
inactivation efficiency) was attained, there would be 100 CPUs remaining after exposure. The acceptabil-
ity of either of these inactivation efficiencies would depend on the specific situation.
Although the specific results vary depending upon the test organism and the test surface, the overall
results of this study indicate that, even at relatively high concentrations of ozone, it is difficult to achieve
significant inactivation of organisms on material surfaces. The high ozone concentrations used in this
study would probably be difficult to maintain near or at the surface of some commonly contaminated
building materials, and even if these concentrations could be maintained in the field, it would be
challenging to achieve a significant reduction of surface biocontamination using ozone.
IV
-------
Ozone Antimicrobial Efficacy
Table of Contents
Section Page
Notice iii
Foreword iii
Abstract iv
List of Figures v
List of Table v
1. Introduction 1
2. Methods and Materials 1
2.1 Ozone Exposure Apparatus 1
2.1.1 Exposure Chamber 1
2.1.2 Air Supply and Humidifier 3
2.1.3 Ozone Source 3
2.2 Test Organisms 4
2.3 Microorganism Challenge 5
2.4 Experimental Procedures 5
2.41 Quantitative Evaluation 6
3. Results 6
3.1 Comparison of Number of CPUs from Exposed and Unexposed Test Pieces 6
3.2 Log Change in CPU's forthe Total (C*T) Ozone Exposure 9
4. Discussion and Conclusions 12
5. References 12
List of Figures
Figure Page
Figure 1. Schematic of the test apparatus 2
Figure 2. Log change in CPU of P. brevicompactum over a range of ozone exposure 10
Figure 3. Log change in CPU of R. mucilaginosa over a range of ozone exposure 10
Figure 4. Log change in CPU of B. atrophaeus over a range of ozone exposure 11
Figure 5. Log change in CPU of S. epidermidis over a range of ozone exposure 11
List of Tables
Table Page
Table 1. Summary of Test Conditions for Each Test Organism and Each Test Surface Type 5
Table 2. CPUs for Control and Exposed Samples on Glass Slides at Low Humidity 6
Table 3. CPUs for Control and Exposed Samples on Gypsum Wallboard at Low Humidity 7
Table 4. CPUs for Control and Exposed Samples on Glass Slides at High Humidity 8
Table 5. CPUs for Control and Exposed Samples on Gypsum Wallboard at High Humidity 9
-------
Ozone Antimicrobial Efficacy
1. Introduction
Ozone is a potent germicide that has been used extensively for water purification (Weavers and
Wickramanayake, 2001). In Europe, 90 percent of the municipal water systems are treated with ozone,
and in France, ozone has been used to treat drinking water since 1903. The germicidal activity of ozone in
water has been reported by many authors (see for example U.S. EPA, 1999); however, there is limited
information on the biocidal activity of ozonated air as a treatment for contaminated surfaces. Ozone has
long been used as an effective deodorant in the remediation of smoke-damaged buildings. As an
important reactive species in the atmosphere, the chemistry of ozone and volatile organic compounds has
also been widely studied (Atkinson and Carter, 1984.)
Understanding of the biocidal capability of ozone against the microorganisms primarily responsible for
indoor air quality biocontaminant problems is still relatively limited. In a previous research project, we
evaluated the biocidal efficacy of three to 10 ppm ozone on selected microorganisms in laboratory
chamber studies under controlled conditions (Foarde et al, 1997). The organism challenge consisted of
vegetative organisms or spores dried on surfaces. The study was conducted in two phases. First, an
extensive series of tests employing glass slides as the test surface were performed under ideal conditions
of ozone exposure in which intensive efforts were made to minimize (or eliminate) ozone losses in the
chambers. Second, a short series of tests was performed using building materials as the test surfaces. We
found that ozone concentrations of 6 to 10 ppm were required to obtain 3-log reductions in colony-
forming units (CPUs). The results from the second phase of the study, where spores ofPenicillium spp.
were deposited on actual building material surfaces, showed no reduction in CPUs after a 23-hr exposure
to 9 ppm of ozone. For the denser materials (ceiling tiles), test levels of ozone were not attained, probably
due to the reaction with the substrate.
The objective of the project discussed here was to expand the early work by testing at much higher ozone
levels. The goal of these experiments was to ascertain the effect of ozone against two bacteria (a
vegetative bacterium and a bacterial spore) and two fungi (a vegetative fungus and a fungal spore) dried
on the surfaces of glass microscope slides and pieces of gypsum wallboard. The target ozone levels were
100, 500, and 1000 ppm. Three exposure times, 1.5 hr, 6 hr, and 24 hr, were used.
In addition, tests were performed at two levels of relative humidity (RH). Many factors, including
temperature, pH, RH, and organic load, affect the susceptibility of microorganisms to ozone (Foegeding
and Busta, 1991). Increasing RH increases the biocidal capability of ozone (Clark and Takacs, 1980). In
addition, materials in buildings are exposed to a range of RHs. Two ranges of RHs were selected for use
in this study: low RH (20-45%) and high RH (80-95%). These two levels bracket the use conditions for
ozone in most buildings and provide information at two extreme air moisture situations.
2. Methods and Materials
2.1 Ozone Exposure Apparatus
2.1.1 Exposure Chamber
A commercially available desiccator cabinet constructed of polished stainless steel and glass was used to
carry out the microorganism exposure experiments. A schematic of the ozone apparatus and setup is
shown in Figure 1. In earlier experiments (Foarde et al., 1997), the all-glass chamber interior was coated
with Teflon: the coating was important to limit ozone reactivity with surfaces at the low ozone concentra-
tions (1 to 10 ppm) being used. The experiments described here used much higher ozone concentrations
(100 to 1000 ppm), and measurements to assess the stability of these higher concentrations within the
chamber demonstrated that there was no need to line the chamber with Teflon for this study because the
1
-------
Ozone Antimicrobial Efficacy
higher levels of ozone, flowing constantly thorough the chamber at various fixed rates from 15 to 27 liters
per minute, could be maintained throughout the chamber interior at all chosen exposure times. The
chamber has interior dimensions of approximately 30 x 25 x 30 (cm, L x W x H) and a total volume of
approximately 22 liters. For use as an ozone exposure chamber, the chamber interior was accessed
through an airtight door lined with a neoprene gasket. Humidified air containing ozone flowed through a
plastic bulkhead fitting mounted in the chamber's rear wall, 3 cm from the bottom. A sample for the
ozone analyzer was withdrawn continuously at a flow rate of 1.0 liter per minute through Teflon tubing
mounted in a bulkhead fitting in the center of the rear wall; the remainder of the flowing ozone-containing
air exited to the laboratory hood through a fitting and tubing attached 3 cm from the top of the chamber's
rear wall. Uniformity of ozone concentrations at various points within the chamber showed the air was
homogeneously mixed. Inoculated microscope slides and glass Petri dishes containing inoculated
wallboard samples were placed on two to four perforated stainless steel shelves within the chamber in
such a way that ozone-bearing air could circulate freely around the samples.
Ozone+
dry air
Dry Air in at 8.6 Lpm
via rotameter—>
Heat tape -
Ozone +
humidified air
im anger
o
JH20
V\fet or dry dilutior
air in at 6 to 18
Lpm
valve
Dry air in via rotameter
Heat tape -
linger
Exposure
chamber
\Test materials on
perforated shelves
Vent to hood
Dry air in via
rotameter
O
C-H'0
Ozone analyzer
Vent to hood
Figure 1. Schematic of the test apparatus.
During set up of the ozone generator and exposure chamber, a TECO Model 49 ambient air ozone
analyzer was used to characterize the system. Ozone-laden air, nominally 500 ppm, flowing at several
liters per minute through the exposure chamber, was diluted 1:1000 with clean air to allow detection by
the TECO 49 analyzer. The ozone-containing air entered through a port at the bottom of the rear wall of
the exposure chamber. Ozone/air vented from the exposure chamber from a port at the center of the
chamber's back wall (and then diluted 1:1000) registered 447 ppb on the TECO 49. Another sample taken
from a port near the top of the chamber also registered 447 ppb. Therefore, we concluded that the mixing
-------
Ozone Antimicrobial Efficacy
and flow conditions within the exposure chamber were such that the ozone concentration was spatially
uniform. In all exposure experiments, ozone entered the chamber at the bottom of the back wall and
exited at the top (i.e., through the vent). The ozone-laden air had to pass around and through the
perforated stainless steel shelves on which the test samples were placed.
2.1.2 Air Supply and Humidifier
The test air for the experiments was ozonized and humidified using the apparatus shown in Figure 1. Dry,
particle-free, compressed house "zero" air was used. The clean air stream was split into two streams using
rotameters and needle valves. One air stream passed through the ozone generator and thence to an all-
glass impinger containing deionized water. The concentration of the ozone and the RH were controlled by
the introduction of dilution air that was also humidified by passage through a separate impinger. The two
airstreams combined at a "tee" and entered the chamber. The exteriors of the glass impingers were
wrapped in heating tape so that the temperature of the water remained constant and thus the amount of
water vapor delivered to the chamber remained nearly constant. The tubing through which sampled air
was pulled to the temperature and RH sensor was placed in the chamber exhaust port before and after
ozone production. Room temperature, controlled by the building's heating, ventilation, and air condition-
ing (HVAC) system was monitored at a point adjacent to the chamber. The air flowed through the
chamber as described above.
Temperature and humidity were measured by a factory-calibrated EdgeTech Model 2000 Series
DewPrime dew point hygrometer. To protect the humidity sensor from high ozone concentrations,
measurements of exposure chamber temperature (T) and RH were made on air withdrawn from the
chamber prior to introduction of ozone and immediately after ozone exposure ceased; all conditions (air
flow, humidification, temperature) were identical except for the production of ozone by the generator.
2.1.3 Ozone Source
A Model GTC-0.5 ozone generator (Ozonia North America, Griffin Division) was used to generate ozone
by corona discharge. The generator's ozone output was controlled by varying the amperage; ozone
concentration was further controlled by varying the dilution air flow.
The ozone concentration in the chamber was measured using a Teledyne Instruments/Advanced Pollution
Instrumentation Ozone Monitor, Model 460M. The factory-calibrated instrument operated on a 0-1000
ppm range. The accuracy and precision of the ozone monitor were estimated to be +/- 10 ppm. The
voltage signal output from the ozone monitor was recorded by a laboratory data acquisition system that
averaged the signal over 10-minute periods and expressed the results as ppmV. The accuracy of the
Teledyne ozone monitor's response was confirmed indirectly as follows:
• First, the response of a TECO Model 49 ambient ozone monitor, operating on the 0 tol ppm (0 to
1000 ppb) range was shown to be accurate and linear by challenging it with ozone generated and
verified by a TECO 49C ozone primary standard calibrator: we challenged the TECO 49 using
the primary standard at seven concentrations ranging from 0 to 879 ppb, and the TECO 49 re-
sponses agreed well with the primary standard photometer designated values (within 1.56 percent
or better at all seven concentrations). The response of the Model 49 (€49, in ppb) was related to
the response of the Model 49C (C49c, in ppb) as follows: C49 = 1.0044* C49c+1.408 (r2 = 1.0000).
• Next, high concentrations of ozone in air were diluted so that they could be accurately sensed by
the TECO Model 49 ambient monitor and simultaneously compared to the values reported by the
Teledyne Model 460M high-concentration ozone monitor. To accomplish this, the exposure
chamber was set to 50% RH, and nominal 800 ppm ozone was produced for an exposure time of
180 minutes. Then, the TECO 49 and Teledyne 460M monitors sampled from the same location
within the chamber.
-------
Ozone Antimicrobial Efficacy
The TECO 49 responses (after accounting for 1:1000 dilution) averaged 725.1 ± 41.5 ppm [5.7 percent
relative standard deviation (RSD)], based on 14 readings taken over the 180 minute period. The Teledyne
460M responses averaged 789.8 ±3.5 ppm (0.44 percent RSD), based on 18 readings taken over the same
time period. The agreement of the Teledyne and the TECO average concentrations was within ~ 8.8
percent, which met the data quality indicator goal of 10 percent accuracy of ozone concentration
determination. The precision of the ozone measurements for various ozone exposure levels and time
frames were most often within ±10 ppm. The higher standard deviation for the TECO 49 responses was
likely due to the dilution process and small variations in flow-regulating components. The excellent
standard deviation shows the temporal stability of ozone in the exposure system as well as the advantages
of an ozone monitor that does not require sample dilution. In conclusion, the factory calibration of the
Teledyne Model 460M was sufficiently accurate. Ozone concentrations for all exposures shown in the
final report were monitored by the Teledyne Model 460M.
Following verification of the accuracy of the Teledyne monitor's response, the chamber was characterized
as follows:
1. Set up the chamber for use, including test gas entry, exit, and sampling ports, placement of sup-
port shelves within the chamber, and placement of test surfaces (glass microscope slides and glass
Petri dishes holding squares of gypsum wallboard).
2. Establish a steady-state ozone concentration at the low end of the test range- i.e., 100 ppm.
3. Measure ozone concentration in the center of the chamber to ensure there is no variability at a
single location.
4. Repeat the measurement at various locations within the chamber to ensure the test gas was well-
mixed as indicated by steady ozone readings.
5. Fine tune the air flow rates, the water-containing impinger conditions, and ozone generator output
to achieve the desired ozone concentration and RH for each experiment.
2.2 Test Organisms
The four organisms selected for testing in this study were Rhodotorula mucilaginosa, Penicillium
brevicompactum, Bacillus atrophaeus, and Staphylococcus epidermidis. B. atrophaeus and S. epidermidis
are bacteria, and R. mucilaginosa and P. brevicompactum are fungi (a yeast and a mold, respectively).
Yeasts are single-celled organisms that reproduce by budding and do not form spores. Molds, however,
are composed of long branching filaments called hyphae (a mass of hyphae is referred to as a mycelium).
Mycelia or hyphae are the vegetative phase of the organism. Molds reproduce by forming spores that are
resistant to unfavorable environmental conditions and can remain dormant for long periods of time. Only
the spores of P. brevicompactum were used in this study.
B. atrophaeus is a gram-positive, spore-forming bacterium. In this study, only the spore of B. atrophaeus
was used. The vegetative bacterium was S. epidermidis. Because vegetative organisms are generally more
susceptible to the effect of biocides than spore-forming organisms, and bacterial spores usually more
resistant than fungal spores, S. epidermidis was expected to be the most susceptible and the B. atrophaeus
spore, the least susceptible.
Three of the test organisms were purchased from the American Type Culture Collection (ATCC): P.
brevicompactum (ATCC #9056), B. atrophaeus (ATCC #9372), and S. epidermidis (ATCC #12228). R.
mucilaginosa, a field isolate from a contaminated building site being maintained in the RTI culture
collection (RTI CC #3435), was identified using the Biolog MicroLog Microbial Identification System
(Biolog, Inc., Hayward, CA.). The Biolog system is a commercially available product for broad-based
rapid identification and characterization. This system uses redox chemistry based on the reduction of
tetrazolium, which responds to the metabolism of specific substrates by the test organism.
-------
Ozone Antimicrobial Efficacy
2.3 Microorganism Challenge
The form of the challenge was a critical element in the design of the study. Typically, biocides are
evaluated using the most difficult test conditions under which they may be expected to function. For this
study, the test organisms were dried on the surface of gypsum wallboard pieces or on glass slides, because
destruction of organisms dried on surfaces is a difficult challenge for a biocide and because building
surfaces such as gypsum wallboard are frequent targets for ozone treatment during remediation. The
number of challenge organisms was based on levels of contamination that have been reported on surfaces
in buildings. Levels ranging from 103 to 107 CFU/cm2 have been isolated from surfaces of contaminated
buildings ranging from ceiling tile to wallboard (Morey, 1993).
2.4 Experimental Procedures
These experiments used two types of test surface: glass microscope slides and gypsum wallboard pieces.
Each test included seven individual test surfaces (glass slides or pieces of gypsum wallboard). For each
test, one of the test organisms was suspended in water and measured volumes were pipetted onto each of
the seven test surfaces and allowed to dry. The seven test surfaces were dried, and four of the surfaces
exposed in the test chamber as described above. The remaining three surfaces, kept on the bench adjacent
to the test chamber, served as unexposed controls (which provided both a baseline and a measure of
inoculum viability). Following exposure, the exposed test surfaces and the control test surfaces were
processed, and the effects of the ozone evaluated by comparing the CPUs of the exposed and unexposed
surfaces. The test conditions are given in Table 1.
Table 1. Summary of Test Conditions for Each Test Organism and Each Test Surface Type
Relative
Humidity
Low
(20-45%)
High
(80-95%)
Ozone
Concentration
1000 ppm
500 ppm
100 ppm
0 ppm
1000 ppm
500 ppm
100 ppm
0 ppm
Exposure Times
1.5hr
1.5hr
1.5hr
1.5hr
1.5hr
1.5hr
1.5hr
1.5hr
6hr
6hr
6hr
6hr
6hr
6hr
6hr
6hr
24 hr
24 hr
24 hr
24 hr
24 hr
24 hr
24 hr
24 hr
The test matrix included the following elements:
• Two test surface types, glass microscope slides and gypsum wallboard
• Seven test surfaces for each run (3 controls and 4 exposed)
• 103 - 107 CPUs/test surface
• Four different microorganisms(fungal spore, bacterial spore, vegetative bacteria, and yeast)
2.4.1 Quantitative Evaluation
Each of the seven test surfaces was placed in a separate container, suspended in sterile phosphate buffered
saline (PBS) containing Tween 80, and shaken for at least five minutes. All necessary dilutions were
made using the same buffer. Aliquots of the suspension were plated on the appropriate media and
incubated for the optimal time and temperature for the test organism. CPUs were counted and calculated
for each test surface piece.
-------
Ozone Antimicrobial Efficacy
3. Results
3.1 Comparison of Number of CPUs from Exposed and Unexposed Test Pieces
The effects of the ozone on the test organisms were evaluated by comparing the number of CPUs
recovered from the exposed slides or wallboard to those recovered from the unexposed test surfaces.
CPUs were counted, numbers per glass slide or wallboard piece computed, and the data transformed to
their logarithmic (base 10) value. The averaged results of the three control or four exposed samples,
including standard deviations, are shown in Tables 2 through 5 by organism. Table 2 shows the results
for glass slides at low RH. Table 3 shows the results for gypsum wallboard at low RH. Table 4 shows the
results for glass slides at high RH, and Table 5 shows the results for gypsum wallboard at high RH. All of
the tables have the same structure: The first column shows the ozone concentration (ppm) in the air
flowing through the test chamber; the second column shows the time that the glass slides or gypsum
wallboard pieces were exposed in the chamber; the third column shows the calculated total ozone
exposure (the ozone concentration [C, in ppm] was multiplied by the time [T, in minutes] to determine the
exposure [C*T, in ppm-minutes]); and the last two columns show the mean and standard deviation of the
number of CPU logio test organisms on the control or exposed test pieces.
The minimum detection limit for the test organism was 150 CPU per sample based on the analysis method
described in Section 2. Therefore, BDL (below detection limit) is used in the tables when fewer than 150
CPUs were isolated from the exposed samples. (Note that the tables present the logio of CPU; the logio of
the 150 CPU detection limit is 2.18.)
Table 2. CPUs for Control and Exposed Samples on Glass Slides at Low Humidity
O3 concentra-
tion (ppm)
Exposure
time (min)
Total
Exposure
(ppm-min)
Control Samples
(CPU Iog10)
Mean ± St. Dev.
Exposed Samples
(CPU Iog10)
Mean ± St. Dev.
P. brevicompactum
119.4 ±22.1
504.6 ±14. 5
516±9
501.8 ±13.6
1007 ±20.7
1010 ±13.4
90
90
360
1440
90
360
10,746
45,414
185,760
722,592
90,630
363,600
5.92 ±0.04
6.61 ±0.05
6. 79 ±0.21
6.34 ±0.08
5.19 ±0.46
5. 80 ±0.17
5.95 ±0.10
6. 10 ±0.31
2.46 ± 0.57
BDL*
4.88 ±0.33
BDL
R. mucilaginosa
11 9.4 ±22.1
504.6 ±14.5
516±9
501. 8 ±13.6
1006 ±20.7
1010 ±13.4
90
90
360
1440
90
360
10,746
45,414
185,760
722,592
90,540
363,600
4.48 ±0.30
7.20 ±0.08
7.46 ±0.03
7.08 ±0.08
7.70 ±0.05
7.70 ±0.05
5. 07 ±0.36
7.10 ±0.04
7.53 ±0.04
6. 77 ±0.10
7.45 ±0.09
6.88 ±0.21
B. atrophaeus
119.4 ±22.1
504.6 ±14. 5
516±9
501.8 ±13.6
1007 ±20.7
1010 ±13.4
90
90
360
1440
90
360
10,746
45,414
185,760
722,592
90,630
363,600
6.47 ±0.25
6.36 ±0.07
6.62 ±0.11
6.40 ±0.06
5.85 ±0.29
6.60 ±0.12
6.71 ±0.36
6.44 ±0.43
6.58 ±0.18
6.06 ±0.34
5. 92 ±0.19
7. 02 ±0.12
-------
Ozone Antimicrobial Efficacy
O3 concentra-
tion (ppm)
1014±17.8
Exposure
time (min)
1440
Total
Exposure
(ppm-min)
1,460,160
Control Samples
(CPU Iog10)
Mean ± St. Dev.
6.43 ±0.17
Exposed Samples
(CPU Iog10)
Mean ± St. Dev.
6.28 ±0.13
S. epidermic/is
11 9.4 ±22.1
504.6 ±14.5
516±9
501.8 ±13.6
1007 ±20.7
1010 ±13.4
90
90
360
1440
90
360
10,746
45,414
185,760
722,592
90,630
363,600
7.00 ±0.28
7.15±0.13
5.47 ±0.13
6. 72 ±0.43
5.72 ±0.32
5.72 ±0.32
6.60 ±0.25
4.54 ±0.53
2.66 ±0.35
BDL
4.49 ±0.57
BDL
* BDL-Below Detection Limit of 2.18 CPU log
10
Table 3. CPUs for Control and Exposed Samples on Gypsum Wai I board at Low Humidity
O3 concentra-
tion (ppm)
Exposure
time (min)
Total
Exposure
(ppm-min)
Control Samples
(CPU Iog10)
Mean ± St. Dev.
Exposed Samples
(CPU Iog10)
Mean ± St. Dev.
P. brevicompactum
119.4 ±22.1
504.6 ±14.5
516±9
501.8 ±13.6
1007 ±20.7
1010 ±13.4
1014 ±17.8
90
90
360
1440
90
360
1440
10,746
45,414
185,760
722,592
90,630
363,600
1,460,160
7.13±0.16
6.61 ±0.05
7. 37 ±0.02
7. 32 ±0.09
5.80 ±0.37
6.56 ±0.15
6. 75 ±0.04
6.61 ±0.36
6.87 ±0.08
7.43 ±0.09
5.61 ±0.66
5.97 ±0.10
6.37 ±0.22
5.46 ±0.52
R. mucilaginosa
119.4±22.1
504.6 ±14.5
516±9
501. 8 ±13.6
1006 ±20.7
1010 ±13.4
90
90
360
1440
90
360
10,746
45,414
185,760
722,592
90,540
363,600
7.05 ±0.18
6.62 ±0.11
7.36 ±0.09
6.62 ± 0.22
7.06 ± 0.20
7.06 ± 0.20
7.27 ±0.19
6.85 ±0.25
7.44 ±0.04
6. 98 ±0.63
6. 76 ±0.27
6.28 ±0.13
B. atrophaeus
119.4 ±22.1
504.6 ±14. 5
516±9
501.8 ±13.6
1007 ±20.7
1010±13.4
1014±17.8
90
90
360
1440
90
360
1440
10,746
45,414
185,760
722,592
90,630
363,600
1,460,160
6.41 ±0.17
6.48 ±0.21
6.54 ±0.10
6.72 ±0.59
6.09 ±0.11
6.63 ±0.16
6. 55 ±0.36
6. 15 ±0.28
6.49 ±0.21
6.94 ±0.54
6.88 ±0.41
5.70 ±0.11
6.29 ±0.14
6.44 ±0.13
S. epidermidis
11 9.4 ±22.1
504.6 ±14. 5
516±9
501.8 ±13.6
90
90
360
1440
10,746
45,414
185,760
722,592
4.58 ±0.46
6. 34 ±0.81
5.21 ±0.50
6.71 ±0.47
4.35 ±0.52
6.20 ±0.51
5.20 ±0.62
5.28 ±0.63
-------
Ozone Antimicrobial Efficacy
O3 concentra-
tion (ppm)
1006 ±20. 7
1010±13.4
Exposure
time (min)
90
360
Total
Exposure
(ppm-min)
90,540
363,600
Control Samples
(CPU Iog10)
Mean ± St. Dev.
6.44 ±0.71
6.44 ±0.71
Exposed Samples
(CPU Iog10)
Mean ± St. Dev.
6.08 ± 0.62
5. 79 ±0.50
Table 4. CPUs for Control and Exposed Samples on Glass Slides at High Humidity
O3 concentra-
tion (ppm)
Exposure
time (min)
Total
Exposure
(ppm-min)
Control Samples
(CPU Iog10)
Mean ± St. Dev.
Exposed Samples
(CPU Iog10)
Mean ± St. Dev.
P. brevicompactum
100.2 ±6. 15
509.6 ±5.9
506 ±5
1003 ±6.9
90
90
360
90
9,018
45,864
182,160
90,270
5.81 ±0.13
6.61 ±0.27
6.19 ±0.06
5.19 ±0.46
3.10±1.41
2.18±0
BDL
BDL
R. mucilaginosa
100.2 ±6. 15
509.6 ±5.9
506 ±5
90
90
360
9,018
45,864
182,160
4.51 ±0.35
4. 18 ±0.00
7.39 ±0.02
4.57 ±0.43
2.88 ±0.92
5.69 ±0.06
B. atrophaeus
100.2 ±6. 15
509.6 ±5.9
506 ±5
1003 ±6.9
966 ± 20.5
90
90
360
90
360
9,018
45,864
182,160
90,270
358,560
6.60 ±0.24
6.42 ±0.15
6.62 ±0.03
5.85 ±0.29
6.63 ±0.19
6.69 ± 0.23
3.94 ± 0.40
BDL
3.67 ±0.17
2.30 ± 0.24
S. epidermic/is
509.6 ±5.9
90
45,864
6. 12 ±0.07
BDL
BDL - Below Detection Limit of 2.18 CPU Iog10
-------
Ozone Antimicrobial Efficacy
Table 5. CPUs for Control and Exposed Samples on Gypsum Wai I board at High Humidity
O3 concentra-
tion (ppm)
Exposure
time (min)
Total
Exposure
(ppm-min)
Control Samples
(CPU Iog10)
Mean ± St. Dev.
Exposed Samples
(CPU Iog10)
Mean ± St. Dev.
P. brevicompactum
100.2 ±6. 15
509.6 ± 5.9
506 ±5
1003 ±6.9
966 ± 20.5
90
90
360
90
360
9,018
45,864
182,160
90,270
358,560
6. 74 ±0.01
6. 79 ±0.56
7.17 ±0.21
5.80 ±0.37
6. 58 ±0.04
6. 34 ±0.80
4.66 ±0.48
BDL
3.62 ±0.31
BDL
R. mucilaginosa
100.2 ±6.15
509.6 ± 5.9
506 ±5
90
90
360
9,018
45,864
182,160
7.13±0.16
7.06 ±0.17
7. 32 ±0.09
6.76 ±0.08
6.56 ±0.18
BDL
8. atrophaeus
100.2 ±6. 15
509.6 ± 5.9
506 ±5
1003 ±6. 9
966 ± 20.5
90
90
360
90
360
9,018
45,864
182,160
90,270
358,560
6. 76 ±0.08
6.54 ±0.39
6.59 ±0.16
6. 09 ±0.13
6. 52 ±0.05
6.40 ±0.08
6.46 ±0.53
3.11 ±1.08
5. 70 ±0.15
BDL
S. epidermidis
509.6 ±5.9
90
45,864
6.12 ±0.07
BDL
BDL-Below Detection Limit of 2.18 CPU log
10
3.2 Log Change in CPUs for the Total (C*T) Ozone Exposure
To quantify the effectiveness of the ozone to inactivate the test organisms, the log change in CPUs was
plotted against the total ozone exposure (C*T). The log CPUs of either the three replicate inoculated,
unexposed controls or the four replicate inoculated exposed slides were averaged and the standard
deviation calculated. The log change was calculated as follows:
Log change = LogCFUc -LogCFUE where
Log CFUE = mean log CPUs of exposed samples (n=4)
Log CFUc = mean log CPUs of control samples (n=3).
Figures 2 through 5 show the results for each of the test organisms for both test materials at both RHs.
The X-axis is the range of ozone exposure (C*T) in ppm ozone-min. The Y-axis is the log change. The
error bars are the combined standard error of the mean of the exposed and control samples for each test.
When the Log CPU for the exposed samples was BDL, the value 2.18 CPU logio was used to calculate the
Log change.
As anticipated, the test organisms were more protected on the gypsum wallboard and at low RH. On glass
slides at low RH, no effect was observed for the B. atrophaeus and R. mucilaginosa, while the S.
epidermidis and P. brevicompactum both decreased at least 4 logs at the maximum ozone exposure.
However, at high RH on both glass slides and gypsum wallboard, all of the organisms but the R.
mucilaginosa were inactivated 4 logs. On the gypsum wallboard at low RH, none of the organisms was
inactivated as much as 2 logs.
-------
Ozone Antimicrobial Efficacy
1.00
0.00
-1.00
I
ro
6-2.00
D)
o
-3.00
-4.00
-5.00
-6.00
1.E+03
•glass slides low RH
-gypsum board low RH
—*—glass slides high RH
-x- gypsum board high RH
1.E+04 1.E+05
C*T (ppm Ozone*min)
1.E+06
1.E+07
Figure 2. Log change in CPU of P. brevicompactum over a range of ozone exposure.
1.00
0.00
-1.00
-2.00
-3.00
-4.00
-5.00
-6.00
1.E+03
-glass slides low RH
gypsum board low RH
-glass slides high RH
gypsum board high RH
1.E+04 1.E+05
C*T (ppm Ozone*min)
1.E+06
1.E+07
Figure 3. Log change in CPU of R mucilaginosa over a range of ozone exposure.
10
-------
Ozone Antimicrobial Efficacy
1.00
-5.00
-6.00
-glass slides low RH
-gypsum board low RH
-glass slides high RH
-gypsum board high RH
1.E+03
1.E+04 1.E+05
C*T (ppm Ozone*min)
1.E+06
1.E+07
Figure 4. Log change in CPU of B. atrophaeus over a range of ozone exposure.
i.oo
0.00
-1.00
0 -2.00
-3.00
-4.00
-5.00
-6.00
- glass slides low RH
- gypsum board low RH
- glass slides high RH
gypsum board high RH
l.E+03
l.E+04 l.E+05
C*T (ppm Ozone*min)
l.E+06
l.E+07
Figure 5. Log change in CPU of S. epidermidis over a range of ozone exposure.
11
-------
Ozone Antimicrobial Efficacy
4. Discussion and Conclusions
The objective of this project was to expand on work from an earlier study (Foarde et al., 1997) by testing
the effect of ozone at much higher levels on a variety of microorganisms. The goal of these experiments
was to ascertain the biocidal efficacy of ozone against four organisms - two bacteria (one spore and one
vegetative organism) and two fungi (one spore and one vegetative organism). In the earlier study, we
evaluated the effects of ozone at levels up to 9 ppm. In this study we evaluated levels ranging from 100 to
1000 ppm (Menetrez et al., 2008).
A series of experiments was performed using either glass slides or gypsum wallboard as the test surface.
The use of the impenetrable, flat-surfaced glass slides was to minimize the loss of the ozone at the surface
where the test organisms were deposited, thereby maximizing the probability of detecting an effect.
However, in order to evaluate ozone under more realistic use conditions, the test organisms were also
inoculated onto the back surfaces of an actual building material, gypsum wallboard.
In earlier experiments at lower levels of ozone, we found that the organisms on glass slides were more
readily killed than organisms on building materials. This series of experiments confirmed those results for
higher levels of ozone. It would be reasonable to assume that the difference in ozone efficacy between the
two test surfaces was due at least in part to the ability of the gypsum wallboard to inactivate the ozone
and thus to protect the spores deposited on its surface.
As stated earlier, increasing RH increases the biocidal capability of ozone. This was found to be true in
the earlier, low-level study and was confirmed in this study at higher ozone levels.
Because adverse health effects differ by organism and susceptibility of the exposure population, no
standard acceptable level of contamination exists, nor does any required level of efficacy for decontami-
nating building materials in the field, therefore, a key issue in evaluating the efficacy of any biocide,
including ozone, is to determine the acceptable number of CPUs remaining after treatment. For example,
in these experiments the inoculum was usually at least 1 x 106 CPU/sample. A 1 log reduction (90%
inactivation efficiency) would mean that 1 x 105 CPU remained after exposure. If a 4 log reduction
(99.99% inactivation efficiency) was attained, there would be 100 CPUs remaining after exposure. The
acceptability of either of these inactivation efficiencies would depend on the specific situation.
Although the specific results vary depending upon the test organism and the test surface, the overall
results of this study indicate that, even at relatively high concentrations of ozone, it is difficult to achieve
significant inactivation of organisms on material surfaces. The high ozone concentrations used in this
study would probably be difficult to maintain near or at the surface of some commonly contaminated
building materials, and even if these concentrations could be maintained in the field, it would be
challenging to achieve a significant reduction of surface biocontamination using ozone.
5. References
Atkinson, R. and W. P. L. Carter. 1984. Kinetics and Mechanisms of the Gas-Phase Reactions of Ozone
with Organic Compounds under Atmospheric Conditions. Chemical Reviews. 84(5): 440-469.
Clark, D.S. and J. Takacs. 1980. Gases as Preservative. In: Microbial Ecology of Foods. Factors Affecting
Life and Death of Microorganisms. International Commission on Microbiological Specification
of Foods. New York Academic Press. 1:170-192.
12
-------
Ozone Antimicrobial Efficacy
Foarde, K., D. VanOsdell, and R. Steiber. 1997. Investigation of Gas-Phase Ozone as a Potential Biocide.
Appl. Occup. Environ. Hyg. 12(8):535-542.
Foegeding, P.M. and F.F. Busta. 1991. Chemical Food Preservatives. Chapter 47 in: Disinfection,
Sterilization and Preservation. Fourth Edition. S.S. Block, ed. Philadelphia: Lea and Febiger.
Menetrez, M.Y., Foarde, K.K., Schwartz, T.D., Dean, T.R., Betancourt, D.A., 2009. An Evaluation of the
Antimicrobial Effects of Gas-Phase Ozone, Journal of Ozone: Science and Engineering.
Morey, P. R. 1993. Use of Hazard Communication Standard and General Duty Clause During Remedia-
tion of Fungal Contamination. In: Indoor Air '93 Proceedings 4:391-395.
U.S. EPA (Environmental Protection Agency). 1999. Wastewater Technology Fact Sheet: Ozone
Disinfection. Report Number EPA 832-F-99-063. Prepared by the Office of Water, Washington,
DC. September. Available at http://www.epa.gov/owmitnet/mtb/ozon.pdf
Weavers, L.K. and G.B. Wickramanayake, 2001. Disinfection and Sterilization by Ozone. Chapter 10 in:
Disinfection, Sterilization, and Preservation. Fifth Edition. S. Block, ed. Philadelphia: Lippincott
Williams and Wilkins.
13
------- |