SCIENTIFIC REVIEWS

FIGURE 1. Experimental setup.

ACF(d, t) )

Cnatural(d, t)

CAP(d, t)

(1)

λ(d, t) ) 1

t

lnC(d, t ) 0)

C(d, t) , (2)

PRR(d, t) ) 1

t

lnCAP(d, t ) 0)

CAP(d, t) - 1

t

lnCnatural(d, t ) 0)

Cnatural(d, t) 

(3)

PRR(d, t) ) 1

t

ln[ACF(d, t)] (4)

CADR(d, t) ) V × PRR(d, t) [m3/h] (5)

VOL. 41, NO. 2, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 607

filteredusing amembrane filter of 0.2 μmporosity andserially diluted so that the nebulizer suspension had 108-109 PFU/ mL (PFU ) plaque forming unit). MS2 phage titer was

determined by following a modified plaque assay protocol of Adams (27); Escherichia coli (ATCC 15597, strain C3000) was used as the host organism. B. subtilis is a gram-positive spore-forming bacterium withrod-shapedspores of approximately 0.7-0.8 μminwidth and 1.5-1.8 μm in length (28). B. subtilis spores have previously been used in laboratory studies as a surrogate of environmentally resistant, pathogenic bacteria (29-31). Freeze-dried bacterial spores of B. subtilis (obtained from the U.S. Army Edgewood Laboratories, Aberdeen Proving Ground, Maryland) were activated at 55-60 C for 25 min and then washed two times with sterile deionized water by vortexing followed by centrifugation at 7000 rpm for 7 min at room temperature. The total bacterial  concentration in suspension was adjusted to 108-109 per mL using a hemacytometer.

The viable bacteria were enumerated by cultivating on trypicase soy agar (TSA) media at 30 C for 18 h; the viable (culturable) concentrationinthenebulizer suspension

wasof the sameorderofmagnitude as the total concentration, i.e., 108-109 CFU/mL (CFU ) colony-forming unit). P. fluorescens bacteria (used in selected tests) are relatively

sensitive to environmental stresses. Prior to aerosolization, vegetative cells of P. fluorescens (ATCC 13525) were cultured in trypticase soy broth at 28 C for 18 h and washed similarly as B. subitilis spores. When testing with biological particles, air samples were collected using Button Samplers (SKC Inc., Eighty Four, PA) equipped with gelatin filters (SKC Inc.) and operated at a flow rate of 4 L/min for 5 min. Eight Button Samplers were utilized in each test generating one blank, one background sample, three samples taken at t ) 0, and the other three taken at a specific time interval; four time intervals were

tested: t ) 10, 15, 30, and 60 min. Additional selected experiments were performed by using a BioSampler (SKC

Inc. Eighty Four, PA) to collect P. fluorescens and B. subtilis.

The BioSampler efficiently collects viable bacteria (29) while

the liquid medium minimizes the desiccation stress. As its

cutoff size is too high to efficiently sample small MS2 virions,

the BioSampler was not used as an alternative to gelatin

filters for collecting MS2 virus.

The samples were analyzed for viable airborne virions

(PFU) andbacteria (CFU) to quantify thepercentages of those

survived over time t. These were obtained with and without

operating the air purifier. Our preliminary tests showed that

the air purifier’s operation considerably reduces the total

bioaerosol concentrationinthe chamberdue to ionemission.

Therefore, the ion emitter was temporarily disabled in the

hybrid unit when testing virus and bacteria inactivation to

ensure sufficient number ofmicroorganisms for determining

the viable count at the end of the test.

An aliquot of 200 μL of dissolved gelatin filter extract was

used for plaque assay to determine the number of airborne

active (viable) virions (PFU/cm3). Similarly, extract was

cultivated on TSAplates to obtain the airborne concentration

of viable bacteria (CFU/cm3).

Additional testing was initiated to examine whether the

biocidal effect of the air purifier took place indeed in the

aerosol phase (and not after microorganisms were collected

onfilters).For thispurpose, aerosolizedmicroorganismswere

collected on eight gelatin filters during 5 min in the chamber

without air purifier. Four filters were analyzed for viable

microorganisms immediately after this test, while the other

four were exposed to the air purifier in the chamber for 10,

15, 30, and 60 min and then analyzed. The comparison of

two sets allowedexamining if themicroorganisminactivation

occurred on filters during the collection process.

The ozone level and the air ion concentration were

monitored in real-time in the chamber using an ozone

monitor (PCI Ozone & Control Systems, Inc., West Caldwell,

NJ) and an air ion counter (AlphaLab Inc., Salt Lake City,

UT), respectively. The air temperature in the test chamber

was 24 ( 20C and the relative humidity ranged from 22 ( 2%

to 28 ( 2% as monitored with a thermo/hygrometer pen

(Fischer Scientific Co., Pittsburgh, PA).

The purifier prototype (Ecoquest International Inc.,

Greeneville, TN) used in the study utilized an ion emitter

and a specially designed RCI cell. The former produces

negative ions into indoor air, where they are acquired by

aerosol particles. It is important to note that this method is

different from air cleaning by charging particles at the

entrance of the purifier and subsequently collecting them

on metal electrodes by electrostatic precipitation. The RCI

cell features a flow optimized target structure comprising

matrices of elongated tubular elements made of polycarbonate

and arranged in a parallel orientation on opposite

sides or alternatively on four sides of a broad-spectrum UV

light source. The UV lamp utilizes argon gas with mercury

and carbide filaments with a spectral output between 100

and 367 nm. Besides, a coating was applied to the target

structure of the cell comprising hydrophilic properties and

containing the following grouping of materials: titanium

dioxide, rhodium, silver, and copper. As a result, a photocatalytic

oxidation forms reactive species, such as hydroxyl

radicals, valence-band holes, superoxide ions, and hydrogen

peroxides.

The tests were conducted in two indoor test chambers,

including a large walk-in chamber (24.3 m3) that simulated

a residential room and a smaller chamber (2.75 m3) that

simulated a confined space (e.g., bathroom, small office area,

or automobile cabin). The particle removal was investigated

in both chambers,whereas the bioaerosol viability testswere

performed in the smaller chamber thatwasmade of stainless

steel and allowed bio-decontamination. The air purifier was

tested in non-ventilated chambers (no air exchange) as it is

known that portable air cleaners are primarily beneficial in

poorly ventilatedspaces (20, 21).Air exchangewas introduced

only when testing the closed-loop ventilation/air-filtration

system equipped with an HVAC filter to compare its

performance to that of the portable air purifier in terms of

CADR.The ventilation/air-filtrationsystemwas alsodeployed

to clean the test chamber between experiments. In most of

the tests, the air purifier operated in the corner of the

chamber, facing the center.Aseparate experimentwas carried

out to examine whether its location and orientation affected

the ACF.

Results and Discussion

Particle Removal from Air. Figure 2 shows the evolution of

the concentration and particle size distribution of NaCl

aerosol when the air purifier operated in the large test

chamber. As seen from this example, the aerosol concentration

of 0.1 μmparticles decreased by a factor of 28 in 1 h and

by a factor of about 250 in 2 h; the corresponding decreases

for 1 μmparticles were approximately 10- and 50-fold. When

testing with smoke particles, the aerosol concentration

decreased even more rapidly. The above levels of the aerosol

concentration reduction are considerably greater than those

predicted by either tranquil or stirred natural decay models

(32). This result was obtained when both the air ion emitter

andtheRCI celloperatedintheunit. Interestingly, statistically

the same particle reduction effect (p > 0.05) was observed

when the RCI cell was turned off and only the ion emitter

operated. The latter finding provides the evidence that the

particle removal was achieved as a result of unipolar ion

emission but not due to photocatalytic reactions.

608 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 2, 2007

The purifier prototype (Ecoquest International Inc., Greeneville,

TN) used in the study utilized an ion emitter and a specially

designed ActivePure cell. The former produces negative ions

in to indoor air, where they are acquired by aerosol particles.

It is important to note that this method is different from air

cleaning by charging particles at the entrance of the purifier and

subsequently collecting them on metal electrodes by electrostatic

precipitation. The ActivePure cell features a flow optimized target

structure comprising matrices of elongated tubular elements

made of polycar-bonate and arranged in a parallel orientation on

opposite sides or alternatively on four sides of a broad-spectrum

UV light source. The UV lamp utilizes argon gas with mercury

and carbide filaments with a spectral output between 100 and

367 nm. Besides, a coating was applied to the target structure

of the cell comprising hydrophilic properties and containing

the following grouping of materials: titanium dioxide, rhodium,

silver, and copper. As a result, a photo-catalyticoxidation forms

reactive species, such as hydroxyl radicals, valence-band holes,

superoxideions, and hydrogen peroxides.

This finding agrees with previously published data on the

effect ofunipolar air ionizationonthe airborne concentration

(18-21). The air purification is particularly efficient at higher

initial aerosol concentrations (>104particles/cm3) that ensure

adequate interaction between the air ions and aerosol

particles. As mentioned above, the effect is expected to be

much more pronounced in non-ventilated environments

than in ventilated ones.

The aerosol reductionwas especially high for the particles

of d e 0.3 μm. E.g., when the air purifier with an ion output

of ∼1012 e/sec continuously operated in a corner of the 24.3-

m3 chamber facing the center for 2 h, ACF reached ∼30-70

for d ) 0.08-0.3 μm and ∼13-16 for d ) 0.8-2 μm (in the

tests conductedwithNaCl and smoke as challenge aerosols).

The same ACF levelsmay be achievedmore rapidly in indoor

environments of smaller volumes and slower inlarger spaces.

The experimental trends agree with the ion-induced aerosol

removal model (20).

The ACF was found to depend not only on the operation

time and the particle size but also onthe location/orientation

of the purifier in the chamber. For example, a corner location

facing the centerof the roomwas foundpreferable asopposite

to the orientation facing the wall. The difference in ACF

obtained for the center and corner locations was significant

and increased with the operation time. The shaded area in

Figure 3 presents the ion-induced Air Cleaning Factor when

the particle size-selective data were integrated over the

measured sizes of NaCl particle up to 2.5 μm and averaged

over the three selected locations/orientations in the 24.3-m3

chamber: in the corner facing the center, in the center, and

at 80 cm from the wall facing it.

Figure 4 presents the CADR values achieved by operating

the tested air purifier for five selected sizes ofNaCl and smoke

particles acting as aerosol contaminants inthenon-ventilated

24.3m3 chamber. TheCADR ranges approximately from42.1

( 0.1 to 62.1 ( 1.8 m3/h for NaCl particles of d ) 0.04-1.99

μm, and from 72.4 ( 0.9 to 115.5 ( 10.8 m3/h for smoke

particles of the same size range. The difference may be

attributed to different ability of NaCl and smoke particles to

acquire electric charges from air ions, which results in their

different mobilities and subsequently different migration

velocities. The above explanation seems valid given that

unipolar ionemissionwas shownto be themajormechanism

causing the aerosol particle concentration reduction.

In addition, Figure 4 presents the CADR values achieved

by the closed-loop air exchange system equipped with a

standard ASHRAE rating 8 HVAC filter at two air exchange

rates, 2.5 and 7.7 ACH. The data suggest that the tested

portable air purifier operating in about 25m3 non-ventilated

room is capable to provide a CADR more than twice greater

than the conventional central HVAC system with the rating

8 filter. Obviously, more efficient particulate filters provide

more rapid reduction of aerosol contaminants and may

perform better than the tested air purifier. For example,

compared to the portable unit, HEPA filter installed in the

closed-loop air exchange system of the 24.3 m3 chamber

provided approximately 4- and 3-fold greater CADRs at 2.5

and 7.7 ACH, respectively, when challenged with NaCl

particles, and 2.2- and 1.4-fold greater when challenged with

smoke particles. However, HEPA filters are rarely used in

residential central HVAC systems because of the highpressure

drop and the loading effect on their performance.

The particle removal from indoor air by the hybrid air

purification technique was also investigated in the smaller

(2.75 m3) chamber, which otherwise was utilized primarily

for assessing the viable microorganism inactivation. The

CADR values obtained with MS2 virions from the WPS

measurements were 73 ( 5 m3/h, which is in the CADRFIGURE

2. Particle concentration and size distribution of NaCl

aerosol as measured with the ELPI in the 24.3 m3 chamber with the

air purifier operating facing the chamber’s center at 1.7 m from the

measurement point. No ventilation in the chamber. The initial total

aerosol concentration ) 1.50 × 105 /cm3.

FIGURE 3. The ion-induced Air Cleaning Factor (ACF) for PM2.5

NaCl asmeasuredwith the ELPI and integrated for different locations

and orientations of the air purifier in the 24.3 m3 chamber. No

ventilation in the chamber. The initial PM2.5 aerosol concentration

) (0.356-1.50) × 105/cm3.

FIGURE 4. Clean Air Delivery Rate (CADR) determined for the NaCl

and smoke aerosols asmeasuredwith the ELPI in the non-ventilated

24.3 m3 chamber. The performance of the air purifier is compared

to that of a standard HVAC filter (ASHRAE rating ) 8) installed in

the closed-loop air exchange system of the chamber.

VOL. 41, NO. 2, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 609

range obtained for NaCl and smoke particles in the large

chamber for the viral sizes. This suggests the feasibility of

using nonbiological particles to determine the ion-induced

aerosol reduction of bio-particles of the same size range.

Furthermore, this finding implies that, at least for the particle

size range representingMS2 virions, PRRdue to ion emission

in indoor air environment is inversely proportional to the air

volume [see eq 5].

Ozone. In both test chambers (non-ventilated), the ozone

concentration gradually increased as the purifier was continuously

operating. In the 24.3-m3 chamber, it increased

from 0.006 to 0.05 ppm in about 35 min, while in a smaller

(2.75-m3) chamber the same increase occurred in approximately

5 min. However, once an air exchange was

introduced (as low as 1 ACH), the ozone concentration in

the 24.3-m3 chamber did not significantly increase as

compared to the initial level (p > 0.05). Our monitoring data

obtained with the tested unit operating in a non-ventilated

roomof ∼100m3 (not presented here) suggest that the ozone

level can be kept below0.05 ppmwhile the unit continuously

operates for many hours.

Some air purifiers utilizing ion emission and, to a greater

extent, the photocatalytic oxidation may cause greater

increase of indoor ozone concentration than the tested one.

The use of such devices in confined occupied air spaces may

not be appropriate as their continuous operation may

eventually lead to excessive ozone levels and, in the presence

of certain chemical compounds, produce nanoparticles (33).

Although the unipolar ion emission has a potential to

suppress this effect, it seems important to keep the ozone

level below existing thresholds. We believe that the solution

can be found by implementing an intermittent regime (as an

alternative to continuous one), which allows the air purifier

operating until the ozone reaches a certain level, after which

the ozone-generating element is automatically turned off to

allow the ozone concentration to drop; then the cycle can

be repeated.

Microbial Inactivation.Table 1 summarizes themicrobial

inactivationresults.Only approximately 10%of initially viable

MS2 virions survived 10-60 min exposure to the purifier in

the chamber and about 90% were inactivated. When the

natural concentration decay of aerosolized MS2 was monitored

in the chamber (with no purifier operating), we found

that the concentration of active viruses was relatively stable:

the decrease did not exceed 20.3 ( 0.9% during 1 h. The data

suggest that the viral inactivation occurs rather quickly since

the percent of survived virions did not show dependence on

the exposure time for t ) 10-60 min. Thus, a relatively short

time may be sufficient to reduce the percent of viable viruses

in an air volume by a factor of 10 while those that survived

showed remarkable resistance to the continuing stress.When

aerosolized virions are exposed to photocatalytic oxidation,

thehydroxyl radicals canaffect theproteincapsidandbinding

sites, thus disabling the virus’s subsequent interaction with

the host and formation of PFUs (34). Additionally, the TiO2

photocatalytic cell may produce oxidative damage to the

virus capsid (35) and the radicals may cause alteration in the

virus’s genetic material (36, 37). Our findings suggest that

the hybrid air purifier may be used continuously for short

time intervals or in intermittent regime to achieve considerable

virus inactivation rate. On the other hand, a prolonged

operation of the air purifier is believed to be advantageous

in environments with a continuous supply of “fresh” active

virions.

Approximately 75% of airborne B. subtilis spores exposed

to the air purifier were inactivated during the first 10 min,

85% during the first 15 min, and about 90% or greater after

30 min (Table 1). Between 30 and 60 min of exposure, we did

not observe significant decrease in the number of survived

spores (similar to the trend found for virions),which suggests

anonlinearityof the effect.Thenaturaldecay inthe culturable

count was not significant (p > 0.05) during 1 h, as measured

using the Button Samplers equipped with gelatin filters.

However, the overall standard deviation of the data obtained

in these control tests was as high as 58% and the CFU counts

from filters were close to the detection limit. To address this

issue, we measured the natural decay of viable B. subtilis

spores with the BioSampler at t ) 0 and at t ) 2 h. It was

confirmed that the viabilitywas constantwithin about (20%

in the absence of the air purifier.

In bacteria, the inactivation process by reactive hydroxyl

radicals can proceed in five reaction pathways:

•oxidation of coenzyme A causing inhibition of cell

respiration and cell death (38);

•destruction of the outermembrane of bacterial cells (12);

•oxidation of unsaturated phospholipid in bacterial cell

membrane (39);

•leakage of intracellular K+ ions (11); and

•detrimental effects on DNA and RNA (36, 37).

One reason that the inactivation of B. subtilis endospores

was time-dependent is their thickmembrane layer containing

peptidoglycans. This is consistent with the study of Matsunaga

et al. (40), who found that photooxidation of coenzyme

A by the TiO2 photocatalyst was not entirely effective against

the algae Chlorella vulgaris in water because of its thicker

cell wall. Some other self-defense mechanisms of bacteria

against the oxidationstress, including synthesis of superoxide

dismutase enzymes, can also slow down the inactivation

process (41).

Although the time was a factor in the bacterial spore

inactivation, the viability loss occurred relatively quickly for

both the MS2 virus and B. subtilis. This can be attributed to

rapid interaction of valence-band holes (h+) (TiO2 + hv f

h+ + e-.) with the organic substances, which are present in

the viral and bacterial outerwalls ormembranes. The abovementioned

interaction likely occurs before considerable

number of hydroxyl radicals (·OH) is generated in the air

volume. Although previous studies (11, 12) emphasized the

roleofhydroxyl radicals (H2O+h+ f ·OH+H+), these radicals

may not be the primary factor in microbial inactivation,

particularly in the air. Furthermore, since our experiments

were conducted in relatively dry air (RH < 30%), water

molecules were not predominant species in contact with the

catalyst, and thus the contribution of hydroxyl radicals was

likely much lower than in liquids. Shang et al. (9) have

concluded that in the gas phase, organic compounds, such

as heptane, can readily interact with photogenerated holes

while the interaction with water vapor molecules is not as

prominent. Alberici and Jardim (8) have reported that the

valence-band holes generated from TiO2 photooxidation are

capable of oxidizing any organic compound. The process

also produces hydrogen peroxide (O2 + e- f O2

•-; O2

•- + H+

fHO2

•; 2HO2

• fO2 +H2O2), which can freely penetrate into

cell membranes and walls and cause microbial inactivation

TABLE 1. Percentage of Airborne Microorganisms Survived

over Time t in the 2.75 m3 Chamber with the RCI-cell

Operating in it, as Measured via PFU Count (for MS2 Virus) or

CFU Count (for Bacillus subtilis Endospores)a

percentage (mean ( SD) of airborne microorganisms surexposure

vived in the chamber with air purifier operating during time t

time,

t (min)

MS2 virus,

[PFU/cm3]t /[PFU/cm3]t)0

Bacillus subtilis endospores,

[CFU/cm3]t/[CFU/cm3]t)0

10 9.3 ( 2.0 (n ) 5) 24.1 ( 3.7 (n ) 2)

15 9.2 ( 4.3 (n ) 12) 15.7 ( 1.7 (n ) 3)

30 8.3 ( 1.1 (n ) 8) 7.9 ( 1.1 (n ) 3)

60 10.3 ( 1.7 (n ) 5) 10.1 ( 1.3 (n ) 3)

a Bioaerosol sampling was conducted with the Button Sampler

equipped with gelatin filters. n ) number of replicates.

610 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 2, 2007

(42). Further biochemical studies on the role of gas-phase

TiO2 oxidation on the airborne microorganisms as well as

studies on the reaction kinetics at the aerosol phase seem

worthwhile to further examine the above interpretations.

Experiments with P. fluorescens revealed CFU counts

belowthedetectionlimitbothinthe test andcontrol samples.

In contrast to B. subtilis endospores, even a very short

exposure to ambient air (RH < 30%) considerably decreased

the viability of aerosolized P. fluorescens vegetative cells,

which are known to be stress-sensitive. Perhaps, microorganisms

sensitive to desiccation stress are more usable for

this kind of test if the test is performed at higher relative

humidity levels.

Additional control experiments were performed to investigate

if the viability decrease found for MS2 virus and B.

subtilis spores occurred in the aerosol phase or on the

sampling filter. For MS2, we found that 1835 ( 270 PFU/mL

and 1855 ( 325 PFU/mL developed when filter extracts were

cultivatedfromunexposedand10-minexposedgelatinfilters,

respectively. For B. subtilis, we observed 1770 ( 275 CFU/

mL and1125 ( 410CFU/mL inextracts takenfromunexposed

and 60-min exposed filters, respectively. No significant

changes in either viral or bacterial viability occurred as a

result of a non-aerosol exposure (p > 0.05). Thus, these

findings confirm that the viral and bacterial inactivation

observed in our tests indeed occurred in the aerosol phase

and was not associated with the inactivation on filters.

CombinedEffect (SampleCalculation). Itwas concluded

that the particle removal took place solely due to unipolar

ion emission, while the inactivation of viable airborne MS2

virions and B. subtilis spores occurred due to the photocatalytic

reactionpromotedby theRCI cell.Bothmechanisms

working simultaneously in a hybrid type air purifier may

result inconsiderabledecrease of the exposure topre-existing

viable aerosol biocontaminants in indoor environment.

Ozone produced by the RCI cell is not believed to cause

significant microbial inactivation because its level was not

sufficient. Tseng and Li (43) referred to 3.43 ppm as an

appropriate level for airborne MS2 virus, and Li and Wang

(44) did not observe any inactivation of airborne B. subtilis

spores at O3 as high as 20 ppm.

The following estimate was made based on the experimental

data obtained in this study. Assuming that the ioninduced

air cleaning removes about 80% of viable airborne

pathogens from a room air in 30 min and the RCI-induced

photoxidation leaves only 10% of the remaining airborne

microorganisms viable, the overall aerosol exposure to the

viable pathogen in this room after 30 min is reduced by a

factor of about 50.

The observedrapidinactivationofmicroorganismsmakes

unnecessary to run the RCI cell continuously. The data

suggest that it can be used “part-time” for 10-30 min and

“rest” for about 1-2 h until the background ozone level is

reached (proposed above as an intermittent regime), while

the ion emission can take place continuously to keep the

aerosol concentration decreasing.

Acknowledgments

The investigation was partially supported by the EcoQuest

International. The participation of Dr. K. Y. Kim was partially

funded by the Korea Research Foundation Postdoctoral

Fellowship Program. This support is greatly appreciated.We

also thank Dr. Taekhee Lee for the technical assistance.

Disclaimer: Reference to any companies or specific commercial

products does not constitute or imply their endorsement,

recommendation or favoring by the authors or the

University of Cincinnati.

Literature Cited

(1) EPA (U.S. Environmental Protection Agency). Ozone generators

that are sold as air cleaners: an assessment of effectiveness and

health consequences; http://www.epa.gov/iaq/pubs/ozonegen.

html; Accessed on May 22, 2006.

(2) ACGIH (American Conference of Governmental Industrial

Hygienists). Threshold Limit Values for Chemical Substances

and Physical Agents & Biological Exposure Indices; ACGIH:

Cincinnati, OH, 2004; p. 44.

(3) OSHA (Occupational Safety & Health Administration). Regulations

(Standards 29 CFR): Air Contaminants, 1915.1000; http://

www.osha.gov/pls/oshaweb; Accessed on May 31, 2006.

(4) FDA (U.S. Food and Drug Administration). Labeling Regulatory

Requirements for Medical Devices; HHS publication FDA 89-

4203;U.S.Governement PrintingOffice: Washington,DC, 1989.

(5) EPA(U.S. Environmental ProtectionAgency). National Ambient

Air Quality Standards (1997); http://www.epa.gov/air/criteria.

html; Accessed on May 24, 2006.

(6) Foarde, K. K.; VanOsdell, D. W.; Steiber, R. S. Investigation of

gas-phase ozone as a potential biocide. Appl. Occup. Environ.

Hyg. 1997, 12, 535-542.

(7) Li, C. -S.; Wang, Y. -C. Surface germicidal effects of ozone for

microorganisms. AIHA J. 2003, 64, 533-537.

(8) Alberici,R.M.;Canela,M.C.;Eberlin,M.N.; Jardim,W.F.Catalyst

deactivationinthe gasphase destructionofnitrogen-containing

organic compounds using TiO2/UV-VIS. Appl. Catal., B 2000,

793, 1-9.

(9) Shang, J.; Du, Y.; Xu, Z. Photocatalytic oxidation of heptane in

the gas-phase over TiO2. Chemosphere 2002, 46, 93-99.

(10) Matsunaga,T.;Tomoda,R.;Nakajima,T.;Nakamura,N.;Komine,

T. Continuous sterilization system that uses photosemiconductor

powders. Appl. Environ.Microbiol. 1988, 54, 1330-1333.

(11) Saito, T.; Iwase, T.; Horie, J.; Morioka, T. Mode of photocatalytic

bactericidal action of powdered semiconductor TiO2 on Streptococci

mutants. J. Photochem. Photobiol. 1992, 14, 369-379.

(12) Sunada,K.;Kikuchi, Y.;Hashimoto,K.; Fujishima,A.Bactericidal

and detoxification effects of TiO2 thin film photocatalysts.

Environ. Sci. Technol. 1998, 32, 726-728.

(13) Goswami, D. Y.; Trivedi, D. M.; Block, S. S. Photocatalytic

disinfection of indoor air. Trans. ASME 1997, 119, 92-96.

(14) Goswami, T. K.; Hingorani, S.; Greist, H.; Goswami, D. Y.; Block,

S. S. Photocatalytic system to destroy bioaerosols in air. J Adv.

Oxidation Tech. 1999, 4, 185-188.

(15) Pal, A.; Min, X.; Yu, L. E.; Pehkonen, S. O.; Ray, M. B.

Photocatalytic inactivation of bioaerosols by TiO2 coated

membrane. Int. J. Chem. Reactor Eng. 2005, 3, A45.

(16) Keller, V.; Keller, N.; Ledoux, M. J.; Lett, M.-C. Biological agent

inactivation in a flowing air stream by photocatalysis. Chem.

Commun. 2005, 23, 2918-2920.

(17) Lin, C. -Y.; Li, C. -S. Effectiveness of titanium dioxide photocatalyst

filters for controlling bioaerosols. Aerosol Sci. Technol.

2003, 37, 162-170.

(18) Grinshpun, S.A.;Mainelis,G.;Trunov,M.;Adhikari,A.;Reponen,

T.;Willeke,K.Evaluationof ionic airpurifiers for reducing aerosol

exposure in confined indoor spaces. Indoor Air 2005, 15, 235-

245.

(19) Lee,B.U.; Yermakov,M.;Grinshpun, S.A.Unipolar ionemission

enhances respiratory protection against fine and ultrafine

particles, J. Aerosol Sci. 2004, 35, 1359-1368.

(20) Mayya, Y. S.; Sapra, B. K.; Khan, A.; Sunny, F. Aerosol removal

by unipolar ionization in indoor environments. J. Aerosol Sci.

2004, 35, 923-941.

(21) Grinshpun, S. A.; Adhikari, A.; Lee, B. U.; Trunov, M.; Mainelis,

G.; Yermakov, M.; Reponen, T. Indoor air pollution control

through ionization. In Air Pollution XII; Brebbia, C. A., Ed.; WIT

Press: Southampton, U.K., 2004; pp 689-704.

(22) Kekez, M. M.; Sattar, S. A. A new ozone-based method for virus

inactivation: preliminary study.Phys.Med.Biol. 1997, 42, 2027-

2039.

(23) O’Connel, K. P.; Bucher, J. R.; Anderson, P. E.; Cao, C. J.; Khan,

A. S.;Gostomski,M.V.;Valdes, J. J.Real-time fluorogenic reverse

transcription-PCR assays for detection of bacteriophages MS2.

Appl. Environ. Microbiol. 2006, 72, 478-483.

(24) Jones, M. V.; Bellamy, K.; Alcock, R.; Hudson, R. The use of

bacteriophageMS2 as amodel systemto evaluate virucidalhand

disinfectants. J Hosp. Infect. 1991, 17, 279-285.

(25) Havelaar, A. H.; van Olphen, M.; Drost, Y. C. F-specific RNA

bacteriophages are adequatemodelorganisms for enteric viruses

in fresh water. Appl. Environ. Microbiol. 1993, 59, 2956-2962.

(26) Allwood, P. B.; Malik, Y. S.; Hedberg, C. W.; Goyal, S. M. Survival

of F-specific RNA coliphage, feline Calicivirus, and Escherichia

coli in water: a comparative study. Appl. Environ. Microbiol.

2003, 69, 5707-5710.

(27) Adams,M.H.Bacteriophages; Interscience Publishers, Inc: New

York, 1959.

VOL. 41, NO. 2, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 611

The following estimate was made based on the experimental data

obtained in this study. Assuming that the ion induced air cleaning

removes about 80% of viable airborne pathogens from a room air

in 30 min and the ActivePure (RCI) -induced photoxidation leaves

only 10% of the remaining airborne micro organisms viable, the

overall aerosol exposure to the viable pathogen in this room after

30 min is reduced by a factor of about 50.

The observed rapid inactivation of micro organisms makes

unnecessary to run the ActivePure cell continuously. The data

suggest that it can be used “part-time” for 10-30 min and “rest” for

about 1-2 h until the background ozone level is reached (proposed

above as an intermittent regime), while the ion emission can take

place continuously to keep the aerosol concentration decreasing.

(28) Johnson, B.; Martin, D. D.; Resnick, I. G. Efficacy of selected

respiratory protective equipment challenged with Bacillus

subtilis subsp. niger. Appl. Environ. Microbiol. 1994, 60, 2184-

2186.

(29) Lin, X; Reponen, T. A.; Willeke, K.; Grinshpun, S. A.; Foarde, K.;

Ensor, D. Long-term sampling of airborne bacteria and fungi

into a non-evaporating liquid. Atmos. Environ. 1999, 33, 4291-

4298.

(30) Lin, X.; Reponen, T.; Willeke, K.; Wang, Z.; Grinshpun, S. A.;

Trunov,M. Survival of airbornemicroorganisms during swirling

aerosol collection. Aerosol Sci. Technol. 2000, 32, 184-196.

(31) Mainelis, G.; Grinshpun, S.; Willeke, K.; Reponen, T.; Ulevicius,

V.; Hintz, P. Collection of airborne microorganisms by electrostatic

precipitation. Aerosol Sci. Technol. 1999, 30, 127-144.

(32) Hinds, W. C. Aerosol Technology: Properties, Behavior, and

Measurement of Airborne Particles, 2nd ed.; John Wiley & Sons,

Inc.: New York, 1999.

(33) Grinshpun, S. A.; Toivola, M.; Lee, S.-A.; Reponen, T. Formation

of Nanoparticles in indoor air at an increased ozone level. In

Abstracts of the 24th AnnualMeeting of the American Association

for AerosolResearch,Austin, Texas,U.S.A.,October 17-21, 2005;

AAAR: Mt. Laurel, NJ, 2005; p 327.

(34) Von Sonntag, C. Disinfection by free radicals and UV-radiation.

Water Supply 1986, 4, 11-18.

(35) Turchi, C. S.; Ollis, D. F. Photocatalytic degradation of organic

water contaminants: mechanisms involving hydroxyl radical

attach. J. Catal. 1990, 122, 178-192.

(36) Dunford, R.; Salinaro, A.; Cai, L.; Serpone, N.; Horikoshi, S.;

Hidaka, H.; Knowland, J. Chemical oxidation and DNA damage

catalyzed by inorganic sunscreen ingredients. FEBS Lett. 1997,

418, 87-90.

(37) Hidaka, H.; Horikoshi, S.; Serpone, N.; Knowland, J. In vitro

photochemical damage to DNA, RNA, and their bases by an

inorganic sunscreen agent on exposure to UVA and UVB

radiation. J. Photochem. Photobiol., A 1997, 111, 205-213.

(38) Vohra, A.; Goswami, D. Y.; Deshpande, D. A.; Block, S. S.

Enhanced photocatalytic inactivation of bacterial spores on

surfaces in air. J. Ind. Microbiol. Biotechnol. 2005, 32, 364-370.

(39) Maness, P. C.; Smolinski, S.; Blake, D. M.; Huang, Z.; Wolfrum,

E. J.; Jacoby, W. A. Bactericidal activity of photocatalytic TiO2

reaction: toward an understanding of its killing mechanism.

Appl. Environ. Microbiol. 1999, 65, 4094-4098.

(40) Matsunaga, T.; Tamoda, R.; Nakajima, T.; Wake, H. Photoelectrochemical

sterilization of microbial cells by semiconductor

powders. FEMS Microbiol. Lett. 1985, 29, 211-214.

(41) Rinco´n, A. G.; Pulgarin, C. Photocatalytical inactivation of E.

coli: effect of (continuous-intermittent) light intensity and of

(suspended-fixed) TiO2 concentration. Appl. Catal., B 2003, 44,

263-284.

(42) Marquis, R. E.; Sim, J.; Shin, S. Y. Molecular mechanism of

resistance toheat andoxidativedamage. J.Appl.Bacteriol. Symp.

Suppl. 1994, 76, 40S-48S.

(43) Tseng, C.-C.; Li, C.-S. Ozone for inactivation of aerosolized

bacteriophages. Aerosol. Sci. Technol. 2006, 40, 683-689.

(44) Li, C.-S.; Wang, Y. C. Inactivation effects of airborne ozone for

bioaerosols. J. Environ. Health 2006, submitted.

Received for review June 8, 2006. Revised manuscript received

September 19, 2006. Accepted October 20, 2006.

ES061373O

612 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 2, 2007