Citation
Inactivation of Microorganisms by Photocatalytic Nanostructures Under Dry Conditions

Material Information

Title:
Inactivation of Microorganisms by Photocatalytic Nanostructures Under Dry Conditions
Creator:
ZHAO, JUE ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Bacteria ( jstor )
Bacterial spores ( jstor )
Disinfection ( jstor )
Irradiation ( jstor )
Lamps ( jstor )
Luminous intensity ( jstor )
Microbial colony count ( jstor )
Microorganisms ( jstor )
Spore dispersal ( jstor )
Spore germination ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Jue Zhao. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
6/30/2007
Resource Identifier:
659806718 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

INACTIVATION OF MICROORGANI SMS BY PHOTOCATALYTIC NANOSTRUCTURES UNDE R DRY CONDITIONS By JUE ZHAO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by Jue Zhao

PAGE 3

This document is dedicated to my parents, YaPing Wu and LiSen Zhao, and my husband, Bin Hua, for always supporting me and sacr ificing so much so that I may pursue my dreams

PAGE 4

iv ACKNOWLEDGMENTS I would like to thank my advisor, Dr. Be n Koopman, for mentoring my research. His patience and generosity were helpful dur ing the whole four-year study. Thanks are also due to Dr. Brij Moudgil, Dr. Samuel Farrah, Dr. Joseph Delfino, and Dr. Wolfgang Sigmund for serving as my committee members and for their valuable recommendations. I would like to express my gr atitude to many individuals with which I have had the pleasure to work during my PhD studies. I thank Dr. Samuel Farrah, Dr. Jerzy Lukasik, Mr. Johnny Davis, Ms. Shannon, and other st aff and students in the Microbiology and Cell Sciences Department for introducing and training me into microbiological world. Mr. Vijay Krishina has to be thanked for his great cooperation during my experiments. I would also like to thank Dr. Samuel Farra h for allowing and helping me to use his facilities during spore puri fication process and thank Dr. Wolfgang Sigmund and Mr. Georgios Pyrgiotakis for allowing me to use th eir facilities during dye tests. Mr. Gill Brubaker, Mr. Gary Scheiffele and Ms. Vane sse Kuder need to be thanked for their training and help in image analysis and many other technical supports. Dr. Seymour Block and Dr. Yogi Goswami mu st be thanked for allowing me to use their equipment for photocatalytic oxidation experiments. Their support and helpful discussions are greatly appr eciated. Thanks go to Dr . Wolfgang Sigmund, Mr. SungHwan Lee, and Mr. Georgios Pyrgiotakis fo r providing photocatalysts and invaluable discussions during photocatalyt ic oxidation experiments.

PAGE 5

v In addition, I would like to thank director of the Particle Engineering Research Center (PERC), Dr. Brij Moudgil, for guiding and encouraging me to produce good work as a PERC graduate student. Thanks go to all staff at PERC for their guiding, training, and discussions throughout the entire work. Also thanks go to Chuck Fender and all staff at UF Water Reclamation Facil ity for their wonderful support. Finally, I would like very much to ac knowledge the contri butions of fellow students to my research. They are alwa ys good friends of mine. Smithi Pumprueg introduced and trained me to the photocatalyt ic inactivation of b acterial endospore area, and Vijay Krishina helped with a lot of work. I would like to acknowledge the financial support of an Alumni Fellowship from the University of Florida. Additional suppor t was provided by the Particle Engineering Research Center at the University of Fl orida, the National Sc ience Foundation grant #EEC-94-02989, and the Industrial Partners of th e Particle Engineering Research Center.

PAGE 6

vi TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................xi LIST OF FIGURES.........................................................................................................xiv ABSTRACT....................................................................................................................xvi i CHAPTER 1 INTRODUCTION........................................................................................................1 2 LITERATURE REVIEW.............................................................................................4 2.1 Bacterial Endospores..............................................................................................4 2.1.1 The Structure and Protective Mech anisms of Bacterial Endospores............4 2.1.2 Sporulation...................................................................................................7 2.1.3 Germination..................................................................................................8 2.2 Semiconductor Photocatalysis................................................................................9 2.3 Photocatalytic Inactivation of Microbes...............................................................11 2.3.1 Systems with Water as Continuous Phase..................................................12 2.3.1.1 Light and photocatalyst concentration effects..................................12 2.3.1.2 Oxygen.............................................................................................16 2.3.1.3 Microorganism load and species......................................................17 2.3.1.4 Hydrogen ion activity and solutes....................................................20 2.3.1.5 Microstructure, size, and immobilization.........................................23 2.3.1.6 Dopants, admixtures, and soluble iron.............................................26 2.3.1.7 Synthesized photocatalyst nanocomposite.......................................27 2.1.3.8 Physiological state and generation of microbes...............................28 2.3.2. Systems with Air as the Continuous Phase...............................................29 2.4 Mechanisms of Photocatalytic Inactivation of Microbes.....................................32 2.4.1 Diffusion of TiO2 or ROS into the Cell......................................................36 2.4.2 Disinfection Action of TiO2 is a Combination Action...............................37 2.5 Kinetics and Modeling of P hotocatalytic Disinfection.........................................37 3 GOAL, HYPOTHESES AND RATIONALE, AND SCIENTIFIC MERIT.............48 3.1 Goal....................................................................................................................... 49

PAGE 7

vii 3.2 Hypotheses and Objectives...................................................................................49 3.3 Scientific Merit of Proposed Research.................................................................50 4 EVALUATION OF BACTERIAL ENDO SPORE PURIFICATION METHODS...51 4.1 Introduction...........................................................................................................51 4.2 Materials and Methods.........................................................................................53 4.2.1 Chemicals...................................................................................................53 4.2.2 Preparation and Storage of Agar Plates......................................................53 4.2.3 Storage and Inoculation of Test Strains......................................................53 4.2.4 Bacterial Culture.........................................................................................54 4.2.5 Spore Purification.......................................................................................55 4.2.6 Spore Analysis............................................................................................57 4.2.7 Data Analysis..............................................................................................59 4.2.8 TEM Sample Preparation...........................................................................59 4.3 Results...................................................................................................................5 9 4.3.1 Purity and Yield..........................................................................................59 4.3.2 Storage Time..............................................................................................60 4.3.3 Spore Integrity............................................................................................60 4.3.4 Time Investment and Complexity..............................................................61 4.4 Discussion.............................................................................................................61 4.4.1 Purity and Yield..........................................................................................61 4.4.2 Spore Integrity............................................................................................62 4.5 Conclusions...........................................................................................................62 5 EFFECTS OF SURFACE TYPE, APPLICATION METHOD AND OTHER FACTORS ON SURFICIAL PA RTICLE DISTRIBUTIONS...................................71 5.1 Introduction...........................................................................................................71 5.2 Background...........................................................................................................72 5.2.1 Pipetting......................................................................................................72 5.2.2 Immersion...................................................................................................74 5.2.3 Filtration.....................................................................................................74 5.2.4 Nebulizing..................................................................................................75 5.2.5 Other Methods............................................................................................75 5.2.6 Summary.....................................................................................................77 5.3 Materials and Methods.........................................................................................78 5.3.1 Washing of Cover Slips and Glass Slides..................................................78 5.3.2 Preparation of Suspensions of S pores, Nanotitania and Mixtures of Spores and Nanotitania....................................................................................78 5.3.3 Suspension Application Methods on Glass, Quartz, Plastic and Modified Glass and Plastic Surfaces................................................................79 5.3.3.1 Dipping method................................................................................79 5.3.3.2 Pipetting method..............................................................................79 5.3.4 Filtration Method........................................................................................80 5.3.5 Image Analysis...........................................................................................80 5.3.6 Assessment of Distribution Quality............................................................81

PAGE 8

viii 5.4 Results and Discussion.........................................................................................82 5.4.1 Effect of Spore Suspension Drying Process on Germination of Spores....82 5.4.2 Examples of Assessment of Qu ality of Particle Distributions...................82 5.4.3 Spore or Mixtures of Spore and T itania Distribution on Glass Surfaces...84 5.4.4 Spore or Mixtures of Spore a nd Titania Distribution on Quartz Surfaces......................................................................................................85 5.4.5 Spore or Mixtures of Spore a nd Titania Distribution on Plastic Surfaces......................................................................................................85 5.4.5.1 Effect of spore concentration on distribution of spores...................85 5.4.5.2 Effect of spore application method on distribution of spores..........87 5.4.5.3 Effect of surface temperat ure on distribution of spores...................88 5.4.5.4 Effect of titania concentrat ion on distribution of titania..................89 5.4.5.5 Effect of titania concentratio n on distribution of spore/titania mixtures....................................................................................................90 5.4.5.6 Effect of spore concentration on distribution of spore/titania mixtures....................................................................................................91 5.4.6 Spore or Mixtures of Spore and T itania Distribution on Modified Glass and Plastic Surfaces.........................................................................................91 5.4.6.1 Glass cover slips Pre-coated with PVA............................................92 5.4.6.2 Glass cover slips Pre-coated with PVA and SDS.............................95 5.4.6.3 Glass Pre-sputter coated by a platinum/gold mixture.......................96 5.4.6.4 Plastic cover slips Pr e-coated with SDS...........................................96 5.4.7 Spore or Mixtures of Spore and Titania Dist ribution on Filter Membrane Applied by Filtration Method........................................................98 5.4.7.1 Preliminary investigation on fact ors affecting spore distribution on membranes...........................................................................................98 5.4.7.2 Mixtures of spore and tita nia distribution on Anodisc membrane...............................................................................................104 5.4.8 Comparison of Spore or Mixture Di stribution on Glass Surface with an Anodisc Filter Membrane Surface.................................................................106 5.5 Summary.............................................................................................................108 5.6 Conclusion..........................................................................................................110 6 ENUMERATION OF VIABLE SPORES ON SURFACES....................................131 6.1 Introduction.........................................................................................................131 6.2 Background.........................................................................................................132 6.2.1 Swab/Wiping Based.................................................................................132 6.2.2 Immersion.................................................................................................134 6.2.3 PVA..........................................................................................................136 6.2.4 Crushing...................................................................................................137 6.2.5 Agar Contact.............................................................................................137 6.2.6 Release of Spores and Cell Fragments to Air Currents............................140 6.2.7 Sonication.................................................................................................141 6.2.8 Mineralization...........................................................................................141 6.2.9 Image Analysis of Spores on Surfaces.....................................................141 6.2.10 Summary.................................................................................................142

PAGE 9

ix 6.3 Materials and Methods.......................................................................................143 6.3.1 Washing of Cover Slips, Glass Slides and Quartz Slides.........................143 6.3.2 Preparation of PVA Suspension...............................................................143 6.3.3 Removal of Spores from Surfaces for Viable Counts by PVA Sampling Method...........................................................................................................143 6.3.4 Preparation of Titania Suspension............................................................144 6.3.5 Preparation of Spore Suspension..............................................................144 6.3.6 Spore, Nanotitania and Spore/Nanotitania Suspension Application Methods on Glass, Quartz, Plastic and Modified Glass and Plastic Surface...........................................................................................................144 6.3.7 Pre-coating Glass Cover Slips with SDS..................................................145 6.3.7.1 Dipping...........................................................................................145 6.3.7.2 Spreading........................................................................................145 6.3.8 Pre-coating Glass Cover Slip with PVA...................................................145 6.3.8.1 Dipping...........................................................................................145 6.3.8.2 Spreading........................................................................................146 6.3.9 Pre-coating Glass Cover Slip with PVA and SDS...................................146 6.3.10 Pre-s putter Coating of Gla ss Cover Slips with Pt/Au.............................146 6.3.11 Pre-coating Plastic Cover Slips with SDS..............................................146 6.3.12 Filtration.................................................................................................146 6.3.13 Enumeration of Spores on Glass Surfaces by Sonication......................147 6.3.14 Enumeration of Spores on Glass Surfaces and Carbon or Pt/Au PreSputter Coated Glass Surface by Immersion.................................................147 6.4 Results and Discussion.......................................................................................148 6.4.1 Recovery of Spores From Glass Surfaces................................................148 6.4.1.1 Polyvinyl alcohol sampling methods.............................................148 6.4.1.2 Sonication method..........................................................................153 6.4.1.3 Immersion method..........................................................................154 6.4.2 Recovery of Spores from Quartz Surfaces by PVA Sampling Method...154 6.4.3 Recovery of Spores from Plastic Surface by PVA Sampling Method.....155 6.4.4 Recovery of spores from modifi ed glass and plastic surface by PVA based method.................................................................................................156 6.4.4.1 Glass cover slip Pre-coated with PVA...........................................156 6.4.4.2 Glass cover slip Pre-coated with SDS............................................158 6.4.4.3 Glass cover slip Pre-co ated with PVA and SDS............................159 6.4.4.4 Presputter Pre-coated gla ss with plantinum and gold....................159 6.4.4.5 Plastic cover slip Pr e-coated with SDS..........................................159 6.4.5 Recovery of Spores from Anodisc Filter Membrane Surface..................160 6.4.5.1 Preliminary investigation of alternative methods for spore recovery..................................................................................................160 6.4.5.2 Sonication method..........................................................................162 6.4.6 Comparison of Spore Recovery from Plastic and Glass Surface by PVA Method and from Anodisc Filter Membrane Surface by Sonication Method...........................................................................................................167 6.5 Summary.............................................................................................................168 6.6 Conclusions.........................................................................................................171

PAGE 10

x 7 EFFECT OF TITANIA POWDER AGAINST B. CEREUS ENDOSPORES ON MEMBRANE SURFACE UNDER 350 NM UV IRRADIATION IN DRY STATE......................................................................................................................185 7.1 Introduction.........................................................................................................185 7.2 Effect of Light Intensity on Photolysis and Photocatalysis................................189 7.3 Materials and Methods.......................................................................................189 7.3.1 Spore Cultivation......................................................................................189 7.3.2 Spore Harvesting and Purification............................................................189 7.3.3 Surface......................................................................................................190 7.3.4 Filtration...................................................................................................190 7.3.4 Titania Coating on the Membrane............................................................190 7.3.5 Application of Spores on Tita nia Coating on the Membrane...................191 7.3.6 Application of Spores on the Membrane..................................................191 7.3.7 Experimental Setup and Procedure..........................................................191 7.3.8 Viable Counting of Spores on the Membrane Surface.............................193 7.3.9 Data Analysis...................................................................................................193 7.4 Results and Discussion.......................................................................................194 7.4 Summary and Conclusions.................................................................................198 APPENDIX A EFFECT OF VARIABLES ON PHOTOCATALYTIC INACTIVATION OF MICROORGANISMS..............................................................................................209 B PRELIMINARY RESULTS OF EFFECT OF TITANIA POWDER AGAINST B. CEREUS ENDOSPORES ON MEMBRANE SURFACE UNDER 350 NM UV IRRADIATION IN DRY STATE............................................................................263 LIST OF REFERENCES.................................................................................................285 BIOGRAPHICAL SKETCH...........................................................................................311

PAGE 11

xi LIST OF TABLES Table page 2-1 Optimum TiO2 concentrations in aqueous system...................................................47 3-1 Important parameters for testing photo catalytic inactivation of dry endospores.....50 4-1 Variation of spore purification methods...................................................................69 4-2 Time investment and complexity of spore purification methods.............................70 5-1 Factors and levels te sted in this chapter.................................................................122 5-2 Approaches for applicat ion of microbes on surfaces.............................................123 5-3 Examples of grading of distribution quality...........................................................125 5-4 Effect of spore concentr ation on their distri bution on plastic cove r slip surface...125 5-5 Effect of mixture concentration of s pore and titania their distribution on plastic cover slip surface....................................................................................................126 5-6 Distribution of spores an d mixtures on glass cover s lip pre-coated by (polyvinyl alcohol) PVA using spreading method...................................................................126 5-7 Effect of sodium dodecyl sulfate (SDS ) coating condition for plastic cover slip on spore distribution...............................................................................................127 5-8 Scanning electronic microscopy (SEM ) bacterial endospore count on three polycarbonate membranes using Anodisc and separate as pad..............................128 5-9 SEM bacterial endospore count on two Anodisc membranes................................128 5-10 Optical microscopy bacterial endospo re count on one glass cover slip.................128 5-11 Surface coverage by mixture of spores and titania on glass cover slip and membrane surfaces.................................................................................................129 5-12 Distribution performance of spores alone on surfaces...........................................129 5-13 Bacterial endospores count on surfaces by image analysis....................................129

PAGE 12

xii 5-14 Distribution performan ce of mixtures on surfaces.................................................130 6-1 Factors and levels te sted in this chapter.................................................................173 6-2 Enumeration of viab le microbes on surfaces.........................................................174 6-3 Effect of PVA source and PVA volume on difficulty of pee ling PVA film from glass cover slips and glass slides............................................................................176 6-4 Recovery percentage of spores from glass.............................................................177 6-5 Effect of PVA source and PVA concen tration on PVA method applied on glass cover slip................................................................................................................178 6-6 Recovery of different titania conc entration from glass cover slips by PVA methods..................................................................................................................179 6-7 Recovery percentage of spores from plastic..........................................................179 6-8 Effect of modified glass surface (P VA pre-coating) on particle removal by peeling PVA pre-coating directly...........................................................................179 6-9 Effect of SDS coating of plastic cove r slip on spore recovery by PVA sampling method....................................................................................................................180 6-10 Viable count results from A nodisc 25 by four sampling method..........................180 6-11 Effect of sonication inte nsity on percentage recoveries of viable spores from Anodisc 25 membrane............................................................................................181 6-12 Effect of sonication inte nsity on percentage recoveries of viable spores from mixtures of spores and titani a applied to Anodisc 25 surfaces..............................181 6-13 Comparison of spore recovery from pl astic and glass cover slip by PVA method with that from membrane by sonication method....................................................182 6-14 Evaluation of spore recovery perfor mance for all the surfaces and methods investigated.............................................................................................................183 6-15 Spore distribution quality versus recove ry performance for all the surfaces and methods investigated..............................................................................................184 7-1 Effect of light intensity on photolysis and photocatalysis......................................207 7-2 Inactivation of B. cereus s pores by UV alone and UV+P25..................................208 A-1 Photocatalytic inactiva tion of microorganisms by TiO2 and other materials.........210

PAGE 13

xiii B-1 Inactivation of B. cereus spores by UV+P 25. Mixture of spores and titania were applied to membrane surface. (trial 1)...................................................................282 B-2 Inactivation of B. cereus spores by UV alone and UV+P25 by applying mixture of spores and titania to membrane surface.............................................................283 B-3 Inactivation of B. cereus spores by UV alone and UV+P25 by adding spores after application of titania to the membrane...........................................................284

PAGE 14

xiv LIST OF FIGURES Figure page 2-1 Structure of a typical bacterial spore........................................................................45 2-2 Stages of sporulation................................................................................................45 2-3 Primary steps for photoelectrochemical mechanism................................................46 2-4 Time scales for photoelectrochemical mechanisms.................................................46 4-1 Optical micrograph of spore suspension after heat shock treatment........................63 4-2 Purity and yield of spore suspensions pur ified after 3 days of culture incubation...64 4-3 Purity and yield of spore suspensi ons purified after 10 days of culture incubation. ..............................................................................................................65 4-4 Effect of storage time on relative pu rity of treated spore suspensions ...................66 4-5 Transmission electron microscopy (TEM ) image of spores from suspension purified by ASTM method.......................................................................................67 4-6 TEM image of spore from suspension purified by daily water wash method..........67 4-7 TEM image of spores from suspension purified by lysozyme method....................68 5-1 Examples of assessment of pa rticle distributions quality:.....................................111 5-2 Optical micrograph of spore, nanot itania and mixture on glass cover slip............113 5-3 Optical micrograph of spore di stribution on plastic cover slip..............................114 5-4 Optical micrograph of spore (105 CF U/mL) distribution on plastic surface by dipping method.......................................................................................................114 5-5 Optical micrograph of mixture (spor e 106 CFU/mL, titania 0.01%) distribution on PVA dip pre-coated glass cover slip with 5% PVA (Sigma) for 10 min with rinse........................................................................................................................11 5 5-6 Optical micrograph of spore and mixtur e distribution on PVA spread pre-coated glass cover slip.......................................................................................................116

PAGE 15

xv 5-7 Scanning electron microscope (SEM) image of spore 106 CFU/mL distribution on glass cover slip sputter pre-coated with Pt/Au..................................................117 5-8 Optical micrograph of mixture (spore, 106CFU/mL; titania, 0.01%) on plastic cover slip pre-coated with 10% SDS for 10 min....................................................118 5-9 SEM images of Bacillus cereus endospores (7 107 CFU) dispersed on Anodisc 25 membrane, quality of distribution = 3+.............................................................118 5-10 Optical images of Bac illus cereus endospores (1.4 107 CFU) dispersed on Anodisc 25, gold pre-coated Anodi sc and polycarbonate membrane....................119 5-11 SEM images of mixture of spor es and titania on Anodisc membrane...................120 5-12 SEM images of mixture of spor es and titania on Anodisc membrane...................121 6-1 Effect of sonication intensity on viab le spore counts from membrane surface......172 6-2 Effect of sonication intensity at 5.0 a nd 6.0 on viable spore counts in mixtures of spores and titania....................................................................................................172 7-1 Photocatalytic expe riment test facility...................................................................200 7-2 Determination of LD90, D value and inactivation rate coefficient from relationship between spore survival ratio and time (Inactivation of B. cereus by UV+P25)................................................................................................................200 7-3 Inactivation of B. cereus spores by UV alone and UV+P25 in trial 1...................201 7-4 Inactivation of B. cereus spores by UV alone and UV+P25 in trial 2...................201 7-5 Inactivation of B. cereus spores by UV alone and UV+P25 in Trial 3..................202 7-6 Inactivation of B. cereus spores by UV alone and UV+P25 in trial 4...................202 7-7 Inactivation of B. cereus spores by UV alone and UV+P25 in trial 5...................203 7-8 Inactivation of B. cereus spores by UV alone and UV+P25 in trial 6...................203 7-9 Inactivation of B. cereus spores by UV alone and UV+P25 at different light intensity..................................................................................................................204 7-10 Models on effect of UVA light intens ity on photocatalysis c ontribution to spore inactivation on surfaces in dry state under UVA irradiation..................................205 7-11 Distribution of spores on membrane surface..........................................................206 7-12 Distribution of spores on tit ania coated membrane surface...................................206

PAGE 16

xvi B-1 SEM image of a mixture of 106 CF U spores and 0.1 mg titania on Anodisc membrane...............................................................................................................273 B-2 SEM image of a mixture of 105 CF U spores and 0.01 mg titania on Anodisc membrane...............................................................................................................273 B-3 SEM image of a mixture of 104 CFU spores and 0.001 mg titania on Anodisc membrane...............................................................................................................274 B-4 SEM image of a mixture of 5 106 CFU spores and 0.05 mg titania on Anodisc membrane at a) 5000 magnification and b) 250 magnification.........................274 B-5 Inactivation of B. cereus spores by UV+P25. The spores and titania were mixed before application to membrane. (Trial 1).............................................................275 B-6 Inactivation of B. cereus spores by UV and UV+P25. The spores and titania were mixed before application to membrane. (Trial 2)..........................................276 B-7 Inactivation of B. cereus spores by UV and UV+P25. The spores and titania were mixed before application to memb rane. (Trial 3, repeat of trial 2)...............276 B-8 Inactivation of B. cereus spores by UV and UV+P25. The spores and titania were mixed before application to membrane. (Trial 4)..........................................277 B-9 SEM image of a mixture of 5 106CFU spores and 0.1 mg titania on Anodisc membrane...............................................................................................................277 B-10 Inactivation of B. cereus spores by UV and UV+P25. The spores and titania were mixed before application to membrane. (Trial 5)..........................................278 B-11 Inactivation of B. cereus spores by UV alone and UV+P25. Spores were added after application of titania to the membrane. (Trial 6)...........................................278 B-12 SEM image membrane pre-coated with 2.5 mg 1 mg titania.................................279 B-13 SEM image membrane pre-coated 1 mg titania.....................................................279 B-14 SEM image of 106CFU spor es distribution on membrane pre-coated with 1 mg titania......................................................................................................................28 0 B-15 Inactivation of B. cereus spores by UV alone and UV+P25. Spores were added after application of titania to the membrane. (Trial 7)...........................................281 B-16 Inactivation of B. cereus spores by UV alone and UV+P25. Spores were added after application of titania to the membrane. (Trial 8)...........................................281

PAGE 17

xvii Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy INACTIVATION OF MICROORGANI SMS BY PHOTOCATALYTIC NANOSTRUCTURES UNDE R DRY CONDITIONS By Jue Zhao December 2006 Chair: Ben Koopman Major Department: Environmental Engineering Sciences The endospores of certain spore-forming bacteria are very resistant to harsh environments and thus difficult to eliminate fr om foods or the local environment. A need therefore exists for development of improved methods for rapid inactivation of bacterial endospores. Many studies have addressed inactiv ation of spores in aqueous systems, but very limited information is available on photocatalytic inactivation of endospores on surfaces in air. Methods to study photocatalyt ic inactivation of endospores on surfaces in air were therefore developed in the presen t research. These methods included spore purification, particle distri bution on surfaces and recovery and enumeration of viable spores from surfaces. Five methods for bacterial endospore pur ification (ASTM E2111-00, ethanol, heat shock, daily water washes, and lysozyme tr eatment) were applied to suspensions of Bacillus cereus . Ethanol treatment was optimum fo r purifying 3-day cultures, whereas the ASTM method was most favorable for 10-day cultures.

PAGE 18

xviii Distribution quality of spores and mixtur e of spore and photocatalyst on surfaces was quantitatively and qualitat ively evaluated. Spores a nd photocatalyst distributed on Anodisc membranes by filtration method demonstr ated the most uniform distributions of spores and mixture on surfaces, whereas the di stribution quality of particles over plastic by a pipetting method was the worst. The di stribution quality of spores and photocatalyst on glass by a pipetting method was inferior to Anodisc membrane s but superior to plastic. Enumeration of viable spores on plastic, glass and membrane surfaces by different surface sampling methods (lifting of spores fr om glass and plastic by polyvinyl alcohol (PVA) or removal of spores from Anodisc membranes by sonication) was compared and investigated. A 100% recovery percenta ge and a high sampling consistency were achieved by sonication in conjunction with A nodisc membranes. Recovery of viable spores from glass by the PVA method gave th e lowest recoveries. Use of PVA to lift spores from plastic gave good recoveries. Enhanced inactiva tion rate by TiO2 has been widely investigated. However, few experiments have been carried out with bacterial spores, even less for dry state inactivation. A controversy ex ists that contacting with tita nia didn’t improve disinfection rate under UVA. In this study, it was found th at contact of UVA irra diated spores with titania did not give higher inactivation rate co efficients than UVA irra diation of spores in the absence of titania at high light intensity. However, at light intensities of 10 to 30 W/m2, significantly higher inactivati on rate coefficients were ac hieved in the presence of titania. Thus, light intensity plays an important role in the difference of inactivation rate under UVA irradiation in the pres ence and absence of titania.

PAGE 19

1 CHAPTER 1 INTRODUCTION Endospores, thick-walled resting forms of bacteria, are formed under unfavorable conditions. The unique structur e of endospores contributes to their high resistance to harsh conditions such as heat, desi ccation, chemicals, UV radiation, and radiation, which enable them to survive in the envir onment for years or decades. Once conditions become favorable, endospores can quickly transform into vegetative, actively growing cells. Because of these characteristics, that is, high resistance to environmental factors, long stability, the ease w ith which they can be grown and processed, and the existence of highly toxic strains (e.g., Bacillus anthracis ), endospores may cause serious environmental and health problems. A n eed therefore exists for development of improved methods for rapid inactivat ion of bacterial endospores. One of the ideas proposed for spore inact ivation is the use of photocatalytic nanoparticles to kill spores. A key technica l question is how eff ective photocatalytic powders are against endospores in the dry state. Another important question is the effectiveness of photocatalysis at low light intensities, such as those present in the indoor environment. As noted in Chapter 3, over one hundred studies have been conducted on the use of TiO2 in photocatalytic inactiv ation of microorganisms and over forty different microbes have been assessed with regard to their susceptibility to this method of disinfection. However, few studies (ten papers) have explored the ability of TiO2 photocatalysis to inactivate re sistant microbial forms such as fungal spores, protozoan cysts or oocysts, or bacterial endospores. Furthermore, only 3 studies have been carried

PAGE 20

2 out on the photocatalytic effects of titania powder against spores in dry (room air) state. This information is very important for the pr actical application of photocatalytic powders against biohazardous agents. Therefore, the goal of the present dissertation was to determine the effectiveness of titania against resistant microb es (bacterial endospores) in dry state in the presence of so lar UV radiation. Specific objectives of the research were to Demonstrate the photocatalytic activity of titania nanoparticles against bacterial endospores in dry state under UVA illumination. Quantify the effect of light intensity on photolytic and photocat alytic inactivation of endospores in dry state under solar UV irradiation. Bacillus cereus was used as a surrogate for Bacillus anthracis in the biological testing due to its close rela tionship to the highly toxic B. anthracis in term of genome sequence. Spores were produced from vege tative cells, which are more susceptible to inactivation process than spores. The existen ce of vegetative cells in spore suspension is unavoidable. Hence, Bacillus cereus spore purification methods in the present research were compared since no evaluation and quantit ative information has been investigated about the effectiveness of these methods on se paration of spores a nd vegetative cells and inactivation results. Uniform distribution of spores and photo catalyst on the surface and enumeration of viable spores on the surface are the two key techniques to conduct photocatalytic inactivation against spores on su rface in dry state. Howeve r, no paper has quantitatively and qualitatively evaluated both spore distribu tions on surface and vi able spores sampling efficiency from surface after inactivation process. Approaches to obtain uniform distribution were explored and the distribution quality was quantified. Recovery percentage and sampling consistency were investigated for methods tested.

PAGE 21

3 In terms of health standard for indo or UVA application, a maximum allowable UVA for photocatalytic disinfection was 10 W/m2, thus, it is important to determine the relationship between photocatalyt ic disinfection rate and UVA in tensity. Results of this study will lead to a better unde rstanding of bacterial spor e inactivation on surfaces by photocatalytic powder in the dry state. Furt hermore, the effectiven ess of photocatalytic disinfection over a range of UV A intensity will be revealed.

PAGE 22

4 CHAPTER 2 LITERATURE REVIEW 2.1 Bacterial Endospores The unique structures of bacterial endos pores impart high resistance to harsh conditions such as heat, desicca tion, chemicals, UV radiation, and radiation, enabling spores to survive in the environment for years or decades. Although no metabolism can be detected in spores, they have sensory monitor mechanisms that allow them to germinate and outgrow into vegetative cells in response to the presen ce of nutrients. The spore forming bacteria Bacillus anthracis produce a potent toxin, causing the disease known as anthrax. Besides Bacillus anthracis , other spore-forming bacteria also cause problems. For example, spores of Bacillus cereus are often present in foods and animal feed and can cause food poisoni ng. Thus, it is important to find quick and effective ways to inactivate spores. In the following, th e structure and protective mechanisms of bacterial endospores are described and the sporulation and germination processes are explained. 2.1.1 The Structure and Protective Mechan isms of Bacterial Endospores Figure 2-1 shows the structure of a typical b acterial spore. The germ cell is the part of a spore that can reproduce to form new vegetative cells. It consists of a core that is surrounded by a plasma membrane and a cell wall. Beyond the germ cell wall is the cortex and the inner and outer spore co ats. In some spores (e.g., those of Bacillus cereus ), an exosporium lies outside the outer spore coat.

PAGE 23

5 The inner and outer spore coats together make up approximately 50-volume percent of the spore. They consist mainly of prot eins. The outer spore coat contains alkaliresistant proteins and thus im parts resistance to bases. Th e inner spore coat has alkalisoluble proteins. These coats are the main sour ce of resistance to chemicals in spores (Block 2001). Inside the spore coats lie two layers: the co rtex and the thin primordial cell wall. Both are composed of pepti doglycan. Chemicals such as lysozyme and nitrous acid can attack the peptidoglycan. The cortex is degraded during spore germination. The primordial cell wall maintains the cellular integrity after germination and serves as a template for peptidoglycan biosynthesis during outgrowth (Atrih and Foster 2002). Between the cell wall and the core lies the plasma membrane. It has a compressed polycrystalline structure. During germination, the plasma membrane is transformed into the membrane of the vegetative cell. The me mbrane of the vegetative cell is less viscous than the plasma membrane of the spore. Ch ange of the inner membrane will affect the germination properties (Atrih and Foster 2002). The inactiv ation of spores by chlorine dioxide is caused by the altera tion of spore inner membrane, so that when treated spores begin to germinate, they die quickly due to the damage of the membrane (Setlow et al. 2002). DNA, RNA, dipicolinic aci d (DPA) and most of the calcium, potassium, magnesium, manganese and phosphorus in the spor es are located in the spore core. The proteins necessary for spore germination are al so present in this region. These small acid soluble proteins (SASPs) exist as / types that are associated with DNA and types that are not associated with any macromolecules. The SASPs are quickly degraded during

PAGE 24

6 germination. They are the most important fact or in spore resistance to UV irradiation and some other biocidal agents, such as hypoc hlorite, hydrogen peroxide, and freeze drying (Sabli et al. 1996). Setlow’s (1994) experi ment supported the hypothesis that saturation of DNA with / SASPs increases its resistance to attack by hydroxyl radicals produced by hydrogen peroxide. This is because the sporic idal action of radicals is considered to be the cleavage of the DNA backbone within th e spore core. The dramatic structural change of spore DNA caused by saturation with / SASPs substantially decreases the reactivity of DNA with chemi cals (e.g., hydrogen peroxide). Different parts of the spore confer resi stance to different physical or chemical agents, as summarized in the following. The spore coats act as a permeability barrier, preventing access of peptidogl ycan-lytic enzymes (e.g., lyso zyme) as well as other chemicals into the spore cortex. However, th e resistance of the spore coats of different species varies. Waites et al. ( 1976) found that the spore coats of Bacillus cereus are much less effective than those of Clostridium bifermentans in acting as a barrier to peroxide. It was found when the protein of the spore coat is removed by chlorine, lysozyme will initiate germination due to its attack on the cortical peptidoglycan (Waites et al. 1976). Riesenman and Nicholson (2000) found that a Bacillus subtilis mutant lacking the spore coats was very sensitive to lysozyme but possessed normal resistance to heat and 254-nm UV light. The cortex is associ ated with resistance to heat (Atrih et al. 1996). Bloomfield and Arthur (1994) found th at the cortex plays a role in spore resistance to chlorine-releasing ag ents and iodine-releasing agents. The spore core has relatively low permeability to hydrophilic molecules (Gerhardt 1972). Dehydration of the spore core occurs dur ing sporulation. Dehydr ation of the spore

PAGE 25

7 core is accompanied by accumulation of minera ls in the core and is contingent on development of the cortex. How the cort ex controls core water content is unknown (Gerhardt and Marquis 1989). Both minera ls accumulated in the spore core and dehydration of spore core are related to s pores’ heat resistance (Bender and Marquis 1985; Marquis and Bender 1985; Slepecky a nd Foster 1959). Increased spore core mineralization is also associated with in creased spore resistance to oxidizing agents (Marquis and Shin 1994). Much of the spore’s pool of divalent cations is likely chelated w ith dipicolinic acid (DPA). However, the role of DPA in spore resistance is unclear. Lack of DPA increases spore core water content and decreases re sistance to heat and hydrogen peroxide (Paidhungat et al. 2000). It wa s found that spores lacking DPA show no decrease or even have increased resistance to UV, which is consistent with th e previous research indicating DPA is a photosensitizer in spores (Setlow and Setlow 1995). Repair of damaged macromolecules is anot her factor contributing to the resistance of spores. Nicholson et al. (2000) found that at least one protein uniquely present in spores could repair, during germination and outgrowth, damage to DNA caused by UV irradiation. 2.1.2 Sporulation The process of bacterial sporulation consists of a series of stages that end with the release of free endospores from cells (Harw ood and Cutting 1990). It can be induced by exhaustion of carbon, nitrogen, or phosphorus sour ces in the environment. The stages of sporulation may be classified by the morphol ogical features of the vegetative cell and developing spore (Fig. 2-2). Stage 0 is a normally growing cell. In Stage I, an asymmetrically sited division septum is formed, creating two unequally sized cell

PAGE 26

8 compartments. Once the septum is formed, spor ulation is irreversible . The larger of the two compartments is referred to as the mother cell or sporangium, whereas the smaller is called the forespore or prespore. In Stag e II, the septum separating the forespore compartment is degraded and the forespore becomes free within the mother cell. Two layers of cytoplasmic membrane with opposite polarity enclos e the free forespore. They are called inner forespore membrane (IFSM ) and outer forespore membrane (OFSM) respectively. During stage III, the germ cell wall and cortex are formed between the two membranes of the forespore. Spore coat pr oteins are deposited on the outside of spores during stage IV. The mother cell lyses and th e mature spore are released during stages V and VI. The unique properties of spores, such as resistance to heat, chemicals, radiation, and mechanical forces, are de veloped during sporulation. 2.1.3 Germination The process of germination is divided in to three stages: ac tivation, germination (initiation) and outgrowth. Activation agents include heat, acidity, sulfhydryl compounds, and pressure (Brooks et al. 1991). Their function is to damage the spore coat. Without activation, spores cannot germ inate or germinate very slowly, even though favorable conditions for germination exist. Once activation occurs, the spore undergoe s germination (initiation), usually a rapid process (on the order of several minutes). The prerequi site for the germination is water and a germination agent. The role of germination agents is to penetrate into the damaged spore coat and trigger the sequence germinant events. They involve the activation of the spore-lytic enzyme, which degrades cortex peptidoglycan and SASP, loss of microscopic refractility of spore, loss of calcium dipicolinate, loss of resistance to heat and chemical, and initiation of core me tabolism (Madigan et al. 2000). Germination

PAGE 27

9 agents can be nutrients or non-nutrients (chemi cal, enzyme, pressure) (Pol et al. 2001). For example, L-analine is the most common nutrient germinant. Dodecylamine is nonnutrient germinant. Germinant agents differ with different species of bacteria. For example, L-alanine triggers germination of Bacillus anthracis , whereas Inosine triggers the germination of Bacillus cereus . Germination (initiation) leads to immedi ate outgrowth when supplied with all nutrients necessary for cell growth. If the suppl y of nutrients becomes depleted during the process of germination, the spore will dehydr ate and formation of a vegetative cell will cease (Davis et al. 1990). During outgrowth, spores swell due to water uptake and new synthesized RNA, proteins and DNA, and a ve getative cell that is composed of spore protoplast with cell wall emer ges from broken spore coat. 2.2 Semiconductor Photocatalysis Several excellent reviews have appeared in the past few years on semiconductor photocatalysis. For example, Hoffmann et al. (1995) reviewed environmental applications of semiconductor photocatalysis, whereas Linseb igler et al. (1995) examined the principles and mechanisms of photocatalysis on TiO2 surfaces. The following discussion draws heavily from these two reviews. Semiconductors can act as photocatalysts by accelerating pro duction of reactive oxygen species, such as hydroxyl radicals (OH ), upon exposure to light. Such materials include ZnO, TiO2, Fe2O3, CdS, and ZnS. Titania (TiO2) has received most attention as a photocatalyst partly because of its low cost, stability in the environment, and lack of toxicity at the micro and macro size ranges.

PAGE 28

10 Electrons from the valence band of a semi conductor can be promoted by excitation into the conduction band, leaving a positive hole in the valence band (Fig. 2-3). The excitation can take place when the semiconducto r absorbs light of sufficient energy. The energetic difference between the two bands is the band gap. For example, the band gap of titania is 3.2 eV and, thus, light of wave length 380 nm or less has sufficient energy to cause excitation. The conductio n-band electron can recombine with the valence-band hole, releasing heat energy. Alternately, th e conduction-band electron and valence-band hole can migrate to the surface of the semi conductor, where they can undergo direct reaction with the electron accep tor and electron donor, respectively. Another possibility is that the conduction-band electron is trappe d in a dangling surfic ial bond to yield Ti (III) and the valence-band hole is tra pped by a titanol group on hydrated TiO2 surface. If the ultimate electron donors are undesirable chemicals or microbes, the end result of photocatalysis is beneficial pur ification of the environment. Hoffmann et al. (1995) proposed the general mechanism of TiO2 photocatalysis, with characteristic times as follows: Primary process Characteristic time Charge-carrier generation cb vbe h h TiO2 (10-15 s) (2-1) Charge-carrier trapping IVIV vbhTiOHTiOH fast (10-8 s) (2-2) IVIII cbeTiOHTiOH shallow trap (10-10 s) (2-3) (dynamic equilibrium) III IV cbTi Ti e deep trap (10-8 s) (2-4) (irreversible) Charge-carrier recombination IVIV cbeTiOHTiOH slow (10-7 s) (2-5)

PAGE 29

11 IIIIV vbhTiOHTiOH fast (10-8 s) (2-6) Interfacial charge transfer IVIVTiOHRedTiOHRed slow (10-7 s) (2-7) IV treOxTiOHOx very slow (10-3 s) (2-8) where >TiOH symbolizes the hydrated surface functionality of TiO2, ecb represents a conduction-band electron, etr represents a trapped conduction-band electron, hvb + represents a valence-band hole, Red represents a reduced electron donor, Red+ represents an oxidized electron donor, Ox repr esents an electron acceptor, Ox represents a reduced electron acceptor, IVTiOH represents a surface-trappe d valence band hole (i.e. surface-bound hydroxyl radical), and IIITiOH represents a surface-trapped conduction band electron. The above reactions do not include direct interaction of valence-band holes with reductant. However, Hoffman et al. (1995) noted that there is substantial evidence that oxidation can also occur directly via the valence-band hole before it is trapped. Two important processes decide the quant um yield (hole-reductant or electronoxidant transfers per photon abso rbed) of interfacial charge tr ansfer (Fig. 2-4). First is the competition between char ge-carrier recombination (p icoseconds; not shown in the figure) and trapping (nanoseconds). Second is the competition between trapped carrier recombination (microseconds) and interfacial charge transfer (milliseconds). Changes to the semiconductor or reaction environment that lengthen the recombination time scales or shorten the interfacial charge transfer time scales should incr ease quantum efficiency. 2.3 Photocatalytic Inactivation of Microbes Despite numerous recent academic review s of semiconductor photocatalysis for destruction of environmenta l pollutants (A ithal et al. 1993; Bahnemann 1993; Hoffmann

PAGE 30

12 et al. 1995; Linsebigler et al. 1995; Ollis et al. 1991; Pichat 1994) , only two considered inactivation of microbes (Bla ke et al. 1999; Mills and Le Hunte 1997), the former being primarily devoted to this topi c. These two reviews collec tively considered 32 papers on photocatalytic microbial inactiv ation appearing in the peer-reviewed literature. Since then, another 42 have been published. The pr esent review gives a comprehensive look at experience on photocatalytic in activation of microbes and provides a detailed discussion of the factors that aff ect inactivation rates. Titania has been studied fo r inactivation of microbes in systems in which either water or air was the continuous phase. In aqueous systems, the effects of oxygen, photocatalyst and microorganism concentrati on, pH, and nature of buffer have been investigated. In air systems, relative humidity and air veloc ity have been studied. Other variables include form of titania (particles vs. coating on surfa ces), phase (rutile, anatase), dopants (Al, Cu) and admixtures (FeSO47H20, Pt), light intensity, wavelength, and microbial species. Most of these have b een studied in systems where water is the continuous phase. The effects of these va riables on photocataly tic inactivation of microbes are summarized in Appendix A. 2.3.1 Systems with Water as Continuous Phase 2.3.1.1 Light and photocatalyst concentration effects Since light initiates the photocatalysis, it becomes important in the photocatalytic reaction. Without irradiation by UV light, TiO2 itself was nontoxic (Block et al. 1997; Dunlop et al. 2002; Jacoby et al. 1998; Mats unaga and Okochi 1995; Matsunaga et al. 1985; Nakagawa et al. 1997). Several author s (Bekbolet 1997; Dunlop et al. 2002; Horie et al. 1996; Koizumi and Taya 2002a; Lee et al. 1997; Matsunaga and Okochi 1995;

PAGE 31

13 Matsunaga et al. 1985; Rincon and Pulgarin 2003) found that the sp ecific inactivation rate was directly proportional to the light intensity. Increased, decreased, or no change in mi crobial inactivation rate was observed under intermittent irradiation compared to c ontinuous irradiation. Pham et al. (1995) found increased specific inactivation rate of Bacillus pumilus spores under intermittent illumination (30 min on following by alternating 15 min off and on periods) with titania compared to continuous illumination. The au thors suggested that reduced activity of TiO2 under continuous illumination was due to the build up of intermediate species (hydroxyl radicals). Laot et al. (1999) found that inactivation of MS2 bacteriophage in the presence of titania power was more eff ective under intermittent illumination (3 min on, 3 min off) and cited the hypothesis of Pham et al. (1995) as a possibl e explanation. In the same study, however, the inactivati on rate of another phage (host: Bacteriodies fragilis) was found to be similar under either continuous or intermittent illumination. Rincon and Pulgarin (2003) observed that specific inactivation rate of Escherichia coli in the presence of Degussa P25 titania was substantially lower under intermittent irradiation than continuous irradiation. The intermitte nt illumination pattern consisted of “on” periods of 2, 5, or 10 min, each followed by a 40 s “off” period. They attributed slower inactivation under intermittent illumination to cell self-repair mechanisms that take place under darkness. Potential for regrowth of bacteria after stopping illumination was tested in five studies. Wist et al. (2002) observed E. coli concentration under da rkened conditions in a suspension of titania that was previously e xposed to a black lamp. No regrowth of bacteria occurred in distilled water over a 24 h period, but a si gnificant increase of

PAGE 32

14 bacteria was observed in natural water. Rinc on and Pulgarin (2003) i rradiated a bacteriacontaminated TiO2 suspension with solar UV lamp. They observed that bacterial numbers continued decreasing for several minut es after turning off the lamp and that no bacterial regrowth occurred over a 3 h period. In comparison, the decline in numbers of bacteria exposed to the solar UV lamp wit hout titania present stopped immediately after the lamp was turned off. Furthermore, the ba cteria grew back to the initial concentration within 3 h in the dark. This result indicat es that damage caused by radicals and other oxidative species produced by titania continue s for a limited period after illumination is stopped and that bacteria are unable to repair the damage caused by the reactive chemical species. Rincon and Pulgarin (2004a) found th e extent of “residua l disinfecting effect” (inactivation continued after stopping illumination) depended on the light intensity applied in irradiation pe riod, but not the dose. Rincon and Pulgarin (2006) studied several advanced photocat alytic oxidation proces s such as UV-vis/TiO2, UV-vis/TiO2/ H2O2, no regrowth of bacteria af ter irradiation was observed w ithin a dark period of 24 h. Effective disinfection time (EDT24) (Time required for total inactivation of bacteria without regrowth within 24 h after ph ototreatment) was 40 min for UV-vis/TiO2 system, 20 min for UV-vis/TiO2/ H2O2 system. Melian et al. (2000) studied regrowth of coliform after photocatalytic in activation, bacterial CFU increment was not observed 2 h later but found 24 and 48 h later. Several investigators have found optimum concentrations for titania powders in aqueous systems, ranging from 0.0025 mg/mL to 2 mg/mL (Table 2-1). This is reasonable, since no photocatalysis can take pl ace in the absence of titania, but at some

PAGE 33

15 point, shading of microbes and adsorbed titania by additional titania in suspension could have a protective effect. The optimum UV wavelength for microbi al activation in the absence of photocatalyst has been found to be 250-260 nm (Block 2001) and, in fact, UV radiation in this wavelength band is referred to as “g ermicidal” UV. Jacoby et al. (1998) have explored the effect of UV wavelength on microbe inactivation in the presence of photocatalyst. They compared morphology of Escherichia coli after a 75 hour period in five different systems: (1) no UV, no photocat alyst, (2) 254 nm UV, no photocatalyst, (3) 356 nm UV, no photocatalyst, (4) 356 nm UV plus photocatalyst, and (5) 254 nm UV plus photocatalyst. UV at the 356 nm wavele ngth had no effect wit hout catalyst, whereas E. coli was decomposed partially under 254 nm UV irradiation without catalyst. In the presence of TiO2, E. coli was almost completely decomposed under either 356 nm or 254 nm UV irradiation. In comparison, ther e was no apparent deterioration of E. coli with no UV and no photocatalyst. Kersters et al. ( 1998) compared disinfection capability of 254 nm UV with 370 nm irradiated TiO2. Inactivation percentage of Aeromonas hydrophila in 254 nm UV treated water is more than 20 times photoactivated TiO2 treated water. Tests conducted outdoors showed same specific inactiv ation rate for water exposed to natural sunlight with and without TiO2. However, Dillert et al . (1998) observed decreased disinfection rate of E. coli with TiO2 compared with TiO2-free system when exposed to UV-C (<280 nm) irradiation. They attributed this effect to sha dowing of bacteria by catalyst particles.

PAGE 34

16 2.3.1.2 Oxygen Oxygen was found to be an important f actor in the rate of photocatalytic degradation of organic chemi cals (Gerischer and Heller 1991 ; Gerischer and Heller 1992; Wang et al. 1992). Oxygen acts as a scavenger of electrons; if the ra te of electron transfer to oxygen is too slow, electrons accumulate and the recomb ination of electrons with holes is favored (Hoffmann et al. 1995). In th e case of photocatalysis with ZnO, Hoffman et al. (1994) demonstrated that oxygen was the precursor to hydrogen peroxide. Wei (1994) showed substantial increase of bactericidal activity of irradiated TiO2 when the composition of gas flowing in the bacteria suspen sion was increased from 100% N2 to 100% O2. Killing efficiency increased with the increase of total oxygen percentage in the gas. At 100% N2, no bactericidal effect was detected. Li et al. (1996) observed enhanced destruction percentage of E. coli when DO concentration is in creased to 4-5 mg/L. With the further increase of DO, destruction percen tage reached maximum at 5 mg/L and kept constant. Sun et al. (2003) found that photomineralization rate of E. coli in a membrane photocatalytic oxidation reactor followed pseudo-first-order ki netics with respect to the dissolved oxygen concentrati on. Interestingly, th e reaction rate was maximized at a dissolved oxygen concentration of 21 mg/L and decreased so mewhat at a concentration of 25 mg/L. They hypothesized that extensive hydroxyla tion of the TiO2 at the highest dissolved oxygen concentration coul d have inhibited adsorption of E. coli. Cho et al. (2004) compared three different dissol ved oxygen conditions on photocatalytic inactivation of E. coli: air equilibrium; O2 saturation and N2 saturation. Results showed a higher inactivation rate under O2 saturation than under air e quilibrium. No inactivation was observed in the absence of oxygen (N2 saturation). They suggested that

PAGE 35

17 recombination of electrons and holes were favored when no electron scavenger (e.g., oxygen) was available, ther efore decreasing the producti on of hydroxyl radicals and reactive oxygen species. Liu and Yang (2003) compared photocatalytic inactivation rates with air and nitrogen purging. Higher bactericidal activity was observed using air as a purging gas. They considered that a more oxidizing environment led to higher generation of superoxide radical (O2 -) and hence higher bactericidal activity. Rincon and Pulgarin (2005) studied effect of oxygen in the disinfection pr ocess in a coaxia l photocatalytic reactor using titania as photocatalyst. A dissolved oxyge n concentration of higher than 8 mg/L can enhance bactericidal activity of phot ocatalyst. Sun et al . (2003) investigated mineralization of Escherichia coliform using TiO2-Fe2O3 membrane photocatalytic oxidation reactor. It was found the DO affect ed the removal efficiency. The ultimate removal efficiency was 99% at DO of 21.34 mg/L and bacterial concentration at 109 CFU/mL. It was proposed that increased DO would probably redu ce the electron-hole recombination rate, which in turn improved th e overall quantum-yield efficiency of the reactor. At a higher DO level than 21.34 mg/L , a decreased of reaction rate constant was deduced. They suggested that this phenomenon was attributed by the fa ct that the titania surface become highly hydroxylated to the extent of inhibiting the bacterial adsorption at the active sites for initiating or participating in reactions. 2.3.1.3 Microorganism load and species Pham et al. (1995) observed that specif ic inactivation rate increased as the microorganism load became greater. The authors explained this effect on the basis of the probability of interactions between hydroxyl radicals and spores. At higher spore concentrations, spores are more likely to collid e with radicals and thus to be inactivated.

PAGE 36

18 At lower concentrations, interaction probability is decreased, and the specific inactivation also decreased. Bekbolet (1997) found that photocatalytic inactiva tion rate decreased with increased initial cell concentration at a Degussa P25 titania load of 1 mg/mL. No explanation was given. Ri ncon and Pulgarin (2004a) observed that initial photocatalytic inactivation rate of E. coli increased with initia l bacterial concentrati on. Shiraishi et al. (1999) plotted initial cell inactiv ation rate vs. initia l cell concentration in the reactor, the relationship was expressed by a straight line for cell concentration lower than 8.8 10 9 CFU/mL. Ibanez et al. (2003) compared photocatalyt ic inactivation resi stance of several representative gram-negative bacteria (Escherichia coli, Pseudomonas aeruginosa, Salmonella typhimurium , Enterobacter cloacae). E. coloacae was found to be most resistant. Escherichia coli and S. typhimurium were slightly less resistant than E. cloacae, whereas P. aeruginosa was much less resistant. Ku hn et al. (2003) tested the resistance of one species of yeast (Candida albicans) and several bacterial species to the photocatalytic activity of TiO2. They found the order of resistance to be: Candiada albicans > Enterococcus faecium > Staphylococcus aureus > Pseudomonas aeruginosa > Escherichia coli. Butterfield et al. (1 997) tested spores of Clostridium perfringens and Escherichia coli. A 3-log10 reduction of E. coli was achieved within 25 min., compared to a 2-log10 reduction of Clostridium perfringens under the same conditions. Rincon and Pulgarin (2005) found the sensitivity of bacteria to photocatalytic treatment was different for each specific group of the microbial community in wastewater. E. coli was the most sensitive species, Enterococcus sp. was the second, all colif orms excluding E. coli was the third and total Gram-negative sp. (other than coliform) was the most resistant to

PAGE 37

19 photocatalytic inactivation. Lonnen et al. (2005) tested inactivation ability of solar and photocatalytic disinfection batchprocess reactor on protozoan (Acanthamoeba polyphaga), fungi (Candida albicans, Fusarium solani) and bacterial microbes (Pseudomonas aeruginosa, Escherichia coli). Both reactors achieved at least a 4 log reduction of protozoa, fungi and bacteria. But only 1.7 log reduction was achieved for spores of B. subtilis and no effect on cyst stage of A. polyphaga. Seven et al. (2004) compar ed photocatalytic disinfec tion efficiency against different microbes: Escherichia coli, Pseudomonas aeruginosa, Staphylococcus aureus, Saccharomyces cerevisiae, Candida albicans, Aspergillius niger using TiO2 and ZnO. It was found all these selected b acteria and fungi were inactiv ated in a short period under sodium lamp in the presence of TiO2 or ZnO. The three stains of bacteria were inactivated faster than the two strains of fungi . No inactivation of A. niger by photocatalytic activity was observed. Sun et al. (2003) found initial photocatalytic inactivation ra te increased 2.4 fold for Escherichia coliform from low concentration of 4.9 108 CFU/mL to high concentration of 1.38 109 CFU/mL. It was explained that an increase of probability for surface interaction at higher bacter ial concentration would increase the relative adsorption availability on photocatalyst TiO2-Fe2O3 surface. Otaki et al. (2000) tested di sinfection of three microbes (E. coli, bacteriophage Q , Cryptosporidium parvum) in a immobilized titania reactor where germicidal UV lamp or black light lamp were used. It was observ ed titiania did not enhance inactivation of E. coli, but promoted Cryptosporidium parvum inactivation under both germicidal UV and black light irradiation. For bacteriophage Q , improved photocatalys is efficiency was

PAGE 38

20 observed under black light irradi ation rather than germicidal UV irradiation. Therefore, it was considered that titania can improve inac tivation of relatively irradiation-resistant microorganisms, especially Cryptosporidium parvum. Rincon and Pulgarin (2004a) observed that Enterococcus sp. was less sensitive than coliforms and other Gram-negative bacteria in wastewater sample to photocatalytic inactivation. Melian et al. (2000) compared photocatalytic disi nfection on two microbial groups: total coliforms and Streptoccocus faecalis. Deactivation of both microbes was similar under UV irradiation in the pr esence or absence of titania. 2.3.1.4 Hydrogen ion activity and solutes Hydrogen ion activity can infl uence electrostatic forces between microbes, as well as aqueous phase chemistry. Block et al. ( 1997) used sulfuric aci d or sodium hydroxide to adjust initial pH of titan ia in distilled water to a ra nge of 3.5.6. Results showed similar inactivation of Serratia marcescens by TiO2 under solar UV illumination at initial pH of 3.5, 5.5 and 6.9. At initial pH of 8.6, th e inactivation rate was decreased. Kikuchi et al. (1997) observed a higher ba cterial survival ratio at in itial pH of 7.4 compared to initial pH of 4. They hypothesized that slower inactivation at the hi gher pH was due to slower production of hydrogen peroxide. The significance of proton availability on hydrogen peroxide formation can be seen from the reactions below: + -+ 2 ++H O,H +e-• 2222 -HOOHOOHO (2-9) Koizumi and Taya (2002b) investigated th e disinfection rate of phage MS2 in the initial pH range of 3 (adjus ted by HCl or NaOH). Specific inactivation rate of MS2 showed a convex feature. At pH 6, maximu m specific inactivation rate was observed. Coincidentally, maximum adsorption of MS2 to titania was found at this same pH. Watts

PAGE 39

21 et al. (1995) observed that the pho tocatalytic inactivation rate of coliform bacteria was unchanged at initial pH between 5 and 8. Cho et al. (2004) adjusted the initial pH of phosphate-buffered titania suspensions to values of 5.6, 7.1, and 8.2. No difference in the rate of photocatalytic inactivation of E. coli was found over this range of pH. The authors suggested that electrostatic interac tions between microbe su rfaces and titania was insignificant in determining photocatalyt ic activity. Rincon and Pulgarin (2004b) modified initial pH between 4.0 to 9.0, inactivation rate was not affected in the absence or presence of titania. This phenomenon was lin ked to the sensibility of bacteria to low or high pH environments and titania agglomer ation state, which is a function of pH. Acid-adapted cells can increase tolerance towards osmotic stress was given as a supplementary reason to the no influence of initial pH on E. coli photocatalytic inactivation. They also found that addition of H2O2 enhanced photocatal ytic inactivation because H2O2 presented as additional electron acceptor accelerate photocatalytic inactivation rate. Gumy et al. (2006) found inactivation of E. coli by Degussa P25 titania under UV irradiation was not affected by initia l pH (4.5.5) of the suspension. This is explained by the observation th at the initial pH of solution at 6.0 and 8.5 dropped 5.2 and 5.9 respectively as photocatalytic reaction star ted. Thus, the titania surface becomes positively charged and E. coli was negative charged, the el ectrostatic attraction was favored and inactivation rate was not affected by the initial high pH value. By comparing different commercialized titania powder, a relationship betw een isoelectric point (IEP) and inactivation activity was investigated. The lower the IEP of titania, the lower the bactericidal activity, which can be explained by electrostatic attract ion between cells and photocatalyst.

PAGE 40

22 Block et al. (1997) reported that the presence of i norganic solutes (phosphate, calcium, magnesium, sodium, potassium) and organic solutes (amino acids) in water decreased the bactericidal activ ity of titania in the presen ce of solar UV illumination. The authors implied that competition between the inorganic salts and microbes for titania adsorption sites could be responsible for this effect, whereas photoreaction of titania with organic compounds could explain th e decrease in bactericidal effect in the presence of amino acids. Koizumi and Taya (2002b) invest igated effect of vari ous inorganic ions on photocatalytic inactivation rate of phage MS2 in titania su spension. It was found that coexistence of NO3 -, SO4 2-, PO4 3-, K+ each at 10 mM decreased the rate constant for phage inactivation, whereas Cl-, Bror Na+ did not decrease rate constant. Ions were considered to block adsorption of phage on titania surface, which decreased photocatalytic activity in microbi al disinfection. This inhibito ry effect can be explained from the direct proportionality between inac tivation rate constants and quantities of the phage particles adsorbed on tit ania surface irrespective the kinds of ionic species. Rincon and Pulgarin (2004b) stud ied inorganic ions (HCO3 -, HPO4 2-, Cl-, NO3 -, SO4 2-) and organic substance (dihydroxybenzenes) natu rally present in water on photocatalytic disinfection. Addition of HCO3 and HPO4 2decreased photocatalytic inactivation ability while Cl-, NO3 -, SO4 2has a week effect. These speci es prefer to retard bacterial inactivation by competing for oxidizing radicals or by blocking active site of titania. Dihydroxybenzenes had negative effect on photo catalytic disinfection due to competing for oxidation site on titania. Rincon and Pulgarin (2006) fount that natural water promoted photocatalytic inactivation of E. coli than deionized water due to chemical matrix of water. Photosensitizers presented in natural water excited the substances that

PAGE 41

23 reacted with oxygen to form reactive oxygen species, which promoted photocatalytic inactivation. Addition of H2O2 in UV-vis/TiO2 system was observed to enhance photocatalytic disinfection since H2O2 is a more efficient electron acceptor than oxygen, which prevents the elec tron-hole recombination. 2.3.1.5 Microstructure, size, and immobilization Important crystalline forms of titania are an atase, brookite, and rutile. Li et al. (2003) determined that anatase has s uperior photocatalytic activity based on decomposition of methylene blue. Ra na et al. (2005) synthesized TiO2-NiFe2O4 composite nanoparticles and compared photocat alytic inactivation ability of anatasetitania-coated nickel ferrite with brookite-titania-coa ted nickel ferrite against E. coli. They found the bacterial concentration decrease d more in anatase-t itania-coated nickel ferrite system than in brookite-titania-coated nickel ferrite system under UV irradiation. Sato and Taya (2006) investigated effect of crystalline structure of titania on bactericidal activity. Mixture of different ratio of anataseand rutile-t ype titania particles was used against Phage MS-2. Maximum specific inac tivation rate was observed at anatase to rutile ratio of 70 wt%. This was caused by an increase in gene ration of reactive oxygen species and quantity of phage adsorbed on titani a. It was also found that close contact of both type of titania, which enabled excha nge of photo-excited electrons and holes between the particles with the balanced acceleration of oxidation and reduction rate, enhanced quantum yield and increased generation of reac tive oxygen species. Trapalis et al. (2003) measured bacterial destruction rates in the presence of thin TiO2 films and concluded that an increase of anatase crystalline structure improved bactericidal effect. Rincon and Pulgarin ( 2003) compared three different forms of TiO2 (rutile, anatase, Degussa P-25) of similar size (approxi mately 21 nm) immobilized on

PAGE 42

24 glass. The P-25, which has an approximate composition of 80% anatase and 20% rutile, had the highest specific inactivation rate for E. coli. Dunlop et al. (2002) compared Degussa P25 to Aldrich titania powder of sim ilar size that consisted of 99.9% anatase. Both powders were coated on electrodes. The P25 was found to have the higher inactivation rate for E. coli. Hoffman et al. (1995) show ed that P25 has one of the highest quantum efficiencies of commercial TiO2 samples tested. They attributed the high efficiency to slow recombination of electrons and holes. Bickley et al. (1991) suggested that the anatase/ru tile structure of P25 promotes charge-pair separation and inhibits recombination. Size and microstructure have been identifie d as important characteristics of titania with regard to photocataly tic activity on the basis of physicochemical measurements (Anpo et al. 1987; van Grieken et al. 2002). However, the effect of size on antimicrobial activity of titania has not been systematica lly studied. Nakagawa et al. (1997) observed that ultra-fine particles (Degussa P 25, average size 21 nm) was the most potent photogenotoxic in vitro genotoxicity assays wh en compared with another three kinds of titania with higher part icle size (anatase, average size 255 nm; rutile, average size 255 nm and rutile, average size 420 nm). But no explan ation was presented. Allen et al. (2005) found the nano anatase (5-10 nm) showed so mewhat greater antibacterial activity on E. coli than normal anatase titania (0.24 m). However, no comment was given for this phenomenon. Immobilized TiO2 was introduced to overcome th e difficulty of separating suspended TiO2 from water. A thorough review about th e supporting material and coating methods for supported titania as a photocatalyst in water decontamination was

PAGE 43

25 reported by Pozzo et al. (1997). Yu et al. (2002) compared photocatalytic activity of transparent anatase titania th in film on glass prepared by reverse micelle method and solgel method. Both films showed signif icant bactericidal activity against E. coli. Titania film prepared by reverse micelle method is better than the film prepared by sol-gel method in gaseous phase oxidation but same in aqueous phase, this is due to the large cell size which is hard to diffuse into porous struct ure of reverse micelle prepared titania film, thus, do not favor degradation of large species. Mc Loughlin et al. (2004b) coated titania on Pyrex rod fixed within the reactor. Very slight improvement of E. coli inactivation was observed by photocatalyst compared to the inactivation under sunlig ht alone. It was explained that the titania loading was not high enough to promote much photocatalytic disinfection with respect to a domin ant solar UV disinfection mechanism. Duffy et al. (2004) coated titania powder on a flexible pl astic sheet and glass bottle side. It was shown the borosilicate glass and PET plastic batch-process solar di sinfection reactor (SODIS) fitted with the TiO2 coated plastic sheet demonstrated 20% and 25% more effective than standard SO-DIS reactor in E. coli inactivation. However, titania coated on glass bottle required 20% more time for complete inactivation than solar UV alone. No explanation was given for this low effici ency. Coronado et al . (2005) imbedded TiO2 into SiO2 fibers and applied them in continuous -operation photoreactors for bacterial inactivation. TiO2/SiO2 fibers showed high inactivati on rate than under UV alone. Kim and Lee (2005) coated titania on Pyrex hollow glass beads which is used for photocatalytic inactivation of selected algae, Anabaena, Melosira and Microcystis. Complete photocataly tic inactivation of Anabaena Microcystis and was achieved in 30 min, whereas photocatalytic in activation efficiency for Melosira was lower than the other

PAGE 44

26 two. It was believed that the presence of the inorganic sili ceous wall around the Melosira cell contributed their resistance to photocatalytic disinfection. Fernandez et al. (2005) coated Degussa P25 powder on glass fiber pape r using an inorganic binder and fixed the paper in the photoreactor. It was found that immobilized TiO2 has lower efficiency than slurry TiO2 for bacterial inactivation. However, no explanation was given. Horie et al. (1998a) immobilized titania on activated charcoal granules (T/AC). The sterilization rate was found to be dependent on the temperature, which affecting amount of adsorbed cells on T/AC. The specific inactiva tion rate constant was correla ted with T/AC concentration which related to adsorption of cells to T/AC granules. Comparative studies indicate that the antimicrobial activity of TiO2 in films is substantially less than that of TiO2 powders (Rincon and Pulgarin 2003; Salih 2002). One explanation for this effect is the decreased interaction between TiO2 and microbes because of the localization of titania on the film surface (Salih 2002). Rincon and Pulgarin (2003) attributed lower E. coli destruction by immobilized TiO2 compared to suspended TiO2 to several factors, including less acce ss of catalyst surface to light and microbes, enhanced electr on hole recombination by TiO2 support, limited oxygen diffusion into TiO2 catalyst, less exposure to chiral and friction forces that could avoid deactivation of TiO2, increase of mean distance between bacteria and TiO2, and inability of immobilized TiO2 to penetrate into bacteria. 2.3.1.6 Dopants, admixtures, and soluble iron Dopants have been used to modify the mi crostructure of titania and improve its photocatalytic activity. Trapal is et al. (2003) achieved a higher inactivation rate of E. coli using a Fe3+ doped titania coating, in comparis on to undoped coating. However, Amezaga et al. (2002) found that undoped titani a films caused higher growth inhibition

PAGE 45

27 of Pseudomonas aeruginosa than copper or aluminum doped titania films. Sokmen et al. (2001) compared photocatal ytic disinfection of E. coli by titania and silver-loaded titania particles. It was found the s ilver loading dramatically de creased irradiation period for complete inactivation. Silver loading possibly increased reaction by trapping the conduction band electrons besides reduc ing the band gap energy of titania. Admixtures of titania and other powders have been evaluated in terms of photocatalytic activity. Mats unaga et al. (1985) ground tita nia (P25, Japan Aerosil) and platinum powder together to form a 10:1 (by weight) admixture. The admixture was more effective than TiO2 powder alone for inactivation of Saccharomyces cerevisiae. Sjogren and Sierka (1994) found that addition of 2M ferrous sulfate to a titania suspension increased the inactivation of phage (MS2). The authors suggested that Fenton reaction enhancement could be responsible for the improved phage inactivation. Rincon and Pulgarin (2006) found that addition of Fe3+ ions in UV-vis/TiO2 system at titania concentration of 0.5 g/L did not affect photocat alytic bactericidal activity, but enhanced photocatalytic disinfection at a lower titania co ncentration. Fe3+ behaved as an electron scavenger, preventing the recombination of el ectron-hole pairs, thus favor photocatalytic inactivation. At a higher titania co ncentration, light absorbance of Fe3+ was screened by titania and thus no beneficial effect was elucidated. 2.3.1.7 Synthesized photocatalyst nanocomposite Several newly invented nanostructured photocatalysts, where titania was coated on the nano structure or pure titania nanotube was produced, has been introduced, their efficiency on microbial inactivation was i nvestigated. Sun et al. (2003) used TiO2-Fe2O3 nanocomposite for photocatalytic inactivation of E. coliform in photocatalytic reactor.

PAGE 46

28 99% removal efficiency was achieved at DO of 21.34 mg/L and bacterial concentration at 109 CFU/mL in the reactor. No comparison of TiO2-Fe2O3 nanocomposite with other photocatalyst was made. Lee et al. (2005) compared TiO2-multiwallnanotube (MWCT) nanocomposites synthesized by wet chemistry and heat treatment with Degussa P-25 on photocatalytic inactivation of B. cer eus spores under solar UV. TiO2-MWCT doubled inactivation efficiency than Degussa P25, which show no significant increase of disinfection activity than unde r UV alone. Rana and Mi sra (2005) synthesized TiO2NiFe2O4 composite nanoparticles by revers e micelle method. They found TiO2-NiFe2O4 composite nanoparticles was more efficient in inactivation of E. coli in the present of UV than UV irradiation alone. Rana (2005) compared bactericidal activity of anatasetitania-coated nickel ferrite with brookite-tit ania-coated nickel ferrite, the previous shows higher inactivation activity than the latte r. Joo et al. ( 2005) synthesized TiO2 nanorods, which demonstrated higher inactivation rate than Degussa P-25 as photocatalyst against E. coli, which is contributed by the high surface area, large amount of hydroxyl radicals and larger band gap of TiO2 nanorods. 2.1.3.8 Physiological state and generation of microbes Rincon and Pulgarin (2004a) found phot ocatalytic inactivation rate of E. coli was affected by physiological state and generation. Bacteria collected at stationary phase were more resistant to photo catalytic inactivation than at exponential growth phase. This is due to the synthesis of a set of proteins, conferring E. coli cells resistance to several stress conditions, at stationaryphase stage. Bacteria harves ted at the third generation of culture were resistant than seventh genera tion to irradiation. This was explained by mutation when successive cultures are made, these mutations were believed to make bacteria more sensitive to photo catalytic attack by their study.

PAGE 47

29 2.3.2. Systems with Air as the Continuous Phase Photocatalytic inactivation of microbes in systems with air as the continuous phase has received much less atten tion than the systems with wa ter as the continuous phase. This relative lack of attention can be ga ged by the number of peer reviewed journal articles in the two areas, 103 and 12, respectively. Conseque ntly, a comprehensive list of the parameters important to the action of photocatalysts in systems with air as a continuous phase has not yet been assembled. However, we can pr esent a partial list based on the research to date. Since formation of aqueous chemical sp ecies such as hydroxyl radicals and hydrogen peroxide through photocatalysis is believed responsible for degradation of chemicals and microbes, it would seem that humidity would be an important variable when photocatalysis is carried out with ai r as the continuous phase. Goswami (1997) found that 50% relative humidity was better than either 30% or 85% for inactivation of Serratia marcescens on TiO2-coated fibre glass air filter under 350 nm UV illumination. The lower relative humidity was thought to li mit availability of water for transfer of hydroxyl free radicals to the ba cteria. Possible explanations for the poorer performance at 85% relative humidity include reactiva tion of bacteria (Riley and Kaufman 1972), hindrance of photoreaction (Dibble and Raupp 1990), or competition between water molecules and other molecules for TiO2 active sites. Wolfrum et al. (2002) applied bacteria to titania covered disk s and measured the amounts of CO2 evolved when the disks was exposed to UV irradiation. They found faster mineralization of several microbes (Escherichia coli, Micrococcus luteus, Bacillus subtilis, Aspergillus niger spores, Bacillus subtilis spores) at 50% relative humidity than at 0% relative humidity. The greatest effect was observed for Bacillus subtilis spores.

PAGE 48

30 Goswami (1997) also investig ated the effect of air velo city on disinfection rate obtained using the titania coated fibreglass air filter. They observed higher disinfection rate at lower air velocities. The author suggested that increa sed retention time of bacteria (on the filter) at lower air velocity contri buted to the higher di sinfection rate. Sato et al. (2003) investigated photocatal ytic deactivation of airborne microbial cells on titania-loaded plate. The titania solution was spread on the plates first and illuminated with the lamps. The cells suspension was sprayed on the surface by nebulizer. A model was established which is in fair agreement with experimental data of E. coli deactivation. Lin and Li (2003b) studied inactivation of B. subtilis spores on photocatalytic surfaces in air. They compared inactiva tion rate with and without titania under UVA irradiation. It was observed that survival fractions of spores on the slide with and without titania under black light irradiation were not statistically significant different. However, no explanation was given. Existence of titani a did not show enhanced disinfection rate under UVA compared with under UVA alone w ithout photocatalyst, this was also observed by Melian et al. (2000) and Lee et al. (2005) in the water system. Thus, a controversy exists whether titania contribu tes inactivation of mi croorganisms on surface under solar UV irradiation since an improve ment of disinfection by contacting with titania under solar UV irradiation was indicate d in other researches. It was found that different light intensities were used in thes e studies; this may explain why contribution by photocatalysis is different in these studies. Vohra (2005) studied enhan cement of photocatalytic in activation on the metal and fabric surface by coating silv er-doped titanium dioxide on. B. cereus spores were dried

PAGE 49

31 on the surface. A complete inactivation of B. cereus spores were observed on the surface with enhanced photocatalyst, whereas inactivation rate of B. cereus spores on surface with titania was lower. Deta ils of the approach for coati ng enhanced photocatalyst and titania on the surface was not described. Surface sampling efficiency was not investigated. Effect of vari ables, such as light intensit y, titania loading, etc on spore inactivation was not studied. Kinetics and modeling of photocatalytic disinfection was not presented. Pal et al. (2005) studied inac tivation of vegetative cells (E. coli, B. subtilis, Microbacterium sp.) on membrane surface and in continuous flow reactor. On the membrane surface, 50 mL titania solution was fi ltered onto the cellulose acetate filter membrane followed by filtration of 50 mL vegetative cells suspension and then exposed to UV irradiation immediately. An irradi ation period ranged from 0 – 20 min. Thus, vegetative cells may still in wet state inst ead of dry state. Distribution quality of vegetative cells was not mentioned. Agar contact method was used for the surface sampling, which limited the applied initial amount of spores on the membrane. Efficiency and consistency of this sampling method for viable microbes on the membrane was not studied. For more detail regardi ng microbial distribution and surface sampling method, refer to Chapter 5 and 6. Al though effect of titania loading (mg/m2) and UV light intensity on microbial inac tivation was studied, influence of relative titania to spore ratio on bacterial inactivation was not inve stigated, which maybe the key factors on surface disinfection by photocatalyst. Profile of survival ratio of bact eria with the time in the absence of UV light was not reported, thus, decrea se of viable vegetative cells may be caused by a loss in microbial viability duri ng this dark period si nce it has been found

PAGE 50

32 some species were not able to tolerate dry conditions after 1 h drying (Moore and Griffith 2002). Kinetics was given based on Chick’s law (1st order kinetics), but no model was proposed to relate inactivati on rate with light intensity and titania loading. In the continuous flow reactor, bacterial aeroso l was generated by bioaerosol nebulizing generator. A polyethersuolfone membrane di sc filter coated with titania by conventional dipping method was wrapped on a steel frame a nd inserted in the reactor in the annual space between UV lamps and the outer cylinder. Air at the outlet and inlet of the reactor was sampled by a single stage impactor. For both batch and continuous reactor, inactivation rate of microbes in creased in the present of titania compared with UV alone. Inactivation rate for E. coli are higher than Bacillus subtilis and Microbacterium sp. It was found that inactivation rate in continuous reactor are hi gher than on the membrane surface due to unaccounted loss of bacteria via adsorption and se ttling on the reactor walls in the flow system. 2.4 Mechanisms of Photocatalytic Inactivation of Microbes In TiO2 photocatalytic inactivation of microbes, the consensus of investigators is that active oxygen species in cluding superoxide anion (O2 ), perhydroxyl radicals (HOO), hydrogen peroxide (H2O2) and hydroxyl radicals (OH) are responsible for microbe destruction. However, the relative significance of these chemical species has been subject of debate. Matsunaga et al. (1985) c oncluded that direct oxid ation of bacteria by TiO2/Pt particles was a possible mechanism of cell in activation. They base d this conclusion on the observation that no bacteria were inactivat ed when the bacteria were separated from TiO2/Pt particles by a dialysis membrane. In addition, they proposed that toxic substance

PAGE 51

33 such as hydrogen peroxide and free radicals formed by photocatalytic activity were not responsible for microbial inactivation becau se they did not find the change of cell viability in the presence of catalase (albumin and cysteine). They attributed the killing action of the photocatalyst to the inhibiti on of respiration caused by decrease of Coenzyme A and formation of its dimer. Saito (1992) observed the rapid leakage of potassium ions and slow release of DNA and protein in streptococ ci exposed to UV (300nmnm) in the presence of TiO2, They proposed that the death of bacteria is caused by the reactive oxygen species that damage the cell membrane. Recently, the microorganism inactivation was attributed to peroxidation of lipid membrane, including memb rane disorder, with consequent loss of necessary cell functions and death (Dunlop et al. 2002; Mane ss et al. 1999; Wei et al. 1994). Huang (2000) found that in photocatal ytic disinfection, cell wall damage occurred first, followed by cytoplasmic membrane da mage, which led to direct intracellular damage. They found the smaller TiO2 particle could cause quic ker intracellular damage. They also observed that photocatal ytic bactericidal activity of TiO2 continued for some time after the UV illumination was terminated. Many studies simply attributed the killi ng effect to hydroxyl radicals without further investigation (Ali et al. 1999; Block et al. 1997; Butte rfield et al. 1997; Cooper et al. 1998; Lee et al. 1997; Pham et al. 1995). Kikuchi et al. (1997) did a thorough study on the role of oxygen species in photocatalytic bactericidal effect. They proposed hydrogen peroxide has a long-ra nge bactericidal e ffect, and that other oxygen species play a cooperative effect. This is based on their observation that increased pH decreased inactivation efficiency. According to equati on (2-9), increased pH causes a decrease of

PAGE 52

34 H2O2 production, which would correlate with the observed decrease of killing efficiency. This was further confirmed by an experiment in which E. coli was separated from titania by a 50 mm thick porous membrane, which w ould deactivate free radicals before they could reach the bacterial surface, but allo w diffusion of hydrogen peroxide into the bacterial suspension. The inac tivation efficiency observed in this configuration was similar to that obtained when E. coli and titania were mixed together. However, mannitol (a hydroxyl radical scavenger) in creased the bacteria survival ratio, which would indicate that OH also contribute to the inactivati on of microbes. Salih (2002) found photocatalytic activity on microbial inactiv ation was inhibited when hydroxyl radical scavengers (dimethyl sulphoxide and cysteamine) were applied. Therefore, he suggested that hydrogen peroxide and radicals are th e germicidal agents. Kuhn et al. (2003) observed same result using dimethyl sulfoxide as a scavenger. Cho et al. (2005) found that biocidal mode of reactive oxygen speci es depends on the microorganism involved. MS-2 phage was mainly inactivated by free hydroxyl radical in the solution bulk while E. coli is inactivated by both the free an d the surface-bound hydroxyl radicals. E. coli might also be inactivated by other re active oxygen species, such as O2 and H2O2. Jacoby (1998) proved that cells could be mineralized via photocatalysis by using 14C-labeled E. coli, which demonstrated that the car bon content of microbes is oxidized to form carbon dioxide with substantial clos ure of the carbon mass balance. Dillert (1998) observed the mineralization of E. coli by measuring the change of total organic carbon in bacteria suspension before and after treatment. Kashige (2001) proposed that steps in the inactivation of Lactobacillus casei phage PL-1 by black-light-catalytic TiO2 film are damage to the capsid protein first, followed

PAGE 53

35 by fragmentation of DNA, and eventually dest ruction of the virus. Active oxygen species produced by the irradiated TiO2 film are responsible for the inactivation. Sokmen et al. (2001) proposed that possi ble killing mechanism by photocatalyst is lipid peroxidation occurred which produced ma londialdehyde, it was further degraded to harmless products. Sunada et al. (2003) inve stigated inactivation of E. coli on Cu/TiO2 film under week UV illumination. Two steps inactivation m echanism was provided. The first is the partial decomposition of the outer membrane in the cell envelope by the photocatalytic process followed by permeation of copper ions in to the cytoplasmic membrane. The second step is the disorder of the membrane caused by copper ions resulting in loss of cell’s integrity. Sun et al. (2003) proposed th at photo mineralization of Escherichia coliform undergoes a combined UV photolysis br eakdown upon cell lysis and a dual-site Langmuir-Hinshelwood mechanism, which involve s the surface controlled adsorption of oxygen and cleavage of cells on electron-ri ch and positive vacant sites respectively. Coronado et al. (2005) e xplained inactivation of Legionella pneumophila by UV irradiated TiO2/SiO2 fibers was due the activation of bacteria secretion systems when contacting the anatase surface, which coul d facilitate the acc ess of photogenerated radicals to key components of the secreti ng system without disrupting outer membrane first. Thus, an irreversible damage in the secretion systems of bacteria occurred, seriously impact the cell viability. Gogniat et al. (2006) sugges ted adsorption of cells onto aggregated titania followed by loss of membrane integrity is crucial to bactericidal effect of titania photocatalysis.

PAGE 54

36 This is based on the observation that adsorption was positively related with reduction or loss of cell membrane integrity by analyzing ba ctericidal effect of illuminated titania in NaCl-KCl or sodium phosphate solution. Immediate cell adsorpti on on catalyst occurred in NaCl-KCl solution, where most E. coli was killed within minutes of illumination. Cell adsorption on catalyst in sodium phosphate solution was delayed, where a delayed inactivation was observed. 2.4.1 Diffusion of TiO2 or ROS into the Cell Montgomery (1985) proposed two importa nt properties of disinfectant that influence their biocidal abil ity: (1) oxidation or damage of cell wall and/or membrane results in the disintegration of cell. (2) Diffusion of disinfectant into the cell to interfere with protein synthesis, inhib it enzymes, and destruct intrac ellular components. Hydroxyl radical, considered one of the main products of photocatalysis of TiO2, has a high standard oxidation potential at 2.7V compared to ozone (2.07 V). However, results (Watts et al. 1995) showed ozone has significantly high in activation rate on coliform bacteria and poliovirus than TiO2 photocatalysis under the same contact time. Therefore, the diffusion of either TiO2 or free hydroxyl radicals into th e microorganism is important. This is consistent with the findings of Do rfman and Adams (1973) that the reaction rate of hydroxyl radical with most biological molecule s occurred at diffusion-controlled rates. Saito (1992) proposed the adsorption of TiO2 onto the bacterial cells might be necessary for photocatalytic bactericidal action of TiO2 because the photocatalytic reaction occurs on the surface of TiO2 and the low lifetime of reactive oxygen species. Koizumi’s (2002a) test shows the enhancemen t of the bactericidal effect of TiO2 by increase the adsorption of phage on TiO2 particles, i.e. increase the diffusion rate.

PAGE 55

37 2.4.2 Disinfection Action of TiO2 is a Combination Action Photocatalytic disinfection of TiO2 is a combination action of both direct action UV on microorganisms and indirect photocatalytic action of TiO2. UV-B (Eladhami et al. 1994) does not grea tly affect the cell respiration, RNA, DNA synthesis. Inhibition of protein synthe sis and killing of bacteria were observed under UV-B irradiation. Experiment results s uggested free radicals me diate the effect of UV-B on bacteria. Lethal UV-A dose (Fernandez and Pizarro 199 6) will cause direct inactivation of bacteria by the damage to the DNA. Sublethal dose of UV-A combined with TiO2 enhance the bactericidal efficiency of UVA itself. Photocatalytic action of TiO2 causes membrane oxidative by the production of ROS. They penetrate into the cell, oxide the protein, lipid, or nucleic acid resulting in i nhibition of respiration or growth of the microorganism. Maness (1999) concluded the death of E. coli under the photocatalysis of illuminated TiO2 is due to the promoted peroxidati on of the polyunsaturated phospholipid component of the lipid membrane followed by destruction of membra ne. As a result, critical functions, such as re spiratory activity dependent on th e integrity of the membrane were lost and the cell was killed. 2.5 Kinetics and Modeling of Photocatalytic Disinfection Several studies have evalua ted kinetics of photocatalytic inactivation of titania against microorganisms. Wei et al. (1994) demo nstrated that the initi al inactivation rate of cells is a first order reaction with respect to initial cell concentration iirkc (2-10)

PAGE 56

38 where ri is the initial rate of ce ll inactivation, k is the first order rate constant and ci is the initial cell concentrati on. Based on data from experime nt carried at different light intensity and titania co ncentrations, they found that the specific inactivation rate (k) was linearly related to the square root of the c oncentration of titania in the range of 0g/L and was also linearly relate d to the incident light in tensity in the range 180 E s-1 m-2. Horie et al. (1996) modeled the kinetics of cell inactivation by photocatalysis using a series-event model, where death of a cell results from a number of reactions 111 OXOXOXOX oiiin M MMMMMdeath (2-11) where Mi is cell concentration at each event leve l i and OX is oxidative radical. Each of the reactions in this model was assumed to obey second-order kinetics with respect to the concentrations of microbial cells and oxidative radicals ge nerated by photocatalysis of titania. / OXdNdtkCN (2-12) where N = number concentra tion of cells at time t, COX = concentration of oxidative radicals and k = sterilization rate constant . The concentration of oxidative radicals was assumed to be constant under quasi-steady state conditions. Ba sed on mass balance on number of viable cells and the concepts e xpressed in equations (2 -11) and (2-12), the following expression for number of viable cells as a function of time was derived ' 11 ' 00 00() exp() !i nn ti ii ttNN kt kt NNi (2-13) where Ni is cell concentration of Mi and k’ is the apparent st erilization rate constant. 'OXkkC (2-14)

PAGE 57

39 Experimental results for phot ocatalytic inactivation of E. coli showed that data were fit by a series reaction model with 8 steps. However, inactivation data for Saccharomyces cerevisiae demonstrated a single reacti on step, which fits single-hit multitarget model. ' 01{1exp()}L t tN kt N (2-15) where L is the number of organisms per clum p, k’ is the specific rate constant. The apparent sterilization rate consta nt for photocatalytic inactivation for E. coli and Saccharomyces cerevisiae was found to vary in linear fash ion with light intensity at a constant titania concentration, whereas the re lationship between steril ization rate constant and titania concentration was nonlinear. A model considering quantum efficiency of titania photocatalyst and adsorption of the phot ocatalyst to bacteria l cells was successful in fitting the collected experimental data. Pham et al. (1997) performed a quantitative analysis of photocat alytic inactivation based on probability theory. The model assu mes that hydroxyl radicals are the primary inactivating species. There are two parts to the model. One is the probability of interaction between hydroxyl radicals and s pores, the other is the probability of inactivation of a spore after interacting with a hydroxyl radical. The product of these two parts represents probabiliti es of spore inactivation. Th e value calculated from model fits the experiment data. Horie et al. (1998a) determined specific in activation rate consta nt of immobilized titania on activated charcoal granules (T/AC) against E. coli based on series-event model. The constant was related with photocatal yst concentration and light intensity by considering the adsorption of cells to the photocatalyst.

PAGE 58

40 ) /(/ 0 / ' ' ' ' AC T t AC T c obsV N C Aq I k (2-16) where k” is apparent ster ilization rate constant, obsI is average light intensity, ” is a function of kinetic parameters in terms of the generation, decomposition and cell deactivation, A is light entrance area of test solution in reactor, qc is amount of cells adsorbed to T/AC granules, CT/AC concentration of a T/AC granule, Nt=0 is initial concentration of viable cells, VT/AC is volume of a T/AC granule. Horie et al. (1998b) related st erilization rate constant w ith light quantities absorbed by titania slurry, which are evaluated by c onsidering both the dependency of absorbance of titania slurry on light wavelengths and the spectral distributions of light rays from the respective lamps. The rate constant was determined on the basis of a series-event model. From the results obtained in experiments usi ng different light sour ce, time profile of E. coli inactivation under sunli ght can be predicted. Koizumi and Taya (2002a) found that the re lationship between survival of phage MS2 exposed to titania under black light and reaction time was: 'exp()A TN kt N (2-17) where k’ is inactivation rate constant w ith respect to microbial concentration, NA is the concentration of virus at reaction time t, NT is the initial virus concentration. A linear relationship was found between the inactivation rate cons tant and light intensity. Although the relationship between inactivation rate constant and titan ia concentration was nonlinear, the relationship became linear when the amount of phage adsorbed on titania particles was taken into account. The authors derived a model relating the

PAGE 59

41 inactivation rate constant to light intensity, titania concentration and adsorbed phage as shown below. '()TOobs T TOLTACI kq dVN (2-18) where is a coefficient, A is entrance area of light rays into reactor, CTO is concentration of titania, obs I is average light intensity, dTO is mean diameter of titania particles, VL is reaction volume, NT is total concentration of virus in reactor and qT is total amount of virus adsorbed on titania surface. Koizumi and Taya (2002b) found a proporti onal relationship between inactivation rate constant and phage quantities adsorbed on the titania surface, which was used to explain inhibitory activity of photocatalytic inactivation by kinds of ions. The inactivation rate constant and quantity of MS2 adsorbed on titania particles was calculated by the following equation. ) exp('t k N No A (2-19) Where N0 and NA are the titer values at the onset and a given irradiation time, k’ is the inactivation rate constant and t is irradiation time. TO e F VC N N q/ ) (, 0 (2-20) Where qV is the quantity of MS2 adso rbed on titania particles, NF,e is the titer of MS2 in the supernatant after adso rption equilibrium and CTO is the titania concentration in the suspension. Koizumi et al. (2002) analyzed the bact erial inactivation w ith photocatalytic reaction time based on series-e vent model (see equation 2-13). The value of n was found to increase with increased initial superoxi de dismutase (SOD) activity. Transition of

PAGE 60

42 intracellular SOD activity expressed was in ag reement with the observed data based on the model where considering the changes in bacterial populations with varied SOD activities. Investigators studying degradation of or ganic compounds have used the LangmuirHinshelwood relationship because it is appl icable for the adsorption of organic compounds on catalyst surfaces. Rincon and Pu lgarin (2003) observed that the kinetics of photocatalytic inactivation of E. coli could be described by the Langmuir-Hinshelwood relationship at titania concentr ations in the range 0.025-0.25g/L, but not over the range of 0.025-1.5g/L. Sato et al. (2003) studied deactivation of E. coli IM303 using titania-loaded quartz glass plates placed in air. E. coli IM303 (a superoxide dismutase –deficient mutant) was used because the deactivation of the cells was found to follow a simple first-order kinetics in terms of viable cell numbers duri ng photocatalytic inactiv ation. Mixture of spores and titania suspension were deposited on quartz plate to determine deactivation rate. A series-event model was used to calcula te apparent deactivati on rate constant. It was found that the apparent deactiv ation rate constant kept cons tant in the ra nge of initial cell number from 1069 cells/m2. Then deactivation profile of cells falling on titanialoaded plate was modeled by combining the empirical equation with the series-event model. The calculated result was in fair agreement with experimental data. ), ( 2 ) ( exp 2' 2 2 max ,t N k m t N r rV F V D F min 4 . 0 t (2-21) where rF is accumulation rate of viable cells falling on titania loaded plate, rD is deactivation rate of viable cel ls on titania loaded plate, max , F VN is maximum number of viable cells falling on titania loaded plate, and m were determined by matching the

PAGE 61

43 empirical equation to experimental data us ing the non-linear squares method, t is the process time, k’ is apparent deactivation rate constant, NV(t) is the number of viable cells on titania loaded plate at time t. Sun et al. (2003) investigate photomin eralization of bacteria using TiO2-Fe2O3 membrane photocatalytic oxidation reactor. The experiment results revealed that photomineralization rate of E. coliform in the reactor followed pseudo-first-order kinetics by the role of dissolved oxygen. An empirical model was derived according the pseudofirst-order kinetic: HRT t C HRTk C HRTkC HRTk ) 1 ( ln ) 1 ( 10 0 (2-22) Where C0 is initial concentration of microbe, HR T hydraulic retention time, k is pseudofirst-order rate constant and t is th e photocatalytic oxidation reaction time. Cho et al. (2004) used the delayed Chick-Wa tson model because it is able to depict the shoulder and tailing of the plot of E. coli concentration versus time when E. coli were exposed to titania under UV-A i rradiation and because it uses CTas the independent variable, allowing it to be compared with CTvalues published for other disinfectants. The delayed Chick-Watson model is given by lnoN N 1 0 if ln()lag oN CtCT kN (2-23) lnoN N lag1 T if ln()lag oN kCtkCCtCT kN where N0 is the initial cell concentration, N is the cell concentration at time t, C is the hydroxyl radical concentration, k is the inactivation rate constant, 0tCCdtt is the

PAGE 62

44 time-average hydroxyl radical concentration and lagCT is the x-intercept of the plot of N/No versus CT. The delayed Chick-Watson model gave good fits to experimental data. Tao et al. (2004) adopted acoustic wave im pedance analysis to study the effect of immobilized titania photocatalyst on the growth of E. coli . A impedance response model was established by relating moti onal resistance variation ( R) to Gopertz model. Thus, a relationship between R and three growth kinetic pa rameters (lag time, maximum specific growth rate an d asymptote) are found. 1 ) ( exp exp' 1t A e A Rm (2-24) Where t is the culture time, A is asymptote, m is maximum specific growth rate and is lag time. Coronado et al. (2005) found relations hip between outle t and inlet cell concentrations inside the continuous pl ug flow photoreactor (CPR). Two main assumptions were made: a) Langmuir-Hin shelwood kinetics describes the cell inactivation rate on catalyst surf ace; b) specific cell inactivati on rate follows a first order kinetics with respect to the fractional cove rage of photocatalyst surface by cells. The relationship was as following: t k C CNRT CPR a a 1 ) exp( exp' 0 (2-25) K k V S k ) (' (2-26) Where 0 aC is cell concentration at the photoreactor outlet just after starting the irradiation, CPR is continuous plug flow photoreactor spatial time, NRT is NRT spatial time, S is catalyst surface, V is suspension volume, k is first order rate constant, K is equilibrium adsorption constant of cells and t is time.

PAGE 63

45 Exosporium Outer Spore Coat Inner Spore Coat Cortex Plasma Membrane Germ Cell Wall Core Exosporium Outer Spore Coat Inner Spore Coat Cortex Plasma Membrane Germ Cell Wall Core Figure 2-1. Structure of a typical ba cterial spore (image redrawn from www.bmb.leeds.ac.uk/mbi ology/ug/ugtech/icu8/ mages/introduction/spore.gif ) Stage 0 Normal Growth Stage I Asymmetric separation Stage II Engulfment Stage III Cortex Synthesis Stage IV Coat Synthesis Stage V Mother cell lysis Stage VI Free Spore Stage 0 Normal Growth Stage I Asymmetric separation Stage II Engulfment Stage III Cortex Synthesis Stage IV Coat Synthesis Stage V Mother cell lysis Stage VI Free Spore Stage 0 Normal Growth Stage I Asymmetric separation Stage II Engulfment Stage III Cortex Synthesis Stage IV Coat Synthesis Stage V Mother cell lysis Stage VI Free Spore Stage 0 Normal Growth Stage I Asymmetric separation Stage II Engulfment Stage III Cortex Synthesis Stage IV Coat Synthesis Stage V Mother cell lysis Stage VI Free Spore Figure 2-2. Stages of sporulation (Redrawn from www.bact.wisc.e du/.../inclusions.html)

PAGE 64

46 CO2,Cl-, H+,H2O Red+ 3 4 5 Ti Ti h+e1 6 2 7 OH Red Ox-Ox .. CO2,Cl-, H+,H2O Red+ 3 4 5 Ti Ti h+e1 6 2 7 OH Red Ox-Ox .. Red+ 3 4 5 Ti Ti h+e1 6 2 7 OH Red Ox-Ox .. Figure 2-3. Primary steps for photoelectro chemical mechanism (redrawn from Hoffmann et al. 1995). 1 formation of charge -carriers by interaction of a photon with titania; 2 recombination of charge-carriers, produc ing heat; 3 reaction of valence-band hole with electron don or; 4 reaction of conduction-band electron with electron acceptor; 5 furt her thermal reactions to mineralize reactant; 6 conduction-band electron trapped on TiO2 surface; 7 valenceband hole trapped on TiOH group at TiO2 surface 100 ns 10 ns 10 ns h h + fs 100 ns TiO2 e Ox ms Ox >TiOH >TiOH + Red >TiOH+ Red + >Ti(IV) ps >Ti(III) >Ti(III) >Ti(IV) recomb. >TiOH>TiOH v.b. c.b. + + + + + + oxidation + reduction recomb. 100 ns 10 ns 10 ns h h + fs 100 ns TiO2 e Ox ms Ox >TiOH >TiOH + Red >TiOH+ Red + >Ti(IV) ps >Ti(III) >Ti(III) >Ti(IV) recomb. >TiOH>TiOH v.b. c.b. + + + + + + oxidation + reduction recomb. Figure 2-4. Time scales for photoelectroch emical mechanisms. Valence-band holes can be trapped by surface TiOH or recombine with trapped conduction-band electron on the order of 10 ns. Trap ped holes can oxidize reductant or recombine with conduction-band electrons on the order of 100 ns. Trapping of conduction-band electrons takes place on the order of ps, whereas reduction of oxidant by conduction-band electrons occurs on the order of ms. Redrawn from Hoffmann et al. (1995)

PAGE 65

47 Table 2-1. Optimum TiO2 concentrations in aqueous system Microbes TiO2 conc. (mg/mL) Form of TiO2 Conditions Reference Escherichia coli 0.0025 Aerosil P-25a Diffuse-light emitting optical fibers (DLEOF) Matsunaga and Okochi (1995) Escherichia coli 1 anatase TiO2 Sunlight Salih (2002) Escherichia coli 1 Degussa P25g Black light PI = 356 nm I = 8 W/m2 Maness et al. (1999) Escherichia coli 1 Degussa P25b Black light = 300 ~ 400 nm PI = 360 nm Bekbolet and Araz (1996) Escherichia coli 0.1 Aerosil P-25a High-pressure mercury lamp PI = 365 nm; 405 nm Horie et al. (1996) Serratia marcescens 0.1 Degussa P25c UV low pressure mercury lamp PI = 350 nm Block et al. (1997) Streptococcus sobrinus AHT 1 Degussa P25d Near-UV light = 300-400 nm; PI = 352 nm Saito et al. (1992) Vibrio parahaemolyticus 1 Aerosil TiO2 e UV lamp PI = 360 nm; I = 4 W/m2 Kim et al. (2003) Bacillus pumilus Spores 2 no specific information PI = 365 nm I = 22 W/m2 Pham et al. (1995) a anatase, Nippon Aerosil Ltd, Tokyo, Japan b anatase type, average size 30 nm, BET surface area 55 m2/g c mainly composed of anatase, BET surface area 50 m2/g, average size 21nm d mean particle size 21nm, 70% anatase, 30% rutile; pI 6.6; P25, Nippon, Aerosil, Japan e from Yakuri pure chemical company, Osaka, Japan f from Sandia National Lab., Albuquerque, NM, deduced from the context, TiO2 is powder g Degussa, 75% anatase, 25% rutile, surface area 50 m2/g

PAGE 66

48 CHAPTER 3 GOAL, HYPOTHESES AND RATIONALE, AND SCIENTIFIC MERIT As reviewed in this proposal, there ar e 115 studies conducted on the photocatalytic inactivation of microbes. Among them, 81 pape rs reported on bacterial inactivation, 15 on viral inactivation, 6 on f ungal inactivation, and 15 on in activation of other microbes such as protozoa, mammalian cells and micr oalgae. Out of the 81 papers on bacterial inactivation, only 11 addresse d inactivation of bacterial endospores by semiconductor photocatalysis. Of these 11, only 5 papers de alt with photocatalytic inactivation against spores with air as th e continuous phase. Finally, only 3 studie s have addressed the activity of a dry photocatalytic powder against dry endospores. One paper reported that on one surface, inactivation of spores was not improved by contact with titania under UVA, whereas on a different surface, photo catalysis contributed significant to inactivation. Data in a second paper also showed a significant contribution of photocatalysis to bacterial in activation. It was found diffe rent light intensities were applied in these two papers. The gap in knowledge identified in this re view is very important for the practical application of photocatalytic powders agai nst biohazardous agents. Simply put, no one knows if a photocatalytic powder could be e ffective in destroyi ng a biohazardous agent that is dispersed throughout th e living environment. Scientif ically, information about the interactions between photocatal ytic powders and biohazardous particles in the dry state could aid in developing models of photocatalytic disinfection that take into account such factors as light intensity, surface loading, and microbe characteristics. A summary of

PAGE 67

49 important parameters for testing a dry syst em for microbe inactivation by photocatalytic powders is given in Table 3-1. 3.1 Goal Up to now, the effectiveness of photocatalytic powders against microbes in the dry state was rarely investigated. Influences of factors such as light intensity on the contribution of photocatalysis to inactivati on of spores on dry surfaces have not been determined. Accordingly, the goal of the proposed re search is to determine the influence of light intensity on the effectiveness of titania powders against bacterial endospores in the dry state under solar UV irradiation. 3.2 Hypotheses and Objectives Specific Hypothesis 1. The rate of inactivation of bacterial endospores in dry state under solar UV irradiation can be increased by contact wi th titania nanoparticles (Degussa P25). A number of studies have demonstrated enhanced inactivation of microbes in dry state under solar UV irradiation when placed on photocatalytic surf aces. Thus, it is expected that photocatalytic na noparticles, in contact with microbes in dry state, would have a similar effect. Thus, Research Ob jective 1 was to compare inactivation of irradiated spores in contact with titania nanopa rticles to inactivation of irradiated spores without titania. Specific Hypothesis 2. There exists a light intensity at which the contribution of photocatalysis to spore inactivation is maximized. Some studies (including thos e on aqueous systems) have failed to demonstrate a contribution of photocatalysis to spore in activation whereas ot her studies show a significant enhancement by photocatalysis. A no table difference between these studies is

PAGE 68

50 the light intensity used in these studies. T hus, it is expected the relative contribution of photocatalysis to inact ivation will change depending on light intensity. Research Objective 2 was, accordingly, to compare inact ivation rates measured in the absence of photocatalyst (i.e., photolytic rate s) to the rates measured in the presence of photocatalyst (i.e., combined rates of photolysis and photocat alysis), over a range of light intensities. 3.3 Scientific Merit of Proposed Research Up to now, there has been no systematic i nvestigation of the f actors that influence inactivation of microorganisms by photocatalytic particles in a dry state. Among these factors, light intensity stands out because of its fundamental role in photocatalysis and the general failure to recognize the significance of light intensity in determining the potential contribution of photocatalysis to photo-induced disinfection. This relationship must be understood in order to appropriately deploy photocatalytic systems for protecting human health. Table 3-1. Important parameters for testing photocatalytic inactivation of dry endospores Parameter Reason Light intensity Light intensity quantifies the available energy Surface loading Tradeoff exists between qua ntity of photocatalyst and self-shading by photocatalytic particles Particle size Affects quantum efficiency of photocatalyst and adsorption to microbes Purity of spore suspension High purity (low vegetative ce lls) is needed to obtain representative results Distribution quality Good distribution is necessary to obtain reproducible results Recovery of spores from surfaces for enumeration A high and consistent recovery of spores from surfaces is necessary to obtain reproducible results

PAGE 69

51 CHAPTER 4 EVALUATION OF BACTERIAL ENDO SPORE PURIFICATION METHODS 4.1 Introduction Considerable research into methods for inactivating b acterial endospores that are released in buildings or distributed in th e environment has recently been carried out (Hamouda et al. 1999; Larson and Marinas 2003; L ee et al. 2005; Lei et al. 2005; Rice et al. 2005; Vohra et al. 2005). Furthermore, contamination of food, animal feeds, and hospitals by bacterial spores is a health c oncern (Bohnel and Gessler 2004; Christiansson et al. 1999; Ehling-Schulz et al. 2004; Rosenquist et al. 2005; Schoeni and Wong 2005). Bacterial endospores have a complex shell structure—not present in vegetative cells— that imparts resistance to disi nfection techniques that rely on heat, chemicals or radiation (Block 2001). Spores of certain bacteria (such as B. cereus ) also have surface appendages—the exosporium and filaments (Hachisuka et al. 1984; Hachisuka and Kuno 1976). Studies of spore inactiv ation methods, particularly those that degrade organic materials, consequently require highly pure spore suspensions (cont aining few vegetative cells) in order to achieve representative results. Spores are produced by vegetative ce lls of certain bacteria (e.g., Bacillus , Clostridium ). Hence, presence of vegetative cell s in spore suspensions is unavoidable. The general approaches for purifying spore suspensions are to repeatedly wash the suspension in order to remove vegetative cells or to apply disinfectants that kill the vegetative cells while minimally harming the spores. Variations of these approaches

PAGE 70

52 have been applied to spores from a nu mber of bacterial species, including B. subtilis, B. cereus and B. anthracis , as shown in Table 4-1. Few comparisons of the various purificati on methodologies have been carried out. Nicholson and Setlow described several sp ore purification met hods and potential problems of these treatments (Harwood 1990). They mentioned that each of the methods could give suspensions contai ning at least 98% free spores. Other than the single value for purity, no quantitative data were provided. Nicholson and Galeano (2003) compared heat treatment alone to lysozyme treatment followed by heat shock for Bacillus anthracis . They mentioned that lysozyme treat ment was effective in removing from suspension the vegetative cells of Bacillus anthracis (which differs from Bacillus cereus only in the genes encoding toxin components (Ivanova et al. 2003; Read et al. 2003)). Based on phase-contrast micrographs, they conc luded that the lysozyme followed by heat treatment gave higher purity than heat shock alone. No quantitative data were provided. Dragon and Rennie (2001) compared eff ectiveness of heat shock and ethanol treatment. They demonstrated that both tr eatments were equally bactericidal against vegetative stocks of B. cereus and did not reduce the viability of B. anthracis spore stock. However, no data regarding purity of spore suspensions achieved were reported. Based on the very limited database of co mparative information, it has up to now been impossible for researchers to rationally select a spore purification method that would enable them to achieve their research objectives. Accordingly, the overall goal of the present study was to compare the majo r techniques for purifying suspensions of bacterial endospores. Given the increasing popularity of B. cereus as a surrogate for B. anthracis (Beuchat et al. 2005; Blatch ley et al. 2005; Ivanova et al. 2003; Read et al.

PAGE 71

53 2003; Rice et al. 2005), this organism was chos en as the model for the present research. As seen from Table 1, the most widely us ed techniques are ethanol , heat shock, daily water washes and lysozyme. Hence, thes e methods were evaluated in the present research. The first objective was to classify met hods according to efficacy, which was considered to include purity, yield (which is used here to indicate the number of spores successfully harvested relative to the number of spores and cells in the original culture), and time and effort required to carry out th e purification. The second objective was to assess the effect of the methods on the integrity of treated spores, since it is plausible that disinfectants used during purification could damage spores and, hence, alter their resistance. 4.2 Materials and Methods 4.2.1 Chemicals All chemicals were obtained from Fi sher Scientific, except as noted. 4.2.2 Preparation and Storage of Agar Plates Plates were made by pouring autoclaved tryptic soy agar (DifcoTM prepared according to manufacturer’s direction) into 100 15 mm sterile plastic Petri dishes (Fisher Scientific) and air dried in a laminar flow hood (LABCONCO purifier class 2 safe cabinet, cat. no. 36209-000 R) for 24 hours. The dr ied agar plates were used immediately or stored in refrigerator at 4 C until use. 4.2.3 Storage and Inoculation of Test Strains Bacillus cereus (ATCC 2) was provided by Dr. Jerzy Lukasik, Depatment of Microbiology and Cell Science, University of Florida. For long-term storage (more than one year), B. cereus was stored in media cont aining 8% glycerol at -70 C(Sambrook et

PAGE 72

54 al. 1989). A single bacterial colony was inoc ulated into 10 mL liquid growth media using a sterile metal loop and incubated overnight at 35 2 C on an orbital incubatorshaker (Model C24, New Brunswick Scientific ) at 250 rev/min. The procedure for liquid growth media preparation is de scribed in the next section. After incubation overnight, 850 L of culture was transferred to a sterile vial containing 85 L sterile 80% glycerol. (The 80% glycerol was made by mixing 8 mL glycerol (Sigma-Aldrich) and 2 mL deionized water and sterilizing the mixed solu tion.) The vial was capped and then the contents of the vial were mi xed thoroughly by vortexing. Th e vials were then immersed for 5 sec in liquid nitrogen and transferred to a -70 C deep freezer. Recovery of bacteria from deep freezer was by scratching the fr ozen culture using sterile wooden stick (Fisherbrand, Cat. # 01-340) and inoculating into 10 mL liquid growth media. The inoculated media was inc ubated overnight at 35 2 C on the orbital in cubator-shaker. According to the concentration of cultured media, suspension was diluted and plated on a sterile plastic Petri dish. The Petri dish was inverted and incubated at 35 2 C in an incubator (Fisher Isotemp incubator, Model 303) and then stored in refrigerator at 4 C. New plates were streaked every two weeks us ing biomass from old plates. The culture was renewed from frozen stock at three-month intervals. 4.2.4 Bacterial Culture Culture media was prepared according to the American Society for Testing and Materials (ASTM) E2111-00 standard using Di fco Columbia broth powder. A volume of 10 mL of Columbia broth was prepared by a dding 0.35 g of Columbia broth powder to 10 mL deionized water in a 125 mL Erlenmeyer flask. A volume of 10 mL of 10 mmol/L MnSO4 was made by adding 0.0169 g MnSO4•H2O to 10 mL deionized water in a 125

PAGE 73

55 mL Erlenmeyer flask. These two solutions were autoclaved. U nder the laminar flow hood, the 10 ml autoclaved Columbia broth wa s diluted to one-tenth strength by adding 90 mL autoclaved deionized water in it. Finally, 1 mL of autoclaved 10 mmol/L MnSO4 was added to 99 mL of 1/10 st rength Columbia broth. Under the laminar flow hood, a loop of colony was inoculated in 100 mL of prepared liqui d growth media contained in 500 mL Erlenmeyer flasks. The flasks we re capped with foam plugs to allow air exchange between the flask and the atmos phere. The media contained 0.1 mM MnSO4, which improves sporulation efficiency and spore stability (Atrih and Foster 2001; Rabinovi and Dasilva 1973). The inoculated growth media was incubated at 35C on an orbital incubator-shaker (Model C24, New Brunswick Scie ntific) at 250 rev/min for 3 days (ASTM, 2001) or 10 days. The vegetative cells and spores were then harvested and the spores were purified as described in the following. 4.2.5 Spore Purification Suspensions were analyzed immediately after completing the purification procedures. Samples for TEM analysis were stored in deionized water at 4 C for a period less than 48 hr. ASTM. ASTM treatment followed the E2111-00 st andard. A suspension of spores and vegetative cells was harvested by tran sferring 70 mL of cu lture to an 85 mL centrifuge tube and then was centrifuged at 10,000g for 10 min at 4C using a Marathon 21000R centrifuge. The supernatant was poured off and the pellet was resuspended in 20 mL of sterile deionized water by vortexing fo r 20 sec. The suspension was centrifuged and the pellet resuspended as before. This washing procedure was repeated two additional times.

PAGE 74

56 Ethanol. Culture was harvested and the pell et was washed once with sterile deionized water as before. The pellet was re suspended in 20 mL of 1:1 sterile deionized water and ethanol. The centrifuge tube was capped and incubated at 22C for 12 hr on an orbital shaker (Lab Line) at 100 rev/min (3-d ay culture) or for 2 hr at 200 rev/min (10day culture). The suspension was then washed twice with sterile deionized water. Heat shock. The heat shock method wa s modified from Peng et al. (Peng et al. 2001). Culture was harvested as before and resuspended in 20 mL sterile deionized water. The suspension was washed three times with sterile deionize d water and then was transferred to a 125 mL Erlenmeyer flask, which was immersed in an 80 2C water bath for 15 min. The flask was transferred i mmediately to an ice water bath. The spores were then washed three times with sterile deionized water. Daily water washes. In the daily water washes pr ocedure, the 3-day culture was harvested as before and washed three times in 20 mL sterile deionized water (Harwood 1990). The suspension was then placed in i ce and agitated at 100 rev/min on the orbital shaker. The suspension was washed once daily with ice cold sterile deionized water over a period of 7 days. The 10-day culture was heat shocked at 80 2C for 15 min after three water washes. Daily water washes were repeated under the same conditions applied to the 3-day culture except that the agita tion speed was 40 rev/min and the period in which washes were carried out was 15 days. Lysozyme. The lysozyme method was carried out according to Xue and Nicholson (1996). Lysozyme and phenylmethylsulfonyl fluoride (PMSF) were stored at 4C and prepared fresh from powder for each use. Because of its low solubility, PMSF was

PAGE 75

57 dissolved in the minimum possible volume of ethanol (< 2 mL) before adding to the mixture. Spore suspensions purified by the alternativ e methods were stored in the dark at 4 C. Each purification method beginning with culture of bacteria was repeated three times for the 10-day culture period and once for 3-day culture period. 4.2.6 Spore Analysis Counting. Serial dilution tubes were prepared by pipetting a volume of 3.0 mL sterile phosphate buffered salin e (PBS; pH 7.2) containing 2 mM sodium dodecyl sulfate (SDS) into 13 100mm glass tubes. (The PBS/SDS solution was made by adding 1.236 g Na2HPO4, 0.18 g NaH2PO4, 8.5 g NaCl, and 0.57 g SDS to 1 L deionized water.) The tubes were capped and then autoclaved. Serial dilution and pour plat ing procedures were carried out under a laminar flow hood. The culture was serially diluted by addi ng 0.33 mL of sample to a volume of 3.0 mL sterile PBS/SDS. Each dilution tube was vortexed for 10 sec. immediately before subsampling for further serial dilutions or to inoculate spread plates . Each pour plate was prepared by pipetting 0.1 mL d iluted culture to a st erile Petri dish and then pouring sterile liquid tryptic soy agar (held at a temperature of 40-50C) into the Petri dish. The mixture of liquid agar and diluted culture were swir led gently by hand and then set down on the benchtop. Three dilutions of each culture we re plated and each dilution was plated in quadruplicate, unless noted otherwise. The pl ated dishes were air dried in laminar hood, inverted, and then incubated at 35C. An incubation time of 12 hr was used to allow visualization of colonies while avoiding overg rowth. Colonies formed on each dish were

PAGE 76

58 counted. Count numbers between 30 and 300 on a plate were preferre d (Madigan et al. 2000). Numbers of colonies below 30 we re accepted for the lowest dilutions. It was also found that dilution with D.I. water does not introduce spore agglomeration. Thus, D.I. water can be used as an alternative to PBS/SDS for dilution of spore suspensions, then c ounted by pour plating. Image analysis. A 20 L drop of purified spore suspension was smeared onto a clean glass slide and air drie d in a laminar flow hood. Samples were imaged under an Olympus BX60 optical microscope using a 50 objective lens. Four images were taken per slide (one from each quadrant) and the counts were averaged. A SPOT digital camera (DIAGNOSTIC instruments, Inc. USA) was used with bright field illumination in transmission mode. Images were digitized by SPOT Advanced Version 4.0.9 software, which allowed capture of 1600 1200 pixel JEPG images at 8 bits per pixel monochrome. Digital images were analyzed using Im age-Pro Plus 4.0/4.1 software (Media Cybernetics). Pixels with a grey level less th an the critical threshold were considered as particle pixels because particles (spores a nd vegetative cells) were darker than the background. The critical threshold was se t to maximize the number of particles successfully identified by the software. The counts by image analysis were within an average of 4% of counts by visual observation. Vegetative cells were relatively scarce and easily distinguishable from spores by their shape and thus could be counted ma nually. The number of spores on each image was found as the difference betw een the total partic le count and the number of vegetative cells.

PAGE 77

59 4.2.7 Data Analysis Purity of the final suspension was calculated as the ratio of the number of spores to the total number of spores and cells. Th e yield of spores was calculated as the concentration of spores in th e purified suspension divided by the total concentration of vegetative cells and spores in the culture befo re purification. (The 3.5x factor due to harvesting 70 mL of culture and resuspendi ng in a 20 mL volume was taken into account in these calculations.) Data were analyzed by one-way ANOVA (Sokal 1994) and Tukey’s post-hoc test (Berthouex 1994). 4.2.8 TEM Sample Preparation Transmission electron microscopy was car ried out with a Hitachi H-7000 TEM. Specimens were prepared on a 300 mesh carbon co ated copper grid with formavar film (EMS, Hatfield, PA). A 10 L drop of spore suspension was pipetted onto the grid. After one minute, the drop was blotted with filter paper. Specimens were subsequently stained with 10 L of 1% uranyl acetate for 30 sec, followed by blotting. 4.3 Results 4.3.1 Purity and Yield Figure 4-1 shows a typical optical image of a purified spore suspension used in enumerating the total number of spores and vegetative cells. Spores were spherical or oval in shape with an average size (longest dimension) of 1 m, whereas vegetative cells were rod shaped with a length of 3-4 m and diameter of 1 m and often occurred in short filaments. Ethanol, daily water wash and lysozyme treatments gave the best purities (97%) for 3-day cultures (Fig. 4-2). Heat shock tr eatment provided the least purity (67%) and

PAGE 78

60 ASTM showed intermediate performance. All the methods gave mean purities of at least 93% when applied to the 10-day cultures (Fig . 4-3). Lysozyme treatment achieved the highest purity (99%), ASTM was intermedia te, and ethanol, daily water wash and heat shock were lowest. Using 3-day culture, the highest yields (20%) were achieved with the ASTM and ethanol methods (Fig. 4-2). At the ot her extreme, a yield of less than 1% was attained with lysozyme treatment. Intermedia te yields (5%) were achieved with the heat shock and water washes methods. The ASTM and ethanol treatments also gave the highest yields (39%) when applied to 10-day culture, whereas heat shock, daily water washes and lysozyme provided yields of 4% or less (Fig. 4-3). 4.3.2 Storage Time The effect of storage time on relative purity (final purity/initial purity) was investigated at storage times of 2, 3 and 4 mo (Fig. 4-4). The relative purity of spore suspensions from all of the treatments evaluate d was 1.0 or higher after 2 mo of storage. Some degradation of relative pu rity was observed at longer st orage times, particularly in suspensions treated by the heat shock met hod. There was no significant change in relative purity with respect to storage time of suspensions treated by lysozyme. 4.3.3 Spore Integrity. The spore morphology for ASTM, ethanol, heat shock and daily water washes showed exosporium and, to a lesser extent, fi laments. Representative images of spores treated by these methods are shown in Figure 4-5 and 4-6. The lysozyme treatment substantially altered spore mor phology, as shown in Figure 4-7.

PAGE 79

61 4.3.4 Time Investment and Complexity Characteristics of the four purification methods in te rms of time investment and complexity are shown in Table 2. Technician time and complexity followed similar general trends, with the simplest method (ASTM) requiri ng the least tech nician time and the most complex method (lysozyme treatment) requiring the longest technician time. When the total time required to obtain the purified suspension is considered, the daily water wash method is most time consuming. 4.4 Discussion 4.4.1 Purity and Yield Yield of purification tec hniques has never been re ported (qualitatively or quantitatively) to our knowledge, while purity is sometimes mentioned based on optical/phase contrast microscopy. Accord ing to Nicholson and Setlow, water wash, french press, lysozyme treatment and urogr afin gradients can give >98% spore purity (Harwood 1990). This conclusion is consiste nt with our experimental results for lysozyme and daily water wash treatments. Since sporulation becomes more complete with time, it can be presumed that longer cult ure times will lead to higher purity. This has been implied (Harwood 1990) but not substan tiated in literature. The presumption is supported by our data. The two methods that involve numerous washes (daily water wash, lysozyme), together with heat shock, had the lowest yields. The washing procedure includes centrifuging the spore suspension with subsequent decantation of supernatant and resuspension of spore pellet in fresh diluen t by vortexing. Some spores (as well as vegetative cells) are lost in each decantation. Thus, the higher the number of washes, the lower the final concentration of spores.

PAGE 80

62 4.4.2 Spore Integrity Prior investigators (Hamouda et al. 1999; Jorge De Lara 2002; Nicholson and Galeano 2003) found that only vegetative cells are damaged during purification. The ASTM, ethanol, heat shock and da ily water washes treatments in the present research had negligible effects on spores, consistent with the literature. However, the lysozyme treatment caused substantial al teration of spore morphology. Lysozyme is an enzyme known to attack peptidoglycan, which makes up the spore cortex (Waites et al. 1976). Bacillus spores have proteinaceous inner and outer coats protecting the cortex from lysozyme. However, the resistance of the spor e coats of different species varies (Waites et al. 1976). Suzuki et al. have shown that certain spores can be damaged by lysozyme, and postulated this effect was dependent on the nature of the spore coats (Suzuki and Rode 1969). Thus, specific testing of suscep tibility of spores to lysozyme should be prerequisite to use of lysozyme tr eatment for spore purification. 4.5 Conclusions The ethanol treatment method applied to 3day culture could be considered optimal on the basis of time investment and complex ity as well as purification efficiency and yield. However, if a longer total time is acceptable, the ASTM method applied to 10-day culture becomes most attractive. Al though the lysozyme treatment achieves a consistently high purification efficiency, its high complexity and time investment, low yield, and damage to spores make this method least attractive.

PAGE 81

63 Endospore Vegetative cells Endospore Vegetative cells Figure 4-1. Optical micrograph of spore su spension after heat shock treatment (bright field illumination; 500); scale bar is 10 m.

PAGE 82

64 a a c a b e c d b a0 20 40 60 80 100 ASTMEtOHHSDWLPercent Purity Yield Figure 4-2. Purity and yiel d of spore suspensions purifie d after 3 days of culture incubation. Means shown with the same letters are not significantly different at = 0.01. Standard deviations for most points are too small to be visible.

PAGE 83

65 b c c b/c a a a b b c0 20 40 60 80 100 ASTMEtOHHSDWLPercent Purity Yield Figure 4-3. Purity and yiel d of spore suspensions purified after 10 days of culture incubation. Means shown with the same letters are not significantly different at = 0.01. Standard deviations for most points are too small to be visible.

PAGE 84

66 0.85 0.90 0.95 1.00 1.05 1.10 EtOHHSDWLFinal purity / initial purity 'asdfg'''' 2 mo 3 mo 4 mob a c c a b a a a b a b 0.85 0.90 0.95 1.00 1.05 1.10 EtOHHSDWLFinal purity / initial purity 'asdfg'''' 2 mo 3 mo 4 mo 0.85 0.90 0.95 1.00 1.05 1.10 EtOHHSDWLFinal purity / initial purity 'asdfg'''' 2 mo 3 mo 4 mob a c c a b a a a b a b Figure 4-4. Effect of storage time on relative purity of tr eated spore suspensions. Means shown with the same letters are not significantly different at = 0.01.

PAGE 85

67 Figure 4-5. Transmission elec tron microscopy (TEM) image of spores from suspension purified by ASTM method Figure 4-6. TEM image of spore from susp ension purified by daily water wash method

PAGE 86

68 Figure 4-7. TEM image of spores from suspension purified by lysozyme method

PAGE 87

69 Table 4-1. Variation of spore purification methods Spore Purification Methodology Variation Reference 60 minutes (Dragon and Rennie 2001) 2 hours (Hamouda et al. 1999) 50 % Ethanol Overnight (Hamouda et al. 2002), (Couvert et al. 1999), (Gaillard et al. 1998) 63 C, 20 minutes (Dragon and Rennie 2001) 3 washes, 65 C, 30 minutes (Ireland and Hanna 2002) Multiple washes, 70 C, 30 minutes (Setlow et al. 2001) No wash, 80 C, 30 minutes (Fujii et al. 2002) One wash, 80 C, 10 minutes (Zs. Cserhalmia 2002) 3 washes, 80 C, 12 minutes (Larson and Marinas 2003) 3 washes, 80 C, 20 minutes (Son et al. 2005) 6 washes, 80 C, 10 minutes (Peng et al. 2001) Heat Shock Multiple washes, 80 C, 10 minutes (Riesenman and Nicholson 2000) Number of washes not mentioned (Schiza et al. 2005), (Vaid and Bishop 1998), (Setlow et al. 2001) Daily water wash, 4-7 days (Paidhungat et al. 2002) 2500 g – 15 minutes, 4 times (Jorge De Lara 2002), (Fernandez et al. 2001) 4000 g – 30 minutes, 5-6 times (Leuschner et al. 2000) 10,000 g – 15 minutes, 8-10 times (Bailey-Smith et al. 2005) Daily Water Wash 12,000 g – 10 minutes, 5 times (Zhang et al. 2005) Sonication Multiple wash followed by sonication for 0.5-1 minutes (Setlow et al. 2003) Lysozyme treatment (Vohra et al. 2005) Lysozyme treatment Lysozyme treatment followed by heat shock at 80 C for 10 minutes (Nicholson and Galeano 2003), (Xue and Nicholson 1996), (Nicholson and Schuerger 2005), (Lee et al. 2005), (Link et al. 2004) Urografin gradient (Rice et al. 2005)

PAGE 88

70 Table 4-2. Time investment and comp lexity of spore purification methods Method Technician time for purification* Total time for 3-day culture** Total time for 10day culture** Complexity ASTM 1.5 hr 73.5 hr 241.5 hr Low Ethanol 3 hr 90 hr 258 hr Moderate Heat shock 3 hr 75 hr 243 hr Moderate Daily water washes 7 hr 247 hr 415 hr Moderate Lysozyme 11 hr 83 hr 251 hr Very high *Includes reagent preparation time, but not culturing period time. **Time from beginning of culture to collection of purified spores

PAGE 89

71 CHAPTER 5 EFFECTS OF SURFACE TYPE, APPLIC ATION METHOD AND OTHER FACTORS ON SURFICIAL PARTICLE DISTRIBUTIONS 5.1 Introduction A uniform distribution of s pores or mixture of spor es and nanotitania is a prerequisite for study of photocatalytic inac tivation of spores on dry surfaces. A nonuniform distribution is undesirable because it will decrease consistency and reliability of experimental results. Spreading of microorganisms on surfaces ha s received little attention in the literature. Only one paper was found that explored the performance of alternative approaches for dispersing microbes on su rfaces. Typically, methods for bacterial dispersion on surfaces were used without quant ification of their effectiveness. Since direct guidance on dispersion of spores or nanotitania on dry surfaces is virtually nonexistent, it was necessary in the present study to find a suitable method of uniformly dispersing spores or mixtures of spores and nanotitania on surfaces. Based on the literature review , it is apparent that the na ture of the surface to which the spores or mixture of spores and titani a are applied has a strong influence on their distribution. The most commonly used surf aces (glass, plastic, quartz, and membrane filters) were chosen for investigation. Surface modifications (sodium dodecyl sulfate, polyvinyl alcohol and platinum/gold sputter co ating) of glass and plastic surfaces were also tested. Selection of sodium dodecyl su lfate (SDS) as pre-coating was due to its hydrophilic and hydrophobic groups wh ich can change surface pr operties. Selection of

PAGE 90

72 polyvinyl alcohol (PVA) and pl atinum/gold pre-coating on glass surface was to decrease adhesion forces between spores and the surfaces in order to improve recovery of spores, as reported in Chapter 6. The spore app lication methods of dippi ng and pipetting have been commonly used and were therefore invest igated. Filtration was reported to achieve good dispersions of spores on membrane filter s and was therefore also investigated. Other factors, such as concen tration of particulate material (spores, nanotitania, mixtures of spore and nanotitania) and surface temperatur e were also evaluated. A summary of the factors tested and levels at which they were tested is give n in Table 5-1. 5.2 Background Several approaches were used to apply microbes to surfaces. They are pipetting method with or without further spreading, immersion and filtration methods. Other methods were also applied but no informa tion on application of spores on surface was given. Only one paper quantitatively asse ssed spore distribution qua lity and consistency of the sample preparation, however, no furthe r quantitatively or qualitatively comparison of spore distribution quality and sample repr oducibility was given for different surfaces by the same spore application method. 5.2.1 Pipetting Jacoby et al. (1998) studied mineraliza tion of bacteria deposited on titanium dioxide coated surface by photocataly tic inactivation. They dispersed E. coli by directly pipetting the bacterial suspension onto frittedglass disk coated with titania and dried under a stream of nitrogen gas. Alternatively, they pipetted 1 mL of titania suspension to a 250 mL gas sampling tube followed by 1 mL E. coli suspension, then dried with house air and moderate heating. An irregular film on the side of the sampling tube was observed. But no comment was given for their distribution on the surface.

PAGE 91

73 Moore and Griffith (2002) a pplied bacteria suspension on the surface by pipetting and spreading the suspension evenly over the surface using a “hockey stick’ shape spreader. However, no image or descripti on was presented on mi crobe distribution on the surface. Lin and Li (2003) tested inactivation of microbes on photo catalytic surface (commercial titania filter (DAIKIN, Japan) and titania coated glass slide) in air. Spore suspension of Penicillium citrinum was directly pipetted ont o the center of the surface containing titania and dried. For the control (no photocatal yst), suspension was pipetted onto nine points uniformly distributed on th e surface. Difference in spore application method on tested surfaces and control surface may be due to different surface property, where titania coated filter favored uniform sp reading of bacteria suspension while control did not. This is mentioned by the author s that spreading of suspension on the photocatalytic surface was uniform because tit ania is hydrophilic. However, no further information about dispersion quality of spor es after the spore suspension was dried was given. Galeano et al. (2003) spotte d bacteria suspension dire ctly on the center of the stainless steel coupons coated with antimic robial substrate to study the inactivation activity of the surface. Samples were placed within humidity chamber to prevent from drying. They also provided no information about distribution perf ormance of spores on the surface. Kuhn et al. (2003) pipetted bacterial suspension onto titania coated Plexiglas surface, then exposed to UVA light for 60 min. Bacterial suspensi on was not dried and distribution of bacteria on th e surface was not investigated.

PAGE 92

74 Foschino et al. (2003) applie d appropriate bacterial susp ension by dripping them onto stainless steel surface, and then withdrew 640 L from the suspension with an 8channel Finnpipette to disp erse in 64 droplets of 10 L each to achieve approximately 110 cells/cm2. The surface was dried under a bios afety cabinet. No assessment of microbe distribution was conducted. 5.2.2 Immersion Faille et al. (2002) investigated adhesion ability and strength of Bacillus spores and E. coli to inert surfaces. Six material (glass, stainless steel, po lyethylene high density, polyamide-6, polyvinyl chloride a nd Teflon) and three microbes ( B. cereus and B. subtilis and E. coli ) were tested. Bacillus spores were applied to surfaces by vertical immersion for 2 hr in a saline spore su spension and then quickly imme rsed into sterile deionized water to remove loosely attached microbes. No more detail on mi crobial distribution on surfaces was described. 5.2.3 Filtration Schiza et al. (2005) demonstrated that spore dispersion on surface can be improved by filtration method compared with the commonly used dropping method. Spore agglomeration was typically observed (image and quantitative analysis not shown in the paper) when spore suspension was dropped onto a gold-coated glass slide and air dried, whereas spores were observed primarily distri buted as individuals (unaggregated) on the gold-coated porous alumina membrane by filtration method. Sonicated Bacillus subtilis spore suspension was vacuum filtered onto the surface and oven dried. The purpose of sonication was to break up spore aggregates , which were previous ly lyophilized and stored dry. The vacuum applied to filtrati on was to obtain even di stribution of individual

PAGE 93

75 spores throughout the membrane surface. Re lative standard devi ation (RSD) of spore count from each sample was found to be 0.037, 0.0037 and 0.061. This is calculated based on counts from ten images taken at random location on each sample. The RSD was 0.026 for the average spore counts from th e triplcate samples. Number of spores on each image was counted by a LabVIEW (version 7.0, National Instruments, Austin, Texas) in-house software program. G ood spore distribution was shown from SEM images, which are in consistent with the low RSD. The distribution quality was not only qualified, also quantified by the author, which was not presented in other research papers. Pat et al. (2005) filtered titania solution on the cellulose acetate filter membrane first followed by filtration of vegetative cells. SEM images were shown to prove that the bacteria can be remained on the surface and might not penetrate a great distance into the membrane. However, no quantitative a nd qualitative investig ation on bacteria distribution on the surf ace was carried on. 5.2.4 Nebulizing Sato et al. (2003) used nebulizer to spray E. coli onto a titania-loaded plate. The cell suspension was poured into nebulizer and th en sprayed for 4 min to yield mists in a chamber where quartz glass plates were fixe d at a certain distance from nebulizer. However, the consistency and distribution quality of cells on the surface was not mentioned. 5.2.5 Other Methods Lindberg and Horneck (1991) conducted research on UV (200 nm) inactivation of B. subtilis on surface in dry state. They applied B. subtilis spore suspension on a quartz disc and dry. No in formation was given how the suspension was

PAGE 94

76 transferred to the surface. An equation was gi ven in the paper to calculate the fraction of unshaded spores on the surface by UV irradiation Fraction of unshaded spores = m me e m 1 (5-1) where m is the ratio of the area covered by spores to the total area of exposure. No reference was given for the equation. They applied 105 CFU spores on a quartz disc with a diameter of 7 mm and calculate d a 3% (we cannot get this number from the equation 5-1) overlapping spores on the su rface by assuming a projected spore surface area of 1 m2. Also dry spores were irra diated at a density of 2.6 108 spores per 13 cm2 area of glass support, which resulted in a sh ading effect of 10% (T his number was same as our calculation based on equation 5-1). However, no image and explanation were presented to demonstrate. Horneck et al. (2001) tested B. subtilis inactivation in space. They mentioned a fairly homogeneous distribution of the spores was achieved by transferring spore suspension onto silanized quart z plates, but no image was pr esented and no detail was given about the production of plates a nd transferring of the suspension. Vohr et al. (2005; Vohra, Goswami et al . 2005) studied photocatalytic inactivation of bacterial spores on surfaces (metal and fa bric substrates coated with silver-doped titaniaum dioxide). It was desc ribed that a volume of 0.1 mL B. cereus spore suspension were spread uniformly on the surface using a st erilized (alcohol and a flame) glass rod. However, the approach that the spore susp ension was transferred on the surface was not mentioned. In addition, no image was s hown how uniform the distribution is and no further assessment for distribution quality of spores on the surface and quantification of distribution was carried out.

PAGE 95

77 5.2.6 Summary The effects of application methods on microbial distributions on surfaces is summarized in Table 5-2. Among all the revi ewed approaches for spreading microbes on the surface, filtration method a pplied by Schiza et al. (2005) appeared to be the most promising method. A very uniform distri bution of bacterial endospores on the goldcoated alumina membrane using vacuum filtration method was demonstrated by SEM image, which other literatures didn’t show im ages to illustrate distribution quality. Although Horneck et al. (2001) claimed a ho mogeneous distribution of spores was obtained on quartz surface, deta ils on how to spread spores on the surface were not presented in the paper, in addition, no im ages were shown and no quantitative analysis were given to prove the dist ribution quality. Other papers simply used methods for application of microbes on su rface without investigation on distribution performance of the microbes. Schiza et al. (2005) are the only inves tigators to quantitatively assess spore distributions on surface. Other investigators either mentioned the distribution quality without giving any detailed information or simp ly ignored distribution quality. Schiza et al. (2005) also investigated re producibility and consistency of the filtration method for distributing spores on membranes. A good repr oducibility and consistency of distribution quality was achieved by filtration method. Re lative standard deviation of spore counts from ten images taken at random location of sample was used to quantify spore distribution quality. The obtai ned low standard deviation in counting matches well with the uniform distribution obs erved from SEM image.

PAGE 96

78 5.3 Materials and Methods The procedures that were applied to severa l aspects of testing are described in the present section. Materials and methods that are specific to a part icular investigated parameter are given together with the results of the investigation. 5.3.1 Washing of Cover Slips and Glass Slides Glass, plastic cover slip (22 22 mm) and glass slide (76.2 25.4mm) were purchased from Fisher. Surfaces were washed with detergent (Sparkleen 1, Fisher), rinsed thoroughly with deionize d water, and then blow dried under an air stream from an air cylinder. Thereafter, th e surfaces were rinsed three times with ethanol followed by three rinses with deionized water. Finally , they were air dried under a laminar flow hood (LABCONCO purifier class 2 safe cabinet, ca t. no. 36209-000 R) at room temperature. 5.3.2 Preparation of Suspensions of Spores, Nanotitania and Mixtures of Spores and Nanotitania Spore suspension of B. cereus applied on tested surfaces ot her than filter membrane was treated by lysozyme method. Spore suspension applied on filter membrane was treated by ASTM (10-d cultu re) (see Chapter 4 for methodology). A suspension of lysozyme treated Bacillus cereus spores was vortexed for 30 s before application to a surface. The maximum concentration of lyso zyme purified spore suspension was on the order of 106 CFU/mL. To achieve lower spore concentrations, the initial spore suspension was diluted with deionized water an d then vortexed. To achieve higher spore concentrations, aliquots of the original spore suspension was centrifuged at 20,000 rcf for 10 min using an Eppendorf 5417C centrifuge, the supernatants were discarded, and the pellets were combined and vortexed for 30 s.

PAGE 97

79 Nanotitania suspension was prepared by addi ng Degussa P25 titania powder into a sterile flask containing sterile deionized water. The suspension was sonicated for 30 min in an ice water bath sonicator (Cole-Parm er 8890). Another soni catior (Sonicator 3000) was used for the titania suspension applied on the membrane, where a intensity level of 5.0 and a sonication period of 10 min was used . Lower concentrations of nanotitania were obtained by diluting the sonicated suspensi on with deionized water. The suspension was vortexed for 30 s before transferring to a surface or mixing with a spore suspension. A mixture of spores and nanotitania was prepared by adding equal volumes of spore suspension and nanotitania suspension to sterile deionized water. The mixture was vortexed for 30 s before application to a surface. 5.3.3 Suspension Application Methods on Glass, Quartz, Plastic and Modified Glass and Plastic Surfaces 5.3.3.1 Dipping method A surface was immersed in a suspension. For a spore suspension and a mixture of spores and titania, a surfaces was immersed fo r 30 s. For other suspensions, immersion time varied, as noted. Then the surface was taken out using sterile forceps and dried overnight in a laminar flow hood. 5.3.3.2 Pipetting method A volume of 20 L suspension was pipetted onto a surface and dried overnight in a laminar flow hood at room temperature. Susp ension was applied to pl astic cover slips at four different points with 20 L at each point. Suspensi on was applied to all other surfaces at a single point. Due to hydrophobili c property of plastic cover slip, spreading size of droplet on the surface is small. Thus, four droplets were able to be applied on the same surface without overlapping. For th e other surfaces, surf ace area was not large

PAGE 98

80 enough to accommodate four droplets because of the larger droplets spreading size on surfaces than on plastic cover slip. Ther efore, one droplet per surface was used. 5.3.4 Filtration Method Four filter membranes, 0.4 m Metricel membrane (GN-6, Gelman Sciences, Inc., ann Arbor, MI), 0.2 m polycarbonate membrane (Poretics Corporation), 0.22 m nylon membrane (MSI Micron Separations Inc.), 0.02 m Anodisc 25 (Whatman, Fisher Scientific), were tested. All the membrane s have the diameter of 25 mm. The filter membranes were transferred to a vacuum filtration system with a holder for 25 mm filters (Model FH225V, Hoefer Scientific Instrument s, Piscataway, NJ, USA). Suspension was vortexed for 30 s and then added to vacuum f iltration system. Volume of the suspension applied on membrane, as noted, was adjusted ac cording to experimental requirement. If necessary, the suspension was diluted to obt ain a higher filtration volume to achieve a better distribution. The vacuum was then appl ied in order to achiev e even distribution of particles over the membrane surface. Applied vacuum was adjusted and an optimum vacuum pressure was chosen to obtain good di stribution of particles on membrane. After filtration, the membrane was transferred to Petri dish a nd dried under laminar flow hood at room temperature overnight. 5.3.5 Image Analysis The surface with par ticles was imaged under an Olym pus BX60 optical microscope using a 50 objective lens. A SPOT digital cam era (DIAGNOSTIC instruments, Inc. USA) under bright field illumination in tran smission or refractive mode, depending on membranes, was used to capture the image. The image was then digitized by The SPOT Advanced Version 4.0.9 software and analy zed by Image-Pro Plus 4.0/4.1 software

PAGE 99

81 (Media Cybernetics). Spore count based on a 314 235 m sample area was measured and used for evaluation of distributi on quality of spores on the surfaces. At higher magnification, the dried membra ne was imaged under a scanning electron microscope (SEM) (JEOL JSM-6400, MA, USA). Samples were mounted on a metal stub with carbon tape and carbon glue, then they were coated under vacuum with a thin gold film for 30s in the chamber of a ion co ater (Eiko IB-2, EIKO Engineering, Ibaraki, Japan) to avoid charging when imaged under SEM. The image was captured and digitized by imaging software (L ink ISIS, Oxford Instruments, England). Image-Pro Plus 4.0/4.1 software was also used to count the bacterial endospores in SEM images. A total of six SEM images were taken at locations along the diameter of the membrane on each filter sample. 5.3.6 Assessment of Distribution Quality Distribution quality of spores, nanotit ania and mixtures of spores and nanotitaniawas rated on a scale of 3to 1and 1+ to 3+, where 3indicates the poorest quality distribution and 3+ indi cates the highest quality distri bution. The distribution was graded on the basis of visual observation. Examples of grading of distribution quality are given in the results section. Image analysis was used to quantitatively evaluate distribution quality. Counts of spores at different locations on each sample were obtained by image analysis, standard deviation was calculated, which was used as on e of the criteria to evaluate uniformity of distribution of spores on surface. For the mixture, surface coverage by particles were calculated by image analysis at different locations on each sample, standard deviation was obtained and used to evalua te the distribution quality.

PAGE 100

82 5.4 Results and Discussion 5.4.1 Effect of Spore Suspension Drying Process on Germination of Spores Under favorable conditions, spores may ge rminate and outgrow into vegetative cells. Existence of germinating spores and vegetative cells on a su rface is undesirable at the present study because these cells have a mu ch lower resistance to disinfectants than spores. Therefore, investigation on whethe r spores will germinate on the surface during the air drying process was carried out. Two surfaces, plastic and glass cover slip, were tested. Pipetting method was used to apply spores with concentration of 106 CFU/mL to the surface. Spores on the surface were checked twice using phase-contrast op tical microscopy: once immediately after drying of spore suspension on the surface, and again after the spore suspension was air dried overnight. Spore suspension on glass cover slip took 2. 5 hr to dry, whereas longer drying time was required on plastic cover slip (4 hr). No germination was observed on either glass or plastic cover slip when they were examined immediately after drying was complete. Furthermore, germination was not detected af ter allowing the spores remain on cover slip overnight in the laminar flow hood. Thus, it may be concluded that the drying process does not induce spore germination. 5.4.2 Examples of Assessment of Qua lity of Particle Distributions Six examples of the grading used to qualita tively assess the qual ity of distributions of spores, titania and spore/titania mixtures on surfaces are summarized in Table 5-3. For glass, plastic and modified glass and plas tic surface, the suspension was applied on the surfaces by pipetting method. For membrane , the suspension was applied by filtration method. Assessment of distribution quality was based on comprehensive observation of

PAGE 101

83 particle distribution at both high and low magnification. Figu re 5-1 gave the example for particle distribution quality assessment, imag es at both low and high magnification were shown for each sample. Figure 5-1 a) and b) shows distribution of spores on the Anodisc membrane surface at magnification 100 and 1000 respectively, where a volume of 1 mL spore suspension with the concentration of 107 CFU/mL was diluted in 4 mL D.I. water and then the total of 5 mL dilu ted spore suspension was applied onto the membrane. A highest grading of 3+ was gi ven, indicating the best distribution was achieved compared to the other studied surface. A very uniform distribution of spores on the surface was demonstrated in the figure, which was in consis tent with result that a low variance of spore counts analyzed by image an alysis was achieved. A Coefficient of Variance (CV) of only 10% was obtained based on the counts from si x different locations on the membrane. Figure 5-1 c) and d) s hows distribution of a mixture with applied spore concentration 106 CFU/mL and titania concentrati on 0.01% on glass cover slip dip coated with PVA at magnification 50 and 200 respectively. A uniform distribution was observed, but many of small agglomerat es can be easily observed. Thus, it was graded 2+. Figure 5-1 e) and f) is the spor e distribution with app lied concentration of 106 CFU/mL on plastic cover slip coat ed with SDS at magnification 50 and 500 respectively. Distribu tion of spores was not uniform. Spores were less densely packed outside layer of droplet than in side of droplet as seen in image. Dried droplet forms a round shape outline. Agglomeration was obvious in the dense area of droplet. Therefore, a grading of 1+ was given. Figure 5-1 g) a nd h) gives the distribution of a mixture with applied spore concentration of 1.5 105 CFU/mL and titania concentration of 0.0001% on plastic surface at magnification 50 and 500 respectively, which was graded 1-.

PAGE 102

84 Particles are well dispersed, however, a big agglomerate was observe d in the right hand portion of the image. Figure 5-1 i) and j) is the mixture of spore a nd titania with applied spore concentration of 1.5 104 CFU/mL and titania concentration of 0.0001% on plastic surface at magnification 50 and 500 respectively with graded 2-. As we can see, spread area of droplet is obvious smaller th an the above samples. Distribution of particles varies significantly from area to area, aggregates exists both inside the droplet and at outer edge. Particle distribution is worse than the above samples based on distribution uniformity, spread si ze of droplet, thus, a grading of 2was given. Figure 51 k) and l) shows the distri bution of a mixture (spore, 106 CFU/mL; titania, 0.01%) on plastic cover slip coated with SDS and rinse at magnification 50 and 500 respectively, with a grading of 3-, indicat ing the worst distribution obser ved in our study. A large aggregate was observed in the center of image, heavier agglomeration observed at the left portion of the big aggregate. There are so me particles dispersed well outside the big aggregate. However, considering the big aggr egate, we evaluate the particle distribution was the worst. 5.4.3 Spore or Mixtures of Spore and Ti tania Distribution on Glass Surfaces Spore and mixture suspension was applied on the glass surface by pipetting method. Figure 5-2 a) and b) shows spore distri bution on glass surface with applied spore concentration of 106 CFU/mL. Figure 5-2 c) and d) shows titania distribution on glass surface with applied concentration of 0.01. Figure 5-2 e) a nd f) shows mixture distribution on glass surface with ap plied spore concentration of 1.3 106CFU/mL and titania concentration of 0.01%. A uniform distribution was observed on glass surface for both spore and titania distribution with some a gglomeration. A grading of 2+ was given.

PAGE 103

85 A lot of agglomeration was observed on mixt ure distribution, 1+ was graded for the distribution. Distribution of titania on glass surface was also studied to help understand the titania concentration effect on their distribution. A volume 20 L titania suspensions with concentration from 0.0001% to 0.1%were applied onto glass c over slip by pipetting method. An optimum dispersion was observed at titania concentra tion with 0.01% as shown in Figure 5-2 c). Titania particles di stributed uniformly. A grading of 2+ was given. However, distribution of titania par ticles at other concentr ation was not uniform. Very irregular distribution of nanotitan ia was observed at the applied titania concentration of 0.0001% and 0. 001% with strips formed by p iled up particles and little or no particles in some area. Heavy agglomer ation was observed for the titania applied at concentration of 0.1%. A grading of 2was given. Concentration of titania applied on the glass surface demonstrated a significan t effect on their distribution on the glass surface. 5.4.4 Spore or Mixtures of Spore and Titani a Distribution on Quartz Surfaces Similar distribution quality of spores or mixture on quartz surface was observed as on glass surface (a grading of 2+ was given). It is reasonable because quartz and glass has similar surface property. 5.4.5 Spore or Mixtures of Spore and Titani a Distribution on Plastic Surfaces 5.4.5.1 Effect of spore concentratio n on distribution of spores Concentration of spore suspension applied on the surface was considered to be an important factor affecting disp ersion of spores in dry state on the plastic cover slip. In our present study, spore suspension with concentration at the level of 106, 105, 104, 103 CFU/mL were applied on the plastic cover sl ip by pipetting method. A comparison of

PAGE 104

86 spore dispersion for different a pplied spore concentrations is summarized in Table 5-4. Spore suspension with a concentration level at 105 and 106 CFU/mL demonstrated a better dispersion than the lower concentration (104 and 103 CFU/mL). Spores were distributed uniform except the agglomerati on at outer edge with spore concentration applied at 106 CFU/mL as shown in Figure 5-3 a), which turned out to be the optimum spore distribution on plastic c over slip among the evaluated spore concentration. At low concentration of 104 and 103 CFU/mL, it was hard to tell spores and agglomeration built up in the middle as well as outer edge. Dropl et size with applied higher concentration at 105 and 106 CFU/mL is larger than 104 and 103 CFU/mL. Distribution area of spores were large at a high ap plied concentration (105 and 106 CFU/mL), whereas spores were concentrated in small area when a lower concentration (104 and 103 CFU/mL) was applied. Distribution of spor es on plastic cover slip is ge nerally not uniform, which is demonstrated by a high CV of 53% counts from three different locations by image analysis on plastic cover slip applied with 105 CFU/mL. In some cases, droplets were applied on plas tic cover slip on two successive days. First, spore suspension w ith concentration of 106 CFU/mL was pipetted onto a plastic coverslip and allowed to dry overnight unde r a laminar flow hood. Then other spore suspensions with concentration of 106 or 105 CFU/mL were transferred to the area where the first droplet was and dried overnight. It was hypothesized that th e variance of spore distribution on the first droplet could be improved by applyi ng the second droplet on the dried first droplet. Thus, this approach wa s tested to check the distribution quality. Figures 5.3 b) shows spore dist ribution applied with two droplets daily on the same area with the second droplet concentration of 106. As shown in Figure 5-3, two droplets with

PAGE 105

87 concentration of 106 CFU/mL did not perfectly overlap , and addition did not improve the uniformity of the spore dispersion. Same phenomena were observed with the second applied droplet with a concentration of 105 CFU/mL shows the same phenomenon as Figure 5-17. Thus, addition of the second dr oplet does not improve spore dispersion on the plastic surface. The optimum dispersion of spores on plas tic cover slips was achieved at applied spore suspension concentrations of 106 CFU/mL (Table 5-4). However, there was spore agglomeration at the outer edge of the droplet. Appling a second droplet to the first droplet was expected to compensate the im balance of spore distribution density on the surface, but even worse agglomeration was obs erved. As demonstrated in the figures, concentrations of spore suspension applied on the surface is an importa nt factor affecting dispersion of spores in dry state on the plastic cover slip. 5.4.5.2 Effect of spore application me thod on distribution of spores Dipping. Plastic cover slip was dipped into sp ore suspension at a concentration of 105, 104, 103 CFU/mL respectively. Figure 5.4 s hows spore distribution on the plastic cover slip by dipping method with applied spore concentration of 105 CFU/mL. A very small amount of spores spore particles are observed on the surface. Spores were well distributed without agglomerati on. For the spore di stribution at the di pping concentration of 104 CFU/mL, only one or two spores can be detected from each image taken under microscopy at magnification of 500 . For plastic cover slip dipped into 8.5 102 CFU/mL spore suspension, no spores were observed on the plastic surface. Spores were very well dispersed by dippi ng method, no agglomeration exists on the plastic surface. However, number of spores on the surface is very low under the

PAGE 106

88 condition used. Longer dipping time or highe r spore suspension concentration may give higher surface concentration. Pipetting. Agglomeration can be observed with in all tested spore concentration (106 to 103 CFU/mL) applied on plastic surface by pipetting method. Even at the optimum concentration of 106 CFU/mL where best spor e distribution quality was observed, agglomeration still existed at the outer edge. Compared to the dipping method, spores applied on the surface by pipetting (Fig.54) are more concentrated. Spore dispersion ob tained by pipetting is generally worse than by dipping method and are more prone to agglomeration. 5.4.5.3 Effect of surface temperatur e on distribution of spores In our experiment, it was found when spor e suspension was pipetted to a glass cover slip surface right after the surface is preheated at alcohol burner flame (Fisher Scientific), contact angle of droplet on the glass surface d ecreased and spreading area of droplet increased dramatically compared with suspension applied on the glass cover slip without heating. Therefore, temperature of the surface may be a f actor affecting spore dispersion on the surface. Our hypothesis is th at increase surface temperature of plastic cover slip could decrease cont act angle of spore suspension and increase spreading area of spore suspension, thus, improve spore dispersion on the surface. Plastic cover slips were placed into incubato r (Fisher Scientific) at a temperature of 60 1 C and 80 1 C for 2h, respectively. Spore susp ension with a concentration of 106 CFU/mL was applied on the plastic cover slip by pipetting method im mediately after the slips were taken out of inc ubator and put in laminar flow hood. The suspension was dried overnight at a laminar flow hood. Control was prepared by pipetting spore

PAGE 107

89 suspension on plastic cover sl ip in laminar flow hood at room temperature and dried overnight. Two samples for each condition were prepared. Droplet size on plastic cover slip pre-heated at 60 C is bigger than control, but distribution of spores was not uniform, aggregates can be detected, a grading of 1was given. A worse distribution of spores was found on plastic cover which was pre-heated at 80 C (a grading of 2) compared with cover slip preheated at 60 C. An agglomeration was observed in the middle of image and very un-uniform distribu tion of spores on the surface. Temperatures higher than 80 C could not be tested since they deformed the plastic cover slip. No improvement of spore dispersion on the plastic surface was observed at high temperature. Agglomerates still existed and distributions of spores were not uniform. Therefore, temperature was not a main factor affecting spore disp ersion on plastic cover slip. 5.4.5.4 Effect of titania concentrat ion on distribution of titania As found in section 5.4.5 a), spore con centration influence spore distribution on plastic surface. It was hypothesized that tit ania concentration will also influence their distribution on surfaces. Theref ore, effect of titania concen tration on their distribution over plastic cover slip was investigated. A volume 20 L titania suspensions with concen tration from 0.000,01% to 0.01% were applied onto plastic cover slip by pipet ting method. Bad distribution of titania on the surface was observed, a grade of 2to 3was given. For titania distribution with applied concentration of 0.000,01%, dried droplet forms an irregular outline and distribution was not uniform. For titania di stribution with applied concentration of

PAGE 108

90 0.0001%, a better distribution was observed than titania distribution with applied concentration of 0.000,01%. Similar droplet sized was observed. Still distribution of titania on the surface was not uniform. Dried dr oplet forms a circle outline. For titania distribution with applied concen tration of 0.001%, aggregates can be observed on the side of the image. Spore distribution was not unifo rm. Droplet size is larger than the above image at a lower titania concentration. For titania distribution with applied concentration of 0.01%, a heavy agglomeration was observed at the outer edge and portion of the image. As a result, agglomeration can be observed at all the applied titania concentration level, distribution of particle s were not uniform. Dispersion of titania on the glass cover slip is much better than on plastic cover slip. Observed agglomeration at different applied titania concentration is due to higher surface area of nanostructure of Degussa P-25 titania, which has higher potential to agglomerate. 5.4.5.5 Effect of titania concentration on distribution of spore/titania mixtures Both titania and spore concentrations a ffect their distribution on surfaces. The effect of titania concentration on the distribut ion of spore/titania mixtures on surfaces is discussed in the present section. The eff ect of spore concentration is presented in the following section. A 20 L volume of the mixture was pipetted onto a plastic cover slip and dried under laminar flow hood overnight. Table 5-5 shows five combinations of titania and spores and their distribution quality. None of the above combinations achieved uniform distribution; in addition, agglom eration was observed at all the levels. Grades from 1to 3were given. For the mixture with an applied spore concentration of 104 CFU/mL and titania concentration of 0.0001%, agglomerati on occurred both inside and along the outer

PAGE 109

91 edge of the droplet. For the mixture with applied spore concentration of 104 CFU/mL and titania concentration of 0.001%, same distribution quality was observed as the mixture with applied sp ore concentration of 104 CFU/mL and titania concentration of 0.0001%. Agglomeration still existed both inside and outer edge of the droplet. Mixture with applied spore concentration of 105 CFU/mL and titania concentration of 0.0001% achieved a better distribution than the above two combinations. But aggregates were observed at the side of image. Mixtur e with applied spore concentration of 105 CFU/mL and titania concentration of 0.001%, aggregates were observed at thecorner of image. For mixture with applied spore concentration of 106 CFU/mL and titania concentration of 0.01%, heavy agglomeration was obser ved in the middle of droplet. Dispersion of the five combin ations on plastic cover slip is not satisfactory. For a given spore concentration, increasing the t itania concentration makes the distribution less uniform. 5.4.5.6 Effect of spore concentration on di stribution of spore/titania mixtures Mixture with final spore concentration of 105 CFU/mL shows better dispersion than final spore at 104 CFU/mL. Same trend was observed at spore dispersion on the plastic cover slip with increased spore concentra tion when spores alone were applied on the surface. Increasing spore concentrations improve s dispersion of the mixture. Effect of spore concentration on mixture di stribution is more significan t than titania concentration on mixture. 5.4.6 Spore or Mixtures of Spore and Titani a Distribution on Modified Glass and Plastic Surfaces Our previous results demonstrated differe nt dispersion quality of particles at different surfaces. Therefore, surface was m odified to improve particles distribution.

PAGE 110

92 Pre-coating was to apply the coating on the su rface before transferring particles to the surface. 5.4.6.1 Glass cover slips Pre-coated with PVA PVA solutions were obtained from Sigma (P8136-250G) and Celanese, Ltd. (Celvol 165, Unisize HA-70, Dallas, TX). PVA was prepared by adding a measured amount PVA powder into deionized water a nd then the suspension was autoclaved. Dissolution of PVA in water was achieved by autoclaving the mixt ure of PVA powder and deionized water. Particles were a pplied on the surface by pipetting method. Dipping method. Our hypothesis is that making PVA pre-coating thinner would favor spore or mixture distribut ion on PVA pre-coated glass cover slip. The thinning of PVA coating was achieved by rinsing the glass cover slip immediately after it was taken out from PVA solution, then dr y the glass cover slip to fini sh coating. Distribution of spores or mixture on rinsed PVA pre-coated gla ss cover slip is expected to be better than no rinsed PVA pre-coated glass cover slip. Washed glass cover slip was dipped into st erile PVA solution for a period from 5 s to 10 min, then removed with sterile metal forceps and placed in plastic Petri dish. The pre-coated glass cover slip was air dried overn ight in laminar flow hood. The thinning of PVA coating was achieved by rinsing the glass cover slip immediately after it was taken out from PVA solution, then dry glass cover slip to finish coating. Spore suspension with concentration of 106 CFU/mL or mixture of spore and ti tania with final concentration of 106 CFU/mL and 0.01% respectively was applied on the surface. Distribution of spores and mixture on glass cover slip without coating is used as control. Mixtures distribution on the following P VA pre-coated glass cover slip were compared: glass cover slip dip coated in 5% PVA (Sigma) for 10 min without rinse,

PAGE 111

93 glass cover slip dip coated in 5% PVA (Sig ma) for 10 min and subsequently rinsed by deionized water and glass cover slip. The be st distribution was observed on the surface with subsequent rinse (Figure 5-5). Mixt ure on PVA (Sigma) precoating without rinse shows worst dispersion. This observation is in consistent with our hypothesis. Distribution of mixture is uni form with some small aggl omeration observed at higher magnification. A grade of 2+ was given. Dr ied droplet size on the control surface is the largest, whereas on PVA pre-coating su rface without rinse is the smallest. Distribution of spore distri bution or mixture of spores and titania on glass cover slip pre-coated in 5% PVA (Celvol 165) by dipping method was also studied. Both spore and mixture distribution on the su rface was not uniform. A con centrated dot at the center of the dried droplet was observed on both s pore and mixture distribution images, area around the concentrated center showed good distribution for mixture but bad for spores. A grade of 1was given to mixt ure and 2was given to spores. Mixture on glass cover slip coated with PVA (5% Sigma for 10 min) pre-coating (with rinse) demonstrates the bette dispersi on than on PVA (5% Celvol 165) pre-coated glass cover slip. This may be caused by th e change of surface property after PVA precoating and different proper ty of PVA pre-coating. Spreading method. Distribution of spores and mixtures on nine different combinations of PVA pre-coating by spr eading method were shown in Table 5-6. Sufficient volume of PVA solution was applied to a glass cover slip and spread by sliding an edge of a second glass cover slip from one si de of the first glass c over slip to the other side. The coating was then allowed to dry overnight under a laminar flow hood. A volume of 20 L spore suspension (106 CFU/mL) or mixture of spore and titania

PAGE 112

94 suspension with final concentration of 106 CFU/mL and 0.01% was pipetted on the different PVA pre-coated surface and dried overnight. All dry droplets on PVA precoated glass cover slip are circle shape, wh ereas the droplet shape is irregular on glass cover slip. Based on visual observation, uniform distribution wa s observed on coating with 10% and 15% Celvol 165, 15% Sigma. A very uniform spore distributi on was obtained for spores (106 CFU/mL) applied on glass cover slip spread pre-coated with 10% Celvol 165 PVA. Figure 5-6 a) and b) show spore distribution on 10% Celvol 165 PVA spread precoated glass cover slip. Very few aggregates can be detected. A grad e of 2+ was given. Distribution of mixture (spore 106 CFU/mL, titania 0.01%) on glass cover sl ip spread coated with 10% Celvol 165 PVA were good but with some obvious aggreg ates, thus, a grade of 1+ was given. Spore (106 CFU/mL) on glass cover slip spr ead pre-coated with 15% Celvol 165 PVA were distributed well but with some aggreg ates. A grade of 1+ was given. For the mixture (spore 106 CFU/mL, titania 0.01%) on 15% Cel vol 165 PVA pre-coated glass cover slip, many aggregates a ll over the surface can be dete cted although distribution is good. A grade of 1was given. A good distribution of particles was obtained for the mixture (spore 106 CFU/mL, titania 0.01%) applied on glass cover slip spread pre-coated with 15% Sigma PVA. Little agglomeration was detected (Figure 5-6 c and d). A grade of 2+ was given. However, for spore distribution on 15% Sigm a pre-coated glass cover slip , it was hard to detect the spore under optical microscopy.An interesti ng phenomenon was observed when the spore suspension was pipetted onto the glass cover s lip pre-coated with Celvol 165 PVA. The contact angle of the droplet on 5% Celvol 165 PVA pre-coated glass cover slip (dipping

PAGE 113

95 method) is the largest, and on non-coating gla ss cover slip is the smallest. The contact angle of the droplet on 10% Celvol 165 P VA pre-coated glass cover slip (spreading method) was intermediate. A relation was found for the dried droplet size with PVA coating. For Celvol 165 coating, droplet size increase with increas ed PVA concentration. 15% Celvol 165 coating has largest droplet size within Celvol pr e-coating. For Sigma coating, the trend is on the opposite way, 5% Sigma PVA pre-co ating has biggest droplet size. In terms of uniformity, droplet size, 10% Celvol 165 was found to be an optimum pre-coating agent. Therefore, glass cover slip pre-coated with PVA affects particles distribution on their surface compar ed with on glass cover slip. 5.4.6.2 Glass cover slips Pre-coated with PVA and SDS Enough volume of 10% Celvol 165 PVA was a pplied to the surface of a glass cover slip and spread with a sterile plastic inoculation loop to even ly coat the cover slip. The coated cover slip was dried for 80 min at 65 C. A volume of 0.1 mL 0.05% SDS was then spread over the cover slip using a sterile plastic inocul ation loop. The coated cover slip was dried overnight under a lamina r flow hood at room temperature. Spore (106 CFU/mL) distribution on glass cove r slip pre-coated with PVA/SDS was not uniform. A very bad distribution wa s observed with heavy agglomeration both at the outer edge and inside. A grade of 3was given. SDS may partially dissolve the PVA pre-coating when the SDS suspension was a pplied onto PVA pre-coating surface, which increased the heterogeneity of the m odified surface. Mixture (spore 106 CFU/mL, titania 0.01%) distribution on glass cover slip precoated with PVA/SDS was dispersed well, with aggregated in the middle. A denser area was in the center, area near the edge is less

PAGE 114

96 concentrated. But in both areas, a good unifo rmity was obtained. Dispersion of mixture is better than spores on th e PVA/SDS pre-coated glass c over slip, a grade of 1+ was given. 5.4.6.3 Glass Pre-sputter coated by a platinum/gold mixture Dry glass cover slips were surface modified by pre-sputter coating for 45 sec in the chamber of a Desk II denton vacuum spu ttering system (Denton Vacuum Inc., Moorestown, NJ, Model No. Carb on Rod Acc) prior to application of suspensions. Subsequently, the spores or mixture of s pores and titania was applied on the coated surface and dried. The surface w ith dried spore, nanotitania or spore/titania was imaged under SEM directly. Dry glass cover slips with spores, titatin a or spore/titania mixture dried on their surface was post-sputter coated 30 sec for im aging by SEM. Post-coating was to apply the coating on the surfa ce after transferring part icles to the surface. Using of post-coating is to avoid charging problem under SEM. Figure 5-7 is SEM image of spore distributi on on pre-sputter coated glass cover slip and control (post-sputte r coated glass cover slip). A better spore distribution on sputter coated surface was observed than control. But agglomeration can be observed at high magnification. Thus, a grade of 2+ was give n. Distribution of mixture on pre-sputter coated glass cover slip was same as on contro l. Although the distribut ion of particles was good, agglomeration can be observed all over the surface. No improvement was observed with pre-sputter coating. A grade of 1was given. 5.4.6.4 Plastic cover slips Pre-coated with SDS Pre-coating plastic cover slip with SDS is expected to improve dispersion of spores or mixture on the plastic cover slip due to property of SDS. Hydrophobic tail of SDS

PAGE 115

97 would stay on the surface of pl astic cover slip, while polar ta il of SDS would points to the air. Thus, surface of coated plastic cover sl ip is hydrophilic, which favors distribution of suspension. Therefore, distribution of spores on SDS pr e-coated plastic cover slip was tested. Concentration of SDS and dipping time of plas tic cover slip in SDS solution was thought to be major coating parameters. A volume 20 L spore suspension with a concentration of 106 CFU/mL was applied on the coated surface by pipetting method. Mixture of spore a nd titania with final concentration of 106 CFU/mL and 0.01% respectively on plastic cover slip pre-coated with 10% SDS for 10 min. No rinse. Table 5-7 shows spore distribution quality on SDS pre-coated plastic cover slip, where SDS with different concen tration and dipping time was used for precoating. Plastic cover slips were dipped into autoclaved SDS solution with a concentration of 0.05% to 10%. The imme rsion time ranged between 5 seconds and 10 min. The plastic cover slips were then ta ken out from the SDS solution and placed in a sterile plastic Petri dish. Th e surfaces were air dried at r oom temperature in a laminar flow hood. However, improvement of spor e dispersion by SDS pre-coating to plastic cover slip is not obvious compared with s pore dispersion on plastic cover slip without coating. Dispersion of spore on plastic cove r slip coated with 0.1% SDS for 30 s and 10 min and 10% SDS for 5 and 10 min turned out to be optimum. Dist ribution performance of spores on the above optimum coating condi tions was same as shown in Figure 5-4. Spores are well distributed with some aggl omeration. A grade of 1+ was given.

PAGE 116

98 Interestingly, mixture of spore and ti tania with final concentration of 106 CFU/mL and 0.01% respectively on plastic cover slip co ated with 10% SDS for 10 min (Figure 5-8) showed a better distribution than spores al one applied on the same surface (Figure 5-1 e and f). Particles were distributed uniformly as shown in Figure 5-8. Droplet spread more on plastic cover slip coated with 10% SDS fo r 10 min than on plastic cover slip without coating as compared Figure 58 with Figure 5-1 e) and f). Mixture distribution on plastic cover slip can be improved by SDS pre-coating on plastic cover slip. With rinse. Plastic cover slips were dipped into SDS solutio n as before and then rinsed with D.I. water before being placed in the Petri dish and air dried. The purpose to rinse SDS is to reduce the h eavy residue of SDS on the plas tic cover slip after dry. Distribution of mixture on pl astic cover slip pre-coated with 10% SDS for 10 min followed by rinse was observed. Particles distribution was not uniform, with heavy aggregates at the corner of the image. W ith the rinse, dispersion of mixture becomes worse than on plastic cover slip withou t coating. A grade of 3was given. As a result, mixture dispersion on SDS pr e-coating without rinse is much better than SDS coating with rinse. This may be caused by too much removal of SDS on the plastic surface in the process of rinse. Surf ace property for pre-coati ng with rinse is close to plastic, where a bad mixture distribution was found before, thus, a worse distribution for mixture was observed on pre-co ating with rinse than the pr e-coating without rinse. 5.4.7 Spore or Mixtures of Spore and Tita nia Distribution on Filter Membrane Applied by Filtration Method 5.4.7.1 Preliminary investigation on fact ors affecting spore distribution on membranes Filter membrane type. Three types of membrane, GN-6 Gelman, polycarbonate and MSI, were first tried to test the perfor mance of spore distribu tion on the surfaces. A

PAGE 117

99 volume of 1 mL spores suspensi on with concentration of 1.4 107 CFU/mL was filtered on each type of membrane and dried overn ight. Each sample was duplicate. Distributions of spores on the membrane we re checked under optical microscopy. Spores cannot be detected on all the three types of membrane under the optical microscope. It was hypothesized that a too low total spore amount on the membrane contributed the difficulty of detection. To test this hypothesis, a vol ume of 10 mL spore suspension was applied to achieve a total amount of 1.4 108 CFU spores on the membrane. Only spores on polycarbonate membranes can be dete cted when the three types of membranes were checked under optical microscopy. Thus, the observation didn’t support the hypothesis, that is, too less spores on the memb rane. Also, if spore distributed uniformly, there will be an average of one spore per 2 2 m area calculated on the basis of the applied spore amount. Since spores were sphe rical or oval in shape with an average size of 1 m, the applied amount should be large eno ugh to be detected. Thus, the hypothesis was rejected. Another possible explanation is that th e structures of membrane cause the difference of spore distribution on the surf ace of membrane. Membrane from Gelman and MSI is made of metricel and nylon re spectively. Although enough amount spores have been applied onto the membrane in or der to be detected under microscopy, spores may pass through the surface of the membrane and be captured within the membrane instead on the surface because their interwoven structure. For polycarbonate membrane, the pores were etched by laser, spores were ab le to be detent at the surface of membrane. Because the spores were able to be tra pped on the surface of polycarbonate membrane, they can be detected under microscope othe r than the other two membrane surfaces.

PAGE 118

100 Since membrane structure plays important role on spore distribution on the surfaces, another type membrane with a honeydew stru cture membrane, Anodisc 25, were tested also. As calculated, a volume of 5 mL s pore suspension was required for filtration to obtain one spore per 9 m2. Thus, a volume of 5 mL s pore suspension (total spore amount 7 107CFU) was filtered through Anodisc 25 and dried overnight. Sample was imaged under both optical microscopy and SEM. When the sample was analyzed under SEM directly, image was blurry due to the surface charging. Thus, sample was postcoated with platinum/gold before SEM imag ing to avoid charging. Figure 5-9 shows SEM image of spore distribution on Anodisc 25 at 100 and 1500 magnification respectively. A uniform distribution was achieved on the surface. An optimum spore distribution was obse rved on the Anodisc membrane among the four alternative membranes. Polycarbonate membrane was the medium, while MSI and Gelman membrane demonstrated the worst di stribution (no spore obs erved on surface). Membrane structure was found to be the main factors contribute to spore distribution on the membrane surfaces. Comparison of spore distribution on membranes with modified condition. In order to further evaluate spore distributi on on the membrane surface, a 1 mL filtration volume of spore suspension (1.4 107 CFU/mL) was used to obtain a lower amount of spores on the surface than observed with 5 mL filtration volume (Fig. 5-9). Another two membranes were tested at the same time, one is polycarbonate because it was expected a good image can be obtained under optical microscope due to its flat surface, one is gold pre-coated Anodisc 25 membrane which was reported to achieve a good distribution for spore dispersion on its surface. For the gold pre-coated Anodisc membrane, the

PAGE 119

101 membrane was sputter coated for 30 s in the chamber of a Desk II denton vacuum sputtering system (Denton Vacuum Inc., M oorestown, NJ, Model No. Carb on Rod Acc) prior to application of suspensions. Subse quently, the surface was filtered with spore suspension and dried overnight, then the sample was ready for image analysis. Duplicate samples were prepared on each type of memb rane surface, one is for optical microscopy analysis, and the other is for SEM analysis. Figure 5-10 shows optical image of spore distribution on Anodisc , gold pre-coated Anodisc and polycarbonate membrane at magnification of 50 and 500 . Optical image for spores on Anodisc and gold pre-coated Anodisc membrane was taken under bright field using transmission light; image for spor es on polycarbonate was taken under bright field using reflective light. At a higher magnification of 500 , a uniform spore distribution was observed on A nodisc and gold pre-coated Anodisc membrane, whereas spore distribution on polycarbona te didn’t show uniform di stribution instead with a certain pattern. At a lower magnification of 50 , it can also be found that spores on Anodisc, gold pre-coated Anodisc achieved uniform distribution. However, some blank area on Anodisc membrane and very less blank area on gold pre-coated Anodisc membrane were observed. This maybe caused by low filtration volume (1 mL). It was observed that a volume of 1 mL spore susp ension was not enough to cover the whole membrane when the spore suspension was filtered. Thus, some areas may be lack of spores. Two modifications to the filtration cond ition were made to improve this defect in the following filtration. One is to dilute s pore suspension to a total volume of 4 mL or 5 mL from 1 mL original spore suspension to ensure the whole membrane area was covered under filtration. Second is to imme rse the filter membranes into sterile D.I.

PAGE 120

102 water for 5 sec to wet the membrane surface to avoid blank spot. Spore distribution on polycarbonate membrane shows the pattern whic h is in accordance with the filter holder structure as shown in Fig. 5.10 e) and f). S pores exist at the places of filter holes where the support drained the water. This may be caused by material property of polycarbonate membrane. Due to its flexibility and fl at surface nature, polycarbonate membrane attached to the filter support under vacuum pressure. Thus, no spore suspension can be drained elsewhere except at holes. Spores were mostly detained at the location of filter holes. This phenomenon was not observed on Anodisc membrane. Because the Anodisc membrane surface is not flat and the materi al is stiff, water can go though all over the membrane. Thus, a pad was tried under pol ycarbonate filter membrane when spore suspension was filtered in order to improve spore distribution in the following study. Two types of pad: Anodisc 25 and separator coming with the polycarbonate membrane in the box. The pad was placed under the polycarbonate membrane to make sure all the spore suspension can drain through the whol e membrane, thus, to achieve a uniform spore distribution on the membrane. Even dist ribution of spores was not observed with either Anodisc 25 or separate as pad. Di stribution was not impr oved by application of pad. However, filtration time was increased wh en separator was used as pad, a filtration time of 20 min was required to drain all the water. SEM images of spore distribution on Anodisc, gold pre-coated Anodisc and polycarbonate membrane were also taken. All the samples were sputter coated with Pt/Au for 30 s in the chamber of a Desk II denton vacuum sputtering system to avoid charging issue under SEM analysis. Perf ormance of spore distribution on three alternative membrane surfaces is consistent with observation from optical microscopy.

PAGE 121

103 Uniform spore distribution was observed on Anodisc 25 membrane and gold coated Anodisc 25 membrane. Spore distribution on polycarbonat e membrane shows the pattern which is in accordance with the filter holder structure. However, spore density (number of spores per area) on Anodisc membrane is lower than on gold pre-coated Anodisc membrane observed SEM image which is not observed under optical microscopy. This may be caused by the error when preparing the sample as found in the following images that a similar spore density was achieved. Quantitative comparison of sp ore distribution on membranes. It has been qualitatively compared in th e previous section that s pores on Anodisc and gold precoated Anodisc membrane can achieve the be st distribution. Th erefore, Anodisc was chosen as membrane surface for spore distribution. Although a uniform spore distribution can also be achie ved on gold pre-coated Anodisc membrane, more steps will be involved to prepare sample , thus, Anodisc was used wher e the same spore distribution performance can be achieved. In order to further prove spores on Anodisc can achieve the best distribution, spore distribution on A nodisc, polycarbonate with Anodisc as pad, polycarbonate with separator as pad were qua ntitatively compared. A volume of 5 mL spore suspension with total amount of 1.2 107 CFU spores were applied on the membrane surface. Samples were duplicat e on each membrane except the polycarbonate with separator as pad. For each sample, six images at six different locations on the membrane were taken using SEM. Numbers of spores were calculated by image analysis software. Average spore count was calculat ed based on the numbers obtained from six locations.

PAGE 122

104 Table 5-8 shows the count result of poly carbonate membrane with Anodisc 25 and separater as pad. A high CV ranged from 30 to 85% was obtained, which means a high variation of spore numbers at different location of the membrane. Thus, spore distribution on the polycarbonate membrane with either Anodisc 25 or separater as filtration pad was not uniform. This result is consistent with our observation that spores were concentrated in some area while th ere is little spores in other area. Table 5-9 shows count result from Anodisc membrane. Sample SEM images were taken at 1000 magnification. A relative low CV (10-38%) was achieved than spore distribution on polycarbonate memb rane with pad. This is also in consistent with the image shown in Figure 5.9 a). Therefore, a uniform distribution of spores was achieved on Anodisc membrane by filtration method. From quantitative comparison, together qualitative assessment from the spore distribution images, distributi on of spores on Anodisc membrane demonstrated a very uniform distribution, whereas spore distribution showed a bad distribution quality (a grade of 2). The quantitative results were consistent with the visual observation for the spore distribution quality. 5.4.7.2 Mixtures of spore and titania distribution on Anodisc membrane Since a uniform spore distribution was achieved on Anodisc membrane, distribution of mixture of spor es and titania was investigat ed. 0.1% titania suspension was prepared by adding 0.01g titania powder in to sterile 20 mL vials containing 10 mL sterile D.I. water. The titania suspension was sonicated for 10 min at intensity of 5.0 using Sonicator 3000. A certain amount of tit ania suspension and spore suspension was added to 150 mL Pyrex beaker containing ster ile D.I. water. The mixture of titania and

PAGE 123

105 spores are stirred on a magnetic stirrer (Nuova II, Thermolyne) with the mixing speed = #5, a volume of 4 mL mixture was pipetted onto the membrane for filtration using Brinkmann Eppendor Repeater Pipetter (Fisher Sc ientific). After filtr ation, samples were dried under laminar flow hood. Before SEM an alysis, samples were post-sputter coated with gold to avoid charging under the microscopy. Three mixture combinations based on same spore to titania ratio were applied on Anodisc membrane and their distributi on quality was observed. They are: 106 CFU for spores and 0.1 mg for titania, 105 CFU for spores and 0.01 mg for titania, 104 CFU for spores and 0.001 mg for titania. Figure 5.11 shows image of mixture distribution of the three combinations at magnification of 5,000 . A uniform distribution can be observed on all these three combinations. The surface was almost fully covered by the mixture for the combination of 106 CFU spores and 0.1 mg titania. Highest surface coverage by the mixture was observed for the combination. It was reasonable because the combination applied largest amount of spor es and titania on the membrane . Mixture applied with 105 CFU spores and 0.01 mg titania has medium surface coverage. While only few particles were detected for the combination of 104 CFU spores and 0.001 mg titania on each image. Another combination with 5 106 CFU spores and 0.05 mg for titania was tested. Distribution of the mixture on the membrane was shown in Figure 5-12. The distribution of mixture in this combination was worse th an the above three combinations. This may be affected by the spore to titania ratio. Application of mixtures of spores and titania direc tly on the Anodisc membrane demonstrated a good distribution. Another sa mple preparing proce dure was tried and the

PAGE 124

106 mixture distribution quality on the Anodisc memb rane was observed. This method is to apply the titania particles on the membrane fi rst followed by application of spores on the membrane instead of applying mixture dir ectly on the membrane. A volume of 4 mL titania suspension with 0.01 mg titania was first applied on the Anodisc membrane and filtered, then a volume of 4 mL spore suspension with a total of 5 106 CFU spores was applied and filtered. Membrane was dried over night under laminar flow hood. Sample was post-sputter coated with gold, and then image was checked under SEM. A good distribution was observed. Titania agglomera tion can be observed. Spores are dispersed on the titania particles. A dist ribution grade of 2+ was given. Good distribution of mixture of spores and titania on Anodisc membrane can be achieved by either directly application of mi xture or application of titania first followed by spores. Spore to titania ratio may be an important factor affec ting mixture distribution uniformity on the Anodisc membrane when mixture was applied directly on the membrane. 5.4.8 Comparison of Spore or Mixture Di stribution on Glass Surface with an Anodisc Filter Membrane Surface Distribution of spores or mixture on th e glass surface was found to be the best among the alternative surfaces ot her than membrane surface. For the membrane surface, best spore or mixture distribution was observed on Anodisc membrane surface. Therefore, the spore or mixture distribut ion on these two surfaces was compared. For spore distribution on the surface s, their performance was compared quantitatively. For each sample, 6 or 4 diffe rent locations on each sample were imaged and amount of spores on each image were calculated by image analysis. Then the average spore count on each image was obt ained, standard de viation and CV was

PAGE 125

107 compared for the two different surfaces. Table 5-10 shows average count of spores on glass cover slip from images taken un der optical microscopy at magnification 500 . Amount of spores on the glass was counted by visual observation because image analysis cannot tell all the individual s pores in the agglomeration. Compared with results for spore count on Anodisc membrane in Table 5-9, there is no statistic significance for variance of count between glass cover slip and Anodisc membrane based on F test at =0.05. However, less agglomeration of spores was observed on Anodisc membrane than on glass cover slip. Th erefore, spore distribution on Anodisc membrane is better than on glass cover slip. A better distribution of mixture was also observed on Anodisc membrane than on glass surface from SEM images. In order to quantitatively compare the distribution quality of mixture on these two surfaces, surface coverage by the mixture on the two surfaces were calculated, standard deviati on and CV was shown in the Table 5-11. Mixture combination demonstrated optimum distribution on individual surfaces was chosen to apply on the surfaces. For the membrane, a total amount of 0.01 mg and 105 CFU spores were applied on the membrane su rface. Images were taken using SEM at magnification 500. A total of five random lo cations on the membrane were imaged and surface coverage was analyzed using image an alysis software. For the glass surface, a total amount of 0.002 mg and 24 CFU spores were pipetted on the surface. Images were taken using optical microscopy magni fication 500. A total of eight random locations on the membrane were imaged and surface coverage was analyzed using image analysis software. Although a lower CV obt ained from Anodisc membrane surface, their variance on the surface coverage was not statistically difference based on F-test at =0.05.

PAGE 126

108 In other words, distribution quality of mixture on membrane and glass are same. Therefore, both spore and mixture distri bution on Anodisc surface demonstrated an optimum performance than on glass cover slip. 5.5 Summary Since our study is to investigate spore in activation by photocatalys t, distribution of spore and mixture is our concern. Thus, Nanotitania distribution was only checked on glass/plastic surface without modification. For other surfaces (m odified glass/plastic surface, quartz and membrane surface), only spore and mixture of spores and titania distribution were studied. The distribution performance for spores alone which can be best achieved on alternative surfaces is shown in Table 5-12. Distribution of spores on glass cover slip and Anodisc 25 membrane filter generally showed the best distribution among all the surfaces tested. Distribution of spores on PVA/SDS pre-coated glass demonstrated the worst distribution. Overall, spor e distribution on surfaces ex cept PVA/SDS pre-coating was graded positive. Table 5-13 shows quantit ative analysis of spore distribution by calculating standard deviation and CV of spor e counts at several different locations on the surface. Spore counts were normalized because different initial spore amount was applied on the surfaces. F-test was used to analyze variance of normalized spore count data between every two groups. The result s howed that variance of spore count on plastic surface was significantly higher than on A nodisc membrane and glass surface at =0.05. No significant difference was found between variance of spore count on glass and Anodisc membrane at =0.05. This statistical result proved that spore distribution on plastic surface was less uniform than on glass and Anodisc surface, which was consistent

PAGE 127

109 with qualitative observations. No differe nce was found for distri bution uniformity on glass and Anodisc membrane surface based on statistical analysis. However, a better distribution can be observed on Anodisc memb rane surface than on glass from images observed under optical microscopy and SEM. This also matches with calculated CV value, where CV was 18 for spore count on glass surface and only 11 for Anodisc membrane. Table 5-14 shows the mixture distribution performance on the surfaces which can be best achieved on alternative su rfaces. The average mixture distribution on the Anodisc 25 membrane was the best, whereas on the plastic cover slip was the worst. Only distribution of mixture on plastic cover slip was graded negative. Interestingly, distribution of the mixture on PVA pre-coated glass and P VA/SDS pre-coated glass are better than spore distribution on the same surfaces. Distribution of spores and mixture on the same surface has different effect on their distribution quality. Distribution of spores and mixtures on glass, quartz, Pt/Au sputter coated glass, PVA pre-coated glass, SDS pre-coated glass surface and membrane demonstrated similar distribution quality. Distribution of mixtures on PVA/SDS precoated glass showed better qua lity than their spore distribu tion, whereas distribution of mixture on plastic cover slip gave worse quality than spores on the plastic surface. This different trend of mixture di stribution with spore distri bution on different surface was probably caused by both surface property of applied particles and the surfaces. Pre-coating glass surface with PVA, P VA/SDS and platinum/gold didn’t improve either spore or mixture di stribution, whereas pre-coat ing plastic surface with SDS improved mixture distribution. This maybe mostly contributed by different surface property of glass and plastic surfaces. Sin ce a good particle distribution on glass surface

PAGE 128

110 and a bad particle distribution on plastic su rface were observed, particles prefers to a hydrophilic surface for a good distribution. Mo dification of glass surface with PVA, PVA/SDS and platinum/gold make it less hydrop hilic, thus, no improve was observed. Modification of plastic surf ace with SDS makes it more hydrophilic, thus, help the distribution of particles. In the investigation of the selected factors on particle distribution, it was found that particle concentration, app lication method and surface modification pre-coated by PVA, SDS or platinum/gold influence particle s distribution on surfaces, whereas surface temperature does not change the distributi on of particles on a surface. Particle distribution also varied with th e nature of the particles, th at is, spores, nanotitania or mixtures of spores and nanotitania demons trate different distri bution performance on the same surface. Filtration method gave a uniform distributi on of spores on the filter whereas other methods showed agglomeration to some exte nt on the alternative surfaces. Moreover, viable count from filter membrane proved to be the most efficient among the alternative surfaces (Details in Chapter 6). Therefore, Anodisc filter membrane was selected as our testing surface. 5.6 Conclusion Among the investigated surface, spore di stribution quality on glass and Anodisc membrane surface was the best, whereas on th e PVA/SDS pre-coated glass cover slip was the worst. Best distribution quality of mixture of spores and titania was observed on membrane surface. Therefore, Anodics me mbrane surface is the optimum surface for spore or mixture of spores and titania distribution.

PAGE 129

111 Figure 5-1. Examples of assessment of particle distributions quality: a) and b): quality of distribution = 3+. SEM of spore (107 CFU/mL) dispersion on Anodisc membrane at magnification 100 and 1000 . c) and d): quality of distribution =2+. Optical micrograph of mixture (spore 106 CFU/mLtitania, 0.01%) on glass cover slip dip pre-coated with 5% PVA (Sigma) for 10 min (bright field illumination) at magnification 50 and 200 . e) and f): quality of distribution = 1+. Optical micrograph of spore dispersion (106 CFU/mL) on plastic cover slip pre-coated with 10% SDS for 10 min (bright field illumination) at magnification 50 and 500 . g) and h): quality of distribution =1-. Optical micrograph of mixture (spore 1.5 105 CFU/mL, titania 0.0001%) distribution on plastic surface (bright field illumination) at magnification 50 and 500 . i) and j): quality of distri bution = 2-. Optical microgr aph of mixture (spore 1.5 104 CFU/mL, titania 0.0001%) distribution on plastic surface (bright field illumination) at magnification 50 and 500 . k) and l): quality of distribution = 3-. Optical micrograph of mixture (spore,106 CFU/mL; titania, 0.01%) on plastic cover slip (bright field illumination) at magnification 50 and at 500 . a b cd

PAGE 130

112 Figure 5-1. Continued f l j e k g h i

PAGE 131

113 Figure 5-2. Optical micrograph of spore, nanotitania and mixture on glass cover slip (bright field illumination): a) and b) Spore (106CFU/mL) dispersion on glass cover slip at magnification 50 and 500 , quality of distribution = 2+, c) and d) Titania (0.01%) dispersion on glass cover slip at magnification 50 and 500 , quality of distribution = 2+, e) and f) Mixture (spore, 1.3 106CFU/mL; titania, 0.01%) on glas s cover slip at magnification 50 and 500 , quality of distribution = 1+. a c d f a b e

PAGE 132

114 Figure 5-3. Optical micrograph of spore distribution on plasti c cover slip (bright field illumination; 50 ): a) Spore (1.3 106CFU/mL) distribution on plastic surface, quality of distribu tion = 1+ and b) Spore (1.3 106 + 1.3 106 CFU/mL) distribution on plastic surface, quality of distribution = 2-. Figure 5-4. Optical micrograph of spore (105 CFU/mL) distribution on plastic surface by dipping method (bright field illumination; 500 ) (quality of distribution = 2+) b a

PAGE 133

115 Figure 5-5. Optical microgr aph of mixture (spore 106 CFU/mL, titania 0.01%) distribution on PVA dip precoated glass cover slip with 5% PVA (Sigma) for 10 min with rinse (bright field illumination): a) at magnification 50 , and b) at magnification 500 , quality of distribution = 2+. a b

PAGE 134

116 Figure 5-6. Optical micrograph of spore a nd mixture distribution on PVA spread precoated glass cover slip (bright fiel d illumination): a) and b) Spore (106 CFU/mL) distribution on glass cover slip spread pre-coated with 10% Celvol 165 PVA at magnification 50 and 500 , quality of distribution = 2+, c) and d) Mixture (spore 106 CFU/mL, titania 0.01%) distribution on glass cover slip spread pre-coated with 15% Sigma PVA, quality of distribution = 2+. a c c d b

PAGE 135

117 Figure 5-7. Scanning electron microsc ope (SEM) image of spore 106 CFU/mL distribution on glass cover slip sputter pre-coated with P t/Au, quality of distribution = 2+, a) and b) Presputter coated, c) a nd d) Postsputter coated (Control) a b c d

PAGE 136

118 Figure 5-8. Optical microgr aph of mixture (spore, 106CFU/mL; titania, 0.01%) on plastic cover slip pre-coated with 10% SDS for 10 min (bright field illumination), quality of distribution = 2+. a) at magnification 50 , and b) at magnification 500 Figure 5-9. SEM images of Bacillus cereus endospores (7 107 CFU) dispersed on Anodisc 25 membrane, quality of distri bution = 3+. a) Magnification was 100 , and b) Magnification was 1500 . a b a a b

PAGE 137

119 Figure 5-10. Optical images of Bacillus cereus endospores (1.4 107 CFU) dispersed on Anodisc 25, gold pre-coated Anodisc and polycarbonate membrane: a) and b) on Anodisc 25 at magnification 50 and 500 , quality of distribution = 3+, c) and d) on gol d precoated Anodisc membrane at magnification 50 and 500 , quality of distributi on = 2+, e) and f) on polycarbonate membrane at magnification 50 and 500 , quality of distribution = 1-. a ef b c d

PAGE 138

120 Figure 5-11. SEM images of mixture of spor es and titania on Anodi sc membrane (spore to titania ratio is same): a) and b) Mixture of spores (106 CFU) and titania (0.1 mg) on Anodisc membrane at magnification 1,000 and 10,000 , quality of distribution = 2+. c) and d) Mixture of spores (105 CFU) and titania (0.01 mg) on Anodisc membrane at magnification 1,000 and 1,000 , quality of distribution = 2+. e) and f) SEM images of mixture of spores (104 CFU) and titania (0.001 mg) on Anodisc me mbrane at magnification 1,000 and 4,000 , quality of distribution = 1-. f d a c b e

PAGE 139

121 Figure 5-12. SEM images of mixture of spor es and titania on Anodisc membrane. a) and b) Mixture of spores (5 106 CFU) and titania (0.05 mg) on Anodisc membrane at magnification at magnification 500 and 5,000 , quality of distribution = 1-. a b

PAGE 140

122Table 5-1. Factors and levels tested in this chapter Surface Factors Level Glass Titania concentration 0.1% – 0.0001% Quartz Spore concentration 1036 CFU/mL Spore application method Dipping method and pipetting method Temperature 25 C,60 C, 80 C Plastic Titania concentration 0.1% – 0.000,01% PVA type Sigma, Celvol 165, Unisize PVA coating method Dipping me thod and spreading method PVA pre-coated glass PVA concentration 5% – 20% PVA/SDS pre-coated glass Pt/Au pre-sputter coated glass SDS concentration 0.05% – 10% Soaking time 5 s – 10 min SDS pre-coated plastic Rinse Type of membrane Gelman GN-6, MSI, Polycarbonate, Anodisc 25 Membrane Modification to filtration condition

PAGE 141

123Table 5-2. Approaches for appl ication of micr obes on surfaces Approaches Microbes Surfaces Result/Conclusion Ref. Bacterial spores Bacillus. subtilis Titania coated filter and filter Titania coated filter favored uniform spreading of bacteria suspension while filter without coating did not 6 Bacillus anthracis , B. subtilis and Bacillus cereus Stainless steel coupons -3 Pipetting Aspergillus. Niger Stainless steel -2 Filtration B. subtilis Gold coated alumina membrane Gold coated glass slide An even distribution was achieved on gold coated alumina membrane Spore agglomeration and non-uniform distribution observed on gol d-coated glass slide by depositing a drop 11 B. subtilis spores Quartz and glass 3% overlapping spores on the surface was calculated by assuming a projected spore surface area of 1 m2 7 Others B. cereus metal and fabric substrates coated with silver-doped titaniaum dioxide -12 Bacterial vegetative cells E. coli Titania coated surface -4 Pipetting Coliform Stainless steel table -8

PAGE 142

124Table 5-2. Continued Approaches Microbes Surfaces Result/Conclusion Ref. E. coli , Pseudomonas aeruginosa , Sraphylococcus aureus , Enterococcus faecium and Candida albicans Titania coated plexiglas -5 E. coli Stainless steel -2 Immersion E. coli Glass, stainless steel, polyethylene high density, polyamide-6, polyvinyl chloride and Teflon -1 Filtration E. coli Cellulose acetate filter membrane SEM image shown to prove that bacteria can be remained on the surface 9 Nebulizing E. coli Titania loaded plate -10 1–Faille et al. (2002), 2–Foschino (2003), 3–Galeano et al. (2003), 4–Jacoby et al. (1998), 5–Kuhn et al. (2003), 6–Lin and Li (2003), 7–Lindberg and Horneck (1991) and (2001), 8–Moore and Griffith (2002), 9–Pat et al. (2005), 10–Sato et al. (2003), 11–Schiza et al. (2005), 12 –Vohr et al. (2005)

PAGE 143

125Table 5-3. Examples of grad ing of distribution quality Figure Particles and surfaces Grade 5-1 a,b Spore (107CFU/mL) on Anodisc membrane 3+ 5-1 c,d Mixture (spore,106CFU/mL; titania, 0.01%) on glass cover slip dip pre-coated with PVA 2+ 5-1 e,f Spore (106CFU/mL) on SDS pre-coated plastic cover slip 1+ 5-1 g,h Mixture (spore 1.5 105 CFU/mL, titania 0.0001%) on plastic surface 15-1 i,j Mixture (spore 1.5 104 CFU/mL, titania 0.0001%) on plastic surface 25-1 k,l Mixture (spore,106CFU/mL; titania, 0.01%) on plastic cover slip 3Table 5-4. Effect of spore concentration on their distribution on plastic cover slip surface Suspension Concentration (CFU/mL) Grade Comment 106 1+ Spores mostly well dispersed; agglomerates present at outer edge 105 1Spores well dispersed in both edge and center except a disordered Distribution (less spore density) at the left area of image 104 2The distribution much worse than 106 and 105 CFU/mL and droplet size is much smaller than droplets with applied spore concentration of 105 and 106 CFU/mL, hard to tell spores located on surface. 103 3Spores not uniformly dispersed, worse than 104 CFU/mL 106+106* 2Spores in most area well dispersed, large amount of agglomeration observed at the edge 106+105* 2Spores in most area well dispersed, large amount of agglomeration observed at the edge * A single droplet was applied on two successive days

PAGE 144

126Table 5-5. Effect of mixture concen tration of spore and titania their di stribution on plastic cover slip surface Titania (%) Spore Concentration (CFU/mL) Grade 0.0001 20.001 1.5 104 20.0001 10.001 1.5 105 10.01 1.3 106 3* Data means final concen tration in the mixture Table 5-6. Distribution of spores and mixt ures on glass cover slip pre-coated by (pol yvinyl alcohol) PVA us ing spreading method Grade PVA coating type Spores Mixture 5% (Celvo165) 1+ 1+ 10% (Celvo165) 2+ 1+ 15% (Celvo165) 1+ 15% (Sigma) 1110% (Sigma) 1115% (Sigma) 1+ 2+ 20% (Sigma) 1+ 1+ 5% (Unsized HA-70) 2210% (Unsized HA-70) 22

PAGE 145

127Table 5-7. Effect of sodium dodecyl sulfate (SDS) coat ing condition for plastic cove r slip on spore distribution SDS Dipping Time Grade (S pore dispersion quality) 0.025% 30 s 110 s 130 s 11 m 10.05% 10 m 130 s 1+ 0.1% 10 m 1+ 2.5% 10 m 15% 10 m 15 m 1+ 10% 10 m 1+

PAGE 146

128 Table 5-8. Scanning electronic microscopy (SEM) bacterial endospore count on three pol ycarbonate membranes using Anodisc and separate as pad Sample Spore average count (N = 6) Standard deviation CV = 100 Standard Deviation/Me an 1 128 108 85 2 161 47 29 3* 11 9 80 * The third sample used separate as pad, sample 1 and 2 were using Anodisc as pad Table 5-9. SEM bacterial endospore count on two Anodisc membranes Filter Spore average count (N = 6) Standard deviation CV = 100 Standard Deviation/Me an 1 428 163 38 2 566 60 10 Table 5-10. Optical microscopy bacterial e ndospore count on one glass cover slip Sample Spore average count (N = 5) Standard deviation CV = 100 Standard Deviation/Me an 1 437 77 18

PAGE 147

129 Table 5-11. Surface coverage by mixture of spores a nd titania on glass cover slip and membrane surfaces Surfaces Surface coverage (%) Standard deviation (%) CV (%) = 100 Standard Deviation/Mean Normalized surface coverage Standard deviation of normalized surface coverage (%) Normalized CV (%) No. of images Glass 4.09 0.85 20.7 1 0.2 21 8 Anodisc membrane 11.2 1.6 14 1 0.14 14 5 Table 5-12 Distribution performan ce of spores alone on surfaces Surfaces Distribution Quality Anodisc 25 membrane 3+ Glass 3+ Pt/Au pre-sputter coated glass 2+ Quartz 2+ Plastic 1+ PVA pre-coated glass cover slip 1+ SDS pre-coated plastic cover slip 1+ PVA/SDS pre-coated glass cover slip 3Table 5-13 Bacterial endos pores count on surfaces by image analysis Surfaces1 Spore average count S.D. CV = 100 Standard Deviation/Mean N ormalize d spore average count2 S.D.of normalized spore count Normalized CV No. of images Plastic 155 82 53 1a 0.53 52 3 Glass 437 77 18 1b 0.18 18 5 Anodisc membrane 566 60 10 1b 0.11 11 6 1F-test for were conducted based on normalized s pore count (spore count/m ean) between two groups 2 Normalized means followed by different letters are significantly different from each other at = 0.05

PAGE 148

130Table 5-14 Distribution performa nce of mixtures on surfaces Surfaces Distribution Quality Anodisc 25 membrane 3+ Glass 2+ PVA pre-coated glass cover slip 2+ Quartz 2+ PVA/SDS pre-coated glass 1+ Pt/Au pre-sputter coated glass 1+ SDS pre-coated plastic cover slip 1+ Plastic 2

PAGE 149

131 CHAPTER 6 ENUMERATION OF VIABLE SPORES ON SURFACES 6.1 Introduction Sampling of the viable spores on surfaces is one of the key techniques to study photocatalytic inactivation of bacterial endospor es on surfaces. In this chapter, methods for recovering viable spores from surfaces are assessed. As revealed by the literature review, nineteen papers discussed and compared the efficiency of sampling methods for microbes on dry surfaces. Thirteen of the papers dealt with bacterial and funga l spores. Most of the papers focused on removing the microbes from the surface for viable counting. The PVA method appeared to be the most promising in terms of recovery percentage. A 100% recovery of Bacillus subtilis from the quartz surface by PVA method was re ported by Horneck et al. (2001b) and approximately 100% recovery of E. coli from steel surface was achieved by Foschino (2003) using Rodac plate, whereas a lower recovery percentage was obtained by other methods achieved. Thus, PVA method was us ed as a main approach to enumerate Bacillus cereus spores from surfaces in the present re search. For the Rodac plate, a long contact time (4 h) was required and a rela tively low recovery of spores (67%) was observed. This method was not tested. F actors affecting effici ency of PVA method on enumeration of Bacillus cereus spores, such as PVA source, PVA concentration, PVA volume, PVA drying temperature and spore co ncentration, were discussed. Although only one reference mentioned the sonication as sampling method, it is a widely used cleaning and material disrup ting method; hence, it was te sted. Immersion method was

PAGE 150

132 used as an alternative method to recaptur e viable spores on the surfaces since it was applied by Kuhn et al. (2003) and Lin and Li (2003). Table 6-1 summarizes the factors and levels tested in this Chapter. 6.2 Background Methods for enumeration of viable spor es on surfaces generally fall into two categories in terms of literature reviewed: recovery of microbes from the surface for viable counting and imaging for viable counts. 6.2.1 Swab/Wiping Based Bredholt et al. (1999) swab b acteria from unsoiled and soiled cold-rolled stainless steel surface and vortexing the swab in diluti on to release cell in suspension for spread plating. It was found conventional cultivation has a low viable cells counting on the surface (data not shown in literature). It was considered to be caused by agglomeration of microbes on the surface. A thorough sc raping of the surface followed by rigorous mechanical shaking and/or ultrasonication of the cotton swab tip to disperse the aggregates was suggested to overcome th e underestimation of total CFU due to aggregates. Although error bar was presented in the microbial count, reproducibility of sampling method was not discussed. Buttner et al. (2001) compared efficien cy of three surface sampling methods: a swab kit, a sponge swipe, a cotton swab. Th ey showed same efficiency (70%) of all sampling methods for spores ( Bacillus subtilis ) on a glass surface, corresponding to an overall loss around 30%. It wa s found that the majority of losses (24%) occurred during processing step (vortexing or hand mixing) rather than sampling loss (7%). Data of standard errors for spore c ount were given for alternative sampling method, but no detail discussion.

PAGE 151

133 Moore and Griffith (2002) compared e fficiency of several surface sampling methods for detecting coliforms on food c ontact surfaces: traditio nal hygiene swabs, dipslides, sampling sponges and newly develope d swab-based techniques. They declared that reliability and reproducibility of met hods for enumeration of microbes on the surface was affected by manual skills of operators and the method of recapture, especially at low cell number on surface. But no data was give n for the sampling reproducibility. Swab method (minimum detection limit 100 CFU/cm-2) was not efficient in detecting bacteria on surface which relies on both the ability of th e swab to remove the bacteria from the surface and removal of bacteria from the swab. It was found all sampling methods resulted in an obvious decrease in detection of bacteria when the inoculated suspension was allowed to dry. Minimum bacterial detection limits for dried microbe suspension on the surface was more than 104 CFU/mL whereas less than 102 CFU/mL with microbe suspension on the surface. Increased adhesi on of bacteria to the surfaces during dry process was suggested to be a reason fo r the reduction in sampling sensitivity. Sanderson et al. (2002) inves tigated effective of samp ling methods for collecting Bacillus anthracis spores from contaminated nonporous surface. These sampling methods were: wipe, wet and dry swab, and HEPA vacuum sock samples. Results from HEPA vacuum is consistent with wipe me thods, but in great difference from swab method. Dry swabs detected 75% of the numbe r of spores that were detected by wipe and HEPA vacuum methods. None of the above methods was able to completely remove spores from the surface. Sampling reproduc ibility was not given since the sampling location was chosen randomly.

PAGE 152

134 Foschino (2003) used swabbing technique by streaking surface with sterile wet cotton wool, then placing the wool in diluti on to resuspend the bacteria for plating. Swabbing technique only had an average of 1% recapture for E. coli and a 26% recovery of A. niger spores. It was claimed that swabbing method was not suitable for low number detection of spores or bacteria on the surface. Also an extreme va riability in counts by swabbing method occurred am ong the replicates where E. coli was applied on the surface, which values of standard deviation exceeded the means. Galeano et al. (2003) sample d the vegetative cell suspension on the stainless steel coupons by adding PBS directly onto th e coupon and resuspended the spores by swabbing the coupon extensively with a sterile inoculation loop. Th en the suspension was plated and counted. Neither the efficiency of spore re moval from the surface nor the sampling reproducibility was described. 6.2.2 Immersion Faille et al. (2002) investigated adhesion ability and strength of Bacillus spores and Escherichia coli to inert surfaces. Six material (gla ss, stainless steel, polyethylene high density, polyamide-6, polyvinyl chlori de and Teflon) and three microbes ( B. cereus and B. subtilis and E. coli ) were tested. Bacillus spores were applied on coupons by vertical immersion for 2 h in a saline spore suspension and quickly immersed in sterile water to remove loosely attached spores. Number of adhering spores were determined by dipping the surface into tube containing Tween soluti on and sonicated for 5 min (Ultrasonic bath, Deltasonic, France, 40 kHz). Then, the detach ed spores were enumerated on agar. No comment was given on the rec overy efficiency and reproduc ibility of this sampling method for spores. They concluded that adhe sion ability of microbe s and their resistance to cleaning procedure (adhesion strength) we re determined by both the microorganisms

PAGE 153

135 and the substrata. It was found that the amount of adhering B. cereus hydrophobic spores and their adhesion strength were 10 times greater than B. subtilis hydrophilic spores, while E.coli was the easiest to be removed from su rface. Moreover, adhesion strength of B. cereus and B. subtilis spores was substantially affect ed by the surface. Hydrophobicity of material seemed to be an important factor on spore adhesion. Spores of B. cereus have stronger adhesion strength on hydrophobic su rface than hydroph ilic surface although inconsistent results were re ported by Busscher et al. (1990) and Boulange Petermann et al. (1993) that microbes preferred to adhere to high wettability surface. It was also suggested that spore exosporium plays a role in spore adhesion, but the mechanism was not clear. Gorny et al. (2003) determined the initial spore concentration on surface contaminated with Streptomyces albus by cutting a 2-cm2 piece of the surface and suspending it in a test tube with 25 mL D.I. water. Then the spores were extracted by vortexing the tube for 10 min. Concentration of spores in the suspension was obtained using brightline hemacytometer (Model La bophot 2A, Nikon, Tokyo, Japan). Recovery efficiency of viable spores from the surface by this method was not mentioned. Kuhn et al. (2003) harvested bacterial suspension ( Escherichia coli , Pseudomonas aeruginosa , Sraphylococcus aureus , Enterococcus faecium and Candida albicans ) on titania (P25, Degussa-Hls AG)) coated surface by placing the sa mple in sterile cup with 0.9% NaCl and shaking for 10min, then plat ed. No comment was given on effectiveness of this method on spore viable counts from the surface and sampling reproducibility. Lin and Li (2003) extracted spores on surface was by immersion of surface into solution containing 0.1% peptone and 0.01% Tween 80 and vortexing for 60 sec. The

PAGE 154

136 solution was then diluted and plated. Howe ver, effectiveness of this method on spore removal from the surface and sampling cons istency was not presented by the author. Sato et al. (2003) spayed E. coli on titani-loaded quartz surface using nebulizer. The number falling on the plates were obtained by immersing the surface into a prescribed volume of saline solution and plating. No information was given on the recovery percentage and sampli ng reproducibility by this method. Vohr et al. (2005) studied phot ocatalytic inactivation of b acterial spores on surfaces (metal and fabric substrates coated with si lver-doped titaniaum dioxide). Metal surface was sandblasted aluminum and fabric surface was polyester. Each of the surface was 50 50 mm. Spores of B. cereus were dried on the surfaces and exposed to UV-A light. After irradiation, the spores on fabric substr ate were collected by vortexing the substrate in about 3 mL sterilezed water, then the water was plated on agar. No comment was given on the recovery percentage of spor es and sampling consistency by the vortex process. 6.2.3 PVA Baltschukat and Horneck (1991) mentioned the use of polyvinyl alcohol (PVA) for spore enumeration on surface. They covered dr y spore layer on the glass plates with 10% polyvinylalcohol (PVA, Kalle) solution, then dry PVA solution. After drying, the PVA film with the spores was stripped off from the glass plates and resuspended in distilled water, a recovery of spore >95% recovery was claimed. However, no detail was given about the resource of this datum. They mentioned a detail desc ription of this technique is presented in the dissertation by Baltschukat (1986 ). But still no detail was found in the dissertation which is written in German. Ho rneck et al. (2001a) also used PVA method

PAGE 155

137 for viable spore counts, however, no further study on application of PVA for enumerate spores on the surface was carried out by the author. Sampling consistency by PVA method was not presented in the paper. Horn eck et al. (Horneck et al. 2001b) listed a table, the recovery percentage can be analyz ed from the data in the table. Although the author didn’t discuss the sampling reproduc ibility, 100% recovery by PVA method was achieved and the result was c onsistent based on the information given in the table. 6.2.4 Crushing Horneck et al. (2001b) enumerate viable dry spores of Bacillus subtilis from the quartz surface by crushing the quart z plates into fragment and shaking with glass beads in distilled water followed by ultrasound treatmen t to release the spores on the surface. Then the spores in suspension were plated. Results were compared with PVA method. A recovery of 20.4% and 53.5% were achiev ed by crushing method while a 100% recovery of spores from surface was claimed us ing PVA method by comparing the counting obtained from PVA film with the initial ap plied spore amounts. It was found that the spore counts by crushing method from two para llel samples varied by more than 1 order of magnitude (3.25 101 vs. 5.4 102 CFU), the author explained that a certain proportion of spores remained on the fragments cause the inco mplete recovery of spores from the surface, therefore, a bad consistency for spore counts by this sampling method. 6.2.5 Agar Contact Bredholt et al. (1999) poured 2,3,5-tr iphenyltetrazolium chloride (TTC) agar directly on the unsoiled and soiled cold-rolled stainless steel surface in the sterile Petri dish, then incubate the agar overnight. Then agar was removed from the test surface and placed with the contact -surface uppermost in a sterile Pe tri dish, and incubated for 24 h for viable counting. Results (data not show n in literature) from TTC agar overlays are

PAGE 156

138 difficult to be explained compared with the result of alte rnative methods (swab, image analysis) used. Reproducibility of sampling was not mentioned. Moore and Griffith (2002) used dipslides to enumerate viable coliforms on food contact surfaces. Dipslides are similar to cont act plates, which are pressed directly onto the sampled surface. Microbes on the surface wi ll contaminate the agar and grow. Both sides of the dipslide (VRBL, Dimanco Ltd, Henlow, UK) were pressed firmly on the surface before incubation at 37 C for 24-48 h. Use of dipslides was the most effective method for detecting the dry bacteria on the surface compared with other methods applied (traditional hygiene swabs, sampling sponges and newly developed swab-based techniques). However, dipslides method b ecome less sensitive in the detection of microbes when the bacterial suspension was dried on surface (minimum detection limits for E. coli 3.3 102 CFU/ cm2) than the suspension was not dried (<1 E. coli CFU/cm2). In their research, microbial detection eff ectiveness by dipslide method was conducted, but sampling recovery and reproduc ibility was not carried out. Foschino (2003) used Rodac plate technique for surface sampling. Rodac plate (a plate with grid) (Bibby Sterilin Ltd, UK) c ontaining agar medium was first prepared by pouring selected agar into Rodac plate. Th e agar was solidified af ter cool down and the meniscus of agar surface was formed. Th en, the entire agar meniscus was pressed carefully to the contaminated su rface for an appropriate time. After contacting, the plates were incubated for counting. TSA (Difco La boratories, Detroit, USA) was used for the capture of E. coli and Rose Bengal Chloramphenicol agar (Unipath Ltd., Bakingstock, UK) was used for recover of Aspergillus niger spores. They f ound Rodac plate can achieve an average 80% recapture for E. coli on steel surfaces with eight different

PAGE 157

139 finishes. A lower recovery percentage (67%) for Aspergilus niger spores than E. coli was achieved by Rodac plate. It was found that contac t time of agar with su rface, type of agar and microbes have large affect on the rec overy of microbes from the surface. Recuperation can be approximately to 100% for E. coli and 85% for A. niger from steel surface finished according to European standa rd 2B (2B glazed by cold rolling) after a contact time of 4 h. It was found that no signi ficant difference between mean value of the control and the mean value of recapture with Roadc plate when a controlled bacterial cells at low number (102 – 103 CFU per 42.25 cm2 area) applied on hard surface. Standard deviation of counts was low. Howeve r, this technique is not industry acceptable since 4 h contact time was required before incubation. Pal et al. (2005) enumerated bacteria on cellulose acetate membrane by placing the membrane face-down on Eosin Methylene Blue agar to grow E. coli or Tryptic Soy Agar to grow Bacillus subtilis and Microbacterium sp. The agar plates were incubated for counting. No information was given on the ba cteria recovery percen tage and consistency for this sampling method. Vohr et al. (2005) dried spores on metal (aluminum) surface with photocatalyst. Enumeration of spores on the surface was by placing the metal in the bottom of sterile Petri dish and then pouring li quid agar over the sample. Th e liquid agar was shaken for 20 s in order to transfer the spores on th e substrate into the agar. Then, the metal substrate was taken out from the agar by lifting them with a sterile needle. The Petri dish was incubated for counting. Whether spores on the surface were completely transferred to the agar during shaking process was not men tioned. Spore recovery was not indicated. Reproducibility of sampling was not discussed. Also, a limitation to the initial test spore

PAGE 158

140 number was inferred in the paper in order to have a reasonable colony counts on agar plates. Agglomeration often occurred on the plates as talked with the author, although this problem was not mentioned in the paper. Moreover, spore rec overy percentage and consistency was not indicated. 6.2.6 Release of Spores and Cell Fragments to Air Currents Gorny et al. (2003) st udied the release of Streptomyces albus propagules from contaminated agar and ceiling tile surface in a newly developed aerosolization chamber. The spores on the surfaces were released by air flow over the surface. The concentration and size distribution of the released Streptomyces albus propagules were measured with an optical particle counter (Model 1.108, Grimm Technologies , Inc., douglasville, GA). This device can measure the concentration of particles in the optical equivalent size range of 0.3-20 m based on lighter scattering, while S. albus propagules were present in a particle size range of 0.3-30 m (optical equivalent diameter). Thus, the released spores from the surfaces can be calculated. They show ed that propagules in the fine particle size range can be released in large amounts from contaminated surfaces (104 CFU/cm-2 released from surface with in itial inoculated amount of 104 CFU/cm-2) and concluded that measurement of S. albus amount in the vicinity of contam inated surface appeared to be a promising method as an alternative to surf ace sampling. However, only a maximum 1% spores was released by air current from the surfaces during a 30-min experimental duration. Therefore, air release was not an efficient way to recovery the spores from surfaces for viable counting. Standard de viation for counting wa s plotted but sampling consistency was not discussed.

PAGE 159

141 6.2.7 Sonication Venkateswaran et al. (2004) placed coupons with spores into sterile water and sonicated 2 min, then the aliquots were transf erred into Petri dish for pour plating. They found it was hard to detect the spore at surface because low level of spores (5.8 103) on the surface. Sampling reproducibility by sonication method was not discussed. 6.2.8 Mineralization Jacoby et al. (1998) meas ured the amount of CO2 produced by mineralization of E. coli dried on titanium dioxide coated surface as an indicator of disinfection efficiency instead of counts of viable bacteria, combined another tw o techniques (scanning electron microscopy and 14C radioisotope labeling). The same parameter was used by Lopez and Jacoby (2002) in their resear ch on titania dioxide coated microfibrous mesh. This method required SEM and radioisotope labe l, increased complex of operation. Reproducibility of sampling method was not given. 6.2.9 Image Analysis of Spores on Surfaces Venkateswaran et al. (2004) filter th e suspension with spores onto a 0.2 m nitrocellulose filter and stained for image analysis to enumerate the amount of spores. It was found to be difficult because the filtration methods required a minimum of 105 spores to visualize s pores under oil immersion ( 2,000 magnification) microscopy. The ability to distinguish viable and non-vi able spore was not mentioned, besides, sampling reproducibility was not discussed. Schiza et al. (2005) used SEM (ESEM FE I Quanta200, FEI, Hillsboro, OR) to enumerate B. subtilis spores on surfaces. Counting of spores on the surface was largely affected by spore agglomeration and non-uni form dispersion. Thus, a gold-coated

PAGE 160

142 alumina filtration membrane was used as the surface where an even spore distribution can be achieved by filtration method. With unifo rm distribution, image analysis program based on SEM images can give precise quantif ication of bacterial endospores. Sample reproducibility was high based on the data shown in the paper. However, the viability of spore cannot be detected under the micros copy if no significant morphology change occurred. 6.2.10 Summary Methods for enumeration of viable micr obes on surface and their efficiency are summarized in Table 6-2. PVA method demonstr ates the best recovery efficiency based on the available data, a recovery pe rcentage of 100% was achieved for B. subtilis spores on quartz and glass surface by PVA method. Agar contact method shows to be the second best, an 80% recovery was observed for E. coli from steel surf aces with eight different finishes. However, a limitation ex its for applied bacterial amount on surface in order to have a detectable colony count on agar plate. Recovery percentage achieved by swab/wip ing based method are relatively low and varied largely from 1% to 70% depending on mi crobial species, surface and other factors. For other methods, such as immersion, sonication and image analysis, no quantitative data was shown on recovery efficiency. PVA method achieved a good sampling reprodu cibility based on the data provided in the paper. Image analysis for enumeration of B. subtilis spores on gold-coated membrane also achieved a very good reproducibility based on analysis of relative standard deviation of spore counts from tripli cate samples. A very low relative standard deviation was obtained. However, it is stil l not feasible to distinguish viable and nonviable spores by image analysis. In the study of swab/wiping and crushing sampling

PAGE 161

143 method, a poor sampling reproducibility was pointed out. Other sampling methods were simply applied by researchers without further investigation for their sampling consistency. 6.3 Materials and Methods 6.3.1 Washing of Cover Slips, Glass Slides and Quartz Slides Glass cover slip (18 18 mm; 22 22 mm; 24 60 mm), plastic cover slip (22 22 mm), glass slide (76.2 25.4mm) and quartz slide (22 22 mm) were purchased from Fisher. Size different from the above will be mentioned specifically in the context. Surfaces were washed with detergent (Spark leen 1, Fisher), rinsed thoroughly with deionized water, and then bl ow dried under an air stream from an air cylinder. Thereafter, the surfaces were rinsed three times with ethanol followed by three rinses with deionized water. Finally, they were air dried under a laminar flow hood (LABCONCO purifier class 2 safe cabinet, ca t. no. 36209-000 R) at room temperature. 6.3.2 Preparation of PVA Suspension PVA suspension was prepared by addi ng PVA powders (Sigma, Celvol 165, Unisize HA-70, Celanese, Ltd. ) into a 20 mL glass scintillation vials (Fisherbrand) containing deionized water. The suspension was then autoclaved at 121 C for 20 min. 6.3.3 Removal of Spores from Surfaces for Viable Counts by PVA Sampling Method The surface was previously dispersed with dr ied spores, titania or mixture of spores and titania by pipetting method as describe d detail in Chapter 6. Sterilized PVA suspension with appropriated type, volume a nd concentration was pipetted to the surface and spread over the entire surface using sterile plastic l oop under a laminar flow hood. The suspension with the surface was placed in incubator (Fisher Science) and dried at 37C, otherwise as noted. As soon as the PVA suspension dried, the surface was taken

PAGE 162

144 out the incubator. The PVA f ilm was peeled off the surface us ing sterile metal forceps. Then, the peeled PVA film and attached s pores was dissolved into 3 mL of sterile PBS/SDS by vortexing for 2 min, and serial diluted as needed. Two dilutions of each culture were plated and each dilution was plat ed in triplicate. The numbers of viable spores were determined by pour plating (Chapt. 3). 6.3.4 Preparation of Titania Suspension Nanotitania suspension was prepared by a dding 0.01 g Degussa P25 titania powder into a sterile flask containing 100 mL ster ile deionized water. The suspension was sonicated for 30 min in an ice water ba th sonicator (Cole-Parmer 8890). Lower concentrations of nanotitania were obtained by diluting th e sonicated suspension using deionized water. Nanotitania suspension wa s vortexed for 30 s befo re transferring to a surface or mixing with spore suspension. 6.3.5 Preparation of Spore Suspension B. cereus spore suspension applied on tested surfaces other than filter membrane was treated by lysozyme method, as detail in Chapter 4. Spore suspension applied on filter membrane was treated by ASTM (10-d culture). 6.3.6 Spore, Nanotitania and Spore/Nanotitania Suspension Application Methods on Glass, Quartz, Plastic and Modi fied Glass and Plastic Surface A volume of 20 L suspension was pipetted onto a surface and dried overnight in a laminar flow hood at room temperature. Susp ension was applied to pl astic cover slips at four different points. Suspension was applie d to all other surfaces at a single point. Otherwise is as noted.

PAGE 163

145 6.3.7 Pre-coating Glass Cover Slips with SDS In our research, both coating was applie d before and after the particles were applied. Pre-coating was to coat a surface be fore application of particles suspensions on the surface. Post-coating was to coat a surface after the particles have been applied on the surface. 6.3.7.1 Dipping Glass cover slips were immersed in a 0.05% SDS suspension for 30 s, and then the cover slip was removed using sterile forceps and dried overnight in a laminar flow hood. 6.3.7.2 Spreading Enough 0.05% SDS suspension to cover the whole surface was pipetted onto the glass cover slip and spread even ly using sterile inoc ulating loop, then dried overnight in a laminar flow hood at room temperature. 6.3.8 Pre-coating Glass C over Slip with PVA Three types of PVA were used, one is pur chased from Sigma, another two (Celvol 165 and Unisize HA-70 from Celanese, Ltd., Dallas, Texas) were provided by Dr. EIMidany, Particle Engineering Research Center, University of Florida. PVA was prepared by adding a measured amount PVA powder into deionized water, then the suspension was autoclaved. PVA is hard to dissolve at high concentration under room temperature. Sterilization at higher temperature helps PVA dissolving. PVA solution was ready for use after autoclaving. 6.3.8.1 Dipping Washed glass cover slip was dipped into st erile PVA solution for a period from 5 s to 10 m, then pulled out by sterile metal forcep s and placed in plastic Petri dish. The precoated glass cover slip was air dried overnig ht in laminar flow hood. PVA pre-coating

PAGE 164

146 with rinse undergoes same pro cedure except that plastic cover slips pre-coated with PVA were rinsed by deionized water before placed in Petri dish. 6.3.8.2 Spreading Enough volume of PVA solution was dropped on a glass cover slip and spread by moping through the whole area using one edge of another glass cover slip. Then, the precoating was allowed to dry overnight under a laminar flow hood. 6.3.9 Pre-coating Glass Cover Slip with PVA and SDS Enough 10% Celvol 165 PVA suspension to cover the glass cover slip was dropped on the surface of glass cover sl ip and spread with a steril e plastic inoculation loop to evenly coat the cover slip. The pre-coated cover slip was dried for 80 m at 65 C. Enough volume of 0.05% SDS suspension was then spread over the cover slip using a sterile plastic inoculation loop. The pre-coated cover slip was then dried overnight under a laminar flow hood at room temperature. 6.3.10 Pre-sputter Coating of Gla ss Cover Slips with Pt/Au A dry glass cover slip was placed in the chamber of a Desk II denton vacuum sputter (Denton Vacuum Inc. Model No. Carb on Rod Acc) and was coated for 40 seconds. The pre-coated gla ss cover slip was taken out the chamber and ready for use. 6.3.11 Pre-coating Plastic Cover Slips with SDS Plastic cover slip was immersed into SDS solution with different concentration for different period of time, and then the surface was taken out using sterile forceps and dried overnight in a laminar flow hood. 6.3.12 Filtration Anodisc 25 with pore size 0.02 m and diameter of 25 mm (Whatman, Fisher Scientific) was used as filtration membrane. The filter membranes were immersed into

PAGE 165

147 sterile D.I. water for 5 sec to wet the me mbrane surface to avoid blank spot after filtration. Then they were transferred to a vacuum filtration system with a holder for 25 mm filters (Model FH225V, Hoefer Scientific Instruments, Piscataway, NJ, USA) using a sterile forceps. Suspension was vortexed for 30 s and then added to vacuum filtration system. Volume of the suspension varied fr om 4 mL to 5 mL, as noted. The vacuum was then applied in order to achieve even distribution of partic les over the membrane surface. Applied vacuum was adjusted and an optimum vacuum pressure was chosen to obtain good distribution of particles on membra ne. After filtration, the membrane was transferred to Petri dish and dried unde r laminar flow hood at room temperature overnight. 6.3.13 Enumeration of Spores on Glass Surfaces by Sonication Surfaces with dried spores were immersed into 40 mL sterile PBS solution containing 2 mM SDS in 100 mL sterile plastic container. The container with solution and cover slip was sonicated for a period of 5 to 15 min at level 1.5 in Misonix Sonictor 3000, depending on spore recovery. The sonicate d cover slip was then taken out of the solution using sterile metal forceps, rinsed wi th deionized water, a nd observed visually or under phase-contrast microscope. 6.3.14 Enumeration of Spores on Glass Surfaces and Carbon or Pt/Au Pre-Sputter Coated Glass Surface by Immersion The surfaces with dried spores were i mmersed into 40 mL sterile PBS solution containing 2 mM SDS or 40 mL sterile D.I. water in 100 mL sterile plastic container. The container with the solution and glass cover slip was put in a cooler with frozen ice bags to maintain a temperature at 4 C. The cooler was fixed to an orbital shaker (Lab Line) and agitated at 100 rev/min overnight. The glass cover slip was taken out using

PAGE 166

148 sterile forceps and rinsed with D.I. water. The residue on glass cover slip was scrubbed using sterile plastic loop. Then, the glass cover slip was checked under phase-contrast microscopy. 6.4 Results and Discussion 6.4.1 Recovery of Spores From Glass Surfaces 6.4.1.1 Polyvinyl alcohol sampling methods Due to its excellent adhesive property, PVA was used to remove the dry spores from surfaces. The spores adhered to PVA film was resuspended when the film was dissolved in dilution solution. A high effici ency of spore recovery from the surface was essential to obtain a reliable and reproducible spore inactivation result. Factors affecting the enumeration of spores on glass su rface using PVA method were studied. Experience with glass surf aces from different sources. Table 6-3 compared the difficulty of peeling PVA film from glass c over slip and glass slide surface when only PVA solution was applied on the surface. No s pores or mixture of spores and titania was pipetted on the surface previous ly. Glass cover slips (22 22 mm; 24 60 mm) and glass slide (76.2 25.4mm) were used to test effect of glass from different sources on PVA film removal. PVA solution was spread out to cover the whole area of glass cover slip with size of 22 22 mm. PVA suspension spread area on glass cover slip with size 24 60 mm and glass slide was same as glass cover slip with size of 22 22 mm. No difference was found on two different sizes of gl ass cover slips in te rms of easy to peel PVA film. Difference was observed between gl ass cover slips and glass slides in terms of easy to peel PVA film when using 0.1 mL 5% Celvol 165 PVA solution. It was found that the dried PVA film has less affinity with glass cover slip than glass slide, which

PAGE 167

149 makes PVA film on glass cover slip is easier to be peeled off than from glass slide. Thus, glass cover slip is chosen as gl ass surface in following research. In order to further investigate the effect of glass from different sources on viable spore count by PVA method quantit atively, another glass cover slip with the size of 18 18 mm were compared with the previous glass cover slip with the size of 22 22 mm for enumeration study. Each glass cover slip was marked 4 points which was located by evenly dividing the cover slip into 9 equal area using 4 lines, the cross points of lines were marked as drop points. At each drop point, a volume of 20 L spore suspension with concentration 106 CFU/mL was pipetted on. The sample was then dried under laminar flow hood overnight. In each trial, sa mples were at least tr iplicate. For each surface, two trials were run. A volume of 100 L 10% PVA (Sigma) solution was used and spread out to cover the whole area of 18 18 mm to achieve 30.9 L PVA volume per cm2 area. Control was made by plating spore suspension from orig inal purified spore solution and the amount of spore was calcula ted. Result is shown in Table 6-4. Recovery percentages of spores from the tw o glass cover slip surf aces are statistically insignificant analyzed by t-test at = 0.05. A low recovery range from 23 to 57% and a higher coefficient of variation than 26% was achieved. This low recovery and high variance of the results were consistent with the observation that PVA film was brittle and easy to break in the process of peeling PVA film, also strong affinity between PVA film and glass cover slip surface applied with spores existed. Thus, PVA film was hard to be removed thoroughly, which resulted in the lo wer recovery efficiency of spores on the glass cover slip and high va riance in recovery results.

PAGE 168

150 Effect of PVA characteristics. Three parameters, PVA volume, type, concentration, were considered in order to find an optimum PVA condition to remove viable spores from glass surfaces for enumeration. Glass cover slips (22 22 mm) were used as the testing surface for the study of PVA source and concentration on spore viab le counting by PVA method. The surface was dispersed with spores or mixture of spores and titaina by pipetting method. Concentration of applied spore suspension is 1.8 106 CFU/mL, and mixture has concentration of 1.8 106 CFU/mL (spores) and 0.01% (titania). For the study of PVA volume effect on PVA method, both glass cover slips and glass slides were used as the testing surface, but no substance was applie d on the surfaces. The results are shown on Table 6-3. Effects of PVA source and PVA concentration on PVA method is shown in Table 6-5. For PVA volume, it was found that incr easing applied PVA volume compensated the worse performance of PVA film peeling fr om glass slide (Table 6-3). No particles were dispersed on the surface in Table 6-3. As we compared 0.15 mL with 0.1 mL 5% PVA (Celvol 165) or 10% PVA (Sigma), an improvement for peeling was observed with increased PVA volume. The main reason is that increasing drop volume reduced very thin area of dried PVA film, which makes the film easy to be peeled. However, the corresponding dry time was increased from 40 min with 0.1 mL PVA solution to 90 min with 0.15 mL PVA solution. It was also cons istent with our previ ous finding that the PVA film formed from 200 L 10% PVA (Sigma) solution was easier to be peeled than from 100 L same PVA solution from glass cover slip, but a long dry time (240 min) was required with PVA volume at 200 L than at 100 L (40 min).

PAGE 169

151 As found in Table 6-5, for PVA source, Ce lvol 165 formed film was generally the easiest to be peeled from glass cover slip am ong the entire three alte rnatives in terms of easy of peeling PVA film. F ilm formed from PVA (Sigma) at the concentration 5% and 10% as well as PVA (Unsize HA-70) at 5% ar e brittle and thin, therefore, hard to be peeled off the glass surface. For PVA concentration, it was observed that PVA suspension become more viscous with the increase of PVA concentra tion. Thus, more vol ume of PVA suspension was required to cover the entire glass cove r slip surface with hi gher concentration, correspondingly, a longer dry time was used. It was hypothesized longer dry time at r oom temperature will favor spore removal from the glass cover slip than shorter dry time at 37 C. Therefore, one more condition, PVA solution dry temperature, was also tested. Two different temperatures, 37 C and room temperature, were adopted. Two type of PVA suspension, 20% (Sigma) and 10% (Celvo165), were indivi dually dried at 37 C and room temperature on the glass cover slip applied with particles. Results of PVA fi lm on spore removal were compared. For both type of PVA, spores still can be observed after PVA film dr ied at room temperature was removed. No improvement on spore enum eration at room temperature than 37 C was observed. Thus, PVA dry temperature is not a main factor affecting spore enumeration.In terms of the criteria of difficulty to peel, dry time and easy of use, PVA of Celvol 165 with the concentration of 5% a nd the volume of 0.1 mL was found to be an optimum choice for PVA method to spore reco very from glass cover slip. Although 0.1 mL 10% Celvol 165 can remove more substance on glass cover slip than 5% Celvol 165, the PVA solution is highly viscous, which made the suspension hard to be transferred and

PAGE 170

152 took long time to dry. However, surprisi ngly, no combination in Table 6-5 can thoroughly remove the substance (either mixtur e or spores) from the glass cover slip. The substance left on the glass cover slip wa s obvious by visual observation. This is probably due to the hydrophilic property of glass, that B. cereus spores has strong adhesion on it as mentioned by Busscher et al . (1990) and Boulange Petermann et al. (1993). This makes it difficult for PVA film to detach the spores from glass surface and remove them. Effect of spore concentration. Spore concentration a pplied on the surface was expected to affect their enumeration by P VA method. All the previous study applied spore suspension with a concentration of 106 CFU/mL on the glass surfaces, other concentration was not considered. Thus, eff ect of spore concentra tion on spore recovery from the surface was studied. Glass cover slips were applied with spore concentration at 1 10, 1 102, 1 103, 1 104, 1 105, 1 106, 1 107 CFU/mL by pipetting method, respectively. Spore suspension was diluted by sterile deionized wa ter or sterile PBS/SDS solution. Samples at same concentration order were duplicated . 0.1 mL 5% PVA (C elvol) was chosen and used due to its good performance in removing spore from surface as discussed in the above section. Image of glass cover slip after PVA peeling was checked under microscopy for the quick detection of enumeration effectiveness. It was found recovery percentage of spor e from the surface varied not only with spore concentration applied on the surface, but also the dilution solutio n. For a range of spore concentration from 10 to 105 CFU/mL diluted by sterile deionized water, spores can be removed thoroughly from glass cover slip by PVA film. No spores can be

PAGE 171

153 detected under optical microscopy. At a high spore concentration of 106, 107 CFU/mL, some spores remained on the glass cover slip which can be detected by visual and microscopy observation. For spore suspen sion diluted by sterile PBS/SDS solution, spore with concentration from 10 to 103 CFU/mL totally come off with PVA film. For the higher concentration from 104 to 107 CFU/mL, a small amount of spores were detected to be on the glass cover s lip under phase-contrast microscopy. Effect of titania concentration . Effect of titania concentration on removal of titania by PVA method is summarized in Table 6-6. Titania suspension with concentration at 0.1%, 0.01%, 0.001%, 0.0001% were dispersed on the glass cover slip surfaces by pipetting method. Sample with same concentration was duplicated. A volume of 0.1 mL 5% PVA (Celvol) was used to recover the titania from glass cover slip. It was observed that with the decrease of c oncentration applied on su rface, fewer particles left on the surface; however, PVA method cannot remove thoroughly the titania particles from the glass cover slip surface for all these applied titania concentr ation. Thus, titania particle was harder to be removed by PVA than spores on the glass cover slip. Mixing titania with spore in the mixture may impact recovery of spores from the surface by PVA method. 6.4.1.2 Sonication method Since the spore recovery pe rcentage from glass cover slip achieved by PVA method was only ranged from 23-57%, the alternativ e method, sonication method, was tried. Spore suspension with a concentration of 106 CFU/mL was applied onto glass cover slip. The recovery ability of the sonica tion to the spores from glass cover slip was observed under intermittent sonication and c ontinuous sonication. For the intermittent sonication, sonication was “on” a period of 5 min., each followed by a 1 min “off”

PAGE 172

154 period. A total time of 15 min was used fo r the “on” period. During each 1 min “off” period, the glass surface was rinsed and checked for the residue on the surface. After the first 5 min sonication, spore residue can be eas ily detected to remain on glass cover slip by visual observation. After the second 5 min sonication, Spor es left on glass cover slip still can be visually observed. No spores left on surface can be observed by visual observation but was detected under optical mi croscopy after third 5 min sonication. For the continuous sonication, sonication wa s turned on for 15 min. Spores still can be observed to remain on the glass surface and did not come off thoroughly by rinsing. No improvement was observed fo r the spore removal from the glass surface compared with intermittent sonication. 6.4.1.3 Immersion method Spore suspension with a concentration of 106 CFU/mL was applied onto glass cover slip. Spores were observed to re main on the glass surface after overnight immersion in both PBS/SDS solution and D.I. water. Scrub and ri nse cannot thoroughly remove spores from the glass surface. Sa me phenomena were observed on glass cover slip pre-coated with Carbon or Pt/Au. Th e carbon pre-coating can be easily removed after immersion whereas Pt/Au kept stable. Results from PVA, sonication and immersi on method used to recovery spores from glass surface demonstrated that none of them can thoroughly remove the spores from the surface, thus, none of them can achieve 100% recovery. In the following section, alternative recovery methods and surfaces were tested. 6.4.2 Recovery of Spores from Quartz Surfaces by PVA Sampling Method A volume of 0.1mL 5% PVA (Celvol 165) was used to recover spores with application concentration of 106 CFU/mL on each quartz slide. Same phenomenon was

PAGE 173

155 observed as on glass cover slip, spores were left on surface after PVA film was peeled off. PVA method didn’t show better recovery ability for sp ores on quartz slide than on glass cover slip. 6.4.3 Recovery of Spores from Plastic Surface by PVA Sampling Method A volume of 0.1 mL 10% PVA (Sigma) was used to recover spores from plastic cover slip with the size of 22 22 mm. Each plastic cover slip was marked 4 points which was located by evenly dividing the cove r slip into 9 equal area using 4 lines, the cross points of lines were marked as drop poi nts. At each drop point, a volume of 20 L spore suspension with concentration 106 CFU/mL was pipetted on. The sample was then dried under laminar flow hood for overnight. Samp le was at least triplicate in each trial. Three trials were run. Result is shown in Ta ble 6-7. An average s pore recovery range of 70.53% to 83.41% spores was achieved. Coeffi cient of variation of plastic cover slip is lower than 25.21%. It was observed that the PVA film was easier to be peeled off from the plastic surface than on glass and quartz surf aces. No spores were detected by visual observation after the film was removed from plastic surface. Spore recovery percentage from plastic co ver slip is significantly higher than on glass cover slip analyzed by t-test at = 0.05. Difference in recovery performance by the same PVA agent is due to the different su rface properties between glass cover slip and plastic cover slip. Surface of plastic cove r slip is hydrophobic and glass cover slip is hydrophilic. Spore suspension are more affinity to glass cover slip th an plastic cover slip and harder to be removed from glass cover slip surface, which corresponds to the low recovery from glass surface. Therefore, plastic surface is a promising surface for photocatalytic inactivation test from aspect of enumeration of viable spores on surface.

PAGE 174

156 However, as studied in Chapter 6, poor disp ersion quality of spores on plastic surface becomes an issue. 6.4.4 Recovery of spores from modified glass and plastic surface by PVA based method 6.4.4.1 Glass cover slip Pre-coated with PVA Pre-coating glass cover slip with PVA is based on the hypothesis that PVA precoating will avoid strong affinity between glass cover slip surface and the PVA postcoating which was used to sample the spore fr om the glass surface. Therefore, the spore recovery from the surface by PVA sampling method can be improved. In our initial test, 10% Celvol 165 was pr e-coated on glass cover slip by spreading method. 5% Celvol 165 was pre-coated on gl ass cover slip by dipping method. Then, spore suspension with applied spore concentration of 106 CFU/ml or mixture of spores and titania with fina l concentration of 106 CFU/ml and 0.01% was applied on the two pre-coated surfaces. PVA sampling method wa s not applied. It was found that 10% Celvol 165 PVA pre-coat on glass by spread me thod can be easily removed together with dried spores or mixture. However, 5% Ce lvol 165 PVA pre-coati ng on glass cover slip by dipping method is not easy to be peeled off thoroughly from the surface because the film was too thin. In this case, PVA samp ling method was applied. A volume of 0.2 mL 5% Celvol 165 PVA was used as sampling ag ent and spread coating on the glass cover slip. It was observed that the post-coating PVA film can be peeled thoroughly together with dried particles and the pre-coat. Thus, the removal of spores from PVA pre-coated surface can be achieved by using PVA sampli ng method in the situation that direct removal of pre-coat was not feasible.

PAGE 175

157 Since it was found that with appropriate P VA pre-coat, all dried spores or mixture on the pre-coated glass cover slip can be direct removed with pre-coat . This turns out to be a promising method for spore recovery from the surface. Thus, PVA solution used for glass cover slip pre-coat was studied in order to find out an optimum PVA pre-coat for spore viable count on the surfaces. Thr ee factors for pre-coating, PVA coating application method, PVA type and concentratio n, were considered to cause main effect on the difficulty for direct removal of PVA pr e-coat together with the particles on it from the glass surface. For the coating applic ation method, spreading and dipping method were used in our study. Spreading method will cover enough PVA solution on the glass cover slip, which make thicker f ilm than dipping method. Thus , pre-coat film is easier to be peeled. A spreading method was used for th e PVA pre-coating. In Table 6-8, effect of nine different combinations of PVA precoating on the ease of direct PVA pre-coating peeling is shown. Spread method was used for PVA pre-coating. Spore suspension (106 CFU/mL) or mixture of spore and titania suspension with final concentration of 106 CFU/mL and 0.01% was applied onto PVA pr e-coated glass cover slip by pipetting method. No PVA sampling method was used. A higher concentration makes the PVA pre-coat easier to be peeled among the same type of PV A. Pre-coat made from Celvol 165 and Unisize PVA was generally easier to be direct removed from the glass cover slip than Sigma PVA. For the pre-coat made by 10% and 15% Sigma PVA, the pre-coat was hard to be peel totally from the glass surface because the film is too thin. Hence, PVA sampling method was used to help recover the spores fr om the pre-coated surface. A volume of 0.1 mL 5% Celvol 165 PVA was used as PVA samp ling agent. It was observed that both

PAGE 176

158 post and pre-coating PVA film together with dried particle can be peeled off for 10% Sigma PVA pre-coated sample, whereas direct removal of pre-coat was hard when 5% Sigma PVA was used for pre-coating because the pre-coated film was too thin. Celvol 165 with a concentration of 10% was found to be an optimum pre-coating agent in terms of uniformity of particle distri bution, droplet spread size when the particle suspension was applied on the surface and difficulty to peel PVA-pre coating. Information about particle dist ribution and droplet spread si ze was discussed in detail in Chapter 6. Pre-coating PVA on glass cover slip im proved spore recovery from the glass surface. However, distribution of spore or mi xture of spores and titania dispersion on the surface was worsen by PVA pre-coating. This may be because that PVA pre-coating on the surface decrease the hydrophil ity of glass cover slip and distribution of particle suspension the surface was impacted. Also, PVA may dissolve in a liquid with spores when the spores were applied on PVA-precoated surfaces. This could be undesirable because PVA could coat the spores and protec t them from photocatalysis, thus, giving misleading results and introducing difficulty in the subsequent photocatalytic activity testing. 6.4.4.2 Glass cover slip Pre-coated with SDS Pre-coating SDS on glass cover slip was based on the hypothesis that SDS coating can reduce affinity between gl ass cover slip and dried spor es, then improve enumeration on glass cover slip by PVA method. A vol ume of 0.15 mL 5% PVA (Celvol 165) was used to sample the spores on surface. S pore concentration applied on each SDS precoated glass cover slip was 106 CFU/mL.

PAGE 177

159 It was hard to peel the PVA film from SDS dipping pre-coated surface. SDS spread pre-coated surface was easier than di pping pre-coated surface to be recovered by PVA film, but still it was difficult to peel a ll the spores off the coated glass surface. Glass cover slip pre-coated with SDS di dn’t show significant improvement to spore enumeration. Strong affinity still ex ists between spores and the surface. 6.4.4.3 Glass cover slip Pre-coated with PVA and SDS Result from glass cover slip pre-coated with SDS showed there is no significant improvement to spore enumeration by PVA sampling method. Strong affinity still exists between spores and surface. It was expected that by pre-coating PVA on glass cover slip, interaction between spores dried on pre-coated glass cover slip and th e coated surface can be decreased. Pre-coating SDS on PVA coa ting may improve mixture distribution on the surface. Both mixture distribution and e numeration can be optimized by pre-coating PVA followed by SDS coating. The concentration of applied spore suspension is 106 CFU/mL and mixture of spores and titania with final concentration of 106 CFU/mL and 0.01%. PVA sampling method was not applied. It was found the dir ect removal of PVA pr e-coat can peel the spores and titania off together with the pre-coat. 6.4.4.4 Presputter Pre-coated glass with plantinum and gold A volume of 0.1 mL 5% Cel vol 165 PVA was used as agen t to recover the spores on the surface. Spore concentr ation applied on the surface is 106 CFU/mL. Not all spores can be removed from the surface by PVA sampling method. 6.4.4.5 Plastic cover slip Pre-coated with SDS Table 6-9 shows effect of SDS used for plastic cover slip pre-coating on viable spore count by PVA sampling method. Recovery performance was rated on a scale of 3-

PAGE 178

160 to 1and 1+ to 3+, where 3+ indicate the hi ghest recovery, while 3indicate the poorest recovery. A volume of 0.1 mL 5% PVA (Celvol 165) was used to recovery spores from the pre-coated plastic surfaces. Recovery of spores from the SDS pre-coated plastic surface by PVA sampling method was observed to have same performance as from plastic cover slip. Thus, plastic cover s lip pre-coated with SDS was a good surface for spore enumeration by PVA sampling method. However, introduction of SDS on the plastic cover slip may cause other concern for photocatalytic test di scussed in the next chapter. 6.4.5 Recovery of Spores from An odisc Filter Membrane Surface 6.4.5.1 Preliminary investigation of alte rnative methods for spore recovery Four alternative sampling methods were used to count viable spores on the filter membrane surface. They are: Polyvinyl alcohol sampling method, sonication method, manual disruption method and agar pour plati ng method. The spore recovery ability by these four alternative methods was compared . A volume of 1 mL spore suspension with the concentration of 1.2 107 CFU/mL was added to 4 mL sterile D.I. water in dilution tube, then this total amount of 5 mL d iluted spore suspension were poured onto a membrane for filtration. The purpose to dilu te spore suspension and filtration a higher volume (>1 mL) is to achieve a better distribu tion of spores on the membrane. Therefore, a total amount of 1.2 107 CFU spores were applied on each membrane except the one used for testing of agar pour plati ng method, which a total number of 1.2 107 CFU were applied. One membrane applied with spores was prepared for each condition of sampling method. This initial spore application number on the membrane is used as control for the calculation of recovery percen tage by alternative methods.

PAGE 179

161 For Polyvinyl alcohol sampling method, 0. 1 mL and 0.15 mL 5% Celvol 165 was used to recover spores from filter membrane. Membrane was fixed on Petri dish by carbon tape to prevent the moving when peeling the post-coated PVA from the membrane. Peeled PVA film was dissolved in dilution tube for plating. Membrane after PVA recovery was incubated with agar to see any leftover on the surface. For sonication method, sample was immersed into 10 mL sterile D.I. water in 50 mL beaker and sonicated for 20 min in an i ce water bath sonicator (Cole-Parmer 8890). Then the sample was diluted and plated. For manual disruption method, sample was immersed into sterile 10 mL sterile D.I. water in 50 mL beaker and disintegrated with sterile forcepts. The membrane was easy to be broken down because the fragile propert y of Anodisc membrane. The sample was stirred for 10 min. The suspension was diluted and plated. For agar pour plating method, two membranes were tested. Each membrane was placed on a sterile plastic Petri dish. Liquid agar (tryptic so y) was poured directly onto membrane with spores and swirled for 5 s. One membrane was taken out of the agar, which was incubated with agar again to see any spores left on the membrane. The other one was leaved inside the agar. The agar was solidated under laminar hood and transferred to incubator. After 12 hr inc ubation, the colonies number was counted. Table 6-10 shows spore viable counting re sults from these four counting methods. Recovery by sonication achieved highest (87% ), whereas disintegration obtained lowest recovery percentage (21%). Results from PVA sampling method di ffered in different PVA applied volume. A higher recovery (30%) was achieved by applied 0.1mL PVA solution than a lower recovery (18%) by 0.15 mL same PVA. It is consistent with the

PAGE 180

162 observation in experiment process that, for 0.1 mL PVA solution applied on the membrane, the whole film was peeled off. Wh ile the film was peeled in several pieces and eventually all come off the surface for a pplied 0.15 mL PVA solution. Results from pour agar plating were not accurate because agglomeration existed. It was also found that membrane after pour agar plating and P VA post-coating, some spores were left on the surface of membrane as observed from growth in agar plates. This is consistent with the recovery results, that is, none of the method achieved 100% recovery. 6.4.5.2 Sonication method Among the above four alternat ive spore recovery methods , sonication demonstrated highest recovery (87%). Ther efore, sonication conditions we re further investigated in order to achieve a better viable spore rec overy from membrane surface. Joyce et al. (2003) demonstrated intensity and time ha ve an important effect on spore viable counting. Thus, sonication intensity was test ed. It was found that sonication time of 1 min is good enough to recovery spores from the membrane, therefore, sonication time was fixed at 1 min. Other e xperimental conditions (i.e. way to sonicate membrane, use of spacer) were also adjusted correspondingly based on experimental results. Sonicator 3000 (Misonix Inc., NY, USA) was used for s onication. Membrane applied with spores or mixture of spores and titania was insert ed into 10 mL sterile D.I. water in 20 mL Scintillation vial (Fisherbrand, Fisher Scientific ) with sterile forceps. Each vial contained one membrane. The vial was shaken to make the membrane immerse into bottom of the vials. Then the vial was fixe d at the center of sonicator wate r bath, a spacer was used to set the distance between vial bottom and horn of sonicator at 5 mm. For all the trials, samples were tested triplicate at each intensity level for 1 min. After sonication, spore

PAGE 181

163 suspension was plated and counted. Recove ry percentage was calculated by counting results divided by control. Effect of sonication intensity. A total of four trials we re carried out to study the effect of sonication intensity on spore r ecovery from membrane surface where only spores were applied on the membrane. A volume of 0.2 mL spore suspension with concentration of 5 107CFU/mL was diluted in sterile 4 mL D.I. water in dilution tube, then the diluted suspension was poured onto th e membrane for filtration. Experimental condition of these four trials has slight diffe rence: a) Spacer was used in all the trials except in the trial 1; b) Only one vial cont aining sample was sonicated in the sonicator every time for all the trials except in the tr ial 1, where three vials with one membrane in each vial were taped together and inserted in sonicator for sonication at each intensity level; c) In trial 1 and 2, control is calculated based on spore volume and concentration applied at filtration membrane, in trial 3 and 4, control is from results using appropriate experiment. Table 6-11 shows spore recovery results by sonication. As seen in trial 1, intensity level at 2.0, 3.0, 4.0, 5.0 and 6.0 were tested. Spore recovery ranged from 6-87%. At intensity 6.0, the highest recovery percentage was achieved. However, a large variation of vi able spore counts on membrane was observed at intensity of 2.0, 5.0 and 6.0. This may be caused by unfamiliar with the operation steps and inappropriate operati on in the procedure arrangement . Statistical analysis was not conducted on trial 1 due to this high varian ce. To minimize this variation, in the following sonication, two conditions were modified . One is to put one vial in water bath at each sonication instead three vials in the trail 1. Second is to using a spacer to make sure the distance between the vial and the horn of sonicator same at each operation.

PAGE 182

164 In trail 2, a significant improvement of spor e recovery and variation of viable spore counts on membrane was achieved compared with tr ial 1. At intensity level of 2.0, spore recovery was increased from 47% to 62%. At intensity level of 5.0, spore recovery was increased from 49% to 99%. The standard devi ation varies from 9.3-20.2% at intensity of 2.0, 5.0 and 6.0, whereas a much highe r variation observed from trial 1 (23.9%46.6%). It was also realized, during the application on filtration membrane, where 0.2 mL spore suspension was diluted in sterile 4 mL D.I. water in dilution tube and poured onto membrane. There are small volume of s pore suspension was left. A correction was calculated based on the volume left in dilution tube. Results shown in Table 6-11 for trail 1 and trial 2 are both corrected. A 99% recove ry was achieved at intensity 5.0. Spore recovery at intensity 2.0, 4.0 and 5.0 are si gnificantly different ba sed on ANOVA test at = 0.05. Among them, spore recovery at in tensity 5.0 achieved a significantly higher percentage than at 2.0, but others are not significant different based on Tukey post hoc test at = 0.05. In trail 3, in order to further explore th e optimum intensity level for spore recovery, sonication intensity at 6.0 in addition to 4. 0 and 5.0 was tested. Control was made by adding 0.2 mL spore suspension with concentration of 5 107CFU/mL into sterile 20 mL vial containing 4 mL D.I. water, then adding 5.8 mL sterile D.I. water to the vial to a total volume of 10 mL. The vial was sonicated at intensity of 5.0 and the suspension was diluted and plated. Control was duplicated. Also, the droplet left in dilution tube was pipetted out and add to the membrane to avoi d result error. As a result, a decrease in standard deviation (<12%) for spore viable counting was observed compared to trial 1 (<48%) and 2 (<21%). A highest recovery was observed at intensity of 4.0. Spore

PAGE 183

165 recovery at 4.0, 5.0 and 6.0 are not stat istically significant based on ANOVA test at = 0.05. In trial 4, same control as trial 3 was used. Spore recovery only at sonication intensity of 5.0 was tested. The droplet left in dilution tube was also pipetted out. A 98.88% spore recovery was achie ved at intensity of 5.0. Figure 6.1 shows spore recovery results from all the trails except trial 1, sonication intensity of 5.0 was found to be the optim um condition for spore recovery from the membrane only dispersed with spores in terms of recovery percentage. For the situation that the membrane applie d with mixture of spores and titania, different sonication intensity condition ma y require to achieve an optimum spore recovery. After an optimum intensity conditi on was obtained, the sa fety of the optimum intensity used for spore recovery was tested , that is, at the obtained optimum intensity level, whether the spores will be partially or totally inactivated by sonication. Effect of sonication intens ity on spore recovery from the membrane applied with mixture of spores and nanotitania was also investigated. 0.1% titania suspension was prepared by adding 0.01g titania powder into sterile 20 mL vials containing 10 mL sterile D.I. water. The titania suspension was soni cated for 10 min at intensity of 5.0 using Sonicator 3000. A certain am ount of titania suspension a nd spore suspension was added to 150 mL Pyrex beaker containing sterile D.I. water. The mixture of titania and spores are stirred on a magnetic stirrer (Nuova II, Thermolyne) with the mixing speed = #5, a volume of 4 mL mixture was pi petted onto the membrane for filtration using Brinkmann Eppendor Repeater Pipetter (Fisher Scient ific). The total amount applied on each membrane is 106 CFU for spores and 0.1 mg for titania. This combination was

PAGE 184

166 commonly used in photocatalytic test. Two tr ials were run. Control in each trial was duplicated. Control in trial 1 was not sonicated and sonicated in trial 2. Samples at each tested intensity were triplicates. Table 6-12 shows spore recovery results from membranes applied with mixture at different intensity level. In trial 1, only one intensity level at 5.0 was tested. Control was made by adding 4 mL mixture into 20 mL vials containing 6 mL D.I. water, then the susp ension was diluted and plated. A recovery percentage of 92.59 12.92 was achieved. In trial 2, intensity at 5.0, 6.0 and 7.0 were tested on spore recovery. Control was made by adding 4 mL mixtur e into 20 mL vials containing 6 mL D.I. water a nd sonicated at intensity of 6.0 for 1 min, then the suspension was diluted and plated. Recove ry percentage was calculated by counting results divided by control results. A 92% reco very percentage was achieved at intensity of 5.0. A 100% recovery pe rcentage was observed at inte nsity of 6.0. At a higher intensity of 7.0, recovery pe rcentage decreased. Sonicat e mixture on the membrane at intensity of 6.0 is promising as a 100% r ecovery achieved. Recovery percentage at intensity of 5.0, 6.0 and 7.0 shows no significant difference based on ANOVA test at =0.05. In order to check the safety of intensity 6. 0 to the viability of spores, control with and without sonication was compared at intens ity of 5.0 and 6.0. Control was prepared same as trial 2 of mixture and duplicated at each intensity level. Results were shown in Figure 6-2. Spore counts from control with and without sonicati on shows no significant difference based on t-test at =0.05 for both sonication intensity level of 5.0 and 6.0. Application of sonication to c ount viable spores on the membrane at intensity level of 5.0 and 6.0 does not inactivate spores, therefor e, cause no misleading understanding to the

PAGE 185

167 photocatalytic test conducted la ter. A sonication intensity of 6.0 was chosen for spore viable counting on the membrane Effect of spore application method. In previous spore application method (method A), spore suspension was diluted in the dilution tube and poured onto membrane for filtration. Leftover in diluting tube af ter pouring spore suspension was need to be empted to improve experimental accuracy . Another method (method B) was tried to avoid this additional step and results were co mpared with the origin al spore application method. In method B, 2 mL s pore with concentration of 5 107CFU/mL was added into 40 mL sterile D.I. water, then vortexted. A volume of 4.2 mL mixture was pipetted onto each membrane for filtration using Eppendor f Maxipettor (Fisher). The following sonication and plating methods are same as me thod A. A sonication intensity at level of 5.0 was used. Each sample was triplicat ed. A recovery percentage of 98.88 7.10 was obtained with method A while 100.33 12.58 was achieved by method B. No significant difference between these two methods in terms of t-test at =0.05, thus, the modified method (method B) was chosen in the followi ng experiment since fewer steps involved for spore application on membrane. In summary, sonication condition was d ecided for the following photocatalytic study. One vial was placed in sonicator at each sonication time. Spacer was used to locate distance and center the vials. An in tensity level of 5.0 and spore application method B were chosen as a recovery of 100% was achieved. 6.4.6 Comparison of Spore Recovery from Plastic and Glass Surface by PVA Method and from Anodisc Filter Membran e Surface by Sonication Method Sonication method can achieve 100% reco very for the spores on the Anodisc membrane. It was also found that recovery for the spores on plastic surface by PVA

PAGE 186

168 method was the best among the other alternat ive surface and method studied. Therefore, the performance of these two recovery method was compared. Results are shown in Table 6-13. A total amount of 2 104 CFU spores were applied on the plastic cover sl ip. Spore recovery percenta ge for plastic cover slip by PVA method ranged from 51% to 86%. Same amount of spores as glass cover slip was applied on glass cover slip. Spore recovery pe rcentage varied from 17% to 70%. A total amount of spores applied on the membrane is 106 CFU spores. Average spore recovery percentage for membrane by sonication method at intensity level of 5.0 ranged from 96% to 100%. A 100% recovery percentage was achieved when modified spore application method was used. Results of spore recove ry from plastic, glass by PVA method and membrane surface by sonication method are significant different from each other based on ANOVA and post-hoc test at =0.05. Spores on membrane surface recovered by sonication method demonstrated best recovery performan ce whereas spores on glass surface recovered by PVA method shows worst re covery percentage, recovery percentage for spores on plastic surface by PVA method wa s medium. This is consistent with our observation that PVA was harder to be peeled off from glass cover slip than from plastic cover slip, this difficulty in peeling cause th e low recovery of spores from glass surface for agar plating, thus, a low recovery pe rcentage was observed. Sonication method applied on recovery of spores from Anodisc filter membrane gave the best performance in all the tested spore recovery methods. 6.5 Summary Recovery of spores from different surfaces (glass, quartz, plastic, modified surfaces and Anodisc membrane) by different methods was compared and their performance is

PAGE 187

169 shown in Table 6-14. 3+ means the best recovery performance, 3means the poorest recovery performance. None of the tested recovery methods, that is, PVA, sonication and immersion method can totally remove the spores from glass surfaces. Same phenomenon was observed from quartz surface by PVA sampling me thod. A low spore recovery (23-57%) was achieved by PVA sampling method on the glass surface, although a higher spore recovery percentage (>95%) from the gl ass plate by PVA method claimed by Baltschukat and Horneck (1991). This may contribute to the difference in species we are using. Spores of B. subtilis were applied in Baltschukat’ s reseach while spores of B. cereus were used in our study. Different surface property of B. subtilis hydrophilic spores and B. cereus hydrophobic spores may be the main reason. Exosporium of B. cereus spores (not around B. subtilis ) may also play a role in spore adhesion on the glass surface. It was concluded by Faille et al. (2002) that the adhe sion strength (resista nce to be cleaned) of microorganism on surfaces relied on the su rface property of both the spores and the substrata. Also Busscher et al. (1990) and Boulange Petermann et al. (1993) demonstrated that microorganisms prefer to adhere to substrata of high wettability. This may explain why a higher recovery of spores from the glass surfaces was not achieved by us. A better spore recovery from plastic surface by PVA sampling method was observed. A recovery range at 70-84% was obt ained, which is higher than spore recovery from glass surface by the same sampling method. Different recovery performance from modified surfaces by different sampling method was observed. PVA sampling method for spores on the glass cover slip pre-

PAGE 188

170 coated with SDS and Platinum /gold presputter-coated glass showed poor recovery. Also, recovery of spores on gold and Plantinum presputter-coated glass immersion method was bad. Both PVA sampling method and di rect removal of PVA pre-coat method demonstrated good recovery for spores from PV A pre-coated glass. Spore recovery from PVA/SDS pre-coated glass surface and SD S pre-coated plastic surface by PVA post sampling method was also good. However, dist ribution of spores or mixture of spores and titania on these modified surface was unf avored as detail discussed in Chapter 5. The best recovery was observed from Anodisc membrane by sonication method. Although the other alternative sampling me thod (PVA, disintegration and agar pour plating) demonstrated a bad recovery of spores on the membrane surface. Optimum sonication condition for spore recovery from the membrane was investigated. A 100% recovery of spores from the surface was achie ved. This study has not been reported and no quantitative data on the viable recovery of spores from the membrane by sonication method has been published. A comprehensive comparison of spore distribution quality and recovery performance for all the surfaces and methods investigated is shown in Table 6-15. Spores applied on Anodisc membrane by filtration method and on glass surface by pipetting achieved the best distribution among all the spore application methods and tested surfaces, while spore distribution on glass pre-coated with PVA and SDS had the worst performance. Recovery of spores from Anodisc membrane by sonication method demonstrated the best performance. App lication of spores on Anodisc membrane by filtration method and recovery of the s pores by sonication method gave the best performance on both distribution quality and recovery performance.

PAGE 189

171 6.6 Conclusions The Anodisc filter membrane surface, together with corresponding spore application and recovery me thod: filtration and sonicati on method, was the optimum on spore distribution quality and spore recovery in all the tested surfaces and methods. Although spore distribution on glass surface by pipetting method achieved same good quality as on Anodisc filter membrane by filtration method, the poor recovery performance from the glass surface makes its application on photocatalytic inactivation study on surface less attractive.

PAGE 190

172 0.00 20.00 40.00 60.00 80.00 100.00 120.00 140.00 1.02.03.04.05.06.07.0 Sonication Intensity LevelSpore Recovery (%) * Figure 6-1. Effect of sonicat ion intensity on viable spore counts from membrane surface 1.0E+05 3.0E+05 5.0E+05 7.0E+05 9.0E+05 1.1E+06 1.3E+06 1.5E+06 4.05.06.07.0 Sonication Intensity LevelViable Spore Counts (CFU) Figure 6-2. Effect of sonicati on intensity at 5.0 and 6.0 on vi able spore counts in mixtures of spores and titania

PAGE 191

173 Table 6-1. Factors and levels tested in this chapter Surface Methods Factors Level Concentration of PVA 5% – 20% PVA volume 0.1 – 0.3 mL PVA type Sigma, Celvol 165, Unisize Spore concentration 10 – 107 CFU/mL PVA sampling Titania concentration 0.1 – 0.0001% Intensity 1.5 Sonication on/off pattern Continuous, intermittent Glass Immersion Quartz PVA sampling Plastic PVA sampling PVA type Sigma, Celvol 165, Unisize Direct removal of pre-coat PVA concentration 5% – 20% PVA pre-coated glass PVA sampling Immersion time in SDS30s, 1m, 10m SDS pre-coated glass PVA sampling SDS concentration 0.025-10% PVA/SDS precoated glass Direct removal of pre-coat PVA sampling Pt/Au pre-sputter coated glass Immersion SDS pre-coated plastic PVA sampling PVA sampling PVA volume 0.1, 0.15 mL Manual disruption Agar pour plating Intensity 2.0 – 7.0 Alumina (Anodisc) membrane Sonication Spore application method Individual dilution, combined dilution

PAGE 192

174Table 6-2. Enumeration of viable microbes on surfaces Methods Microbes Recovery percentage Result/Conclusion Ref. Bacterial spores Bacillus subtilis subsp. niger 70% Overall sampling efficiency by swab kit, cotton swab and sponge swipe were not significant different 3 Bacillus thuringiensis -Total viable cells on surfaces were underestimated in the order of 102 – 103 CFU by swab method comp ared with image analysis 2 Bacillus anthracis -Cannot sample all the spores on the surface 14 Aspergillus. niger 1% Swab is not suitable to de tect surface with low number microbes 5 Swab/wiping B. anthracis , B. subtilis and Bacillus cereus --6 B. cereus and B. subtilis --4 Streptomyces albus --7 B. subtilis --11 Immersion B. cereus --18 B. subtilis > 95% -1 PVA B. subtilis 100 % Sampling consistency is high 8 Crushing B. subtilis 20.4% and 53.5% Variability of sampling is high 8 Agar contact B. cereus --18 Release of spores and cell fragments to air currents Streptomyces albus 1% Air release was not efficient to recover spores and cell from surface 7 Sonication B. subtilis -Hard to detect spores on surfa ce due to low level of spores 17 B. subtilis -a minimum of 105 spores was required to be imaged under oil microscopy 17 Image analysis B. subtilis -Enumeration was affected by spore agglomeration 16

PAGE 193

175Table 6-2. Continued Methods Microbes Recovery percentage Result/Conclusion Ref. Bacterial vegetative cells Pseudomonas fragi and Listeria monocytogenes -Total viable cells on surfaces were underestimated in the order of 102 – 103 CFU by swab method comp ared with image analysis 2 Coliform -Minimum detection limit was 100 CFU/cm-2, spore suspension allowed to dry on the surface was harder to be sampled than the one without drying 12 Swab/wiping based E. coli 26% Variability in counts was high 5 E. coli --4 E. coli , Pseudomonas aeruginosa , Sraphylococcus aureus , Enterococcus faecium and Candida albicans --10 Immersion E. coli --15 Pseudomonas fragi and Listeria monocytogenes -Results were hard to be explained (data not shown in the paper) 2 Coliform -The most efficient method compared with swab and swab-based sampling method 12 E. coli 80% Contact time of agar with surface, type of agar and microbes affect microbial recovery from the surface 5 Agar contact E. coli --13 Mineralization E. coli --9 1–Baltschukat and Horneck (1991), 2–Bredholt et al. (1999), 3–Buttn er et al. (2001), 4–Faille et al. (2002), 5–Foschino (2003), 6–Galeano et al. (2003), 7–Gorny et al. (2003), 8–Horneck et al. (2001b), 9–Jacoby et al. (1998), 10–Kuhn et al. (2003), 11–Lin and Li (2003), 12–Moore and Griffith (2 002), 13–Pal et al. (2005), 14–Sanderson et al. (2002), 15–Sato et al. (2003), 16–Schiza et al. (2005), 17–Venkateswaran et al. (2004), 18–Vohr et al. (2005)

PAGE 194

176 Table 6-3. Effect of PVA source and PVA vol ume on difficulty of peeling PVA film from glass cover slips and glass slides PVA source PVA volume (mL) Surfaces Ease of peeling PVA film Comment Glass cover slip (22 22 mm) Easy PVA film very thin, not sticky to glass Glass cover slip (24 60 mm) Easy PVA film very thin, not sticky to glass 0.1 Glass slide Hard PVA film too thin and sticky to glass Glass cover slip (22 22 mm) Easy PVA film less affinity to glass 5% Celvol 165 0.15 Glass slide Easy PVA film easy to peel except thin edge Glass cover slip (22 22 mm) Hard PVA film too thin and sticky to glass 0.1 Glass slide Hard PVA film too thin and very sticky to glass Glass cover slip (22 22 mm) --10% Sigma 0.15 Glass slide Moderate PVA film very sticky to glass * No spore/titania/mixture applied.

PAGE 195

177 Table 6-4. Recovery percentage of spores from glass Surface 1,2 Trial Average Recovery (%) No. of Samples 1 --3 -Glass cover slip (18 18 mm) 2 37.8 16.04 3 1 56.7 14.7 3 Glass cover slip (22 22 mm)5 2 22.8 5.7 3 1A volume of 100 L 10% PVA (Sigma) was used to recover spores from surfaces 2 PVA volume/area = 30.9 L/cm2, for the larger size of cover slip, a PVA covering area was applied to be same as the smaller cover slip. 3 Only a small part of PVA film was able to be peeled from the surface. 4 mean 1.0 S.D. 5 The high variation of data for spore recovery from glass cover slip is due to the difficulty to peel the whole PVA film on the surface off

PAGE 196

178 Table 6-5. Effect of PVA source and PVA concentration on PVA method applied on glass cover slip PVA source PVA concentratio n PVA volume / area ( L/cm2) Applied particles Ease of peeling PVA film Comment Spores Hard 5% 20.7, 31.0 Mixture1 Hard Spores Hard 10% 31.0 Mixture Hard PVA film was thin Spores Easy 15% 31.0 Mixture Easy PVA film was elastic Spores Easy Sigma 20%2 31.0 Mixture Easy PVA film was very elastic Spores Easy 5% 20.7, 41.3 Mixture Easy Spores Easy 10%2 62.0 Mixture Easy Spores Easy Celvol 165 15% Spread enough Mixture Easy Spores Hard 5% 20.7 Mixture Moderate Spores Moderate Unisize HA-70 10% 31.0 Mixture Easy PVA film was thin 1 means mixture of spores and titania 2 PVA was dried both at 37 C and room temperature; others were dried at 37 C 3 None of the above PVA film can remove all the particles from the glass surface 4Glass cover slip is 22 22 mm

PAGE 197

179 Table 6-6. Recovery of different titania concentration from glass cover slips by PVA methods Applied titania concentration (%) Result 0.1 Moderate recovery1 0.01 Moderate recovery 0.001 Moderate recovery 0.0001 High recovery 1 Moderate recovery means that some particles remain ed on the surface; high recovery means that very little particles left on the surface after PVA recovery Table 6-7. Recovery percentage of spores from plastic Surface 1,2 Trial Average Recovery (%) No. of Samples 1 70.5 16.2 3 4 2 83.1 21.0 3 Plastic cover slip (22 22 mm) 3 84 16.0 4 1A volume of 100 L 10% PVA (Sigma) was used to recover spores from surfaces 2 PVA volume/area = 30.9 L/cm2 (PVA solution was spread on a 18 18 mm area on the cover slip) 3 mean 1.0 S.D. Table 6-8. Effect of modified glass surf ace (PVA pre-coating) on particle removal by peeling PVA pre-coating directly PVA pre-coating source PVA pre-coating concentration Ease of pealing PVA pre-coating Comment 5% Easy Easy to peel pre-coating together with particles 10% Very easy a little affinity between film and glass cover slip(GCS)) Celvo165 15% Easy strong affinity between film and GCS 5% Moderate PVA film is thin 10% Moderate PVA film is thin 15% Easy PVA film is hard not elastic, feel affinity between film and GCS Sigma 20% Easy Less affinity between film and GCS than 15% Sigma film 5% Very easy -Unisized HA-70 10% Very easy -*Spores or mixture of spores and titania was applied on the pre-coating

PAGE 198

180 Table 6-9. Effect of SDS coating of plas tic cover slip on spore recovery by PVA sampling method SDS Immersion time Spore recovery 0.025% 30 s 2+ 10 s 2+ 30 s 2+ 1 m 2+ 0.05% 10 m 2+ 30 s 2+ 0.1% 10 m 2+ 2.5% 10 m 2+ 5% 10 m 2+ 10% 10 m 2+ * 2+ means a good recovery by PVA sampling method (Recovery performance was rated on a scale of 3to 1and 1+ to 3+, where 3+ indicate the highest recovery, while 3indicate the poorest recovery) Table 6-10. Viable count results fr om Anodisc 25 by four sampling method Sampling method Level Counts (CFU) Recovery percentage (%) 0.1 mL PVA 3.7 106 30.12 PVA sampling 5% Celvol 165 0.15 mL PVA 2.2 106 18.32 Sonication 1.1 106 87.30 Manual disruption 2.5 106 20.57 Control 1.2 107 a1 326 26.72 Pour agar plating b2 N/A ---3 Control 1.2 103 1Membrane was taken out 2Membrane was not taken out 3A lot of agglomeration fails to count 4A variance of recovery percentage for PVA method is due to difficulty in peel the PVA film

PAGE 199

181 Table 6-11. Effect of sonicati on intensity on percenta ge recoveries of viable spores from Anodisc 25 membrane Intensity level Trial 2.0 3.0 4.0 5.0 6.0 11,3,4,7 47.71 48.32 6.44 8.80 85.36 10.47 49.97 36.02 87.10 24.94 26 62.11 9.71 b -85.54 8.65 a/b 99.24 21.04 a --5 3 --104.19 6.05 a 96.82 12.37 a 94.97 12.04 a 4 ---98.88 7.10 -1 Control in trial 1 and 2 is calculated by multipling the applied spore volume with the spore concentration. Control in trials 3 and 4 is from results using appropriate experiment. 2mean 1.0 S.D. (N = 3) 3 Three vials with one sample in each vial were tape d together and sonicated once in trail 1. In the following trails, one vial with one sample was sonicated each time. 4 No spacer used in the trial 1, but used in all the other trials 5 experiment under that condition was not carried out. 6 means followed by different letters are significantly different from each other at = 0.05 7 Statistical analysis was not conducted on trial 1 due to the unreliability of the experimental data Table 6-12. Effect of sonicati on intensity on percenta ge recoveries of viable spores from mixtures of spores and titania applied to Anodisc 25 surfaces Intensity level 5.0 6.0 7.0 Spore recovery %1 92.77 11.112100.16 11.9491.35 10.88 1 means are not significantly different from each other at = 0.05 2 mean 1.0 S.D. (N = 6) at intensity level 5.0, N=3 at intensity level 6.0 and 7.0

PAGE 200

182 Table 6-13. Comparison of spore recovery from plastic and glass cover slip by PVA method with that from membrane by sonication method Surfaces 1 (corresponding spore recovery method) Recovery percentage (%)2,4 Glass cover slip (PVA sampling method) 39.12 18.51 a Plastic cover slip (PVA sampling method) 79.69 15.44 b Anodiscs memebrane3 (sonication method) 98.3 8.20 c 1Only spores were a pplied on the surfaces 2 Mean 1.0 S.D. 3Spore application method on filter membrane is method A 4means followed by different letters are si gnificantly different from each other at = 0.05

PAGE 201

183 Table 6-14. Evaluation of spor e recovery performance for all the surfaces and methods investigated Surfaces corresponding spore recovery method Spore Recovery1 PVA sampling 1Sonication 1Glass Immersion 1Quartz PVA sampling 1Plastic PVA sampling 2+ Glass cover slip precoated with SDS PVA sampling 1Post-coating PVA sampling 2+ Glass cover slip precoated with PVA Direct removal of pre-coated 2+ Glass pre-coated with PVA and SDS Direct removal of pre-coated 2+ PVA sampling 1Presputter pre-coated glass with plantinum and gold Immersion 1Modified Surfaces Plastic cover slip pre-coated with SDS PVA sampling 2+ PVA sampling 1Manual disruption 1Agar pour plating 1Anodisc memebrane sonication 3+ 1Recovery performance was rated on a scale of 3to 1and 1+ to 3+, where 3+ indicate the highest recovery, while 3indicat e the poorest recovery

PAGE 202

184 Table 6-15. Spore distribution quality versus recovery performance for all the surfaces and methods investigated Surfaces1 Corresponding Spore Recovery Method Spore Distribution2 Spore Recovery3 Glass PVA sampling/ Sonication/ Immersion 3+ 1Quartz PVA sampling 2+ 1Plastic PVA sampling 1+ 2+ Glass cover slip precoated with SDS PVA sampling 1+ 1Glass cover slip precoated with PVA Post-coating PVA sampling/ Direct removal of pre-coated 1+ 2+ Glass pre-coated with PVA and SDS Direct removal of precoated 32+ Presputter pre-coated glass with plantinum and gold PVA sampling/ Immersion 2+ 1Modified Surfaces Plastic cover slip precoated with SDS PVA sampling 1+ 2+ PVA sampling/ Manual disruption/ Agar pour plating 3+ 1Anodisc membrane sonication 3+ 3+ 1Corresponding spore application me thod for Anodiscs membrane is f iltration, for other surfaces is pipetting method 2 Distribution performance was rated on a scale of 3to 1and 1+ to 3+, where 3+ indicate the best distribution, while 3indicate the p worst distribution 3 Recovery performance was rated on a scale of 3to 1and 1+ to 3+, wher e 3+ indicate the highest recovery, while 3indicat e the poorest recovery

PAGE 203

185 CHAPTER 7 EFFECT OF TITANIA POWDER AGAINST B. CEREUS ENDOSPORES ON MEMBRANE SURFACE UNDER 350 NM UV IRRADIATION IN DRY STATE 7.1 Introduction As shown in literature review, a total of 10 papers investigated bacterial endospore inactivation by titania. Five of them studied photocatalytic in activation against spores in systems with air as the continuous phase, whereas the other five investigated the photocatalytic inactivation of spores in aqueous media. Among the five that investigated spores with air as continuous phase, one paper (Lin and Li 2003a) tested photocatalytic activity against spores by introducing aerosol ized spore suspension into photoreactor. One paper (Goswami 2003) applied spore suspension on the photocatalyst coating and qualitatively evaluated photo catalytic activity by visual izing spore morphology changes using scanning electron microscopy (SEM). Another three papers (Lin and Li 2003b; Vohra et al. 2005; Wolfrum et al. 2002) dealt with photocatalytic inactivation of spores by drying spores on surfaces coated with titania. Further discussion of the papers dealing with photocatalytic inac tivation spores with air as continuous phase is provided below. Wolfrum et al. (2002) applied bacteria suspension to qua rtz disks pre-coated with titania (Degussa P25). After the bacterial suspension was dried in air stream, the disks was exposed to solar (300-400 nm) UV irradiation at intensity of 104 W/m2 for 24 h and the amounts of CO2 evolved was measured. They f ound that mineralization of several microbes ( Escherichia coli , Micrococcus luteus , Bacillus subtilis , Aspergillus niger spores, Bacillus subtilis spores) was faster at 50% relative humidity than at 0% relative

PAGE 204

186 humidity. Significant effect of humidity on microbial inactivation was observed for Micrococcus luteus and Bacillus subtilis cells and spores. A range of 70% of the expected CO2 is produced at 50% relative hum idity whereas 30% the expected CO2 is produced at 0% relative humidity after 12 h exposure to UVA irradiation in the presence of titania. However, no results were shown on evolution of CO2 exposed to solar UV at absence of titania coating. Thus, it was unknown whether inactivation of spores was improved by the presence of titania u nder UVA irradiation compared with UVA irradiation alone Goswami (2003) qualitative obs erved the destruction of Bacillus subtilis and Aspergillus niger spores using SEM. Purified spore suspension was applied to photocatalyst-coated aluminum disks, and then exposed to UVA. Spores began to show morphology change by 11.75 hr when the spores were exposed to UVA irradiation in the presence of photocatalyst, and by 36 hr al l the spores showin g some morphological change. No images were given on morphology change of spores with time when spores were exposed to UVA irradiation in the abse nce of photocatalyst. Thus, inactivation activity of photocatalyst was unknow n under their studied condition. Lin and Li (2003a) aero solized spores of B. subtilis into photoreactor using a collision nebulizer. A commercial titania filter was used as th e photocatalyst in the photoreactor. A sampling chamber was located at the downstream of the photoreactor. A one-stage viable sampler was used for bioa erosol sampling and c ounting. Fluorescent black lights were used as photon sources. Th e filter surface was exposed to the light intensity of 74 or 318 W/m2. Three different values of relative humidity%, 55% and 85%–were tested. The effective of photocatal ytic activity against spores was evaluated

PAGE 205

187 by calculating the bioaerosol penetration th rough the filter. The penetration of bioaerosols is calculated as the ratio of the bi oaerosol concentration in the absence of the filter to the bioaerosol concentration downstrea m of the filter. There were no statistically significant differences of spore penetrations among the three tested levels of relative humidity. Under the two test ed intensity (74 and 318 W/m2), no significant difference was found for the filter penetration with and without UVA irradiation. Since there is no improvement in limiting penetration with th e UVA on, it follows that the contribution of photocatalysis to inactiv ation of aerosolized microbes was negligible. Lin and Li (2003b) inves tigated inactivation of B. subtilis spores on photocatalytic surfaces in air. Spore suspension was pipetted onto a glass slide with titania coating or onto a commercial titania filter surface, and then the suspension was dried. The sample was tested at four intensity le vels of 2.4, 7.4, 14 and 21 W/m2 for an irradiation period of 4 h. Enumeration of spores on the surface was by immersing filter or slide into extraction solution containing 0.1% peptone, 0.01% Tween 80 and plated onto agar plates. It was found that the survival fraction of spores de clined exponentially with the light intensity increase. A higher intensity resulted in highe r inactivation rate. The inactivation rate of B. subtilis spores on the titania-coated slide was highe r than that on the titania filter under UVA irradiation. However, survival fractions for the slide with and without titania under black light irradiation were not statistical ly significant different based on multiple regression analysis. The inactivation rate constant for the filter with titania was significantly greater than the constant for the filter without titania, where both received UVA irradiation.

PAGE 206

188 Vohra et al.(2005) tested inactivation activity of a dry photocatalytic powder against dry spores under UVA irradiation. Ti tania was coated on metal (aluminum) and fabric (polyester) surface by applying slurry of photocatalyst onto the surface and drying. Spore suspension was spread onto the coated surfaces by a sterilized glass rod and dried in dark for 12 h. The sample was exposed to 350 nm UVA light at intensity of 50 W/m2 for 24 h. Enumeration of spores on alum inum substrate was by pouring agar on the samples placed in the bottom of Petri dish and incubating for count. Enumeration of spores on fabric was by immersing the surface in water and plating the water in plate count agar. For spores on metal substrate, 76% B. cereus spores was destructed in 4 h in the presence of titani a under UVA irradiation, whereas only 22% spores was inactivated after 4 h in the absence of titania under UVA irradiation. After 24 h exposure to UVA, 99% of spores were inactivated in the presence of titania, whereas only 55% spores was inactivated in the absence of titania. For spor es on fabric surface, a destruction of spores to 96% in 24 hr were observed in the pr esence of photocatalys t under UVA irradiation, whereas only 56% spores was inactivated afte r 24 h exposure to UVA irradiation without titania. During the investigation of photo catalyst activity against spores, important techniques, such as surface sampling effectiveness was not mentioned by the author. Furthermore, the effects of va riables such as titania loadi ng and titania surface coverage were not studied. As shown in the review above, only tw o papers to date have compared effectiveness of bacterial e ndospore inactivation on surfaces in the presence and absence of photocatalyst under UVA irradiation. Lin and Li (2003b) observed no difference for inactivation of B. subtilis spores on glass slide with or w ithout photocatalyst in the

PAGE 207

189 presence of UVA, whereas Vohra (2005) f ound increased inactivation efficiency by applying photocatalyst u nder UVA irradiation for B. cereus spores on metal and fabric surfaces. Since little is know n about the inactivation abil ity of photocatalytic powder against dry spores, the photocatalytic activity of titania powder against dry B. cereus spores was tested in the present study. Because of the limitation on allowable UVA irradiation intensities in th e indoor environment (Lin and Li, 2003), the effect of light intensity on photocatalysis is important, a nd was therefore stud ied as independent variable in the work. 7.2 Effect of Light Intensity on Photolysis and Photocatalysis Kinetic studies on the destru ction of chemicals or micr obes in the presence of photocatalyst under UVA irradiation were revi ewed, as summarized in Table 7-1. Possible relationships between reaction rate coefficients a nd light intensity are linear, power and other. Many studies have found e ither linear or power relationships. Other studies have found that the relationship depends on the range of light intensity: linear at low light intensities, non-linear (power rela tionship) at medium light intensities and independent of intensity at high light intensities. 7.3 Materials and Methods 7.3.1 Spore Cultivation Bacillus cereus was inoculated in liquid media and incubated for 10 days as detail described in chapter 4. The ve getative cells and spores were then harvested and purified. 7.3.2 Spore Harvesting and Purification The culture was harvested by transferring to a 85 mL Nalgene Oak Ridge polycarbonate centrifuge tube and then was centrifuged at 10,000g for 10 min at 4C using a Beckman J2-HS centrifuge. Then the pellets were resuspended in 20 mL sterile

PAGE 208

190 deionized water. The suspension was washed three times with sterile deionized water and then was transferred to a 150 mL Pyre x beaker, which was immersed in an 80 2C water bath for 15 min. The spore suspensions were centrifuged and then the spores were washed once with sterile deionized water. The purified spore suspension was stored at 4 C in the refrigerator. 7.3.3 Surface Anodisc filter membrane (Whatman, Fish er Scientific) was chosen as surface where spores or titania and spores were ap plied. The Anodisc used were 25 mm in diameter with 0.02 m pore size. 7.3.4 Filtration The Anodisc membranes were immersed into sterile D.I. water for 5 sec to wet the membrane surface to avoid blank spot in the pr ocess of filtration and then transferred to a vacuum filtration system with a holde r for 25 mm filter (Model FH225V, Hoefer Scientific Instruments, Piscataway, NJ, USA) using a sterile forceps. A volume of 4 mL titania suspension was added to each membrane in the vacuum filtration system using Brinkmann Eppendorf Repeater pipetter to achieve 1 mg titania on each membrane surface. A vacuum of 6 inches of Hg was app lied in order to achiev e even distribution of particles over the membrane surface. After 4 min. filtration, vacuum was turned off and the membrane was transferred to sterile 60 15 mm plastic Petri dish (Fisher Scientific). 7.3.4 Titania Coating on the Membrane Titania suspension was prepared by addi ng required amount of titania (Degussa P25) into beaker containing sterile D.I. wate r, and sonicated for 10 min at level of 5.0 (Misonix Sonictor 3000). Then the titania suspension was applied on the membrane by

PAGE 209

191 filtration as described detail in section 7.3.3. A magnetic stirrer (Nuova II, Thermolyne) with the mixing speed = #5 on magnetic stirre r was used to keep mixing the remaining titania suspension before the titania suspension was applied onto the membrane. The samples were covered using alumina foil to avoid illumination by indoor fluorescence light and dried under laminar flow hood at room temperature overnight. 7.3.5 Application of Spores on Tita nia Coating on the Membrane The purified spore suspension was sonicated fo r 2 min at level of 6.0. Afterwards, the spore suspension was pipetted on 10 spot s on titania coating on the membrane, with 20 L at each spots. The eight spots were previous marked by equally dividing the perimeter of membrane into eight pieces; the nine spots were marked in between any two adjacent marks, the tenth spot is the center of membrane. Th e spore droplets were dried under laminar flow hood at room temperature fo r 2 hr. During the drying process, the samples were covered using alumina foil to avoid illumination by indoor fluorescence light. Then, the membrane with dried spor es on titania coating were used for UVA and photocatalytic test immediately. 7.3.6 Application of Spores on the Membrane Application of spores on the membrane is same as section 7.3.5 except no titania coating was previously applied on the membrane. 7.3.7 Experimental Setup and Procedure The UVA and photocatalytic inactivation studies were carried out in a UVA chamber (Figure 7-1) with 16 solar UV la mps (Southern New England Ultra Violet Company, Branford, Connecticut) with peak intensity at 350 nm. Samples were placed on vertically adjustable platform. By ad justing distance between lamp and samples, UVA intensity applied on the samples can be changed. The UVA intensity at which the

PAGE 210

192 samples located (for distance between lamp array bottom and sample surface at 5 and 10 cm) was measured using a radiometer (Model 30526, Eppley Laboratories Inc.) periodically. The intensity was calculated using the voltage read from a multimeter and a conversion factor of 1.56 mV/(mWcm-2) (Vijayaraghavan and Goswami 2002). An intensity at which the samples we re located at 10 cm was 104.5 W/m2 and at 5 cm was 129.5 W/m2. Intensities at distance of 20, 30, 40 and 50 were calculated in terms of intensity measured at 10 cm. This calcula tion is based on the a ssumption that light intensity was in reverse proportiona l with square of distance. First, light intensity of each lamp at 10 cm distance was assumed, the sample is assumed to be placed at a distance of 10 cm from the center of UV array to sample surface. Then intensity applied on the sample from each individual lamp was calcula ted and summed up. The distance between the lamp and the sample surface can be calcu lated based on geometry of the lamp since the distance between lamp and center of arra y can be measured and the distance between array bottom and sample surface was 10 cm. The total light intensities obtained by adding the intensities from 16 lamps should eq ual the measured intensity at a distance between lamp array and sample surface of 10 cm. By solving this equation, the light intensity of each lamp at 10 cm distance was obt ained. With this data, the light intensity from UV array at other distance can be calcu lated by the same method, that is, adding up the intensities contributed from each indivi dual lamp. A fan with ice pack around was used to pass cooled air over the Petri dish to limit the heating of the membrane surface. UVA lights were turned on for 30 min before ex periment to obtain stable intensity. Then the UVA lights were turned off. A total of 6 sterile 60 15 mm plastic Petri dishes were placed on the platform. The Petri dishes were symmetrically placed from the

PAGE 211

193 middle of the light with 3 Petri dishes on each side. This arrangement is to make sure all the samples in the Petri dishes are exposed to same intensity. Each Petri dish contains 3 membranes applied with spores or spores a nd titania particles. Samples on one array were used as control where only spores are applied on the membrane, whereas samples another array were applied both spores and titania. All the sa mples were placed under UVA chamber except the samples used for initial counts at time zero. No cover was applied on Petri dish. At each sampling time, the two symmetric Petri dishes containing three samples each were taken out and viab le spores on the samples were counted. Sampling intervals were adju sted depending on trials. The UVA light was on continuously until the experiment is over. The distance between membrane and the bottom of solar UV lamp was adjusted to 10, 20, 30, 40 and 50 cm. A Fisherbrand traceable humidity/temperature pen was used and placed near the sample to monitor the humidity and temperature in the chamber. The humidity/temperature also displays highest and lowest readings during experimental period. 7.3.8 Viable Counting of Spores on the Membrane Surface Each sample was immersed into 20 mL sterile Scintillation vial with 10 mL sterile D.I. water using sterile forceps. The susp ension with membrane was sonicated for 1 min at level 6.0 using Sonicator 3000. After s onication, the suspension was diluted, if necessary, and plated duplicate at appropriate dilution order as detail in Chapter 4. Count of spores on membrane was mean of two plates. 7.3.9 Data Analysis Log(Survival ratio) vs. time under UVA and UVA+P25 at each light intensity was plotted based on experimental results. Each data point in Log (Sur vival ratio) vs. time

PAGE 212

194 figure was derived from mean of th ree samples. Error bars indicate 1 S.D. Linear regression line was derived based on the data point excluding data point time zero in the figure. In this region, time period required for a 1.0 log10 reduction in survival is the D value. Slope of the linear regression line is defined as inactivati on rate. The length of exposure time, starting at time zero, required to inactivate 90% of the spores is termed the LD90. The difference between slopes of regr ession lines for the UVA and UVA+P25 test were analyzed by F-test us ing GraphPad Prism 4 for Windows software (GraphPad software, Inc). Experiment results at different light intensities were integrated by plotting inactivation rate under UVA and UVA+P25 vs. light intensity. The relationship between inactivation rate under UVA and UVA+P25 was modeled. 7.4 Results and Discussion A series of preliminary experiments was carried out to determine the results of irradiating spores or mixtures of spores and titania on surfaces that were partially or fully coated with titania. Titania loading, spore preparation, a nd two levels of UVA intensity were also evaluated. The results of thes e experiments are summarized in Appendix B. Experimental parameters for final testing effect of light intensity on photolysis and photolysis of spores on surfaces was determined as follows: target microbes are B. cereus spores, culture incubation time is 10 days, cu lture purification met hod is ASTM and heat shock; surfaces are Anodisc membrane; ap plication methods for spores are pipetting method and for titania were filtration met hod; applied titania concentration on the membrane is 2 g/mm2 and spore concentration on membrane is 2 103 CFU/mm2; recovery of spores is sonication methods.

PAGE 213

195 Results of UVA intensity on the inactivation of spores lying atop either on a bare membrane filter surface or tita nia-coated membrane filter surface are presented in the following. Effect of light intensity on photolysis and photocatalysis of spores on surfaces. The kinetics of spore inactivation was charac terized using a semi-log plot of survival ratio vs. time, as shown in Figure 7.2. The exposure time to achieve 90% inactivation of the spores is the LD90. The time required for 1 log reduc tion of survival ratio in the region of the plot in which the log surv ival vs. exposure time relationship is approximately linear is the D value. The slope of this linear trend gives the inactivation rate coefficient. Figures 7.3.8 show the spore inactivation re sults over a range of light intensities. The D value and LD90 of each trial, in addition to the pa rameters varied in the trials, are summarized in Table 7-2. Figure 7-3 is the experiment al result of trial 1, in which the distance between sample and light is 10 cm. A total of 90% spores were inactivated in around 120 min with UVA alone and UVA+P25. A D valu e of 127 and 176 min was achieved for UVA and UVA+P25 respectively. Based on the F-test at = 0.05, results from solar UVA and UVA +P25 was not statistically si gnificant different. Trial 2 wa s the repeat of trial 1. Results were shown in Fig. 7-4. Based on F-test at = 0.05, results from UVA and UVA +P25 were not statistica lly significant different. In trial 3 (Figure 7-5), th e distance between lamp and sample was increased to 20 cm. Slopes of trendlines under UV and UV+P25 was not statistically significant different

PAGE 214

196 based on F-test at = 0.05. Therefore, the inacti vation rate of UV A and UVA+P25 was not significant different at this light intensity. In trials 4, 5, 6, the distance was increased to 30, 40 and 50 cm respectively. In each of these trials, the in activation rate coefficient with UVA+P25 was significantly greater than inactivation rate coefficient with UVA alone. Furthermore, the ratio of inactivation rate coefficients progressivel y increased as the distance was increased. Figure 7-9 summarizes inactivation rate coefficients at different light intensities. For the lower light intensities (distances of 30, 40 and 50 cm), inactivation rate coefficients were averaged and plotted. As shown in the figure, at a lower light intensity less than 20 W/m2, inactivation rate coefficient by UVA+P25 is higher than UVA. The difference of inactivation rate coefficien ts by UVA+P25 and UVA is significant based on F-test at = 0.05. At a medium light intensity, inactivation rate coefficient by UVA is almost same as UVA+P25. The inactivation ra te coefficient is not significant different based on F-test at = 0.05. At a light intensity range from 35 to 110 W/m2, inactivation rate coefficient in the presence of titania is lower than in the absence of titania under UVA. Based on F-test at = 0.05, there is no significant difference for the inactivation rate coefficient with and without titania under UVA irradiation. For UVA alone, inactivation rate coeffici ents increased with increased light intensities. A linear relati onship was observed between inac tivation rate coefficient and light intensity giving a R2=0.99. For UVA + P25, an increase of inactivation rate coefficient at low light intensity was observed with the decrease of light intensities. At a lower light intensity (< 20 W/m2), the significant higher inactivation rate coefficient achieved by UVA+P25 than UV indicates a greater contribution from

PAGE 215

197 photocatalyst. At medium and high intensitie s, no significant difference in inactivation rate coefficients by UVA and UVA+P25, i ndicating a negligible contribution by photocatalysis. In addition, the observed lo wer inactivation rate coefficient in the presence of titania than in the absence of titania under UVA at higher light intensities (35 to 110 W/m2) may be contributed by shading effect due to agglomeration of spores on titania coating. As shown in Figure 7-11 a nd 7-12, which shows distribution of spores on titania coated membrane and bare membrane respectively, agglomerations were observed for spores on titania coated membrane, whereas spores on bare membrane were uniformly distributed. Thus, spores on titan ia coated membrane are shaded from UVA irradiation due to agglomeration, resulting in a decreased inactivation rate coefficients than spores on bare membrane under UVA. Fu rthermore, this shading effect may be greater than the contribution by photocatalysis activity, theref ore, the inactivation rate coefficients by UVA+P25 is lower than UVA. The linear relationship between inactivation rate coefficient and light intensity under UVA observed in the study is c onsistent with that reported in literatures. Increased inactivation rate coefficient at low light intensity by UVA+P25 may be caused by weaken of spore resistance due to long-term storage of spore suspensions. Difference in inactivation rate coeffici ents by UVA and UVA+P25 varies with the light intensities. This phenomenon is expl ained by the model illustrated in Figure 7-10. Distribution of spores on bare membrane is uniform, thus the inac tivation of spores on bare membrane surfaces under UVA is by phot olysis without shading effect. The relationship between inactivation rate coeffici ents and light intensities is represented by pink solid line (photolysis wit hout shading). Photolysis is in linear relationship with

PAGE 216

198 increased light intensity. S pore agglomerations were obser ved on titania coating, thus, photolysis of spores on titania coated membra ne is lower than spores on bare membrane due to shading effect. Performance of spore inactivation on titania coating under UVA over a range of light intensities is the combin ed activity by photolysis (with shading) and photocatalysis, which is represented by pink dashed line and blue dashed line respectively. For the photolysis (with shading), the linear relationship between inactivation rate coefficient and light intensities is assumed. For the photocatalysis, it is assumed that the inactivation rate coefficients increases with increased light intensities and saturated at a certain point in term s of Langmuir-Hinshelwood mechanism. Pink square dots shows inactivation results of spores on bare membrane surface under UVA at certain light intensities, which is the photolysis without shading effect. Blue round dots demonstrate inac tivation results of spores on titania coated membrane surface under UVA by adding the photolysis and photocatalysis (with shading) at the certain light intensities. At higher intensity , contribution by photocat alysis is traded off by the shading effect of photolysis, no signi ficant difference can be observed for spore inactivation with and without titania under UVA. At lower intensity, contribution of photocatalysis is much higher than the shad ing effect of photolys is, thus, significant contribution of photocatalysis to spore inactiv ation on titania coated membrane surface is observed. Results of this model match well with experimental result. 7.4 Summary and Conclusions Inactivation rate of B. cereus spores under UVA irradiati on alone increased with increase of light intensity. Difference for in activation rate in the pr esence and absence of titania under UVA irradiation was significantly affected by light intensity. At higher light intensity, photocatalysis contribution is negligible, inactivation rate with and

PAGE 217

199 without titania under light irra diation was not statistically significant different. Whereas at lower intensity, photo catalysis contribution is significan t, this is relevant for indoor application. At higher light intensity, in crease of titania amount in the mixture of spores and titania didn’t show improvement in spore inactivation under UVA irradiation compared with UVA in the absence of titania. Inst ead, obvious shading effect observed with increased titania amount. For the sp ore application method that depositing B. cereus spores on titania coating, although shadi ng effect was eliminated, no significant difference in spore inactivation rate was observed between UVA and UV+P25. The high light intensity was thought to contribute this insignificant difference for UVA irradiation system with and without photocatalyst. The high inactivation ra te by UVA irradiation alone makes contribution by photocatalysis negligible.

PAGE 218

200 UV Lamp Array Fan Platform RH and Temperature Probe Petri Dish Ice Pack UV Lamp Array Fan Platform RH and Temperature Probe Petri Dish Ice Pack Figure 7-1. Photocatalytic experiment test facility 60 y = -0.0052x -0.3603 R2= 0.8589 -3 -2 -1 0 0 120180 Time (min)Log (Survival ratio) * 1 log10 LD90 Inactivation rate coefficient D value 240 300 60 y = -0.0052x -0.3603 R2= 0.8589 -3 -2 -1 0 0 120180 Time (min)Log (Survival ratio) * 1 log10 LD90 Inactivation rate coefficient D value 240 300 Figure 7-2. Determination of LD90, D value and inactivation rate coefficient from relationship between spore survival ratio and time (Inactivation of B. cereus by UV+P25)

PAGE 219

201 y = -0.0079x 0.0584 R2 = 0.9998 y = -0.0052x 0.3603 R2 = 0.8589 -3 -2 -1 0 060120180 Time (min)Log (Survival ratio) * UV+P25 UV Figure 7-3. Inactivation of B. cereus spores by UV alone and UV+P25 in trial 1. Inactivation rate coefficients for UVA and UVA+P25 were not statistically different at = 0.05. y = -0.0051x 0.7241 R2 = 0.9704 y = -0.0044x 0.5005 R2 = 0.9994 -3 -2 -1 0 060120180 Time (min)Log (Survival ratio) * UV+P25 UV Figure 7-4. Inactivation of B. cereus spores by UV alone and UV+P25 in trial 2. Inactivation rate coefficients for UVA and UVA +P25 were not statistically different at = 0.05.

PAGE 220

202 y = -0.0029x + 0.1751 R2 = 0.9882 y = -0.0025x 0.3876 R2 = 0.9797 -6 -5 -4 -3 -2 -1 0 090180270360 Time (min)Log (Survival ratio) * UV+P25 UV Figure 7-5. Inactivation of B. cereus spores by UV alone and UV+P25 in Trial 3. Inactivation rate coefficients for UVA and UVA+P25 were not statistically different at = 0.05. y = -0.0011x 0.0398 R2 = 0.908 y = -0.0029x 0.0354 R2 = 0.8582 -6 -5 -4 -3 -2 -1 0 090180270360 Time (min)Log (Survival ratio) * UV+P25 UV Figure 7-6. Inactivation of B. cereus spor es by UV alone and UV+P25 in trial 4. Inactivation rate coefficients fo r UVA and UVA+P25 were statistically different at = 0.05.

PAGE 221

203 y = -0.0013x + 0.0772 R2 = 0.9652 y = -0.003x + 0.1416 R2 = 0.986 -6 -5 -4 -3 -2 -1 0 090180270360 Time (min)Log (Survival ratio) * UV+P25 UV Figure 7-7. Inactivation of B. cereus spores by UV alone and UV+P25 in trial 5. Inactivation rate coefficients fo r UVA and UVA+P25 were statistically different at = 0.05. y = -0.0006x + 0.0468 R2 = 0.6348 y = -0.0041x + 0.3364 R2 = 0.9911 -6 -5 -4 -3 -2 -1 0 1 090180270360 Time (min)Log (Survival ratio) * UV+P25 UV Figure 7-8. Inactivation of B. cereus spores by UV alone and UV+P25 in trial 6. Inactivation rate coefficients fo r UVA and UVA+P25 were statistically different at = 0.05.

PAGE 222

204 0 2 4 6 8 10 12 14 020406080100120 Light intensity (W/m2)Inactivation rate co efficient (1/d) * UV+P25 UV Figure 7-9. Inactivation of B. cereus spores by UV alone and UV+P25 at different light intensity

PAGE 223

205 0 2 4 6 8 10 12 14 020406080100120Calculatedintensity (W/m2) Photolysis (with shading) + photocatalysisInactivation rate coefficient (1/day)Photolysis (with shading) Photolysis (without shading) Photocatalysis Photolysis (without shading) 0 2 4 6 8 10 12 14 020406080100120Calculatedintensity (W/m2) Photolysis (with shading) + photocatalysisInactivation rate coefficient (1/day)Photolysis (with shading) Photolysis (without shading) Photocatalysis Photolysis (without shading) Figure 7-10. Models on effect of UVA light intensity on photocat alysis contribution to spore in activation on surfaces in dry st ate under UVA irradiation

PAGE 224

206 Figure 7-11. Distribution of spores on membrane surface Figure 7-12. Distribution of spores on titania coated membrane surface

PAGE 225

207Table 7-1. Effect of light intens ity on photolysis and photocatalysis Reaction rate coefficient dependence on light intensity Reference Continuous phase = water Linear Daneshvar et al. (2004); Dionysiou et al. (2002); El-Dein et al. (2003); Emeline et al. (2000); Haeger et al. ( 2004a); Kundu et al. (2005); Mills et al. (2006); Ollis (2005); Sauer et al. (2002); Tokumura et al. (2006); Uyguner and Bekbolet (2004); Wu et al. (2006); Wu and Chern (2006); Yamazaki et al. (1999) Power (exponent < 1.0) Behnajady et al. (2006); Chen and Ray (1999); Fu et al. (2006); Hwang et al. (2003); Inel and Okte (1996); Mehrot ra et al. (2003); Mehrotra et al. (2005); Meng et al. (2002); Mills et al. (2006); Ollis (2005); Shang et al. (2002) and Yatmaz et al. (2004) Linear at low light intensity Square root of light intensity at medium light intensity Independent at high light intensity Muruganandham and Swaminathan (2006) and Ollis et al. (1991) Continuous phase = gas Linear Demeestere et al. (2004); Haeger et al. (2004a); Haeger et al. (2004b) and Ollis 2005 Power (exponent < 1.0) Hwang et al. ( 2003); Ollis (2005) and Wang et al. (1999) aThe magnitudes of “low”, “medium” and “high” intensity were not specified

PAGE 226

208Table 7-2. Inactivation of B. cer eus spores by UV alone and UV+P25 Trial1 T( C) Relative Humidity (%) Distance (cm) System LD90 (min) D Value (min) Ratio of LD90 Ratio of D value UV 120 127 1 26-30 42-50 10 UV + P25125 176 0.96 0.72 UV 70 195 2 25-27 52-65 10 UV + P25112 228 0.62 0.86 UV 431 339 3 24-28 43-61 20 UV + P25241 384 1.79 0.88 UV 921 959 4 23-30 42-68 30 UV + P25336 349 2.74 2.75 UV 827 768 5 25-29 38-54 40 UV + P25375 329 2.20 2.33 UV 1854 1771 6 25-29 46-67 50 UV + P25329 248 5.64 7.15 1 Titania quantity = 1.0 mg 2Slope refers to slope of regression lin e from UV and UV+P25 inactivation result

PAGE 227

APPENDIX A EFFECT OF VARIABLES ON PHOTOCATALYTIC INACTIVATION OF MICROORGANISMS

PAGE 228

210 Table A-1. Photocatalytic inactivation of microorganisms by TiO2 and other materials Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Bacterial spores TiO2 f TiO2-multiwallnanotube (MWCT) nanocomposites UV lamp PI = 350 nm I = 92 W/m2 Cont. phase = water Degussa P25 didn’t enhance spore inactivation than UV alone TiO2-MWCT nanocomposites doubled inactivation efficiency than UV alone Lee et al. (2005) Bacillus cereus Spores Aluminum coated with silver-doped titanium dioxide Fabric (polyester) coated with silver-doped titanium dioxide TiO2 e UV-A lamp PI = 350 nm I = 50 W/m2 Cont. phase = air Enhanced spore inactivation was observed on aluminum and fabric coated with silver-doped titanium dioxide than TiO2 (Degussa P25) photocatalysis and UV alone Vohra et al. (2005)

PAGE 229

211Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Bacillus pumilus Sporesa TiO2 b PI = 365 nm I = 22 W/m2 Cont. phase = water Viability of spores decreased exponentially with time No reduction of spores occurred in the dark or in the absence of TiO2 Intermittent illumination gave higher specific inactivation rates than continuous irradiation Maximum specific inactivat ion rate of spores was observed at 2 mg/mL TiO2 Higher initial spore concentrations resulted in higher specific inactivation rates Pham et al. (1995) Degussa P25 coated quartz disk Low-pressure nearUV lamp PI = 365 nm Cont. phase = air Higher relative humidity increased spore oxidation rate, as measured by production rate of CO2 Wolfrum et al. (2002) Photocatalystcoated aluminum disk UV-A Disinfection of the bacteria in the air Cont. phase = air Photocatalytic technolog y showed disinfection activity Goswami (2003) Bacillus subtilis Spores Commercial TiO2 filterc Black light PI = 365 nm I = 7.1 mW/cm2 I = 31.8 mW/cm2 Cont. phase = air Germicidal capability for airborne microorganism was almost zero for TiO2 photocatalyst filter media Lin and Li (2003 a)

PAGE 230

212Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Bacillus subtilis Spores TiO2 i immobilized on acetate sheet Xenon arc solar simulating lamp I = 870 W/m2 (300 nm – 10 m) Cont. phase = water Both solar and solar pho tocatalytic disinfection batch process reactor achieved 1.7 log reduction in 8 h Lonnen et al. (2005) Bacillus subtilis Sporesa TiO2 dispersed in activated carbons Black light Cont. phase = water 99% spore killed in 15 min. High crystallinity of TiO2 gave high photocatalytic inactivation Tamai et al. (2002) Clostridium perfringens Sporesa 1 mm thick TiO2 film on titanium plate UV-B lamp Electric field enhancement applied Cont. phase = water 2 log10 reduction achieved in 25 min. Butterfield et al. (1997) Bacterial vegetative cells Aeromonas hydrophila AWWX1 Veg. cell TiO2 pellet Low mercury fluorescent lamp = 300 ~ 460 nm PI = 365 nm Cont. phase = water 4 log 10 reduction observed w ithin 50 min. when exposed to mercury lamp Addition of titania to solar irradiated water did not improve inactivation activity Kersters et al. (1998)

PAGE 231

213Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Actinomyces viscosus Veg. cell TiO2 d Fluorescent lamp PI = 578 nm Cont. phase = water Viability decreased when TiO2 exposed to light Nagame et al. (1989) Bacillus pumilus Veg. cell TiO2 coated on stainless steel UV lamp I = 630 m/cm2 PI = 365 nm Cont. phase = air 50% inactivation observed within 100 min. Yu et al. (2003) Bacillus subtilis Veg. cell Membrane filtered with TiO2 r (batch inactivation) Polyethersulfone membrane disc filter coated with TiO2 r (continuous flow reactor) UV-A lamp I = 1.82, 4.28, 6.28 mW/cm2 PI = 365 nm Cont. phase = air Significant bacterial inac tivation occurred under UV alone (without titania) Enhanced inactivation ra te was observed in the presence of titania Inactivation rate in the continuous system were higher than in batch reactor Pal et al. (2005) Coliform bacteria Veg. cells Anatase TiO2 (Fisher Scientific) Sunlight Black light Cont. phase = water 2 log10 reduction of Poliovirus 1 achieved in 120 min. No difference for disinfection detected among the pH 5-8 Watts et al. (1995)

PAGE 232

214Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Coliform bacteria Veg. cells Degussa P25 UV lamp Solar light Cont. phase = water Solar disinfection result s are similar to those obtained with UV lamp Similar inactivation yields observed as Streptococci faecalis TiO2 enhance inactivation only at pH 5 Bacteria regrowthed after stop irradiation Melian et al. (2000) Deinococcus radiophilus Veg. cells Degussa P25 UV lamp I = 7 mW/cm2 PI = 350 nm Cont. phase = water 1-log10 reduction observed within 40 min. Laot and Narkis (1999) Enterobacter cloacae Veg. cells Degussa P-25e UVA light PI = 365 nm I = 5.5 mW/cm2 Cont. phase = water 4 log10 reduction observed after 40 min. irradiation Ibanez et al. (2003) Enterococcus faecium Veg. cells Degussa P25 coated on Plexiglas UVA light PI = 356 nm Cont. phase = water 6-log10 reduction observed in 60 min. Inactivation of bacteria s uggested to be damage of cell wall caused by hydroxyl free radicals Kuhn et al. (2003)

PAGE 233

215Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference TiO2-NiFe2O4 nanocomposite UV spectrophotomete r = 270 nm Cont. phase = water Bacterial inactivation is more efficient under UV and TiO2-NiFe2O4 composite nanoparticles than UV light alone Rana and Misra (2005) TiO2-NiFe2O4 nanocomposite UV spectrophotomete r = 270 nm Cont. phase = water Anatase-titania-coated nickel ferrite is superios to brookite-titania-coated nick el ferrite in bacterial inactivation under UV Rana et al. (2005) Escherichia coli Veg. cells Membrane filtered with TiO2 r (batch inactivation) Polyethersulfone membrane disc filter coated with TiO2 r (continuous flow reactor) UV-A lamp I = 1.82, 4.28, 6.28 mW/cm2 PI = 365 nm Cont. phase = air Significant bacterial inac tivation occurred under UV alone (without titania) Enhanced inactivation ra te was observed in the presence of titania Inactivation rate in the continuous system were higher than in batch reactor Pal et al. (2005)

PAGE 234

216Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference TiO2 i immobilized on acetate sheet Xenon arc solar simulating lamp I = 870 W/m2 (300 nm – 10 m) Cont. phase = water Solar disinfection batch process reactor achieved 5.4 log inactivation in 2.5 hr Photocatalytic disinfecti on batch process reactor achieved 5.4 log inactivation in 1.5hr Lonnen et al. (2005) Degussa P25 Black light = 300-420 nm I = 7.9 E/L/s Cont. phase = water Inactivation is by both the free and the surfacebound hydroxyl radicals, may also be inactivated by other reactive oxygen species Cho et al. (2005) Degussa P25, ZnO Sahara desert dust Sodium lamp Cont. phase = water Complete inactivation occurred in 40 min Sahara desert dust has no microbicidal effect under photolysis Seven et al. (2004) Escherichia coli Veg. cells Copper deposited titania thin film White fluorescent light bulb I = 40, 7, 1 W/cm2 Cont. phase = water A decrease survival rate was observed with a very weak UV intensity (1 W/cm2) on Cu/TiO2 film Sunada et al. (2003)

PAGE 235

217Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference SbTPP/SiO2 fluorescent lamp I = 21 W/cm2 Cont. phase = water SbTPP/SiO2 showed much superior bactericidal activity to the commercially available TiO2 Yokoi et al. (2003) TiO2 Natural sunlight collected by compound parabolic collector Cont. phase = water Addition of TiO2 enhanced bacterial inactivation only at 3 mg/L dose McLoughlin et al. (2004a) Escherichia coli Veg. cells Degussa P25 Hanau Suntest lamp Cont. phase = water Present of TiO2 accelerate inactivation action of light Bacterial recovery was not observed during 24 hr after stopping sunlight exposure in the presence of TiO2 but occurred in the absence of TiO2 Rincon and Pulgarin (2004c)

PAGE 236

218Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Degussa P25 Hanau Suntest lamp Cont. phase = water Inactivation rate was a ffected by physiological state, generation and in itial concentration of bacteria Initial inactivation rate increased as initial bacterial concentration increased Enterococcus sp. was less sensitive than coliforms and other Gram-negative bacteria in wastewater Wastewater samples from two different date showed different respon se to photocatalytic treatment Rincon and Pulgarin (2004a) Escherichia coli Veg. cells Degussa P25 Hanau Suntest lamp Cont. phase = water Initial pH between 4.0 to 9.0 does not affect inactivation rate in the absence or presence of titania Addition of H2O2 increase inactivation rate Addition of inorganic has different affect on inactivation rate Dihydroxybenzenes impacted photocatalytic inactivation Rincon and Pulgarin (2004b)

PAGE 237

219Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference TiO2 electrode UVA lamps = 300-475 nm I = 8 10-10 Ein/s cm2 Cont. phase = water Application of a sma ll potential bias can significantly increased bacteria photocatalytic disinfection rate Nature of titania electrode affected disinfection rate and order of the reaction Catalytic activity per unit surface area of catalyst is many orders of magnitude greater than titania slurries Harper et al. (2001) Degussa P25 Black light = 320-420 nm I = 67.9 E/s m2 Cont. phase = water Complete inactivation observed within 60 min. Optimum TiO2 concentration was 1 mg/mL Increased light intensity increase inactivation activity Bekbolet (1997) Degussa P-25f Black light = 300 ~ 400 nm PI = 360 nm Cont. phase = water Optimum TiO2 concentration for bacteria inactivation is 1 mg/mL. Complete inactivation achieved in 60 min. Bekbolet and Araz (1996) Escherichia coli Veg. cells Immobilized Degussa P25 layer UV lamp = 300-400 nm I = 0.9-6.2 10-9 Einstein/cm2 s2 Cont. phase = water 99.6% bacteria was inactivated within 6hr. at maximum light intensity Salts found to decrease inactivation rate. Microbe viability was directly proportional to the initial microbe concentration Increased intensity improved inactivation activity Belhacova et al. (1999)

PAGE 238

220Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference 1 mm thick TiO2 film on titanium plate UV-B lamp Electric field enhancement applied Cont. phase = water Over 3 log10 reduction achieved in 25 min. Butterfield et al. (1997) Degussa P25 Black light = 300420 nm Cont. phase = water OH radicals is found to be the main agent for the bacteria inactivation Concentration of OH radicals was quantitative demonstrated to be linearly related to bacteria inactivation Cho et al. (2004) Escherichia coli Veg. cells TiO2-Fe2O3 membrane UV lamp Cont. phase = water Photomineralization rate followed pseudo-firstorder kinetics with respect to dissolved oxygen concentration Surface interaction between adsorbed cleavage bacterial cells and hydroxyl radicals considered to be rate-controlling step 99% removal percentage observed at high concentration Darren et al. (2003)

PAGE 239

221Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Degussa P-25 Mercury highpressure lamp TQ718-Z4g Mercury highpressure lamp TQ718h Cont. phase = water 5-log10 reduction of bacteria observed in1 hr. irradiation by TQ 718 lamp. TQ718-Z4 showed lower specific inactivation rate. Dillert et al. (1998) TiO2 electrode (Degussa P25i ; Aldrichj) Xenon arc lamp Electric field enhancement applied Cont. phase = water Disinfection was not observed in the absence of UVA and/or TiO2. Percentage killing rate increased with increase light intensity. Percentage killing rate increased with increased initial cell density. 99.996% inactivation achieved after 120 min. irradiation Regrowth of bacteria not detected Degussa P25 electrode showed higher percentage killing rate than Aldrich electrode Electric field enhanced photocatalysis increased 40% percentage killing rate than Degussa P25 electrode alone and 80% than Aldrich electrode alone. Dunlop et al. (2002)

PAGE 240

222Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Aerosil P25k High-pressure mercury lamp PI = 365 nm; 405 nm Cont. phase = water Specific inactivation rate increased with increased light intensity Optimum TiO2 concentration for disinfection is 0.1kg/m3 Horie et al. (1996) TiO2 particle (10 nm) and TiO2 film Sunshine Cont. phase = air No inactivation observed in the absence of TiO2 Cell viability decreased monotonically with TiO2 concentration increase Cell viability decreased monotonically with light intensity Huang et al. (1998) Degussa P25l Black light PI = 365 nm I = 8 W/m2 Cont. phase = water 12% survival achieved within 20 min. Intermittent illumination resulted same survival rate as continuous illumination within 60 min. TiO2 has continued bacteric idal activity after UV irradiation terminated. Huang et al. (2000) TiO2 coated fiberglass mesh UV lamp with coaxially wrapped TiO2 coated fiberglass mesh =300 400 nm Cont. phase = water Little or no inactivation observed in thiosulfate system 9 to 10 log10 reduction achieved in system without thiosulfate Ireland et al. (1993)

PAGE 241

223Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference TiO2 (Degussa P25) film UV ( = 254nm) Near UV ( = 356 nm) Cont. phase = water No bactericidal effect of TiO2 observed without irradiation Inactivation of bacteria observed with UV alone No bactericidal effect of TiO2 observed with Near UV alone CO2 evolution observed due to the photocatalytic oxidation of bacteria under irradiation Jacoby et al. (1998) TiO2 nanoparticles Fluorescent light Cont. phase = water Inactivation of bacteria was proportional to the anatase mass fraction of TiO2 and inversely to the particle size. Jang et al. (2001) Silver coated TiO2 m Preparation of Ag-coated TiO2 made by exposure to = 354 nm for 2hr. Inactivation experiment done without light Cont. phase = water Minimum inhibitory concentration (MIC) for Agcoated TiO2 was 6.4 g/mL. MIC for Ag metal was 2500 g/mL. MIC for AgNO3 was 3.9 g/mL. Keleher et al. (2002)

PAGE 242

224Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference 0.4 m thick TiO2 film on glass plate Black light PI = 360 nm I = 1.0 mW/cm2 Cont. phase = water 50% reduction observed in 4 hr. irradiation (UV alone) LD90 = 0.5 hr. in mixture system (bacteria and TiO2 film mixed and I = 1 mW/cm2) LD90 = 3.5 hr. for separated system (bacteria and TiO2 separated by membrane and I = 0.4 mW/cm2) Mannitol (Hydroxyl radical scavenger) decreased bacteria killing percentage in the mixture system. It didn’t affect killing percentage in separated system. Hydrogen peroxide thought to be the main agent for bacteria killing That bacteria survival ratio at pH 7.4 higher that at pH 4-5 thought to be caused by the slow production of hydrogen peroxide. Kikuchi et al. (1997) TiO2 (Degussa P25) coated Plexiglas UVA light PI = 356 nm Cont. phase = water 6-log10 reduction observed in 60 min. irradiation Inactivation of bacteria s uggested to be the result of cell wall damage caused by hydroxyl free radicals Kuhn et al. (2003)

PAGE 243

225Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference TiO2 n Black light PI = 350 nm Cont. phase = water Inactivation rate affected by DO, pH, temperature, light intens ity and irradiation time Light intensity was proportional to bacteria destruction rate Increased DO increased specific inactivation rate and reached an optimum concentration range 4-5 mg/L Lower cell loading has higher removal percentage Li et al. (1996) TiO2 powder ZnO powder UV light PI = 365 nm I = 20 W/m2 Cont. phase = water ZnO found to be effective on bacteria inactivation than TiO2 under similar experimental conditions Inactivation efficiency with air as purging gas higher than nitrogen as purging gas Liu and Yang (2003) Commercial TiO2 filterc Black light PI = 365 nm I = 7.1 mW/cm2 I = 31.8 mW/cm2 Cont. phase = air Germicidal capability for airborne microorganism was almost zero for TiO2 photocatalyst filter media Lin and Li (2003 a) Nickel microfibrous mesh coated with TiO2 PI = 365 nm I =11 mW/cm2 Cont. phase = air Microfibrous mesh coated with titania used to filter bacteria. Coated mesh recovered by exposing to UV. Lopez and Jacoby (2002)

PAGE 244

226Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference TiO2 coated glass slide High-pressure mercury lamp > 300 nm I = 0.7 mW/cm2 Cont. phase = water Potassium ion (K+) was observed Cell wall and cell membrane decomposition observed Lu et al. 2003 Degussa P25o Black light PI = 356 nm I = 8 W/m2 Cont. phase = water TiO2 optimum concentration is 1 mg/ml TiO2 photocatalysis was considered to promote peroxidation of the polyu nsaturated phospholipid component of lipid membrane and induce membrane disorder, thus, cause the cell death Maness et al. (1999) Plantium-loaded titania (TiO2/Pt)p Metal halide lamp I = 4600Einstein/(m s) Cont. phase = water 20% survived after 60 min illumination Bacteria sterilized in 120 min. irradiation with initial concentration 103 CFU/mL Direct photochemical oxidation of CoA thought responsible for the death of bacteria Matsunaga et al. (1985) TiO2 immobilized membrane Mercury lamp I = 1100 microeinsteins/m2 s Cont. phase = water Bacteria (102 cells/ml) inactivated to less than 1% for within 16 min. residence time Surviving ratio decreased with increased light intensity Surviving ratio decreased with increased TiO2 amount Matsunaga et al. (1988)

PAGE 245

227Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Aerosil P-25q Diffuse-light emitting optical fibers (DLEOF) Cont. phase = water TiO2 itself was nontoxic to the bacteria. Bacteria with concentration 2 103 CFU/mL disinfected in 2 hr. Optimum TiO2 concentration for disinfection was 2.5 g/mL. Survival ratio decreased with increased light intensity. Light supply showed to be the limiting for the disinfection at high TiO2 concentration Cell percentage killing by DLEOFs increased four times as the conventional optical fibers at high TiO2 concentration (250 g/mL). Matsunaga and Okochi (1995) Degussa P-25 Suntest lamp I = 80 mW/cm2 Cont. phase = water 7 log10 reduction observed within 20 min by photocatalytic inactivation. 7 log10 reduction observed within 70 min by UV alone. Rincon et al. (2001)

PAGE 246

228Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Degussa P25r; Degussa P25 coated Nafion membrane; Degussa P25 immobilized on Pyrex glass; TiO2 (Bayer, Germany); TiO2 (Aldrich) Lamp approximating solar spectrums I = 400-1000 W/m2 Cont. phase = water No disinfection detected without irradiation. Decreased specific inactivation rate observed under intermittent illumination compared to continuous illumination No bacterial regrowth observed after stopping irradiation Specific inactivation rate of bacteria increased with increased light intensity Specific inactivation rate of Gram-positive bacteria increased with increased temperature. Specific inactivation rate of Gram-negative bacteria decreased with increased temperature. Water turbidity decreased specific bacteria removal rate. Higher specific inactivation rate for suspended TiO2 than fixed TiO2 Degussa P25 achieved the highest specific inactivation rate among three commercial TiO2 Rincon and Pulgarin (2003)

PAGE 247

229Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Anatase TiO2 powder (SigmaAldrich); Immobilized TiO2 on glass slide Sunlight Cont. phase = water Immobilized TiO2 showed lower specific inactivation rate than suspended TiO2. Optimum TiO2 concentration for disinfection is 1mg/mL. Cell percentage killing increased at high concentration of TiO2. Dimethyl sulphoxide and cysteamine, hydroxyl radical scavengers, eliminated bactericidal of TiO2. Free radicals considered to be involved in the inactivation of bacteria Salih (2002) Degussa P25 Fluorescent lamp I = 270 E/s m2 Cont. phase = water 50% inactivation achieved within 60 min. in recirculation reactor with combined photocatalytis and sonolysis 50% inactivation achieved within 24 min. in reactor with static solution under combined photocatalytis and sonolysis Sonolysis found to enhance photocatalytic inactivation of microbe Improvement by sonolysis was more modest for batch recirculation reactor than reactor with static solution Stevenson et al. (1997)

PAGE 248

230Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference 0.4 m thickness TiO2 film Black light I = 0.4 mW/cm2 Cont. phase = water Complete inactivation occurred within 2 hr. Sunada et al. (1998) TiO2 film I = 1.0 mW/cm2 Cont. phase = water Inactivation of bacteria was observed to be an initial lower rate photok illing step followed by a higher rate one Cell wall is considered to be major barrier to photocatalytic inactivation Sunada et al. (2003) TiO2-Fe2O3 membrane UV lamp PI = 253.7 nm Cont. phase = water photomineralization rate followed pseudo-firstorder kinetics with respect to the dissolved oxygen concentration Maximum inactivation rate observed at dissolved oxygen concentration of 21 mg/L Sun et al. (2003) TiO2 based nanostructured Fe3+ doped coating Mercury lamp Cont. phase = water Appearance of rutile structure in coating showed decrease of bact eria inactivation Single layer coating has lower inactivation activity than double layer Trapalis et al. (2003) TiO2 dispersed in activated carbons Black light Cont. phase = water 99% bacteria killed in 15 min. High crystallinity of TiO2 gave high photocatalytic inactivation Tamai et al. (2002) Charged TiO2WO3 film UV light (used to charge TiO2-WO3 film) Cont. phase = air Survival ratio for charged film was 25%. Survival ratio for discharged film was 58% Tatsuma et al. (2003)

PAGE 249

231Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Ti(IV) doped hydroxyapatite film Black light I =1 mW/cm2 PI = 360 nm Cont. phase = water TiO2 itself showed no inactivation activity in dark Ti (IV) doped hydroxyapatite had bactericidal activity in the dark. Inact ivation activity enhanced under the irradiation of UV. Wakamura et al. (2003) Degussa P-25t Fluorescent lamp and halide lamp Cont. phase = water Oxygen found to be a prerequisite for the TiO2 disinfection. Complete killing observe d within 30 min. Wei et al. (1994) Degussa P-25 Black lamp Cont. phase = water 4 log10 reduction of bacteria achieved with 1.5 h irradiation Bacteria concentration for nature water increased 24 h after terminati on of irradiation. Bacteria concentration fo r distilled water showed no increase 24 h after term ination of irradiation. TiO2 lack of residue for the disinfection Wist et al. (2002) Degussa P25 coated quartz disk Low-pressure near-UV lamp PI = 365 nm Cont. phase = air Higher relative humidity increased the cell oxidation rate, as measured by CO2 production Wolfrum et al. (2002) P25 coated glass spring Black light PI = 365nm Cont. phase = water 1 log10 reduction achieved within 30 min. Zeng et al. (2003)

PAGE 250

232Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Silver-loaded TiO2 Low pressure mercury lamp PI = 294 nm I =5.8W/m2 Cont. phase = water Silver loading (1%, w/w) dramatically decreased irradiation period for complete inactivation. Sokmen et al. (2001) Anatase titania films coated on glass UVlamp PI = 365 nm I = 0.63 mW/cm2 Cont. phase = water Film showed significant bactericidal activity toward the bacteria Titania film prepared by reverse micelle method is better than the film prepared by sol-gel method in gaseous phase oxidation but same in aqueous phase Yu et al. (2002) TiO2 immobilized at reactor bottom Germicidal lamp Black light lamp I = 0.2 mW/cm2 (for germicidal UV at 254 nm) I = 7.0 mW/cm2 (for black light at 360 nm) Cont. phase = water No significant difference in E. coli inactivation with and without titania under either UV or black light irradiation Otaki et al. (2000)

PAGE 251

233Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference TiO2 coated Pyrex rod Real sunlight Cont. phase = water Compound parabolic reflect or in the reactor promoted inactivation than the parabolic and Vgroove profiles used in the photolysis reactor Using TiO2 coated Pyrex rod in the reactor yielded a slight enhancement in the compound parabolic reactor but no im provement in either the parabolic or V-groove reactors McLoughlin et al. (2004b) TiO2 coated plastic sheet Natural sunlight Cont. phase = water Borosilicate glass and PET plastic bactch-process solar disinfection reactor (SO-DIS) fitted with flexible TiO2 coated plastic sheet showed to be 20% and 25% more effective than standard SODIS reactor Duffy et al. (2004) TiO2 Solar irradiation I = 25W-UV global/m2 Cont. phase = water 4-log reduction was achieved in a low-cost compound parabolic concentr ator prototype with 0.05% (w/w) titania particles Vidal et al. (1999) TiO2 Xenon lamp I = 550 W/m2 PI = 310 – 800 nm Cont. phase = water Bactericidal effect of titania photocatalysis involves adsorption of cells onto aggregated titania followed by loss of membrane integrity Bacterial adsorption was influenced by solution composition and illumination period Gogniat et al. (2006)

PAGE 252

234Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference 13 commercialized TiO2 Suntest solar light simulator I = 70, 100, 140 W/m2 Cont. phase = water Surface charge of titania significantly affected inactivation kinetics Degussa P25 particles in suspension are in partial contact with cell walls The lower isoelectric point of titania, the lower the bactericidal activity Inactivation rate increased with light intensity Gumy et al. (2006) Degussa P25 Suntest solar light simulator Cont. phase = water Addition of titania was more efficient in bacterial inactivation than solar light alone Addition of H2O2 in UV-vis/TiO2 system enhanced photocatalytic disinfection Addition of Fe3+ ions in UV-vis/TiO2 system with low titania concentration enhanced photocatalytic disinfection Effective disinfection time was 40 min for UVvis/TiO2 system, 20 min for UV-vis/TiO2/ H2O2 system Natural water promoted photocatalytic inactivation than deionized water due to chemical matrix of water Rincon and Pulgarin (2006)

PAGE 253

235Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference TiO2 nanorods Black light blue lamps I =7.9 10-6 Einstein/L s Cont. phase = water Inactivation rate using TiO2 nanorods was higher than Degussa P-25 as photocatalyst Joo et al. (2005) Degussa P-25 TiO2 immobilized on inert matrix of reactor Natural solar light Cont. phase = water Immobilized TiO2 has lower efficiency than slurry TiO2 for bacterial inactivation Fernandez et al. (2005) TiO2 af Black light fluorescent lamp I =4-14 W/m2 Cont. phase = water Viability of superoxide dismutase (SOD)deficient mutant decreased linearly with photoreaction time, Viability of wild-type strain showed a curved form Bacterial viability with time were analyzed on series-event model Koizumi et al. (2002) TiO2 immobilized activated charcoal granules (T/AC) Black light fluorescent lamp I = 23 W/m2 Cont. phase = water Sterilization rate depends on temperature Magnitude of sterilizatio n rates corresponds to the order of absorbed amount of cells to the granules Specific inactivation rate constant decided on the basis of the series-event model can be related to T/AC concentration Horie et al. (1998a)

PAGE 254

236Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference TiO2 film on glass tube Germicidal lamp = 254 nm Cont. phase = water Location of titania film in reactor affected inactivation efficiency Inactivation of microbe depended on circulation rate of the solution in the reactor Initial inactivation rate was in positively linear relationship with initial cell concentration Shiraishi et al. (1999) Aerosil P-25k Black light fluorescent lamp I = 0 – 23W/m2 Blue actinic fluorescent lamp I = 0 – 15 W/m2 Daylight fluorescent lamp I = 0 – 1.2 W/m2 Sunlight I = 15, 25 W/m2 Cont. phase = water Inactivation rate increased with light intensity Apparent sterilization rate constants decided from series-event model can be related with light quantity absorbed by titania slurry Time profiles of cell inac tivation by sunlight can be predicted using relationship between rate constant and light quantity absorbed by the titania slurry Horie et al. (1998b) Anatase normalah Nano anataseai Degussa P-25 High pressure Hg lamp Cont. phase = water Degussa P-25 achieved highest destruction activity, whereas Anatase normal showed small effect. Nano anatase showed medium inactivation activity Allen et al. (2005)

PAGE 255

237Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Escherichia coliaj Veg. cells TiO2 ag Black light fluorescent lamp I =15W/m2 Cont. phase = water Apparent rate constant for cell deactivation was constant at 0.526 min-1 for the initial cell density range of 106-109 cells/m2 Sato et al. (2003) Escherichia coli and Bacillus sp. Veg. cells TiO2 r Black actinic light PI = 366nm Cont. phase = water E. coli is more sensitive than Bacillus sp. to photocatalytic inactivation in a mixture of E. coli and Bacillus sp. Bacterial inactivation de pended on organic matter and dissolved oxygen Presence of oxygen enhanced inactivation of the mixed culture E. coli was the most sensitive to photocatalytic treatment compared to other bacteria groups (Enterococcus sp., total Gram-negative and other coliforms) No bacterial recovery was observed during the subsequent 24 h in the dark Rincon and Pulgarin (2005) Escherichia coliform Veg. cells TiO2-Fe2O3 nanocomposite Low-presure mercury light = 600nm Ev = 500 Lux Cont. phase = water Dissolved oxygen, hydraulic retention time and bacterial concentrate affected photocatalytic inactivation efficiency in reactor Photomineralization rate followed pseudo-firstorder kinetics by the role of dissolved oxygen Derived empirical modes were consistent with the proposed reaction pathways Sun et al. (2003)

PAGE 256

238Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Lactobacillus acidophilus Veg. cells Plantium-loaded titania (TiO2/Pt) p Metal halide lamp I = 4600Einstein/(m s) Cont. phase = water Bacteria sterilized unde r metal halide lamp illumination for 60min. Decrease of coenzyme A (CoA) in cell observed under irradiation Direct photochemical oxidation of CoA thought responsible for the death of bacteria Matsunaga et al. (1985) Lactobacillus helveticus Veg. cells TiO2 powder ZnO powder UV light PI = 365 nm I = 20 W/m2 Cont. phase = water ZnO found to be effective on bacteria inactivation than TiO2 under similar experimental conditions Inactivation efficiency with air as purging gas higher than nitrogen as purging gas Liu and Yang (2003) TiO2 filter UV germicidal light (UVGL) PI = 253.7 nm Black light PI = 365 nm I = 7.4 mW/cm2 Cont. phase = air 5 log10 reduction achieved with UV germicidal irradiation dose 289 to 860 W sec/cm2 for UVGL alone Bacteria penetration was 0.1 with black light irradiation and relative humidity 55% Li et al. (2003) Legionella pneumophila Veg. cells TiO2 / SiO2 fiber Low-pressure Hg lamp PI = 254 nm I = 100 W/m2 Cont. phase = water Viable microbes were not detected after 3 h irradiation with TiO2 / SiO2 fiber while 91.4% bacteria survived after 7 h irradiation with UV light alone Coronado et al. (2005)

PAGE 257

239Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Listeria monocytogenes Veg. cells Aerosil TiO2 u UV lamp PI = 360 nm I = 4 W/m2 Cont. phase = water 87% reduction of bacteria achieved after 3h irradiation Kim et al. (2003) Microbacterium sp. Veg. cells Membrane filtered with TiO2 r (batch inactivation) Polyethersulfone membrane disc filter coated with TiO2 r (continuous flow reactor) UV-A lamp I = 1.82, 4.28, 6.28 mW/cm2 PI = 365 nm Cont. phase = air Significant bacterial inac tivation occurred under UV alone (without titania) Enhanced inactivation ra te was observed in the presence of titania Inactivation rate in the continuous system were higher than in batch reactor Pal et al. (2005) Micrococcus Iuteus Veg. cells Degussa P25 coated on quartz disk (3.8 cm) Low-pressure near-UV lamp PI = 365 nm Cont. phase = air Higher relative humidity increased the cell oxidation rate, as measured by CO2 production Wolfrum et al. (2002) Pseudomonas aeruginosa Veg. cells TiO2 (Degussa P25) coated Plexiglas UVA light PI = 356 nm Cont. phase = water 6-log10 reduction observed in 60 min. irradiation Inactivation of bacteria s uggested to be the result of cell wall damage caused by hydroxyl free radicals Kuhn et al. (2003)

PAGE 258

240Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Undoped TiO2; Cu doped TiO2; Al doped TiO2; Black light lamp PI = 356 nm Cont. phase = water Cell percentage killing ranged from 28% to 96% for different film. Amezaga et al. (2002) Degussa P-25e UVA light PI = 365 nm I = 1.4 mW/cm2 Cont. phase = water 3 log10 reduction observed after 40 min. irradiation Ibanez et al. (2003) 98 nm thick TiO2 film 192 nm TiO2 thick film UV light PI = 365 nm Cont. phase = air 32% inactivation observed within 120min. Amezaga-Madrid et al. (2003) TiO2 nanoparticles Fluorescent light Cont. phase = water Inactivation of bacteria was proportional to the anatase mass fraction of TiO2, and inversely to the particle size. Jang et al. (2001) TiO2 coated glass UVC light Cont. phase = water TiO2 coated glass reduced adhesion of bacteria onto the slides Bacteria was inactivated when exposed to TiO2 coated glass with UVC light Ali et al. (1999) Degussa P25 UV low pressure mercury lamp I = 29 W/m2 Cont. phase = water Disinfection achieved in 4 hr. at 0.01% TiO2 Cooper et al. (1998)

PAGE 259

241Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference TiO2 i immobilized on acetate sheet Xenon arc solar simulating lamp I = 870 W/m2 (300 nm – 10 m) Cont. phase = water Total inactivation occurred at 2 h in both solar and photocatalytic disi nfection batch process reactor Lonnen et al. (2005) Degussa P25, ZnO Sahara desert dust Sodium lamp Cont. phase = water Complete inactivation occurred in 40 min Sahara desert dust has no microbicidal effect under photolysis Seven et al. (2004) Pseudomonas fluorescens R2f Veg. cell TiO2 pellet Low mercury fluorescent lamp = 300 ~ 460 nm PI = 365 nm 1 log10 reduction observed w ithin 60 min. when exposed to mercury lamp 4 log 10 reduction observed w ithin 50 min. when exposed to mercury lamp Addition of titania to solar irradiated water did not improve inactivation activity Kersters et al. (1998) Pseudomonas stutzeri Veg. cells TiO2 v Black light PI = 370 nm I = 8.1 W/m2 Specific inactivation rate increased with increased titania concentraion Biguzzi and Shama (1994)

PAGE 260

242Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Salmonella choleraesuis subsp. Veg. cells Aerosil TiO2 u UV lamp PI = 360 nm I = 4W/m2 Cont. phase = water Complete killing achieved in3 h i rradiation. Kim et al. (2003) Degussa P-25e UVA light PI = 365 nm I = 5.5 mW/cm2 Cont. phase = water 4 log10 reduction observed after 40 min. irradiation Ibanez et al. (2003) Salmonella typhimurium Veg. cells 21 nm anatase (P25) w 255 nm anatase (WA)x 255 nm rutile (WR)y 420 nm rutile (TP3)z Sunlight simulator with filter that allows 50% transmission at = 335 nm. UVA: UVB = 25:1 Intensity = 0.2,0.4, 0.8, 1.6 mW/cm2 Cont. phase = water TiO2 showed no or weak genotoxicity without irradiation. TiO2 exhibited strong genotoxicity under UV illumination. WR shows no inactivation on the cells. Other two anatase and one rutile form exhibited photoenhanced damage to the cell. P-25 gave the highest cel l percentage killing Nakagawa et al. (1997)

PAGE 261

243Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Degussa P25aa UV low pressure mercury lamp PI = 350 nm Cont. phase = water TiO2 had no bactericidal e ffect without UV light. Optimum concentration of TiO2 for killing bacteria is 0.01% Stirring increased inactivation rate. Alkaline solution reduced the bactericidal activity of TiO2. Block et al. (1997) Degussa P25 UV low pressure mercury lamp I = 29 W/m2 Cont. phase = water Disinfection achieved in 4 hr. at 0.01% TiO2 Cooper et al. (1998) Degussa P25aa Low-pressure mercury lamp PI = 350 nm Disinfection of the bacteria in the air Cont. phase = water Relative humidity < 30% or > 85% showed lower percentage killing rate. Relative humidity of 50%, complete inactivation was observed. High UV radiation intensity caused high percentage killing rate Goswami et al. (1997 a) Serratia marcescens Veg. cells Immobilized TiO2 UV light = 300 400 nm PI = 365 nm Cont. phase = air 99.9% bacteria destructi on occurred within 3 min in a recirculation loop. Goswami et al. (1997 b)

PAGE 262

244Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Photocatalystcoated aluminum disk UV-A Disinfection of the bacteria in the air Cont. phase = air Photocatalytic technolog y showed disinfection activity Goswami (2003) Silver coated TiO2 ab Preparation of Ag-coated TiO2 made by exposure to = 354 nm for 2hr. Inactivation experiment done without light Cont. phase = water MIC for Ag-coated TiO2 was 3.9 g/mL. MIC for Ag metal was 4900 g/mL. MIC for AgNO3 was 2.5 g/mL. Keleher et al. (2002) Degussa P25 coated Plexiglas UVA light PI = 356 nm Cont. phase = water 6-log10 reduction observed in 60 min. irradiation Inactivation of bacteria s uggested to be damage of cell wall caused by hydroxyl free radicals Kuhn et al. (2003) Staphylococcus aureus Veg. cells Degussa P25, ZnO Sahara desert dust Sodium lamp Cont. phase = water Complete inactivation occurred in 120 min Sahara desert dust has no microbicidal effect under photolysis Seven et al. (2004)

PAGE 263

245Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Streptococcus cricetus HS-6 Veg. cell TiO2 d Fluorescent lamp PI = 578 nm Cont. phase = water Viability decreased when TiO2 exposed to light Optimum inactivation concentration of TiO2 was 0.1% (w/v) Nagame et al. (1989) Streptoccocus faecalis Veg. cell Degussa P25 UV lamp Solar light Cont. phase = water Solar disinfection result s are similar to those obtained with UV lamp Similar inactivation yields observed as E. coli Bacteria regrowthed after stop irradiation Melian et al. (2000) Streptococcus rattus BHT Veg. cell TiO2 d Fluorescent lamp PI = 578 nm Cont. phase = water TiO2 did not affect bacteria viability in both dark and light. Nagame et al. (1989) Streptococcus rattus FA-1 Veg. cell TiO2 d Fluorescent lamp PI = 578 nm Cont. phase = water Viability of bacteria affected equally by TiO2 in both dark and light. Nagame et al. (1989) Streptococcus mutans Veg. cells TiO2 ac Mercury lamp 99.9% bacteria inactivated within 80 min. TiO2 dispersed by ultrasonic oscillation inactivated 99.9% bacteria within 20 min. Morioka et al. (1988)

PAGE 264

246Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Streptococcus sobrinus AHT Veg. cells TiO2 ad Near-UV light = 300-400 nm PI = 352 nm Cont. phase = water Concentration of TiO2 was considered one of the important factors that de cided the photocatalytic bactericidal action of TiO2. Co-aggregation of bacteria at high bacterial initial concentration and TiO2 observed. Co-aggregation decreased with irradiation. pH of the spore and TiO2 mixture decreased to 4.5 after 120min irradiation. Destruction of bacteria observed under TEM after reaction time 60 min. 1 mg/mL TiO2 is the optimum concentration for the inactivation of bacteria 105 CFU/mL cells killed completely within 1 min. Leakage of potassium ions from the bacteria happened parallel to inactivation of bacteria. Saito et al. (1992) Vibrio parahaemolyticus Veg. cells Aerosil TiO2 u UV lamp PI = 360 nm I = 4W/m2 Cont. phase = water Complete killing achieved in 3 hr. irradiation. Optimum concentration of TiO2 was 1mg/mL. Kim et al. (2003)

PAGE 265

247Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Viruses Bacteroides fragilis bacteriophages Virus Degussa P25 UV lamp I = 7 mW/cm2 PI = 350 nm Cont. phase = water Inactivation rate was similar under either continuous or intermittent illumination Laot and Narkis (1999) Bacteriophage NM 1149 Virus Immobilized Degussa P25 layer UV lamp = 300-400 nm I = 0.9-6.2 10-9 Einstein/cm2 s2 Cont. phase = water Bacteriophage was inactiva ted to below detection limit (<10 PFU) after 3 hr. Salts found to decrease inactivation rate. Microbe viability was directly proportional to the initial microbe concentration Increased intensity improved inactivation activity Belhacova et al. (1999) Lactobacillus casei Phage PL-1 Virus 100 nm thick TiO2 (primary anatase) film Black light I = 1.9-2.0 W/m2 PI = 365 nm Cont. phase = water Genome DNA inside the phage fragmented by the photocatalytic action of TiO2 That Active oxygen species caused damage to the capsid protein, then the DNA inside the phage was thought to be the mechanism. Kashige et al. (2001) Lactobacillus phage PL-1 Virus Silica sand coated with mixture of SiO2, Al2O3, TiO2 and Ag Fluorescent lamp Cont. phase = water Inactivation of phage show ed first-order reaction kinetics. Dose of coated silica sand increased the inactivation rate. Kakita et al. (1997)

PAGE 266

248Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Degussa P25 UV lamp I = 7 mW/cm2 PI = 350 nm Cont. phase = water Inactivation rate was more effective under intermittent illumination than continuous irradiation Laot and Narkis (1999) Degussa P25 Black light = 300-420 nm I = 7.9 E/L/s Cont. phase = water MS-2 phage is mainly inactivated by free hydroxyl radical in solution bulk Cho et al. (2005) TiO2 P25af Black light fluorescent lamp I = 23 W/m2 Cont. phase = water Coexistence of NO3 -, SO4 2-, PO4 3-, K+ each at 10100 mM decreased the rate constant for phage inactivation, whereas Cl-, Bror Na+ did not decrease rate constant Inhibitory effects of ions was elucidated by the relation between the rate constants and quantities of phage on titania regardless the kinds of existing ions Koizumi and Taya (2002b) MS2 Virus Mixture of anatase and rutile-type TiO2 Black light fluorescent lamp I = 14, 23 W/m2 Cont. phase = water Specific inactivation rate was maximized at anatase ratio of 70 wt% Close contact of both type of titania enhanced quantum yield and increased generation of reactive oxygen species Sato and Taya (2006)

PAGE 267

249Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference TiO2 P25 ae Black light I = 8,14,23 W/m2 Cont. phase = water Viability of phage MS2 decreased linearly with irradiation time. Specific inactivation rate proportional to the light intensity and concentration of TiO2 Specific inactivate rate of MS2 showed a convex feature in the initial pH range of 3-10. Koizumi and Taya (2002 a ) TiO2 P25 af Black light I =23 W/m2 Cont. phase = water Coexistence of NO3 -, SO4 2-, PO4 3-, K+ or Ca 2+ decreased photocatalytic inactivation rate of MS2. Koizumi and Taya (2002 b ) Mixture of TiO2 (P25) and iron (FeSO4 7H20) UV lamp PI = 365 nm I = 2 mW/cm2 Cont. phase = water 90% inactivation of Phage MS2 increased to 99.9% after the addition of 2 m ferrous sulfate. Hydroxyl radical oxida tion and Fenton reaction enhancement was thought the probably agent for the inactivation of virus. Sjogren and sierka (1994) Phage Q Virus TiO2 coated tile Near-UV black light (I = 3.6-3 W/m2 , = 300400 nm); Germicidal lamp ( PI = 254 nm) Cont. phase = water 2.2-log10 reduction achieved in1 hr. Specific inactivation rate constant is proportional to the light intensity Lee et al. (1997)

PAGE 268

250Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Degussa P25 UV lamp PI = 254 nm I = 0.4 mW/cm2 Cont. phase = water 3.5 log10 reduction achieved within 2 min. by photocatalytic activation 2 log10 reduction achieved within 2 min. by UV alone Lee et al. (1998) TiO2 immobilized at reactor bottom Germicidal lamp Black light lamp I = 0.2 mW/cm2 (for germicidal UV at 254 nm) I = 7.0 mW/cm2 (for black light at 360 nm) Cont. phase = water No significant differenc e in phage inactivation with and without titania under germicidal UV irradiation Inactivation was enhanced by titania under black light irradiation Otaki et al. (2000) Poliovirus 1 Virus Anatase TiO2 (Fisher Scientific) Sunlight Black light Cont. phase = water 2 log10 reduction of Poliovirus 1 achieved in 30 min. No difference between disinfection for pH 5~8 Specific inactivation rate by sunlight were lower than Black light Watts et al. (1995)

PAGE 269

251Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Other microbes Acanthamoeba polyphaga Tropho zoite stage, cyst stage TiO2 i immobilized on acetate sheet Xenon arc solar simulating lamp I = 870 W/m2 (300 nm – 10 m) Cont. phase = water Both solar and solar pho tocatalytic disinfection can achieved at least 4-l og reduction in viability of the trophozoite stage of A. polyphaga Both solar and solar pho tocatalytic disinfection were not effective against cyst stage of A. polyphaga . Lonnen et al. (2005) Airborne microorganism TiO2 UV Cont. phase = air 98% destruction achieved using combination of TiO2/UV compared 30% inactivation by UV alone Kondo et al. (2003) Anabaena Veg. cells TiO2 coated pyrex hollow glass beads Black light UV lamp PI = 370 nm I = 0.6 mW/cm2 Cont. phase = water String of cells was completely separated into individual spherical one, and most of them lost photosynthetic activity Complete photocatalytic inactivation was achieved in 30 min Kim and Lee (2005) Artemia salina plankto n TiO2 Black light PI = 365 nm I = 2 mW/cm2 Cont. phase = water A. salina stopped moving and the body was fully covered by titania powder after 1 hr irradiation Matsuo et al. (2001)

PAGE 270

252Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Degussa P25 coated quartz disk Low-pressure near-UV lamp at peak intensity ( PI) = 365 nm Cont. phase = air Higher relative humidity increased the spore oxidation rate, as measured by CO2 production Wolfrum et al. (2002) Aspergillus niger Sporesa Degussa P25, ZnO Sahara desert dust Sodium lamp Cont. phase = water Inactivation was not occu r under either Degussa P25 or ZnO Sahara desert dust has no microbicidal effect under photolysis Seven et al. (2004) Aspergillus niger Spores Photocatalystcoated aluminum disk UV-A Disinfection of the bacteria in the air Cont. phase = air Photocatalytic technolog y showed disinfection activity Goswami (2003) Bacteria/Fungi Veg. cells TiO2 plasma air cleaning filter Pulsed discharge plasma = 310-380 nm Cont. phase = air Bacteria and fungi removed rapidly in airtight room Lee (2003) Bacteria Veg. cells TiO2 UV lamp and electrolysis system PI = 253.7 nm Cont. phase = water 90% bacteria inactivated within 5 min. Chen et al. (2003)

PAGE 271

253Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Veg. cells Degussa P25 coated Plexiglas UVA light PI = 356 nm Cont. phase = water 2-log10 reduction observed in 60 min. irradiation Inactivation of yeast sugg ested to be damage of cell wall caused by hydroxyl free radicals Resistance to the photocat alytic activity of TiO2 was in the order: Candida albicans > Enterococcus faecium > Staphylococcus aureus > Pseudomonas aeruginosa > Escherichia coli Kuhn et al. (2003) Veg. cell TiO2 d Fluorescent lamp PI = 578 nm Cont. phase = water TiO2 did not affect bacteria viability in both dark and light. Nagame et al. (1989) Veg. cells TiO2 i immobilized on acetate sheet Xenon arc solar simulating lamp I = 870 W/m2 (300 nm – 10 m) Cont. phase = water Total inactivation occurred at 4 h in photocatalytic disinfecti on batch process reactor and at 6 h in solar disinfection batch process reactor Lonnen et al. (2005) Candida albicans Veg. cells Degussa P25, ZnO Sahara desert dust Sodium lamp Cont. phase = water Complete inactivation occurred in 120 min Sahara desert dust has no microbicidal effect under photolysis Seven et al. (2004)

PAGE 272

254Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Candida famata var. flareri Veg. cells Commercial TiO2 filterc Black light PI = 365 nm I = 7.1 mW/cm2 I = 31.8 mW/cm2 Cont. phase = air Germicidal capability for airborne microorganism was almost zero for TiO2 photocatalyst filter media Lin and Li (2003 a) Chattonella antiqua flagellat e TiO2 Black light PI = 365 nm I = 2 mW/cm2 Cont. phase = water Deformation of the body was induced in 20 min irradiation, cells burst and came to annihilation after prolonged UV irradiation Matsuo et al. (2001)

PAGE 273

255Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Chinese hamster CHL/IU cells Veg. cells 21 nm anatase (P25) w 255 nm anatase (WA)x 255 nm rutile (WR)y 420 nm rutile (TP3)z Sunlight simulator with filter that allows 50% transmission at = 335 nm UVA:UVB= 25:1 Intensity = 0.2, 0.4, 0.8 ,1.6 mW/cm2 Cont. phase = water TiO2 showed no or weak genotoxicity without irradiation TiO2 exhibited strong genotoxicity under UV illumination Three of the four materials tested caused photoenhanced damage to cells. WR did not inactivate cells. P-25 gave the highest cel l percentage killing Nakagawa et al. (1997) Chlorella vulgaris Veg. cells Plantium-loaded titania (TiO2/Pt)p Metal halide lamp I = 4600Einstein/(m s) Cont. phase = water 85% algae survived after 60 min. illumination. 55% algae survived after 120 min. irradiation. Decrease of coenzyme A (CoA) in cell observed under irradiation Direct photochemical oxidation of CoA thought responsible for the death of microbe Matsunaga et al. (1985)

PAGE 274

256Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Cryptosporidium parvum oocytes TiO2 immobilized at reactor bottom Germicidal lamp Black light lamp I = 0.2 mW/cm2 (for germicidal UV at 254 nm) I = 7.0 mW/cm2 (for black light at 360 nm) Cont. phase = water Inactivation was enhanced by titania under both germicidal UV and black light irradiation Otaki et al. (2000) Fusarium solani Fungi TiO2 i immobilized on acetate sheet Xenon arc solar simulating lamp I = 870 W/m2 (300 nm – 10 m) Cont. phase = water Total inactivation occurred at 4 h in photocatalytic disinfecti on batch process reactor and at 8 h in solar disinfection batch process reactor Lonnen et al. (2005) Hansenula polymorpha Veg. cells Degussa P25 Fluorescent lamp I = 270 E/s m2 Cont. phase = water 50% inactivation achieved within 120 min. in reactor with static solution Sonolysis found to enhance photocatalytic inactivation of microbe Improvement by sonolysis was more modest for batch recirculation reactor than reactor with static solution Stevenson et al. (1997)

PAGE 275

257Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Human bladder cell 24 Veg. cells 30 nm anatase (P25) ag Xenon lamp Cont. phase = water Increase of Ca2+ concentration in cell observed at irradiation 4 min., followed by second elevation at irradiation 6 min. Distribution of TiO2 in T24 cell observed with TEM Sakai et al. (1994) Microcystis Veg. cells TiO2 coated pyrex hollow glass beads Black light UV lamp PI = 370 nm I = 0.6 mW/cm2 Cont. phase = water Colonies of cells was completely separated into individual spherical one, and most of them lost photosynthetic activity Complete photocatalytic inactivation was achieved in 30 min Kim and Lee (2005) Melosira Veg. cells TiO2 coated pyrex hollow glass beads Black light UV lamp PI = 370 nm I = 0.6 mW/cm2 Cont. phase = water Photocatalytic inactivation efficiency was lower for Melosira than Anabaena and Microcystis Kim and Lee (2005)

PAGE 276

258Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Mouse lymphoma L5178Y cells Veg. cells 21 nm anatase (P25) w 255 nm anatase (WA)x 255 nm rutile (WR)y 420 nm rutile (TP3)z Sunlight simulator with filter that allows 50% transmission at = 335 nm. UVA:UVB = 25:1 Intensity = 0.2,0.4, 0.8 ,1.6 mW/cm2 Cont. phase = water TiO2 showed no or weak genotoxicity without irradiation. TiO2 exhibited strong genotoxicity under UV illumination. WR shows no inactivation on the cells. Other two anatase and one rutile form exhibited photoenhanced damage to the cell. P-25 gave the highest cel l percentage killing Nakagawa et al. (1997) Black light PI = 365 nm I = 7.1 mW/cm2 I = 31.8 mW/cm2 Cont. phase = air Germicidal capability for airborne microorganism was almost zero for TiO2 photocatalyst filter media Lin and Li (2003 a) Penicillium citrinum Spores Commercial TiO2 filterc Fluorescent black light PI = 365 nm I = 240 W/cm2 I = 740 W/cm2 I = 1400 W/cm2 I = 2100 W/cm2 Cont. phase = air Increased light intensity increased inactivation efficiency 90% inactivation achieved in 8.35 hr. under light intensity of 740 W/cm2 Lin and Li (2003 b)

PAGE 277

259Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Photosynthetic bacteria Veg. cells TiO2 film Low-pressure mercury lamp Fluorescent lamp Cont. phase = air Cell density decreased with increase of UVdosage Hong and Otaki (2003) Saccharomyces cerevisiae Veg. cells Plantium-loaded titania (TiO2/Pt)p Metal halide lamp Xenon lamp White fluorescent lamp Cont. phase = water Bacteria sterilized unde r xenon lamp illumination in 120min. with ini tial concentration 103 CFU/mL Inactivation of bacteria not observed under illumination without TiO2/Pt. Specific inactivation rate increased with the increase of the light intensity for metal halide and xenon lamps pH value of the solution did not change during the irradiation. Destruction of cell wall not observed. Result suggested hydrogen peroxide and free radicals were not respon sible for the inactivation of bacteria. Decrease of coenzyme A (CoA) in cell observed under irradiation Direct photochemical oxidation of CoA thought responsible for the death of bacteria Matsunaga et al. (1985)

PAGE 278

260Table A-1. Continued Microbe Type Photocatalyst Other experimental conditions Result/Conclusion Reference Degussa P25, ZnO Sahara desert dust Sodium lamp Cont. phase = water Complete inactivation occurred in 120 min with Degussa P25 and 40 min with ZnO Sahara desert dust has no microbicidal effect under photolysis Seven et al. (2004) Strongyloides Stercolaris Veg. cells TiO2 Low pressure mercury vapor lamp = 254 nm. Cont. phase = water Complete inactivation was achieved in UV/TiO2 process. Al-Bastaki (2004) Viable cell Veg. cells TiO2 and apatite coated fibrous ceramics Black light I = 1.0 mW/cm2 Cont. phase = water 57% was inactivated within 6 hr. Nonami et al. (2003) Specific inactivation rate = dN1 dtN , where N = number concentration of cells and t = time aNot clear from reference whether a purified suspension of spores or a mixture of spores and vegetative cells was used bFrom Sandia National Lab., Albuquerque, NM, deduced from the context, TiO2 is powder cTiO2 filter: DAIKIN, Air filter no. 1119589, Japan d99.98% rutile, particle size 1.44 m, specific gravity 4.2, Kobe Steel Co. Japan eDegussaP25: 80% anatase, 20% rutile from e xperimental analysis; Degussa A. G. Germany

PAGE 279

261fDegussa P25: anatase type, average size 30 nm, BET surface area 55 m2/g gQ718-Z4 lamp: sheathed by Jena glass (approximate photon flux: 0.2mmol/h<280n m, 18 mmol/h 280-315 nm, 390 mmol/h 315-380nm) hTQ718 lamp: sheathed by Solidex glass (approximate photon flux: 5mmol/h<280nm, 150 mmol/h 280-315 nm, 220 mmol/h 315-380nm) iDegussa P25: 70% anatase, 30% rutile jAldrich: 99.9% anatase kP25: anatase form, average size 0.021 m, Nippon-Aerosil, Tokyo, Japan lDegussa AG, Germany, surface area 50 m2/g, primary particle size 20 nm mDegussa P25: anatase P25, average size 30 nm, BET surface area 505m2/g, Degussa Corp.; Preparation of Ag-coated TiO2 made by exposure to = 354 nm for 2h. nfrom BDH with the grade of GPR oDegussa, 75% anatase, 25% rutile, surface area 50 m2/g pTiO2/Pt: made by mixing TiO2 and Pt, grounding in a mortar. TiO2, P-25, 99.99% anatase, Aerosil, Japan; Pt black (Nakarai Chemicals) Weight ratio TiO2/Pt = 10:1 qanatase, Nippon Aerosil Ltd, Tokyo, Japan rDegussaP25: primary anatase, specific surface are 50 m2/g s0.5% emitted photons at < 300 nm, 7% emitted photons at = 300 nm ~ 400 nm, emission at = 400 nm ~ 800 nm follows the solar spectrum. tDegussa P25: 75% anatase, 25% rutile, specific surface area 60 m2/g , Degussa A. G. Germany ufrom Yakuri pure chemical company, Osaka, Japan vGrade NP92, comprising principally the anatase from, fr om Tioxide specialities Ltd, Billingham Cleveland wP25: anatase form, average size 0.021 m, Nippon-Aerosil, Tokyo, Japan xWA: anatase form, average size 0.255 m, Wako, Tokyo, Japan yWR: rutile form, average size 0.255 m, Wako zTP-3: rutile form, average size 0.42 m, Fuji titan, Kanagawa, Japan aaDegussa P25: mainly composed of anatase, BET surface area 50 m2/g, average size 21nm

PAGE 280

262abDegussa P25: anatase P25, averag e size 30 nm, BET surface area 505m2/g, Degussa Corp.; Preparation of Ag-coated TiO2 made by exposure to = 354 nm for 2hr. ac99.98% rutile, particle size 1.48 m, Fujichitan Co., Osaka, Japan admean particle size 21nm, 70% anatase, 30% ru tile; pI 6.6; P25, Nippon, Aerosil, Japan aeDegussaP25: mainly composed of anatase, BET surface area 50 m2/g, average size 21nm afanatase to rutile: 70 to 30; mean diamet er 21 nm, from Japan Aerojil Co., Japan aganatase, Nippon Aerosil Ltd, Tokyo, Japan ah anatase, particle size 0.24 m, BET surface area 10.1 m2/g ai anatase, particle size 5-10 nm, BET surface area 329.1 aj Mutant strain of E. coli, E. coli IM303 (a superoxide dismutase-deficient mutant) was used because the deactivation pr ofile of the cells follows a simple first-order kinetics in terms of viable ce ll number during the photoreacti on in titania suspension.

PAGE 281

263 APPENDIX B PRELIMINARY RESULTS OF EFFECT OF TITANIA POWDER AGAINST B. CEREUS ENDOSPORES ON MEMBRANE SURFACE UNDER 350 NM UV IRRADIATION IN DRY STATE Spores were applied on the membrane in tw o ways: spores were mixed with titania followed by filtration of mixture on membrane, or spores were applied on the membrane pre-applied with titania coating. Inactivation of B. cereus spores on membrane surface by titania and solar UV was studied respectivel y for these two different spore application methods. B.1 Inactivation of B. cereus spores on Membrane Surface by Applying Mixture of Spores and Titania on the Surface In trial 1, mixture with the combination of 5 106CFU spores and 0.05 mg titania was filtered onto the membrane surface. This combination of titania and spore amount selected in the 1st trial is based on the SEM images taken from different combination described in Chapter 5. Figure B-1 shows distribution of a mixture of 106 CFU spores and 0.1 mg titania on Anodisc membrane. Figur e B-2 is the distribut ion of a mixture of 105CFU spores and 0.01 mg titania on A nodisc membrane. Figure B-3 shows distribution of a mixture 104CFU spores and 0.001 mg titania on Anodisc membrane. A uniform distribution can be obs erved on all these three combinations. However, surface coverage by titania was very high for the mixture of 106CFU spores and 0.1 mg titania in Figure B-1, besides, only 67 spores can observed at a 5000 magnification. For the mixture of 104CFU spores and 0.001 mg titania (Figur e B-3), a very low surface coverage by mixture of spores and titania powder was achieved, and spores were hard to be

PAGE 282

264 detected. Figure B-2 gives medium titaina coverage on the membrane surface among the three combinations; still, spores were hard to be detected. Thus, in order to obtain an appropriate titania coverage on the membrane, at the same time, to ensure an enough contact between titania powder and spores, a re lative higher spore to titania ratio than the ratio of the above three combinations was chosen: a mixture with a total amount of 5 106 CFU spores and 0.05 mg titania (Figure B-4) was used for UV and photocatalytic inactivation test. As shown in Figure B-4, a medium surface coverage was achieved and spores are contacting well with titania powder, which matches our expectation. Because no results were reported on phot ocatalytic inactivation of B. cereus spores on membrane surface by Degussa P25 under solar UV, an irradi ation period of 24 hr was used to make sure that inactivation activity of Degussa P25 and solar UV against B. cereus spores on membrane surface can be observed. Samples were taken every 8 hr. Result is shown in Table B-1. It can be found that the temperature in chamber remained stable and as low as room temper ature, which was contri buted by the running of cooling fan. Data under the dilution order means it was the spore counts observed after the viable spores on membrane was dilu ted the corresponding orde r. An at least 3log reduction after 8 hr and a complete in activation after 24 hr exposure to UVA light was achieved in the presence of titania (Figure B-5). The test was repeated and experiment conditions such as irradiation time, titania amount we re adjusted in the following trials. In trial 2, same condition was duplicated as in the trial 1 except the irradiation period was shortened to 6 hr since a rapid inactivation was observed, which facilitate observation of detailed inactivation information. Starting from trial 2, control that only

PAGE 283

265 spores were applied on membrane and exposed to solar UV irradiation was run at the same time. In the control, a total amount of 5 106CFU spores were distributed on each Anodisc membrane by filtration and dry overn ight, then the samples were exposed to solar UV irradiation. For photocatalytic test, a mixture of 0.05 mg titania and 5 106 CFU spores were filtered on the membrane surface. Samples were taken every 2 hr. Figure B-6 shows spore survival ratio vs. time for both UV and UV+P25 inactivation in trial 2. After 120 min irradiation, more than 90% spores were inactivated under both solar UV and UV+P25 system. A LD 90 of 98 min and 71 min and D value of 128 and 132 were achieved by UV in the presence of titania and UV inact ivation respectively (Table B-2). Slopes between two linear re gression line (Figure B6) derived from UV and UV+P25 test respectively are not si gnificant different ba sed on F-test at = 0.05. Thus, inactivation results against B. cereus spores with and without titania under solar UV irradiation are not statistically significant. Trial 3 is the repeat of trial 2. Figure B-7 shows spore survival ratio vs. time under UVA irradiation in the presence and absence of titania in trial 2. The result was summarized in Table B-2. A LD 90 of 126 min and D value of 70 min were achieved by UVA irradiation in the presence of titania, whereas a less LD 90 (93 min) and D (62 min) value were observed under solar UV alone. Sa me trend was observed in trial 2. A less LD 90 and D value were observed under solar UV alone than in the presence of titania powder. Based on F-test at = 0.05, inactivation result under solar UV with and without titania for both two trials are statistically not significant. In another word, in the present condition, the presence of titania particle under solar UV didn’t improve spore

PAGE 284

266 inactivation. A trade off between shielding of light by titania partic le and photocatalytic activity of titania coul d be a possible reason. We hypothesized that the rate of inactiva tion of bacterial endos pores in dry state under solar UV irradiation can be increased by adding tita nia nanoparticles (Degussa P25). But no improvement was observed from the above three trials. Thus, conditions were adjusted to test this hypothesis. One idea is to adjust UV intensity, which is one of the important factors in cell inactivation kinetics. In trial 4, the dist ance between UV lamp and sample was shortened from 10 cm to 5 cm, corresponding to an in crease of light intensity from 104.5 W/m2 to 129.5 W/m2. Other conditions were kept same as the trial 2 and 3. Inactivation of B. cereus spores by UV+P25 and UV alone was tested. Results were shown in Fig. A-8. A rapid destruction of B. cereus spores were observed, more than 1 log reduction was achieved after 120 min irradiation for both w ith and without titania on the membrane surface. Inactivation rate by UV+P25 is sta tistically insignificant from by UV alone based on F-test at = 0.05. Possible reason is that sh ading effect by titania was large, photocatalytic activity of tit ania was still compromised although a higher intensity was applied. Another idea is to increase applied titani a amount in the mixture. An increase of contacting possibility between spores and titania was expected by increased titania amount, thus, inactivation activ ity may be improved. In trial 5, Titania amount was increased from 0.05 mg to 0.1 mg, other conditio n keeps same as trial 2 and 3. Figure B4 and B-9 shows SEM image of mixture distri bution with applied tit ania amount of 0.05 mg and 0.1 mg respectively, applied spore amount is 5 106CFU. As observed from Fig.

PAGE 285

267 B-4 and B-9, a complete surface coverage was achieved when 0.1 mg titania was applied on the surface, where only a partial coverage was achieved when 0.005 mg titania was used in the mixture. Thus, contact area be tween spore and titania was increased, which may favor photocatalytic inactivation. An LD90 of 90 min and a D value of 70 min were obtained for UV alone and an LD90 of 120 min and a D value of 82 min were obtained for UV and photocatalyst (Fig. A-10). There is no significant difference for the inactivation activity between solar UV alone and titania with solar UV based on F-test at = 0.05. This result was same as the previous tr ials. No improvement was observed by the presence of titania under UV compared to UV irradiation alone. A maximum contacting area between spores and titania may have been reached in the combina tion tested in trial 1, 2 and 3, thus, increasing titania amount in the mixture in this tr ial didn’t show any improvement for inactivation of B. cereus spores by photocatalyst compared that without photocatalyst when irradiated to solar UV light. Results of inactivation of B. cereus spores by UV alone and UV+P25 by applying mixture of spores and titania to membrane surface are summarized in Table B-2, data under temperature and relative humidity re presents the range monitored during the inactivation test. It can be found the temperature and relative humidity keeps stable for each trial. For all the five trials, no statistically differen ce was found for the results of inactivation of B. cereus spores from UV (control) and UV+P25 in each trial based on Ftest. The possible reason is that photocatalytic activity of titania was traded off by shading effect of titania. In terms of LD90 and D value, a shorter time was observed when inactivation of dry spores on membrane was tested at a higher intensity under UV and UV+P25 in trial 4.

PAGE 286

268 Lin and Li (2003b) also found that higher intens ity resulted in higher inactivation rate for B. subtilis spores under UVA irradiation in the presence photocat alyst on both glass slide and filter surface, where the spores were app lied on titania coating on the slide and filter surface. No difference was found for trial 5 with trial 1, 2 and 3, although double amount of titania was applied in the mixture. It ’s possible that increased contacting area by adding more titania was balanced by increased sh ading effect of titania. Also, it could be maximum contacting area of spores and titani a have been met before increasing titania amount, thus, increase in titania amount di d not promote photocatalytic activity. B.2 Inactivation of B. cereus spores on membrane surface by applying spores on titania coating Enough titania powder was applied on membrane by filtration in order to achieve a complete coverage on membrane. Then, deposition of spores on the filter precoated by titania was checked. A total amount of 10 mg and 50 mg titania particles were tested on the membrane respectively, corresponding to a filtration time of 8 min and 30 min. For further filtration of spore su spension through the membrane pr e-filtered with titania, a filtration time of 5 min was required for 10 mg titania particle loading, whereas spore suspension didn’t go through totally after 30 min fitration for 50 mg titania particle loading. Thus, a total amount of 10 mg tita nia particles was chosen to filter on the membrane surface followed by filtering 5 106 CFU spores on the titania coating. Only a total amount 5 106 CFU spores was filtered on the me mbrane for control (UV alone). Result of trial 6 is shown in Figure B11. For UV alone, spore survival ratio decreased to 0.1 within 86 min. After 1 log reduction, spor e survival ratio decreased to 0.01 in 53 min. For UV+P25, 1-log reduction of spores was achieved in 59 min, then spore survival ratio continued to decrease rapidly to the leve l of 0.03 in 120 min, but the

PAGE 287

269 survival ratio at 240 min of exposure time ro se up to approximately 0.05. With the further increase of exposure time, the survival ratio decreased again to the level of 0.001 after 360 min irradiation. In activation activity of solar UV alone and photocatalyst with UV against spores are statistically significant analyzed by F-test at = 0.05. The shading effect of spores by titaina pa rticles may contribute to this phenomenon. Since the spores were filtered though membrane right after the titania solution was filtered, the titania layer may be disturbed by the application of spore suspension, thus, some spores may be imbedded under the titania particles. This matches our observation that titania coating with 10 mg titania loading on surface resuspe nded in the D.I. wate r when the surface was immersed into the water. This amount loading of titania made a loose titana coating. To avoid this possibility, three approaches were used in the following trials. One is to dry the titania coating under laminar flow hood for 3 h after filtration to achieve a firm titania coating. The second is to apply as thin as possible titania coating layer as long as complete surface coverage can be achieved. The purpose was same as the first one, to achieve a firm titania coating. The third is to pipette spore suspension at 10 spots on the titania coating so the spores are sitting on the top of coating, then dry under laminar flow hood for 2 h. For the second approach, SEM image of titania coating was taken to decide appropriate titania amount applied on the membrane surface. The membrane surface was applied with titania amount of 2.5 mg and 1 mg respectively, SEM image was taken and compared. Figure B-12 is SEM image of titani a coating with applied titania amount of 1 mg. Figure B-13 is SEM image of titania coa ting with applied titani a amount of 2.5 mg.

PAGE 288

270 Both images show good distribution and co mplete surface coverage, thus, 1 mg was chosen as titania coating lo ad in the following trials. For the third approach, sonication of spor e suspension for 2 min at level 6.0 was tried in order to achieve a better distribu tion. Fig. A-14 shows SEM image of spore distribution without sonicati on and b) shows SEM image of spore distribution with sonication. A total of 1 mg titania was appl ied on the membrane surface by filtration and dry, then the spore suspension was pipetted on the titania coating and dry. A better distribution was observed for the spore susp ension sonicated before pipetting on the membrane. Sonication helps break spore agglom eration, thus, sonication was used before pipetting spores on the titania coating. Start from trial 7, this spore application method used in third approach was adopted and run in the subsequent trials. Figure B-15 shows experimental results for trial 7. As mentioned above, sample was prepared by filtering 1 mg titania on the Anodsic membrane, then spore suspension was pipetted at 10 spots (20 L/spot) on the titania coating to a total spore amount of 5 106 CFU spores and dry. Samples were then exposed to solar UV irradiation for 240 min. Control was prepared by pipet ting only spore on the surface. Results were shown in Fig. A-15. 90% spores were inactivated in around 120 min under UV and UV+P25. A D value of 127 and 176 was achieved for UV and UV+P25 respectively. Based on F-test = 0.05, results from solar UV and UV +P25 was not statistically significant different. Trial 8 was the repeat of trial 7 except th at the spore suspension was heat shock and washed once with sterile D.I. water right befo re inactivation test, a lthough the heat shock and water wash has been done in the purifica tion process. The purpos e for the heat shock and water wash right before test is to make sure that the effect of vegetative cells and

PAGE 289

271 germinating spores was eliminated. Results we re shown in Fig. A-16. Based on F-test at = 0.05, results from solar UV and UV +P25 wa s not statistically significant different. In trial 7 and 8, no significant difference was observed for inactivation efficiency by UV and UV+P25. Thus, conditi on was further adjusted. Results of inactivation of B. cereus spores by UV alone and UV+P25 by adding spores after application of titania to membrane surface (tri al 6-8) were summarized in Table B-3. Distance was measured from the bo ttom of UV light to the surface of sample. For trial 6, inactivation activity of UV is significant higher than UV+P25, which indicated a large shading e ffect based on F-test at = 0.05. For trial 7 and 8, there is no significant difference between UV and UV+P25 in activation results. The shading effect in these two trials was avoided by modifi cation of spore application method. The insignificance between UV and UV+P25 inactiva tion activity is probably due to the high intensity applied in the test. The high inactivation efficiency by UV alone may make photocatalytic activity become ne gligible. Vohra et al. (2005) found that inactivation rate was improved by the adding of titania under solar UV irradiation. A destruction of 22% spores on fabric or metal surface were inactivated by solar UV irradiation alone was observed after 4 hr irradiation, wh ereas 76% and 80% destruction of B. cereus spores were achieved on metal and fabric surface with titania coating respectively. At an irradiation period of 24 hr, 55% spores were inactivated by solar UV alone, whereas 99% and 96% B. cereus on metal and fabric surface were de structed by UV in the presence of photocatalyst. In Vohra’s research , a much lower intensity (50 W/m2) was used as compared with our research (104.5 W/m2, 129.5 W/m2). However, at a very low intensity (7.4 W/m2), Lin and Li (2003b) observed no difference between UV and

PAGE 290

272 UV+P25 for inactivation of B. subtilis. Thus, an appropriate UV intensity may require for photocatalysis to achieve its optimum ac tivity on bacterial endospore inactivation. Also, a 10 wt% titania slurry (not menti oned in paper, communicated with author) was painted on the surface by Vohra et al. (2005) whereas a total of 1 mL 0.1 wt% titania solution was filtered on the membrane in our research. This may contribute to an obvious photocatalytic activity. Surface can also be a factor affecting phot ocatalytic inactivation of spores under UV irradiation. As found by Lin and Li (2003b), no signifi cant difference was found for inactivation of B. subtilis spores on glass slides under UV and UV+P25, whereas inactivation rate constant is higher for inactivation of B. subtilis spores on titania filter than on filter without photocat alyst under UV irradiation. B.3 Summary Results of preliminary experi ments indicate that light inte nsity is important factor affecting photocatalytic contri bution to spore inactivation ra te. Titania loading plays important role on the shading effect in spore inactivation on surfaces.

PAGE 291

273 Figure B-1. SEM image of a mixture of 106 CFU spores and 0.1 mg titania on Anodisc membrane Figure B-2. SEM image of a mixture of 105 CFU spores and 0.01 mg titania on Anodisc membrane

PAGE 292

274 Figure B-3. SEM image of a mixture of 104 CFU spores and 0.001 mg titania on Anodisc membrane Figure B-4. SEM image of a mixture of 5 106 CFU spores and 0.05 mg titania on Anodisc membrane at a) 5000 magnification and b) 250 magnification a b

PAGE 293

275 y = -0.0018x 3.7881 R2 = 0.996 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 04809601440 Time (min)Log (Survival ratio) * UV+P25 Figure B-5. Inactivation of B. cereus spores by UV+P25. The spores and titania were mixed before application to membrane. (Trial 1)

PAGE 294

276 y = -0.0076x 0.5673 R2 = 0.9968 y = -0.0078x 0.2529 R2 = 0.9997 -6 -5 -4 -3 -2 -1 0 060120180240300360 Time (min)Log (Survival ratio) * UV+P25 UV Figure B-6. Inactivation of B. cereus spores by UV and UV+P25. The spores and titania were mixed before application to membrane. (Trial 2) y = -0.016x + 0.5583 R2 = 0.9769 y = -0.0145x + 0.8601 R2 = 1 -6 -5 -4 -3 -2 -1 0 060120180240300360 Time (min)Log (Survival ratio) * UV+P25 UV Figure B-7. Inactivation of B. cereus spores by UV and UV+P25. The spores and titania were mixed before application to memb rane. (Trial 3, repeat of trial 2)

PAGE 295

277 y = -0.0104x 0.9018 R2 = 0.4455 y = -0.0166x + 0.3983 R2 = 0.936 -6 -5 -4 -3 -2 -1 0 060120180240300360 Time (min)Log (Survival ratio) * UV+P25 UV Figure B-8. Inactivation of B. cereus spores by UV and UV+P25. The spores and titania were mixed before application to membrane. (Trial 4) Figure B-9. SEM image of a mixture of 5 106CFU spores and 0.1 mg titania on Anodisc membrane a) 5000 magnification and b) 250 magnification a b

PAGE 296

278 y = -0.0143x + 0.394 R2 = 0.9758 y = -0.0123x + 0.403 R2 = 0.996 -6 -5 -4 -3 -2 -1 0 060120180240300360 Time (min)Log (Survival ratio) * UV+P25 UV Figure B-10. Inactivation of B. cereus spores by UV and UV+P25. The spores and titania were mixed before appli cation to membrane. (Trial 5) y = -0.0188x + 0.7216 R2 = 0.9758 y = -0.0049x 0.6975 R2 = 0.5812 -6 -5 -4 -3 -2 -1 0 060120180240300360 Time (min)Log (Survival ratio) * UV+P25 UV Figure B-11. Inactivation of B. cereus spores by UV alone and UV+P25. Spores were added after application of titani a to the membrane. (Trial 6)

PAGE 297

279 Figure B-12. SEM image membrane pre-coat ed with 2.5 mg 1 mg titania at a) 5000 magnification and b) 250 magnification Figure B-13. SEM image membrane pre-coated 1 mg titania at a) 5000 magnification and b) 250 magnification a b a b

PAGE 298

280 Figure B-14. SEM image of 106CFU spores distribution on me mbrane pre-coated with 1 mg titania a) spore suspension was pipette d on titania coating directly b) spore suspension was sonicated before pipetting on titania coating a b

PAGE 299

281 y = -0.0079x 0.0584 R2 = 0.9998 y = -0.0052x 0.3603 R2 = 0.8589 -3 -2 -1 0 060120180 Time (min)Log (Survival ratio) * UV+P25 UV Figure B-15. Inactivation of B. cereus spores by UV alone and UV+P25. Spores were added after application of titani a to the membrane. (Trial 7) y = -0.0051x 0.7241 R2 = 0.9704 y = -0.0044x 0.5005 R2 = 0.9994 -3 -2 -1 0 060120180 Time (min)Log (Survival ratio) * UV+P25 UV Figure B-16. Inactivation of B. cereus spores by UV alone and UV+P25. Spores were added after application of titania to the membrane. (Trial 8)

PAGE 300

282 Table B-1. Inactivation of B. cereus spores by UV+P25. Mixture of spores and titania were applied to membrane surface. (trial 1) Average viable spores on membrane Time (hr) T( C) Sample No. 10-1 dilution 10-2 dilution 10-3 dilution 10-4 dilution 10-5 dilution 10-6 dilution 0 22.6 3 276.7 39.8 4 8 25.4 3 0.3 0 0 16 25.0 3 0.5 0 0 24 26.0 3 0.2 0 0 0

PAGE 301

283 283Table B-2. Inactivation of B. cereus spores by UV alone and UV+P25 by applying mixture of spores and titania to membrane surface Trial System T( C) Relative humidity (%) Distance (cm) Total amount titania (mg) LD90 (min) Ratio of LD90 D Value (min) Ratio of D value Significance difference in slopes1 1 UV + Degussa P25 TiO2 22-26 -10 0.05 237 -490 --UV 71 132 2 UV + Degussa P25 TiO2 24-26 -10 0.05 98 0.72 128 1.03 Insignificant UV 93 62 3 UV + Degussa P25 TiO2 25-26 34-40 10 0.05 126 0.74 70 0.89 Insignificant UV 42 33 4 UV + Degussa P25 TiO2 25-29 44-57 5 0.05 80 0.53 60 0.55 Insignificant UV 90 70 5 UV + Degussa P25 TiO2 22-24 66-76 10 0.1 120 0.75 82 0.85 Significant 1 Slope refers to slope of regression line from UV and UV+P25 inactivation result

PAGE 302

284Table B-3. Inactivation of B. cereus spores by UV alone and UV+P25 by adding spores after application of titania to the membrane Trial System T( C) Relative Humidity (%) Distance (cm) Total amount titania (mg) LD90 (min) Ratio of LD90 D Value (min) Ratio of D value Significance difference in slopes1 UV 86 53 61 UV + Degussa P25 TiO2 34-35 48-57 10 10 59 1.46 204 0.26 Significant UV 120 127 7 UV + Degussa P25 TiO2 26-30 42-50 10 1 125 0.96 176 0.72 Insignificant UV 70 195 83 UV + Degussa P25 TiO2 25-27 52-65 10 1 112 0.62 228 0.86 Insignificant 1 Slope refers to slope of regression line from UV and UV+P25 inactivation result 2 Spores were filtered on the membrane in tr ial 6, where spore suspension was pipetted on membrane in trial 7 and 8. Titania co ating in trial 6 was not dried wh ile dried in trial 7 and 8 3 Spores were heat shocked and washed with steril e D.I. water once right before UV inactivation test

PAGE 303

285 LIST OF REFERENCES Aithal, U. S., Aminabhavi, T. M., and Shukl a, S. S. (1993). "Photomicroelectrochemical detoxification of hazardous materials." Journal of Hazardous Materials, 33(3), 369. Al-Bastaki, N. M. (2004). "Performance of advanced methods for treatment of wastewater: UV/TiO2, ro and uf." Chemical Engineering and Processing, 43(7), 935. Ali, Z., Gimblett, R., Lax, D., and Meakins, D. "Atmospheric deposition of TiO2 films on glass substrates for antibacterial activity." Part of SPIE Conference on Environmental Monitoring and Remediation Technologies II, Boston, Massachusett, 409. Allen, N. S., Edge, M., Sandoval, G., Verran, J., Stratton, J., a nd Maltby, J. (2005). "Photocatalytic coatings for environmental applications." Photochemistry and Photobiology, 81(2), 279. Amezaga-Madrid, P., Nevarez-Moorillon, G. V., Orrantia-Borunda, E., and MikiYoshida, M. (2002). "Photoinduced bactericidal activity against Pseudomonas aeruginosa by TiO2 based thin films." Fems Microbiology Letters, 211(2), 183– 188. Amezaga-Madrid, P., Silveyra-Morales, R., Co rdoba-Fierro, L., Neva rez-Moorillon, G. V., Miki-Yoshida, M., Orra ntia-Borunda, E., and Solis, F. J. (2003). "TEM evidence of ultrastructural alteration on Pseudomonas aeruginosa by photocatalytic TiO2 thin films." Journal of Photochemistry and Photobiology B-Biology, 70(1), 45. Atrih, A., and Foster, S. J. (2001). "Analysi s of the role of bact erial endospore cortex structure in resistance properties and de monstration of its conservation amongst species." Journal of Applied Microbiology, 91(2), 364. Atrih, A., and Foster, S. J. (2002). "Bact erial endospores the ultimate survivors." International Dairy Journal, 12(2), 217. Atrih, A., Zollner, P., Allmaier, G., and Fost er, S. J. (1996). "Str uctural analysis of Bacillus subtilis 168 endospore peptidoglycan and its role during differentiation." Journal of Bacteriology, 178(21), 6173.

PAGE 304

286 Bahnemann, D. W. (1993). "Ultrasmall metaloxide particles preparation, photophysical characterization, and photocatalytic properties." Israel Journal of Chemistry, 33(1), 115. Bailey-Smith, K., Todd, S. J., Southworth, T. W., Proctor, J., and Moir, A. (2005). "The ExsA protein of Bacillus cereus is required for assembly of coat and exosporium onto the spore surface." Journal of Bacteriology, 187(11), 3800. Baltschukat, K. (1986). "Wirkung sehr schwerer ionen auf sporen von Bacillus subtilis inaktivierung, reparatur von strahl enschaden und mutationsauslosung," Dissertation, University of Frankfurt, Frankfurt. Baltschukat, K., and Horneck, G. (1991). "Respons es to accelerated heavy-ions of spores of Bacillus-subtilis of different repair capacity." Radiation and Environmental Biophysics, 30(2), 87. Baltschukat, K., Horneck, G., Bucker, H., Faci us, R., and Schafer, M. (1986). "Mutationinduction in spores of Bacillus-subti lis by accelerated very heavy-ions." Radiation and Environmental Biophysics, 25(3), 183. Behnajady, M. A., Modirshahla, N., and Hamzavi, R. (2006). "Kinetic study on photocatalytic degradation of ci acid yellow 23 by zno photocatalyst." Journal of Hazardous Materials, 133(1), 226. Bekbolet, M. (1997). "Photocatalytic bactericidal activity of TiO2 in aqueous suspensions of E-coli." Water Science and Technology, 35(11), 95. Bekbolet, M., and Araz, C. V. (1996). "Inactivation of Escherichia coli by photocatalytic oxidation." Chemosphere, 32(5), 959. Belhacova, L., Krysa, J., Geryk, J., a nd Jirkovsky, J. (1999) . "Inactivation of microorganisms in a flow-through photoreact or with an immob ilized TiO2 layer." Journal of Chemical Te chnology and Biotechnology, 74(2), 149. Bender, G. R., and Marquis, R. E. ( 1985). "Spore heat-res istance and specific mineralization." Applied and Environmental Microbiology, 50(6), 1414. Berthouex, P. M. a. B., L.C. (1994). "S tatistics for environmental engineers." Beuchat, L. R., Pettigrew, C. A., Tremblay, M. E., Roselle, B. J., and Scouten, A. J. (2005). "Lethality of chlorine, chlori ne dioxide, and a commercial fruit and vegetable sanitizer to vegeta tive cells and spores of Bacillus cereus and spores of Bacillus thuringiensis." Journal of Industrial Mi crobiology & Biotechnology, 32(7), 301. Bickley, R. I., Gonzalezcarreno, T., Lees, J. S ., Palmisano, L., and Tilley, R. J. D. (1991). "A structural investigation of titanium-dioxide photocatalysts." Journal of Solid State Chemistry, 92(1), 178.

PAGE 305

287 Biguzzi, M., and Shama, G. (1994). "Effect of titanium-dioxide concentration on the survival of Pseudomonas-stutzeri during irra diation with near-ultraviolet light." Letters in Applied Microbiology, 19(6), 458. Blake, D. M., Maness, P. C., Huang, Z., Wolf rum, E. J., Huang, J., and Jacoby, W. A. (1999). "Application of the photocatalytic chemistry of titanium dioxide to disinfection and the kil ling of cancer cells." Separation and Purification Methods, 28(1), 1. Blatchley, E. R., Meeusen, A., Aronson, A. I., and Brewster, L. ( 2005). "Inactivation of Bacillus spores by ultraviolet or gamma radiation." Journal of Environmental Engineering-Asce, 131(9), 1245. Block, S. S. (2001). Disinfection, steriliza tion and preservation., Lippincott Williams & Wilkins, Philadelphia. Block, S. S., Seng, V. P., and Goswami, D. W. (1997). "Chemically enhanced sunlight for killing bacteria." Journal of Solar Energy Engi neering-Transactions of the Asme, 119(1), 85. Bloomfield, S. F., and Arthur, M. (1994). "M echanisms of inactivation and resistance of spores to chemical biocides." Journal of Applied Bacteriology, 76, S91–S104. Bloomfield, S. F., and Megid, R. (1 994). "Interaction of iodine with Bacillus-subtilis spores and spore forms." Journal of Applied Bacteriology, 76(5), 492. Bohnel, H., and Gessler, F. (2004). "From bact erial spore to death of the patient the development cascade of botulism." Tierarztliche Umschau, 59(1), 12-+. Boulange-Peterman, L., Baroux, B., and Bell on-Fontaine, M. N. (1993). "The influence of metallic surface wettabili ty on bacterial adhesion " J. Adhes. Sci. Technol. , 7, 221. Bredholt, S., Maukonen, J., Kujanpaa, K., Alanko, T., Olofson, U., Husmark, U., Sjoberg, A. M., and Wirtanen, G. (1999). "Microbial methods for assessment of cleaning and disinfection of food-processi ng surfaces cleaned in a low-pressure system." European Food Research and Technology, 209(2), 145. Busscher, H. J., Bellon-fontaine, M. N., Mozes, N., van der Mei, H. C., Sjollema, J., Cerf, O., and Rouxhet, P. G. (1990). "Adhesi on and motility of gliding bacteria on substrata with different surface-free energies." Appl. Environ. Microbiol. , 56, 2529. Butterfield, I. M., Christensen, P. A., Curtis, T. P., and G unlazuardi, J. (1997). "Water disinfection using an immob ilised titanium dioxide film in a photochemical reactor with electric field enhancement." Water Research, 31(3), 675.

PAGE 306

288 Buttner, M. P., Cruz-Perez, P., and Stetzenb ach, L. D. (2001). "Enhanced detection of surface-associated bacter ia in indoor environmen ts by quantitative pcr." Applied and Environmental Microbiology, 67(6), 2564. Carraway, E. R., Hoffman, A. J., and Hoffma nn, M. R. (1994). "Phot ocatalytic oxidation of organic-acids on quantum-sized semiconductor colloids." Environmental Science & Technology, 28(5), 786. Chen, D. W., and Ray, A. K. (1999). "Phot ocatalytic kinetics of phenol and its derivatives over UV irradiated TiO2." Applied Catalysis B-Environmental, 23(2-3), 143. Chen, J. S., Liu, M. C., Zhang, L., Zhang, J. D., and Jin, L. T. (2003). "Application of nano TiO2 towards polluted water treatment combined with electro-photochemical method." Water Research, 37(16), 3815. Cho, M., Chung, H., Choi, W., and Yoon, J. (2004). "Linear correlation between inactivation of E-coli and oh radical concentratio n in TiO2 photocatalytic disinfection." Water Research, 38(4), 1069. Cho, M., Chung, H. M., Choi, W. Y., and Y oon, J. Y. (2005). "Different inactivation behaviors of MS-2 phage and Escherichia coli in TiO2 photocatal ytic disinfection." Applied and Environmental Microbiology, 71(1), 270. Christiansson, A., Bertilsson, J., and Svensson, B. (1999). "Bacillus cereus spores in raw milk: Factors affecting the contamina tion of milk during the grazing period." J Dairy Sci, 82(2), 305. Ciarciaglini, G., Hill, P. J., Davies, K., McClure, P. J., Kilsby, D., Brown, M. H., and Coote, P. J. (2000). "Germination-induced bioluminescence, a route to determine the inhibitory effect of a combination pr eservation treatment on bacterial spores." Applied and Environmental Microbiology, 66(9), 3735. Cooper, A. T., Goswami, D. Y., and S.S., B. (1998). "Solar photoc hemical detoxification and disinfection for water treatment in tropical developing countries." J. Adv. Oxid. Technol. , 3(2), 151. Coronado, J. M., Soria, J., Conesa, J. C., Bellod, R., Adan, C., Yamaoka, H., Loddo, V., and Augugliaro, V. (2005). "Phot ocatalytic inactivation of legionella pneumophila and an aerobic bacteria consortium in wa ter over TiO2/SiO2 fibr es in a continuous reactor." Topics in Catalysis, 35(3), 279. Couvert, O., Leguerinel, I., and Mafart, P. (1999). "Modelling the ove rall effect of ph on the apparent heat resistance of Bacillus cereus spores." International Journal of Food Microbiology, 49(1), 57.

PAGE 307

289 Cremers, P. T. J., Busscher, D. L. T ., and Macfarlane, J. D. (1990). "Ultrasound demonstration of a superior mesenteric-art ery aneurysm in a patient with ehlersdanlos syndrome." British Journal of Rheumatology, 29(6), 482. Daneshvar, N., Rabbani, M., Modirshahla, N., and Behnajady, M. A. (2004). "Kinetic modeling of photocatalytic degradation of acid red 27 in UV/TiO2 process." Journal of Photochemistry and Photobiology a-Chemistry, 168(1), 39. Darren, D. S., Tay, J. h., and min, T. k. (2003). "Photocatalytic degradation of E. coliform in water." Water research, 37, 3452. Demeestere, K., De Visscher, A., Dewulf, J., Van Leeuwen, M., and Van Langenhove, H. (2004). "A new kinetic model for titan ium dioxide mediated heterogeneous photocatalytic degradation of tr ichloroethylene in gas-phase." Applied Catalysis BEnvironmental, 54(4), 261. Dibble, L. A., and Raupp, G. B. (1990). "K inetics of the gas-solid heterogeneous photocatalytic oxidation of trichloroethylene by near uv illuminated titaniumdioxide." Catalysis Letters, 4(4), 345. Dillert, R., Siemon, U., and Bahnemann, D. (1998). "Photocatalytic disinfection of municipal wastewater." Chemical Engineering & Technology, 21(4), 356-+. Dionysiou, D. D., Suidan, M. T., Baudin, I ., and Laine, J. M. (2002). "Oxidation of organic contaminants in a rotating disk phot ocatalytic reactor: Reaction kinetics in the liquid phase and the role of mass transfer based on the dimensionless damkohler number." Applied Catalysis B-Environmental, 38(1), 1. Donnella.Je, and Setlow, R. B. (1965). "T hymine photoproducts but not thymine dimers found in ultraviolet-irradi ated bacterial spores." Science, 149(3681), 308–&. Dorfman, L. M., and Adams, G. E. (1973). Reactivity of the hydroxyl radical in aqueous solutions, National bureau of standards, Washington, U.S.A. Dragon, D. C., and Rennie, R. P. (2001). "Eva luation of spore extr action and purification methods for selective recovery of viable Bacillus anthracis spores." Letters in Applied Microbiology, 33(2), 100. Duffy, E. F., Al Touati, F., Kehoe, S. C., McLoughlin, O. A., Gill, L. W., Gernjak, W., Oller, I., Maldonado, M. I., Malato, S., Ca ssidy, J., Reed, R. H., and McGuigan, K. G. (2004). "A novel TiO2-assisted solar pho tocatalytic batch-process disinfection reactor for the treatment of biological and chemical contaminants in domestic drinking water in developing countries." Solar Energy, 77(5), 649. Dunlop, P. S. M., Byrne, J. A., Manga, N., and Eggins, B. R. (2002). "The photocatalytic removal of bacterial pollutants from drinking water." Journal of Photochemistry and Photobiology a-Chemistry, 148(1), 355.

PAGE 308

290 Ehling-Schulz, M., Fricker, M ., and Scherer, S. (2004). "Bacillus cereus, the causative agent of an emetic type of food-borne illness." Molecular Nutrition & Food Research, 48(7), 479. Eladhami, W., Daly, S., and Stewart, P. R. (1994). "Biochemical-studies on the lethal effects of solar and artifici al ultraviolet-radiation on Staphylococcus-aureus." Archives of Microbiology, 161(1), 82. El-Dein, A. M., Libra, J. A., and Wiesma nn, U. (2003). "Mechanism and kinetic model for the decolorization of the azo dye r eactive black 5 by hydrogen peroxide and UV radiation." Chemosphere, 52(6), 1069. Emeline, A. V., Ryabchuk, V., and Serpone, N. (2000). "Factors affecting the efficiency of a photocatalyzed process in aqueous metal-oxide dispersions prospect of distinguishing between two kinetic models." Journal of Phot ochemistry and Photobiology a-Chemistry, 133(1), 89. Ezzell, J. W., and Welkos, S. L. (1999). "The capsule of Bacillus anthracis, a review." Journal of Applied Microbiology, 87(2), 250. Faille, C., Jullien, C., Fontaine, F., Be llon-Fontaine, M. N., Slomianny, C., and Benezech, T. (2002). "Adhesion of Bacillus spores and Escherichia coli cells to inert surfaces: Role of surface hydrophobicity." Canadian Journal of Microbiology, 48(8), 728. Fernandez, A., Ocio, M. J., Fernandez, P. S., and Martinez, A. (2001). "Effect of heat activation and inactivati on conditions on germination and thermal resistance parameters of Bacillus cereus spores." International Journal of Food Microbiology, 63(3), 257. Fernandez, P., Blanco, J., Sichel, C., and Ma lato, S. (2005). "Water disinfection by solar photocatalysis using compound parabolic collectors." Catalysis Today, 101(3), 345. Fernandez, R. O., and Pizarro, R. A. (1996). "Lethal effect induced in Pseudomonas aeruginosa exposed to ultraviolet-a radiation." Photochemistry and Photobiology, 64(2), 334. Fernandez, R. O., and Pizarro, R. A. (1999). "Pseudomonas aeruginosa UV-a-induced lethal effect: Influence of salts, nutritional stress and pyocyanine." Journal of Photochemistry and Photobiology B-Biology, 50(1), 59. Foschino, R., Picozzi, C., Civardi, A., Bandi ni, M., and Faroldi, P. (2003). "Comparison of surface sampling methods and cleanability assessment of stai nless steel surfaces subjected or not to shot peening." Journal of Food Engineering, 60, 375. Fox, M. A., and Dulay, M. T. ( 1993). "Heterogeneous photocatalysis." Chemical Reviews, 93(1), 341.

PAGE 309

291 Fu, J. F., Ji, M., Zhao, Y. Q., and Wang, L. Z. (2006). "Kinetics of aqueous photocatalytic oxidation of fu lvic acids in a photocatalys is-ultrafiltration reactor (pur)." Separation and Purification Technology, 50(1), 107. Fujii, K., Ohtani, A., Watanabe, J., Ohgoshi, H., Fujii, T., and Honma, K. (2002). "Highpressure inactivation of Bacillus cereus spores in the presence of argon." International Journal of Food Microbiology, 72(3), 239. Fujishim.A, and Honda, K. (1972). "Elect rochemical photolysi s of water at a semiconductor electrode." Nature, 238(5358), 37–&. Gaillard, S., Leguerinel, I., and Mafart, P. (1998). "Model for combined effects of temperature, ph and water activ ity on thermal inactivation of Bacillus cereus spores." Journal of Food Science, 63(5), 887. Galeano, B., Korff, E., and Nic holson, W. L. (2003) . "Inactivation of ve getative cells, but not spores, of Bacillus anthracis, B-cereus, and B-subtilis on stainless steel surfaces coated with an antimicrobial silverand zinc-containing zeolite formulation." Applied and Environmental Microbiology, 69(7), 4329. Gerhardt, P., Scherrer, R. and Black, S.H. (1972). "Molecular sieving by dormant spore structures." Spores v. , H. O. Halvor son, Hanson R. and Campbell L. L., ed., American Society for Microbiol ogy, Washington, D. C., 68. Gerhardt, P., and Marquis, R. E. (1989) . "Spore thermoresistance mechanisms." Regulation of prokaryotic development, I. Smith, Slepecky, R.A. and Setlow P. , ed., American Society for Micr obiology Washington, D.C., 43. Gerischer, H., and Heller, A. (1991). "The role of oxygen in photooxidation of organicmolecules on semiconductor particles." Journal of Physical Chemistry, 95(13), 5261. Gerischer, H., and Heller, A. (1992). "Photocat alytic oxidation of organic-molecules at TiO2 particles by sunlight in aerated water." Journal of the Electrochemical Society, 139(1), 113. Gogniat, G., Thyssen, M., Denis, M., Pu lgarin, C., and Dukan, S. (2006). "The bactericidal effect of Ti O2 photocatalysis involves ad sorption onto cat alyst and the loss of membrane integrity." Fems Microbiology Letters, 258(1), 18. Gorny, R. L., Mainelis, G., Grinshpun, S. A ., Willeke, K., Dutkiewicz, J., and Reponen, T. (2003). "Release of Streptomyces albus propagules from contaminated surfaces." Environmental Research, 91(1), 45. Goswami, D. Y. (1997). "A review of engin eering developments of aqueous phase solar photocatalytic detoxification and disinfection processes." Journal of Solar Energy Engineering-Transactions of the Asme, 119(2), 101.

PAGE 310

292 Goswami, D. Y. (2003). "Decontamination of ventilation systems using photocatalytic air cleaning technology." Journal of Solar Energy Engi neering-Transactions of the Asme, 125(3), 359. Goswami, D. Y., Trivedi, D. M., and Block, S. S. (1997). "Photocatal ytic disinfection of indoor air." Journal of Solar Energy Engin eering-Transactions of the Asme, 119(1), 92. Goswami, D. Y., Vijayaraghavan, S., Lu, S., and Tamm, G. (2004). "New and emerging developments in solar energy." Solar Energy, 76(1), 33. Guillard, C., Lachheb, H., Houas, A., Ksibi, M., Elaloui, E., and Herrmann, J. M. (2003). "Influence of chemical structure of dyes, of ph and of inor ganic salts on their photocatalytic degradation by TiO2 compar ison of the efficiency of powder and supported TiO2." Journal of Photochemistry and Photobiology a-Chemistry, 158(1), 27. Gumy, D., Morais, C., Bowen, P., Pulgarin, C., Giraldo, S., Haj du, R., and Kiwi, J. (2006). "Catalytic activ ity of commercial of TiO2 pow ders for the abatement of the bacteria (E-coli) under solar simulated light: Influe nce of the isoelectric point." Applied Catalysis B-Environmental, 63(1), 76. Hachisuka, Y., Kozuka, S., and Tsujikawa, M. (1984). "Exosporia and appendages of spores of Bacillus species." Microbiology and Immunology, 28(5), 619. Hachisuka, Y., and Kuno, T. (1976). "Filamentous appendages of Bacillus-cereus spores." Japanese Journal of Microbiology, 20(6), 555. Haeger, A., Kleinschmidt, O., and Hesse, D. (2004). "Kinetics of photocatalyzed gas reactions using titanium dioxi de as the catalyst part i: Photocatalyzed total oxidation of olefines with oxygen." Chemical Engineering & Technology, 27(2), 181. Haeger, A., Kleinschmidt, O., and Hesse, D. (2004). "Kinetics of photocatalyzed gas reactions using titanium dioxi de as the catalyst part ii: Photocatalyzed total oxidation of alkanes with oxygen." Chemical Engineering & Technology, 27(9), 1019. Hamouda, T., Hayes, M. M., Cao, Z. Y., Tonda , R., Johnson, K., Wright, D. C., Brisker, J., and Baker, J. R. (1999). "A novel su rfactant nanoemulsion with broad-spectrum sporicidal activity against Bacillus species." Journal of Infectious Diseases, 180(6), 1939. Hamouda, T., Shih, A. Y., and Baker, J. R. (2002). "A rapid staining technique for the detection of the initiation of germination of bacterial spores." Letters in Applied Microbiology, 34(2), 86.

PAGE 311

293 Harper, J. C., Christensen, P. A., Egerton, T. A., Curtis, T. P., and Gunlazuardi, J. (2001). "Effect of catalyst type on the kinetics of the photoelect rochemical disinfection of water inoculated with E-coli." Journal of Applied Electrochemistry, 31(6), 623– 628. Harwood, C. R. a. C., S.M. (1990). Molecular biology methods for Bacillus, John Wiley & Sons Ltd., London. Henriques, A. O., and Moran, C. P. (2000). "Structure and assembly of the bacterial endospore coat." Methods, 20(1), 95. Herz, R. K. (2004). "Intrinsic kinetics of first-order reactions in photocatalytic membranes and layers." Chemical Engineering Journal, 99(3), 237. Hoffman, A. J., Carraway, E. R., and Ho ffmann, M. R. (1994). "Photocatalytic production of h2o2 and organic peroxi des on quantum-sized semiconductor colloids." Environmental Science & Technology, 28(5), 776. Hoffmann, M. R., Martin, S. T., Choi, W. Y., and Bahnemann, D. W. (1995). "Environmental applications of semiconductor photocatalysis." Chemical Reviews, 95(1), 69. Hong, J. L., and Otaki, M. (2003). "Effects of photocatalysis on biol ogical decolorization reactor and biological ac tivity of isolated photo synthetic bacteria." Journal of Bioscience and Bioengineering, 96(3), 298. Horie, Y., David, D. A., Taya, M., and Tone , S. (1996). "Effects of light intensity and titanium dioxide concentrati on on photocatalytic sterili zation rates of microbial cells." Industrial & Engineeri ng Chemistry Research, 35(11), 3920. Horie, Y., Taya, M., and Tone, S. (1998). "E ffect of cell adsorpti on on photosterilization of Escherichia coli over titanium dioxide-activated charcoal granules." Journal of Chemical Engineering of Japan, 31(6), 922. Horie, Y., Taya, M., and Tone, S. (1998). "Eva luation of photocatalytic sterilization rates of Escherichia coli cells in titanium dioxide slurry irradiated with various light sources." Journal of Chemical Engineering of Japan, 31(4), 577. Horneck, G., Rettberg, P., Reitz, G., Wehner, J ., Eschweiler, U., Stra uch, K., Panitz, C., Starke, V., and Baumstark-Khan, C. (2001) . "Protection of ba cterial spores in space, a contribution to the discussion on panspermia." Origins of Life and Evolution of the Biosphere, 31(6), 527. Horneck, G., Stoffler, D., Eschweiler, U., a nd Hornemann, U. (2001) . "Bacterial spores survive simulated meteorite impact." Icarus, 149(1), 285.

PAGE 312

294 Huang, N. P., Xiao, Z. D., Huang, D., and Yuan, C. W. (1998). "Photochemical disinfection of Escherichia coli with a TiO2 colloid solution and a self-assembled TiO2 thin film." Supramolecular Science, 5(5), 559. Huang, Z., Maness, P. C., Blake, D. M., Wolfru m, E. J., Smolinski, S. L., and Jacoby, W. A. (2000). "Bactericidal mode of titaniaum dioxide photocatalysis." Journal of photochemistry and Photobi ology A: chemistry, 130, 163. Hwang, S., Lee, M. C., and Choi, W. (2003) . "Highly enhanced phot ocatalytic oxidation of co on titania deposited with pt na noparticles: Kinetics and mechanism." Applied Catalysis B-Environmental, 46(1), 49. Ibanez, J. A., Litter, M. I., and Pizarro, R. A. (2003). "Photocatalytic bactericidal effect of TiO2 on enterobacter cloacae. Comparative study with ot her gram (–) bacteria." Journal of Photochemistry and Photobiology a-Chemistry, 157(1), 81. Inel, Y., and Okte, A. N. (1996). "Photocatalyt ic degradation of ma lonic acid in aqueous suspensions of titanium dioxide: An initial kinetic investigation of co2 photogeneration." Journal of Photochemistry and Photobiology a-Chemistry, 96(1– 3), 175. Ireland, J. A. W., and Hanna, P. C. (2002) . "Amino acidand purine ribonucleosideinduced germination of Bacillus anthracis delta sterne endospor es: Gers mediates responses to aromatic ring structures." Journal of Bacteriology, 184(5), 1296. Ireland, J. C., Klostermann, P., Rice, E. W ., and Clark, R. M. (1993). "Inactivation of Escherichia-coli by titanium-dioxide phot ocatalytic oxidation." Applied and Environmental Microbiology, 59(5), 1668. Ivanova, N., Sorokin, A., Anderson, I., Ga lleron, N., Candelon, B., Kapatral, V., Bhattacharyya, A., Reznik, G., Mikhailova , N., Lapidus, A., Chu, L., Mazur, M., Goltsman, E., Larsen, N., D'Souza, M., Walunas, T., Grechkin, Y., Pusch, G., Haselkorn, R., Fonstein, M., Ehrlich, S. D., Overbeek, R., and Kyrpides, N. (2003). "Genome sequence of Bacillus cereus and comparative analysis with Bacillus anthracis." Nature, 423(6935), 87. Jacoby, W. A., Maness, P. C., Wolfrum, E. J., Blake, D. M., and Fennell, J. A. (1998). "Mineralization of bacterial cell ma ss on a photocatalytic surface in air." Environmental Science & Technology, 32(17), 2650. Jang, H. D., Kim, S. K., and Kim, S. J. (2001). "Effect of particle size and phase composition of titanium dioxide nanoparticles on the photocatalytic properties." Journal of Nanoparticle Research, 3(2), 141. Joo, J., Kwon, S. G., Yu, T., Cho, M., Lee, J., Yoon, J., and Hyeon, T. (2005). "Largescale synthesis of TiO2 nanorods via nonhydrolytic sol-gel ester elimination reaction and their application to photocatalytic inactivation of E. coli." Journal of Physical Chemistry B, 109(32), 15297.

PAGE 313

295 Jorge De Lara, P. S. F., Paula M. Periago, Al fredo Palop. (2002). "Irra diation of spores of Bacillus cereus and Bacillus subtilis with electron beams." Innovative Food Science and Emerging Technologies, 3, 379. Joyce, E., Phull, S. S., Lorimer, J. P., a nd Mason, T. J. (2003). "The development and evaluation of ultrasound for the treatment of bacterial suspen sions. A study of frequency, power and sonication time on cultured Bacillus species." Ultrasonics Sonochemistry, 10(6), 315. Kakita, Y., Kashige, N., Miake, F., and Wa tanabe, K. (1997). "Photocatalysis-dependent inactivation of lactobacillus phage pl-1 by a ceramics preparation." Bioscience Biotechnology and Biochemistry, 61(11), 1947. Kashige, N., Kakita, Y., Nakashima, Y ., Miake, F., and Watanabe, K. (2001). "Mechanism of the photocat alytic inactivation of lactobacillus casei phage pl-1 by titania thin film." Current Microbiology, 42(3), 184. Keleher, J., Bashant, J., Heldt, N., Johnson, L., and Li, Y. Z. (2002). "Photo-catalytic preparation of silver-coated TiO2 particles for antibacterial applications." World Journal of Microbiology & Biotechnology, 18(2), 133. Kersters, I., De Keyser, T., and Verstraete , W. (1998). "Sensitivity of bacteria to photoactivated titanium dioxide in comparison with UV irradiation." Indian Journal of Engineering and Materials Sciences, 5(4), 211. Kikuchi, Y., Sunada, K., Iyoda, T., Hash imoto, K., and Fujishima, A. (1997). "Photocatalytic bactericidal effect of TiO2 thin films: Dynamic view of the active oxygen species responsible for the effect." Journal of Photochemistry and Photobiology a-Chemistry, 106(1-3), 51. Kim, B., Kim, D., Cho, D., and Cho, S. (2003) . "Bactericidal effect of TiO2 photocatalyst on selected food-borne pathogenic bacteria." Chemosphere, 52(1), 277. Kim, S. C., and Lee, D. K. (2005). "Prepa ration of TiO2-coated hollow glass beads and their application to the control of algal growth in eutrophic water." Microchemical Journal, 80(2), 227. Koizumi, Y., and Taya, M. (2002). "Kinetic evaluation of biocidal activity of titanium dioxide against phage ms2 consideri ng interaction betw een the phage and photocatalyst particles." Biochemical Engineering Journal, 12(2), 107. Koizumi, Y., and Taya, M. (2002). "Photocatal ytic inactivation rate of phage ms2 in titanium dioxide suspensions cont aining various ionic species." Biotechnology Letters, 24(6), 459.

PAGE 314

296 Koizumi, Y., Yamada, R., Nishioka, M., Ma tsumura, Y., Tsuchido, T., and Taya, M. (2002). "Deactiva tion kinetics of Escherichia coli cells correlated with intracellular superoxide dismutase activity in photorea ction with titanium dioxide particles." Journal of Chemical Te chnology and Biotechnology, 77(6), 671. Kondo, M. M., Orlanda, J. F. F., Ferreira, M. D., and Grassi, M. T. (2003). "Proposition of a photocatalytic reactor to in activate airborne microorganisms." Quimica Nova, 26(1), 133. Krishna, V., Pumprueg, S., Lee, S. H., Zhao, J., Sigmund, W., Koopman, B., and Moudgil, B. M. (2005). "Photo catalytic disinfection with titanium dioxide coated multi-wall carbon nanotubes." Process Safety and Environmental Protection, 83(B4), 393. Kuhn, K. P., Chaberny, I. F., Massholder, K ., Stickler, M., Benz, V. W., Sonntag, H. G., and Erdinger, L. (2003). "Disinfection of surfaces by photocatalytic oxidation with titanium dioxide and UVA light." Chemosphere, 53(1), 71. Laot, N., Narkis, N., Neeman, I., Bila novic, D., and Armon, R. (1999). "TiO2 photocatalytic inactivatin of selected microorganisms under various conditions. Sunlight, intermittent and variable irra diation intensity, cds augmentation and entrapment of TiO2 into sol-gel." J. Adv. Oxid. Technol., 4, 97. Larson, M. A., and Marinas, B. J. (2003). "Inactivation of Bacillus subtilis spores with ozone and monochloramine." Water Res, 37(4), 833. Lee, S., Nakamura, M., and Ohgaki, S. ( 1998). "Inactivation of phage q beta by 254nm UV light and titanium dioxide photocatalyst." Journal of Environmental Science and Health Part a-Toxic/Hazardous Subs tances & Environmental Engineering, 33(8), 1643. Lee, S., Nishida, K., Otaki, M., and Ohgaki , S. (1997). "Photocatalytic inactivation of phage q beta by immobilized titanium dioxide mediated photocatalyst." Water Science and Technology, 35(11), 101. Lee, S. H., Pumprueg, S., Moudgil, B., a nd Sigmund, W. (2005). "Inactivation of bacterial endospores by photo catalytic nanocomposites." Colloids and Surfaces BBiointerfaces, 40(2), 93. Lee, W., Shen, H. S., Dwight, K., and Wo ld, A. (1993). "Effect of silver on the photocatalytic activity of TiO2." Journal of Solid State Chemistry, 106(2), 288– 294. Lei, H., Li, D. Q., Lin, Y. J., Evans, D. G., and Xue, D. (2005). "Influence of nano-mgo particle size on bacter icidal action against Bacillus subtilis var. Niger." Chinese Science Bulletin, 50(6), 514.

PAGE 315

297 Leuschner, R. G. K., Ferdinando, D. P., and L illford, P. J. (2000). "Structural analysis of spores of Bacillus subtilis during germination and outgrowth." Colloids and Surfaces B-Biointerfaces, 19(1), 31. Li, C. S., Tseng, C. C., Lai, H. H., and Chang, C. W. (2003). "Ultraviolet germicidal irradiation and titanium dioxide photocatalyst for controlling legionella pneumophila." Aerosol Science and Technology, 37(12), 961. Liau, L. C. K., and Tung, M. T. (2006). "Kine tic investigation of phot ocatalytic effects on poly(vinyl butyral) photodegradation." Industrial & Engineering Chemistry Research, 45(7), 2199. Lin, C. Y., and Li, C. S. (2003). "Effectiven ess of titanium dioxide photocatalyst filters for controlling bioaerosols." Aerosol Science and Technology, 37(2), 162. Lin, C. Y., and Li, C. S. (2003). "Inactiva tion of microorganisms on the photocatalytic surfaces in air." Aerosol Science and Technology, 37(12), 939. Lindberg, C., and Horneck, G. (1991). "Ac tion spectra for survival and spore photoproduct formation of Bacillus-subtilis irradiated with sh ort-wavelength (200300 nm) uv at atmospheric-pressure and invacuo." Journal of Photochemistry and Photobiology B-Biology, 11(1), 69. Link, L., Sawyer, J., Venkateswaran, K., and Nicholson, W. (2004). "Extreme spore UV resistance of Bacillus pumilus isolates obtained from an ultraclean spacecraft assembly facility." Microb Ecol, 47(2), 159. Linsebigler, A. L., Lu, G. Q., and Yates, J. T. (1995). "Photocatalysis on TiO2 surfaces principles, mechanisms, and selected results." Chemical Reviews, 95(3), 735. Liu, H. L., and Yang, T. C. K. ( 2003). "Photocatalytic inactivation of Escherichia coli and lactobacillus helveticus by zno and TiO2 activated with ultraviolet light." Process Biochemistry, 39(4), 475. Lonnen, J., Kilvington, S., Kehoe, S. C., Al-Touati, F., and McGuigan, K. G. (2005). "Solar and photocatalytic disinfection of protozoan, fungal and bacterial microbes in drinking water." Water Research, 39(5), 877. Lopez, J. E. O., and Jacoby, W. A. (2002). "Microfibrous mesh coated with titanium dioxide: A self-sterilizing, self-cleaning filter." Journal of the Air & Waste Management Association, 52(10), 1206. Madigan, M. T., Martinko, J. M., and Parker, J. (2000). Brock biology of microorganisms, Upper Saddle River, Prentice Hall, NJ.

PAGE 316

298 Maness, P. C., Smolinski, S., Blake, D. M., Hu ang, Z., Wolfrum, E. J., and Jacoby, W. A. (1999). "Bactericidal activity of photocat alytic TiO2 reaction: Toward an understanding of its killing mechanism." Applied and Environmental Microbiology, 65(9), 4094. Marquis, R. E., and Bender, G. R. (1985) . "Mineralization and heat-resistance of bacterial–spores." Journal of Bacteriology, 161(2), 789. Marquis, R. E., and Shin, S. Y. (1994). "Miner alization and responses of bacterial-spores to heat and oxidative agents." Fems Microbiology Reviews, 14(4), 375. Matos, J., Laine, J., and Herrmann, J. M. ( 2001). "Effect of the type of activated carbons on the photocatalytic degradation of aque ous organic pollutants by UV-irradiated titania." Journal of Catalysis, 200(1), 10. Matsunaga, T., and Okochi, M. (1995). "TiO 2-mediated photochemical disinfection of Escherichia-coli using optical fibers." Environmental Science & Technology, 29(2), 501. Matsunaga, T., Tomoda, R., Nakajima, T., a nd Wake, H. (1985). "Photoelectrochemical sterilization of microbial-cel ls by semiconductor powders." Fems Microbiology Letters, 29(1), 211. Matsuo, S., Anraku, Y., Yamada, S., Honjo, T., Matsuo, T., and Wakita, H. (2001). "Effects of photocatalytic reactions on marine plankton: Titanium dioxide powder as catalyst on the body surface." Journal of Environmental Science and Health Part a-Toxic/Hazardous Substances & Environmental Engineering, 36(7), 1419. McLoughlin, O. A., Ibanez, P. F., Gernjak, W., Rodriguez, S. M., and Gill, L. W. (2004). "Photocatalytic disinfection of wate r using low cost compound parabolic collectors." Solar Energy, 77(5), 625. McLoughlin, O. A., Kehoe, S. C., McGuigan, K. G., Duffy, E. F., Al Touati, F., Gernjak, W., Alberola, I. O., Rodriguez, S. M., and Gill, L. W. (2004). "Solar disinfection of contaminated water: A comparison of three small-scale reactors." Solar Energy, 77(5), 657. Mehrotra, K., Yablonsky, G. S., and Ray, A. K. (2003). "Kinetic studi es of photocatalytic degradation in a TiO2 slurry system : Distinguishing working regimes and determining rate dependences." Industrial & Engineering Chemistry Research, 42(11), 2273. Mehrotra, K., Yablonsky, G. S., and Ray, A. K. (2005). "Macro kinetic studies for photocatalytic degradation of benz oic acid in immobilized systems." Chemosphere, 60(10), 1427.

PAGE 317

299 Melian, J. A. H., Rodriguez, J. M. D., Su arez, A. V., Rendon, E. T., do Campo, C. V., Arana, J., and Pena, J. P. (2000). "The photocatalytic disinfection of urban waste waters." Chemosphere, 41(3), 323. Meng, Y. B., Huang, X., Wu, Y. X., Wang, X. M., and Qian, Y. (2002). "Kinetic study and modeling on photocatalytic degradation of para-chlorobenzoate at different light intensities." Environmental Pollution, 117(2), 307. Mills, A., and LeHunte, S. (1997). "An overview of semiconductor photocatalysis." Journal of Photochemistry and Photobiology a-Chemistry, 108(1), 1. Mills, A., and Wang, J. S. (1999). "The kine tics of semiconductor photocatalysis: Light intensity effects." Zeitschrift Fur Physikalische Ch emie-International Journal of Research in Physical Chemistry & Chemical Physics, 213, 49. Mills, A., Wang, J. S., and Ollis, D. F. (2006). "Kinetics of liquid phase semiconductor photoassisted reactions: Supporting observati ons for a pseudo-steady-state model." Journal of Physical Chemistry B, 110(29), 14386-14390. Minero, C. (1995). "A rigorous kinetic appro ach to model primary oxidative steps of photocatalytic degradations." Solar Energy Materials and Solar Cells, 38(1), 421. Moir, A., Corfe, B. M., and Behravan, J. (2002). "Spore germination." Cellular and Molecular Life Sciences, 59(3), 403. Moir, A., Kemp, E. H., Robinson, C., and Corf e, B. M. (1994). "The genetic-analysis of bacterial spore germination." Journal of Applied Bacteriology, 76, S9–S16. Moir, A., and Smith, D. A. (1990). "The ge netics of bacterial spore germination." Annual Review of Microbiology, 44, 531. Montgomery, J. M. (1985). Water treatment pr inciples and design, John Wiley & Sons, New York, NY. Moore, G., and Griffith, C. (2002). "A comparison of surface sampling methods for detecting coliforms on food contact surfaces." Food Microbiology, 19(1), 65. Morioka, T., Saito, T., Nara, Y., and Onoda, K. (1988). "Antibacterial action of powdered semiconductor on a serotype-g streptococcus-mutans." Caries Research, 22(4), 230. Muruganandham, M., and Swaminathan, M. (2006). "TiO2-UV photo catalytic oxidation of reactive yellow 14: Effect of operational parameters." Journal of Hazardous Materials, 135(1-3), 78-86.

PAGE 318

300 Nakagawa, Y., Wakuri, S., Sakamoto, K., a nd Tanaka, N. (1997). "The photogenotoxicity of titanium dioxide particles." Mutation Research-Genetic Toxicology and Environmental Mutagenesis, 394(1-3), 125. Nicewonger, R., and Begley, T. P. (1997) . "Synthesis of the spore photoproduct." Tetrahedron Letters, 38(6), 935. Nicholson, W. L., and Galeano, B. (2003). "UV resistance of Bacillus anthracis spores revisited: Validation of Bacillus subtilis spores as UV surrogates for spores of banthracis sterne." Applied and Environm ental Microbiology, 69(2), 1327. Nicholson, W. L., Munakata, N., Horneck, G., Melosh, H. J., and Setlow, P. (2000). "Resistance of Bacillus endospores to extreme terres trial and extraterrestrial environments." Microbiology and Molecular Biology Reviews, 64(3), 548–+. Nicholson, W. L., and Sc huerger, A. C. (2005). "Bacillus subtilis spore survival and expression of germination-induced bi oluminescence after prolonged incubation under simulated mars atmospheric pressure and composition: Implications for planetary protection and lithopanspermia." Astrobiology, 5(4), 536. Nonami, T., Hase, H., and Funakoshi, K. (2003). "Apatite-coated titaniaum dioxide photocatalyst." Materials science forum ( 2003 trans tech publications. switzerland), 439, 337-343. Ohko, Y., Fujishima, A., and Hashimoto, K. (1998). "Kinetic analysis of the photocatalytic degradation of gas-phase 2-propanol under mass transport-limited conditions with a TiO2 film photocatalyst." Journal of Physical Chemistry B, 102(10), 1724. Ohko, Y., Hashimoto, K., and Fujishima, A. ( 1997). "Kinetics of phot ocatalytic reactions under extremely low-intensity UV illumina tion on titanium dioxide thin films." Journal of Physical Chemistry A, 101(43), 8057. Ohko, Y., Ikeda, K., Rao, T. N., Ha shimoto, K., and Fujishima, A. (1999). "Photocatalytic reaction ki netics on TiO2 thin films under light-limited and mass transport-limited conditions." Zeitschrift Fur Physikalische Chemie-International Journal of Research in Physical Chemistry & Chemical Physics, 213, 33. Ollis, D. F. (2000). "Photocatalytic purificati on and remediation of contaminated air and water." Comptes Rendus De L Academie Des Scie nces Serie Ii Fascicule C-Chimie, 3(6), 405. Ollis, D. F. (2005). "Kinetic disguises in heterogeneous photocatalysis." Topics in Catalysis, 35(3), 217. Ollis, D. F. (2005). "Kinetics of liquid pha se photocatalyzed reactions: An illuminating approach." Journal of Physical Chemistry B, 109(6), 2439.

PAGE 319

301 Ollis, D. F., Pelizzetti, E., and Serpone, N. (1991). "Photocatalyzed destruction of water contaminants." Environmental Science & Technology, 25(9), 1522. Oppezzo, O. J., and Pizarro, R. A. (2001). "Subl ethal effects of ultr aviolet a radiation on enterobacter cloacae." Journal of Photochemistry and Photobiology B-Biology, 62(3), 158. Otaki, M., Hirata, T., and Ohgaki, S. ( 2000). "Aqueous microorganisms inactivation by photocatalytic reaction." Water Science and Technology, 42(3), 103. Paidhungat, M., Setlow, B., Dani els, W. B., Hoover, D., Papa fragkou, E., and Setlow, P. (2002). "Mechanisms of induc tion of germination of Bacillus subtilis spores by high pressure." Applied and Environmental Microbiology, 68(6), 3172. Paidhungat, M., Setlow, B., Driks, A., and Setlo w, P. (2000). "Characterization of spores of Bacillus subtilis which lack dipicolinic acid." Journal of Bacteriology, 182(19), 5505. Paidhungat, M., and Setlow, P. (2000). "Role of ger proteins in nutrient and nonnutrient triggering of spore germination in Bacillus subtilis." Journal of Bacteriology, 182(9), 2513. Pal, A., Min, X., Yu, L. E., Pehkonen, S. O., and Ray, M. B. (2005). "Photocatalytic inactivation of bioaerosols by TiO2 coated membrane." International Journal of Chemical Reactor Engineering, 3, –. Palop, A., Rutherford, G. C., and Marquis, R. E. (1996). "Hydroper oxide inactivation of enzymes within spores of Bacillus megaterium atcc19213." Fems Microbiology Letters, 142(2), 283. Palop, A., Rutherford, G. C., and Marquis, R. E. (1998). "Inactivation of enzymes within spores of Bacillus megaterium atcc 19213 by hydroperoxides." Canadian Journal of Microbiology, 44(5), 465. Peng, J. S., Tsai, W. C., and Chou, C. C. (2001). "Surface characteristics of Bacillus cereus and its adhesion to stainless steel." International Journal of Food Microbiology, 65(1), 105. Pham, H. N., Mcdowell, T., and Wilkins, E. (1995). "Photocatalytically-mediated disinfection of water using TiO2 as a catalyst and spore-forming Bacillus-pumilus as a model." Journal of Environmental Science and Health Part a-Environmental Science and Engineering & Toxic and Hazardous Substance Control, 30(3), 627– 636.

PAGE 320

302 Pham, H. N., Wilkins, E., Heger, K. S., and Kauffman, D. (1997). "Quantitative analysis of variations in initial Bacillus pumilus spore densities in aqueous TiO2 suspension and design of a photocatalytic reactor." Journal of Environmental Science and Health Part a-Environmental Scienc e and Engineering & Toxic and Hazardous Substance Control, 32(1), 153. Pichat, P. (1994). "Partial or complete hete rogeneous photocatalytic oxidation of organiccompounds in liquid organi c or aqueous phases." Catalysis Today, 19(2), 313-333. Pol, I. E., van Arendonk, W. G. C., Mastw ijk, H. C., Krommer, J., Smid, E. J., and Moezelaar, R. (2001). "Sensitivities of germ inating spores and carvacrol-adapted vegetative cells and spores of Bacillus cereus to nisin and puls ed-electric-field treatment." Applied and Environmental Microbiology, 67(4), 1693. Pozzo, R. L., Baltanas, M. A., and Cassano, A. E. (1997). "Supported titanium oxide as photocatalyst in water decontam ination: State of the art." Catalysis Today, 39(3), 219. Rabinovi, L., and Dasilva, S. M. (1 973). "Studies on sporulation of a Bacillus strain .4. Further evidence mn-2+ ion activit y in endotrophic sporulation." Memorias Do Instituto Oswaldo Cruz, 71(1-2), 149. Rana, S., and Misra, R. D. K. (2005). "The an ti-microbial activity of titania-nickel ferrite composite nanoparticles." JOM, ABI/INFORM Trade & Industry, 57(12), 65-69. Rana, S., Rawat, J., and Misra, R. D. K. (2005). "Anti-microbial active composite nanoparticles with magnetic core an d photocatalytic shell: TiO2-nife2o4 biomaterial system." Acta Biomaterialia, 1(6), 691. Ray, A. K., and Beenackers, A. (1997). "Nove l swirl-flow reactor fo r kinetic studies of semiconductor photocatalysis." Aiche Journal, 43(10), 2571-2578. Read, T. D., Peterson, S. N., Tourasse, N., Baillie, L. W., Paulsen, I. T., Nelson, K. E., Tettelin, H., Fouts, D. E., Eisen, J. A., Gill, S. R., Holtzapple, E. K., Okstad, O. A., Helgason, E., Rilstone, J., Wu, M., Kolonay, J. F., Beanan, M. J., Dodson, R. J., Brinkac, L. M., Gwinn, M., DeBoy, R. T ., Madpu, R., Daugherty, S. C., Durkin, A. S., Haft, D. H., Nelson, W. C., Peterson, J. D., Pop, M., Khouri, H. M., Radune, D., Benton, J. L., Mahamoud, Y., Jiang, L., Han ce, I. R., Weidman, J. F., Berry, K. J., Plaut, R. D., Wolf, A. M., Watkins, K. L., Nierman, W. C., Hazen, A., Cline, R., Redmond, C., Thwaite, J. E., White, O ., Salzberg, S. L., Thomason, B., Friedlander, A. M., Koehler, T. M., Hanna, P. C., Kolsto, A. B., and Fraser, C. M. (2003). "The genome sequence of Bacillus anthracis ames and comparison to closely related bacteria." Nature, 423(6935), 81-6.

PAGE 321

303 Rettberg, P., Eschweiler, U., Strauch, K., Reitz, G., Horneck, G., Wanke, H., Brack, A., and Barbier, B. (2002). "Survival of microorganisms in space protected by meteorite material: Results of the experime nt 'exobiologie' of the perseus mission." Space Life Sciences: Extraterrestrial Organic Chemistry, Uv Radiation on Biological Evolution, an d Planetary Protection, 30(6), 1539. Rice, E. W., Adcock, N. J., Sivaganesan, M ., and Rose, L. J. (2005). "Inactivation of spores of Bacillus anthracis sterne, Bacillus cereus, and Bacillus thuringiensis subsp. Israelensis by chlorination." Applied and Environmental Microbiology, 71(9), 5587. Richardson, S. D., Thruston, A. D., Collette, T. W., Patterson, K. S., Lykins, B. W., and Ireland, J. C. (1996). "Identification of TiO2/UV disinfection byproducts in drinking water." Environmental Science & Technology, 30(11), 3327. Riesenman, P. J., and Nicholson, W. L. (2000). "Role of the spore coat layers in Bacillus subtilis spore resistance to hydrogen peroxide, artificial UV-c, UV-b, and solar UV radiation." Applied and Environmental Microbiology, 66(2), 620-626. Rincon, A. G., and Pulgarin, C. (200 3). "Photocatalytical inactivation of E. coli: Effect of (continuous-intermittent) light intensit y and of (suspended-fixed) TiO2 concentration." Applied Catalysis B-Environmental, 44(3), 263-284. Rincon, A. G., and Pulgarin, C. (2004). "Bactericidal action of illuminated TiO2 on pure Escherichia coli and natural bacterial consortia: Post-irradiation events in the dark and assessment of the effective disinfection time." Applied Catalysis BEnvironmental, 49(2), 99. Rincon, A. G., and Pulgarin, C. (2004). "Eff ect of ph, inorganic ions, organic matter and h2o2 on E-coli k12 photocatalytic inactivation by TiO2 implications in solar water disinfection." Applied Catalysis B-Environmental, 51(4), 283. Rincon, A. G., and Pulgarin, C. (2004). "Field solar E-coli inactivation in the absence and presence of TiO2: Is UV solar dose an appr opriate parameter for standardization of water solar disinfection?" Solar Energy, 77(5), 635. Rincon, A. G., and Pulgarin, C. (2005). "Use of coaxial photocatalyt ic reactor (caphore) in the TiO2 photo-assisted treatment of mixed E-coli and Bacillus sp and bacterial community present in wastewater." Catalysis Today, 101(3), 331. Rincon, A. G., and Pulgarin, C. (2006). "Comparative evaluation of fe3+ and TiO2 photoassisted processes in solar photo catalytic disinfection of water." Applied Catalysis B-Environmental, 63(3-4), 222. Rincon, A. G., Pulgarin, C., Adler, N., a nd Peringer, P. (2001) . "Interaction between Ecoli inactivation and dbp-p recursors dihydroxybenzen e isomers in the photocatalytic process of drinking -water disinfection with TiO2." Journal of Photochemistry and Photobiology a-Chemistry, 139(2), 233.

PAGE 322

304 Rosenquist, H., Smidt, L., Andersen, S. R ., Jensen, G. B., and Wilcks, A. (2005). "Occurrence and significance of Bacillus cereus and Bacillus thuringiensis in ready-to-eat food." Fems Microbiology Letters, 250(1), 129-136. Sabli, M. Z. H., Setlow, P., and Waites, W. M. (1996). "The effect of hypochlorite on spores of Bacillus subtilis lacking small acid-soluble proteins." Letters in Applied Microbiology, 22(6), 405-407. Saito, T., Iwase, T., Horie, J., and Mo rioka, T. (1992). "Mode of photocatalytic bactericidal action of powdered semic onductor TiO2 on mutans Streptococci." Journal of Photochemistry and Photobiology B-Biology, 14(4), 369-379. Sakai, H., Ito, E., Cai, R. X., Yoshioka, T ., Kubota, Y., Hashimoto, K., and Fujishima, A. (1994). "Intracellular ca2+ concentration cha nge of t24 cell under irradiation in the presence of TiO2 ultrafine particles." Biochimica Et Biophysica Acta-General Subjects, 1201(2), 259-265. Salih, F. M. (2002). "Enhancement of solar inactivation of Escherichia coli by titanium dioxide photocatalytic oxidation." Journal of Applied Microbiology, 92(5), 920926. Sambrook, J., Fritsch, E., and Maniatis, T. (1989). Molecular cloning–a laboratory manual, Cold spring harbour laboratory press, New York. Sanderson, W. T., Hein, M. J., Taylor, L., Curwin, B. D., Kinnes, G. M., Seitz, T. A., Popovic, T., Holmes, H. T., Kellum, M. E., McAllister, S. K., Whaley, D. N., Tupin, E. A., Walker, T., Freed, J. A., Small, D. S., Klusaritz, B ., and Bridges, J. H. (2002). "Surface sampling methods for Bacillus anthracis spore contamination." Emerging Infectious Diseases, 8(10), 1145-1151. Sato, T., Koizumi, Y., and Taya, M. (2003). "Photocatalytic deactiv ation of airborne microbial cells on TiO2-loaded plate." Biochemical Engineering Journal, 14(2), 149-152. Sato, T., and Taya, M. (2006). "Enhancement of phage inactivation using photocatalytic titanium dioxide particles with di fferent crystalline structures." Biochemical Engineering Journal, 28(3), 303-308. Sauer, T., Neto, G. C., Jose, H. J., and Mo reira, R. (2002). "Kine tics of photocatalytic degradation of reactive dyes in a TiO2 slurry reactor." Journal of Photochemistry and Photobiology a-Chemistry, 149(1-3), 147-154. Schiza, M. V., Perkins, D. L., Priore, R. J ., Setlow, B., Setlow, P., Bronk, B. V., Wong, D. M., and Myrick, M. L. (2005). "Improve d dispersion of bacterial endospores for quantitative infrared sampling on gold coated porous alumina membranes." Appl Spectrosc, 59(8), 1068-74.

PAGE 323

305 Schoeni, J. L., and Wong, A. C. L. (2005). "Bacillus cereus food poisoning and its toxins." Journal of Food Protection, 68(3), 636-648. Setlow, B., Cowan, A. E., and Setlow, P. (2003). "Germination of spores of Bacillus subtilis with dodecylamine." J Appl Microbiol, 95(3), 637-48. Setlow, B., Loshon, C. A., Genest, P. C., Cowan, A. E., Setlow, C., and Setlow, P. (2002). "Mechanisms of killing spores of Bacillus subtilis by acid, alkali and ethanol." Journal of Applied Microbiology, 92(2), 362-375. Setlow, B., Melly, E., and Setlow, P. (2001). "Properties of spores of Bacillus subtilis blocked at an intermediate stage in spore germination." Journal of Bacteriology, 183(16), 4894-4899. Setlow, P. (1994). "Mechanisms which contribute to the long-term surv ival of spores of Bacillus species." Journal of Applied Bacteriology, 76, S49-S60. Setlow, R. B., and Carrier, W. L. (1966). "Pyr imidine dimers in ul traviolet-irradiated dnas." Journal of Molecular Biology, 17(1), 237-&. Seven, O., Dindar, B., Aydemir, S., Metin, D ., Ozinel, M. A., and Ic li, S. (2004). "Solar photocatalytic disinfection of a group of bacteria and fungi aqueous suspensions with TiO2, zno and sahara desert dust." Journal of Photochemistry and Photobiology a-Chemistry, 165(1-3), 103-107. Shang, J., Du, Y. G., and Xu, Z. L. ( 2002). "Kinetics of gasphase photocatalytic oxidation of heptane over TiO2." Reaction Kinetics and Catalysis Letters, 75(2), 259-265. Sharon, M., and Pal, B. (1996). "Reactivity of nano-size partic les vis-a-vis large particles and their applications." Bulletin of Electrochemistry, 12(3-4), 219-233. Shin, S. Y., Calvisi, E. G., Beaman, T. C., Pa nkratz, H. S., Gerhardt, P., and Marquis, R. E. (1994). "Microscopic and thermal charac terization of hydrogen-peroxide killing and lysis of spores and protection by transition-metal ions, chelators, and antioxidants." Applied and Environmental Microbiology, 60(9), 3192-3197. Shiraishi, F., Toyoda, K., Fukinbara, S., Obuc hi, E., and Nakano, K. (1999). "Photolytic and photocatalytic treatment of an aqueous solution containing microbial cells and organic compounds in an annular-flow reactor." Chemical Engineering Science, 54(10), 1547-1552. Sjogren, J. C., and Sierka, R. A. (1994). "Inactivation of phage ms2 by iron-aided titanium-dioxide photocatalysis." Applied and Environm ental Microbiology, 60(1), 344-347.

PAGE 324

306 Sjollema, J., Vandermei, H. C., Uyen, H. M. W., and Busscher, H. J. (1990). "The influence of collector and bacterial-cell surface-properties on the deposition of oral Streptococci in a parallel plate flow cell." Journal of Adhesion Science and Technology, 4(9), 765-777. Skomurski, J. F., Racine, F. M., and Vary , J. C. (1983). "Steady-state fluorescence anisotropy changes of 1,6-diphenyl-1,3,5,-hexatriene in membranes from Bacillusmegaterium spores." Biochimica Et Biophysica Acta, 731(3), 428-436. Slepecky, R., and Foster, J. W. (1959). "Alte rations in metal cont ent of spores of Bacillus-megaterium and the effect on some spore properties." Journal of Bacteriology, 78(1), 117-123. Slieman, T. A., and Nicholson, W. L. (2001). "Role of dipicolinic ac id in survival of Bacillus subtilis spores exposed to artificial and solar UV radiation." Applied and Environmental Microbiology, 67(3), 1274-1279. Sokal, R. R. a. R., F.J. (1994). Biometry: The principles and practice of statistics in biological research. Sokmen, M., Candan, F., and Sumer, Z. (2001). "Disinfection of E-coli by the agTiO2/UV system: Lipidperoxidation." Journal of Photochemi stry and Photobiology a-Chemistry, 143(2-3), 241. Son, H., Cho, M., Kim, J., Oh, B., Chung, H. M., and Yoon, J. (2005). "Enhanced disinfection efficiency of mechanically mixed oxidants with free chlorine." Water Research, 39(4), 721. Stevenson, M., Bullock, K., Lin, W. Y ., and Rajeshwar, K. (1997). "Sonolytic enhancement of the bactericidal activity of irradiated titanium dioxide suspensions in water." Research on Chemical Intermediates, 23(4), 311. Subramanian, V., Kamat, P. V., and Wolf, E. E. (2003). "Mass-transfer and kinetic studies during the photocatalytic degradation of an azo dye on optically transparent electrode thin film." Industrial & Engineering Chemistry Research, 42(10), 2131– 2138. Sun, D. D., Tay, J. H., and Tan, K. M. ( 2003). "Photocatalytic degradation of e-coliform in water." Water Research, 37(14), 3452. Sunada, K., Kikuchi, Y., Hashimoto, K., a nd Fujishima, A. (1998). "Bactericidal and detoxification effects of TiO2 thin film photocatalysts." Environmental Science & Technology, 32(5), 726. Sunada, K., Watanabe, T., and Hashimoto, K. (2003). "Bactericidal activity of copperdeposited TiO2 thin film unde r weak UV light illumination." Environmental Science & Technology, 37(20), 4785.

PAGE 325

307 Suzuki, E., Hayashi, Y., Shimomura, Y., Yoshida, S., Usami, H., Nakasa, A., and Fujimatsu, H. (2005). "Kinetics study on photocatalytic hydroge n generation from hydrogen sulfide." Journal of Chemical Engineering of Japan, 38(10), 824. Suzuki, Y., and Rode, L. J. (1969). "Eff ect of lysozyme on resting spores of Bacillus megaterium." Journal of Bacteriology, 98(1), 238–&. Tamai, H., Katsu, N., Ono, K., and Yasuda , H. (2002). "Simple preparation of TiO2 particles dispersed activat ed carbons and their phot o-sterilization activity." Journal of Materials Science, 37(15), 3175. Tao, H., Wei, W. Z., and Zhang, S. F. ( 2004). "Photocatalytic inhibitory effect of immobilized TiO2 semic onductor on the growth of Escherichia coli studied by acoustic wave impedance analysis." Journal of Photochemistry and Photobiology a-Chemistry, 161(2), 193. Tatsuma, T., Takeda, S., Saitoh, S., Ohko, Y., and Fujishima, A. (2003). "Bactericidal effect of an energy storage Ti O2-wo3 photocatalyst in dark." Electrochemistry Communications, 5(9), 793. Tokumura, M., Znad, H. T., and Kawase, Y. (2006). "Modeling of an external light irradiation slurry photoreactor: UV li ght or sunlight-photoassisted fenton discoloration of azo-dye orange ii wi th natural mineral tourmaline powder." Chemical Engineering Science, 61(19), 6361. Tone, S., Taya, M., Kato, S., Horie, Y., Ashi kaga, Y., and Joo, H. K. (1993). "Kineticanalysis of photochemical ster ilization of thermoduric bact erial-spores in slurry of semiconductor catalyst part icles with aeration." Kagaku Kogaku Ronbunshu, 19(6), 1149. Trapalis, C. C., Keivanidis, P., Kordas, G., Zaharescu, M., Crisan, M., Szatvanyi, A., and Gartner, M. (2003). "TiO2(fe3+) nanostruc tured thin films w ith antibacterial properties." Thin Solid Films, 433(1-2), 186. Uyen, H. M. W., Schakenraad, J. M., Sjolle ma, J., Noordmans, J., Jongebloed, W. L., Stokroos, I., and Busscher, H. J. (1990). "Amount and surface-st ructure of albumin adsorbed to solid substrata with different wettabilities in a parallel plate flow cell." Journal of Biomedical Materials Research, 24(12), 1599. Uyguner, C. S., and Bekbolet, M. (2004). "Photo catalytic degradation of natural organic matter: Kinetic considerations and light intensity dependence." International Journal of Photoenergy, 6(2), 73-80. Vaid, A., and Bishop, A. H. (1998). "The dest ruction by microwave radiation of bacterial endospores and amplification of the released DNA." Journal of Applied Microbiology, 85(1), 115.

PAGE 326

308 Varghese, A. J. (1970). "5-thyminyl-5,6-di hydrothymine from DN A irradiated with ultraviolet light." Biochemical and Biophysical Research Communications, 38(3), 484-&. Venkateswaran, K., Chung, S., Allton, J., a nd Kern, R. (2004). "Evaluation of various cleaning methods to remove Bacillus spores from spacecraft hardware materials." Astrobiology, 4(3), 377. Vidal, A., Diaz, A. I., El Hraiki, A., Romero, M., M uguruza, I., Senhaji, F., and Gonzalez, J. (1999). "Solar photocatalysis for detoxification and disinfection of contaminated water: Pilot plant studies." Catalysis Today, 54(2), 283. Vijayaraghavan, S., and Goswami, D. Y. (2002). "On the calibration of a solar UV radiometer to measure broadband UV radiation from blacklight lamps." Journal of Solar Energy Engineering–Transactions of the Asme, 124(3), 317. Vohra, A., Goswami, D. Y., Deshpande, D. A., and Block, S. S. (2005). "Enhanced photocatalytic inactivation of bact erial spores on surfaces in air." Journal of Industrial Microbiol ogy & Biotechnology, 32(8), 364. Vorontsov, A. V., Savinov, E. N., Kurkin, E. N., Torbova, O. D., and Parmon, V. N. (1997). "Kinetic features of the steady st ate photocatalytic co oxidation by air on TiO2." Reaction Kinetics and Catalysis Letters, 62(1), 83. Vorontsov, A. V., Stoyanova, I. V., Kozlov, D. V., Simagina, V. I., and Savinov, E. N. (2000). "Kinetics of the photo catalytic oxidation of gase ous acetone over platinized titanium dioxide." Journal of Catalysis, 189(2), 360. Waites, W. M., Wyatt, L. R., King, N. R., a nd Bayliss, C. E. (1976). "Changes in spores of clostridium-bifermentans caused by treatment with hydr ogen-peroxide and cations." Journal of General Microbiology, 93(Apr), 388. Wakamura, M., Hashimoto, K., and Watana be, T. (2003). "Photocatalysis by calcium hydroxyapatite modified with ti(iv): Al bumin decomposition and bactericidal effect." Langmuir, 19(8), 3428-3431. Wang, C. M., Heller, A., and Gerischer, H. (1992). "Palladium cat alysis of o2 reduction by electrons accumulated on TiO2 partic les during photoassisted oxidation of organic-compounds." Journal of the American Chemical Society, 114(13), 5230– 5234. Wang, K. H., Hsieh, Y. H., Lin, C. H., and Chang, C. Y. (1999). "The study of the photocatalytic degradation kinetics fo r dichloroethylene in vapor phase." Chemosphere, 39(9), 1371. Wang, K. H., Tsai, H. H., and Hsieh, Y. H. (1998). "The kinetics of photocatalytic degradation of trichloroet hylene in gas phase over TiO2 supported on glass bead." Applied Catalysis B-Environmental, 17(4), 313.

PAGE 327

309 Watts, R. J., Kong, S. H., Orr, M. P., Miller, G. C., and Henry, B. E. (1995). "Photocatalytic inactivation of coliform bacteria and vi ruses in secondary wastewater effluent." Water Research, 29(1), 95. Wei, C., Lin, W. Y., Zainal, Z., Williams, N. E., Zhu, K., Kruzic, A. P., Smith, R. L., and Rajeshwar, K. (1994). "Bactericidal activ ity of TiO2 photocatalyst in aqueousmedia toward a solar-assisted water disinfection system." Environmental Science & Technology, 28(5), 934. Wist, J., Sanabria, J., Dierolf, C., Torres, W., and Pulgarin, C. (2002). "Evaluation of photocatalytic disinfection of crude water for drinking-water production." Journal of Photochemistry and P hotobiology a-Chemistry, 147(3), 241. Wolfe, R. L. (1990). "Ultraviolet disinfecti on of potable water current technology and research needs." Environmental Science & Technology, 24(6), 768. Wolfrum, E. J., Huang, J., Blake, D. M., Mane ss, P. C., Huang, Z., Fiest, J., and Jacoby, W. A. (2002). "Photocatalytic oxidation of bacteria, bacterial and fungal spores, and model biofilm components to car bon dioxide on titanium dioxide-coated surfaces." Environmental Science & Technology, 36(15), 3412. Wu, C. H., Chang, H. W., and Chern, J. M. (2006). "Basic dye decomposition kinetics in a photocatalytic slurry reactor." Journal of Hazardous Materials, 137(1), 336. Wu, C. H., and Chern, J. M. (2006). "Kin etics of photocatalytic decomposition of methylene blue." Industrial & Engineering Chemistry Research, 45(19), 6450– 6457. Xue, Y., and Nicholson, W. L. (1996). "The two major spore DNA repair pathways, nucleotide excision repair and spore phot oproduct lyase, are sufficient for the resistance of Bacillus subtilis spores to artificial UV-c and UV-b but not to solar radiation." Appl Environ Microbiol, 62(7), 2221. Yamakata, A., Ishibashi, T., and Onishi, H. (2003). "Kinetics of the photocatalytic watersplitting reaction on TiO2 and pt/TiO2 studied by time-resolved infrared absorption spectroscopy." Journal of Molecular Catalysis a-Chemical, 199(1), 85. Yamazaki, S., Tanaka, S., and Tsukamoto, H. (1999). "Kinetic studies of oxidation of ethylene over a TiO2 photocatalyst." Journal of Photochemi stry and Photobiology a-Chemistry, 121(1), 55. Yatmaz, H. C., Akyol, A., and Bayramoglu, M. (2004). "Kinetics of the photocatalytic decolorization of an azo reactive dye in aqueous zno suspensions." Industrial & Engineering Chemistry Research, 43(19), 6035.

PAGE 328

310 Yokoi, H., Shiragami, T., Hirose, J., Kawauc hi, T., Hinoue, K., Fueda, Y., Nobuhara, K., Akazaki, I., and Yasuda, M. (2003). "Bacter icidal effect of a silica gel-supported porphyrinatoantimony(v) complex unde r visible light irradiation." World Journal of Microbiology & Biotechnology, 19(6), 559. Yoshinari, M., Oda, Y., Kato, T., a nd Okuda, K. (2001). "Influence of surface modifications to titanium on antibacterial activity in vitro." Biomaterials, 22(14), 2043. Yu, J. C., Ho, W. K., Lin, J., Yip, K. Y., and Wong, P. K. (2003). "Photocatalytic activity, antibacterial effect, and photoinduced hydrophili city of TiO2 films coated on a stainless steel substrate." Environmental Science & Technology, 37(10), 2296– 2301. Yu, J. C., Tang, H. Y., Yu, J. G., Chan, H. C., Zhang, L. Z., Xie, Y. D., Wang, H., and Wong, S. P. (2002). "Bacteri cidal and photocatalytic activ ities of TiO2 thin films prepared by sol-gel and reverse micelle methods." Journal of Photochemistry and Photobiology a-Chemistry, 153(1-3), 211. Zeng, C. T., Chen, A. P., Chen, A. H., Liu, W., and Dai, Z. M. (2003). "Photocatalytic disinfection by TiO2 immob ilized on glass springs." Chinese Journal of Catalysis, 24(7), 520. Zhang, X., Young, M. A., Lyandres, O., and Van Duyne, R. P. (2005). "Rapid detection of an anthrax biomarker by surf ace-enhanced raman spectroscopy." J Am Chem Soc, 127(12), 4484. Zhang, Y., Crittenden, J. C., and Hand, D. W. (1994). "The solar photocatalytic decontamination of water." Chemistry & Industry(18), 714. Zheng, H., Maness, P. C., Blake, D. M., Wolfru m, E. J., Smolinski, S. L., and Jacoby, W. A. (2000). "Bactericidal mode of titanium dioxide photocatalysis." Journal of Photochemistry and Photobiology a-Chemistry, 130(2), 163. Zhou, S. H., and Ray, A. K. (2003). "Kinetic studies for photocatal ytic degradation of eosin b on a thin film of titanium dioxide." Industrial & Engineering Chemistry Research, 42(24), 6020. Zs. Cserhalmia, I. V. a., J. Becznerb, B. Czukorb. (2002). "Inactivation of saccharomyces cerevisiae and Bacillus cereus by pulsed electric fields technology." Innovative Food Science & Emerging Technologies, 3, 41. Zuo, Y. J., Xi, H. L., Zhang, J. H., Li, Z. J., and Zhou, F. (2001). "Foundation of kinetics model for TiO2-photocatalyzed degrada tion of organic compounds in suspending system." Chinese Journal of Catalysis, 22(2), 198.

PAGE 329

311 BIOGRAPHICAL SKETCH Jue Zhao was born on November 9, 1976, in Wuxi, China. She earned her Bachelor of Science in water supply and drainage engine ering from Yang Zhou University, China, in 1998. She received he r Master of Science degree in environmental engineering from Southeast University in China in 2002. Jue came to the United States to study for a PhD degree at the University of Florida in the Fall of 2002 and received her PhD degree in Fall of 2006.