measuring the susceptibility and adhesion of microorganisms to light-activated antimicrobial
TRANSCRIPT
Measuringthesusceptibilityand
adhesionofmicroorganismstolight‐
activatedantimicrobialsurfaces
AthesispresentedtoUniversityCollegeLondoninpartialfulfilmentofthe
requirementsforthedegreeofDoctorofPhilosophy
ZoieAlexandraAiken
DivisionofMicrobialDiseasesUCLEastmanDentalInstitute
Supervisedby
DoctorJonathanPratten
DivisionofMicrobialDiseasesUCLEastmanDentalInstitute
ProfessorMichaelWilson
DivisionofMicrobialDiseasesUCLEastmanDentalInstitute
2012
2
Declaration
I Zoie Alexandra Aiken confirm that the work presented in this thesis is my own
WhereinformationhasbeenderivedfromothersourcesIconfirmthatthishasbeen
indicatedinthethesis
3
AbstractThe prevention of healthcare‐associated infections (HCAIs) is a major challenge
currently being faced by hospitals in both the UK and worldwide The hospital
environment acts as a reservoir for nosocomial organisms contributing towards the
transmissionofbacteriaand thus thecolonisationand infection ratesof the patient
populationThereforeitisdesirabletoimplementmeasurestodecreasethemicrobial
load within the hospital environment as a whole and particularly on frequently
touchedsurfacesAntimicrobialcoatingscouldbeappliedtothesesurfacesandused
asanadjuncttootherinfectioncontrolpoliciestoreducetheincidenceofHCAIs
Novelnitrogen‐dopedsulfur‐dopedandsilver‐coatedtitaniumdioxidephotocatalytic
thin films were generated by sol‐gel or chemical vapour deposition The materials
exhibitedantibacterialpropertiesafterexposuretoawhitelightcommonlyusedinUK
hospitalsHoweveritwasdifficulttosynthesisereproduciblethinfilmsusingtheCVD
method of deposition An additional antibacterial material was generated with the
potential tobeused inendotracheal tubesto reducethe incidenceofHCAIssuchas
ventilator‐associated pneumonia The novel polymer was impregnated with a
photosensitiserusingaswellencapsulationmethodandactivatedwithlaserlightthe
antibacterialandanti‐adhesivepropertieswerethenassessed
Sampling the test surfaces by swabbing and subsequently performing viable counts
was shown to provide an adequate estimate of concentration of bacteria on a test
surfaceThenitrogen‐andsulfur‐dopedtitaniumdioxidecoatingsdisplayedsignificant
photocatalyticactivityagainstEscherichia coliafterexposure toawhite light source
4
whichdemonstratedashiftinthebandgapfromtheUVtothevisibleregionofthe
electromagnetic spectrum Visible light photocatalysis was confirmed on the silver‐
coated titania thin films when a UV filter was used to block out the minimal UV
componentofthewhitelightsourceintheformofphoto‐oxidationofstearicacida
reduction in thewater contactangleandphotocatalyticactivityagainstanepidemic
strain of meticillin resistant Staphylococcus aureus (EMRSA‐16) This is the first
example of unambiguous visible light photocatalysis and photo‐induced
superhydrophilicity alongside a titanium dioxide control that shows no activation A
reduction in the viability of EMRSA‐16 adhered onto the surface of the irradiated
silver‐coatedtitaniathinfilmswasalsodemonstrated
AsignificantreductionintherecoveryofPseudomonasaeruginosaStenotrophomonas
maltophilia Acinetobacter baumannii and Candida albicans was observed on TBO‐
impregnated polymers after irradiation with a HeNe laser light A recently isolated
clinicalstrainofPaeruginosashoweddecreasedsusceptibilitytothephoto‐activityof
the TBO‐impregnated polymers compared with a laboratory type strain Finally a
significant reduction in the adhesion of P aeruginosa on the TBO‐impregnated
polymers was demonstrated after a 3‐step irradiation schedule A photo‐bleaching
effect was noted after light exposure that reduced the antibacterial activity of the
polymerswhichdemonstratestherequirementforfurthermodificationtoretainthe
photosensitiserwithinthepolyurethanematrix
These novel materials have the potential to be used as anti‐microbial surfaces in
healthcareenvironments
5
AcknowledgementsIwould liketothankmysupervisorsDrJonathanPrattenandProfessorMikeWilson
fortheirsupportoverthelastfouryearsIthasbeenaneventfuljourneyandIthank
youforalltheknowledgeandwisdomyouhavesharedwithmeThankstoDrCharlie
DunnillandDrGeoffHyett forsynthesisingtheCVDthinfilmsandtoCharlie forthe
assistancewith theoretical concepts especially duringmywriting up period ndash it has
beeninvaluableThankstoProfessorIvanParkinDrKristopherPageandDrStefano
PerniforteachingmaterialschemistrytoamicrobiologistndashitcanrsquothavebeeneasyI
wouldliketoacknowledgetheEngineeringandPhysicalSciencesResearchCouncilfor
financial support Dr Aviva Petrie for providing statistical assistance and Dr Nicky
Mordan forpreparing samples forSEManalysis andhelp inanalysing thegenerated
images
IwouldliketothankthestafffromtheDivisionofMicrobialDiseasesatTheEastman
Dental Institute past and present who made the experience more enjoyable
especially Mike Brouwer (for motivating tea breaks Body Combat Stroopwafels
Bastongne amp beer) Dr Sarah Tubby Linda Dekker Dr Katherine McCurrie Salim
IsmalDrLenaCiricDrRachaelWhealanDrFlorentChangPiDrJohnWrightandDr
GilShalomThankstoDrTomMorganandDrWillKoningforchallengingmyviewson
statisticalanalysis
ThankyoutoKerryWilliamsRebeccaGortonMichelleCairnsandDrCassiePopefor
yourfriendshipsupportloveandscientificadvicendashmy(other)LondonfamilyThanks
to Samantha KaiserHelen Castle CatrionaWright AliceOrsquoSullivan andBeccaOwen
6
for your continued friendship and patience during the tough times Thanks to Emiel
Aiken Dad Mike Nelson and also to the Derbyshire family for your laughter and
continuedsupport
IrsquomindebtedtoDrTimMcHughforhelpingmetobelievethatIcoulddoaPhDandto
Dr Clare Ling and Simon Rattenbury for hiring me as a trainee Clinical Scientist all
those years ago supporting me since and allowing me to pursue a career in
MicrobiologyThankstoDrMathewDiggleandDrKatrinaLeviforyourflexibilityand
understandingduringmywrite‐upperiod
Finally thank you tomyMum for being a constant support inmy life Irsquove enjoyed
sharingmypositiveresultswithyouandyourwordsofencouragementhavekeptme
goingthroughthebadtimesIcouldnothavedonethiswithoutyouThisisdedicated
toyouandtoNannyGrandadandAuntieAnnwhowouldhavelovedtobearoundto
readthis
7
TableofcontentsDeclaration 2
Abstract 3
Acknowledgements 5
Tableofcontents 7
Listoffigures 13
Listoftables 19
1 Introduction 20
11 Healthcare‐associatedinfections 20
111 OrganismscausingHCAIs 22
12 RelevanceoftheenvironmentinHCAIs 26
121 Bacterialsurvivalofdesiccation 31
122 Cleaningfrequencyandstandards 32
123 Levelofsurfacecontamination 34
124 Frequencyofsurfacere‐contaminationpost‐cleaning 36
125 Frequencyofcontactwiththehand‐touchsurface 37
126 Hygienepracticesofstaffpatientsandvisitors 39
13 Antimicrobialcoatings 40
131 Silverasanantimicrobialagent 41
132 Copperasanantimicrobialagent 48
133 Titaniumdioxidephotocatalyticthinfilms 49
14 Relevanceofsurfacesinventilator‐associatedpneumonia 63
141 Photodynamictherapy 66
15 Methodsofproducinglight‐activatedantimicrobialmaterials 70
151 Chemicalvapourdeposition 72
152 Sol‐gel 71
153 Swellencapsulation 72
16 Measuringenvironmentalcontamination 73
161 Swabbing 73
8
162 Dipslides 73
163 Airsampling 74
164 ATPbioluminescence 75
165 Stainingtechniques 77
166 Summaryofenvironmentalsamplingtechniques 78
17 Methods of characterising and assessing the functionality of light‐activatedantimicrobialmaterials 79
171 UV‐visible‐IRspectroscopy 79
172 Photooxidationofstearicacid 79
173 Contactanglemeasurements 81
174 Standardmethodsofassessment 82
18 Overviewandprojectaims 84
2 Materialsandmethods 86
21 Targetorganisms 86
22 Growthconditions 87
23 Preparationofthebacterialinoculum 87
24 Lightsources 87
241 Whitelightsource 87
242 Ultraviolet(UV)lightsources 88
243 Laserlightsource 89
25 Generalsamplingmethodology 89
26 ATPbioluminescence 90
261 Luminometer‐specificmethodologies 91
27 DirectvisualisationofbacteriandashLiveDeadstaining 93
28 Effectofwhitelightonbacterialsurvival 93
29 Optimisationofthesamplingtechnique 94
210 Preparationoflight‐activatedantibacterialmaterials 95
2101 Thinfilmsgeneratedbychemicalvapourdeposition 95
2102 Thinfilmsgeneratedbysol‐geldeposition 99
2103 Toluidine Blue O‐containing polymers generated by swell encapsulation 101
211 Characterisation and functional assessment of light‐activated antibacterial materials 102
9
2111 UV‐visible‐IRspectroscopy 102
2112 Contactanglemeasurements 103
2113 Photooxidationofstearicacid 103
212 Microbiologicalassessmentoflight‐activatedantimicrobialmaterials 105
2121 Decontaminationofthethinfilms 105
2122 Measuringtheeffectof lightonthethinfilmsgeneratedbyAPCVDor sol‐gel 105
2123 Measuring the effect of light on Toluidine Blue O‐impregnated polymersgeneratedbyswellencapsulation 107
213 Statisticalanalysis 108
3 Development of protocols used to assess the activity of thephotocatalyticthinfilms 110
31 Introduction 110
32 Materialsandmethods 112
321 Optimisationofthesamplingtechnique 112
322 ATPbioluminescence 113
323 Measuringtheeffectofwhitelightonbacterialsurvival 114
33 Results 115
331 Optimisationofthesamplingtechnique 115
332 ATPbioluminescence 117
333 Measuringtheeffectofwhitelightonbacterialsurvival 122
34 Discussion 130
341 Optimisationofthesamplingtechnique 130
342 ATPbioluminescence 132
343 Theeffectofwhitelightonbacterialsurvival 135
35 Conclusions 138
4 AssessmentofnovelCVD‐synthesisedlight‐activatedantibacterialmaterialsforuseinthehospitalenvironment 139
41 Introduction 139
42 Materialsandmethods 140
421 Synthesisofthethinfilms 140
10
422 Measuringtheantibacterialeffectofthethinfilms 140
423 Assessmentofthedecontaminationregimen 141
424 Effectofthecoveringmaterialonthinfilmactivity 141
43 Results 142
431 Photocatalyticactivityoftitaniumdioxidethinfilms 142
432 Photocatalytic antibacterial activity of nitrogen‐containing titanium dioxidethinfilmsTiON‐1andTiON‐2 144
433 Photocatalytic antibacterial activity of nitrogen‐doped titanium dioxidethinfilmsN1N2andN3 149
434 EffectofchangingthedecontaminationregimenonthinfilmN1 153
435 Effectofcoveringmaterialonthinfilmactivity 154
436 Photocatalytic antibacterial activity of sulfur‐based titanium dioxide thinfilms 157
44 Discussion 161
441 UVlight‐inducedphotocatalyticactivity 161
442 Whitelight‐inducedphotocatalyticactivity 162
443 Limitationsoftheexperimentalwork 166
45 Conclusions 168
5 Assessment of novel sol‐gel synthesised light‐activatedantibacterialmaterialsforuseinthehospitalenvironment 170
51 Introduction 171
52 Materialsandmethods 171
521 Thinfilmsynthesis 171
522 Characterisationandfunctionalassessmentofthethinfilms 171
523 Antibacterialassessmentofthethinfilms 172
53 Results 173
531 Characterisationandfunctionalassessmentofthethinfilms 175
532 AntibacterialactivityagainstEcoliATCC25922 184
533 AntibacterialactivityagainstEMRSA‐16 189
54 Discussion 195
541 Synthesisofthesilver‐dopedtitaniathinfilms 196
542 Characterisation and functional assessment of the silver‐doped titania thinfilms 197
11
543 Antibacterialactivityofthesilver‐dopedtitaniathinfilms 200
55 Conclusion 203
6 Assessment of a novel antibacterial material for use inendotrachealtubesinintubatedpatients 204
61 Introduction 204
62 Materialsandmethods 206
621 Materialsynthesis 206
622 Measuring the antibacterial photo‐activity of the TBO‐impregnated polymers 206
63 Results 207
631 Assessmentoftheantibacterialphoto‐activityoftheTBO‐ impregnated polymersagainstPaeruginosaPAO1atypestrain 207
632 Assessmentoftheantibacterialphoto‐activityoftheTBO‐ impregnated polymersagainstaclinicalstrainofPaeruginosa 213
633 Assessmentoftheantibacterialphoto‐activityoftheTBO‐ impregnated polymersagainstaclinicalstrainofAbaumannii 217
634 Assessmentoftheantibacterialphoto‐activityoftheTBO‐ impregnated polymersagainstaclinicalstrainofSmaltophilia 220
635 Assessmentoftheantibacterialphoto‐activityoftheTBO‐ impregnated polymersagainstaclinicalstrainofCalbicans 223
64 Discussion 226
641 TBO‐mediatedphotodynamicbacterialinactivation 226
642 Limitationsoftheexperimentalwork 230
643 Novelmaterials for potential use as antimicrobial endotracheal tubes 232
65 Conclusions 234
7 Assessment of the disruptive and anti‐adhesive properties ofnovellight‐activatedmaterials 235
71 Introduction 235
72 Materialsandmethods 236
721 Silver‐dopedtitaniumdioxidethinfilms 236
722 TBO‐impregnatedpolymers 240
73 Results 243
731 Silver‐dopedtitaniumdioxidethinfilms 243
12
732 TBO‐impregnatedpolymers 251
74 Discussion 256
741 AssessmentofinitialattachmentofEMRSA‐16 256
742 DisruptionofanimmaturebiofilmofEMRSA‐16 258
743 PreventionofinitialPaeruginosaPAO1attachment 260
744 Limitationsoftheexperimentalwork 262
75 Conclusions 263
8 Concludingremarksandfuturework 265
9 Publicationsarisingfromthiswork 270
91 Peer‐reviewedPublications 270
92 Posterpresentations 271
93 Otherpublications 271
10 References 272
13
ListoffiguresFigure11TheWHOFiveMomentsforHandHygiene 27
Figure12Transmissionroutesofpathogenswithinahospitalenvironment 28
Figure13Schematicofaconductionbandinaconductor 49
Figure14Freemovementofelectronswithinaconductor 50
Figure15Schematicofaconductionbandinaninsulator 50
Figure16Schematicdisplayingthebandgapwithinasolidstatematerial 51
Figure 17 Promotion of an electron from the valence band (VB) to the conductionband(CB)inasemiconductorafterlightabsorption 52
Figure18n‐typesemiconductors 53
Figure19p‐typesemiconductors 53
Figure110Electronicexcitationofasemiconductormolecule 55
Figure111Generationofsingletoxygen 68
Figure112SchematicrepresentationofaCVDapparatus 71
Figure113Chemicalstructureofstearicacid 80
Figure21Spectralpowerdistributiongraphforthewhitelightsource 88
Figure22Experimentalsetupofthemoisturechamber 94
Figure23Thesol‐geldippingapparatus 100
Figure24Whitelightirradiationofnitrogen‐dopedthinfilms 106
Figure31ComparisonofdifferentswabtypestoincreasetherecoveryofEcoliandEfaecalis 115
Figure32ComparisonofdifferentsamplingmethodsusedtoincreasetherecoveryofEcoli 116
Figure33ComparisonofSaureusdetectionmethods 118
Figure34ComparisonofEcolidetectionmethods 120
Figure35EffectofthewhitelightsourceonthesurvivalofSaureusNCTC6571 123
Figure36EffectofthewhitelightsourceonthesurvivalofEcoliATCC25922 124
14
Figure37EffectofthewhitelightsourceonthesurvivalofEfaecalis 125
Figure38EffectofthewhitelightsourceonthesurvivalofSpyogenesATCC12202 126
Figure39EffectofthewhitelightsourceonthesurvivalofEMRSA‐16 127
Figure310EffectofthewhitelightsourceonthesurvivalofEMRSA‐15 128
Figure311EffectofthewhitelightsourceonthesurvivalofMRSA43300 128
Figure312Effectofthewhite lightsourceonthesurvivalofSaureusNCTC8325‐4 129
Figure41Photo‐activityoftheTiO2thinfilms 142
Figure42PhotocatalyticactivityofPilkingtonActivTMonEcoli 143
Figure43EffectofthethinfilmTiON‐2againstEcoliafterexposureto1hour254nmlightand4hours365nmlight 145
Figure44EffectofthethinfilmTiON‐1againstEcoliafterexposureto1hour254nmlightand4hours365nmlight 146
Figure 45 Effect of the thin film TiON‐2 on the survival of E coli Thin films wereexposedtowhitelightfor24hoursthebacterialdropletwasaddedthenthesamplewasexposedasecondlightexposureperiodofeither618or24hours 148
Figure 46 Effect of the thin film TiON‐1 on the survival of E coli Thin films wereexposedtowhitelightfor24hoursthebacterialdropletwasaddedthenthesamplewasexposedasecondlightexposureperiodofeither618or24hours 149
Figure47EffectofthethinfilmN1onthesurvivalofEcoliThinfilmswereexposedto white light for 24 hours the bacterial droplet was added then the sample wasexposedasecondlightexposureperiodof24hours 150
Figure48EffectofthethinfilmN2onthesurvivalofEcoliThinfilmswereexposedto white light for 24 hours the bacterial droplet was added then the sample wasexposedasecondlightexposureperiodof24hours 152
Figure49EffectofthethinfilmN3onthesurvivalofEcoliThinfilmswereexposedto white light for 24 hours the bacterial droplet was added then the sample wasexposedasecondlightexposureperiodof24hours 153
Figure 410 Light‐activated antimicrobial killing of E coli on thin film N1 and afterinactivation 154
Figure411ConcentrationofEcoliremainingonthethinfilmTiON‐2usingaclingfilmcovering 155
15
Figure 412 UV‐visible light transmission trace of the petri dish lid and the clingfilmcovers 157
Figure413EffectofthethinfilmS2onthesurvivalofEcoliThinfilmswereexposedto white light for 72 hours the bacterial droplet was added then the sample wasexposedasecondlightexposureperiodof24hours 158
Figure414EffectofthethinfilmS1onthesurvivalofEcoliThinfilmswereexposedto white light for 72 hours the bacterial droplet was added then the sample wasexposedasecondlightexposureperiodof24hours 160
Figure415EffectofthethinfilmS3onthesurvivalofEcoliThinfilmswereexposedto white light for 72 hours the bacterial droplet was added then the sample wasexposedasecondlightexposureperiodof24hours 160
Figure51PhotographoftheAg‐TiO2thinfilms 174
Figure52TransmissiondataoftheAg‐TiO2andTiO2thinfilmsdepositedontoaquartzsubstrateobtainedbyUV‐visible‐IRspectrometry 176
Figure53 Tauc plotsof theUV‐visible‐IRdata taken for the (a)Ag‐TiO2and (b) TiO2thinfilmspreparedonquartzsubstrates 177
Figure54UV‐VisspectrumfortheOptivextradeUVfiltershowingthecut‐offforradiationbelow400nminwavelength 179
Figure55IRabsorptiondatashowingthephoto‐oxidationofstearicacidmoleculesonthesurfaceofthethreematerialsover72hoursusinga254nmlightsource 181
Figure56IRabsorptiondatashowingthephoto‐oxidationofstearicacidmoleculesonthesurfaceofthethreematerialsover96hoursusingawhitelightsource 182
Figure 57 Raw data showing the photo‐oxidation of stearic acid molecules on thesurface of the three samples over 500 hours using a white light source and theOptivextradeUVfilter 183
Figure 58 Effect of the thin film Ag‐TiO2 on the survival of E coli Thin films wereirradiatedwithwhitelightorincubatedinthedarkfor2hours 185
Figure 59 Effect of the thin film Ag‐TiO2 on the survival of E coli Thin films wereirradiatedwithwhitelightorincubatedinthedarkfor6hours 185
Figure 510 Effect of the thin filmAg‐TiO2 on the survival ofE coli Thin filmswereirradiatedwithwhitelightorincubatedinthedarkfor12hours 187
Figure 511 Effect of the thin filmAg‐TiO2 on the survival ofE coli Thin filmswereirradiatedwithwhitelightfilteredwiththeOptivextradeglassorincubatedinthedarkfor12hours 187
16
Figure 512 Effect of the thin filmAg‐TiO2 on the survival ofE coli Thin filmswereirradiatedwithwhitelightorincubatedinthedarkfor18hours 189
Figure513EffectofthethinfilmAg‐TiO2onthesurvivalofEMRSA‐16Thinfilmswereirradiatedwithwhitelightorincubatedinthedarkfor6hours 190
Figure514EffectofthethinfilmAg‐TiO2onthesurvivalofEMRSA‐16Thinfilmswereirradiatedwithwhitelightorincubatedinthedarkfor12hours 191
Figure515EffectofthethinfilmAg‐TiO2onthesurvivalofEMRSA‐16ThinfilmswereirradiatedwithwhitelightfilteredwiththeOptivextradeglassorincubatedinthedarkfor12hours 192
Figure516EffectofthethinfilmAg‐TiO2onthesurvivalofEMRSA‐16Thinfilmswereirradiatedwithwhitelightorincubatedinthedarkfor18hours 193
Figure517EffectofthethinfilmAg‐TiO2onthesurvivalofEMRSA‐16ThinfilmswereirradiatedwithwhitelightfilteredwiththeOptivextradeglassorincubatedinthedarkfor18hours 194
Figure 61 A catheter tube impregnated with the photosensitising agent methyleneblue 205
Figure 62 Antibacterial activity of TBO‐impregnated polymer against P aeruginosaPAO1after30seconds 208
Figure 63 Antibacterial activity of TBO‐impregnated polymer against P aeruginosaPAO1after60seconds 208
Figure 64 Antibacterial activity of TBO‐impregnated polymer against P aeruginosaPAO1after90seconds 209
Figure 65 Antibacterial activity of TBO‐impregnated polymer against P aeruginosaPAO1after120seconds 209
Figure 66 Antibacterial activity of TBO‐impregnated polymer against P aeruginosaPAO1after150seconds 210
Figure 67 Antibacterial activity of TBO‐impregnated polymer against P aeruginosaPAO1after180seconds 210
Figure 68 Antibacterial activity of TBO‐impregnated polymer against P aeruginosaPAO1after210seconds 211
Figure 69 Antibacterial activity of TBO‐impregnated polymer against P aeruginosaPAO1after240seconds 211
Figure610AntibacterialactivityofTBO‐impregnatedpolymeragainstaclinicalstrainofPaeruginosaafter90seconds 214
17
Figure611AntibacterialactivityofTBO‐impregnatedpolymeragainstaclinicalstrainofPaeruginosaafter180seconds 214
Figure612AntibacterialactivityofTBO‐impregnatedpolymeragainstaclinicalstrainofPaeruginosaafter240seconds 215
Figure613AntibacterialactivityofTBO‐impregnatedpolymeragainstaclinicalstrainofAbaumanniiafter90seconds 218
Figure614AntibacterialactivityofTBO‐impregnatedpolymeragainstaclinicalstrainofAbaumanniiafter180seconds 218
Figure615AntibacterialactivityofTBO‐impregnatedpolymeragainstaclinicalstrainofAbaumanniiafter240seconds 219
Figure616AntibacterialactivityofTBO‐impregnatedpolymeragainstaclinicalstrainofSmaltophiliaafter90seconds 221
Figure617AntibacterialactivityofTBO‐impregnatedpolymeragainstaclinicalstrainofSmaltophiliaafter180seconds 221
Figure618AntibacterialactivityofTBO‐impregnatedpolymeragainstaclinicalstrainofSmaltophiliaafter240seconds 222
Figure619AntimicrobialactivityofTBO‐impregnatedpolymeragainstaclinicalstrainofCalbicansafter90seconds 223
Figure620AntimicrobialactivityofTBO‐impregnatedpolymeragainstaclinicalstrainofCalbicansafter180seconds 224
Figure621AntimicrobialactivityofTBO‐impregnatedpolymeragainstaclinicalstrainofCalbicansafter240seconds 224
Figure71Theflowcellchamberusedtoassessbacterialattachment 237
Figure72Microtitreplatelayoutforthebiofilmdisruptionassays 241
Figure73AttachmentofEMRSA‐16toeitheran(a)uncoatedslideor(b)Ag‐TiO2thinfilmafter0hexposuretothewhitelightsource 244
Figure74AttachmentofEMRSA‐16toeitheran(a)uncoatedslideor(b)Ag‐TiO2thinfilmafter6hexposuretothewhitelightsource 244
Figure75AttachmentofEMRSA‐16toeitheran(a)uncoatedslideor(b)Ag‐TiO2thinfilmafter18hexposuretothewhitelightsource 244
Figure76ConfocalmicrographofEMRSA‐16inPBSontheAg‐TiO2thinfilmafter24hoursgrowthat37degCinthedarkand24hoursexposuretowhitelightat22degC 246
18
Figure77ConfocalmicrographofEMRSA‐16inPBSontheAg‐TiO2thinfilmafter24hoursgrowthat37degCinthedarkand24hoursincubationat22degCinthedark 247
Figure78ConfocalmicrographofEMRSA‐16 inBHIontheAg‐TiO2thinfilmafter24hoursgrowthat37degCinthedarkand24hoursexposuretowhitelightat22degC 249
Figure79ConfocalmicrographofEMRSA‐16 inBHIontheAg‐TiO2thinfilmafter24hoursgrowthat37degCinthedarkand24hoursincubationat22degCinthedark 250
Figure710AbilityoftheTBO‐impregnatedpolymerstopreventtheinitialattachmentofPaeruginosaPAO1 252
Figure711 SEM imageofPaeruginosaPAO1on the surfaceofaTBO‐impregnatedpolymerafter3hoursirradiationwiththelaserlight 253
Figure712 SEM imageofPaeruginosaPAO1on the surfaceofaTBO‐impregnatedpolymerafter3hoursincubationintheabsenceoflaserlight 254
Figure 713 Effect of photo‐bleaching on the anti‐P aeruginosa activity of the TBO‐impregnatedpolymers 256
19
Listoftables
Table21Bacterialandfungalstrainsusedinthesestudies 86
Table22Nomenclatureusedduringmicrobiologicalassessmentofthethinfilms107
Table 31 Definitions of the terms used to compare the luminometer‐specificmethodologies 110
Table32ReproducibilityoftheATPbioluminescenceassay‐Saureus 118
Table33ReproducibilityoftheATPbioluminescenceassay‐Ecoli 121
Table34Effectofwhitelightonbacterialsurvival 130
Table41Summaryofthephotocatalyticactivityofthenitrogenandsulfurdopedthinfilms 161
Table51WatercontactanglesoftheAg‐TiO2thinfilmsandthecontrolsamples 178
Table 52 Photo‐oxidisation of stearic acid during irradiation by the different lightsources 184
Table 61 Nomenclature used during microbiological assessment of the TBO‐impregnatedpolymers 207
Table62SummaryofPaeruginosaPAO1experiments 212
Table63ComparisonofthetwoPaeruginosaexperiments 217
Table64TBO‐impregnatedpolymers‐Summaryofresults 226
Table71Confocalscanninglasermicroscope‐samplesdescriptions 240
Table72Resultsofthebacterialattachmentassays 251
20
1 Introduction
11 Healthcare‐associatedinfections
Healthcare‐associated infections (HCAIs)aredefinedbytheDepartmentofHealthas
ldquoany infection by any infectious agent acquired as a consequence of a personrsquos
treatmentinhealthcarerdquo(DepartmentofHealth2008)andtheyareamongthemost
commonadverseevents inhospitalisedpatients (Leapeetal 1991)Organisms that
cause HCAIs are able to cause disease in the susceptible host and survive in the
hospital environment for long periods of time (Dancer 2011) The prevention and
control of HCAIs within healthcare institutions both in the UK and worldwide is a
majorpriorityandtherecentlyreviseddocumentfromtheDepartmentofHealthlsquoThe
Health Act 2006 Code of Practice for the Prevention and Control of Healthcare‐
AssociatedInfectionsrsquodetailsstandardsrequiredtoachievetheseaims(Departmentof
Health 2008) Mandatory surveillance of certain infections such as orthopaedic
surgical site infections and those caused by specific bacteria such as meticillin‐
resistantStaphylococcusaureus(MRSA)andClostridiumdifficilehavebeenintroduced
becauseofthemorbidityandmortalityassociatedwiththoseinfections(Reportbythe
Comptroller and Auditor General ‐ HC Session 2003‐2004) Surveillance data are
updatedfortnightlyandareavailableatwwwdatagovukThemandatorysurveillance
schemewasextendedinJune2011to includeratesofEscherichiacoliandmeticillin‐
sensitive S aureus bacteraemia (Health Protection Agency 2011a) Government
targetsarealsoinplacetoreducetheincidenceofinfectionscausedbySaureusand
CdifficileBothoftheseorganismscanresideinharmonywithinhealthyhumanhosts
but cause serious problemswhen growth is uncontrolled or permitted outside their
usualniches
21
Approximately 17 million HCAIs are acquired annually in the American healthcare
environment resulting in nearly 99000 deaths a year greater than the number of
casesofanynotifiablediseasewithanassociatedcostperpatientofbetween$16359
and $19430 (Scott II 2009)When this figure is scaled up it amounts to a cost of
between$284to338billiondollarsperannum(Klevensetal2007ScottII2009)In
responsetotherisingcostof in‐patientcaretheCentersforMedicareandMedicaid
Services which provide health insurance for certain sections of the American
populationhavediscontinuedpaymenttohospitalsifthepatientisafflictedbyoneof
eight lsquopreventable complicationsrsquo during their stay (Rosenthal 2007) The HCAIs
included in this list are catheter associated urinary tract infections and vascular
catheter‐associated infections An estimated 13000 deaths were caused by urinary
tractinfectionalonein2002(Klevensetal2007)
InEnglandapproximately1 in10patientshaveanHCAIatanyonetimeaccounting
for100000casesand5000deathsperannum(ReportbytheComptrollerandAuditor
General‐HC230Session1999‐2000ReportbytheComptrollerandAuditorGeneral‐
HCSession2003‐2004)PatientsthatacquireanHCAIarerequiredtostayinhospital
for an average of eleven additional days and incur treatment costs of nearly three
timesthatofanuninfectedpatienttheyarealsoseventimesmorelikelytodiethan
patientsthatdidnotacquireanHCAI(Plowmanetal2000ReportbytheComptroller
and Auditor General ‐ HC Session 2003‐2004 2004) The financial cost of HCAIs in
Englandhasbeencalculatedtobeapproximatelypound1billionperannumandupto30
oftheseinfectionscanbeprevented(Plowmanetal2000)Introducingpreventative
measurescostslessthantreatingtheinfectionitselfsointensiveeffortsareinplaceto
22
reduce infection rates (Report by the Comptroller andAuditorGeneral ‐ HC Session
2003‐20042004)
111 OrganismscausingHCAIs
1111 Meticillin‐resistantSaureus(MRSA)
S aureus is found in the anterior nares of 20 of the population (Report by the
Comptroller andAuditorGeneral ‐ HC Session 2003‐2004 2004 Alekshun and Levy
2006)butcausesinfectioninwoundswhichcanleadtoosteomyelitisifitreachesthe
boneabscessesif itpenetratesdeepintothetissuesbacteraemiaandsepticaemiaif
itgetsintothebloodstreamandfromthispointitcouldseedintoanyorganandcause
disseminateddiseaseMeticillin‐resistantSaureus(MRSA)isresistanttotheβ‐lactam
group of antibiotics which was the first line therapy before the widespread
development of resistance This resistance decreases the number of available
treatmentoptionsrequiringtheuseofantibioticswithgreatersideeffectswhichcan
prolongthedurationoftreatmentandthetimespentinhospital
MRSA ismost commonly transmittedbetweenpatientsvia contaminatedhandsbut
thepersistenceoftheorganismintheenvironmentalsoprovidesanimportantsource
AdditionallythepresenceofMRSAinthenasalpassagesofcolonisedpatientsenables
spreadviarespiratorydropletnucleiForthesereasonsthenearpatientenvironment
is often contaminatedwith bacteria and themost likely sources ofMRSAmeticillin‐
sensitive (MSSA) contamination in colonised patients are the floor and bedframe
followedbythepatientlockerandtheoverbedtable(Mulveyetal2011)
23
1112 Glycopeptide‐resistantenterococci
Glycopeptide‐resistant enterococci (GRE) predominantly cause infections of the
bloodstreamabdomenpelvisoropenwoundsinimmunocompromisedpatientsThis
patient group is likely to have had previous antibiotic treatment and a prolonged
hospital stay due to significant co‐morbidities such as liver or renal disease
haematologicalmalignanciesordiabetes(Hanetal2009)usuallyinaspecialistward
such as intensive care or a renal unit (Health Protection Agency 2011b) GRE are
resistant to the glycopeptide group of antibiotics which includes vancomycin and
teicoplaninInfectionsareusuallyeithernosocomialorduetoendogenousinoculation
andaredifficulttotreatduetothelackoftreatmentoptionsandthevulnerabilityof
theaffectedpatient
The first reportsofglycopeptide resistantenterococciweredocumented in themid‐
1980s(Uttleyetal1988)andtherehasbeenasignificantincreaseintheincidenceof
bothGREcolonisationand infectionsincebetween1989and1995theproportionof
glycopeptide‐resistant strains of enterococci isolated in the United States rose from
03to104(Gaynesetal1996)TheemergenceofGREcoincidedwithanincrease
in the use of vancomycin (Ena et al 1993) and it is possible that sub‐inhibitory
concentrationsofvancomycinweregeneratedinthetissuesofthesepatientssothat
vancomycin‐resistance was selected alongside an overgrowth of the resistant
Enterococcusfaecalis(Uttleyetal1988)Arecent10‐yearstudycalculatedthe60‐day
mortalityofpatientswithGREbacteraemiaat57andasstandardempiricaltherapy
oftendoesnot includecover forGREsuitableantimicrobial therapy isoftendelayed
whichfurtherincreasesmortality(Hanetal2009)
24
GRE have increased tolerance to environmental conditions and therefore have an
improved survival rate compared withMRSA However transmission of GRE is less
frequent because the colonisation site is usually the gastrointestinal tract whereas
MRSAcommonlycolonisesthenasalpassagesallowingfortransmissionviarespiratory
droplets (Dancer 2002) Unwashed hands remain an important fomite in the
transmissionofGRE
1113 Cdifficile
C difficile can be found in small numbers in the large intestines of some healthy
humansHoweverwhenthenormalmicrobiotaofthegut iscompromisedeitherby
theuseofbroadspectrumantibioticssuchascephalosporinsduetoco‐morbiditiesor
oldagethecolonisationresistanceeffectofthegutisdepletedwhichallowsCdifficile
to proliferate (Wilcox 1996) The clinical presentation ranges from asymptomatic
carriage through to profuse diarrhoea and in serious cases toxic megacolon and
pseudomembranous colitis which carries a significant mortality rate (Alekshun and
Levy2006)Cdifficile produces toxinsduringgrowthwhichdamage the integrityof
thecolonandthisdamagecontributestotheclinicalsymptomsCdifficile iscapable
of entering a dormant phase during which the bacterial cells sporulate and these
spores have increased resistance to harsh environmental conditions such as
desiccationextremesintemperatureanddisinfectantsSporesareoftenfoundinhigh
numbers in the areas surrounding C difficile positive patients (Dancer 1999) and
elimination of this environmental source has been cited as a contributing factor in
haltingtheonwardtransmissionofinfection(Samoreetal1996)
25
1114 Organismscausingventilator‐associatedpneumonia
Ventilator‐associatedpneumonia(VAP)isanosocomialbacterialinfectionofthelungs
withamultifactorialetiologyAnendotrachealtube(ETT)isplacedalongthetrachea
andisconnectedtoaventilatortoallowmechanicallyassistedbreathingThephysical
presenceofthetube interfereswiththenormalclearingofsecretionssuchasmucus
from the upper airways and allows micro‐aspiration of contaminated subglottic
secretionsintothelungsThesesecretionsarecontaminatedwithcommensalbacteria
which provide a source for a pulmonary infection The lumen of the ETT itself can
become colonised with bacteria providing an additional source of infection The
organisms most commonly implicated are S aureus Pseudomonas aeruginosa
Acinetobacter species and Stenotrophomonas maltophilia (Johanson et al 1972
Weberetal2007Bouadmaetal2010)theseorganismsarenotusualcommensals
of the upper respiratory tract but the normal flora of hospitalised patients tends to
containagreaterproportionofGram‐negativebacilliwhicharealso likelytodisplay
multidrugresistancephenotypesVAPisthemostcommonHCAIintheintensivecare
unitaccountingfor30‐50ofinfectionsandisassociatedwithincreaseddurationof
intubationand increased lengthofhospital stay (Kollefetal 2008Bouadmaetal
2010)
TheestimatednumberofinfectionscausedbyVAPintheUnitedStatesis52543with
anattributablecostofbetween$14806and$27520perpatient(Klevensetal2007)
Whenallnosocomialpneumoniaswereconsideredtherewerenearly36000deaths
intheUnitedStatesandofthepatientsthatsurvivedtheextra lengthofstay inthe
hospitalwas9days(Wenzel1995)
26
12 RelevanceoftheenvironmentinHCAIs
Dr Ignac Semmelweis dubbed the lsquoFather of Infection Controlrsquo first described the
importanceofcleanhandsinthepreventionofinfectionin1861(Semmelweis1861)
Henoticedanincreasedrateofpuerperalfeverinalabourwardattendedexclusively
by clinicians compared toaneighbouringwardattendedexclusivelybynursing staff
Thecliniciansperformedautopsiesoncadaversbeforeattendingtoparturientpatients
butdidnotwashtheirhandsaftertheinvestigationsthusallowingthetransferofthe
lsquocadavericparticlesrsquotothewomeninlabourSemmelweisproposedthatallexaminers
should wash their hands in a solution of chlorinated lime to destroy the cadaveric
materialadheringtothehandsByintroducingthismeasurehereducedtheratesof
childhoodmortalityfrom114in1846to18in1848(Semmelweis1861)
MorerecentlytheNHSNationalPatientSafetyAgencylaunchedthelsquocleanyourhandsrsquo
campaignwiththeaimtoimprovethehandhygieneofhealthcareworkersinorderto
reduce the incidence of HCAIs (NHS National Patient Safety Agency 2004) Hand
hygiene plays an essential role in preventing the transmission of microorganisms
(CasewellandPhillips1977Haydenetal2006Dancer2010)anditisrecommended
both in the scientific literature and by the World Health Organisation that hands
should be decontaminated before and after touching a patient before any aseptic
procedureandafterexposuretobodyfluidsasdetailedinFigure11
27
Figure 11 The World Health Organisation Five Moments for Hand Hygienerecommend hand decontamination after touching the near patient environment(Pittetetal2009)
The guidelines also recommend that hands should be decontaminated after contact
withtheenvironmentsurroundingapatientasevidenceshowsthatsitesclosetothe
patientcanbeheavilycontaminatedwithbacteriaorbacterialspores(Samoreetal
1996WeberandRutala1997Devineetal2001BoyceandPittet2002Oieetal
2007 Dancer et al 2008 Pittet et al 2009) The role of the environment in the
transmission of HCAIs has been demonstrated in the scientific literature and is
illustratedinFigure12
28
Figure 12 Transmission routes of pathogens within a hospital environment Boldarrows indicate potential routes of pathogen transfer and red crosses denote adisruptionintransmission
Two independent routes have been described (Talon 1999 Boyce and Pittet 2002
Boyce2007Dancer2008)
1 A healthcare worker (HCW) contaminates their hands by touching the
environmentthentouchesapatientleadingtomicrobialtransferor
2 Asusceptiblepatienttouchesacontaminatedsurfaceandthemicroorganisms
aretransferreddirectlyfromtheenvironmenttothesamepatient
Surfaces that are frequently touched by people in the hospital environment are
termedlsquohand‐touchsurfacesrsquoandthosethathavebeenstudiedinthemostdetailto
determine levels of microbial contamination include the bed‐frame bedside tables
doorhandlestoiletrailsandtoiletseats(Dancer2004Dentonetal2004Boyceet
29
al2008Danceretal2008Huslageetal2010)Hand‐touchsurfacesinthehospital
environment are being increasingly implicated in the transmission of nosocomial
pathogenspatientcolonisationbytheseorganismsandoutbreaksofHCAIs(Boyceet
al1994WeberandRutala1997Bartleyetal2001DepartmentofHealth2001
Ramplingetal2001Frenchetal2004Johnstonetal2006Dancer2010Dancer
and Carling 2010) In reality adherence to hand washing practices has remained
substandard but even exemplary hand hygiene cannot stop transmission if the
environment has a high bacterial load (Dharan et al 1999 Boyce and Pittet 2002
Dancer20042010Erasmusetal2010)
The risk of acquiring MRSA GRE or C difficile has been demonstrated to be
significantlyhigherinpatientsadmittedtoaroomwhosepreviousoccupanthadbeen
MRSAGREorCdifficilepositive(McFarlandetal1989Huangetal2006Dancer
2009CarlingandBartley2010Shaughnessyetal2011)Dreesetal(2008)showed
patientswhoacquiredGREduringtheirhospitalstayweremorelikelytobeinaroom
inwhichaGRE‐positivepatienthadpreviouslyoccupiedandGREwas isolated from
the near‐patient environment in 25 of cases Bacteria are frequently found to
contaminate hand‐touch surfaces even after cleaning and organisms commonly
foundincludeMRSAGREandothercausesofHCAIssuchasMSSAandAcinetobacter
baumannii (Dentonetal 2004 Lewisetal 2008Boyceetal 2009Mulveyetal
2011)
Theenvironmenthasalsobeenshowntoplaya role inthetransmissionof infection
outsideahospitalsettingAnAmericanstudyshowedanincreasedrateofdiarrhoeal
diseaseinchildrenattendingdaycarecentreswheretheenvironmentwasfoundtobe
30
contaminatedwithfaecalcoliforms(Labordeetal1993)Theenvironmentalsources
implicatedweremoistsitessuchassinksandtapsandatwo‐foldincreaseintherate
of diarrhoea was found in children attending these facitilites compared to centres
withanuncontaminatedenvironmentInaseparatestudyofhouseholdcasesofinfant
salmonellosistheserotypeofSalmonellaexcretedbytheinfectedindividualwasalso
isolated from the environment (van Schothorst et al 1978) Chopping boards have
beencommonly implicated inthespreadofgastroenteritisForexample inadequate
cleaning of a chopping board contaminated with juices from raw turkeys led to an
outbreak of gastroenteritis when the chopping board was later used to prepare
sandwiches Additionally an individual investigating the outbreak also developed
symptomsaftertouchingthechoppingboardbeforesmoking(Sanborn1963)
31
The riskofacquiringan infection fromacontaminatedenvironment ismultifactorial
anddifficulttodirectlyassess(Boyce2007Lewisetal2008)Howeveritislikelyto
belinkedto
bull theabilityoftheorganismtosurvivedesiccation
bull thefrequencyandlevelofcleaning
bull thelevelofsurfacecontamination
bull thefrequencyofrecontaminationaftercleaning
bull the frequencyof contactwith thehand‐touch surfacebyhealthcareworkers
patientsandvisitors
bull thehygienepracticesofthehealthcareworkerspatientsandvisitors
121 Bacterialsurvivalofdesiccation
Somebacterialstrainsaremoreresilienttodesiccationbecauseoftheecologicalniche
theyoccupyForexamplestaphylococcalspeciesarewelladaptedforsurvivalonthe
arid environment of the skin and on environmental surfaces which is likely to be
linkedtomatricand ionicstressresistance(ChaibenjawongandFoster2011)MRSA
has been shown to survive for over 2months on a cotton‐blanket (Duckworth and
Jordens 1990) GRE has been shown to survive for up to 4 months on a polyvinyl
chloride surface (PVC) (Wendt et al 1998) andA baumanniiwas recovered froma
patients room 6months after discharge (Zanetti et al 2007) ConverselyNeisseria
gonorrhoeaethrivesinthemoisture‐richenvironmentofthegenitalandbuccaltracts
but is not so well adapted for survival on the predominantly dry hospital surfaces
32
(Griffith et al 2000) Furthermore some epidemic strains of MRSA (EMRSA) have
beenshowntohaveanincreasedsurvivalrateandcansurviveintheenvironmentat
higherconcentrationsthansporadicstrains(Farringtonetal1992Wagenvoortetal
2000)Thisprovidesaselectiveadvantageandcontributestowardsitspersistenceand
endemicityinthehospitalenvironment(Talon1999)Cdifficilesporescansurvivein
the environment formany years and spores are resistant to hand decontamination
products such as alcohol hand gels which further contributes to the persistence of
theseorganismsintheenvironment(BAPS1994)
122 Cleaningfrequencyandstandards
Thepurposeofcleaningistwofoldthemicrobiologicalpurposeistoreduceboththe
microbial load and any nutrientswhich support bacterial growth or substances that
inhibittheactivityofdisinfectantsthenon‐microbiologicalpurposeisaestheticandis
torestoretheappearanceofthematerialandpreventdeterioration(Collins1988)As
thoroughcleaningcanreducethemicrobialloadthenitcanassistinbreakingthecycle
of transmissionof infectionwithin thehospitalenvironment (Dancer2002 Lewiset
al 2008) Indeed regular disinfection of surfaces has been shown to reduce the
transmission of hospital pathogens by 40 and enhanced cleaning of the patient
environment reduces acquisition of bacteria known to cause HCAIs (Hayden et al
2006 Boyce 2007 Carling and Bartley 2010) Despite this the frequency and
standard of cleaning has decreased in recent years due to out‐sourcing of contracts
andlimitationsoncleaningbudgets(Dancer1999Carlingetal2008Dancer2008)
33
Cleaningwithadetergentsolutionisusuallysufficientbuttheuseofdetergentalone
hasbeenshowntoleadtoanincreaseinbacterialcontaminationofhospitalsurfaces
(Dharanetal1999Dancer2011)Asporicidalagentsuchasachlorinecontaining
formulationisrequiredwhentheenvironmentiscontaminatedwithCdifficile(Weber
andRutala2011)
UsingATPtoassessthecleaningprocessisaneffectivetoolasthetotalorganicsoiling
ofasurfacecanbedetermined(HawronskyjandHolah1997)Asurfacecouldbefree
from microbial contamination but could still contain a high level of organic soil
originating from food residues which would provide nutrients to support microbial
growth(Whiteheadetal2008)Deadbacteriaandviablebutnon‐cultivable(VBNC)
organismscanalsobedetectedusingATPbioluminescenceandwouldbemissedby
traditional culturing methods (Poulis et al 1993) ATP bioluminescence has been
shown to be a good indicator of the cleanliness of a surface and of likely bacterial
contamination(Griffithetal2000Maliketal2003Andersonetal2011)
The Department of Health has drawn up a set of lsquoStandard Principles for the
PreventionofHealthcare‐AssociatedInfectionsrsquoforhospitalstoadhereto(Department
ofHealth 2001NHS Estates2004)The first guidelinecovers themaintenanceofa
clean hospital environment and describes the potential link between inadequate
environmentalhygieneandthespreadofmicroorganismscapableofcausingHCAIsIt
recommends that the hospital environment should be visibly clean and free from
soilageanddustbutnomicrobiologicalguidance isprovided(DepartmentofHealth
2001)Morerecentguidancestatesthathospitalsalsohavetoprovideandmaintaina
clean and appropriate environment for healthcare (Department of Health 2008)
34
althoughnospecificrecommendationonthecleanlinessoftheenvironmentisgiven
TheAmericanbasedCenters forDiseaseControlandPreventionhaveacknowledged
this association in a set of guidelines which recommend cleaning or disinfection of
environmentalsurfacesonaregularbasisinadditiontowhenvisiblysoiled(Rutalaet
al 2008) and more frequent cleaning and disinfection of high‐touch surfaces than
minimaltouchsurfaces(Sehulsteretal2003)Theserecommendationsareallbased
onvisualassessmenttodeterminethecleanlinessoftheenvironmentwhichisapoor
indicationoftheefficiencyofthecleaningprocess(Maliketal2003)
However proposed cleaning standards are not always adhered to This is
demonstratedbyanenvironmentalauditofarenalunitinanAustralianhospitalthat
showed just 43 of theminimum standardswere beingmet during an outbreak of
GRE (Bartley et al 2001) The epidemic was terminated with a combination of
measures including enhanced environmental cleaning and isolation of colonised
patientstopreventonwardtransmission
123 Levelofsurfacecontamination
Thelevelofenvironmentalcontaminationispartlydependentonthepatientsrsquositeof
colonisationorinfectionpatientswithMRSAintheurinestoolsorinawounddisplay
higher levelsofenvironmentalcontaminationthanpatientswithMRSA isolatedfrom
other body sites (Rutala et al 1983 Boyce et al 1997 2007 2007 2008) The
environment surrounding a GRE‐positive patient was seven times more likely to be
contaminatedwithGREthananun‐colonisedpatient(Haydenetal2006)andwhen
the routine environmental cleaning regimen was improved a decrease in
35
environmentalcontaminationwasobservedCertainlycontaminatedroomsarearisk
factor for the acquisition of nosocomial pathogens (Hota 2004) and a positive
correlationhasbeendemonstratedbetweenthelevelofAbaumanniienvironmental
contamination and the number of patients colonised or infectedwithA baumannii
(Dentonetal2004)
Theminimumlevelofcontaminationonasurfaceneededtoinitiatecolonisationofa
patientwhich could lead to an infection has not been quantified and is difficult to
measureMicrobiological standards have been proposed for hand‐touch surfaces in
hospitalsinanattempttodeterminewhetherthemicrobialcontaminationofagiven
surface presents a risk of infection for any patients in that vicinity (Dancer 2004
Mulvey et al 2011) It was proposed that an integrated and risk based approach
should be used encompassing visual assessment rapid assays to detect organic soil
and microbiological testing The standards for the microbiological assessment were
splitintotwosections(i)thepresenceofindicatororganismsand(ii)thetotalaerobic
colonycount
Indicator organisms are pathogens that pose a significant threat to patients and
include MSSA MRSA C difficile Salmonella species multi‐drug resistant Gram‐
negativebacilliGREanda numberofotherorganisms thatare important in certain
clinical situations such as Aspergillus species in a ward for severely
immunocompromised patients This standard was set at less than 1 cfu cm2 The
secondstandardwassettoprovideanindicationofthecompletemicrobialloadona
given surfaceasahighmicrobial loadonahand‐touch surface is likelyto represent
poorenvironmentalcleaningandtheheavygrowthofotherorganismsmayshieldthe
36
presenceofanindicatororganism(Dancer2004)Thisstandardwassetatlessthan5
cfucm2Thesestandardshavesincebeentestedandadapted indifferenthospitals
using various detection systems to validate the set benchmarks and are still under
review(Griffithetal2000Maliketal2003Ayciceketal2006Griffithetal2007
Oieetal 2007Danceretal 2008 Lewisetal 2008Dancer2011Mulveyetal
2011)
124 Frequencyofsurfacere‐contaminationpost‐cleaning
Thehospitalenvironmentisrapidlyre‐contaminatedaftercleaning(WeberandRutala
1997)andhospitalfloorscanbecomere‐contaminatedtothesamelevelasbeforethe
cleaning event within 2 hours (Collins 1988 Dettenkofer and Spencer 2007)
Benchmarkscouldbeusedtoestablishhowlongittakesforasurfacetobecomere‐
contaminated after cleaning so that the frequency of cleaning could be optimised
(Lewis et al 2008) Bed occupancy rates also have an effect on the microbial
contaminationofthehospitalenvironmentandtheriskof infectionwithMRSAOne
studydemonstratedgreaterbacterialcontaminationofsampledhand‐touchsurfaces
whenbed occupancy rateswere above 95 comparedwith bed occupancy rates of
below80anda separate study showed the riskof cross‐infectionwithMRSAwas
increasedforpatientsinafive‐beddedbaycomparedwiththoseinafour‐beddedbay
(Kibbleretal1998Danceretal2008)Bedmakinghasalsobeenshowntoincrease
airborne levelsofSaureuswhich thenhave thepotential to settleonnear‐patient
surfacesand further contaminate theenvironment (Shiomorietal 2002Hansenet
al 2010) Re‐contamination of the patient environment is not surprising given that
viable skin colonising microorganisms are carried on skin squames one million of
37
which are shed from healthy skin each day efficiently transferring bacteria into the
immediatesurroundings(Noble1975)
Bacterial contamination of the environment is not necessarily detrimental to a
patientrsquos health Bacterial contamination of the hospital environment is ubiquitous
even though the environment is dry and free from substances that encourage
microbial growth (Collins 1988 Dettenkofer et al 2011) Gram‐positive cocci are
most commonly found and more than 99 are likely to be coagulase negative
commensals and thus unlikely to cause serious disease To create an environment
completely free from bacteria would require sterilisation which is both impractical
andunnecessaryItwouldhoweverbeadvantageoustocreateanenvironmentwhere
thebacterialpopulationpresentdoesnotcontainpathogensand isunlikelytocause
infection(Collins1988)
125 Frequencyofcontactwiththehand‐touchsurface
Bydefinitionhandcontactuponhand‐touchsurfaces is frequentsothenumbersof
occasions for thepotential transferof pathogens fromcontaminatedhands to these
surfaces or vice versa is high The near‐patient environment contains numerous
hand‐touchsurfacesonanintensivecareunitforexampletherearevariousitemsof
instrumentationsuchasventilatorsandmonitorsthatcouldbepotentialreservoirsof
infection (Dancer 2008) Nursing staff rather than domestic staff are usually
responsible forcleaningthesesurfacesand it isoftena lowprioritytask in factonly
40 of these surfaces were shown to be cleaned adequately (Dancer et al 2008
Dancer 2009 Carling and Bartley 2010) Ten hand‐touch surfaceswere sampled in
38
two surgical units over a one year period and itwas found that near‐patient hand‐
touch sites cleaned by trained nursing staff were most likely to fail microbiological
hygiene standards as opposed to surfaces cleaned by domestic staff (Dancer et al
2008)Dentonetal(2004)clearlydefinedtheresponsibilityforcleaningthesehand‐
touch surfaces to thedifferent staff groupsduringanoutbreak ofAbaumannii and
this measure along with a number of others assisted in terminating the outbreak
Andersonetal(2011)demonstratedmorerecentlythatsurfacescleanedbydomestic
staff are more likely to pass defined hygiene standards than surfaces which are
cleanedbyotherstaffsuchasnursesandclinicalsupportworkers
Hands are an important fomite implicated in the transfer of pathogens between
patientsandimprovementsinroutinecleaningregimenshavebeenassociatedwitha
decrease in the contamination on the hands of healthcare workers (Hayden et al
2006) An association has been demonstrated between positive cultures from the
hands of healthcare workers and C difficile environmental contamination which
impliesthattheenvironmentcanplayarole in contaminatingthehandsofthestaff
(Samore et al 1996 Weber and Rutala 2011) Bhalla et al (2004) showed the
transfer of pathogens from the near‐patient environment to the hands of the
investigatorsinoverhalfofthesamplingoccasionsandsurprisinglypathogentransfer
occurred inoccupiedpatientroomsregardlessofthecolonisationor infectionstatus
of the patient These examples demonstrate the importance of adhering to defined
cleaningstandardswithdefinedrolesandresponsibilitiesforstaffmembers
39
126 Hygienepracticesofstaffpatientsandvisitors
There isa largevariation in thehandhygienepracticesofhealthcareworkersanda
recent systematic review of 96 studies reported hand hygiene compliance rates
ranging from 4 ndash 100with an overall average rate of 40 (Erasmus et al 2010)
Compliancewaslowerintheintensivecareunitsettingamongstcliniciansandbefore
patient contact even though this is the first of TheWorld Health Organisation Five
Moments for Hand Hygiene (Pittet et al 2009 Erasmus et al 2010) Intervention
campaigns to improve hand‐washing compliance are often effective during and
immediatelyafterthecampaign(Chengetal2011)butcomplianceratesoftendrop
inthemonthsaftertheintervention
Educating staff about the importance of cleaning the hospital environment has
resulted in improvements in the quality of cleaning as assessed by a number of
methodsUVpowdersandgelshavebeenappliedtosurfacestoassesstheefficiency
ofthecleaningregimenandanincreaseincleaningrateswasachievedafterfeedback
of surveillance results (Carling et al 2008 Munoz‐Price et al 2011) ATP
bioluminescencehasalsobeenusedtoassesscontaminationonhand‐touchsurfaces
and a reduction in the relative light unit (RLU) values was observed after a similar
education programmes (Poulis et al 1993 Griffith et al 2007 Boyce et al 2009
Mulvey et al 2011) Patient and visitor involvement in hand decontamination also
decreasesbacterialcontaminationofthehealthcareenvironment
40
13 Antimicrobialcoatings
Antibacterial materials could be used to supplement cleaning of the hospital
environmentandTheCentresforDiseaseControlandPreventionrecommendfurther
evaluating implementation of antimicrobial materials for use in the hospital
environment(Rutalaetal2008)Ithasbeenshownthatbacteriacanbespreadfrom
acontaminatedareatoanon‐contaminatedareaduringthecleaningprocess(Dharan
et al 1999) Recontamination of the hospital environment also occurs readily after
cleaning events (Collins 1988) and cleaning has often been found to be inadequate
with studies showing only 34 compliancewith policies (Carling and Bartley 2010
Carlingetal2010)
Ifhospitalsurfaceswerecoatedwithanantibacterialmaterialthenthecontaminated
areaswouldbesusceptibletothekillingeffectofthecoatinganddecontaminationof
theaffectedareascouldoccur inbetweencleaningeventsContinuousprotectionof
thehospitalenvironmentinthiswayhasbeenproposedbyanumberofauthorsasan
adjunct to other infection control procedures (Casey et al 2010) Reducing the
bacterial load in the environment can help to prevent person‐to‐person spread of
bacteriaandthedevelopmentofinfection
MRSAhasbeen isolatedfromcomputerkeyboardswithinahospitalward (Devineet
al 2001) howeverwhen self‐cleaning keyboardswere used in a surgicalward in a
Scottish hospital sampled surfaces were consistently below the defined ATP
benchmarks and passed the hygiene standards in the cleanliness audit (Anderson et
al2011)
41
131 Silverasanantimicrobialagent
Silver has a broad spectrum of activity and is active against Gram‐negative and ‐
positive bacteria fungi viruses and protozoa (Davies and Etris 1997 Martinez‐
Gutierrezetal2010)Theantibacterialeffectofsilverhasbeenknownforcenturies
andwas used by the ancient Egyptians and Greeks to treat infectious ailments For
exampleHippocratesdescribedtheuseofasilverpowdertotreatulcers(Hippocrates
400 BC) and at around the same time Alexander the Great kept his drinkingwater
clean by the use of silver water vessels (White 2002) Silver was re‐introduced for
topical applications in the 1960s in the forms of silver nitrate or silver sulfadiazine
especiallyinthepreventionofwoundinfections(Moyeretal1965Foxetal1969)
Inmore recent times silver has been coated ontomany substrates or impregnated
throughoutsubstancestoprovideantibacterialprotection(MelaiyeandYoungs2005)
Theuseofsilvernanoparticlesisincreasingduetotheirhighantibacterialactivityand
smallsizewhichprovidesalargesurfaceareatovolumeratio(Rupareliaetal2008
Lvetal2010)
1311 Mechanismofaction
Themechanismbehindtheantibacterialactivityofsilverandothermetalionsisdueto
theoligodynamiceffectfirstdescribedbyKarlWilhelmvonNaumlgeliasthelethaleffect
thatsmallmetalionsexertonlivingcells(Kraemer1905)Silverbindstothiolgroups
on the bacterial proteins including the ribosome and NADH dehydrogenase which
inhibitstheexpressionofenzymesrequired inATPproductionandpreventselectron
transfer and respiration respectively (Davies and Etris 1997 Plowman et al 2000
Percivaletal2005Yamanakaetal2005Kimetal2008Liuetal2010)Oxidation
42
ofkeycomponentsoftherespiratorypathwayinhibitsbacterialrespiration(Braggand
Rainnie1974)andsilveralsoreactswithmicrobialDNAtocausethefreeDNAtoform
a condensedAg‐DNA complex in the centre of the cellwhich results in a loss in its
replicative function (Feng et al 2000Melaiye and Youngs 2005) Externally silver
targetsthebacterialcellmembraneandonceboundcausespittingand interference
of membrane function which has been visualised by electron microscopy (Clement
andJarrett1994Linetal1996Percivaletal2005Kimetal2007)Interactions
withthecellmembranealsocauseacollapseintheprotonmotiveforceleadingtothe
leakageofH+de‐energisationof themembraneandcelldeath (Dibrovetal 2002)
Silver nanoparticles have also been shown to form silver‐sulfur aggregates on the
surfaceofbacterialcellswhich interfereswiththegenerationof freeradicalswhich
cancausedamagetobacterialcellmembranes(Kimetal2007)
Serious adverse effects of silver in humans is limited to neurotoxicitywhich is only
experienced if theblood‐brainbarrier isbreechedand invitro toxicitytomammalian
cells has not been replicated in the treatment of wound infections (Melaiye and
Youngs2005Tayloretal2009)
Zone of inhibition or agar pour plate tests were used to demonstrate the diffusible
antibacterialactivityofsilver‐basedcompoundsagainstarangeofbacteriaincludingE
coli Klebsiella pneumoniae P aeruginosa Streptococcus mutans S epidermidis S
aureusBacillusanthracisAcinetobacterbaylyiMycobacteriumfortuitumandCandida
albicans(Furnoetal2004Ebyetal2009DurucanandAkkopru2010Gerasimchuk
etal2010Pollinietal2011Riveroetal2011)Thisdiffusibleantibacterialactivity
wouldbeadvantageousforimplantsorsurgicalinstrumentstogiveaninitialhighdose
43
of silver to the surrounding environment which would decrease the likelihood of
resistancedeveloping(Stobieetal2008)Thereleaseofsilver fromthesurfacecan
be further controlled bymodifying the composition of the coating (Liu et al 2010)
Combiningsilverwithanantibioticagentcanfurtherenhancetheantibacterialactivity
(Fox1968Shahverdietal2007Kimetal2008)
1312 Resistancetosilver
Silver isabiocideandassuchhasmultiplemodesofactionunlikeanantibioticthat
tendstotargetaspecificsite(Percivaletal2005)Biocidesthereforehaveabroader
spectrum of activity and resistance is less likely to occur Silver resistance was not
detectedinanybacterialstrainscausingurinarytractinfectionsinpatientswithsilver‐
coated catheters in situ over a 12‐month period (Rupp et al 2004) However
resistance has been identified inmany species of bacteriamainly from burns units
where silver‐based dressings are used to prevent bacterial infection (Clement and
Jarrett1994Silver2003)
A strain of silver‐resistantSalmonellawas isolated froma hospital inMassachusetts
andtheresistancedeterminantwasfoundtobea180kbplasmidpMG101(McHughet
al 1975) Much work has since been performed on this plasmid to elucidate the
molecular basis for resistance and the sequenced region is available on Genbank
(Gupta et al 1999) The gene cluster includes a periplasmic silver‐specific binding
protein(SilE)andtwoparalleleffluxpumps(SilPandSilCBA) (Guptaetal1999)and
amplification of these genes provides a rapidmethodof identifying resistant strains
(Percival et al 2008) Genotypic resistance does not typically translate directly into
phenotypic resistance three strains of Enterobacter cloacae isolated from burn
44
woundswerefoundtocarrytheseresistantgenesbutstilldemonstratedsusceptibility
to therapeutic levels of silver in vitro (Percival et al 2005) The widespread
developmentofresistancetosilver isunlikelyasbacteriahavebeenexposedtosub‐
inhibitory concentrations of this metal ion for centuries however greater use will
increasethelikelihoodofresistancedeveloping(Percivaletal2008)
1313 Applicationsofsilverasanantimicrobialmaterial
13131 Centralvenouscatheters
Silver‐coatedcathetershavebeendevelopedwiththeaimtoreducetheprobabilityof
developingline‐associatedinfectionswhichareacommoncauseofHCAIs(Noimarket
al2009Syedetal2009)Experimentallysilver‐coatedpolyurethanecatheterswere
inserted intoaratmodelandbacteriacouldnotbe isolatedfromthesurfaceofthe
linesafter6weeksimplantationintheinternaljugularvein(Bambaueretal1997)A
significant reduction in E coli adhesion on silver‐coated polyurethane catheterswas
demonstrated in vitroandofthosebacteriathatdidadhereagreaterproportionof
cells found on the silver‐containing polymer were non‐viable compared to the
uncoatedcontrols(Grayetal2003)
13132 Urinarycatheters
The American‐based Healthcare Infection Control Practices Advisory Committee
publishedguidelinesdetailingbestpractices inthepreventionofcatheter‐associated
urinarytractinfectionandtheuseofantimicrobialcathetersweretobeconsideredif
othermethodsofdecreasingratesofinfectionwerefailing(Gouldetal2010)Inthe
USAa trialon theuseof silverhydrogel coatedcatheterswas conductedcompared
45
with standard siliconehydrogel urinary catheters and the incidence of catheter‐
associated urinary‐tract infections fell from63 infections per 1000 catheter days to
26infectionsper1000catheterdaysachievinga57reductionoverall(Ruppetal
2004) In a separate study a 60 reduction in catheter‐associated urinary‐tract
infectionswasachievedfollowingintroductionofsilvercoatedcathetersachievingan
annual saving estimated to be in the region of pound38000 and the release of 192 bed
days(ReportbytheComptrollerandAuditorGeneral‐HCSession2003‐2004)
1314 Endotrachealtubes
An endotracheal tube (ETT) containing silver nitrate and sodium hydroxide reduced
adhesionofPaeruginosa(Monteiroetal2009)andanumberofotherstudieshave
demonstrated clinical efficency of silver coated ETTs this is further discussed in
Section 14 Silver coated endotracheal tubes have been approved for clinical use in
the USA but the increased cost and risk of breakthrough events of VAP have
preventeditsrsquowidespreaduse(Raadetal2011)
1315 Environmentalsurfaces
Silver‐based compounds can also be employed on inanimate surfaces which could
potentiallybeaddedtohand‐touchsurfacessol‐geldepositionwasusedtosynthesise
silver‐doped phenyltriethoxysilane films that prevented S epidermidis adhesion and
biofilm formation over a 10‐day period (Stobie et al 2008) Silver‐doped TiO2 and
titaniumnitridethinfilmscausedsignificantdecreases intheviabilityofSaureusE
coliStreptococcuspyogenesandAbaumannii(Kellyetal2009Wongetal2010)P
aeruginosa appeared more sensitive to the titanium nitride films and growth was
46
inhibitedforupto7dayssupportingthehypothesisthatGram‐positivebacteriaare
more resistant to the antibacterial effects of silver This could be due to the larger
amount of negatively‐charged peptidoglycan in the thicker Gram‐positive cell wall
whichcouldbind silver thus reducing the silveravailable toactupon the interiorof
thecell to causedamage (Schierholzetal 1998Kawaharaetal 2000Grayetal
2003Monteiroetal 2009)Howeverothergroupshave shown thatGram‐positive
and ‐negative strains possess similar susceptibility to silver (Ruparelia et al 2008
Wongetal2010)Inarecenthospitalstudyarangeofsilver‐coatedproductswere
placed in ward areas to monitor the effect on bacterial contamination of the
environment and up to 98 fewer bacteria were recovered from the environment
compared with a control ward which contained uncoated products (Taylor et al
2009)Theantimicrobial activity lasted for thedurationof the12‐month testperiod
andadverseeffectstosilverwerenotreported
1316 Otherapplications
Surgicalmaskshavebeenimpregnatedexperimentallywithtitaniumdioxide(TiO2)and
silvernanoparticlesandnoviableSaureusorEcoliwasdetectedafter48hoursNo
adversereactionswereobservedinhumanvolunteers(Lietal2006)Silverhasbeen
incorporated intodental composite resinsanda slowand sustained releaseof silver
intothesurroundingenvironmenthasbeendemonstratedwitha6‐logreductioninS
mutans growth after 12 hours (Kawashita et al 2000) These composites could
potentiallyreduceinfectivecausesofsurgicalimplantfailure(Floresetal2010)Silver
nanoparticleshavebeen incorporatedwith lysozymeandcoatedonto stainless steel
surgical blades and needles and significant antibacterial activity against a panel of
47
Gram‐positiveandGram‐negativebacteriawasobserved(Ebyetal2009)Silverwas
added to an ethanol‐based disinfectant to generate additional residual antibacterial
activitypost‐application(Bradyetal2003)Silvernanoparticleshavealsobeenused
inenvironmentalsettingssuchasinwastewatertreatment(Linetal1996Daviesand
Etris1997)
132 Copperasanantimicrobialagent
TheantibacterialactivityofcopperhasalsobeenknownforcenturiesandHippocrates
describeditasacureforulcers(Hippocrates400BC)Awiderangeofmicroorganisms
aresusceptibletocopperincludingSaureusEcoliCdifficileEfaecalisEfaecium
Mycobacterium tuberculosisAspergillus fumigatusCalbicansand influenzaAH1H1
(Grassetal2010)Copper‐dopedTiO2coatingswereappliedtoatitaniumalloyasa
model formetal implants used for total joint arthroplasty and a 6‐log reduction in
MRSAgrowthwasobservedafter24hourscomparedwiththeTiO2coatingswithout
the copper ions (Haenle et al 2010) Noyce et al (2006) inoculated MRSA onto
coppersurfacesandwereunabletorecoverviablebacteriafromthesurfacesafter45
minutesincubationatroomtemperatureSignificantreductionswerealsoachievedat
4degC and frombrasswhich contains 80 copper although extended exposure times
wererequired
Coppersurfaceshavebeenassessedfortheiruseinthehealthcareenvironmentinthe
UKUSAChileandJapan(Pradoetal2010Schmidtetal2011KeevilandWarnes
2011)Copper‐containingtapsdoorpushplatesandtoiletseatswere installed inan
acute medical ward in the UK and compared with non‐copper containing control
48
surfaces and the level of bacterial contamination found on the copper‐containing
surfaceswassignificantly lowerthanthatfoundonthecontrolsurfaces(Caseyetal
2010)Thetoiletseatandtaphandlesurfacespassedthebenchmarkmicrobiological
standards proposed by Dancer (2004) for hand‐touch surfaces whereas 50 of the
controlsurfacesfailedHoweverthecleanlinessofthesurfaceaffectscopperactivity
and cumulative soiling and cleaning of copper surfaces was shown to inhibit
antibacterial activity this decrease in antibacterial activity was not observed on
stainlesssteelcontrolsurfaces(AireyandVerran2007)
The mechanism of activity of copper has been shown to be predominantly due to
disruption of cellular respirationDNAdamage by the generation of reactive oxygen
and ionic copper species which cause damage to bacterial enzymes and proteins
(Yoshidaetal1993Noyceetal2006Weaveretal2010)Thecellmembranemay
also be damaged during exposure to copper which leads to rupture and loss of
membranepotential (Grassetal2010)althoughthis isnotthemainmechanismof
celldeath(WarnesandKeevil2011)
133 Titaniumdioxidephotocatalyticthinfilms
Titanium dioxide has inherent light‐activated antibacterial activity and its
functionalitieshavealreadybeencommerciallyexploitedTiO2 coatingsareavailable
as self‐cleaning glasses with Pilkington Activtrade and Saint Gobain BIOCLEANtrade as the
marketleadersTheglasscanbeusedinwindowsconservatoriesandglassroofsand
requires less frequent cleaning because of the dual photocatalytic and
superhydrophilic activities of TiO2 Modified TiO2 has the potential for use in
49
healthcare institutions to reducebacterial contamination of theenvironmentbut to
understand how the TiO2 thin films are activated by light to exert an antibacterial
effect it is firstnecessarytogainabasicunderstandingofbandtheoryofsolidstate
materials
1331 Bandtheoryofsolids
Solid state materials can be split into three categories conductors insulators and
semiconductors (West1999)Their characterisationwithinthesegroupsdependson
theband structurewhich in turn dependson thepositioningof theelectronswithin
theatomsandmoleculesastheycometogethertomakeasolidmaterialElectronsare
arrangedintobandsthatcontainspaceorlsquoholesrsquofortheelectronstoexistinNotwo
electronscanoccupythesamespaceanditispreferentialfortheelectronstoexistin
pairsThecategoryofthesoliddependsuponthenumberofspacesavailableandhow
manyelectronstherearetofillthesespaces
13311 Conductors
Materialscharacterisedasconductorshaveanlsquounfilledconductionbandrsquo(Figure13)
Figure13Schematicofaconductionbandinaconductor
Electronhole
Electronlyingwithinahole
50
Theelectronsinconductorsarefreetomovefromoneholetoanotherwithnoenergy
inputandahole isleftinthespacefromwhichtheelectronhasmoved(Figure14)
The electrons are able to transport charge because of this free movement and
therefore the material is an electronic conductor Metallic materials fall into this
category
Figure14Freemovementofelectronswithinaconductor
13312 Insulators
If theconductionbandofamaterial is full (Figure15) theelectronsarenotableto
moveandsoconductionofelectricitywillnotbepossibleThismaterialisclassifiedas
aninsulator
Figure15Schematicofaconductionbandinaninsulator
Electronhole
Electronlyingwithinahole
51
13313 Semi‐conductors
Inadditiontothepreviouslydescribedbandsanadditionalsetofelectronholesalso
exists above the conduction band and there is a further set found above that
However an input of energy is required in order to promote an electron from the
valence band (highest band occupied by electrons) to the conduction band (lowest
bandwithspacesforelectrons(Figure16))Thisenergyinputiscalledthebandgap
Figure16SchematicdisplayingthebandgapwithinasolidstatematerialwhereCB=conductionbandandVB=valenceband
The band gap of insulators like rubber is very high and a large input of energy is
required to promote the electron to the conduction band Semiconductors however
have an accessible band gap (Figure 17) a small amount of energy is required to
promoteanelectron to theconductionbandand thus createa conductoroutofan
insulator (Carp et al 2004) Once the electron has been promoted conduction can
occurviatwopossiblerouteseitherwithinthevalencebandusingthepositiveholes
createdorwithintheconductionbandsthroughthemovementofelectrons
Electronhole
Electronlyingwithinahole
Bandgap
CB
VB
52
Figure17Promotionofanelectron fromthevalenceband (VB) to theconductionband(CB) inasemiconductorafterabsorptionof lightwithawavelengthmatchingthebandgapenergyofthematerial
Theexcitedelectroncansubsequentlyfallfromtheconductionbandintoaholeinthe
valencebandwhichresultsintheemissionoflightenergyofthesamewavelengthas
theabsorbedincidentrayAlternativelysemi‐conductormaterialssuchasTiO2canbe
dopedwithelementssothattheseparationoftheholeandelectroncanbestabilised
andtheabsorbedenergycanbeutilised
13314 DopedSemiconductors
Doped semiconductors can be classified into one of two groups depending on the
chemical properties of the dopant material n‐type semiconductors or p‐type
semiconductorsInann‐typesemiconductorthedopantmaterialhasavalenceband
which isslightly lower inenergythantheconductionbandofthesemiconductorbut
higherinenergythanthevalencebandofthesemiconductor(Figure18)(Carpetal
2004)Conductionoccurswhenanelectronispromotedfromthevalencebandofthe
dopanttotheconductionbandofthesemiconductorwhichrequireslessenergythan
thenormalelectronictransition
Electronhole
Electronlyingwithinahole
Lightin
CB
VB
53
Figure18n‐typesemiconductors‐positioningofthedopantvalencebandinrelationtothesemiconductorconductionband(CB)andvalenceband(VB)
Alternativelyinap‐typeconductorthedopantmaterialhasaconductionbandwhich
isslightlylowerinenergythantheconductionbandofthesemiconductor(Figure19)
Electronsaretrapped inthedopantconductionbandandconductionoccursthrough
the positive holes The number of electrons should always equal the number of
positiveholesbecausetheproductionofasinglefreeelectronresultsinthecreation
ofasinglepositivehole
Figure 19 p‐type semiconductors ‐ positioning of the dopant conduction band inrelationtothesemiconductorconductionband(CB)andvalenceband(VB)
Anumberofprocessescanoccuronthesemiconductorafterelectronicexcitationand
thesearesummarised inFigure110(MillsandLeHunte1997)Anelectron(‐)anda
positivehole(+)aregeneratedandasmentionedpreviouslyTheelectroncouldreturn
Normaltransition
Dopantmaterialwithlower
conductionband
CB
VB
Normaltransition
Dopantmaterialwithhighervalenceband
CB
VB
54
to the valence band of the semiconductor which is termed electron‐hole
recombinationThisprocesscouldoccuronthesurfaceofthesemiconductor (Figure
110 i) or within the bulk of the semiconductor (Figure 110 ii) Alternatively the
electroncouldreduceanelectronacceptor ina redoxreactiononthesurfaceofthe
semiconductor(Figure110iii)orthepositiveholecouldoxidiseanelectrondonoron
thesurfaceofthesemiconductor(Figure110iv)
55
Figure110Diagramtoillustratethemainreactionstakingplaceonasemiconductormoleculeafterexposure toa light sourcecausingelectronicexcitation (i)electronholerecombinationatthesurface (ii)electron‐holerecombination inthebulk (iii)reductionofanelectronbyanelectronacceptorat the surface (iv)oxidationofapositive hole by an electron donor at the surface Figure amended from thesemiconductorreviewbyMillsandLeHunt(MillsandLeHunte1997)
1332 Titaniumdioxideasasemiconductor
Titanium dioxide (TiO2) is commonly used as a semiconductor as it is inexpensive
chemically stable non‐toxic possesses a high refractive index and has excellent
transmission inthe infraredandvisibleregions(DoboszandSobczynski2003Parkin
andPalgrave2005Dunnilletal2011)TiO2existsinmanypolymorphsandthemost
abundant are anatase and rutile (Parkin and Palgrave 2005) Pure anatase tends to
display greater photocatalytic properties than rutile due to the faster electron‐hole
recombinationrateofrutiletitania(MillsandLeHunte1997Allenetal2005Brook
56
etal2007b)WhenTiO2intheanatasecrystallineformisexposedtowavelengthsof
lightbelow385nmitbehavesasann‐typesemiconductor(Carpetal2004)andfree
electronsandpositiveholesarecreatedinthefollowingreaction
TiO2 h+vb+e‐cb
The positive holes react with water present on the surface of the thin films in the
followingreactionstogeneratehydroxylfreeradicals
h+vb+H2Oadsorbed OH+H+
h+vb+‐OHsurface OH
Thefreeelectronsparticipateinthefollowingreactionstogeneratethesuperoxideion
andsubsequentlyhydroxylfreeradicals
e‐cb+O2 O2‐
2O2‐+2H2O 2HO+2OH‐+O2
Thegeneratedreactiveoxygenspeciescanreactwithorganicmaterialonthesurface
ofthesemiconductorwhichundergooxidationorreductionreactionsPhotoreactions
occurring on the surface of a catalyst such as TiO2 are termed heterogeneous
photocatalysis(MillsandLeHunte1997)
ThegenerationoffreeelectronsandpositiveholesinTiO2wasfirstdescribedin1972
whenwaterwasdecomposedafterexposuretoUVlight(FujishimaandHonda1972)
λlt385nm
57
Thiswasfollowed in1979byresearchdemonstratingthegenerationofthehydroxyl
radical by electron spin resonance after irradiation of TiO2 by UV light (Jaeger and
Bard1979)Theheterogeneousphotocatalyticprocessisdependentonthepresence
ofwateronthesurfaceofthecatalystandoxygenasanelectronacceptor(Figure110
iii)
1333 Titaniumdioxide‐basedantibacterialphotoactivity
The bactericidal activity of the TiO2 photocatalyst increases proportionately as the
concentration of oxygen is increased from 0 to 100 (Wei et al 1994) Near UV
lightwithwavelengthsbetween300and400nmisthe lightsourcemostcommonly
used for bacterial photoinactivation experiments becauseUV lightwithwavelengths
under300nmareabsorbedbynucleicacidsandcancausemajordamagetoorganisms
(Saitoetal1992)NearUVlightisnotabsorbedbynucleicacidsandsoanyobserved
damagecanbeattributedtothephotoactivityofthecatalystandnottheincidentlight
source
13331 Demonstratingthelossofcellviability
Theseminalpaperinthefieldofphotocatalysisdescribedthephotoinactivationofthe
Gram‐positive bacterium Lactobacillus acidophilus the Gram‐negative bacterium E
coli the yeast Saccharomyces cerevisiae and the green alga Chlorella vulgaris
(Matsunagaetal1985)Asuspensionofplatinum‐loadedtitaniumoxidewasadded
toeachmicrobialsuspensionbeforeaUVlightsourcewasappliedareductioninthe
viability of all organisms was observed The concentration of coenzyme A (CoA)
generatedthroughoutthecourseoftheexperimentwasmonitoredandadecreasein
58
CoAconcentrationwasassociatedwithalossofcellviabilityTheypostulatedthatthe
mechanismofactionwasthephotoelectrochemicaloxidationofCoAwhichresulted
inadecreaseinthemetabolicactivityofthecellsandsubsequentcelldeath
Thegroup followeduptheseexperimentsby immobilisingtheTiO2particleswithina
membraneinacontinuousflowsystemwhichwasusedtosterilisewaterspikedwith
Ecoli(Matsunagaetal1988)AdecreaseinCoAconcentrationwasagainobserved
and reactive oxygen specieswere implicated in the photoinactivation ofE coli The
electrondonorCoAwasoxidisedbythepositively‐chargedholesinthevalenceband
A similarexperimental rigwasusedby Irelandetal (1993) to furtherelucidate the
mechanism of the photocatalytic bactericidal activity of TiO2 E coli in an aqueous
suspension was photoinactivated and after a 9 minute exposure time a 9 log10
reductionwasobservedWhenhydrogenperoxide(H2O2)wasaddedtothesystemit
actedasanirreversibleelectronacceptorandparticipatedinthefollowingreactions
H2O2+e‐cb OH+OH‐
H2O2+O2‐ OH+OH‐+O2
Thegenerationofhydroxylradicalswaspromotedwhich inturnreducedtherateof
electron‐holerecombinationwhichwasaccompaniedbyanincreaseinphotocatalytic
activity Photoinactivation of Streptococcus sobrinus was also demonstrated after
exposureto21nmdiameterparticlesofTiO2andUVlighta5log10decreaseinviable
bacteria was seen after just 1 minute at a bacterial concentration of 105 cfu mL
Photocatalytic activity was reduced when the bacterial inoculum was higher and it
59
took 60minutes to achieve a 5 log10 decrease in S sorbrinus when a 109 cfu mL
inoculumwasused(Saitoetal1992)
A combination of reactive oxygen species is necessary to exert a photocatalytic
bactericidaleffectwith thehydroxyl radical as theprimary radical actingdirectlyon
the cell (Yan et al 2009) Hydrogen peroxide has also been postulated to directly
contribute towards the bactericidal activity as an increase in the concentration of
catalase which degrades hydrogen peroxide to water and oxygen increased the
survival rate of E coli (Kikuchi et al 1997) Therefore hydrogen peroxide could
provide a source of hydroxyl radicals and act as a direct attacking agent (Yan et al
2009)
Viruses have also been shown to be susceptible to the photocatalytic effect of
irradiated TiO2 The non‐enveloped polio virus was spiked intowastewater samples
containingastocksolutionofanataseTiO2andarapid inactivationofthepoliovirus
wasobserved(Wattsetal1995)A2log10decreaseinviablepolioviruswasdetected
after30minutes comparedwitha150minutesexposure time toachieve the same
reductionofEcoliTheincreasedsusceptibilityofthepoliovirustophotoinactivation
waspostulatedtobeduetothelowsurfacetovolumeratiocomparedwithbacteria
whichprovidedahigherrateofhydroxylradicalreactionwiththeextracellularprotein
capsidofthevirus(Wattsetal1995)
60
13332 Detectingchangesinthebacterialcellarchitecture
The activity of the hydroxyl radical is limited by diffusion through the outer and
cytoplasmic membranes (Watts et al 1995 Sunada et al 1998) therefore
compromiseofthesebarrierswillallowgreateractivityofthereactiveoxygenspecies
Potassium ion (K+) leakage was used to demonstrate increased cell membrane
permeability as an indicator of damage to the integrity of the cell membrane An
increaseintheextracellularK+concentrationwasdetectedafterlightirradiationwith
TiO2presentasapowderwhichoccurredinparallelwiththelossincellviability(Saito
etal1992Luetal2003)TheleakageoflargermoleculessuchasRNAandprotein
hasalsobeendetectedaccompaniedbyalossincellviability(Saitoetal1992)
Using transmission electronmicroscopy (TEM) the internal changes associatedwith
photocatalysis couldbevisualisedand thedestructionof thecytoplasmicmembrane
andintracellularcontentswasobservedafter60ndash120minuteslightirradiation(Saito
et al 1992) The reactive oxygen species generated initially damaged the bacterial
peptidoglycan layerbeforeattacking thecytoplasmicmembrane causing irreversible
damageChangesintheoutermembranestructureofEcoliinoculatedontoTiO2thin
films has been demonstrated by atomic force microscopy (AFM) (Lu et al 2003
Sunadaetal2003)After10minutescellviabilityhaddecreasedandacompleteloss
inintegritywasseenafter60minutesWhenbacterialspheroplasts(which lackacell
wall)wereinoculatedontoTiO2thinfilmstherateofbactericidalactivitywasgreater
than thatobserved for the intact cells suggesting that thecellwall hasaprotective
effect on E coli and is the initial site of photocatalytic attack (Sunada et al 2003)
Quantumdots(QD)havealsobeenusedasamarkerofchangesinthepermeabilityof
61
thecellmembraneQDarelightemittingcolloidalnanocrystallinesemiconductorsand
after 20minutes irradiation QDwere shown to enter E coli cells demonstrating a
changeincellmembranepermeability(Luetal2003)
Lipid peroxidation has been demonstrated to occur at the surface of E coli during
photoinactivation inthepresenceofTiO2 (Manessetal1999Soumlkmenetal 2001)
Lipidperoxidationisaprocessinwhichfreeradicalsremoveelectronsfromlipidssuch
as those within the bacterial cell membranes which results in a reduction in the
integrityofthemembraneandthuscellviabilityMalondialdehyde(MDA)aproduct
oflipidperoxidationwasusedasamarkerandanaccumulationofMDAwasdetected
withanaccompanyingdecrease incellularrespiratoryactivityTheauthorsproposed
that reactive oxygen species were generated on the TiO2 surface and attacked the
polyunsaturatedphospholipidspresentintheoutermembrane(Manessetal1999)
TiO2particlesalsointeractwiththeoutermembranecausingreversibledamagewhich
doesnotaffecttheviabilityofthecells(Huangetal2000)Oxidativedamagefollows
whichincreasesthepermeabilityofthecellcausingeffluxofintracellularcomponents
Once thecytoplasmicmembranehasbeen severely compromisedTiO2particles can
enter the cell and directly attack intracellular components Intracellular components
arethenabletoleakoutofthecellandtheo‐nitrophenol(ONP)assaycanbeusedto
detectthisAnincreaseinONPlevelswasobservedinEcoliwhichsignifiedincreased
permeability of the cellmembranes (Huang et al 2000) Bacterial endotoxin is also
degraded in the photocatalytic process and occurs simultaneously with E coli cell
death(Sunadaetal1998)
62
13333 Photoinducedoxidativebacterialdecomposition
InterestinglybacteriacanundergooxidativedecompositionuponthesurfaceofTiO2
thinfilmsuponexposureto356nmlight(Jacobyetal1998)AsuspensionofEcoli
was inoculatedonto irradiatedTiO2thinfilmsandSEMandcarbondioxideevolution
was used tomonitor photocatalytic oxidation After 75 hours exposure to UV light
decompositionofthebacterialcellswasevidentinstarkcontrasttotheuncoatedglass
slidesusedascontrolsAconcomitantincreaseintheconcentrationofcarbondioxide
(CO2)wasalsodetectedPhotocatalyticoxidationofBacillussubtilisvegetativecellsB
subtilissporesandAspergillusnigersporeswasalsodemonstratedandincreasedCO2
concentrations were used as markers of microbial decomposition (Wolfrum et al
2002) The rate of oxidationwas slower forA niger cells comparedwith the other
testedorganismsThishasimportanttranslationalimplicationsasitprovidesevidence
that the coatings are self‐cleaning and do not require a physical removal step after
photoinactivation organic matter present on the surface of the catalyst can be
mineralisedifexposedtothelightsourceforanadequatetimeperiodprovidingmore
spaceforphotocatalyticreactionstotakeplace
1334 Enhancingthepropertiesoftitaniumdioxidethinfilms
AdditionalelementscanbeaddedtoTiO2toalterthechemistryofthematerialTiO2
can be dopedwith substances such as nitrogen or sulfur to cause a batho‐chromic
shiftwhichalters thebandonsetenergy (Section13314) so thatphotonsof light
withalowerfrequencyareabsorbedandareabletoexcitetheelectronstoahigher
energystate(Asahietal2001Carpetal2004)Transitionmetalionssuchasiron
leadandcoppercanalsobeusedasdopantstoenhancethephotocatalyticproperties
63
ofTiO2(ThompsonandYates2006)Theaimofthisdopingistogenerateamaterial
that can be activated by visible light such as indoor lighting conditions which
broadens the commercial applications of the material A ten‐fold increase in the
numberofphotonsavailable forphotocatalysiswouldbegeneratedbyashift inthe
TiO2bandonsetofjust40ndash50nm(DunnillandParkin2009)
The exact mechanisms governing visible light photocatalysis are poorly understood
althoughitisgenerallyagreedthatnitrogendopingcausesincreasedphotocatalysisat
lower photon energies and localised nitrogen 2p states above the valence band are
generatedbytheadditionofnitrogen(ThompsonandYates2006)Itisnotyetagreed
whether substitutional or interstitial nitrogen binding provides the most favourable
visiblelightdrivenphotocatalyticproperties
14 Relevanceofsurfacesinventilator‐associatedpneumonia
Ventilator‐associated pneumonia (VAP) is a serious healthcare‐associated infection
that affects patients on ventilators predominantly in the intensive care unit The
intubatedpatientusuallyhasseriousco‐morbiditiessuchthattheyrequireassistance
with theirbreathingand thephysicalpresence of theendotracheal tube (ETT)both
compromisesthenormalactionoftherespiratorytractandallowsmicro‐aspirationof
contaminatedsubglotticsecretions
AnumberofclinicalmeasurescanbeappliedtopreventVAPaspreventionrequiresa
multifactorial approachand research into the subject includes theuseofalternative
ETTmaterials (Balk2002Pneumatikosetal 2009Torresetal 2009Bouadmaet
al 2010 Berra et al 2011 Blot et al 2011 Coppadoro et al 2011 Rewa and
64
Muscedere 2011) Bacteria originating from the oropharynx colonise the ETT and
produceabiofilmonthelumenofthetubewhichisdifficulttoremoveandprovidesa
potentialsourceofcolonisationandinfectionofthelowerairways(Sottileetal1986)
Therefore the prevention of bacterial adhesion to the surface of the ETT and the
destructionandremovalofboundorganismsisofclinicalinterest(Berraetal2003)
Polyurethane cuffed ETTs are being used in preference to the traditional
polyvinylchlorideETTsas theyaremore flexibleandabetter seal isproducedat the
base of the tube which prevents leakage of oropharngeal contents into the lower
airways (Berra et al 2008b Miller et al 2010) An alternative novel way to
decontaminate theETT isbyusing theMucusShaverwhichphysically removesboth
mucus and bacterial biofilms from the inner lumen of the tubing (Kolobow et al
2005)
ETTs can also be impregnated with antibiotics or other antibacterial compounds to
preventtheinitialbiofilmformationstageortokilltheadherentorganismsSilverions
have been added to polyurethane ETTs and a series of in vitro studies have
demonstrated reduced adherence of MRSA P aeruginosa Enterobacter aerogenes
andAbaumanniitothesilver‐coatedmaterials(Berraetal2008aRelloetal2010)
Colonisationof silver‐coated ETTsbyPaeruginosawas shown tobe lowerand take
longerthanonuncoatedcontrolETTswithlowerlevelsoflungcolonisationobserved
inventilateddogsasa consequence (Olsonetal 2002Relloetal 2010)A similar
study used silver‐sulfadiazine and chlorhexidine coated ETTs in ventilated dogs and
demonstratedareductionintrachealcolonisationandanabsenceoflungcolonisation
(Berraetal2004)
65
Whensilver‐coatedETTswereusedinastudyinvolvingninepatientsnoneoftheETTs
werecolonisedwithpathogens therewas lesscolonisationofcommensalorganisms
andtherewasadecreaseinbiofilmformationcomparedwiththenon‐coatedcontrol
ETTs(Relloetal2010)AdelayedETTcolonisationtimeandpositivetrachealaspirate
culture time was demonstrated in an earlier study using the same coated material
(Relloetal 2006)andnobacterial growthorbiofilmproductionwasdetectedona
silversulfadiazinecoatedpolyurethaneETTused inacohortof46 intubatedpatients
(Berra et al 2008b) A reduced incidence of VAPwithin 10 days of intubationwas
observedintheNASCENTtrialwhichrecruitedover2000patientssilver‐coatedETTs
were used in the test group and were compared with non‐coated equivalents that
wereusedinthecontrolgroup(Kollefetal2008)
A number of silver‐coated ETTs are now commercially available butwidespread use
has been hindered by the pricewhich is up to 45 timesmore than uncoated ETTs
however a theoretical cost‐analysismodel showed silver‐coated ETTswere actually
associatedwithfinancialsavingsofover$12000peravertedcaseofVAP(Shorretal
2009Torresetal2009)
Chlorhexidinehasbeencombinedwiththedyebrilliantgreenorgentianviolettoform
the novel compounds gardine and gendine respectively These compounds have
displayedsignificantantibacterialactivity invitroand inanelegantbiofilmdisruption
assaydemonstratedsuperioritytosilvercoatedETTsThesecompoundsarerelatively
cheap to produce and the authors propose clinical use after thorough in vivo
assessment(Chaibanetal2005Hannaetal2006Hachemetal2009Reitzelet
al2009Raadetal2011)Thesestudies illustratethebenefitsofantibacterialand
66
novel ETTmaterials and to further improve the incidence of VAP and other device‐
relatedinfectionsfurtherresearchshouldbeconducted
141 Photodynamictherapy
AdifferentmethodofgeneratinganantibacterialeffectonthesurfaceoftheETTsis
viaaprocess calledphotodynamic inactivation (PDI)Phototherapywas firstusedby
theNobelPrizewinnerNielsFinsentotreatatuberculosisskinconditioncalled lupus
vulgaris in the 1890rsquos by applying light directly onto the lesions (Bonnett 1995
Dolmansetal2003)Photodynamictherapy(PDT)evolvedfromthisinitialworkand
involves the use of a photosensitising agent and a light source to generate toxic
reactive oxygen species (Wainwright 1998) The procedure can be used in the
targetedtreatmentofcanceroustumours(MarcusandMcIntyre2002Dolmansetal
2003) in ophthalmology to treat age‐related macular degeneration (Bressler and
Bressler2000)atherosclerosis(Rocksonetal2000)andinthelocalisedtreatmentof
bacterial infectionsparticularlyindentistry(Wainwright2003)WhenPDTisusedto
killbacteriaitistermedphotodynamicinactivation(PDI)(HamblinandHasan2004)
There are two types of photosensitisation reactions type I and type II and the
pathwaysinvolvedingeneratingthesereactionsareillustratedinFigure111Whena
photosensitisermolecule is irradiatedwith lightofanappropriatewavelength itcan
undergoanelectronictransitiontoformthesingletexcitedstatewithpairedelectron
spinsThemoleculetheneitherundergoeselectronicdecayandreturnstotheground
stateortheenergycanbetransferredsothatthemoleculeundergoesanelectronic
transitiontothetripletexcitedstateTheelectronspinsatthispointareunpairedThe
67
molecule could once again lose the energy depending on the environmental
conditions and the structure of the molecule itself and return to the ground state
Alternatively ifoxygen ispresent theenergycouldbetransferredandusedtodrive
redoxreactionsandgenerateradicalions(typeI)ortogeneratesingletoxygen(typeII
reaction) Themajor pathway involved in generating the bactericidal effect in PDI is
the production of singlet oxygen (Wakayama et al 1980) To be an efficient
photosensitiseramoleculemustbeefficientatproducingsingletoxygenandthat in
turn isdependentonthegenerationofa largepopulationof long‐livedmolecules in
thetripletstate(Wainwright1998)
68
Figure111FlowdiagramtodemonstratethegenerationofsingletoxygenTheboldarrows indicate the pathway to the Type II reaction (Bonnett 1995 Wainwright1998)
The reactiveoxygen species‐drivenbactericidaleffect is similar to thatgeneratedby
TiO2 thin films upon irradiation with suitable wavelengths of light Singlet oxygen
speciesexertadirecteffectonmicrobialcellsbyoxidisingcellconstituentssuchasthe
cellwall cellmembrane or intracellular components such as nucleic acidswith the
cytoplasmicmembraneastheprimarytargetPDIcausesalossofmembraneintegrity
suchthattheintracellularcontentsleakoutofthecellcontrolledtransportofsolutes
across themembrane is compromised and the cell loses viability due to the lack of
essential constituentsneeded foranabolicandcatabolicpathways (Jorietal 2006)
69
The reactiveoxygen speciesare thenable toaccess the intracellularDNAandcause
further damage (Dunipace et al 1992 Salmon‐Divon et al 2004 Chi et al 2010)
Singlet oxygen has a diffusion distance of approximately 20 nm therefore if the
bacterial species are in contactwith the light‐activatedmaterial then the generated
singlet oxygen should be active against both the bacterial cell wall and underlying
membrane
Anadvantageous featureofPDI is thatmulti‐drug resistantstrainsofbacteriawhich
are resistant to a number of different antibiotic classes do not show enhanced
resistancetoPDIcomparedwiththeequivalentantibioticsensitivestrains(Maliketal
1990) The susceptibility of 60 multi‐drug resistant strains of P aeruginosa to the
photosensitiser toluidine blue and red laser light were comparedwith 19 antibiotic
sensitivestrainsandnodifference insusceptibilitywasobserved(Tsengetal2009)
InadditionthegrowthphaseofPaeruginosadoesnotimpactonitssusceptibilityto
TBO‐mediatedphotosensitisation(KomerikandWilson2002)unlikesomeclassesof
antibioticswhichhaveselectiveactivityforbacteriaintheexponentialphaseofgrowth
(Tuomanenetal 1986)Duetothemulti‐siteactivityofthereactiveoxygenspecies
generated during light irradiation it is unlikely that resistant phenotypes will be
selected(HamblinandHasan2004)
1411 Typesofphotosensitisers
There are a number of different aromatic compounds which can act as
photosensitiserswhenirradiatedbyspecificwavelengthsoflightThecompoundsare
usually coloured as they reflect light in the visible part of the electromagnetic
spectrum An ideal photosensitiser would contain an overall cationic charge as
70
bacterial cells carry an overall anionic charge because of the presence of the
cytoplasmic membrane (Hamblin and Hasan 2004) Examples of photosensitisers
whichhavebeenusedforPDIarethephenothiazinestoluidineblue(Wakayamaetal
1980Paardekooperetal1992Wainwrightetal1997Pernietal2009bRagaset
al 2010) and methylene blue (Decraene et al 2009 Perni et al 2009a) the
halogenated xanthene rose bengal (Decraene et al 2006) and acridines such as
acridineorange(Wainwrightetal1997)
Photosensitiserscanbeusedinsolutionandappliedtothetreatmentareaorcanbe
impregnatedintoapolymerwhichcanbeusedinavarietyofsettingsForexamplea
solution of photosensitiser can be injected into a periodontal pocket before the
applicationof laserlighttoexertPDIonthepathogenspresent(Wilson19931996)
Alternatively the photosensitiser could be immobilised in a polymer used in as a
cathetermaterialsothatanybacteriapresentinthelumenorexteriorofthetubing
would be exposed to the reactive oxygen species generated during PDI upon
applicationofthelightsource(Pernietal2011)
15 Methodsofproducinglight‐activatedantimicrobialmaterials
151 Chemicalvapourdeposition
Thin films of TiO2 are commonly synthesised using the chemical vapour deposition
(CVD)technique indeed itisthemethodusedindustriallybyPilkingtontosynthesise
theirPilkingtonActivtradeself‐cleaningglasses(Millsetal2003)Thedepositionprocess
requiresheatingtoahightemperature(gt500degC)thereforethechoiceofsubstrateis
limited as the substrate has to withstand the rise in temperature this constraint
71
makesglassan idealchoicePrecursormoleculescontainingtitaniumandoxygenare
heated into a gaseous phase and transported via the nitrogen carrier gas into the
reaction chamber The precursormolecules are adsorbed onto the heated substrate
anddecompose theelementsof choice remainadhered to the substrateandwaste
productsareremovedfromthesystembythenitrogencarriergas(West1999Carp
etal2004Page2009)AschematicofatypicalCVDrigisdisplayedinFigure112
Figure112Schematic representationofaCVDapparatusThe setupshown in thisdiagram was used to deposit thin films of titanium oxynitride as discussed inChapter4(Aikenetal2010)
152 Sol‐gel
The sol‐gel technique is considered to be more reproducible than CVD and the
production of a uniform film is possible on a small scale (Carp et al 2004) To
synthesiseTiO2 thin filmsby the sol‐gelmethodahomogenous solution isprepared
containing thecationic reactants required for the synthesis analkoxide isusedasa
72
sourceofTiO2waterisrequiredtohydrolysethealkoxideandanalcoholisaddedto
catalyse the reaction (West 1999 Rampaul et al 2003 Page 2009) A viscous gel
develops containing colloidal particleswhich grows further as the solution is left to
age During this time the water and alcohol trapped in the matrix of the polymer
evaporate and so the resultant aged sol is transparent and homogenous with no
crystallinephasesorprecipitatesTheglasssubstratecanthenbedippedintothesol
andthesoladherestothesurfaceoftheglassitisremovedataconstantratesothat
thethinfilmproducedisofaconsistentthicknessalongthelengthofthematerialThe
sol dries readily but is mechanically weak so is sintered at a high temperature to
removeanyorganicmatterandadensecrystallineoxidecoatingisproduced
153 Swellencapsulation
Swell encapsulation is a chemical method used to impregnate polymers with an
organic compound and can be modified to add a photosensitiser molecule to a
polymer in order to generate a light‐activated antibacterial material When an
elastomer is immersed in an organic solution containing a photosensitiser the
photosensitiserisabletopenetratethepolymerastheelastomericmatrixswellsThe
elastomer is removed from the photosensitiser‐containing solution after a defined
periodandthepolymerrevertsbacktoitsoriginalsizeasthesolventevaporatesThe
photosensitiserremainsembeddedintheelastomericmatrixduringevaporationand
thefinalconcentrationofphotosensitisercanbeadjustedbyvaryingtheconcentration
intheorganicsolution(Pernietal2009aPernietal2011)
73
16 Measuringenvironmentalcontamination
Accuratemethodsarerequiredtomonitormicrobialcontaminationofenvironmental
surfacestoassesscleaningregimensandtodetectanybacteriapresent (Manheimer
andYbanez1917SaloandWirtanen1999MooreandGriffith2002Verranetal
2002Hedinetal2010Verranetal2010a)
161 Swabbing
Bacterial culture is a widely used method as any viable bacteria present can be
detected quantified and identified at a relatively low cost The test surface can be
sampled using a swab or spatula which can be made from a variety of materials
includingcotton viscosenylon orman‐madesubstances suchas thebrush‐textured
nylon flock Samples can then either be streaked directly onto an agar plate or re‐
suspendedintoagrowthenhancingbrothbeforesubcultureontosolidmedia(Moore
andGriffith2007) If thebacterial inoculum ishigh thesamplecanbeserialdiluted
before plating out to allow enumeration of the single colonies on the culture plate
ensuring a more accurate estimation of the original bacterial inoculum Pathogenic
yeastsandfungicanalsobedetectedinthiswayHoweverthetechniquereliesupon
theabilityof the swab to collectallmicrobial contaminationon the surfaceand the
releaseoftheorganismsfromtheswabheadduringprocessing(Faveroetal1968)
162 Dipslides
Environmental surfaces can alternatively be directly sampled by placing a section of
agar directly onto the surface by use of a RODAC (replicate organismdetection and
counting)plateorasimilarsamplingdeviceandenumerationofthecoloniesafteran
74
incubation period Dipslides have a greater sensitivity and reproducibility compared
with swabbingwithout enrichment culturewhen sampling surfaces especially if the
surface isdry (Mooreetal2001MooreandGriffith2002FoodStandardsAgency
2004Obeeetal2007)Howeverquantificationcanbedifficultifthesurfacelevelof
contamination is too high as the microbial load on the surface cannot be diluted
resulting in confluentgrowth on theagarwhichmakes colonycounting impractical
Growth is instead classified instead as moderate or heavy based on the surface
coverageoftheslideandcomparisonwithvisualimagesofcontrols
163 Airsampling
Air sampling devices are used to sample the microbial contamination of the
surroundingairAdefinedvolumeofairisdrawnintothedeviceandispassedoveran
agar plate so that microorganisms found in the air are inoculated onto the plate
surfaceAirbornesporesarealso inoculatedontotheplatesandgrowthoccursafter
germination These units have been employed in the healthcare environment to
monitor efficiency of cleaning schedules and terminal decontamination regimens
(Jeanesetal2005Wongetal2011)thefungalcontaminationofairduringbuilding
work(Goodleyetal1994)andthequalityofairinoperatingtheatres(Whyteetal
1982Hambraeus1988Landrinetal2005)Ariskfactorforsurgicalsiteinfectionsis
microbial contamination of the air in operating theatres so knowledge of the air
quality isessential toensureairhandlingunitsare functioningcorrectlyandprevent
theseinfectionsoccurring(Whyteetal1982Hambraeus1988Whyteetal1992)
Microbialcontaminationoftheaircanalsobemonitoredusingsettleplateswhichare
large agar plates that are placed in the test environment Airbornemicro‐organisms
75
which fall onto the plates are then detected by colony counting after incubation
However droplet nuclei stay suspended in the air so cannot be detected using this
methodandtheplatesrequire longerperiodsofsampling(circa24hours)compared
withamechanicaldevicethattakesminutestoobtainasample
164 ATPbioluminescence
All of themethodsdescribedabovehave thedisadvantage that theyaredependent
upontheabilityoforganismstogrowonsolidmediasobacteriaintheviablebutnon‐
cultivable (VBNC) state would not be cultured Alternative sampling methods that
overcometheselimitationswouldthereforebeuseful(MooreandGriffith2007)ATP
bioluminescence is a process based upon a naturally occurring light‐generating
reactionfoundintheNorthAmericanfireflyPhotinuspyralis(HawronskyjandHolah
1997) Both themale and female fireflies use the generation of light to locate one
anotherandasmatingsignals(EncyclopediaBritannica2011)Theluciferaseenzyme
isolated from P pyralis can be used in the laboratory to catalyse the oxidation of
luciferinusingATPastheenergysourceandthereactionisasfollows
ATP+D‐luciferin+O2 AMP+PPi+oxyluciferin+CO2+light
The light produced during the reaction can be quantified by a luminometer and the
output is given in relative light units (RLU) (Lundin 2000) The generated light is
directlyproportionaltotheamountofATPpresentintheinitialsampleasonephoton
oflightisemittedpermoleculeofATP
luciferase
76
ATP is found inall living organismsand isalsopresentas freeATP (Hawronskyjand
Holah1997)Luminometerscanbeusedtoprovidedataontheleveloforganicdebris
andmicrobialcontaminationonasurface(Davidsonetal1999)EukaryoticATPand
ATPfromextracellularsourcescanbedegradedpriortothelysisofthebacterialcells
withcertainmodels (HawronskyjandHolah1997)enablingthenumberofbacterial
cellstobecalculatedfromtheamountoflightemittedResultscanbeavailablefrom
fivetothirtyminuteseliminatingthetime‐consumingovernightincubationofculture
plates
ATP bioluminescence has been used for the last decade in the food industry and is
especiallyusefulincomplyingwithspecificfoodregulationswhichservetoreducethe
riskoffoodspoilageandcontamination(HawronskyjandHolah1997Davidsonetal
1999Wagenvoortetal2000)Qualitativemeasurementsareusuallytakensothata
surfacewill eitherpass if anacceptablenumberofbacteriaarepresentor fail if the
numberofbacteria is aboveapredetermined level (Cooperetal 2007)Theuseof
ATPbioluminescenceinthesesituationsisadvantageousastheresultsareavailablein
minutes so if the surface contaminationwas deemed too high then it could be re‐
cleanedre‐testedandfoodproductioncouldcontinueifitsubsequentlypassed
ThereareanumberofcommerciallyavailableluminometersincludingtheClean‐Trace
(BioTraceBridgendUK)aportableluminometerwhichdetectsATPbioluminescence
ofbothmicrobialandnon‐microbialoriginThissystemiscommonlyusedtoassessthe
effectiveness of cleaning regimens as organic debris is also detected The easily
transportable BioProbe (Hughes Whitlock Gwent UK) and the Junior (Berthold
TechnologiesGmbHBadBadwildGermany)luminometersrequireadditionalreagents
77
to generate RLU readings as does the Lumat luminometer (Berthold Technologies
GmbH) The Microbial ATP Kit (BioThema AB Sweden) can be used to degrade
exogenousATPbefore thebacterial cells are lysed soamoreaccurate indication of
theactualnumberofbacteriapresentonthetestsurfacecanbeobtained(BioThema
AB2006)Thesemethodologiesarenot commonlyused in thehealthcareora food
environmentastheyrequireasamplepreparationstepandtakeslightlylonger(upto
30 minutes) These methodologies can be used for molecular experiments such as
reporter gene assays where a higher sensitivity is required (Dyer et al 2000
McKeatingetal2004BioThemaAB2006)
165 Stainingtechniques
Staining techniques could alternatively be used to estimate the level of bacterial
contaminationonasurfaceAcridineorangeisacommonlyuseddyeusedtoperform
direct counts on test surfaces although no indication of bacterial viability is given
Fluorescentprobessuchascyanoditolyltetrazoliumchloride(CTC)andrhodamine123
canbeusedasindicatorsofcellviabilityCTCisreducedtocrystallineCTC‐formazan
present as red crystals within bacterial cells and rhodamine 123 is concentrated in
functioningmitochondriaandcellsfluorescegreen(YuandMcFeters1994Pyleetal
1995)Visualisationrequirestheuseofappropriateexcitationandemissionfiltersona
fluorescentmicroscope(YuandMcFeters1994)TheLiveDeadBacLighttradeBacterial
Viability stain (Molecular Probes Inc) is a fluorescent dye which can differentiate
betweenviableandnon‐viablebacterialcellsThekitcontainstwodyesSYTO9and
propidiumiodideSYTO9emitsat500nmandstainsallcellsgreenwhereaspropidium
iodide is a red stain that emits at 635 nmand penetrates cellswith a damaged cell
78
membrane(Boulosetal1999AireyandVerran2007)Allgeneratedimagescanbe
capturedonacameraattachedtoafluorescentmicroscopetoenableenumerationof
the organisms present using computer software such as ImageJ
(httprsbwebnihgovijindexhtml) Direct visualisation techniques can also detect
thepresenceofnon‐microbialcontaminationsuchasorganicsoil thatcouldprovide
sustenanceforbacterialgrowth(Verranetal2002)
166 Summaryofenvironmentalsamplingtechniques
Thereiscurrentlynostandardisedtechniqueforsamplingenvironmentalsurfacesina
hospital environment so a variety of methods are used (Hedin et al 2010) ATP
bioluminescence provides a snapshot of bacterial contamination and can detect the
presence of organic soil Viable bacteria can be enumerated by performing viable
counts which is cheap and easy to perform and improvements in the swab head
material and sampling diluent have been shown to increase sampling efficiency
althoughtheimprovementsobservedwereminimal(Hedinetal2010)Visualisation
techniquesrequiremorespecialisedequipmentandstainsbutintactbiofilmscanbe
observedwithoutdisruptionandnon‐viablebacteria included in thebacterial count
Thesetechniquesallpossessinherentadvantagesanddisadvantagessoarebestused
with clear knowledge of these limitations especially when interpreting any data
generated(Verranetal2010a)
79
17 Methodsof characterisingandassessing the functionalityof light‐
activatedantimicrobialmaterials
171 UV‐visible‐IRspectroscopy
UV‐visible‐IRspectroscopycanbeusedtopredictthelikelyphotocatalyticactivityofa
potentialantibacterialmaterialbycalculatingthebandonset(Section13313)When
incidentlightwithawavelengthbetween200nmand700nmisappliedtoacandidate
materialthreereadingscanbetaken(i)thetransmissionoflightthroughthesample
(ii) the absorption of light by the sample and (iii) the reflectance of light from the
sampleThesereadingscanbeusedtoestimatethebandgapAplotof(αhv)12against
hv isthengeneratedwherehvequalstheincidentlightandaequalstheabsorbance
coefficient(a=‐logTT0whereTequalsthetransmissionreadingofthesampleandT0
equals the transmission of the substrate)When thecurve isextrapolatedalong the
linearportionofthecurvethebandgapcanbereadfromthexaxis(Tauc19681970
Sharmaetal2009)ThisiscalledaTaucplotThetransmissiondatacanalsobeused
tocalculatethethicknessofthethin filmsusingtheSwanepoelmethod (Swanepoel
1983)
172 Photooxidationofstearicacid
Thephotodegradationoftheorganicmoleculestearicacid(Figure113)canbeusedto
quantify the photocatalytic self‐cleaning ability of candidate antibacterial materials
andisbasedonthefollowingequation(Millsetal2002)
CH3(CH2)16CO2H+26O2 18CO2+18H2Ohvgebandgapenergyofsemiconductor
80
Carbondioxideandwater isgeneratedfromorganicmolecules inacoldcombustion
reaction(ParkinandPalgrave2005)Theprocessisrelativelysimpletoperformandso
a large number of thin films can be screened for potential photocatalytic activity
Infrared (IR) spectroscopy is used to monitor the degradation of the stearic acid
molecules The thin films that show the greatest activity by this method can then
selectedforantibacterialtesting
Figure113ChemicalstructureofstearicacidC18H36O2
Infraredspectroscopyisananalyticalmethodusedtoobservethevibrationalenergies
of molecular bonds Photons of light from the IR portion of the electromagnetic
spectrum interact withmolecular bondswithin the sample The incident light has a
lower frequency than UV or visible light and causes molecular bonds to bend and
stretchastheyabsorblightAbsorptionofthephotonofIRlightcausesanincreasein
thevibrationalenergyofthebondraising it toahighervibrationalenergy levelThe
modeofvibrationvariesdependingupontheconstituentatomsinthebondandthese
chemicalstretchesandbendsareidentifiableontheIRspectragenerated(McCarthy
1997)
TheIRmeasurementsareplottedonagraphofwavenumberagainsttransmittanceor
absorption The changes in the vibrational energies of the molecular bonds are
detected as inverted peaks on the resultant IR spectra as the transmittance of the
incident light decreases because of the absorbance of the light by the molecular
81
bondsTheseinvertedpeaksaretermedabsorptionbandsandarecharacteristicofthe
IR vibrations of specific molecular bonds Stearic acid has three modes which are
visibleintheIRspectrumthesymmetricCndashHstretch(CH2)hasanabsorbanceband
of2923 cm‐1 theCndashH stretch (CH3)hasanabsorbancebandof 2958cm‐1and the
asymmetric C ndash H stretch (CH2) has an absorbance band of 2853 cm‐1 The
concentrationofstearicacidcanbeapproximatedbyintegratingtheareaofthelatter
twopeaks the firstpeak isof low intensityand is generallynotusedAn integrated
areaof1cm‐1equatestoapproximately97x1015molecules(MillsandWang2006)
andsothedestructionofstearicacidcanbemonitoredovertimebynormalisingthe
concentrationofstearicacidmoleculesonthetestsurfaceasCxC0readingswhere
C0istheinitialconcentrationandCxistheconcentrationofstearicacidatagiventime
point
173 Contactanglemeasurements
Photo‐inducedsuperhydrophilicitycanbeinducedonphotocatalyticthinfilmssuchas
TiO2 after irradiationwith light possessing band gap energy (Mills et al 2002) The
hydrophilicity or indeed hydrophobicity of a substrate can be calculated by
determiningthecontactangleofadropletofwaterinoculatedontothesurfaceofthe
materialAhydrophilicmaterialwillpossessalowwatercontactangleasthedroplet
will spread flat on the lsquowater‐lovingrsquo hydroxylated surface with an accompanying
increase in the diameter of the droplet Conversely a hydrophobicmaterialwill not
have an affinity for the droplet of water so the diameter of the droplet will be
reduced resulting in a highwater contact angle (Page 2009)Hydrophobic surfaces
82
havewatercontactanglesabove90deghydrophobicsurfaceshavewatercontactangles
below90degandsuperhydrophilicsurfaceshavewatercontactanglesapproaching0deg
During photo‐induced superhydrophilicity on a TiO2 semiconductor light exposure
causes the trapping of holes at lattice sites near the surface of thematerial and a
concomitant reduction of Ti4+ to Ti3+ (Carp et al 2004) The bonds between the
titanium and oxygen within the lattice are weakened by the trapped holes which
enable the release of oxygen atomswhich in turn creates oxygenvacancies and an
increaseinthehydroxylationstateofthesurfaceHydroxylgroupsareadsorbedonto
thesurfacewhichbindwiththewaterinoculatedontothesurfaceduetoanincrease
inthevanderWaalsforcesandhydrogenbonding(Carpetal2004)
174 Standardmethodsofassessment
International standards have been developed to assess the activity of novel
antimicrobial products such as the Japanese Industrial Standard JIS Z 2801 which
measuresantibacterialactivityandefficiencyandnumerousISOstandardsdeveloped
by the International Organisation for Standardisation (International Organisation for
Standardisation 2011) Antibacterial activity can be calculated using the following
formula R = log(BC) where R is a measure of the antibacterial activity B is the
averagenumberofviablecellsofbacteriaonanuntreatedsampleafter24hoursand
Cistheaveragenumberofviablebacteriaontheantibacterialsampleafter24hours
If a test sample has a value of greater than 20 then it is denoted an antibacterial
materialaccordingtoJISZ28012006
83
The methylene blue reduction test can also be used for the assessment of
photocatalytic surfaces and has recently been adopted as an ISO standard (ISO
106782010)Whenmethyleneblueisinoculatedontoatestsurfacephotogenerated
electronsreduceatmosphericoxygentoproducesuperoxidewhichdegradesthedye
or photogenerated holes either directly oxidisemethylene blue or generate reactive
oxygenspeciesthatdirectlyattackthedye(AthertonandNewlander1977Zitaetal
2009)These reactions result inadecrease in the intensityof thecolouration of the
dye and this colour change can be monitored on a spectrophotometer over time
comparedwithanuntreatedcontrolsampletodeterminetheabilityofUV‐activated
surfaces to photodegrade dissolved organic molecules Therefore this would be a
useful tool toscreena largenumberofdifferentphotocatalystsbeforefocusingona
smallernumberofsamplestotestagainstbacterialsuspensionsHowevertheassayis
notvalidatedtouseonsurfacesactivatedbyvisible lightoragainstbacterial targets
AcidOrange7isanotherdyethatisoxidisedduringphotocatalysisanddegradationof
themoleculecanbemonitoredasamethodofdeterminingphotocatalyticactivityA
morerecentdevelopmentistheuseofaninkResazurinwhichisdescribedasafaster
and simpler method (Mills andMcGrady 2008) During photocatalysis the positive
holes generated are trapped by glucose which is containedwithin the preparation
and thephotogeneratedelectrons reduceResazurin (Zitaetal 2009)Thecolourof
theinkchangesfrombluetopinkwhichoccursinsecondscomparedwiththehours
requiredfortheformermethodsandthecolourchangecanbedetectedbyeyewhich
providesaninexpensivesemi‐quantitativemeasureofphotocatalyticactivity
84
18 Overviewandprojectaims
A multi‐disciplinary approach is required to prevent HCAIs as the acquisition and
transmissionofinfectionisrarelycausedbyanisolatedeventbutasaconsequenceof
anumberoffailuresinprocedure(Dettenkoferetal2011)Handhygieneisviewedas
themost important and effectivemethod for preventing the transmission ofHCAIs
Adequate isolation facilities need to be available and high‐risk patients need to be
transferred into these areas promptly This requires sensitive specific and rapid
detection of the infective organisms so that these scarce resources are used
appropriately (Cheng et al 2011) Prudent antibiotic prescribing is important to
preventtheemergenceofresistantorganismsandhasbeenshowntoreducetherates
ofCdifficile infection (Mearsetal 2009)The patientenvironment shouldbekept
free of pathogens by methods as basic as regular scheduled cleaning and hand
decontamination after each patient contact This has been shown to significantly
reduce the transmission of microorganisms and prevents the transfer of organisms
from patient‐to‐patient and from the environment‐to‐patient (Devine et al 2001
Rampling et al 2001 Dancer 2004 Johnston et al 2006 Department of Health
2008Danceretal2009)Novel technologiescouldalsobeemployedaspartofthe
armoury of interventions used to prevent the transmission of infectious
microorganismswithinhospitalsascurrentlyemployedmethodssuchascleaningand
handhygienealonearenotprovingtobesufficient(Ramplingetal2001Frenchet
al 2004) Recontamination of surfaces occurs readily after disinfection of areas
surrounding an infected patientwhich allows further transmission of the organisms
(Collins1988WeberandRutala1997Bradyetal2003)Self‐cleaningsurfacescould
potentially lower the bacterial load in the near‐patient environment and reduce re‐
85
colonisation rates as organisms shed in‐between cleaning events would be killed
breakingthecycleofre‐colonisationAntimicrobialpolymerscouldbeusedtoproduce
ETTsandcatheters to reduce theadherenceof bacteriawithin the lumenof tubing
andpotentiallydecreasetheincidenceofdevice‐relatedHCAIs
Thepurposeofthisprojectwastogenerateandassesstheantibacterialactivityofa
rangeoflight‐activatedmaterialswiththepotentialtobeusedinahealthcaresetting
toreducethetransmissionandacquisitionofHCAIs
86
2 Materialsandmethods
21 Targetorganisms
Bacterial typestrainsused inthesestudiesare listed inTable21Allof thebacterial
strainswerestoredat‐80degCinbrainheartinfusionbroth(BHI)containing10glycerol
andmaintainedbyweeklysubcultureonto5Columbiabloodagarplates (allmedia
fromOxoidLtdBasingstokeUK)AclinicalisolateofCalbicanswasalsoused(Table
21)andwas stored onaSabourauddextroseagar slopeat22degCandmaintainedby
weeklysubcultureontoSabourauddextroseagarplates
Table21Bacterialandfungalstrainsusedinthesestudies
Bacterialfungalstrain Referencenumber
Escherichiacoli ATCC25922
Staphylococcusaureus NCTC6571
Staphylococcusaureus ATCC8325‐4
Epidemicmeticillinresistant‐Staphylococcusaureus16 Clinicalisolate
Epidemicmeticillinresistant‐Staphylococcusaureus15 Clinicalisolate
Meticillinresistant‐Staphylococcusaureus ATCC43300
Streptococcuspyogenes ATCC12202
Enterococcusfaecalis Clinicalisolate
Pseudomonasaeruginosa PAO1
Pseudomonasaeruginosa Clinicalisolate
Acinetobacterbaumannii Clinicalisolate
Stenotrophomonasmaltophilia Clinicalisolate
Candidaalbicans Clinicalisolate
87
22 Growthconditions
Bacteria were grown aerobically in either nutrient broth (P aeruginosa E coli S
maltophiliaandAbaumannii)orBHIbroth(SaureusSpyogenesSepidermidisand
E faecalis) and incubated for 18 hours at 37degC in an orbital incubator (Sanyo BV
Loughborough UK) at a speed of 200 rpm C albicans was grown aerobically in
Sabourauddextroseliquidmediafor18hoursat37degCinanorbitalincubator
23 Preparationofthebacterialinoculum
A1mLaliquotoftheovernightculturewascentrifugedat12000rpmandthepellet
was re‐suspended in 1 mL phosphate buffered saline (PBS) (Oxoid Ltd) An optical
densityof005Aatawavelengthof600nmwasachievedbyaddinganaliquotofthe
re‐suspendedsolutionto10mLPBSwhichequatestoapproximately107cfumLFor
C albicansexperiments the entire re‐suspendedpelletwas added to 10mL PBS to
achieveanopticaldensityof1100Awhichcorrespondedtoapproximately107cfu
mL
ForexperimentsinvolvinganalginateswabthePBSwassubstitutedwith3mLCalgon
ringerrsquos solutionand for thoseusing LiveDead stains1mLbufferedpeptonewater
(BPW)wasused
24 Lightsources
241 Whitelightsource
Forwhite light photocatalysis experiments aGeneral Electric 28WBiax 2D compact
fluorescentlamp(GELightingLtdEnfieldUK)wasusedThislampiscommonlyfound
88
inUK hospitals and emits light across the visible spectrum the spectral distribution
chartisshowninFigure21Forexperimentalpurposesthelampwasaffixedinsidea
cooled incubator tomaintain a constant temperature of 22degC (LMS Series 1 Cooled
Incubator Model 303 LMS Ltd Sevenoaks Kent) The intensity of the light was
measured using a luxmeter (LX101 Luxmeter Lutron Electronic Enterprise Co Ltd
Taiwan) and readings were recorded in lux units The term visible light indicates
wavelengths of light in the visible portion of the electromagnetic spectrum namely
between 400 ndash 700nm however the terms white light and visible light are used
interchangeablyinthisthesisandindicateuseofthisfluorescentlightsource
Figure21Spectralpowerdistributiongraphforthelightsourceused inthevisiblelightphotocatalysisexperiments(Technicalpublicationforthe2Dserieslamp2005)
242 Ultraviolet(UV)lightsources
2421 365nmlightsource
For theUV light photocatalysis experiments aUV fluorescent lampwas used (Vilber
LourmatVL‐208BLB LeicestershireUK)The light sourceemitted lightprimarilyata
89
wavelength of 365 nmand the intensity of the lightwasmeasured using aUV light
meterSolarmeterModel50(SolartechIncHarrisonTownshipMichiganUSA)with
the readings recorded inmWcm‐2 Experimentswereconducted ina cabinet (Philip
HarrisLtdShenstoneUK)fittedwithaUVsafetyscreen
2422 254nmlightsource
AsecondUVlightsourcewasused(VilberLourmatVL‐208GVWRLtdLeicestershire
UK)eitherasamethodfordecontaminatingtheusedsamplesortoactivatetheTiO2
slidesbeforeexposuretothe365nmlightsourceThisgermicidalUVfluorescentlamp
emitted lightprimarilyatawavelengthof254nmExperimentswereconducted ina
cabinet(PhilipHarrisLtdShenstoneUK)fittedwithaUVsafetyscreen
243 Laserlightsource
AHeNelaserlightsource(ChangchunNewIndustriesOptoelectronicsTechCoLtd
Changchun China) was used for the photodynamic therapy experiments The light
sourceemitted lightprimarilyatawavelengthof660nmanda light intensityof230
mW
25 Generalsamplingmethodology
Asuspensionofbacteriacontaining107cfumLbacteriaasdescribedinSection23
wasdilutedtenfoldinPBStoproduceaseriesofbacterialconcentrationsrangingfrom
107 ‐ 104 cfu mL The standard volume of bacterial suspension used in these
experimentswas25microLwhichoccupiedanareaofapproximately1cm2uponthetest
samplesthereforethefinalbacterialpopulationrangedfrom25x105ndash25x102cfucm‐
90
2 A standard volume (25 microL) of bacterial suspension was inoculated onto a clean
microscope slide of dimensions 76 x 26 x 08 ndash10mm (length xwidth x thickness)
(VWR International Ltd Lutterworth UK) and was sampled using a cotton‐tipped
swab The surface was swabbed for 20 seconds in three directions with continual
rotation of the swabhead ina standardisedmanner before inoculation intoabijou
containing1mLofPBSThebijouwasvortexedfor2minutestoremovetheadherent
bacterialcellsandpriortopreparationoftenfoldserialdilutionsTwentymicrolitresof
eachdilutionwasplatedoutontoeitherMacConkeyagar forE coli ormannitol salt
agar for S aureus and the plates were incubated at 37degC for up to 48 hours The
aerobic colony count (ACC) was calculated by counting the resultant colonies to
determinethenumberofcolonyformingunitspersquarecentimetre(cfucm2)
26 ATPbioluminescence
AseriesofluminometerswereusedtomeasureATPbioluminescenceasanalternative
methodofdetectingandquantifyingbacteriafromthetestsurfacesAllluminometers
were programmed to capture luminescence readings every 1 second and themean
reading in relative light units (RLU)was reported after 10 seconds Test tubeswere
requiredforthedetectionofATPusingcertainmodelsofluminometerandtodestroy
any exogenous ATP before use theywere placed under the 254 nm germicidal UV
lamp(Section2422)for30minuteswithinsealedplasticbagsThebagwasinverted
atthehalfwaypointtoprovideevenexposuretothelightsource
91
261 Luminometer‐specificmethodologies
2611 Juniorluminometer
The cotton‐tipped swabwas added to a test tube containing 50 microL ATP Eliminating
Reagent from theMicrobialATPKit (BioThemaABHandenSweden)post sampling
Thetubewasincubatedfor10minutesatroomtemperatureaccordingtothereagent
kit instructions before 50 microL Extractant BS was added and the covered tube was
vortexedfor5secondstothoroughlymixthesolutionFourhundredmicrolitresofATP
ReagentHSwas finally added and the light generatedwas quantified by placing the
tubeintotheJuniorLB9509luminometer(BertholdTechnologiesGmbHampCoKGBad
WildbadGermany)AnATPstandardwasusedoneachrunand10microLofthepremixed
100nmolLATPstandardwasaddedtothefinalsolutionsothattheequivalentof1
pmolATPwasaddedtothetestsolutionTheATPbioluminescenceofthetestsample
plustheATPstandardwasthenquantifiedbytheJuniorluminometer
Foreachbacterialconcentrationonasurfacethreeindependentswabswereusedto
generate an ATP bioluminescence reading and one swab was used for ACC
measurements with each dilution plated out in duplicate Each experiment was
performedatleastintriplicatetodemonstratereproducibility
2612 Lumatluminometer
The Lumat LB9507 luminometer (Berthold Technologies GmbH amp Co KG) is a more
sensitivebutlessportablemodelthantheJuniorluminometerThemethodologyused
tomeasureATPbioluminescenceemittedfromtestsamples incombinationwiththe
Lumat luminometerwas as described for the Junior luminometer in Section 2611
92
with the exception that the test tubewas placed in the Lumat luminometer for the
bioluminescencereadings
2613 BioProbeluminometer
TheBioProbeluminometer(HughesWhitlockLtdGwentUK)wasusedincombination
withtheMicrobialATPKitasinthepreviouslydescribedmethodologiesHoweverthe
ATP bioluminescence generated from the bacterial suspension could be measured
directlyfromthetestsurfacesothereagentswereapplieddirectlytothetestsurfaces
andtheunnecessaryswabbingstagewasomittedInsteadtheBioProbeluminometer
wasplacedabovethetestsurfacecreatingasealbetweenthe inoculated laboratory
benchandtheluminometerTheluminescencegeneratedwasthenquantifiedbythe
BioProbeluminometer
2614 Clean‐TraceNGluminometer
TheMicrobialATPKitwasnotrequiredforthedetectionassayutilisingtheClean‐Trace
NG luminometer (3M Bracknell UK) This luminometer was designed for use with
custom‐made pre‐moistened swabs which after sampling in the standard manner
were returned to thecasingand immersed ina reagent solution locatedat itsbase
The entire swab casing was placed in the luminometer for quantification after
vortexingfor5secondsApositivecontrolwasusedoneveryrunThiswasa freeze‐
driedpowdercontaining5pmolATPwhichwassampledwiththepre‐moistenedswab
andhandledusingthesamemethodologyasthetestsamples
93
27 DirectvisualisationofbacteriandashLiveDeadstaining
Slideswereexaminedunderthefluorescentlightmicroscopepost‐samplingusingthe
LiveDeadBacLightBacterialViabilityKit (InvitrogenLtdPaisleyUK)tovisualiseany
remaining bacterial cells and to determine their viability The kit consisted of two
stains SYTO9tradewhichpenetrated themembranesofall cells andpropidium iodide
which penetrated bacterial cells with damaged membranes (Boulos et al 1999)
Viable cells appeared green under the fluorescent microscope whereas non‐viable
cellsgeneratedaredfluorescenceImageswerecapturedonacameraattachedtothe
microscopeandbacterialcellswereenumeratedandtheproportionofviableandnon‐
viable cells was noted The final bacterial populationwas compared to the starting
inoculumvaluetoevaluatetheefficiencyofthesamplingprocess
28 Effectofwhitelightonbacterialsurvival
Glass microscope slides were placed in a moisture chamber which was custom‐
designed topreventevaporationof thebacterial inoculumduringexposure towhite
light (Figure 22) Filter paper 150 mm in diameter (Whatman plc Maidstone UK)
soakedinsteriledistilledwaterwasusedtolinethebaseofasquare24cmx24cm
petridishWoodenstickswereplacedontopofthefilterpapertoresttheslideson
Anadditionalmoisturechamberwascovered infoiltopreventlightpenetrationand
slideswhichweretobeincubatedintheabsenceoflightwereplacedinthismoisture
chamber for the exposure period as a dark control The moisture chambers were
placedinthecooledincubatorandtheuncoveredchamberwasplacedonashelf20
cmfromthelightsourcewiththefoilcoveredchamberontheshelfdirectlybelow
94
Figure 22 Experimental set up of the moisture chamber used during white lightexperimentswhereA=woodenswabsB=glassslidesC=moistenedfilterpaperD=bacterialinoculum
Theeffectof thewhite light source on theviabilityofanumberofbacterial species
was investigated A suspension of bacteria was inoculated onto a microscope slide
priortoincubationunderthewhitelightsourcefor24hoursAnydecreaseintheACC
aftertheirradiationperiodwascalculatedasapercentageandlogreduction
29 Optimisationofthesamplingtechnique
To increasetheproportionofbacteriathatwererecovered fromthetestsurfacesa
seriesofexperimentswereperformedandasinglevariablewaschangedUncoated
cleanmicroscopeslideswereinoculatedwithasuspensionof25microLofaGram‐negative
bacterium(Ecoli)oraGram‐positivebacterium(Efaecalis)andtheneither
(i)sampledusingarangeofdifferentswabtypes
(ii) sampledwith a cotton swab and either vortexed or sonicated to remove
adherentbacteria
A
B
C
D
95
(iii)sampledwithuptothreedifferentcottonswabswhichwerere‐suspended
intothesamebijou
(iv)sampledwithuptothreedifferentcottonswabswhichwerere‐suspended
intoseparatebijoux
Total bacterial numberswere calculated by serially diluting the bacterial suspension
within thebijouand inoculatingduplicate20microLaliquotsonto 5bloodagar plates
TheACCwascalculatedafterupto48hoursgrowthat37degCtodeterminethecfumL
andthisvaluewascomparedwiththeACCrecoveredfromthestartinginoculum
210 Preparationoflight‐activatedantibacterialmaterials
2101 Thinfilmsgeneratedbychemicalvapourdeposition
Novelantibacterialthinfilmsweregeneratedbyoneoftwopost‐doctoralresearchers
based at the UCL Department of Chemistry The thin films were prepared by
atmospheric pressure chemical vapour deposition (APCVD) (Section 151) The
depositionswerecarriedoutontheSiO2surfaceofslidesofstandardfloatglassfrom
Pilkingtonofdimensions220x85x4mm(lengthxwidthxthickness)coatedonone
sidewithabarrierlayerofSiO2topreventiondiffusionfromtheglasstothefilmThe
glasswaswashedpriortoinsertionintotheAPCVDreactorusingsequentialwashings
ofwateracetonepetroleumether(60‐80)andpropan‐2‐olgivingacleanandsmear
freefinish
96
21011 Nitrogen‐containingtitaniathinfilmsTiON‐1andTiON‐2
The nitrogen containing thin films TiON‐1 and TiON‐2 were prepared by Dr Geoff
Hyett with anhydrous ammonia (BOC Ltd) as the nitrogen source titanium (IV)
chloride (TiCl4 999 Sigma‐Aldrich Ltd) as the titanium source and ethylacetate
(EtAc990BOCLtdGuildfordUK)astheoxygensource(Hyettetal2007Aiken
etal2010)Depositionswerecarriedoutat550degCfor60secondsandtheresulting
filmswerecutintosevenequallysizedsectionsof32mmx89mmoncecooled
AnitrogencarriergaswasusedfortheTiCl4andEtAcataflowrateof2LminThe
TiCl4bubblerwasheatedto61degCandtheEtAcbubblerto44degCataflowrateof05L
minwhichproducedamolarmassflowratioof12TheTiCl4andEtAcwerecarriedto
a singlemixing chamber through gas delivery lineswhichweremaintained at 200degC
andheatedto250degCwithanadditionalflowofnitrogencarriergasatarateof12L
min The glass substratewas dopedwith nitrogen by flowing ammoniawithout the
carrier gas through the reservoir at a flow rate of 026 L min The TiCl4 and EtAc
mixture and the ammonia gas were introduced just before contact with the glass
substrateandtheTiCl4EtAcammoniamassflowratiooftheresultantthinfilmwas
28541TheresultantthinfilmwasTiON‐1thetitaniumoxynitrideThinfilmTiON‐
2waspreparedusingthesamemethodologyexceptthedepositionwascarriedoutat
450degCinsteadof550degC
21012 Nitrogen‐dopedtitaniumdioxidethinfilmsN1N2andN3
ThenitrogencontainingthinfilmsN1N2andN3werepreparedbyDrCharlesDunnill
witht‐butylamine(995FisherScientificUKLtdLoughboroughUK)asthenitrogen
97
sourcetitanium(IV)chloride(TiCl4999Sigma‐AldrichLtd)asthetitaniumsource
andethylacetate(EtAc990BOCLtdGuildfordUK)astheoxygensource(Dunnill
et al 2009bDunnill et al 2009cDunnill and Parkin 2009) The resultant coatings
werenitrogen‐dopedtitaniumdioxide(N‐dopedTiO2)thinfilmsanddepositionswere
carriedoutat500degCfor30seconds
AnitrogencarriergaswasusedfortheTiCl4andEtAcwhichwaspreheatedto150degC
ataflowrateof05LminTheTiCl4bubblerwasheatedto70degCandtheEtAcbubbler
to 40degC which produced a molar mass flow ratio of 12 The TiCl4 and EtAc were
carried to a singlemixing chamber and heated to 250degC with an additional flow of
nitrogencarriergaspreheatedto150degCatarateof6LminTheglasssubstratewas
doped with nitrogen by flowing the carrier gas preheated to 60degC through the t‐
butylamine reservoir set at 5degC the temperature of the t‐butylamine reservoirwas
controlledusingawaterbathcontainingwaterandethyleneglycolinequalpartsThe
TiCl4andEtAcmixtureandthet‐butylaminegaswere introduced justbeforecontact
withtheglasssubstrateat100degCwithanadditional flowofcarriergasat1Lmin
TheTiCl4 EtAc t‐butylaminemassflowratiooftheresultantthin filmwas1 25
03Sectionsofthesamesheetofthegeneratedfilmweredivided into25x25cm
samplesoncecooledanddividedintothreegroupsrepresentingthinfilmsN1N2and
N3
21013 Sulfur‐dopedtitaniumdioxidethinfilms
Threesetsofsulfurcontainingthinfilms(S‐dopedTiO2)werepreparedbyDrCharles
Dunnillusingtitaniumtetrachloride(TiCl4Sigma‐AldrichLtd)asthetitaniumsource
ethylacetate(EtAc990BOCLtd)astheoxygensourceandcarbondisulfide(CS2
98
999AlfaAesarHeyshamUK)asthesulfursource(Dunnilletal2009a)Anitrogen
carriergaswasused for theTiCl4 andEtAcwhichwaspreheated to150degCata flow
rateof05LminTheTiCl4bubblerwasheatedto70degCandtheEtAcbubblerto40degC
whichproducedamolarmassflowratioof12TheTiCl4andEtAcwerecarriedtoa
singlemixingchamberandheatedto250degCwithanadditionalflowofnitrogencarrier
gas preheated to 150degC at a rate of 6 L min The glass substratewas dopedwith
sulfurbyflowingthecarriergaspreheatedto60degCthroughtheCS2reservoirsetata
temperaturebetween0and10degCthetemperatureoftheCS2reservoirwascontrolled
usingawaterbathcontainingwaterandethyleneglycol inequalpartsTheTiCl4and
EtAc mixture and the CS2 gas were introduced just before contact with the glass
substrate at 100degC with an additional flow of carrier gas at 1 L min Depositions
were carried out at 500degC for 30 seconds and three thin filmswere producedwith
different TiCl4 EtAc CS2 mass flow ratios which varied dependent upon the
temperatureoftheCS2reservoirduringsynthesis
(i) during synthesis of sample S1 the reservoir was set at 0degC generating a
massflowratioof12509
(ii) during synthesis of sample S2 the reservoir was set at 5degC generating a
massflowratioof12512
(iii)during synthesisof sampleS3 the reservoirwas setat10degCgeneratinga
massflowratioof12516
Theresultingfilmswerecutintosevenequallysizedsectionsof32mmx89mmonce
cooled
99
21014 Controlthinfilms
ThinfilmsofTiO2weresynthesisedusingAPCVDwiththesamesyntheticconditionsas
that described above but omitting the addition of the dopant (ie ammonia t‐
butylamine or carbon disulfide) Uncoated glass of the same size was used as an
additionalcontrol
2102 Thinfilmsgeneratedbysol‐geldeposition
Thesilver‐titaniathinfilmsweregeneratedinatwo‐stepprocess(Dunnilletal2011)
glassslideswereinitiallycoatedwithtitaniumdioxideandannealedbeforeacoating
ofsilvernitratewasadded
21021 Titaniumdioxidesolpreparationandthinfilmsynthesis
TheTiO2 solwaspreparedbyadding25246gofacetylacetone (002526mol99+
Sigma‐AldrichLtd) toa250mLglassbeakercontaining32cm3butan‐1‐ol (035mol
994 Sigma‐Aldrich Ltd) This produced a clear and colourless solution to which
1750 g titanium n‐butoxide (005 mol 970 Fluka) was added The solution was
vigorously stirred for 1 hour before 364 mL distilled water dissolved in 905 g
isopropanol (015 mol analytical grade Fisher Scientific) was added to the stirring
titanium n‐butoxide solution The yellow colouration of the sol deepened but
remained clear and itwas stirred for a further hour Lastly 166 g acetonitrile (004
mol99minFisonsScientificUKLtd)wasaddedtothesolutionanditwasstirredfor
an hour The deep yellow coloured sol was covered with parafilm and left to age
overnightinthedark
100
21022 Titaniumdioxidethinfilmsynthesis
On the following day clean single cavity ground glass slides (Jencons Scientific Ltd
EastGrinsteadUK)ofdimensions76x26x1mm (lengthxwidthxthickness)were
attachedtothedipcoatingapparatusinbatchesof4(Figure23)
Figure23ThedippingapparatususedtoproduceaxerogelonthemicroscopeslidesPhotographreproducedwithpermissionfromDrKristopherPage
Thecavityslideswereloweredintoaglassbeakercontainingtheagedsolandafter20
secondsthecavityslideswerewithdrawnbytheapparatusatasteadyrateof120cm
min The first coat was allowed to dry before the process was repeated The
deposited xerogel films required calcination in order to adhere the coating to the
cavityslideandtobecomecrystallineThereforethecoatedcavityslideswereplaced
insideamufflefurnaceandfiredat500degCfor1hourwithaheatingrateof10degCmin
101
andacoolingrateof60degCminThethinfilmswerethenleftinthefurnaceovernight
to cool and stored in a dark container until required The resultant coatings are
referredtoasTiO2thinfilms
21023 Silver‐titaniumdioxidethinfilmsynthesis
Asolutionofsilvernitratewaspreparedbyadding042gsilvernitrateto500mLof
methanol(bothFisherScientificUKLtd)toproduceafinalconcentrationof5x10‐3mol
dm3TheTiO2thin filmswereattachedtothedipcoatingapparatusdipped inthe
silvernitrate solutionandwithdrawnata rateof120cm minThe thin filmswere
thenexposedtothe254nmUVlampfor5hourswithinacustommadelightboxand
were stored in the dark for at least 72 hours before bacteriological testing
Photodepositionoccursquickly (lt30min)butanexcessoftimewasusedtoremove
the time of irradiation as a variable and ensure that the filmswere fully clean and
activatedpriortoinitialcharacterizationTheresultantcoatingsarereferredtoAg‐TiO2
thinfilms
2103 ToluidineBlueO‐containingpolymersgeneratedbyswellencapsulation
Toluidine Blue O (TBO) was incorporated into polyurethane polymers by swell
encapsulation To achieve this 125mg of TBOwas added to 25mL distilledwater
beforetheadditionof225mLacetoneforminga91ratioofacetonetodistilledwater
(H2O10vv)Thesolutionwasplaced inasonicatingwaterbathfor15minutesto
ensure the TBO was evenly distributed throughout the suspension To prevent
interaction of the solution with light the container was covered in foil during
sonicationTenmillilitrealiquotsof theTBOsolutionwasdispensed intoglass screw
102
capped bottles and a 1 cm2 square of polyurethane was added The bottles were
stored horizontally in the dark for 24 hours The polyurethane squares were then
removedandlaidtodryonapapertowelandcoveredfor1hourAfterthistimethey
wererinsedwithsteriledistilledwateruntiltheexcessTBOadheredtothesurfaceof
the polymers had detached and thewater remained clear The polymerswere then
driedandstoredinthedarkforafurther24hoursbeforeuseBatchesof24polymers
were made and control polymers were also prepared without the addition of TBO
(Pernietal2009b)
211 Characterisation and functional assessment of light‐activated
antibacterialmaterials
2111 UV‐visible‐IRspectroscopy
UV‐visible‐IR spectroscopy was employed to determine the band onset of the thin
filmsandassessthe likelyphotocatalyticactivityofthematerialThethinfilmswere
decontaminated by exposure to the 254 nm germicidal UV lamp for 12 hours and
storedinthedarkfor72hoursThethinfilmwasthenplacedinsidetheUV‐Visible‐IR
spectrophotometer (Perkin Elmer λ950 Massachusetts USA) and percentage
transmission readings were measured which were converted to absorption and
absorbanceusingthereflectancetogaugethethicknessofthefilmsbytheSwanepoel
method(Swanepoel1983)DataweretransformedandaTaucPlotwasgeneratedto
determinetheopticalbandgapofthethinfilmsbyextrapolatingthe linearcurveto
theabscissaATaucplotcanbecalculatedusingtheformula(axhv)12againstenergy
whereadenotes the absorbance of thematerial andhvdenotes the energy of the
103
photon of light (Tauc 1968 1970) Measurements were also taken of the titanium
dioxidethinfilmanduncoatedglassslidesothatthereadingscouldbecompared
2112 Contactanglemeasurements
Water droplet contact angles were measured to determine the potential photo‐
induced hydrophilicity of the thin films The thin films were decontaminated by
exposuretothe254nmgermicidalUVlampfor12hoursandstoredinthedarkfor72
hours A FTA 1000 droplet analyserwas used tomeasure the diameter of a 86 microL
dropletofdeionisedwaterinoculatedontothethinfilmusingasidemountedcamera
The dropwas formed and dispensed by gravity from the tip of a gauge 27 needle
Readings were taken before and after irradiation with UV light (Section 2421) or
filteredwhitelight(Section241)between200and2500nmAnuncoatedglassslide
and titanium dioxide thin film were used as controls Results were entered into a
computer programme to calculate the contact angles based upon the volume‐
diameterdataAnaverageof5readingsweretakenateachexposuretimesothatthe
resultsobtainedwerereproducible
2113 Photooxidationofstearicacid
Thestearicacidtestwasusedtoquantifythephotocatalyticactivityofthethinfilmsas
a preliminary indicator of their potential antibacterial activity The destruction of
stearicacidwasmeasuredbyFourierTransform InfraredSpectroscopy (FTIR)usinga
PerkinElmerSpectrumRX1FTIRspectrometer
The thin filmsweredecontaminatedbyexposure to the254nmgermicidalUV lamp
for12hoursandstoredinthedarkfor72hoursThethinfilmswerethenattachedto
104
an IR sample holder comprised of a sheet of aluminiumwith a circular hole in the
centre beforea 10 microL dropofa001Msolutionof stearicacid inmethanol (Fisher
ScientificUKLtd)wasappliedtotheexposedportionofthethinfilmAcharacteristic
white smearwas observed once the droplet had evaporated and the sampleswere
thenstoredonceagain inthedarkforat72hourspriortothebaselinereadingat0
hours FTIR spectrawereobtained for the stearicacid layerbetween2800and3000
cm‐1 andanuncoatedglass slidewasusedasacontrol for thebackground readings
Baselinereadings(C0)weretakenofthethinfilmsandblankcontrolsthenallsamples
were placed in the custom‐made light box and were exposed to the light source
Readings (Cx)were takenat24hour intervalsand the sampleswere returned to the
lightboxaftereachreadingForeachtimepointtheareaofthepeakswereintegrated
andthevaluescombinedtogiveanapproximateconcentrationofstearicacidonthe
surfacewhere1cmminus1intheintegratedareabetween2700and3000cmminus1corresponds
to approximately 97times1015 molecules cm2 (Mills and Wang 2006) A graph was
plottedofthenormalisedconcentrationofstearicaciddetectedonthesurface(CxC0)
againsttimewhichallowedthedestructionofstearicacidtobeobserved
The light sources were attached to the lids of the custom‐made light boxes which
were suspended 25 cm from the surface of the thin films Three lighting conditions
wereexaminedaUVlightsource(Section2422)awhitelightsource(Section241)
andthewhitelightsourcefittedwithaUVfilterTheUVfilterusedwasa3mmthick
sheet of Optivextrade glass which is described to cut off all radiation below 400nm
(InstrumentGlasses2000)Thefilterwaspositioned1cmabovethesamplesandwas
setupsuchthatalllightarrivingatthesampleshadpassedthroughthefilter
105
212 Microbiological assessment of light‐activated antimicrobial
materials
2121 Decontaminationofthethinfilms
Priortomicrobiologicalassessmentcoatedsamplesweresoakedin70isopropanol
for 30 minutes to kill and remove any adherent contaminants rinsed with fresh
isopropanolandair‐driedThesampleswerethen incubated inahotairoven(Weiss
Gellenkamp oven BS Leicestershire UK) for 1 hour at 160degC to kill any residual
organisms and stored in the dark until required This process was repeated after
microbiologicalassessmentinpreparationforfurthertesting
The decontamination procedure was later amended and after microbiological
assessment the slides were rinsed with sterile distilled water and air‐dried before
exposuretothe254nmgermicidalUV lamp(Section2422) for18hourstokillany
remainingadherentorganismsTheslideswerethenplacedinthedarktoreversethe
activating effect of theUV light Sampleswere then ready for re‐use after 72 hours
dark storage Thin films were re‐used due to the lack in reproducibility of the
depositionmethod
2122 Measuringtheeffectof lightonthethinfilmsgeneratedbyAPCVDor
sol‐gel
Thethinfilmswereplacedina24x24cmpetridishlid20cmfromthelightsourcefor
theactivationstep(designatedA+)forthedesiredtimeperiodThethinfilmswerenot
coveredduringthislightexposureperiodAsacontrolduplicatethinfilmswerealso
106
placed inthecabinetbutwithina foil‐encased24x24cmpetridishtopreventlight
penetration(designatedA‐)
Thethinfilmswerethenpositionedwithinthemoisturechamber(Figure24)beforea
25microLdropletofbacterialsuspensionwasaddedThelidwasaddedtopreventdroplet
evaporation and the moisture chamber was placed under the light source at a
distanceof20cmfortheirradiationstep(designatedL+)andexposedforthedesired
periodoftimebeforesamplingControlduplicatethin filmswere incubatedwithina
foil‐encasedmoisturechamberduringthewhitelightexposureperiod(designatedL‐)
ThenomenclatureusedforthelightexposureexperimentsissummarisedinTable22
Figure 24 Irradiation of the nitrogen‐doped thin films to white light with thesamplesplacedwithinthecustomdesignedmoisturechamber
107
Table22Nomenclatureusedduringmicrobiologicalassessmentofthethinfilms
Nomenclature Description
A+L+Sample exposed to first light dose bacterial droplet addedthensampleexposedtosecondlightdose
A‐L+Sample stored in the dark bacterial droplet added thensampleexposedtosecondlightdose
A+L‐Sample exposed to first light dose bacterial droplet addedthensamplestoredinthedark
A‐L‐Sample stored in the dark bacterial droplet added thensamplestoredinthedark
Bacteria were recovered by sampling the thin films as described in Section 25
Experiments were performed in at least duplicate and repeated on a minimum of
threeseparateoccasionsforeachtypeofthinfilmandexposuretime
2123 Measuring the effect of light on Toluidine Blue O‐impregnated
polymersgeneratedbyswellencapsulation
Newly synthesised polymers (described in Section 2103) were used for each
experimentandwerediscardedaftereachuseApolymerwasplacedinawellwithina
6‐wellmicrotitreplatebeforea25microLdropletofthemicrobialsuspensionwasadded
Aglasscoverslipwascarefullyplacedontoptospreadthedropletevenlyacrossthe
surfaceofthepolymerandtheplatewastransferredtoaraisedplatform24cmfrom
thelaserlightsourceThelightemittedfromthelaserpassedthroughabeamdiffuser
tospreadthelightbeamsothattheentirepolymerwasexposedtothelaserlightand
thepolymerwasexposedtothelaserlightfortherequiredperiodoftime
108
Oncetheexposuretimehadendedthecoverslipwasasepticallyremovedandplaced
insidea50mLtubecontaining135microLPBSA10microLaliquotofthemicrobialdroplet
wasremovedfromthepolymerandinoculateddirectlyontoanappropriateagarplate
andspreadusinganL‐shapedspreaderTheremaining15microLofmicrobialsuspension
was recovered placed in the 50mL tube and briefly vortexed before tenfold serial
dilutions were prepared Twenty microlitres of each dilution was inoculated and
spread onto an appropriate agar plate in duplicate As controls TBO‐containing
polymerswereinoculatedwiththemicrobialsuspensionforthesamelengthoftimein
the absence of laser light (L‐S+) or polymers preparedwithout the addition of TBO
were inoculated with the microbial suspension and exposed to identical periods of
laser light (L+S‐) or not exposed to the laser light (L‐S‐) The sampling process was
repeated three times for each polymer type and exposure time and the entire
experimentwasrepeatedonatleastthreeseparateoccasionsforeachorganismand
exposuretime(Pernietal2009b)
213 Statisticalanalysis
Inordertodeterminethesignificanceofanydecreases inthecfuobservedbetween
the light‐activated antibacterialmaterials exposed to different conditions theMann
WhitneyUtestwasusedThenumberofsurvivorsrecoveredfromthetestgroup(ie
thelight‐activatedmaterialexposedtolight)wascomparedtothenumberofsurvivors
fromthecontrolgroups(ie the light‐activatedmaterialsnotexposedto lightorthe
uncoated samples)Median valueswere taken because the datawere not normally
distributedand thevalueswere transformed to log10 fornormalisationAp valueof
less than 005 was considered statistically significant Statistical significance is
109
diagrammaticallyrepresentedontheboxandwhiskerplots intheresultssectionsas
asterisksoneasteriskdenotesapvaluelt005twoasterisksdenotesapvaluelt001
andthreeasterisksdenotesapvaluelt0001Allstatisticalanalyseswereperformed
usingtheSPSSstatisticalpackage(version160SPSSIncChicagoILUSA)
110
3 Developmentofprotocolsusedtoassesstheactivityofthephotocatalyticthinfilms
31 Introduction
The purpose of the work described in this chapter was to develop a reproducible
method of testing the antibacterial photocatalytic activity of thin films Initially the
sampling technique was examined to determine the sampling efficiency and an
optimised regimen was developed Researchers from our laboratory had previously
used swabs (Page et al 2007) to remove bacteria from the test surface in order to
detectchangesinthebacterialconcentrationpost‐exposuretoantibacterialcoatings
Othergroupshaveuseddipslidesasadirectdetectionmethodbutthisisunsuitable
for accurately quantifying high concentrations of bacteria as it results in confluent
growth which only generates an estimate of the bacterial load The recovery of
bacteriafromglasssurfaceswasinitiallycomparedusingarangeofswabswithswab
headscomprisedofdifferentmaterialsusingadifferingnumberofswabspersample
and using sonication as a method of releasing bacterial cells from the swab head
There are however inherent problems with swabbing as bacteria are either left
behindonthesurfaceafterswabbingorgetcaughtwithinthemeshoftheswabhead
andarenotreleasedintothediluentaftersampling(Davidsonetal1999)
Antimicrobial coatings are generally assessed using the viable count technique and
bacterialsurvivalisdeterminedbycountingcoloniesoriginatingfrom(i)serialdilutions
ofthebacterialsuspensiononthecoating(Wilson2003Decraeneetal2006Page
et al 2007) (ii) those grown on an agar overlay applied to the entire coating
(Decraeneetal2008b)(iii)serialdilutionsofthebacterialsuspensionaftertheentire
111
coating has been immersed in a sterile fluid and agitated to remove adherent
organisms(Decraeneetal2008a)or(iv)acombinationofthese(Pernietal2009a)
These techniques have proven to be effective at determining the activity of novel
antimicrobial coatingsbut the turnaround time for results is around48hours soan
alternativefastermethodisstilldesirable
ATPbioluminescencehasbeenusedasarapiddiagnostictesttodetectbacteriafrom
urinesamples(Selanetal1992)andmorerecentlyhasbeenappliedinthehospital
environment to rapidly assess the efficiency of cleaning regimens in hospitals as
described in Section 164 (Griffith et al 2000 Malik et al 2003 Dancer 2004
Ayciceketal2006Griffithetal2007Willisetal2007Lewisetal2008Boyceet
al2009Mulveyetal2011)followingonfromthesuccessfuluseofthismethodin
the food industry for the monitoring of surface cleanliness (Poulis et al 1993
HawronskyjandHolah1997Ayciceketal2006)Thecleanlinessofasurfacecanbe
rapidlyassessedand if the levelofATP isaboveanacceptable level thenthesurface
canbere‐cleanedandretested
ATPbioluminescenceutilisesthefirefly luciferaseenzymetocatalysetheconversion
ofATPintoAMPresultingintheemissionoflight(Lundin2000)Theamountoflight
emitted is quantified by a luminometer and is directly proportional to the initial
amountofATPinthesampleIftheeukaryoticATPisremovedfromthesurfacebefore
sampling then this value is in turn proportional to the amount of bacteria in the
startingsampleasonephotonoflightisgeneratedpermoleculeofATPForthisstudy
themethodwasevaluatedforitspotentialuseasatooltoassesstheeffectivenessof
novel antibacterial coatings by quantifying bacteria present on a surface before and
112
after light exposure The generation of quantitative data especially at low bacterial
concentrationswouldbeusefulanditwaspostulatedthatATPbioluminescencecould
supersede swabbing as the first choice for bacterial detection from surfaces in this
project
Alsoassessedinthischapterwastheeffectoftheincidentlightsourceonthesurvival
ofbacteriaCertainspecificwavelengthsofwhite lightareknownto inactivatesome
Gram‐positivestrainsofbacteria(Macleanetal20082009)soitwasimportanttobe
aware of the effect of the light source used to activate the novel thin films Any
decreaseinthebacterialconcentrationcouldthenbeattributedtotheactivityofthe
thinfilmsandnottoincidentlightsource
32 Materialsandmethods
321 Optimisationofthesamplingtechnique
BacterialstrainsweremaintainedasdescribedinSection21andbacterialsuspensions
ofEcoliandEfaecaliswerepreparedasdetailedinSection23resultinginastarting
inoculumofapproximately107cfumlAnumberofstrategieswereemployed inan
attempttoimprovebacterialrecoveryfromthesurfaceofuncoatedmicroscopeslides
as described in Section 29 Three different cotton swabs were used (all Fisher
ScientificUKLtd)woodstickcottontippedswabs‐CottonAcottonswabssterilisedby
ethyleneoxide‐CottonBandcottonswabssterilisedbyUVlight‐CottonCAlginate
andviscoseswabswerealsousedinthecomparison
113
322 ATPbioluminescence
BacterialstrainsweremaintainedasdescribedinSection21andbacterialsuspensions
ofEcoliandSaureuswerepreparedasdetailedinSection23resultinginastarting
inoculum of approximately 107 cfu ml ATP bioluminescence was used to detect
bacteria inoculated onto the surface of uncoated microscope slides as described in
Section261Anumberofcommercialluminometerswereusedwithoutputgivenin
relativelightunits(RLU)andtheamountofATPpresentinthesampleswascalculated
usingthefollowingformula(HughesWhitlockLtd1995)
ATPsample=RLUsample(RLUsample+standardndashRLUsample)
The number of bacteria present in each sample was then calculated based on
previously documented studies which estimate that each bacterial cell contains
approximately2x10‐18molATP(Lundin2000BioThemaAB2006)Itwasimportantto
determinetheinitialamountsofATPpresentasotherwisetheRLUreadingsobtained
fromdifferent luminometerscouldnotbedirectlycompared(HawronskyjandHolah
1997)Toassessthesensitivityoftheassayusingeach instrumentone‐tailedt‐tests
were performed where the sensitivity was the lowest concentration that was
significantlydifferentfromthenegativecontrolwith95confidenceThecoefficient
ofvariation(CV)wascalculatedasapercentageforeachdilutiontodemonstratethe
reproducibilityofeach luminometerwheregreater reproducibility is representedby
lower CV values particularly below 100 (Griffith et al 1994) The luminometer‐
specific methodologies were assessed to determine the precision accuracy and
sensitivityofeachassayusingthedefinitionsdescribedinTable31
114
Table 31 Definitions of the terms used to compare the luminometer‐specificmethodologies
Parameter Definition
PrecisionA measure of the reproducibility of the luminometer‐specificmethodAssessedbycalculatingthecoefficientofvariation(CV)
SensitivityThe lowest concentrationofbacteria that is significantlydifferenttothenegativecontrolAssessedbyperformingone‐tailedt‐tests
AccuracyHow close the value generated by the luminometer‐specificmethod is to the true value Assessed by comparison with theinoculumlevelestimatedbyviablecolonycount
323 Measuringtheeffectofwhitelightonbacterialsurvival
BacterialstrainsweremaintainedasdescribedinSection21andbacterialsuspensions
of S aureus NCTC 6571 E coli ATCC 25922 E faecalis S pyogenes ATCC 12202
EMRSA‐16 EMRSA‐15 MRSA 43300 S aureus NCTC 8325‐4 and S epidermidis 01
were prepared as detailed in Section 23 resulting in a starting inoculum of
approximately 107 cfu ml equating to approximately 25 x 105 cfu sample The
effectofthewhitelightontheviabilityofbacterialstrainswasdeterminedusingthe
methodologydescribedinSection28andFigure22TheMannWhitneytestwasused
to determine the statistical significance of any differences observed as described in
Section213
115
33 Results
331 Optimisationofthesamplingtechnique
The use of different swabs during sampling did not result in a notable increase in
bacterial recovery (Figure 31) the greatest recovery of E coli and E faecaliswas
observedusing thealginate swabbut there remaineda973and 992 respective
loss compared with the starting inoculum Recovery of E coli and E faecalis using
cottonswabCresultedina989and996lossofbacteriarespectivelyandtheuse
ofcottonswabAresultedina989and997lossofbacteriarespectivelyOverall
recoveryofEcoliwasbetterthanrecoveryofEfaecalis
Figure31ComparisonofdifferentswabtypestoincreasetherecoveryofEcoliandEfaecalisTheuseofanyoftheswabtypesresultedinalossofmorethan97ofbacteriaduringtheswabbingprocessBarsindicatemeanvalues(n=8)anderrorbarsrepresentstandarddeviations
116
ThereforeEcoliwasusedtoassessfurtherimprovementsinthesamplingtechnique
withcottonswabASonicatingtheswabsaftersamplingthesurfacedidnotresultina
greater recoveryofE colinor did theuseofmore than one swab (Figure32) The
methodwhichresultedinthegreatestrecoveryofbacteriawasthe2‐swabin1bijou
methodbuttherewasstilla98differencebetweenthestartingconcentrationofE
coli and the concentration recovered All nine methods tested resulted in losses of
morethan98ofEcoliThereforethe1‐swabtechniquewithcottonswabAanda
120secondvortexwasusedforallsubsequentexperimentsThedifferenceinrecovery
betweenthevarioustechniqueswasnotsubstantialandthechosenmethodwasthe
leastlabourintensiveandmostcosteffective
Figure32ComparisonofdifferentsamplingmethodsusedtoincreasetherecoveryofEcoliAllsamplingmethodstrialledresultedinlossesofmorethan98ofEcoliBarsindicatemeanvalues(n=8)anderrorbarsrepresentstandarddeviations
117
332 ATPbioluminescence
3321 Saureus
Themost accurate prediction of the concentration ofS aureuswas producedwhen
the BioProbe luminometer was used to detect ATP bioluminescence a starting
inoculumof 25x105 cm2was reportedas67x105 cm2 (Figure 33)However the
highest dilutions of bacteriawere not always detected andwere falsely reported as
negativewhichresultedinlargestandarddeviationsandacoefficientofvariation(CV)
of over 100 for the lowest concentration of bacteria (Table 32) Furthermore the
methodology was not the most sensitive the calculated sensitivity of the BioProbe
assaywas 25x104 cm2 (plt005)whichmeant that lower bacterial concentrations
couldnotbedifferentiatedfromthenegativecontrolAnaccurateestimateoftheS
aureus concentrationwas also producedwhen the Junior luminometerwas used to
detectATPbioluminescenceHoweveratthelowesttestconcentrationthevariance
ofthedatawasverylargewhichsimilarlyresultedinaCVvalueabove100
118
Figure33Comparisonofthefivedifferentmethodsemployedforthedetectionofsurface‐associated S aureus Data points represent mean values and error barsrepresentstandarddeviations(Aikenetal2011)
Table 32 Reproducibility of the ATP bioluminescence assay using the fourluminometerstodetectSaureusdisplayedascoefficientsofvariation(CV)wherealower CV represents a greater reproducibility All values are expressed aspercentagesThesensitivityofeachassayismarkedwithanasterisk
cfucm2
SaureusLumat Junior BioProbe
Clean‐Trace
25x105 16 62 52 21
25x104 20 64 70 29
25x103 27 51 62 35
25x102 44 158 137 133
The most precise estimate of the bacterial load on the test surface was generated
when the Lumat luminometer was used to detect ATP bioluminescence (p lt001)
whereprecisionisanindicationofthereproducibilityofthemethodThepresenceof
119
25x102cm2(thelowestdilutionfactortested)ofSaureuswasconsistentlydetected
(Figure33)and low levelsofbacteriawerenotmisreportedasnegativewhichwas
confirmedbythelowCVvaluesobtained(Table32)foralldilutionfactorsHowever
theaccuracyofthedevicewaspoorasthedetectedconcentrationofbacteriawasat
leastafactorof10lowerthantheinoculumaddedtothetestsurface
When the Clean‐Trace luminometer was used to detect ATP bioluminescence an
inaccurate result was always generated although the data produced was always
reproducibleTheconcentrationofSaureuswasunderestimatedbyalmostafactorof
10 at each dilution factor At low bacterial concentrations an absence of ATP was
commonlyreportedresultinginlargestandarddeviationsandaCVvalueover100at
thelowestbacterialconcentration
Reproducible estimateswere obtained using the viable countmethod however the
bacterial loadwasunderestimatedbyuptoa factorof10andwas lowerthanthose
values generated by the ATP bioluminescence assays using the BioProbe or Junior
luminometersA largevariation in thevaluesobtainedathigher concentrationswas
alsoseenalthoughthepresenceofbacteriawasnevermisreported
3322 Ecoli
ThemostaccuratepredictionoftheconcentrationofEcoliwasproducedwhenthe
BioProbe luminometer was used to detect ATP bioluminescence and a starting
inoculumof 25x105 cm2was reportedas22x105 cm2 (Figure 34)However the
highest dilutions of bacteriawere not always detected andwere falsely reported as
negativewhich resulted in large standarddeviationsandCVvaluesofover100A
120
lessaccuratepredictionoftheconcentrationofEcolipresentonthetestsurfacewas
providedwhentheJunior luminometerwasusedtodetectATPbioluminescenceFor
examplewhen the starting inoculumwas 25x105 cm2 the bacterial concentration
was underestimated by a factor of 10 and at the lowest bacterial concentration no
bacteria were detected on any of the six replicates performed (Figure 34) The
reproducibilityoftheassaywaspoorwhichwasreflectedbythehighCVvaluesaCV
valueof0wasobtainedwhenthestarting inoculumwas25x102 cm2butthiswas
onlybecauseoftheinabilityoftheassaytodetectthepresenceofEcoli
Figure34Comparisonofthefivedifferentmethodsemployedforthedetectionofsurface‐associated E coli Data points represent mean values and error barsrepresentstandarddeviations(Aikenetal2011)
121
Table 33 Reproducibility of the ATP bioluminescence assay using the fourluminometers to detect E coli displayed as coefficients of variation (CV)where alower CV represents a greater reproducibility All values are expressed aspercentagesThesensitivityofeachassayismarkedwithanasterisk
cfucm2Ecoli
Lumat Junior BioProbeClean‐Trace
25x105 14 85 52 32
25x104 23 67 32 36
25x103 15 254 58 54
25x102 13 0 98 104
ThemostsensitiveandreproducibleestimateofthenumberofEcolipresentonthe
test surface was generated when the Lumat luminometer was used to detect ATP
bioluminescence (Figure 34) Low levels of bacteria were always detected and not
misreportedasnegativeand therewasvery little variationobserved in the readings
generatedwhichwasconfirmedbythe lowCVvaluesobtainedforallconcentrations
ofbacteriatested(Table33)Howevertheaccuracyoftheestimatewaspooraswas
alsoseenintheSaureusassayandthedetectedconcentrationofbacteriawasatleast
afactorof10lowerthantheinoculumlevelForexamplejust74x103cm2ofEcoli
wasdetectedbythismethodwhenthestartinginoculumwas25x105cm2
When the Clean‐Trace luminometer was used to detect ATP bioluminescence an
accuratepredictionoftheconcentrationofEcoliatthelowestdilutionswasprovided
(Figure 34) However there was little differentiation between the highest two
dilutionsofbacteriatestedForexampleastartingconcentrationofEcoliof25x103
cm2 was reported as 34x102 cm2 and a starting concentration of 25x102 cm2
122
reported as 17x102 cm2 and this problem was compounded by the fact that the
highestdilutionsofeitherbacteriawerenotalwaysdetectedandthusfalselyreported
asnegativeresultinginlargestandarddeviationsandCVvaluesofover100
Theviable countmethodwassuperiortoallothermethodsforEcolidetectionFor
examplewhenthestarting inoculumofEcoliwaseither25x105 cm2or25x102
cm2 respective concentrations of 11x105 cm2 and 14x102 cm2 were obtained
(Figure 34) The presence of bacteria was always reported even at low
concentrationswhichwasnotshownforalltheluminometer‐basedmethods
333 Measuringtheeffectofwhitelightonbacterialsurvival
3331 Comparisonof4bacterialstrainsonaglasssubstrate
White lightwasobservedtohaveanantibacterialeffecton the survivalofSaureus
NCTC6571 onaglass surface (Figure 35)After24hoursexposure towhite light a
statisticallysignificantreductioninviableorganismswasseen(56log10cfusample)
comparedwiththecontrolconditionswithoutwhitelightexposureThemediancount
wasbelowthedetectionlimitoftheassaybuttherewasawiderangeincountsand
valuesbetween0and47log10cfusamplewereobtained(plt0001)
White light did not have an effect on the survival of E coliATCC 25922 on a glass
surface (Figure36)After24hoursexposure towhite light anegligible reduction in
viableorganismswasseen(02log10cfusample)comparedwiththecontrolsample
which was not exposed to white light Although when the data were statistically
analysedahighly significantdifference in countswasobserved thiswasdue to the
123
very smallerrorbars in this seriesofexperimentsattributed to the little variation in
counts obtained on each experimental repeat Such small differenceswould not be
consideredmicrobiologicallydifferent
log 10cfusample
Exposureconditions
log 10cfusample
Exposureconditions
Figure35EffectofthewhitelightsourceonthesurvivalofSaureusNCTC6571onaglasssurfaceA25microlbacterialsuspensionwas inoculatedontoaglassslidebeforeexposetowhite lightfor24hours(L+n=29)Asacontrol inoculatedglassslideswerealsoincubatedinthedarkfor24hours(L‐)Thethickhorizontallinesindicatemedianvaluesthebaseandtopofeachboxrepresentsthe25and75quartilesrespectivelyandtheerrorbars the10and90percentilesandthesmallcirclesareoutliersThedottedhorizontallineindicatesthedetectionlimitofthesamplingmethod14log10cfusample
124
log 10cfusample
Exposureconditions
log 10cfusample
Exposureconditions
Figure36EffectofthewhitelightsourceonthesurvivalofEcoliATCC25922onaglasssurfaceSampleswereexposedtowhitelightfor24hoursafteradditionof25microlofabacterialinoculum(n=10)
TheeffectofwhitelightonthesurvivalofEfaecalisonaglasssurfacecanbeseenin
Figure37After24hoursexposure towhite light a smallbut statistically significant
reduction in viable organismswas seen (01 log10 cfu sample) comparedwith the
controlsamplethatwasnotexposedtowhite light(plt005)Awiderange incounts
was obtained with values between 22 and 54 log10 cfu sample observed on the
surfaceexposedtolight
125
log 10cfusample
Exposureconditions
log 10cfusample
Exposureconditions
Figure 37 Effect of thewhite light source on the survival ofE faecalis on a glasssurfaceSampleswereexposedtowhitelightfor24hoursafteradditionof25microlofabacterialinoculum(n=6)
White lightwasalsoobservedtohaveaneffectonthesurvivalofSpyogenesATCC
12202 inoculatedontoaglasssurface(Figure38)After24hoursexposuretowhite
lighta13 log10cfusamplereduction inviableorganismswasseencomparedwith
thecontrolconditionswithoutwhitelightexposurewhichwasstatisticallysignificant
(plt005)Therewasawiderangeincountsandvaluesbetween0and45log10cfu
samplewereobtained
126
Figure38EffectofthewhitelightsourceonthesurvivalofSpyogenesATCC12202onaglasssurfaceSampleswereexposedtowhitelightfor24hoursafteradditionof25microlofabacterialinoculum(n=4)
3332 ComparisonofSaureusstrainsonaglasssubstrate
Thedata collected in the previous sections suggested thatSaureusNCTC6571was
particularlysusceptibletothewhitelightusedforthisseriesofexperimentssoitwas
decided toexamineotherSaureus strains to seewhether theyshare this increased
sensitivity towhite light inactivation This was particularly important as it would be
usefultoassesstheactivityofthelight‐activatedantimicrobialcoatingsagainststrains
ofSaureus especially theepidemic strainsEMRSA‐15and EMRSA‐16because they
areacommoncauseofHCAIstheyhavebeenthepredominantcirculatingstrainsof
MRSAintheUKandarecitedasthecauseofmorethan95ofMRSAbacteraemias
(Johnsonetal2001Ellingtonetal2010)
AreductionintherecoveryofbothEMRSA‐16(Figure39)andEMRSA‐15(Figure310)
wasseenfromtheglasssurfacesexposedtothewhitelightsourcecomparedtothat
127
recoveredfromthesurfacesnotexposedtowhitelightTheobservedreductionswere
statistically significantandwere09 log10 cfu sampleand15 log10 cfu sample for
EMRSA‐16 (p lt001) and EMRSA‐15 (plt001) respectively indicating that EMRSA‐16
waslesssusceptibletothewhitelightcomparedwithEMRSA‐15
WhitelightwasobservedtohaveamuchgreatereffectonthesurvivalofMRSA43300
inoculatedontoaglasssubstrate(Figure311)After24hoursexposuretowhitelight
a statistically significant reduction in viable organisms was seen (46 log10 cfu
sample) compared with the control conditions without white light exposure The
mediancountwasbelowthedetectionlimitoftheassaybuttherewasawiderangein
countsandvaluesbetween0and46log10cfusamplewereobtainedTheseresults
were similar to thoseobservedafterSaureusNCTC6571wasexposed to the same
lightconditions(Figure35)
Exposureconditions
log 10cfusample
Exposureconditions
log 10cfusample
Figure 39 Effect of thewhite light source on the survival of EMRSA‐16on a glasssurfaceSampleswereexposedtowhitelightfor24hoursafteradditionof25microlofabacterialinoculum(n=8)
128
Exposureconditions
log 10cfusample
Exposureconditions
log 10cfusample
Figure310Effectof thewhite light sourceon the survivalofEMRSA‐15onaglasssurfaceSampleswereexposedtowhitelightfor24hoursafteradditionof25microlofabacterialinoculum(n=12)
Exposureconditions
log 10cfusample
Exposureconditions
log 10cfusample
Figure311EffectofthewhitelightsourceonthesurvivalofMRSA43300onaglasssurfaceSampleswereexposedtowhitelightfor24hoursafteradditionof25microlofabacterialinoculum(L‐n=10L+n=12)
129
Theeffectofwhite lightonthesurvivalofSaureusNCTC8325‐4 isshown inFigure
312A33log10cfusamplereductioninbacterialcountwasobservedcomparedwith
thecontrolgroupwhichwasnotexposedtowhitelightandthisreductionwashighly
statistically significant The survival of S aureus NCTC 8325‐4 also appeared to be
affectedbytheexperimentalsetupasareductionintherecoveryofbacteriafromthe
control groupwas seen whichwas also statistically significant at the 01 level S
aureusNCTC8325‐4appearedtobeslightlymoretoleranttotheeffectsofthewhite
lightcomparedwithSaureusNCTC6571(Figure35)
Exposureconditions
log 10cfusample
Exposureconditions
log 10cfusample
Figure312EffectofthewhitelightsourceonthesurvivalofSaureusNCTC8325‐4onaglasssurfaceSampleswereexposedtowhitelightfor24hoursafteradditionof25microlofabacterialinoculum(n=8)
130
Table34SummaryofresultsfromtheseriesofexperimentsexaminingtheeffectofwhitelightonbacterialsurvivalDataareexpressedasmedianvalues
BacterialstrainReductioninbacterialrecovery
(log10cfusample)
SaureusNCTC6571 56
EcoliATCC25922 02
Efaecalis 01
SpyogenesATCC12202 13
EMRSA‐16 09
EMRSA‐15 15
MRSA43300 46
SaureusNCTC8325‐4 33
34 Discussion
341 Optimisationofthesamplingtechnique
Accurateassessmentoftheactivityofthelightactivatedcoatingsisdependentupona
reliable and reproduciblemethod of detecting bacteria found on the surface of the
coatings both before and after light exposure (Verran et al 2010a) Therefore the
sampling technique used previously in this laboratory was examined to determine
whetheritcouldbefurtherimprovedDifferenttechniqueswereusedtomeasurethe
levelofmicrobialcontaminationonuncoatedsurfacesSwabsarethemostcommonly
used technique for measuring surface contamination but it has been well reported
that the rate of bacterial recovery using thismethod is poor (Davidson et al 1999
MooreandGriffith2007)Cotton‐tippedswabsareoftenusedbecausetheyabsorba
large volumeof the bacterial suspension on the surface so the surface appears dry
after sampling However bacteria become entangled within the meshwork of the
131
cellulose fibres of the swab head and are not readily released during vortexing
resulting in a low count during enumeration (Favero et al 1968) Viscose is a
derivativeof cottonsowouldbe likelytoabsorb liquidtothesamedegreeAlginate
swabshavebeenreportedtoimprovetherecoveryofbacteriafromsurfaces(Pageet
al2007)butthesedatashowthatthisimprovementwasnotsubstantialandthatthe
bacterial recovery was comparable to the other swab head materials Swab heads
comprisedofman‐madefibressuchasnylondonotretainliquidtothesamedegree
and so any organisms taken up by the swab are readily released However fewer
bacteria are taken up by the initial sampling event so a similarly low count is
generated(Davidsonetal1999)Detergentbasedsamplingsolutionshavebeenused
to increase sampling efficiency and could have been used instead of PBS in these
studiestoimprovebacterialrecovery(SaloandWirtanen1999)
Other factors to consider when interpreting data generated from viable counts are
thateachcolonyformingunitcountedonaplatedoesnotnecessarilycorrespondto
one bacterial cell as a clump of numerous cells will form one colony as will one
bacterial cell Light exposure causes bacterial stress which in turn causes bacterial
clumping and a concomitant reduction in the number of organisms recovered
Furthermore both the swabbing and vortexing processes used to remove adherent
organismsfromthesurfaceandswabheadrespectivelycandamagethe integrityof
thebacterialcellwallwhichwouldalsoresultinasmallernumberofviablecellsanda
lower viable count (Obee et al 2007) To detect the presence of residual
microorganisms remaining on the surface post‐sampling microscopic examination
132
could be employed and any remaining bacteria could be stainedwith a differential
viabilitystain(Verran2010Verranetal2010a)
342 ATPbioluminescence
Samplingasurfacewithaswabcangiveagoodindicationofthepresenceofbacteria
but does not provide an exact concentration of the bacteria present on the surface
(MooreandGriffith2007Verranetal2010a)Luminometersareusedfrequentlyin
thefoodindustry(Davidsonetal1999Storgardsetal1999)andincreasinglyinthe
healthcareprofession(Griffithetal2000Dancer2004Lewisetal2008)todetect
thepresenceofmicrobialcontaminationandorganicsoilFourdifferentluminometers
were tested as alternative sampling methods to swabbing and performing viable
counts
Previousstudieshaveshownthatitisnotpossibletodetectlownumbersofbacteria
fromatestsurfaceusingATPbioluminescence(Saloetal1999)specificallylt103cfu
cm2(Davidsonetal1999Mooreetal 2001MooreandGriffith2002) Improved
more sensitive luminometers such as the Lumat and the Junior were used in this
chapter inaddition toan improveddetection reagent thateliminatednon‐microbial
ATPandclaimedtobeabletodetectasfewasfivebacterialcells(BioThemaAB2006)
soanincreasedsensitivitywasexpected
However this study supports previous findings and has demonstrated that ATP
bioluminescencewasnotsuitableforaccuratelydetectingthenumberofbacteriaona
test surface over a range of concentrations (Aiken et al 2011) The methodology
utilising the BioProbewas able to detect higher concentrations of both E coli or S
133
aureus but no one method was able to reproducibly detect both organisms at all
bacterial concentrations At lower concentrations of bacteria the BioProbe‐based
assayeitherdidnotdetectthepresenceofbacteriaormadenodistinctionbetween
the suspensions containing 25x103 cm2 and 25x102 cm2 The BioProbe
methodologywaslikelytohaveproducedthebestresultsbecausetheinstrumentwas
specificallydesigned fordetectingbacteriadirectly froma flat surfaceHowever the
BioProbe is no longer commercially available so the use of this instrument was
unsuitable for future studies The methods employing the Junior Clean‐Trace and
LumatluminometersandindeedviablecountsallincorporateaswabbingstepForthe
organisms to be detected by these methods they therefore needed to be both
capturedbytheswabfromthetestsurfaceandreleasedfromtheswabheadintothe
diluentpriortoquantification(MooreandGriffith2002)whichlimitstherecoveryof
bacteriafromthesurface
TheLumat luminometerwasstatisticallythemostsensitivemodeltested(plt001at
25x102 cm2 for both E coliandS aureus)andproduced consistent data at every
dilution tested However the estimate although reproducible was not always
accurateandwasuptotenfoldlowerthanboththeknownconcentrationofbacteria
inoculated onto the test surface and the estimates made using alternative
luminometersThiswasdisappointingasunderoptimumconditionstheinstrumentis
abletodetect1amolATPwhichcorrespondstolessthanonebacterialcell(BioThema
AB2006BertholdTechnologiesGmbHampCoKG2007)The instrument isdesigned
forexperimentssuchasgenereporterassaysandluminescentimmunoassays(Dyeret
134
al2000McKeatingetal2004)andthisworksuggeststhatthepublishedsensitivity
cannotbetransferredtothequantificationofbacteriafromsurfaces
Inthepresent laboratorystudyacorrelationbetweencolonyformingunitsandRLU
wasmadebutithaspreviouslybeendifficulttodemonstrateahighdirectcorrelation
between these parameters outside of laboratory conditions because ATP
bioluminescence detects all ATP present on the sampled surface including organic
material of bacterial origin food residues human secretions and dirt (Poulis et al
1993)GenerallyofthetotalATPisolatedfromahandtouchsurface33ismicrobial
in origin therefore it is likely that theRLUvaluesobtainedwillbehigher than that
expectedifonlymicrobialATPwasdetected(Griffithetal2000)Howeveranumber
ofgroupshavedemonstratedacorrelationbetweentheseparameters
Selanetal(1992)usedATPbioluminescencetodetecturinarypathogensfromeither
bacterial culture or patient samples and employed the NRB Lumit PM kit At high
bacterialconcentrations(gt105cfuml)acorrelationbetweencfumlandRLUwas
observedwhere105cfumlEcolicorrespondedto10ndash500RLUand109cfumlE
coli corresponded to an RLU of around 87000 A statistically significant but low
correlationbetweencfumlandRLUvalueswasdemonstratedwhenthe3MClean‐
Trace ATP system was used to monitor the effectiveness of cleaning in a hospital
(Boyceetal2009)Othergroupshavedemonstratedaweakcorrelationbetweenthe
ATPscoreandmicrobialgrowthwhendifferentATPsystemswereusedtoassessthe
cleanlinessofhospitalwards(Ayciceketal2006Mulveyetal2011)Inaseparate
cleaning study sites which were considered unsatisfactory by ATP bioluminescence
werealsoshowntobeunsatisfactorybymicrobiologicalswabbing(Willisetal2007)
135
Articles in the literaturehavequestioned thevalue in correlating theaerobic colony
count and ATP bioluminescence RLU values because they measure different
parameterstheformermeasuresthenumberofviablemicroorganismsandthelatter
measurestheresidualorganicsoilwhichcouldbeofmicrobialornon‐microbialorigin
(Lewisetal2008) Inthischaptera relationshipbetweentheviablecountandATP
bioluminescence readings was sought and this was valid because the test surfaces
weredecontaminatedbeforeuse so itwasassumed thatno residualATP remained
Additionallythereagentkitthatwasusedcontainedan initialstepwhicheliminated
non‐microbialATPwhichfurtherincreasesthelikelihoodthatanyATPdetectedonthe
surfaceswasofbacterialoriginandnotfromanotherexogenoussourceHoweverthis
questionisperhapsinvalidwithinthecontextofassessingthecleanlinessofahospital
environment
An important limitation of ATP bioluminescence is that no information about the
bacterialspeciesisgiven(HawronskyjandHolah1997)Withinahospitalenvironment
itwouldbeadvantageoustodifferentiatebetweenbacterialspeciesforexamplethe
presenceofMRSAonapatientrsquosbed‐railwouldbeofmuchgreater interestclinically
thanthepresenceofcoagulase‐negativestaphylococcionthesamesurfaceMolecular
techniques such as the polymerase chain reaction (PCR) or culture‐based methods
wouldberequiredtospeciatethebacteriapresent
343 Theeffectofwhitelightonbacterialsurvival
Finally the effect of white light on the viability of a range of microorganisms was
investigated to ensure that any reduction in bacterial counts observed on the novel
136
lightactivatedthinfilmstobetestedwasattributeddirectlytotheintrinsicactivityof
thecoatingsandnotduetothelightexposureitselfWhenEcoliandEfaecaliswere
inoculated onto uncoated glass surfaces and then exposed to white light an
insubstantialreductionincellnumberwasobservedAreductionintherecoveryofE
coli has previously been observed after irradiation with 458 and 488 nm light
(Vermeulenetal2008)althoughazenonarclampwasusedwhichgenerateslightof
amuchgreaterintensityInterestinglythiswasnotthecasewithSaureusNCTC6571
An average reduction of 56 log10 cfu sample was observed on an uncoated glass
surfaceThiseffectwasalsoseentoa lesserextent inadifferentstrainofSaureus
ATCC 8325‐4 and an average reduction of 33 log10 cfu sample was observed S
aureusNCTC6571haspreviouslybeenshowntobeunaffectedby6hoursexposureto
the samewhite light source (Decraene et al 2006 2008b) implying that the killing
occursafteraprolongedirradiationtimeIndeedMacleanetal(2009)demonstrated
that longer exposure times were required for photoinactivation of certain bacterial
species suchasE coliandE faecalis Thisgroupandothers haveused lightwitha
wavelengthofbetween400ndash420nmtophotoinactivatearangeofbacterialspecies
(GuffeyandWilborn2006Macleanetal200820092010)
Themechanism of action is proposed to be due to photo‐excitation of endogenous
intracellularporphyrinsresultinginthegenerationofcytotoxicsingletoxygenspecies
(Hamblin and Hasan 2004 Lipovsky et al 2009) It is proposed that the observed
reductionsinbacterialviabilitydescribedinthesestudiesarelikelytobecausedbythe
samemechanismbutthishasnotbeeninvestigatedfurtherThevariationinbacterial
countsobserved in someof theexperiments couldalsobedue todifferences in the
137
intracellular concentration of porphyrins but the reason for this variation is unclear
(Hamblinetal2005)
InterestinglytheepidemicstrainsofMRSAdidnotshowthesamelevelofsensitivity
to the effect of the white light source EMRSA‐16 appears to show an increased
tolerancetotheinhibitoryeffectofthewhitelightsourcecomparedtoothertestedS
aureusstrainsasa09 log10cfusampledecreaseintherecoveryofEMRSA‐16was
seenafter24hoursexposuretothewhitelightcomparedwitha15log10cfusample
decrease when EMRSA‐15 was used and much greater reductions for meticillin‐
sensitivestrainsVariations inthesensitivityofSaureustotheeffectsofwhite light
hasbeendescribedpreviouslyandwasproposedthatthedifferencesinsusceptibility
were due to increased production of porphryns increased generation of reactive
oxygenspeciesanddecreasedproductionofcarotenoidsinthelight‐sensitivestrains
(Lipovskyetal2009)Amutationcouldbepresentinepidemicstrainswhichconfers
increasedtolerancetowhite lightbyoverproductionofthecarotenoidsantioxidants
ordecreasedproductionofporphyrinsAmplificationofthegenesflankingeitherthe
S aureus‐specific porphyrin coproporphyrin or golden pigment carotenoid and
sequencingofthePCRproductcouldconfirmthishypothesis
The observed decreased susceptibility to white light could contribute towards the
persistence of epidemic strains such as EMRSA‐16 in the hospital environment
ThereforewhenchoosinganepidemicMRSAstraintouseforassessmentofthelight‐
activatedantimicrobialcoatingsitwouldbelogicaltoselectthestrainthatislesslight
sensitiveandthesestudiesshowthistobeEMRSA‐16
138
35 Conclusions
Samplingthetestsurfacesbyswabbingandsubsequentlyperformingviablecountshas
been shown toprovideanadequateestimateof concentrationofbacteriaona test
surfaceDatageneratedinthischaptersuggestthatamethodincorporatingtheuseof
ATP bioluminescence for testing novel antimicrobial coatings would not be
appropriateThesuperiorityoftheviablecounttechniquewasespeciallyapparentat
lowbacterial concentrationswhen theATPbioluminescencebased techniqueswere
unable to consistently confirm the presence of small numbers of bacteria Two
meticillin‐sensitive strains of S aureus were shown to be susceptible to
photoinactivation by white light alone whereas the meticillin‐resistant strains of S
aureustestedshowedincreasedtoleranceindicatingapossiblevirulencefactorfound
inEMRSA‐16EcoliandEfaecalisalsodisplayedtolerancetotheinhibitoryeffectsof
thewhitelightsourcesoEcoliwillbeinitiallyusedtoassesstheantibacterialactivity
ofthelight‐activatedcoatings
139
4 Assessment of novel APCVD‐synthesised light‐activatedantibacterialmaterialsforuseinthehospitalenvironment
41 Introduction
Presentedinthischapterarethefindingsfromaseriesofnovelantimicrobialcoatings
thatwereactivatedbyeithervisibleorultravioletlightThefilmsweregeneratedusing
aprocesscalledAPCVD(Section151)wheredopantswereaddedduringthesynthesis
of the TiO2 thin films in order to alter the photochemical properties TiO2 is awell‐
described photocatalyst both as a powder and when immobilised within thin films
(Matsunagaetal1985)andisnormallyactivatedbyultraviolet(UV)lightTheaimof
thecurrentworkwastoshiftthebandwidthofnovelTiO2filmssothatlightofalower
frequencywas able to initiate photocatalysis (Section 133)E coliwas used as the
test organism for the initial screening as it has been demonstrated that it is not
affected by the white light used for activation unlike some of the staphylococcal
speciestested(Section333)whichhavepreviouslybeenshowntohaveanincreased
resistance to theactivityofphotocatalysis (Decraeneetal 2006Pageetal 2007)
Pure TiO2 thin films were also tested to demonstrate the difference between the
dopedandun‐dopedmaterialsTheantibacterialactivityofthematerialswasassessed
usingaswab‐basedmethodologyandnotanATPbioluminescencebasedtechniqueas
viablecountsproducedthemostreproducibleresultsinChapter3thepresenceofE
coliwasalwaysreportedevenatlowconcentrations
140
42 Materialsandmethods
421 Synthesisofthethinfilms
Thetitanium(IV)oxynitridefilms(Ti285O4N) (TiON‐1)wereproducedbyAPCVDusing
ammoniaas thenitrogen sourceasdescribed inSection21011Anitrogen‐doped
thin film (TiON‐2) was also synthesised using ammonia as the nitrogen source as
described in Section 21011 The nitrogen‐doped TiO2 films N1 N2 and N3 were
producedbyAPCVDusingt‐butylamineasthenitrogensourceasdescribedinSection
21012andwerecutfromdifferentareasofasinglesheetofcoatedglassThesulfur
containingthinfilmsS1S2andS3werepreparedwithcarbondisulfideasthesulfur
sourceandtitaniumtetrachloride(TiCl4)asthetitaniumsourceasdescribedinSection
21013TiO2thinfilmswerepreparedascontrolsasdescribedinSection21014
Theconditions chosen forall experimentsallowed for the rapid deposition ofa thin
filmwhichremaineddefect‐andpinhole‐freebyeyeThefilmswereallwelladhered
tothesubstrateandresistanttoabrasionThethinfilmswerecharacterisedandthe
functionalactivityassessedasdescribedpreviously(Dunnilletal2009a2009bAiken
etal2010)
422 Measuringtheantibacterialeffectofthethinfilms
BacterialstrainsweremaintainedasdescribedinSection21andbacterialsuspensions
ofEcoliATCC25922werepreparedasdetailed inSection23resultinginastarting
inoculum of approximately 107 cfu ml equating to approximately 25 x 105 cfu
sampleTheeffectofthephotocatalyticthinfilmsontheviabilityofbacterialstrains
was determined using the swab‐basedmethodology described in Section 2122 and
141
Figure22SamplesweredenotedCforthenitrogenorsulfur‐containingsamplesTfor
theTiO2thinfilmsandGfortheuncoatedglassTheMannWhitneytestwasusedto
determine the statistical significance of any differences observed as described in
Section213
423 Assessmentofthedecontaminationregimen
Priortomicrobiologicalassessmentthethinfilmsweredecontaminatedasdescribed
in Section 2121 The decontamination procedurewas later amended and stored in
thedarktodeactivateandusedonlyafteraperiodof72hours
424 Effectofthecoveringmaterialonthinfilmactivity
To prevent dehydration of the bacterial inocula the effect of thematerials used to
coverthemoisturechamberwasinvestigatedThethinfilmswereincubatedunderthe
whitelightfor24hourswitharangeofcoveringswhichstillallowedlightpenetration
ontothebacterialsuspensioninoculatedontothethinfilmThefollowingcoverswere
used(i)glasscoverslips(ii)quartzcoverslips(iii)thepetridishlid(iv)clingfilmAUV‐
visiblelighttracewasalsogeneratedtomeasurethetransmissionoflightthroughthe
petri dish lid and the clingfilm The intensity of light generated by the lamp was
quantifiedusinga lightmeter (LX101LuxmeterLutronElectronicEnterpriseCoLtd
Taiwan)
142
43 Results
431 Photocatalyticactivityoftitaniumdioxidethinfilms
The activity of the TiO2 films was initially examined to check whether any
photocatalyticactivitywasobservedusingwhite lightasthesourceof incident light
TiO2thinfilmspreparedin‐housewereassessedalongsidecommerciallyproducedthin
filmsWhentheTiO2thinfilmswereassessedforphotocatalyticantibacterialactivity
againstEcoli(Figure41)nostatisticaldifferenceinbacterialrecoverywasobserved
from the thin films after a 24 hour exposure period compared with the bacterial
recoveryfromtheglassslides (pgt005) thereforetheseTiO2thinfilmswereusedas
controlsfortheremainingexperimentswherenecessary
log 10cfuthinfilm
Exposureconditions
log 10cfuthinfilm
Exposureconditions
Figure41ActivityoftheTiO2thinfilmspreparedin‐houseAnaliquotofEcoliwasaddedtothethinfilmsbeforeexposuretothewhite lightsourcefor24hours(L+)Alternativelythinfilmswereincubatedinthedarkthroughout(L‐)UncoatedglasssidesandTiO2thinfilmsaredenotedbyGandTirespectivelyThethickhorizontallinesindicatemedianvaluesthebaseandtopofeachboxrepresentsthe25and75quartilesrespectivelyandtheerrorbarsthe10and90percentilesandthesmallcirclesareoutliersThedottedhorizontal line indicates thedetection limitofthesamplingmethod14log10cfusample
143
ThecommerciallyproducedTiO2thinfilmPilkingtonActivTMwasalsoassessedforany
photocatalytic activity using the white light source and a 03 log10 cfu sample
reduction in the recovery of E coli was observed compared with the thin film
incubated in the absence of light (Figure 42) This small decrease was statistically
significant (plt 0001)which is likely to be due to the small level of variance in the
viable count recovered from the thin films in the control group rather than to a
differencefromthenumberofbacterialcoloniesobservedinthetestgroupandsuch
smalldifferenceswouldnotbeconsideredmicrobiologicallydifferent
$ amp$$()$$+-$(-
0123)45$6-+-3
Figure42Effectof thecommerciallyproducedTiO2 thin filmPilkingtonActivTMonthesurvivalofEcoliThinfilmswereexposedtowhite lightfor24hours(A+) thebacterialdropletwasaddedthen thesamplewasexposedasecond lightexposureperiodof24hours(L+)Alternativelythinfilmswereexposedtojustthelatterlightdose(A‐L+)thefirstlightdoseonly(A+L‐)orincubatedinthedarkthroughout(A‐L‐)The asterisk denotes statistical significance compared with an uncoated controlincubatedunderthesamelightingconditionsasdescribedinSection213
144
432 Photocatalytic antibacterial activity of nitrogen‐containing titanium
dioxidethinfilmsTiON‐1andTiON‐2
4321 Photocatalyticactivityafterexposuretoultravioletlight
Theactivityofthenitrogen‐dopedthinfilmsTiON‐2wereassessedinitiallyusingtwo
UVlamps(254nm365nm)asthelightsourcesWhenthethinfilmTiON‐2waspre‐
exposed to 1 hour of 254 nm light inoculated with E coli and then subjected to 4
hoursof365nmlight(CA+L+)a14log10cfusample(955)reductioninbacteria
was observed compared with the uncoated control exposed to the same light
conditions (GA+L+)Thisdifference is statistically significant (plt001)and is shown
graphically alongwith the bacterial counts for a number of the other conditions in
Figure43
Exposingtheuncoatedslidestobothlightincubationsteps(GA+L+)orjustthelatter
light incubation step (GA‐L+) resulted ina05 log10 cfu sample reductionofE coli
comparedwiththeslidesincubatedintheabsenceoflight(GA‐L‐)asthisdifference
wasstatisticallysignificant(plt001)theGA+L+slidewasusedasthenegativecontrol
throughout
The pre‐inoculation activation step did not substantially enhance the activity of the
thin films when they were subsequently exposed to the 365 nm light A similar
decreaseinbacterialrecoverywasobservedwhetherthethinfilmswerepre‐activated
(14 log10cfu samplereduction)ornot (11 log10cfusamplereduction)andthese
valueswerenot statisticallydifferent (pgt005) Therewasno significantdecrease in
the number of bacteria recovered from thin films which were exposed to just the
activationstep(CA+L‐)andnosignificantdecreaseinthenumberofrecoverableEcoli
145
was observed from the thin films which were incubated in the absence of light
throughout (CA‐L‐) in fact the bacterial recoverywasgreater from these thin films
thanfromthenegativecontrol
$ amp$$()$$+-$(-
0123)45$6-+-3
Figure43ConcentrationofEcoliremainingonthethinfilmTiON‐2afterexposureto1hour254nmlightand4hours365nmlight(CA+L+)orjustthelatterlightdose(C A‐L+) Thin films were also exposed to the activation step only (C A+L‐) orincubated in the dark throughout (C A‐L‐) Uncoated glass slideswere exposed tobothlightconditions(GA+L+)orneither(GA‐L‐)
Whenthetitanium(IV)oxynitridefilmTiON‐1waspre‐exposedto1hourof254nm
lightinoculatedwithEcoliandthenexposedto4hoursof365nmlight(CA+L+)a
41 log10cfusample(9999)reduction inbacterialcountwasobservedcompared
withtheuncoatedcontrolexposedtothesamelightconditions(GA+L+)(Figure44)
Thisdifferencewashighlystatisticallysignificant(plt001)
Thepre‐inoculationactivationstepwasfoundtoenhancetheactivityofthethinfilms
TherecoveryofEcoli fromtheoxynitridethinfilmswhichwereexposedtothe365
nmlightforfourhourswithoutprioractivationwasnotsignificantlydifferentfromthe
146
recoveryfromtheuncoatedcontrolslides(pgt005)Similarlynosignificantdecrease
inthenumberofbacteriarecoveredfromthethinfilmswasobservedwhentheywere
justactivated(CA+L‐)orwhenthethinfilmswereincubatedintheabsenceoflight(C
A‐L‐)
IncomparisonwhentheTiO2thinfilmswereexposedto365nmlightwitha254nm
activationsteptherewasa41 log10cfusamplereduction inbacterialcount Itwas
converselyfoundthatfortheTiO2thinfilmstheactivationstepwasunnecessaryand
exposure to 365 nm light alone led to a 41 log10 cfu sample reduction after four
hoursoflightexposure(datanotpresented)
$ amp$$()$$+-$(-
0123)45$6-+-3
Figure44ConcentrationofEcoliremainingonthethinfilmTiON‐1afterexposureto1hour254nmlightand4hours365nmlight(CA+L+)orjustthelatterlightdose(C A‐L+) Thin films were also exposed to the activation step only (C A+L‐) orincubated in the dark throughout (C A‐L‐) Uncoated glass slideswere exposed tobothlightconditions(GA+L+)orneither(GA‐L‐)
147
4322 Photocatalyticactivityafterexposuretowhitelight
Thephotoactivityofthesethinfilmswassubsequentlyassessedusingvisible lightas
theactivatinglightsourceAswhitelighthasalowerfrequencythanultravioletlight
the sampleshad tobeexposed to thewhite light fora longer timeperiodThe thin
films were exposed to the white light for 24 hours as an lsquoactivatingrsquo step then
inoculatedwithEcoliandexposedtothewhitelightforeither618or24hoursThe
thin film TiON‐2 did not display any significant photoactivity after 6 18 or 24 hours
exposure to thewhite light (Figure 45) The greatest decrease in bacterial recovery
was exhibited after 24 hours where just a 05 log10 cfu sample reduction was
observedcomparedwith the thin films incubated in theabsenceof light throughout
the duration of the experiment (A‐L‐) However the effect of the light source alone
should be incorporated into this reduction to show that any reduction in bacterial
recoverywasduetothephotoactivityofthethinfilmsandnotanartefactcausedby
thelightsource
Itwasdemonstrated inSection3331andFigure36that24hoursexposuretothe
whitelightresultedina02log10cfusampledecreaseintherecoveryofEcoliThis
figurewassubtractedfromthereductionsseeninthissectionandthisvaluewasused
astheoverallnegativecontrol(GA+L+)Thereforethegreatestdecreaseinbacterial
recoveryforthenitrogen‐dopedthinfilmwasjust02log10cfusampleafterexposure
toboth24hourlightincubationstepswhichwasnotstatisticallysignificant
148
log 10cfuthinfilm
Exposureconditions
log 10cfuthinfilm
Exposureconditions
Figure 45 Effect of the thin film TiON‐2 on the survival ofE coli Thin filmswereexposedtowhite lightfor24hours(A+) thebacterialdropletwasaddedthenthesamplewasexposedasecondlightexposureperiodofeither618or24hours(L+)Alternatively thin filmswere exposed to just the latter light dose (A‐L+) the firstlightdoseonly(A+L‐)orincubatedinthedarkthroughout(A‐L‐)
Whenthetitanium(IV)oxynitridefilmTiON‐1wasexposedtothewhitelightforeither
6or18hours therewasno significant reduction in the recoveryofE coliHowever
after24hours irradiationareductiveeffectwasseenandtheaveragerecoveryofE
colifromthethinfilm(A+L+)was06log10cfusamplelowerthantherecoveryfrom
theuncoatedglassslidesexposedtothesamelightconditions(GA+L+)asdisplayedin
Figure 46 This result was statistically significant (p lt 001) However the observed
effect was not consistent demonstrated by the variability of the A+L+ 24h data
showninFigure46Evenafterfiveexperimentalrepeatsaconsistentresultcouldnot
beachievedandreductionsinthebacterialcountrangedfrom49log10cfusampleto
05log10cfusamplewithanaveragereductionofjust06log10cfusample
149
$ amp$$()$$+-$(-
0123)45$6-+-3
Figure 46 Effect of the thin film TiON‐1 on the survival ofE coli Thin filmswereexposedtowhite lightfor24hours(A+) thebacterialdropletwasaddedthenthesamplewasexposedasecondlightexposureperiodofeither618or24hours(L+)Alternatively thin filmswere exposed to just the latter light dose (A‐L+) the firstlightdoseonly(A+L‐)orincubatedinthedarkthroughout(A‐L‐)
Theanti‐Ecolieffectoftitanium(IV)oxynitridethinfilmTiON‐1wasgreaterthanthe
nitrogen‐doped thin filmTiON‐2 underboth lighting conditionswhichdemonstrates
thattheformerthinfilmwasamoreeffectivephotocatalystunderthetestconditions
433 Photocatalyticantibacterialactivityofnitrogen‐dopedtitaniumdioxide
thinfilmsN1N2andN3
4331 Photocatalyticactivityafterexposuretowhitelight
Theactivityofasecondsetofnovelnitrogen‐containingthinfilmswasassessedusing
whitelightastheactivatingsourceof irradiationThethinfilmswereexposedtothe
whitelightfor24hourstheninoculatedwithEcoliandre‐exposedtothewhitelight
for24hoursThegreatestreduction inbacterial recoverywasseenwhenEcoliwas
150
inoculated onto thin film N1 and a 28 log10 cfu sample (999) reduction was
observed(Figure47)comparedwiththethinfilms incubated intheabsenceof light
throughout the duration of the experiment (A‐L‐)When the uncoated glass sample
exposedtobothlightconditionswasusedasacontrol(GA+L+)theoverallreduction
inEcolicauseddirectlybytheactivityoftheN‐dopedthinfilmN1wasapproximately
25log10cfusample(997)whichwashighlystatisticallysignificant(plt0001)
log 10cfuthinfilm
Exposureconditions
log 10cfuthinfilm
Exposureconditions
Figure47EffectofthethinfilmN1onthesurvivalofEcoliThinfilmswereexposedtowhitelightfor24hours(A+)thebacterialdropletwasaddedthenthesamplewasexposed a second light exposure period of 24 hours (L+) Alternatively thin filmswereexposed to just the latter lightdose (A‐L+) the first lightdoseonly (A+L‐)orincubatedinthedarkthroughout(A‐L‐)
Exposingthethinfilmstojustthesecondlightcondition(A‐L+)resulted ina09log10
cfu sample reduction in the recovery of E coli (p lt 005) compared with the
uncoated control incubated under the same conditions (G A+L+) Exposing the thin
filmstotheinitialactivatinglightdoseonly(A+L‐)didnothaveasignificanteffecton
151
therecoveryofEcolinordidexposuretothethinfilmsintheabsenceoflightinfact
a higher recovery of E coli was observed in this control group Hence an additive
effectwasobservedwherebyexposure toeither the second lightdoseor both light
doses resulted in a significant reduction in bacterial recovery with the greatest
decreaseobservedafterbothlightexposureperiods
WhenthethinfilmN2wasexposedtowhitelightforboth24hourperiodsa16log10
cfu sample reduction was observed (Figure 48) compared with the thin films
incubatedinthedarkthroughoutWhentheuncoatedglassslideexposedtothesame
lightconditionswasusedasthecontrolthentherecoveryofEcoliwasreducedto11
log10cfusampleNostatisticalsignificantdifferencewasseenbetweenthetestand
control groups as the data sets were small No decrease in bacterial recovery was
observedwhen the thin filmswere exposed to thewhite light for 24 hourswithout
pre‐activation(A‐L+)whenthethinfilmswere justpre‐activated(A+L‐)orwhenthe
thinfilmswere incubated intheabsenceof light (A‐L‐)comparedwiththeuncoated
controlexposedtobothlightdoses
152
log 10cfuthinfilm
Exposureconditions
log 10cfuthinfilm
Exposureconditions
Figure48EffectofthethinfilmN2onthesurvivalofEcoliThinfilmswereexposedtowhitelightfor24hours(A+)thebacterialdropletwasaddedthenthesamplewasexposed a second light exposure period of 24 hours (L+) Alternatively thin filmswereexposed to just the latter lightdose (A‐L+) the first lightdoseonly (A+L‐)orincubatedinthedarkthroughout(A‐L‐)
AlargevariationintherecoveryofEcoliwasobservedfromthesetofthinfilms(N3)
displayedinFigure49Onaveragethereductioninbacterialrecoveryfromthepre‐
activatedthinfilmsincubatedunderwhitelightfor24hourswas09log10cfusample
whencomparedwiththethinfilmsincubated inthedarkthroughoutthedurationof
the experiment The reduction drops to a 05 log10 cfu sample reduction when
compared with the uncoated control incubated exposed to both light doses These
reductions were not statistically different The recovery of E coli from these films
rangedfrom58log10cfusampletobelowthelimitofdetectiondemonstratingthe
wide spectrum of activity that these thin films displayed under the experimental
conditionsWhetherthethinfilmN3wasexposedtojustthesecondlightdosewhilst
inoculatedwithEcolijustthepre‐activatingwhitelightdoseorneithertherewasno
153
significant reduction in bacterial recovery compared with the uncoated control
exposedtobothperiodsoflight
log 10cfuthinfilm
Exposureconditions
log 10cfuthinfilm
Exposureconditions
Figure49EffectofthethinfilmN3onthesurvivalofEcoliThinfilmswereexposedtowhitelightfor24hours(A+)thebacterialdropletwasaddedthenthesamplewasexposed a second light exposure period of 24 hours (L+) Alternatively thin filmswereexposed to just the latter lightdose (A‐L+) the first lightdoseonly (A+L‐)orincubatedinthedarkthroughout(A‐L‐)
434 EffectofchangingthedecontaminationregimenonthinfilmN1
The effect of themodified decontamination regimewas evaluated by repeating the
white lightexposureexperimentson the thin filmdesignatedN1However the thin
films could not be reproduced to the samespecifications and had therefore already
been exposed to the original decontamination regime before the newmethod was
usedTheactivityofthethinfilmwasmaintainedforthefirstfourreplicateswhenthe
new decontamination regimen was used (Figure 410a) a statistically significant
reduction in bacterial recovery was observed (p lt 001) and the new regime was
thought to be successful However the photocatalytic activity of the thin filmswas
154
thenlostwhentheexperimentwasrepeatedonasubsequentthreeoccasions(Figure
410b)andnostatisticallysignificantreductionintherecoveryofEcoliwasobserved
WhenthethinfilmswerestainedusingtheLiveDeaddifferentialstainafluorescent
greensmearwasseenonsurfaceofthefilmsbutnoviableornon‐viablebacterialcells
werepresent
(a) (b)
log10
cfuthinfilm
Exposureconditions
log10
cfuthinfilm
Exposureconditions
log10
cfuthinfilm
Exposureconditions
log10
cfuthinfilm
Exposureconditions
log10
cfuthinfilm
Exposureconditions
log10
cfuthinfilm
Exposureconditions
Figure410Light‐activatedantimicrobialkillingofEcolionthinfilmN1(a)andafterinactivation (b) The thin film was exposed to first light dose (A+) the bacterialdropletwas added and then the thin filmwas exposed to second light dose (L+)Alternatively thin filmswere exposed to just the latter light dose (A‐L+) the firstlightdoseonly(A+L‐)orincubatedinthedarkthroughout(A‐L‐)
435 Effectofcoveringmaterialonthinfilmactivity
Theeffectofthematerialusedtocoverthemoisturechamberwasinvestigatedwith
regardtobacterialviabilityGlassorquartzcoverslipswereusedtocoverthebacterial
inoculumduringexposuretothewhitelightsourcebutafter24hoursincubationthe
dropletshadevaporateditwasnotpossibletoculturetheorganismsontosolidagar
using the viable count technique and the cells had become non‐viable This was
confirmedbyvisualisationusingtheLiveDeadstain(datanotincluded)whichshowed
100ofcellsweredeadAbathofwaterwasplacedatthebaseofthe incubatorto
155
saturate the environment with moisture to prevent evaporation but the bacterial
inoculumhadonceagaindriedoutafterthe24hourincubationperiod
Whenthemoisturechamberwascoveredwithaplasticpetridishlidorclingfilmthe
bacterialdropletsdidnotdryoutthereforetheeffectivenessofthesecoveringswas
assessedE coli inoculated onto thin film TiON‐2 showed a greater susceptibility to
killingbyUVlightwhenthemoisturechamberwascoveredwithclingfilm(Figure411)
comparedtowhenitwascoveredwiththepetridishlid(Figure43)A49log10cfu
samplereductioninviableorganismswasseenwiththeclingfilmcoveringcompared
witha14log10cfusamplereductionwhentheplasticpetridishcoverwasused
$ amp$$()$$+-$(-
0123)45$6-+-3
Figure 411 Concentration of E coli remaining on the thin film TiON‐1 using aclingfilmcoveringThethinfilmswereexposedto1hour254nmlightand4hours365 nm light (C A+L+) or just the latter light dose (C A‐L+) Thin films were alsoexposedtotheactivationsteponly(CA+L‐)orincubatedinthedarkthroughout(CA‐L‐)Uncoatedglassslideswereexposedtobothlightconditions(GA+L+)orneither(GA‐L‐)
156
AUV‐visible lighttransmissiontracewasproducedtohighlightanydifferencesinthe
transmissionoflightthroughandthereflectancefromthetwocoveringmaterialsThe
UV‐visiblelighttransmissiontrace(Figure412)showedthataround90oflightfrom
the visible portion of spectrum (with a wavelength between 400 and 700 nm)
penetrated through both the petri dish and the clingfilm coverings Less than 2of
lightwithawavelengthbelow280nmwasabletopenetratethroughthepetridishlid
However more than 80 of light of this wavelength could penetrate through the
clingfilm covering This finding indicates that this coveringwould not be suitable for
the series of experiments evaluating the effect of the light activated antimicrobial
coatingsasbacteriaareinactivatedbylightofthiswavelengthandbelow(Saitoetal
1992)Thegreaterreductioninbacterialrecoveryshownwhentheclingfilmwasused
to cover the moisture chamber suggests that wavelengths of light with a higher
frequencywereabletopassthroughtheclingfilmresultinginthegreatersusceptibility
ofE coliobservedwhen inoculatedonto the thin filmTiON‐2which suggests there
couldbe some leakageof sub‐365nmUV light from the light source that caused the
observedincreaseinphotoactivityThereforethepetridishlidwasusedtocoverthe
moisturechamberinalllight‐activationexperiments
157
$
amp
(
)
amp $ $amp amp amp ampamp amp (
+-012345406
78096
990454lt
=284gt934-8
01A6
)06
06
Figure412UV‐visible lighttransmissiontraceofthepetridish lidandtheclingfilmcoversThewavelengths280nmand400nmareindicatedbyverticaldottedlines
436 Photocatalytic antibacterial activity of sulfur‐based titanium dioxide
thinfilms
The photocatalytic activity of a series of novel sulfur‐doped thin filmswas assessed
Thethinfilmswereexposedtowhitelightfor72hoursbeforeasuspensionofEcoli
wasaddedThethinfilmswerethenre‐incubatedunderthewhite light fora further
24hoursbeforesamplingThephotocatalyticactivityofthinfilmS2isshowninFigure
413whereasignificantdecreaseinbacterialrecoverywasobserved(plt001)A25
log10 cfu sample decreasewas observed comparedwith the sulfur‐doped thin film
incubatedinthedarkthroughoutthedurationoftheexperimentTheoveralldecrease
in bacterial recovery when compared to a TiO2 thin film exposed to the same light
conditionswas22log10cfusamplewhichremainsstatisticallysignificant(p=001)
158
AlargevariationinbacterialrecoverywasobservedwhenthethinfilmS2wasexposed
to thewhite light for 24 hourswithout prior activation ranging from62 log10 cfu
sample to below the limit of detection with an average recovery of 41 log10 cfu
sample indicating that the activation step did not have a significant effect on the
photoactivity of the S‐doped thin film No statistically significant decrease in the
recovery of E coli was observed under these conditions when the thin film was
exposedtotheactivating lightdosealoneorwhen incubated intheabsenceof light
entirely
$ amp$$()$$+-$(-
0123)45$6-+-3
Figure 413 Effect of the thin film S2 on the survival of E coli Thin films wereexposedtowhite lightfor72hours(A+) thebacterialdropletwasaddedthenthesamplewasexposedasecond lightexposureperiodof24hours(L+)Alternativelythinfilmswereexposedto justthe latter lightdose(A‐L+) thefirst lightdoseonly(A+L‐) or incubated in the dark throughout (A‐L‐) TiO2 controls were exposed toeitherbothlightdoses(TiA+L+)orneither(TiA‐L‐)
ThethinfilmsS1andS3werelesseffectiveatreducingtheEcolibacterialloadafter
exposuretothewhitelightTherewasnosignificantdecreaseintherecoveryofEcoli
fromthesurfaceofpre‐activatedthinfilmS1afterthe24hourexposureperiod(Figure
159
414)comparedwitheithertheTiO2controlexposedtothesamelightingconditions
or the sulfur‐doped thin film incubated in the absence of light Similarly the pre‐
activatedthinfilmS3didnotproduceasignificantreductiveeffectintherecoveryofE
coli from the surface of the thin films after the 24 hour exposure period when
comparedwitheithertheTiO2controlexposedtobothlightdosesorthesulfur‐doped
thinfilmnotexposedtowhitelight(Figure415)Howeveraninconsistenteffectwas
seenontheS3thinfilmswhichwerenotpre‐exposedtothewhitelightfor72hours
but incubated under the white light for 24 hours after addition of the bacterial
suspension This result was not reproducible demonstrated in the box andwhisker
plotbythelargesizeofboththeboxanderrorbarsA09log10cfusamplereduction
was seen comparedwith the thin film incubated in the absence of light (p lt 005)
HoweverthemedianreductionwaslowerwhencomparedwiththeTiO2thinfilm(06
log10 cfu sample) or the uncoated glass control (01 log10 cfu sample) and these
reductionswerenotstatisticallysignificant
160
log 1
0 cfu
t
hin
film
Exposure conditions
Figure 414 Effect of the thin film S1 on the survival of E coli Thin films wereexposedtowhite lightfor72hours(A+) thebacterialdropletwasaddedthenthesamplewasexposedasecond lightexposureperiodof24hours(L+)Alternativelythinfilmswereexposedto justthe latter lightdose(A‐L+) thefirst lightdoseonly(A+L‐) or incubated in the dark throughout (A‐L‐) TiO2 controls were exposed toeitherbothlightdoses(TiA+L+)orneither(TiA‐L‐)
Exposure conditions
log 1
0 cfu
t
hin
film
Figure 415 Effect of the thin film S3 on the survival of E coli Thin films wereexposedtowhite lightfor72hours(A+) thebacterialdropletwasaddedthenthesamplewasexposedasecond lightexposureperiodof24hours(L+)Alternativelythinfilmswereexposedto justthe latter lightdose(A‐L+) thefirst lightdoseonly(A+L‐) or incubated in the dark throughout (A‐L‐) TiO2 controls were exposed toeitherbothlightdoses(TiA+L+)orneither(TiA‐L‐)
161
Table41SummaryofthephotocatalyticactivityofthenitrogenandsulfurdopedthinfilmsassessedinthischapterThinfilmswereexposedtowhitelightfor24or72hoursforN‐dopedandS‐dopedsamplesrespectivelyThebacterialdropletwasaddedbeforethesamplewasexposedasecondwhitelightexposureperiodof24hoursBacterialcountsobtainedwerecomparedwithuncoatedglassslidesexposedtothesamelightingconditions
SamplenameWhitelight‐induced
photocatalyticactivitylog10cfupersample
Statisticalsignificance
TiON1 06 plt001
TiON‐2 02 Nil(pgt005)
N1 25 plt0001
N2 11 Nil(pgt005)
N3 05 Nil(pgt005)
S1 Nodecrease Nil(pgt005)
S2 17 pgt0001
S3 Nodecrease Nil(pgt005)
44 Discussion
441 UVlight‐inducedphotocatalyticactivity
Thedatapresentedinthischapterhasdemonstratedtheantibacterialphotoactivityof
anumberofnoveldopedTiO2thinfilmsgeneratedbyAPCVDThethinfilmsthatwere
initially assessed were doped with nitrogen and exposed to UV light in order to
demonstrateequivalencewithpuretitaniaThetitanium(IV)oxynitridethinfilmTiON‐
1 demonstrated greater photoactivity than theN‐doped thin film TiON‐2 and a 41
log10cfusamplereductionwasachievedonthepre‐activatedtitanium(IV)oxynitride
sampleafterjust4hoursexposuretothelightsourceTheseresultsalsoshowthatthe
162
titanium(IV)oxynitridethinfilmsdemonstratednoanti‐bacterialactivitywithoutUV
exposure after the inoculation of the bacterial suspension therefore the mode of
actionisunlikelytoberelatedtothediffusionofionsontothesurfaceandisgenuinely
photo‐activated
442 Whitelight‐inducedphotocatalyticactivity
Thephotocatalyticactivityofthethinfilmswasthenassessedusingwhitelightasthe
activatinglightsourceWhitelightwasusedasanactivatingsourcelightsourceasUV
light is known to have a bactericidal effect (Vermeulen et al 2008) and the
applications of the resultant thin film would be wider using a lower energy light
source Any reduction in bacterial count observed under these conditions would
indicate a shift in the band gap of the material caused by the doping process
demonstratingthatactivationby lightofa lowerwavelength ispossible (Dunnilland
Parkin2009)A reductionofup to49 log10 cfu sampleofE coliwasobservedon
thinfilmTiON‐1(Ti285O4N)butthiswasnotconsistentandtheaveragereductionwas
just06 log10 cfu sampleHowever thisdoes provideapromisingbasis for further
dopingexperiments
The photocatalytic activity of the N‐doped thin films N1 N2 and N3were assessed
next using white light as the activating light source Thin film N1 displayed the
greatestphotocatalyticactivityanda25log10cfusampledecreaseintherecoveryof
E coli was observed after exposure to both light incubation steps These findings
confirm the chemical characterisation tests performed on these samples such as
photooxidation of stearic acid and contact anglemeasurements and these data are
163
published elsewhere (Dunnill et al 2009b 2009c 2010) A 09 log10 cfu sample
decrease was observed when the 24 hour activating step was omitted which
demonstrated that the activation stepwas required to increase the photoactivity of
the thin films This increase in activity is attributed to the pre‐cleaning effect of the
treatmentThelackofactivityonthethinfilmsthathadbeenactivatedbutthennot
exposedtothesecondlightstepindicatedtheshortlifetimeofthereactivespecieson
the surface of the thin films that are responsible for killing the bacterial cells It is
unlikely that the oxygen radicals generated in the presence of light survive long
enoughtokillthebacteriathatwereappliedaftertheactivationstephasendedgiven
that singletoxygenhasahalf lifeof just1 micros (Pernietal 2009a)Thevariability in
photocatalytic activity observed on the N1 N2 and N3 thin films which were
synthesised on the same sheet of float glass demonstrates the inherent lack of
reproducibility in the composition of coatings produced using this deposition
technique
TheactivityoftheN‐dopedthinfilmN1wasgreaterthanthatseenforthetitanium
(IV) oxynitride thin film TiON‐1 The two thin films were synthesised with different
precursors the N‐doped thin films were synthesised using t‐butylamine as the
nitrogensourceandammoniawasusedforthetitanium(IV)oxynitridethinfilmsThe
chosen nitrogen precursor was introduced into the titanium (IV) chloride and ethyl
acetatevapoursatthepointofentrytothedepositionchamberresultinginthermal
decompositionofthenitrogenprecursoronthesurfaceoftheglasssubstrateduring
formationofTiO2(DunnillandParkin2009)Pre‐reactioncomplexesweremorelikely
toformwhenammoniawasusedasthenitrogensourceratherthant‐butylamineand
164
thesecomplexescancausecontrollineblockageswhichcanaffecttheconcentration
of nitrogen deposited onto the surface of the glass The activity of the thin films is
dependentupontheconcentrationofnitrogen intheTiO2thinfilm(Irieetal2003)
so perhaps the greater control of nitrogen deposition displayed when t‐butylamine
was used as the nitrogen precursor conferred the increased photocatalytic activity
observed
The S‐doped thin film S2 also displayed significantwhite light driven photocatalytic
activityanda22 log10 cfu sampledecrease in the recoveryofE coliwasobserved
after a 24 hour exposure periodOnce again themicrobiological findings confirmed
the initialchemicalcharacterisationscreeningtestsandthethinfilmwiththefastest
rateofstearicacidphotodegradationdemonstratedthemostsignificantantibacterial
activity (Dunnill et al 2009a 2010) However the N‐doped thin films displayed
greaterphotocatalyticactivitythantheS‐dopedthinfilmsevenwhentheinitialwhite
lightactivationtimewasextendedfrom24to72hours
Reports in the literature have described the antibacterial properties of white light
activated N‐ and S‐doped thin films but direct comparison is difficult due to
differences in the method of synthesis used (Asahi et al 2001 Mills et al 2002
Diwaldetal2004ThompsonandYates2006)Indeedthethinfilmsdescribedinthis
chapterarethefirstpublishedthinfilmswith interstitialnitrogen‐orsulphur‐doping
possessingwhitelightactivatedantibacterialpropertiesN‐dopedthinfilmshavebeen
shown to generate a greater photocatalytic effect against E coli compared with
carbon‐doped thin films (Wong et al 2006) However the reduction in bacterial
recovery was minimal (less than a 1 log10 reduction) and when these films were
165
characterised the nitrogen doping was shown by XPS to be substitutional with an
ionisation peak at 396 eV (Yang et al 2004) in contrast to the interstitial‐doped
nitrogen described in this chapterwith an ionisation peak at 400 eV (Dunnill et al
2009c)Thisdoeshoweverdemonstratethatnitrogenisabetterchoiceofdopantthan
carbon if photocatalytic properties are desired Titanium oxide doped with both
nitrogen and carbon was shown to exhibit enhanced photocatalytic properties and
reductionsofmorethan3log10cfumLwereobserved(Lietal2007)butahalogen
bulbwas used as the light sourcewhich has a higher intensity than thewhite light
sourceusedinthischapterandsoagreaterphotocatalyticeffectwouldbeexpected
Additionallypowdershaveagreatersurfaceareapervolumeratiothansolidsfurther
boostingthepredictedlevelofphotocatalysis
Thequantityofnitrogenpresentinthethinfilmisofparamountimportanceandsome
groups show high levels of nitrogen doping can result in the production of poor
photocatalysts (Irie et al 2003) whereas other groups show increased levels of
photocatalysis when the nitrogen concentration is higher (Li et al 2007) When
nitrogen concentrations are higher less TiO2 reduction occurs and there are more
oxygenvacanciesthatactasrecombinationsitesforpositiveholesandelectronsthus
reducing the overall photocatalytic activity The concentration of nitrogen in the N‐
doped thin film N1 was 013 at and reports in the literature surmise that
concentrationsaround1ndash2atisfavourablealthoughtheoptimallevelisstillunder
debate(Irieetal2003Dunnilletal2011)ConverselywhenTiO2powderwasdoped
withsulfurincreasedlevelsofthedopantledtoahigherlevelofphotocatalysisandan
166
increasedbactericidaleffectwasobservedagainstMicrococcuslylae(Yuetal2005)
Theoptimallevelofdopingisthereforedebatable
443 Limitationsoftheexperimentalwork
Problemswere experienced in synthesising reproducible thin films using the APCVD
apparatusTheprecursorgasesusednamelytitanium(IV)chlorideandethylacetate
werechosenastheyareusedindustriallyintheproductionofTiO2‐basedself‐cleaning
glassbutthesetupofthedepositionchambersusedinthisprojectweredifferentIn
an industrial setting general mass flow controllers would be used to deliver the
reactantsandthegasoutletswouldbestablewiththeglasssheetsmovingunderneath
the float at 500 ‐ 600degC (Dunnill et al 2009b) These conditions result in a more
consistentreactiononthesurfaceoftheglassandamorehomogenouscoatingwhich
is essential for a commercial product The flow rate of the precursor gases are also
more tightly regulated which was more difficult to control using the in‐house
apparatus overall this meant that the resultant thin films varied in their chemical
composition with differences observed between batches of samples samples
synthesised during the same run and even on different areas on the same piece of
floatglass Forexample theN‐dopedsamplesN1N2andN3wereall cut from the
samepieceof floatglassandyetdisplayeda largevariation inphotocatalyticactivity
against E coli This inconsistency is an inherent disadvantage of the APCVD
methodologyandmadeitverydifficulttoassessthethinfilmsmicrobiologicallyasfor
accurate assessment the samples should at least be identical and tested at least in
triplicate for each light exposure condition on three separate occasions for each
bacterialspecies
167
Asaresultthethinfilmsweredecontaminatedaftereachmicrobiologicalassessment
toenablere‐useItwaspostulatedthatbacterialcellsremainingonthesurfaceofthe
thinfilmswouldbeinactivatedbytheisopropanolandheattreatmentswhichwould
restorethethinfilmstotheirnativestateIthasbeenshownpreviouslythattherewas
no residual antimicrobial effect when isopropanol treatment was used to
decontaminate thin films so any activity observed after decontamination can be
attributed to the activity of the coatings alone (Page 2009) However the
photoactivityofthethinfilmsdecreasedaftereachroundofmicrobiologicaltestingso
thedecontaminationregimenwasamendedsothatastageincludingexposuretoUV
light was incorporated Any remaining bacterial cells were postulated to undergo
photoinduced oxidative decomposition (Section 13333) and non‐bacterial debris
wouldalsobedegradedaftertheextendedlightexposureperiodThethinfilmswere
thenincubatedinthedarkforatleast48hourssotoallowoxygenintheairtoreact
withthehydroxylspeciestonegatetheactivatingeffectoftheUVlight(ONeilletal
2003)
Amendment of the decontamination regimen did not prevent the decrease in
antibacterial activity observed on the thin films after sequential use and the exact
mechanismforthis loss inphotoactivitywasnotestablishedBacterialcellswerenot
presentonthethinfilmafterdecontaminationbutafluorescentsmearwasobserved
whichwasnotseenontheunusedthinfilmsIn‐depthmicrobiologicalassessmentof
thesethinfilmswasthereforenotpossibleandanalternativereproduciblemethodof
synthesiswassoughtwhichwillbeexploredinthefollowingchapterHoweverre‐use
168
ofthethinfilmsdiddemonstratethedurabilityofthecoatingsandtheintegrityofthe
coatingwasnotcompromisedafterrepeateduseanddecontaminationcycles
Another limitation of the testmethodwas the choice ofmedia used to recover the
bacterialstrains fromthetestsurfacesTheselectivemediumMacConkeywasused
to culture E coli because round discrete colonies were formed which made
enumeration easier to perform than when the counts were performed on a non‐
selectivesolidmediumsuchasbloodagarHoweverbacteriarecoveredwerelikelyto
besubletallydamagedbyexposuretothephotocatalyticeffectsofthethinfilmsand
cultivationonselectivemediahasbeenshowntoinhibittherepairofthesedamaged
strains (Sandel and McKillip 2004) A non‐selective agar overlay could have been
poured over the selective medium after inoculation to increase the recovery of
damagedcells(SandelandMcKillip2004)
45 Conclusions
Twosetsofnitrogenbasedthinfilmsweresynthesisedbychemicalvapourdeposition
namely N‐doped TiO2 and titanium oxynitride These coatings displayed significant
photocatalyticactivityagainstEcoliafterexposuretoUVlightandimportantlyawhite
light sourcewhich demonstrates a shift in the band gap from theUV to the visible
region of the electromagnetic spectrum TheN‐doped thin films displayed a greater
photocatalyticactivitycomparedwiththetitanium(IV)oxynitridethinfilmsAseriesof
sulfur‐doped thin films were synthesised using the same apparatus which also
displayed significant photocatalytic activity against E coli after exposure to awhite
light source The N‐doped thin film N1 displayed the greatest photoactivity The
169
reproducibilityofthethinfilmssynthesisedusingAPCVDwaspoorandadecrease in
the photocatalytic activity of the thin films was observed after repeated use An
alternativemethodofdepositionwillbeexploredinthenextchapter
170
5 Assessment of novel sol‐gel synthesised light‐activatedantibacterialmaterialsforuseinthehospitalenvironment
51 Introduction
InthepreviouschapteraseriesofTiO2basedthinfilmsweresynthesisedbychemical
vapourdeposition(APCVD)whichdisplayedphotocatalyticpropertieswhenexposed
tovisiblelightThethinfilmsweredopedwitheithernitrogenorsulfurwhichcaused
a shift in the band gap energy of the coating so that lower energy photons of light
could cause excitation of electrons from the valence band to the conduction band
resultingintheproductionofreactiveoxygenspeciesthataretoxictobacteriaThere
were however issueswith the reproducibility of the thin filmswhichmeant itwas
difficulttosynthesisealargenumberoffilmswithidenticalcompositionsInaddition
theactivityofthethinfilmsdecreasedovertimesomicrobiologicalassessmentofthe
usedthinfilmsgeneratedresultswithalargevariation
Analternativemethodofsynthesiswasthereforesoughtandsol‐geldepositionwas
chosenAlargenumberofsamplescouldbesynthesisedfromthesamehomogenous
solandthereislittlevariationintheconstitutionofdifferentbatchesofpreparedsols
so the composition of the resultant films are easier to control However sol‐gel
synthesisedfilmsaregenerallythickerlessmechanicallyrobustandrequiredsintering
aftercoatingtoannealthefilmtothesubstratecomparedwithAPCVDgeneratedthin
films (Brook et al 2007b) Therefore the synthesis methodology included a post‐
coating annealing step and the thickness and robustness of the thin films was be
examinedtodeterminewhetherthiswasdetrimentaltothephotocatalyticactivity
171
Silver ions were added to the titania base layer to improve the photocatalytic and
photo‐activatedantibacterialpropertiesoftitaniaSilverhasbeenusedextensivelyin
antibacterialmaterialsbecauseof itsintrinsicactivity(Silver2003Silveretal2006
Noimark et al 2009) silver ions can move from the surface of the antibacterial
materialthroughthecellmembraneofbacteriawheretheyareabletoelicitapotent
toxiceffect(Kawashitaetal2000Page2009Pageetal2009)
52 Materialsandmethods
521 Thinfilmsynthesis
The thin films were synthesised using sol‐gel deposition in a two‐step process
describedinSection2102ThesilvercoatedTiO2thinfilmsweredenotedAg‐TiO2and
TiO2 thin films and uncoated glass microscope slides were used as controls The
adherence of the TiO2 and Ag‐TiO2 thin films to the glass substrates was tested by
scratchingwith(i)fingernails(ii)aHBpencil(iii)a2Hpencil(iv)asteelscalpel(v)a
diamondtippencilandapplicationandremovalofscotchtapeThestabilityofthethin
filmswereassessedbyimmersioninthefollowingliquidsfor2hours(i)methanol(ii)
acetone(iii)distilledwater(iv)2MHCl(v)2MNaOH
522 Characterisationandfunctionalassessmentofthethinfilms
Thin films of TiO2 and Ag‐TiO2 were prepared on both glass and quartz substrates
beforecharacterisationusingUV‐visiblespectroscopyasdescribed inSection2111
The reflectance datawas used to calculate the thickness of the thin films using the
SwanepoelmethodandtoestimatethebandonsetofthethinfilmsusingaTaucplot
172
Further methods employed to characterise the thin films included XRD Raman
spectroscopyAFMandXPSasdescribedinDunnilletal(2011)
5221 Contactanglemeasurements
Waterdropletcontactanglemeasurementsweretakenofadropletofdeionisedwater
inoculated onto both the Ag‐TiO2 and TiO2 thin films and uncoated glass control as
describedinSection2112Measurementsweretakenafter(i)incubationinthedark
for72hours(ii) irradiationwiththeUVlightsourcefor30minutes(Section2421)
(iii) irradiation with the filtered white light source for 30 minutes (Section 241)
(InstrumentGlasses2000)
5222 Photo‐oxidationofstearicacid
A solution of stearic acidwas inoculated onto both the thin films and the uncoated
glass control slides to assess the rate of photo‐oxidisation as described in Section
2113 The rate of photo‐activity was determined after exposure to three lighting
conditions (i)254nmUV light source forup to 72hours (Section2422) (ii)white
lightsourcefor96hours(Section241)(iii)thesamewhitelightsourcewithafilter
attachedthatabsorbedvirtuallyallsub‐400nmradiation(InstrumentGlasses2000)
523 Antibacterialassessmentofthethinfilms
BacterialstrainsweremaintainedasdescribedinSection21andbacterialsuspensions
ofEcoliATCC25922andEMRSA‐16werepreparedasdetailedinSection23excepta
50 microL bacterial droplet was inoculated onto the surface resulting in a starting
inoculumofapproximately5x105cfusampleTheeffectofthephotocatalyticthin
films on the viability of bacterial strains was determined using the methodology
173
described in Section 2122 and Figure 22 except the activation stepwas omitted
WhenrequiredaUV light filterwaspositioned25cmabovethemoisturechamber
The Mann Whitney test was used to determine the statistical significance of any
differencesobservedasdescribedinSection213
53 Results
ThinfilmsofAg‐TiO2andTiO2weresuccessfullysynthesisedusingthesol‐gelmethod
ofdeposition(Figure51)Controlthinfilmsconsistingofjustsilvernanoparticleswere
alsoproducedbutthesecoatingswereunstabledemonstratingtheessential roleof
theTiO2under‐layer foradherenceof the silvernanoparticles to theglass substrate
The TiO2 and Ag‐TiO2 thin films were well adhered to the glass substrates after
applicationandremovalofscotchtapeandwereresistanttoscratchingbyfingernails
aHBpencila2HpencilandasteelscalpelBoththinfilmswereeasilyscratchedwitha
diamondtippencilThethinfilmswerestableafterimmersioninmethanolacetone
distilledwateror2MHClfor2hoursbutweredissolvedin2MNaOH
174
Figure51PhotographoftheAg‐TiO2thinfilmsThepurplecolouredthinfilm(left)wasstoredinthedarkandtheorangecolouredthinfilm(right)wasirradiatedwithUVlighttoinducethecolourchange
Thethinfilmswereuniformlyadheredtotheglassmicroscopeslidesandwereorange
incolourandtransparentwhensynthesisedAfterstorage inthedarkforat least72
hoursthethinfilmsturnedpurplereversiontotheorangecolourcouldbeinducedby
irradiationwith UV light for 10minutes or standard indoor lighting conditions for 1
hourThereversiblephoto‐inducedcolourchangecanbedescribedusingthefollowing
formula
Silveroxide(purple) silver(orange)+oxygen
To confirm this orange and purple thin films were placed inside separate Schlenk
flasksandtheairwasevacuatedThepurplesamplewasirradiatedwithUVlightinthe
createdvacuumandturnedorangeHoweverwhentheorangethinfilmswerestored
in the dark for 72 hours the orange colour remained indicating that oxygen was
hv+TiO2
air
175
required for the backward reaction and light exposurewas needed for the forward
reaction
531 Characterisationandfunctionalassessmentofthethinfilms
5311 UV‐visiblespectroscopy
ThinfilmsofAg‐TiO2andTiO2werepreparedusingquartzastheunderlyingsubstrate
inplaceofglassasitallowedbettermeasurementofthebandonsetusingaTaucplot
withouttheinterferenceoftheunderlyingglassbandonsetexpectedatabout33eV
TheUV‐visible‐IRspectroscopyresultsaredisplayedinFigure52andtheAg‐TiO2and
TiO2arevery similar TheAg‐TiO2 thin filmshoweda smalldecrease in transmission
due to silver ions on the surface and a minimal red shift compared with TiO2 The
uncoatedquartzslideshowednofeaturesabove300nm
176
0
10
20
30
40
50
60
70
80
90
100
200 700 1200 1700 2200
Wavelength
T
Qaurtz
TiO2
Ag-TiO2
Figure 52 Transmission data of the Ag‐TiO2 and TiO2 thin films deposited onto aquartzsubstrateobtainedbyUV‐visible‐IRspectrometry
ThethicknessoftheAg‐TiO2andTiO2thinfilmswereestimatedat211nmand196nm
respectivelyusingtheSwanpoelmethodwhich indicatedthatadditionofsilverhad
littleeffectonthethicknessofthethinfilmsThethicknessofthinfilmssynthesised
from the same sol can vary by 10 nm suggesting that the difference observed
betweentheAg‐TiO2andTiO2thinfilmswasunsubstantial
ThebandonsetoftheAg‐TiO2andTiO2thinfilmswereestimatedusingtheUV‐visible‐
IRdatatoproduceTaucplots(Figure53)Theincorporationofsilverontothesurface
of the TiO2 caused a shift in the bandonset towards lower energy radiationwith a
shift from 32 eV for titania to 29 eV for the silver‐doped titania This indicates an
interactionbetweensilverandthetitaniasubstratecausingashifttowardsactivation
inthevisibleregionofthespectrum
177
0
20
40
60
80
100
120
140
160
180
200
00 05 10 15 20 25 30 35 40Energy eV
(ah
v)1
2320 eV29 eV
0
50
100
150
200
250
00 05 10 15 20 25 30 35 40Energy eV
(ah
v)1
2
320 eV
Figure53TaucplotsoftheUV‐visible‐IRdatatakenforthe(a)Ag‐TiO2and(b)TiO2thinfilmspreparedonquartzsubstrates
5312 Contactanglemeasurements
When the Ag‐TiO2 thin film was exposed to UV light the water contact angle
decreasedfrom60degto8degasthesurfacebecamesuperhydrophilic(Table51)Asimilar
decreaseinwatercontactanglewasobservedontheTiO2thinfilmafterexposureto
UVlight(64degto8deg)Thewatercontactangleontheuncoatedglassslidedidnotchange
afterirradiationwithUVlightalthoughtheinitialreadingwascomparativelylow
ThesamplesweresubsequentlyexposedtowhitelightusingtheOptivexUVfilterto
eliminate any higher energy photons of light and the UV‐visible IR spectrum of this
178
filter isdisplayed inFigure54which showsalmost zero transmissionof lightbelow
400nmThedecreaseinwatercontactangleontheAg‐TiO2thinfilmwasthesameas
thatobservedafterUV irradiation(Table51)Thefilteredwhite lightsourcedidnot
haveaneffectontheTiO2thinfilmandtherewasnosubstantialchangeinthewater
contactangleTheseresultsclearlydemonstratethevisiblelight‐inducedhydrophilicity
oftheAg‐TiO2thinfilms
Table51ThewatercontactanglesoftheAg‐TiO2thinfilmsandthecontrolsamplesMeasurementsareaccuratetoplusmn2deg
Samplename Lightsource Watercontactangle
Uncoatedglassslide None 25(2)deg
UV 24(2)deg
TiO2 None 64(2)deg
UV 8(2)deg
Filteredwhitelight 60(2)deg
Ag‐TiO2 None 60(2)deg
UV 8(2)deg
Filteredwhitelight 8(2)deg
179
0
10
20
30
40
50
60
70
80
90
100
200 300 400 500 600 700 800 900 1000 1100
Wavelength nm
T
Figure 54 UV‐Vis spectrum for the Optivextrade UV filter showing the cut‐off forradiationbelow400nminwavelength
5313 Photo‐oxidationofstearicacid
Theeffectofthe lightsourcesontheconcentrationofstearicacidonthesurfaceof
theuncoatedglassslide is illustrated inFigure55aFigure56aandFigure57aThe
heightsofthelinesonthegraphrepresenttimewiththehighestpeakscorresponding
to the shortest irradiation timeTheuncoatedglass slidesdidnot showany signsof
photo‐activityafterexposuretoanyofthethreelightingconditionsandtherewasno
appreciabledecrease in the concentrationof stearicacid detectedon the surfaceof
the samples after the exposure periods Significant destruction of stearic acid was
demonstratedontheTiO2andAg‐TiO2thinfilmsafterexposuretothe254nmUVlight
source(Figure55bandFigure55c)andafter29hoursthepeakshaddisappearedThe
rateofstearicaciddestructionforboththeTiO2andAg‐TiO2thinfilmswascalculated
tobeapproximately11x1014moleculescm2perhourbasedupontheassumption
that1unitofintegrationbetween2700and3000cmequatedtoapproximately97x
180
1015moleculescm2(MillsandWang2006)Thereforesilverdopingdidnothavean
effectonthephoto‐oxidisationofstearicacidafterirradiationwithUVlight
Whenthewhitelightwasusedastheirradiationsourceasignificantdecreaseinthe
stearicacid concentrationwasdemonstratedon theAg‐TiO2 thin films (Figure56c)
whereasaminimal reductionwasobservedon theTiO2 thin films (Figure56b)The
rateofstearicaciddestructionfortheTiO2andAg‐TiO2thinfilmswerecalculatedto
be approximately 16 x 1014 and 42 x 1014 respectively (Table 52) However TiO2
shouldnotdisplayanyphoto‐activityafterirradiationwiththewhite lightsourceand
activationshouldonlyoccurafterexposuretowavelengthsoflightbelow385nmas
thebandonsetofTiO2 is32eVTherefore theOptivextradeUVfilterwasfittedtothe
light box to eliminate any higher energy photons of light The photo‐oxidation of
stearic acid on the TiO2 thin film was seriously compromised and only a negligible
changeintheconcentrationofthecompoundwasobserved(Figure57b)Incontrast
thephotocatalyticactivitywasretainedontheAg‐TiO2thinfilms(Figure57c)which
was shown to be 200 timesmore effective at destroying stearic acid than the TiO2
control(Table52)Thisisthefirstunequivocalevidenceofvisiblelightphotocatalytic
destructionofstearicacid(Dunnilletal2011)
181
-002
000
002
004
006
008
010
28002850290029503000
Wavenumber cm-1
Absorb
tion
0
5
24
29
48
53
72
-002
000
002
004
006
008
010
28002850290029503000
Wavenumber cm-1
Absorb
tion
0
5
24
29
-002
000
002
004
006
008
010
28002850290029503000
Wavenumber cm-1
Absorb
tion
0
5
24
29
Figure55IRabsorptiondatadisplayingthephoto‐oxidationofstearicacidmoleculeson the surface of the threematerials over 72 hours using a 254 nm light sourcewherea)uncoatedglassslideb)TiO2andc)Ag‐TiO2Linetimesareshowninorderof height on the graph and in all cases the area under the curve indicates theamountofstearicacidremainingonthesurface
a
b
c
182
-002
000
002
004
006
008
010
012
28002850290029503000
Wavenumber cm-1
Absorb
tion
0 h
24 h
48 h
72 h
96 h
-002
000
002
004
006
008
010
012
014
016
018
28002850290029503000
Wavenumber cm-1
Absorb
tion
0 h
24 h
48 h
72 h
96 h
-002
000
002
004
006
008
010
012
014
28002850290029503000
Wavenumber cm-1
Absorb
tion
0 h
24 h
48 h
72 h
96 h
Figure56IRabsorptiondatashowingthephoto‐oxidationofstearicacidmoleculesonthesurfaceofthethreematerialsover96hoursusingawhitelightsourcewherea) uncoated glass slide b) TiO2 and c) Ag‐TiO2 Line times are shown in order ofheightonthegraphandinallcasestheareaunderthecurveindicatestheamountofstearicacidremainingonthesurface
a
b
c
183
Figure 57 Rawdata showing the photo‐oxidationof stearic acidmolecules on thesurface of the three samples over 500 hours using a white light source and theOptivextrade UV filter where (a) uncoated glass slide (b) TiO2 and (c) Ag‐TiO2 Linetimes are shown in order of height and in all cases the area under the curveindicatestheamountofstearicacidremainingonthesurface
a
b
c
184
Table52Thenumberofmoleculesofstearicacidphoto‐oxidisedduringirradiationbythedifferentlightsourcesRatesaregivenasmoleculescm2perhourExposuretimestotheUVwhitelightandfilteredwhitelightwere29hours96hoursand500hoursrespectively
TiO2 Ag‐TiO2
Lightsource Moleculesoxidised
RateMoleculesoxidised
Rate
UVndash254nm 332x1016 114x1015 330x1016 114x1015
Whitelight 149x1016 155x1014 405x1016 422x1014
Filteredwhitelight 149x1016 299x1011 312x1016 625x1013
532 AntibacterialactivityagainstEcoliATCC25922
Theantibacterial activityof the thin filmswasassessedagainstEcoliAfter2hours
irradiationwithwhitelighta09log10cfusampledecreasewasobservedcompared
withboth the uncoated controlsand theTiO2 controlsexposed to the same lighting
conditions (Figure58) Thedecrease inbacterial recoverywasmuchgreaterafter6
hours irradiationwith thewhite light sourceE coliwasnot recovered from theAg‐
TiO2thinfilmsafterthe6hourexposureperiodonanyoftheexperimentalrepeats
Thisreductioncorrespondstoa48 log10cfusampledecreaseinbacterialrecovery
comparedwiththeglasscontrolsexposedtothesamelightingconditions(plt0001)
ThedecreaseinrecoverywasslightlylesswhencomparedtotheTiO2thinfilmsbuta
statistically significant 44 log10 cfu sample decreasewas still achieved (p lt 0001)
However E coli could not be recovered from the Ag‐TiO2 thin films which were
incubated in thedark for the6 hour incubation period indicating that theobserved
antibacterialactivityobservedwasnotlight‐dependent
185
Figure58 Effectof the thin filmAg‐TiO2on the survivalofE coli Thin filmswereirradiatedwithwhitelight(L+)orincubatedinthedarkfor2hours(L‐)TheuncoatedglassslidesTiO2andAg‐TiO2arerepresentedbylsquoUnrsquolsquoTirsquoandlsquoAgrsquorespectively
Figure59 Effectof the thin filmAg‐TiO2on the survivalofE coli Thin filmswereirradiatedwithwhitelight(L+)orincubatedinthedarkfor6hours(L‐)TheuncoatedglassslidesTiO2andAg‐TiO2arerepresentedbylsquoUnrsquolsquoTirsquoandlsquoAgrsquorespectively
186
TheantibacterialactivityoftheAg‐TiO2thinfilmswasfurtherassessedtheexposure
periodwasextendedto12hoursandonceagainitwasnotpossibletorecoverEcoli
fromtheAg‐TiO2thinfilmsaftertheincubationtimeandthiseffectwasindependent
of light exposure (Figure 510) Interestingly the activity of the TiO2 thin films
increasedwithextendedexposuretowhitelightanda24log10cfusampledecrease
inbacterial recoverywasobserved comparedwith theglass controlexposed to the
samelightingconditionsThisfindingsupportstheresultsfromthefunctionaltesting
whichdemonstratedphoto‐oxidationofstearicacidafterexposuretothiswhitelight
sourceThereforetheOptivextradeUVfilterwasplacedabovethemoisturechamberto
eliminatetheUVcomponentofthewhitelightsourceTheantibacterialactivityofthe
TiO2thinfilmswaseliminated(Figure511) the reductionobservedontheTiO2thin
filmswasnegligible (002 log10cfusampledecrease)The light‐independentactivity
of the Ag‐TiO2 thin films was retained and the decrease in bacterial recovery was
maintained at 49 log10 cfu sample on the Ag‐TiO2 thin films in the presence and
absenceoffilteredlight
187
Figure510EffectofthethinfilmAg‐TiO2onthesurvivalofEcoliThinfilmswereirradiated with white light (L+) or incubated in the dark for 12 hours (L‐) Theuncoated glass slide TiO2 and Ag‐TiO2 are represented by lsquoUnrsquo lsquoTirsquo and lsquoAgrsquorespectively
Figure511EffectofthethinfilmAg‐TiO2onthesurvivalofEcoliThinfilmswereirradiatedwithwhitelightfilteredwiththeOptivextradeglass(L+)orincubatedinthedarkfor12hours(L‐)TheuncoatedglassslideTiO2andAg‐TiO2arerepresentedbylsquoUnrsquolsquoTirsquoandlsquoAgrsquorespectively
188
The antibacterial activity of theAg‐TiO2 thin filmswere further determined after 18
hours exposure to thewhite light source The light‐independent activity of the thin
filmswasmaintainedanda46 log10cfu sampledecrease intherecoveryofEcoli
was observed compared with the glass controls exposed to the same lighting
conditions (p lt0001)No re‐growthofE coliwasobservedoneither the thin films
incubated in the presence or absence of light indicating a sustained antibacterial
effect Aminimal decrease in the recovery ofE coliwas observed on the TiO2 thin
filmsafterthe18hourincubationperiod(03log10cfusample)whichparadoxically
wasmuchlessthanthatseenafter12hoursThisdifferencewashoweverstatistically
significant(plt001)ThewhitelightalonedidnothaveaneffectonthesurvivalofE
coliontheuncoatedcontrolslidesandnosignificantdifferenceinbacterialrecovery
wasobservedonthesesamplesafterincubationinthepresenceorabsenceofwhite
lightwhichimpliesthatthephoto‐activityobservedontheTiO2thinfilmswasnotdue
totheeffectofthewhitelightsourcealone
189
Figure512EffectofthethinfilmAg‐TiO2onthesurvivalofEcoliThinfilmswereirradiated with white light (L+) or incubated in the dark for 18 hours (L‐) Theuncoated glass slide TiO2 and Ag‐TiO2 are represented by lsquoUnrsquo lsquoTirsquo and lsquoAgrsquorespectively
533 AntibacterialactivityagainstEMRSA16
TheantibacterialactivityofthethinfilmswasassessedagainstEMRSA‐16A03log10
cfu sample decrease in the recovery of EMRSA‐16 was observed after 6 hours
irradiation with white light (Figure 513) compared with the uncoated glass slides
exposedtothesamelightingconditionswhichdidnotreachstatisticalsignificance
190
Figure 513 Effect of the thin filmAg‐TiO2 on the survival of EMRSA‐16 Thin filmswere irradiatedwithwhite light (L+)or incubated in thedark for6hours (L‐) Theuncoated glass slide TiO2 and Ag‐TiO2 are represented by lsquoUnrsquo lsquoTirsquo and lsquoAgrsquorespectively
TheAg‐TiO2thinfilmsweresubsequentlyexposedto12hourswhite lightanda26
log10 cfu sample decrease in the recovery of EMRSA‐16 was observed (p lt001)
comparedwith the uncoated glass slides (Figure 514)Negligible photo‐activitywas
observedontheTiO2thinfilmsandtherewasan insignificantdifferenceobserved in
the recovery from the irradiated TiO2 thin films compared to those incubated in the
dark (02 log10 cfu sampledecrease) Theantibacterialeffectappeared to be light‐
dependentandtherewasa23log10cfusampledifferenceintherecoveryofEMRSA‐
16 from the irradiated Ag‐TiO2 thin films comparedwith the non‐irradiated Ag‐TiO2
thinfilms(plt001)anda26log10cfusampledifferenceintherecoveryofEMRSA‐16
fromtheuncoatedirradiatedsamples(plt0001)
191
Figure 514 Effect of the thin filmAg‐TiO2 on the survival of EMRSA‐16 Thin filmswere irradiatedwithwhite light (L+)or incubated inthedarkfor12hours(L‐)Theuncoated glass slide TiO2 and Ag‐TiO2 are represented by lsquoUnrsquo lsquoTirsquo and lsquoAgrsquorespectively
TheexperimentwasrepeatedwiththeOptivextradeUVfilterinsitutoeliminateanystray
photons of sub 400 nm light and the antibacterial activity of theAg‐TiO2 thin films
decreased(Figure515)A11log10cfusamplereductionintherecoveryofEMRSA‐
16 was observed compared with the uncoated sample irradiated with the same
filteredlightsource(plt0001)Theminimalphoto‐activityobservedontheTiO2thin
films in the presence of unfilteredwhite light wasmaintained and a 02 log10 cfu
sampledecreasewasdetectedcomparedwiththeuncoatedsamples irradiatedwith
filteredwhitelightThisdifferencewasnotstatisticallysignificant(pgt005)
192
Figure 515 Effect of the thin filmAg‐TiO2 on the survival of EMRSA‐16 Thin filmswereirradiatedwithwhitelightfilteredwiththeOptivextradeglass(L+)orincubatedinthe dark for 12 hours (L‐) The uncoated glass slides TiO2 and Ag‐TiO2 arerepresentedbylsquoUnrsquolsquoTirsquoandlsquoAgrsquorespectively
TheAg‐TiO2thinfilmsweresubsequentlyirradiatedwithwhitelightfor18hoursand
theresultsareshowninFigure516A34log10cfusamplereductionintherecovery
of EMRSA‐16was observed comparedwith the glass controls exposed to the same
lighting conditions (p lt 0001) The light‐dependent activity of the thin films was
sustainedanda29log10cfusampledecreaseinbacterialrecoverywasobservedon
the irradiated Ag‐TiO2 thin films compared with those incubated in the dark (p lt
0001) However significant photo‐activity was detected on the TiO2 thin films
althoughthiseffectwasextremelyinconsistentasindicatedonthegraphbythelarge
errorbarsandwasalsolessstatisticallysignificant(plt005)A34log10cfusample
decrease in the recovery of EMRSA‐16was observed comparedwith the uncoated
glasscontrolsexposedtothesamelightingconditionsNoactivitywasdetectedonthe
TiO2thinfilms incubated inthedark indicatingthattheactivitywas lightdependent
andcouldonceagainbeduetotheUVcomponentofthewhitelightsource
193
Figure 516 Effect of the thin filmAg‐TiO2 on the survival of EMRSA‐16 Thin filmswere irradiatedwithwhite light (L+)or incubated inthedarkfor18hours(L‐)Theuncoated glass slides TiO2 and Ag‐TiO2 are represented by lsquoUnrsquo lsquoTirsquo and lsquoAgrsquorespectively
Therefore the Optivextrade filter added and the samples were irradiated with filtered
white light (Figure 517) The antibacterial activity of the Ag‐TiO2 thin films was
retained but at a reduced rate the average decrease in bacterial recovery dropped
from34 log10cfusampleto23 log10cfusampleusingtheunfilteredandfiltered
whitelightsourcesrespectivelyThisresultmirrorsthatseenafter12hoursirradiation
with the filtered light sourceand remainedhighly statistically significant (plt0001)
ThelightdependentactivityoftheAg‐TiO2thinfilmswasalsoreplicatedand14log10
cfu sample decrease in bacteriawas observed on the irradiatedAg‐TiO2 thin films
comparedwiththoseincubatedinthedark(plt005)butagainthisreductionwasless
thanthatobservedwhentheunfilteredlightsourcewasused
194
Figure 517 Effect of the thin filmAg‐TiO2 on the survival of EMRSA‐16 Thin filmswereirradiatedwithwhitelightfilteredwiththeOptivextradeglass(L+)orincubatedinthedarkfor18hours(L‐)TheuncoatedglassslideTiO2andAg‐TiO2arerepresentedbylsquoUnrsquolsquoTirsquoandlsquoAgrsquorespectively
Themostsurprisingresultwastheretainedphoto‐activityoftheTiO2thinfilms(Figure
517) the photo‐activity was reduced when filtered white light was used as the
irradiationsourcebutastatisticallysignificant31log10cfusampledecreaseinviable
bacteriawasstillobserved(plt001)whichwasagreaterdecreasethanthatseenon
theAg‐TiO2thinfilmsAwiderangeofbacterial recoverywasobservedindicatedby
the large box on the graph on occasion no bacteria were recovered at all and on
otherexperimentalreplicatesthenumberofcoloniespresentequalledthatobserved
from the control samples incubated in the dark The bacterial recovery from the
control samples Ag‐TiO2 and TiO2 which were incubated in the dark was also
significantly lower than theuncoatedglass samples incubated in thedark (plt001)
Furthermore the values obtained from the TiO2 thin film incubated in the darkwas
significantly lower than that obtained in the previous 18 hour experiment (Figure
516)
195
54 Discussion
Silverhasbeenshownboth inthischapterand inthe literature to improvetitanium
dioxide photo‐activity and this is achieved through three mechanisms The first
involvesreductionofsilverionstosilverbyphoto‐excitedelectronsTheelectronsare
furtherattractedtosilverparticlesinthefollowingreactionwherethesilverparticles
actaselectrontraps(Herrmannetal1997Heetal2002Brooketal2007b)
(Ag)+e‐ e‐Ag
The electrons move to the interior of the thin film and the holes move to the
interfacial region which enhances their separation and inhibits electron‐hole
recombination The photo‐generated holes then react with surface hydroxyl groups
and water to form hydroxyl radicals and other reactive species which possess
antibacterial activity (Sclafani et al 1991 Herrmann et al 1997 Stathatos et al
2001 He et al 2002) Secondly the electric field around the silver particles is
increased by surface plasmon resonance effects which further enhance photo‐
excitationoftheelectronsandelectron‐holeseparation(Zhaoetal1996)Finallythe
surface roughness of the titaniumdioxide thin film changes upon silver addition so
that the titanium dioxide particle size in the resultant thin films is smaller which
exposes a greater surface area available for photo‐reactionwhich further increases
photo‐activity(Herrmannetal1997Heetal2002Martinez‐Gutierrezetal2010)
Therefore thepropertiesofaphotocatalyst can beadaptedby reducing theparticle
sizetocoupletheintrinsicbandonsetpropertiestoallowlowerenergyphotocatalysis
(Herrmannetal1997Heetal2002Dunnilletal2011)
196
541 Synthesisofthesilver‐dopedtitaniathinfilms
Analogoustonitrogenandsulfurdopingoftitaniathesilverconcentration iscritical
and a decrease in the photo‐activity of the thin films will occur if the silver
concentrationexceedsanoptimumlevel(Sclafanietal1991DoboszandSobczynski
2003 Brook et al 2007b) This is due to the lsquoscreening effectrsquo where the silver
depositedonthesurfaceofthethinfilmmasksthephoto‐reactivesitessothatthey
are inaccessible for interaction with photons (Dobosz and Sobczynski 2003) In
additionthenegativelychargedsilverparticlesonthethinfilmcouldattracttheholes
beforeanyinteractionwithwaterwhichwoulddecreasetheconcentrationofreactive
oxygenspeciesgeneratedandtheobservedphoto‐activity(Heetal2002)
Sol‐geldepositionwasusedtosynthesisethethinfilms inthischapter incontrastto
APCVDwhichwasused togenerate the thin filmsassessed in theprevious chapter
APCVD was initially chosen as a deposition method as the resultant coatings are
transparentrobustandstronglyadheredtothe substrateSol‐gel filmsaregenerally
thicker less mechanically robust and require sintering after coating to anneal the
coating to the substrate (Brook et al 2007b) A post‐coating annealing step was
includedinthesol‐gelmethodofsynthesissothethinfilmsgeneratedinthischapter
were well adhered to the substrate and as mechanically stable as the APCVD
generatedthinfilms
197
542 Characterisationand functionalassessmentof thesilver‐dopedtitania
thinfilms
The silver‐coated titania thin films exhibited photo‐chromic behaviour which was
causedbyachangeintheoxidationstateofthesilvernanoparticlesfromsilveroxide
tometallicsilver(Ohkoetal2003Paramasivametal2007Gunawanetal2009)
BothUVandvisiblelightwereabletoinducethemorecolouredorangemetallicstate
and the less coloured purple oxide state occurred after storage in the dark Excited
electronsgeneratedduring lightexposurephoto‐reactedwith the silver ionspresent
withinthepurplefilmandthefilmsturnedorangeasthesilveroxidewasreducedto
silvermetal(Ohtanietal1987)Whenthefilmsweresubsequentlystoredinthedark
inthepresenceofairthephoto‐reducedsilverwasoxidisedformingsilveroxideand
the films reverted to the purple colour due to a decrease in light absorbance
(Paramasivametal2007)Thesechangesarecausedbysurfaceplasmonresonance
effects which in turn are influenced by the nanoparticle size shape and the local
refractiveindex(Jinetal2001Mocketal2002Ohkoetal2003Gunawanetal
2009)
Thebandonsetofthesilver‐coatedtitaniathinfilmshadshiftedto29eVtowardsthe
visible regionof theelectromagnetic spectrumwhich in theabsenceofparticle size
modification indicated doping of silver nanoparticles within the titanium dioxide
structureWehadpreviouslyshownthatdopingtitaniathinfilmswitheithernitrogen
orsulfurcausedashiftinthebandonsetto29eVand30eVrespectivelyindicating
thatthesethinfilmswouldmakebetterwhitelightphotocatalyststhantitaniaaloneA
lowerbandonsetfromsilver‐dopedtitaniasampleshasbeenreportedabandonset
198
of 26 eV was estimated by Medina‐Ramirez et al (2011) although these were
nanoparticulatecompositesandnotthinfilmsTheobservedshifttowardsthevisible
spectrum could also be partly due to mixing of the band onsets silver oxide at
approximately1eVforAgOand14eVforAg2O(Idaetal2008Rajuetal2009)
Thewatercontactangleofthethin filmswasmeasuredtodetermineanychange in
the hydrophilicity of the surface after irradiation with the different light sources
Superhydrophilicity occurs after photo‐oxidation of hydrocarbons adsorbed onto the
substrate which results in the production of a hydroxylated surface (Zubkov et al
2005) Predictably thewater contact angle of the titania thin films decreased after
irradiation with the UV light source (Mills and LeHunte 1997 Parkin and Palgrave
2005)andthewatercontactangleofthesilvercoatedtitaniathinfilmsalsodecreased
byasimilaramountTheadditionofsilvernanoparticlestothesurfaceofthetitania
thinfilmwaspredictedtoresult inanalterationofthehydrophilicityofthethinfilm
prior to light exposure as the surface roughness of the thin film had changed and
largercontactanglesareusuallyfoundonroughersurfaces(Wenzel1936Cassieand
Baxter 1944) but these data show this effect is insignificant even though silver
coverageofthesurfacereached64(Dunnilletal2011)IrradiationwithUVlightdid
nothaveaneffecton thewater contactangleon theuncoatedglass slidealthough
thewatercontactangleontheslidewasinitiallylowTheexpectedcontactangleona
glasssurfaceisapproximately70degandthelowreadingobservedintheseexperiments
indicatedthattheglasssubstratewasinaverycleancondition(Zubkovetal2005)
Thevisiblelight‐inducedhydrophilicityofthethinfilmswasdeterminedbyirradiation
withwhitelightfilteredwithasheetofOptivexglasstoeliminateanystrayhigher
199
energy photons of light with awavelength of less than 400 nm Thewater contact
angle on the silver‐coated titania thin film decreased to the same degree as that
observed after UV irradiation In contrast no change in water contact angle was
observedonthetitaniathinfilmsThis clearlydemonstratesthevisible‐light induced
natureofthesilvercoatedtitaniathinfilms
The photo‐oxidisation of stearic acid has been used extensively in the literature to
indicate the photocatalytic activity of novel thin films and estimate their potential
antibacterial activity (Mills et al 2002 Mills andWang 2006 Brook et al 2007a
2007bPageetal2007)TherateofstearicaciddegradationwascalculatedfortheN‐
dopedandS‐dopedthinfilmsassessed inthepreviouschapterafterexposuretothe
white light source The N‐doped sample (N1) displayed a rate of destruction of
approximately 14 x 1014 molecules cm2 per hour and the S‐doped sample (S2)
demonstrated a similar rate of 11 x 1014 molecules cm2 per hour (Dunnill et al
2010)Thesilver‐coatedtitaniathinfilmsgeneratedinthischapterdemonstratedrate
of destruction of approximately 42 x 1014molecules cm2 per hourwhich is three
timesmoreefficientthantheN‐dopedandS‐dopedthinfilmsandtwiceasefficientas
thetitaniumdioxidethinfilmsThisimpliesthatsurfacesilverdopingdoesnotinduce
asmuch electron‐hole recombination as that observed in theN‐doped and S‐doped
titaniawhichresultsinimprovedphotocatalysis
The anatase titanium dioxide thin film should not exhibit any photo‐activity after
irradiationwiththewhitelightsourceandactivationshouldonlyoccurafterexposure
towavelengthsoflightbelow385nmasthebandonsetoftitaniumdioxideis32eV
The photo‐activity observed suggests that therewas light of an increased frequency
200
emitted from the white light source The emission spectrum for the light source is
shown in Figure 21 and no emission is detectable below 410 nm however the
spectrumstartsat380nm so theprofileat lowerwavelengths isnotknownWhite
light sources suchas the fluorescent lampused in theseexperiments can leakvery
small amounts of higher energy photons of light as they age due to the release of
phosphor from the inside of the fluorescent tubing which could explain the photo‐
activitygeneratedonthetitaniumdioxidethinfilm
TheOptivexUVfilterwasemployedoncemoreandthephoto‐activityofthesilver‐
coatedtitaniathin filmswasretainedandthephoto‐activityofthetitaniathinfilms
was terminated This demonstrated the true visible light driven photo‐oxidation of
stearicacidonthesilver‐coatedtitaniathinfilmsTherateofstearicaciddegradation
wasslowerwhentheUVfilterwasemployedpartlybecausetheintensityofthewhite
lightwasreducedasonlyaround80ofemitted lightwasabletotransmitthrough
the glass shield and partly due to the loss of the UV part of the electromagnetic
spectrum
543 Antibacterialactivityofthesilver‐dopedtitaniathinfilms
Theantibacterialpropertiesofthesilver‐coatedtitaniathinfilmswereassessedusing
E coliand EMRSA‐16as representative strainsGram‐negative strains suchasE coli
havebeendemonstratedtobemoredifficulttokillusinglight‐activatedantimicrobial
coatingsthanGram‐positivestrainssuchasMRSA(Decraeneetal2006Pageetal
2009) However in these experiments E coli was eradicated from the silver‐coated
titaniathinfilmsataquickerratethanEMRSA‐16AreductionintherecoveryofEcoli
201
fromthesilver‐coatedtitaniathinfilmswasobservedafterjust2hoursandnoviable
bacteriacouldberecoveredfromthesamplesafter6hoursincubationHoweverthe
observedantibacterialeffectwasindependentoflightexposureasasimilarreduction
in bacterial recovery was observed on the silver‐coated titania incubated in the
absenceoflightwhichillustratestheactivitywasduetothetoxicityofthesilverions
ratherthanalightinducedeffectwhichhasbeendemonstratedintheliterature(Feng
etal2000Kimetal2007Jungetal2008)TheincreasedsusceptibilityofGram‐
negative bacteria to the silver containing thin filmwas postulated to be due to the
thinnerpeptidoglycanlayerinthecellmembranewhichallowsincreaseduptakeinto
the interior of the bacterial cell (Schierholz et al 1998) Conversely Kowal et al
(2011) showed a greater susceptibility of MSSA and MRSA to silver‐doped titania
nanopowderscomparedwithEcoli
EMRSA‐16 has been responsible for a significant proportion of the healthcare‐
associatedcasesofMRSAbacteraemiaoverthelastdecadeandwasshowninChapter
3tobealighttolerantstrainofMRSA(Johnsonetal2001Ellingtonetal2010)The
antibacterial activity of the silver‐coated titania thin films increasedwith prolonged
exposuretowhitelightwiththelargestreductioninbacterialrecoveryobservedafter
18 hours irradiation Enhancement of the photocatalytic properties of the light‐
activatedsurfacebythesilverparticlesandtheenhancementofthetoxicpropertiesof
thesilverbytitaniawasobservedonthesilver‐coatedtitaniawhichdemonstrateda
synergisticrelationshipbetweenthetwocomponentsofthethinfilmThiseffectwas
muchgreaterthanthatobservedwhenthesilver‐coatedtitania filmswereincubated
intheabsenceoflightorwheneitherthetitaniaoruncoatedsampleswereirradiated
202
with white light The silver ions alone appeared to have an effect on EMRSA‐16
especially after a prolonged incubation time but this was less significant than the
effect seenafter lightexposureThe lack ofactivityobservedon theuncoatedglass
slidesdemonstratedthatthewhitelightsourcedidnothaveaninhibitoryeffectonthe
viability of EMRSA‐16 The lack of activity observed on the titania thin film in the
presenceof6or12hourswhite light indicatedthattheUVcomponentofthewhite
lightsourcewasnotsufficienttophoto‐activatethetitaniafilmsHoweverthispattern
wasnotmaintainedandasignificantdifferenceintherecoveryofEMRSA‐16fromthe
irradiatedTiO2thinfilmswasobservedcomparedwiththeuncoatedglassslidesafter
18hoursThiseffectwasnoteliminatedwhentheOptivextradeUVfilterwasappliedThe
significantdecreaseinrecoveryofEMRSA‐16observedontheTiO2thinfilmincubated
inthedarksuggeststhatalight‐independentmechanismofactionwasinvolved
It is possible to conclude that the photo‐induced destruction was due to reactive
oxygenproducedbytitaniadrivenbywhitelightphotocatalysisinducedbythesilver
These effects did not occur in the absence of white light or silver An alternative
explanationcould involvephoto‐assisted releaseof silver ions from the silver‐coated
titaniawhichinturncausedtheantibacterialeffect
Amajor limitation of the experimentswas that the test conditionswere laboratory‐
controlledanddidnottakeintoaccountfactorssuchasorganicsoilwhichwouldbe
presentonhand‐touch surfaces Substancessuchas sebaceousoilsbloodandother
humansecretionswouldbe likelytocontaminatethethinfilms if theywereusedas
antibacterial coatings in a patient environment and the effect of these substances
should be investigated as they are likely to cause an inhibition in the photocatalytic
203
activity of the thin films (Furno et al 2004)Organic soiling of a surface is likely to
precedebacterialcontamination(Verranetal2002)soifthethinfilmswereableto
photo‐degrade any organic soil present it would keep the surface hygienically clean
andeliminateapotentialnutrientsourceofanycolonisingbacteria
55 Conclusion
Thischapterhasdemonstratedthattheantibacterialactivityoftitaniathinfilmscan
be significantly enhanced by the addition of surface‐bound silversilver oxide
nanoparticles The thin films displayed photochromic behaviour and were found as
either silver oxide or pure silver depending on the storage conditions oxidation of
silvertosilveroxideoccurredafterstorageinthedarkandapurplecolourationwhilst
exposuretoindoorlightingconditionscausedphoto‐reductionofthesilveroxideback
to silver and an orange coloured film White light induced photocatalysis was
generatedbyashiftinthebandonsetofthethinfilmscausedbytheadditionofsilver
nanoparticlesVisiblelightphotocatalysiswasdemonstratedwhenaUVfilterwasused
to block out the minimal UV component of the white light source and this was
observed in the form of photo‐oxidation of stearic acid a reduction in the water
contactangleandphotocatalyticactivityagainstEMRSA‐16Thisisthefirstexampleof
unambiguous visible light photocatalysis and photo‐induced superhydrophilicity
alongsideatitaniumdioxidecontrolthatshowsnoactivation
204
6 Assessment of a novel antibacterial material for use inendotrachealtubesinintubatedpatients
61 Introduction
Ventilator‐associatedpneumonia(VAP)isaHCAIassociatedwithsignificantmorbidity
and mortality Intubated patients have an endotracheal tube (ETT) in situ to allow
mechanicallyassistedbreathingwhichcompromises thenormal clearanceofmucus
and other upper airway secretions and allows micro‐aspiration of contaminated
subglotticsecretionsintothelungsThesesecretionscontaincommensalbacteriathat
provide a source for pulmonary infection In addition the lumen of the ETT itself
becomes colonised with bacteria which provides a secondary source of infective
organisms (Deem and Treggiari 2010) A number of studies investigating the
microbiology of VAP have shown that Gram‐negative bacilli are isolated more
commonly in patients with VAP compared with patients with hospital‐acquired
pneumonia (ie pneumonia acquired in hospital in the absence of mechanical
ventilation) P aeruginosa Acinetobacter species and S maltophilia are the most
commonly observed Gram‐negative pathogens causing VAP (Johanson et al 1972
Richards et al 1999 Weber et al 2007 Bouadma et al 2010) Both meticillin‐
sensitive and resistant S aureus have also been isolated but were observed more
frequentlyinnon‐intubatedpatients(Weberetal2007)
It is advantageous to reduce microbial load and decrease biofilm formation in the
lumenoftheETTasthiswouldeliminatethebacterialreservoirand lowertheriskof
developing VAP The use of antimicrobial silver ETTs has been recommended in
combinationwithadditionalclinicalmeasures inthepreventionofVAP(Torresetal
205
2009 Coppadoro et al 2011) and it would be desirable to expand on the pool of
antimicrobialETTsavailablePhotodynamicinactivation(PDI)ofbacteriahasprovento
beaneffectivemethodofreducingthebacterialloadonsurfacesandthistechnology
has the potential to be applied to an ETT A laser light could be inserted along the
length of the ETT and switched on periodically to activate the surface and kill any
bacteriapresentFigure61showshowthismaybeachievedinacathetertube
Figure61Acathetertube impregnatedwiththephotosensitisingagentmethyleneblueItissuggestedthatlightfromalasercouldbeprojectedthroughthetubewiththeuseoffibreopticsPhotographcourtesyofProfWilson(UCL)
This chapter describes the development of a polyurethane polymer which was
impregnatedwiththephotosensitisingagenttoluidineblueO(TBO)Theantibacterial
effect of the impregnated polymers after irradiation with laser light was observed
206
againstaseriesofpathogensknowntocauseVAPBothclinicalandtypestrainswere
tested to assess any difference in susceptibility to PDI The published literature
describedabovewasusedtoguidethechoiceofbacteriaandmaterialtypeassessed
inthischapter
62 Materialsandmethods
621 Materialsynthesis
Thepolyurethanepolymersrequiredforthisseriesofexperimentsweresynthesisedas
described inSection2103PolymerswerepreparedcontainingTBO(S+)andcontrol
polymerswerepreparedinparallelwithouttheadditionofTBO(S‐)
622 Measuring the antibacterial photo‐activity of the TBO‐impregnated
polymers
BacterialstrainsweremaintainedasdescribedinSection21andbacterialsuspensions
of P aeruginosa PAO1 and clinical strains of P aeruginosa A baumannii and S
maltophiliawerepreparedasdetailedinSection23resultinginastartinginoculumof
approximately107cfumlwhichequatedtoaconcentrationofapproximately106cfu
polymerasdescribedinSection2123AsuspensionofCalbicans(107cfuml)was
alsopreparedasdescribed inSection23TheMannWhitneyUtestwasusedforall
statistical analyses to determine the statistical significance of any differences
observed as described in Section 213 The nomenclature used during this series of
experimentsisdetailedinTable61
207
Table 61 Nomenclature used during microbiological assessment of the TBO‐impregnatedpolymers
63 Results
631 Assessmentoftheantibacterialphoto‐activityoftheTBO‐impregnated
polymersagainstPaeruginosaPAO1atypestrain
TheactivityoftheTBO‐impregnatedpolyurethanepolymerswasfirstassessedagainst
atypestrainofPaeruginosaPAO1Thepolymerswereexposedtothelaserlightfor
timeperiodsofbetween30secondsand240secondsandtheresultsareillustratedin
Figure62throughtoFigure610
Nomenclature Description
L+S+ TBO‐impregnatedsampleexposedtolaserlight
L+S‐ TBO‐impregnatedsampleNOTexposedtolaserlight
L‐S+ NonTBO‐impregnatedsampleexposedtolaserlight
L‐S‐ NonTBO‐impregnatedsampleNOTexposedtolaserlight
208
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure62AntibacterialactivityofTBO‐impregnatedpolyurethanepolymeragainstPaeruginosaPAO1after30secondsThedottedhorizontallineindicatesthedetectionlimitofthesamplingmethod080log10cfupolymer
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure63AntibacterialactivityofTBO‐impregnatedpolyurethanepolymeragainstPaeruginosaPAO1after60secondsThedottedhorizontallineindicatesthedetectionlimitofthesamplingmethod080log10cfupolymer
209
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure64AntibacterialactivityofTBO‐impregnatedpolyurethanepolymeragainstPaeruginosaPAO1after90secondsThedottedhorizontallineindicatesthedetectionlimitofthesamplingmethod080log10cfupolymer
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure65AntibacterialactivityofTBO‐impregnatedpolyurethanepolymeragainstPaeruginosa PAO1 after 120 seconds The dotted horizontal line indicates thedetectionlimitofthesamplingmethod080log10cfupolymer
210
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure66AntibacterialactivityofTBO‐impregnatedpolyurethanepolymeragainstPaeruginosa PAO1 after 150 seconds The dotted horizontal line indicates thedetectionlimitofthesamplingmethod080log10cfupolymer
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure67AntibacterialactivityofTBO‐impregnatedpolyurethanepolymeragainstPaeruginosa PAO1 after 180 seconds The dotted horizontal line indicates thedetectionlimitofthesamplingmethod080log10cfupolymer
211
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure68AntibacterialactivityofTBO‐impregnatedpolyurethanepolymeragainstPaeruginosa PAO1 after 210 seconds The dotted horizontal line indicates thedetectionlimitofthesamplingmethod080log10cfupolymer
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure69AntibacterialactivityofTBO‐impregnatedpolyurethanepolymeragainstPaeruginosa PAO1 after 240 seconds The dotted horizontal line indicates thedetectionlimitofthesamplingmethod080log10cfupolymer
Highly statistically significant reductions in the numberof viablePaeruginosa PAO1
recoveredfromtheTBO‐impregnatedpolymerswasobservedatalltimepointstested
212
(allplt0001)Thereductioninbacterialcountfollowedadose‐dependentresponse
whereby as the dose of laser light was increased the antibacterial activity of the
impregnatedpolymers increasedwhich resulted ina lower recoveryofbacteria For
examplea141log10cfupolymerdecreasewasobservedafter90secondsexposure
to the laser light (Figure65) rising toa294 log10 cfu polymerdecreaseafter180
seconds(Figure67)anda333log10cfupolymerdecreaseafter240seconds(Figure
69)TheresultsfromalloftheexperimentsaresummarisedinTable62
Table62SummaryofthedataobtainedfromthePaeruginosaPAO1experimentsThestatedreductions inbacteriaarecalculatedbycomparingthemedianbacterialrecoveryfromtheL‐S‐samplewiththeL+S+sample
ExposuretimesecondsLogreductioncfuper
polymerPercentagereduction
cfuperpolymer
30 044 639
60 049 679
90 141 961
120 209 992
150 282 9985
180 294 9989
210 305 9991
240 333 9995
Theobservedreductions inbacterial recoverywerehighlystatisticallysignificant (plt
0001) at all time points (L‐S‐ comparedwith L+S+)which demonstrates the potent
light‐dependent antibacterial activity of the TBO‐impregnated polymers When the
twogroupsofTBO‐impregnatedpolymerswerecomparedandtheeffectofthe laser
213
lightwas investigated (L‐S+ and L+S+) the recovery ofP aeruginosa from the TBO‐
impregnatedpolymersexposedtolightwassignificantlylowerthanrecoveryfromthe
TBO‐impregnated polymers incubated in the dark This difference was highly
statisticallysignificant(plt0001)foralltimepointsabove60secondsthedifference
wasalsostatisticallysignificantafter30secondswithapvalueofplt001Thesedata
further confirm the photocatalytic nature of the TBO‐impregnated polymers There
wasno statisticaldifference in thebacterial recoveryobtained from the twosetsof
polymers incubated in the dark (L‐S‐ compared with L‐S+) which demonstrates the
intrinsic lackofantibacterialactivityofTBO intheabsenceof lightofanappropriate
wavelength
632 Assessmentoftheantibacterialphoto‐activityoftheTBO‐impregnated
polymersagainstaclinicalstrainofPaeruginosa
The photo‐activity of the TBO‐impregnated polyurethane polymers was assessed
againstaclinicalstrainofPaeruginosatoassesswhethertherewereanydifferences
in the susceptibility of the laboratory type strain compared with a strain recently
isolatedfromapatientwithclinicallyconfirmedVAPThepolymerswereexposedto
thelaserlightfortimeperiodsof90seconds180secondsand240secondsusingthe
sameinitialbacterialinoculumofapproximately106cfubacteriaperpolymer
214
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure610AntibacterialactivityofTBO‐impregnatedpolyurethanepolymeragainsta clinical strain of P aeruginosa after 90 seconds The dotted horizontal lineindicatesthedetectionlimitofthesamplingmethod080log10cfupolymer
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure611AntibacterialactivityofTBO‐impregnatedpolyurethanepolymeragainsta clinical strain of P aeruginosa after 180 seconds The dotted horizontal lineindicatesthedetectionlimitofthesamplingmethod080log10cfupolymer
215
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure612AntibacterialactivityofTBO‐impregnatedpolyurethanepolymeragainsta clinical strain of P aeruginosa after 240 seconds The dotted horizontal lineindicatesthedetectionlimitofthesamplingmethod080log10cfupolymer
Ahighlysignificantreduction intherecoveryoftheclinicalstrainPaeruginosa from
theTBO‐impregnatedpolymersafterexposuretothelaserlightwasachievedafter90
seconds (Figure610)180 seconds (Figure611)and240 seconds (Figure 612) This
reductionwas highly statistically significant for all time points tested (p lt 0001) A
highly statistically significant decrease (p lt 0001) was observed on the TBO‐
impregnatedpolymersexposedtothelaserlightcomparedwiththosenotexposedto
thelaserlightAlackofantibacterialactivitywasdemonstratedintheabsenceoflaser
lighttherewasnostatisticaldifferenceintherecoveryofPaeruginosafromthetwo
sets of polymers which were not exposed to the laser at any light exposure time
Combining these data illustrates the laser light‐induced antibacterial nature of the
polymers
216
ThedirecteffectofthelaserlightontheviabilityofPaeruginosawasdeterminedby
comparingthebacterialcountsfromthenon‐impregnatedpolymerswiththebacterial
counts from the TBO‐impregnated polymers irradiated with laser light A small
decreasecanbeobservedontheboxplotswhichwasstatisticallysignificant(plt0001
at90sand240splt005at180s)howeverthisreductionwasnotsubstantial(lt05
logcfupolymerreduction)anditismorelikelythatthisisduetothesmallvariation
in the bacterial count rather than a genuine effect of the laser To reinforce this
statement the bacterial count of P aeruginosa from the non TBO‐impregnated
polymersexposedtothelaserlight(L+S‐)wascomparedwiththatobtainedfromthe
TBO‐impregnated polymers exposed to the laser light (L+S+) large reductions in
bacterial countswere observed for all three timepoints tested (088 151 and 129
log10cfupolymerdecreasesafter90180and240secondsrespectively)whichwere
allhighlystatisticallysignificant(plt0001)
Thedifference in the susceptibilityof the twoPaeruginosa strainswas investigated
and summarised in Table 63 It was immediately evident that the laboratory type
strainofPaeruginosaPAO1wasmoresusceptibletothephotodynamiceffectofthe
TBO‐impregnatedpolymerscomparedwiththeclinical isolateAgreaterrecoveryof
bacteriawas obtained during the experimentswith the clinicalP aeruginosa isolate
compared with the type strain and this was demonstrated after 90 180 and 240
seconds
217
Table 63 Comparison of the data obtained from the two sets of P aeruginosaexperiments The stated reductions in bacteria are calculated by comparing themedianbacterialrecoveryfromtheL‐S‐samplewiththeL+S+sample
ClinicalstrainofPaeruginosa PaeruginosaPAO1
Exposuretimeseconds
Logreductioncfuperpolymer
Percentagereductioncfuperpolymer
Logreductioncfuperpolymer
Percentagereductioncfuperpolymer
90 106 913 141 961
180 170 980 294 9989
240 155 972 333 9995
633 Assessmentoftheantibacterialphoto‐activityoftheTBO‐impregnated
polymersagainstaclinicalstrainofAbaumannii
The activity of the TBO‐impregnated polyurethane polymers was subsequently
assessedagainstarecentlyisolatedclinicalstrainofAbaumanniiandtheresultsare
displayedinthefollowingthreefiguresThepolymerswereexposedtothelaserlight
for time periods of 90 seconds 180 seconds and 240 seconds using the same
concentrationofapproximately106cfubacteriaperpolymer
218
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure613AntibacterialactivityofTBO‐impregnatedpolyurethanepolymeragainstaclinicalstrainofAbaumanniiafter90secondsThedottedhorizontallineindicatesthedetectionlimitofthesamplingmethod080log10cfupolymer
$amp())+-
01+2()amp3456532
Figure614AntibacterialactivityofTBO‐impregnatedpolyurethanepolymeragainsta clinical strain of A baumannii after 180 seconds The dotted horizontal lineindicatesthedetectionlimitofthesamplingmethod080log10cfupolymer
219
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure615AntibacterialactivityofTBO‐impregnatedpolyurethanepolymeragainsta clinical strain of A baumannii after 240 seconds The dotted horizontal lineindicatesthedetectionlimitofthesamplingmethod080log10cfupolymer
AreductionintherecoveryofAbaumanniifromtheTBO‐impregnatedpolymerswas
achieved after 90 seconds (Figure 613) 180 seconds (Figure 614) and 240 seconds
(Figure615) irradiationwiththe laserlightdemonstratingthephotocatalyticactivity
of the TBO‐impregnated polymers These reductions were all highly statistically
significant (p lt 0001) There was no statistical difference in the recovery of A
baumanniifromthetwosetsofpolymerswhichwerenotexposedtothelaserlight(L‐
S‐ and L‐S+) confirming the light dependent properties of the TBO‐impregnated
materialWhen theeffect of the laser lightalonewas investigated (L‐S‐andL+S‐) a
statistically significant differencewas observed at 180 seconds (p lt 0001) and 240
seconds(plt005)andnotat90secondsbutthefiguresshowthatthisreduction is
minimal and this is likely to be a consequence of the small amount of variation in
bacterialcountsseeninthesetwogroupsFurthermorehighlystatisticallysignificant
220
reductions (plt0001)wereachievedwhen the recovery from the TBO‐impregnated
polymers exposed to the laser light were compared with the irradiated non‐
impregnated polymers further emphasising the requirement for both the laser light
andthephotosensitisertoexertahighlysignificantconsistentantibacterialeffect
634 Assessmentoftheantibacterialphoto‐activityoftheTBO‐impregnated
polymersagainstaclinicalstrainofSmaltophilia
The activity of the TBO‐impregnated polyurethane polymerswas assessed against a
newly isolated clinical strain of S maltophilia and the results are displayed in the
followingfiguresThepolymerswereexposedtothelaserlightfortimeperiodsof90
seconds 180 seconds and 240 seconds using the same concentration of
approximately106cfubacteriaperpolymer
221
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure616AntibacterialactivityofTBO‐impregnatedpolyurethanepolymeragainsta clinical strain of S maltophilia after 90 seconds The dotted horizontal lineindicatesthedetectionlimitofthesamplingmethod080log10cfupolymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Figure617AntibacterialactivityofTBO‐impregnatedpolyurethanepolymeragainsta clinical strain of S maltophilia after 180 seconds The dotted horizontal lineindicatesthedetectionlimitofthesamplingmethod080log10cfupolymer
222
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure618AntibacterialactivityofTBO‐impregnatedpolyurethanepolymeragainsta clinical strain of S maltophilia after 240 seconds The dotted horizontal lineindicatesthedetectionlimitofthesamplingmethod080log10cfupolymer
The TBO‐impregnated polymers exerted a significant antibacterial effect on S
maltophiliaafterexposuretothelaserlightfor90seconds(Figure616)180seconds
(Figure 617) and 240 seconds (Figure 618) This reduction was highly statistically
significant (p lt 0001) for all of the three exposure times Comparison of the two
groupsofTBO‐impregnatedpolymersshowedastatisticallysignificantdecreaseinthe
recoveryofSmaltophilia fromthepolymersexposedtothe laser lightcomparedto
that recovered from those polymers not exposed to the laser light There was no
statisticaldifference in the recoveryofSmaltophilia from the twosetsofpolymers
incubated in the absence of laser light (L‐S‐ and L‐S+) demonstrating the light
dependent activity of the polymers A small but statistically significant reduction in
bacterialcountswasobservedwhenthedirecteffectofthelaserlightwasinvestigated
bycomparingvaluesobtainedfromrecoveryfromthetwogroupsofnon‐impregnated
223
polymers but the effect of the laser light in combination with the impregnated
photosensitiserwasmuchlargerThisfindingmirrorsthedataobtainedintheprevious
experimentalsectionsassessingtheactivityoftheTBO‐impregnatedpolymersagainst
Abaumannii(Section633)andPaeruginosa(Sections0and632)
635 Assessmentoftheantibacterialphoto‐activityoftheTBO‐impregnated
polymersagainstaclinicalstrainofCalbicans
The activity of the TBO‐impregnated polyurethane polymerswas assessed against a
recently isolated clinical strain of C albicans and the results are displayed in the
followingfiguresThepolymerswereexposedtothelaserlightfortimeperiodsof90
seconds 180 seconds and 240 seconds using the same concentration of
approximately106cfubacteriaperpolymer
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure619AntimicrobialactivityofTBO‐impregnatedpolyurethanepolymeragainstaclinicalstrainofCalbicansafter90secondsThedottedhorizontal line indicatesthedetectionlimitofthesamplingmethod080log10cfupolymer
224
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure620AntimicrobialactivityofTBO‐impregnatedpolyurethanepolymeragainstaclinicalstrainofCalbicansafter180secondsThedottedhorizontallineindicatesthedetectionlimitofthesamplingmethod080log10cfupolymer
log 10cfupo
lymer
Exposureconditions
log 10cfupo
lymer
Exposureconditions
Figure621AntimicrobialactivityofTBO‐impregnatedpolyurethanepolymeragainstaclinicalstrainofCalbicansafter240secondsThedottedhorizontallineindicatesthedetectionlimitofthesamplingmethod080log10cfupolymer
225
A decrease in the recovery ofC albicans from the TBO‐impregnated polymerswas
noted after exposure to the laser light for 90 seconds (Figure 619) 180 seconds
(Figure 620) and 240 seconds (Figure 621) The observed reduction was highly
statisticallysignificant(plt0001)forallofthethreeexposuretimesThefindingswere
similar to thoseobtained from theexperiments involvingbacterial causesofVAP in
that a decrease in the recovery of C albicans was not detected from the TBO‐
impregnatedpolymerswhenincubatedinthedark(L‐S‐comparedwithL‐S+pgt005)
MoreoverthelaserlighthadnoeffectontherecoveryofCalbicansafter90seconds
or 180 seconds irradiation and although a statistically significant decrease was
observedafter240secondsthedifferenceisrathersmallinabsoluteterms(031log10
cfu polymer) When the effect of the laser light in combination with TBO was
comparedwith theTBOaloneahighly statistically significantdecrease in countwas
observeddemonstratingthelight‐activatednatureoftheTBO‐impregnatedpolymers
The data from this chapter are summarised below in Table 64 It is immediately
evidentthattheTBO‐impregnatedpolymers incombinationwiththe laser lightexert
anantimicrobialeffectagainstalltheorganismstestedafter90seconds180seconds
and 240 seconds The TBO‐impregnated polymers were most effective against A
baumannii where a reduction of over 4 log10 cfu polymerwas achieved after 240
seconds and was least effective against C albicans but a significant reduction
approaching2log10cfupolymerwasstillobservedafter240secondsAsmentioned
previouslytheclinicalisolateofPaeruginosawaslesssusceptibletothephoto‐active
nature of the TBO‐impregnated polymers and a smaller reduction was observed
comparedwiththelaboratorytypestrain
226
Table 64 Summary of the data obtained from the experiments investigating theactivity of the TBO‐impregnated polymers The stated reductions in bacteria arecalculatedbycomparing thebacterial recoveryfromtheL‐S‐ samplewith theL+S+sample
Logreductioncfuperpolymer
Exposuretimeseconds
Paeruginosa
PAO1
Paeruginosa
clinicalisolate
Abaumanniiclinicalisolate
Smaltophilia
clinicalisolate
Calbicansclinicalisolate
90 141 106 172 096 054
180 294 170 190 282 148
240 333 155 416 312 179
64 Discussion
641 TBO‐mediatedphotodynamicbacterialinactivation
The assessment of novel antimicrobial materials for use in endotracheal tubes is a
timely and pertinent task Therefore in this chapter polyurethane polymers were
impregnatedwiththephotosensitiserTBOandexposedtowavelengthsoflightknown
tocausephotoactivityPolyurethaneisamaterialcommonlyusedinETTs(Berraetal
2008a2008bRelloetal2010)andthepolymerswereimpregnatedwithTBOrather
thancoatedastheprocessallowsapplicationoftheantibacterialagentonboththe
inner and outer surfaces of the catheter which can increase overall antibacterial
activity (Furnoetal 2004)TheTBO‐impregnatedpolymerswereassessedagainsta
rangeofbacterialspeciescommonlyisolatedfrompatientswithVAPandtheyeastC
albicans which has also been cultured from this patient group (Weber et al 2007
Bouadma et al 2010) Previous work in our laboratory has shown that the TBO‐
impregnated polymers produced photodynamic inactivation (PDI) of a meticillin‐
227
resistant strainofSaureus (EMRSA‐16)andE coli (Pernietal 2009b)Thecurrent
studyexpandedonthesedatatoinvestigatethephotoactivityofthepolymersagainst
themostcommoncausesofVAP
These experiments have shown that the TBO‐impregnated polymers exerted a
significantantimicrobialeffectonallorganismstestedafterirradiationwithlaserlight
Thereductionsfollowedadose‐dependentresponsesothatthegreatestreductionsin
bacterial (or yeast) numbers were observed after the longest irradiation time A
baumanniiwasshowntobemostsusceptibletophotodynamic inactivationwiththe
TBO‐impregnated polymers and a reduction of over 4 log10 cfu polymer was
achieved after a 4minute irradiation time Reductions of over 3 log10 cfu polymer
werealsoachievedintherecoveryofPaeruginosaPAO1andSmaltophiliaafterthe
sameirradiationtime
Many groups have reported photodynamic inactivation of a range of planktonic
bacteriaandyeasts inthepresenceofanaqueoussolutionofTBOand laser lightE
coliwasfirstshowntobesusceptibletoa25microMsolutionofTBOinthepresenceofa
tungstenlampatalightintensityof5400luxThegenerationofsingletoxygenduring
irradiationwasconfirmedastheadditionofthesingletoxygenquencherα‐tocopherol
reduced thephotoactivityof thedye (Wakayamaetal 1980)A2 ‐ 3 log10 cfu ml
decreaseintherecoveryofAbaumanniiwasdescribedafterexposureto635nmlight
at a concentration of 2 microM and 225 J cm2 energy (Ragas et al 2010) but a pre‐
sensitisation step of 30 minutes was required to achieve this level of
photoinactivationMRSAwas shown to be susceptible to a suspension of TBO after
exposuretoaHeNe laser light for just30seconds(WilsonandYianni1995)andthe
228
susceptibilityofE faecalisB cereusandPaeruginosawasdemonstratedagainsta
variety of phenothiazinium dyes including TBO after 60 minutes light exposure
(Wainwrightetal1997)
Gram‐negative bacteria have been shown to be less susceptible than Gram‐positive
bacteria to the photoactivity of the TBO‐impregnated polymers (Perni et al 2009b)
andtophotodynamictherapyusingotherphotosensitiserssuchasmethyleneblueand
rose bengal (Phoenix et al 2003 Decraene et al 2006 Perni et al 2009a) The
cytoplasmicmembrane is the primary target of the singlet oxygen generated during
irradiationwith the laser light (Wakayama et al 1980 Jori et al 2006) which has
been demonstrated in E coli and S cerevisiae (Ito 1977 Ito and Kobayashi 1977)
Gram‐negative bacteria have a reduced rate of uptake of singlet oxygen due to the
presenceoftheoutermembrane(Jorietal2006)whichpreventsdirect interaction
of the singlet oxygen with the underlying cytoplasmic membrane It also acts as a
permeabilitybarrierpreventingthediffusionofsmallmoleculesintothecytoplasmof
thecellConverselyGram‐positivebacteriaaresurroundedbyarelativelyporouslayer
of peptidoglycan and aremore likely to be susceptible to the action of the reactive
oxygen species generated on the surface of the polymers DNA damage occurs in
Gram‐positiveandGram‐negativebacteriaandinyeastcellsoncethepermeabilityof
the externalmembrane has been compromised and the reactive oxygen species are
abletopenetratetheinteriorofthecells(Dunipaceetal1992Chietal2010)The
susceptibility of Gram‐negative bacteria to the effect of the TBO‐impregnated
polymerssuggeststhatthemechanismofactivityistheTypeIIpathway(Figure111)
The photosensitiser was immobilised in the polymer and was not able to interact
229
directly with the bacterial cell wall and so the phototoxic effect occurred via the
generationofsingletoxygenwhichoxidisedmoleculesintheoutermembraneItwas
hypothesised that reactive oxygen species generated by the Type I pathway
wereunabletocauselethaldamagetotheoutermembraneandrequiredpenetration
ofthemembraneinordertoexertlethalPDI(Jorietal2006)
It was hypothesised that the reductions observed for the Gram‐negative organisms
usedintheseexperimentswouldbelessthanthatobservedforSaureus(Pernietal
2009b)Although these results support the hypothesis the data cannot be directly
comparedwiththepublishedworkasalargerstartinginoculumwasusedinthisseries
ofexperimentsandcellsaremoresusceptibletoPDIwhenalowerinoculumisused
(Soetal2010)TheinitialbacterialconcentrationusedinthePernistudyequatedto
approximately4x104cfupolymerandinpreliminaryexperimentsa354log10cfu
polymerreductioninPaeruginosaPAO1wasdetectedwhichwasbelowthedetection
limitof theexperiment(datanotshown)Thereforeahigher initialbacterial loadof
106 cfu polymerwas selected so that colonieswere always detectable on the test
(L+S+) plates and the values obtained were within the detectable limits of the
experimental design Alternatively the exposure time to the laser could have been
decreased to ensure the recovered bacteria werewithin the detection limits of the
assay For reference the Perni et al (2009a) study showed a gt4 log10 cfu ml
reduction in EMRSA16 after a 1 minute irradiation time and a gt4 log10 cfu ml
reductioninEcoliATCC25922aftera2minuteirradiation
These data also show that C albicans was less susceptible to TBO‐mediated
photodynamic inactivation than the Gram‐negative bacteria S maltophilia A
230
baumanniiandPaeruginosaPAO1IthaspreviouslybeenshownthatCalbicanswas
susceptible toPDIusinga solutionof TBOand irradiationwith red light (Wilsonand
Mia 1993) and an increased tolerance to these conditionswas displayed compared
with the Gram‐negative oral bacteria Fusobacterium nucleatum Actinobacillus
actinomycetemeomitans and Porphyromonas gingivalis (Wilson et al 1993 Wilson
andMia1994)Yeastcellsaremuchlargerinsizethanbacterialcellsthediameterof
aCalbicanscellisapproximately3to4microm(MerzandRoberts1999)comparedwith
Abaumanniiwhichisapproximately1to15by15to25microminsize(Schreckenberger
and von Graevenitz 1999) and S aureus which is approximately 05 to 15 microm in
diameter (Kloos and Bannerman 1999 Sandel and McKillip 2004) Therefore the
yeastcellislikelytorequirealargerdoseofreactiveoxygenspeciestoexertasimilar
photodynamiceffect (Jorietal2006)Thestructureoftheyeastcellwallcouldalso
contributetowardsincreasedtolerancetoPDT(BowmanandFree2006)
642 Limitationsoftheexperimentalwork
The clinical strain of P aeruginosa was shown to be the least susceptible to the
photoactivityoftheTBO‐impregnatedpolymersaftera4minuteirradiationtimeand
the reduction in bacteria observed was substantially less than that seen in for the
laboratory strain ofP aeruginosa PAO1P aeruginosaPAO1was originally isolated
fromawoundinMelbourneAustraliain1955(Holloway1955)Sincethenithasbeen
serially passaged for many years and shared with laboratories around the world
where further passages have taken place (Fux et al 2005) The PAO1 strain was
selectedbecauseitrsquosubiquitoususeallowsthedatageneratedintheseexperimentsto
becomparedwithresultsgeneratedbygroupsaroundtheworldonthesensitivityof
231
P aeruginosa to the TBO‐laser combination However itrsquos limitations should be
acknowledged and it is probable that the PAO1 strain in use today has lost
characteristicsfoundintheoriginalstrainasaresultofserialpassage(Fuxetal2005)
Theconditionsthatbacteriaareexposedtoduringlaboratoryculturearesubstantially
differentfromthoseexperiencedwithinthehostileenvironmentofthehumanbody
An abundance of nutrients are present in laboratory media to encourage bacterial
growth and incubation conditions are optimal for rapid replication Therefore the
genesthatarerequiredforcolonisationandsurvivalwithinthehumanhostaresurplus
to requirement For example in E coli genes required for flagella production are
inactivatedafterserialpassagersquos(Edwardsetal2002)whichbenefitsthelaboratory‐
adaptedstrainasflagellaproductionisanenergy‐richprocessthatrequireshighlevels
ofaminoacidproduction If thesegenesare inactivated the replication timewillbe
shorterwhichwillgivethelaboratory‐adaptedstrainafitnessadvantageoverthewild
typestrain
Theabilityofthe laboratoryadaptedcells toadhereandformbiofilmscouldalsobe
reduced(Fuxetal2005)MucoidstrainsofPaeruginosaarecommonlyisolatedfrom
patientswithcysticfibrosisandthisphenotypeisoftenlostduringlaboratoryculture
due to a series of point mutations and a non‐mucoid rough colony morphology
predominates(Govan1975DrenkardandAusubel2002)Mucoidstrainsproducea
greater quantity of alginate (Simpson et al 1989) a known scavenger of reactive
oxygen species such as singlet oxygen which is produced in abundance during the
photodynamicreactionontheTBO‐impregnatedpolymers(Wakayamaetal1980)A
possible reason for the decreased susceptibility of the clinical isolate to the
232
photoactivityofTBO‐impregnatedpolymerscouldthereforeberelatedtoanincreased
production of alginate which is a defencemechanism against the respiratory burst
released by macrophages within the human hostWong et al (2006) showed that
clinical isolates exposed to the visible‐light driven photocatalytic effect of N‐doped
TiO2 thin films displayed increased tolerance to killing compared with a laboratory
strainofEcoliOP50and itwassuggestedthatthemechanismbehindthiswasalso
linkedtoresistancetoreactiveoxygenspecies
Thebacterial isolatesused in this seriesofexperimentswerecultured inbrainheart
infusion (BHI) liquid media and subsequently re‐suspended in PBS which is a low
protein saline solution It has been shown that the PDI effect is reduced by the
presence of proteins in the medium and so it is possible that the inhibitory effect
observed in these experiments would be reduced under in vivo conditions as the
trachealsecretionscontainhighlevelsofproteins(WilsonandPratten1995Nitzanet
al 1998) These proteins could absorb light which would reduce the number of
photonsavailablewhichwouldinturndecreasetheconcentrationofreactiveoxygen
species generated (Komerik and Wilson 2002) The proteins may also be used as
alternativetargetsbythesingletoxygenspeciesandshieldbacteriafromthecytotoxic
effectsgenerated
643 Novelmaterialsforpotentialuseasantimicrobialendotrachealtubes
Numerous invitrostudieshavebeenconductedonmaterialswhichcouldbeusedas
novel antibacterial ETTs Methylene blue was incorporated into silicone and the
photodynamic effect with and without the addition of gold nanoparticles was
233
investigated (Perni et al 2009a) A significant level of photoactivity was observed
againstEcoliandMRSAafter5minutes irradiationwithared laser lightwhichwas
enhanced with the addition of gold nanoparticles Berra et al (2008a) coated
polyurethaneETTswithsilversulfadiazineandchallengedthetubeswithPaeruginosa
PAO1 The silver coated ETT was examined by both scanning electron microscopy
(SEM)andconfocal laser scanningmicroscopy (CLSM)and sectionsof the tubewere
culturedafteraperiodof72hoursadhesionofPaeruginosaPAO1tothesubstrate
hadbeenpreventedandthegrowthratewasalsoreducedThesilvercoatedETTwas
subsequentlyused inaventilated sheepmodelNobacteriawerecultured from the
coatedETTsafter24hoursandathinnerlayerofmucuswaspresentonthelumenof
the tube compared with the uncoated control where bacterial colonisation was
present(Berraetal2008a)
Rello et al (2010) coated a proprietary hydrophilic polymer with silver ions and
investigated the adherence of 18 organisms after an exposure time of 4 hours A
reducedlevelofbacterialattachmentwasobservedforrespiratorystrainsofMRSAP
aeruginosaandEaerogenesbuttheattachmentofanumberofotherorganismssuch
asCalbicansandKpneumoniaewasnotpreventedTheantibacterialactivityofthe
silverion‐coatedETTwasthenassessedinarabbitmodelwhichwaschallengedwitha
respiratoryisolateofPaeruginosaAfter16hoursareducedlevelofETTcolonisation
wasobservedonthesilverion‐coatedtubesandPaeruginosawasnotisolatedfrom
thelungsoftherabbitsIncomparisonPaeruginosawasculturedfromallnon‐coated
ETTsand from the lungsofall rabbits intubatedwith the control tubes (Relloetal
2010)
234
A large‐scale randomised trial published in 2008 aimed to ascertain whether silver
coatedETTscouldreducetheincidenceofVAPinhumans(Kollefetal2008)Nearly
10000patientswerescreenedfortheireligibilityintothestudyandsuitablepatients
wereassignedasilver‐coatedETToranon‐coatedtubeAreduction inthe incidence
of VAP was observed in patients with silver‐coated tubes These findings were
extremelypromisingastheyshowedthatbysimplyusingadifferentventilatortube
theincidenceofVAPcouldbereducedanditrequirednoadditionalinvolvementfrom
themedical team treating the patientHowever some authors have questioned the
meritofreducingbacterialloadontheETT(Balk2002Spronketal2006)asthereis
no direct evidence to demonstrate that antibacterial ETTs can reduce length of
hospital stay ormortality rates and the silver coated ETTs cost over $100 per tube
compared with less than $1 for a traditional uncoated tube (Deem and Treggiari
2010)
65 Conclusions
The antibacterial photodynamic inactivation of P aeruginosa S maltophilia and A
baumanniiwasassessedonTBO‐impregnatedpolymersafter irradiationwithaHeNe
laser light A significant reduction in the recovery of all bacterial strains testedwas
observed after 90 180 and 240 seconds A recently isolated clinical strain of P
aeruginosa showed decreased susceptibility to the photo‐activity of the TBO‐
impregnated polymers compared with a laboratory type strain Significant
photodynamicinactivationofCalbicanswasalsoobservedafterexposuretothesame
lightsourcedemonstratingthatthelight‐inducedeffectisnotrestrictedtobacteria
235
7 Assessment of the disruptive and anti‐adhesive propertiesofnovellight‐activatedmaterials
71 Introduction
Theanti‐adhesivepropertiesoftwoofthenovellight‐activatedantibacterialmaterials
generatedinthisthesiswasexploredinthischapterusingarangeoftechniquesThe
silver‐doped titanium dioxide thin films were examined to determine whether in
addition to the photo‐activated bactericidal effects already demonstrated initial
bacterialadhesiontothesurfacecouldbepreventedandwhethertheformationofan
immaturebacterialbiofilmcouldbedisruptedTheinitialattachmentofbacteriatothe
TBO‐impregnated polyurethane polymers was assessed after irradiation with the
HeNe laser which prompted the examination of the photo‐bleaching effect of the
laserontheantibacterialactivityoftheTBO‐impregnatedpolymers
Demonstratingareductionintherecoveryofviablebacteriainoculatedontothenovel
surfacesafterlightexposureisausefulinitialmethodofestablishingtheantibacterial
activityofthenovelmaterialsHoweveritwouldalsobeadvantageoustopreventthe
initialattachmentofbacteriatothesurfaceDuringthe initialadhesioneventsthere
willbea lowerbacterial loadsophotoinactivationmayoccuratafasterrateAlsoin
the clinical environment the risk of onward transmission of bacteria from a hand‐
touch surface via the hands of patients or healthcare workers would be further
reduced due to the smaller inoculum present An additional measure which would
provebeneficialintheclinicalenvironmentwouldbethedetachmentandinactivation
ofbacteriaalreadyboundtothesurfacebeforelightexposure
236
72 Materialsandmethods
721 Silver‐dopedtitaniumdioxidethinfilms
7211 AssessmentofinitialattachmentofEMRSA‐16
BacterialattachmenttotheAg‐TiO2thinfilmswasmeasuredusingtwosinglechannel
transmissionFC81‐PCflowcells(BioSurfaceTechnologiesCorporationMontanaUSA)
Two flowcell chambers (50x13x235mm)were joined togetherwith tapebefore
autoclavingandrinsingwithwaterTheflaskwaspreparedbyconstituting500mLPBS
ina1000mLconicalflaskwithamagneticstirreraddedarubberstopperwasloosely
placedonandcoveredwith foil The two female connectorswerewrappedwith foil
andsealedwithautoclavetapeClampswereattachedtotheendsofbothtubesby
the male connectors and on either side of the air filters and the entire unit was
autoclavedfor15minutesat121degC
237
Figure 71 The flow cell chamber used to assess bacterial attachment TheAg‐TiO2thin film was placed within the chamber and adhesion was assessed by lightmicroscopyasabacterialsuspensionflowedacrossthematerial
Theflowcellchamberwasassembledandasealantwasappliedbetweeneachlayerto
preventthe leakageof liquidAcoverslipwasplacedontheclearplastic lidandthe
entry and exit points in the flow cell chamber were cleaned with an isopropanol‐
containingwipetoensuretherewasnoobstructioncausedbysealantTheuncoated
glassslidedenotedS‐wasplacedintheridgeontheclearplasticlidandscrewswere
addedtothetopandnottightenedTheAg‐TiO2thinfilmscouldnotbeautoclavedso
thesewere not added at this point The screwswere loosely positioned on top and
coveredwithtapeFoilwasaddedtothetopofthebubbletrapandtheendsofthe
twomale connectors Clamps were affixed to the ends of both tubes by the male
connectorsTheflowcell chamberwasthen laid flat inanautoclavebagandsealed
thenplacedintoasecondautoclavebagsealedandlabelledThebagwassterilisedby
autoclavingat121degCfor12minutes
238
Afterautoclavingtherubberstopperonthetopoftheconicalflaskwassecuredand
theclamps fromeither side of theair filterwere removedThe flaskofPBSand the
flow cell chamberswere allowed to cool before the Ag‐TiO2 slide denoted S+ was
placed into the flow cell chamber and all screws on the flow cell chamber were
tightened to prevent any leakages The clamps from the end of each tube were
removedand the flowcell chamberwas joined to the flaskbyplacing themaleand
female connectors together Finally a 045 nm filter (Nalgenereg Labware Roskilde
Denmark) was added to the top of the bubble trap A culture of EMRSA‐16 was
preparedinBHIasdescribedinSection22
After24hoursgrowth5mLoftheovernightculturewasdispenseddirectly intothe
flaskcontaining500mLPBSprovidingadilutionofapproximately1in100Theflow
cell chamber and bubble trap was placed into a large white tray and the narrow
section of tubing was passed through the peristaltic pump (Watson‐Marlow Pumps
GroupFalmouthUK)toachievealowflowrateThewholesystem(peristalticpump
flask and tubing)was transferred into the 22degC incubator containing thewhite light
sourcealongwithamagneticstirrerTheperistalticpumpwasthenswitchedonand
thespeedsetto30equatingtoashearrateof40s‐1Thevalveonthebubbletrap
waskeptopenuntiltheliquidhadreachedthehalfwaymarkatwhichpointthevalve
wasclosedandtheliquidcouldpassthroughthesystembacktotheconicalflask
After06and18hourstheflowcellsystemwasmovedtothelightmicroscopesothat
theattachmentofbacteriaonthesurfaceofthethinfilmscouldbevisualisedThex40
objectivelens(OlympusULWDCDPlan40)wasusedandatleasttenrandomfieldsof
viewwereexaminedpersampleandrepresentativeimageswerecaptured
239
7212 DisruptionofanimmaturebiofilmofEMRSA‐16
BacterialstrainsweremaintainedasdescribedinSection21andbacterialsuspensions
ofEMRSA‐16werepreparedinPBSasdetailedinSection23Alternativelyanaliquot
of the re‐suspendedpelletofbacteriawasadded toa10mL ofBHIand the optical
densitywasmeasuredonthespectrophotometerInbothcasestheresultingbacterial
suspensioncontainedapproximately107 cfu mL Silver‐doped titaniumdioxide thin
filmsoruncoatedcontrolswereplacedinthemoisturechambersdescribed inFigure
22before50microLofthebacterialsuspensionwasaddedandthemoisturechambers
wereincubatedinthedarkfor24hourstoallowanimmaturebiofilmtodevelop
Themoisture chamberswere subsequently transferred to the cooled incubator and
incubated at 22degC for 24 hours under thewhite light source The Live Dead stain
(Molecular Probes)was prepared by adding 20 microL of both SYTO9trade and propidium
iodidetoafoil‐covereduniversalcontaining40mLPBSandwasincubatedinthedark
for 30minutes before use The Live Dead stainwas poured into a petri dish the
sampleswere immersed inthepetridishand incubated inthedark for5minutesto
allow the stain to penetrate the bacterial cells before viewing Two slides were
examinedforeachexposureconditionasdetailedinTable71andatleasttenfieldsof
view were examined per sample and representative images were captured The
sampleswereexaminedontheconfocal laserscanningmicroscope(CLSM)usingthe
x40 lenswithabluefilterand lateranalysedusingthe ImageJcomputerprogramme
which can be accessed for free from httprsbwebnihgovij The experimentwas
repeatedtodemonstratereproducibility
240
Table 71 Description of the samples examined under the confocal scanning lasermicroscope
Samplereference Sampletype Exposureconditions Inoculum
K2K3 Ag‐TiO2 light EMRSAinPBS
K4K5 Ag‐TiO2 dark EMRSAinPBS
K6K7 Ag‐TiO2 light EMRSAinBHI
K8K9 Ag‐TiO2 dark EMRSAinBHI
K10K13 Ag‐TiO2 light Nobacteria
K14K17 Ag‐TiO2 dark Nobacteria
B1B2 Uncoatedslide light Nobacteria
B3B4 Uncoatedslide dark Nobacteria
722 TBO‐impregnatedpolymers
7221 PreventionofinitialPaeruginosaPAO1attachment
BacterialstrainsweremaintainedasdescribedinSection21andbacterialsuspensions
ofPaeruginosaPAO1weregrownandpreparedinPBSasdetailedinSection22and
Section23resultinginabacterialsuspensioncontainingapproximately107cfumL
Thedescribedmethodwasadapted fromapaperbyChrzanowskietal (2010)The
testsampleswerepreparedandplaced ina24wellmicrotitreplateas illustrated in
Figure 72 Empty wells were filled with foil to prevent laser light penetrating into
adjacent wells One millilitre of bacterial suspension was added to the test well
ensuring the polymer did not float to the surface and the remaining wells were
covered with a sheet of black paper The well was irradiated with the HeNe laser
source described in Section 243 for the designated exposure time and theemitted
light was passed through a beam diffuser to ensure that the entire polymer was
241
exposed to the laser light The process was repeated for each appropriate sample
beforestaticincubationat37degCforthedesignatedtimeperiodbeforere‐exposureto
the laser source After three hours each sample was placed into a separate bijou
containing3mLPBSandincubatedat22degCfor5minorpreparedforscanningelectron
microscopyThepolymerwassubsequentlytransferredtoabijoucontaining1mLPBS
and 5 glass beads each with a diameter of 3 mm and vortexed for 1 min Twenty
microlitresofthebacterialsuspensionwasthenremovedseriallydilutedandspread
ontoMacConkey agar plates before incubation at 37degC for 48 hours The resultant
colonieswerecountedandcomparedwiththecontrolstocalculatethelevelofbiofilm
disruption
Figure 72 The layout of themicrotitre plate during the biofilm disruption assayswhere++correspondstoaTBO‐impregnatedpolyurethanepolymerexposedtothelaserlight‐+correspondstoaTBO‐impregnatedpolyurethanepolymernotexposedto the laser light +‐ corresponds to a polyurethane polymer exposed to the laserlightand ‐‐ corresponds toapolyurethanepolymernotexposed to the laser lightShadedcirclesrepresentwellsfilledwithfoil
7222 Scanningelectronmicroscopy
Afterthreehoursincubationat37degCthesampleswerepreparedforSEManalysisby
DrNickyMordanThesamplesunderwentaseriesof10minutesdehydrationstages
242
in increasing concentrations of alcohol (20 50 70 90 and 3x 100) before
immersioninhexamethyldisilazane(HMDS)(TAABLaboratoriesLtdReadingUK)for5
min followedbydryingon filterpaper for2 ‐3 hours toensure that theHMDShad
completely evaporated The samples were then fixed onto alumininum SEM stubs
(Agar Scientific) using carbon conducting cement (Neubauer Chemikalien Munster
Germany) as an adhesive before sputter‐coating with goldpalladium in a Polaron
E5000 Sputter Coater (Quorum Technologies Ltd Newhaven UK) A Cambridge
Stereoscan90B (LEO ElectronMicroscopyLtdCambridgeUK)wasused toview the
specimensoperatingat15kVandatleasttenfieldsofviewwereexaminedThei‐scan
2000software(ISSGroupManchesterUK)wasusedtocapturerepresentativedigital
imagesforeachsample
7223 Photo‐bleachingeffects
TheTBO‐impregnatedpolymerswereirradiatedwiththeHeNelasersourcedescribed
inSection243foreither90180or240secondsbeforeincubation inasterilepetri
dishfor24hoursat22degCThepolymerswerethenprocessedasdescribed inSection
2123polymerswhichhad been initially irradiated for 90 secondswereexposed to
another90 second laserdosepolymers irradiated for 180 secondswere re‐exposed
for180secondsandpolymersirradiatedfor240secondsweretreatedwithafurther
240 second light doseNaiumlve TBO‐impregnated polymerswere used as controls ie
TBO‐impregnated polymers that had been stored in the dark during the initial
irradiationstepThreeTBO‐impregnatedpolymersweretestedforeachexposuretime
andtheexperimentwasrepeatedthreetimestodemonstratereproducibility
243
73 Results
731 Silver‐dopedtitaniumdioxidethinfilms
7311 Assessmentofbacterialattachment
The attachment of EMRSA‐16 to the surface of the Ag‐TiO2 thin filmswas assessed
using the flowcellmodelBacteriawere observed in thecirculatingbrothafter zero
hours in low numbers in Figure 73(a) and Figure 73(b) the cocciwere in constant
motionmoving in the direction of the flow suggesting that attachment had not yet
occurredAsimilarnumberofbacteriawerefoundontheAg‐TiO2thin filmsandthe
uncoated control slides After 6 hours the number of bacteria observed on both
coating typeshad increased substantiallyanda near complete coverageof the slide
was observed (Figure 74a and Figure 74b) Again there was no difference in the
attachment of bacteria to the irradiatedAg‐TiO2 thin film and the uncoated control
exposedtothesame lightconditionsAfter18hoursexposuretothewhite lightno
reductioninthenumberofbacteriawasobservedontheAg‐TiO2thinfilmsexposedto
thewhitelightandtherewasnovisualdifferenceinthenumberofbacteriaobserved
ontheAg‐TiO2thinfilmcomparedwiththeuncoatedcontrol (Figure75aandFigure
75b)
TheshrinkcrackswhichcanbeclearlyseenontheAg‐TiO2thinfilmsareafeatureof
the coating and are a result of the annealing process There was no greater than
bacterial attachment observed in these areas than on the non‐cracked areas of the
thinfilm
244
Figure73AttachmentofEMRSA‐16toeitheran(a)uncoatedslideor(b)Ag‐TiO2thinfilmafter0hexposuretothewhitelightsource
Figure74AttachmentofEMRSA‐16toeitheran(a)uncoatedslideor(b)Ag‐TiO2thinfilmafter6hexposuretothewhitelightsource
Figure75AttachmentofEMRSA‐16toeitheran(a)uncoatedslideor(b)Ag‐TiO2thinfilmafter18hexposuretothewhitelightsource
245
7312 DisruptionofanimmaturebiofilmofEMRSA‐16
Astherewasnodifference intheattachmentofEMRSA‐16totheAg‐TiO2thinfilms
theviabilityofEMRSA‐16wasexaminedafterirradiationwithwhitelightItispossible
thatthephoto‐activatedthinfilmswerenotpreventingbacterialattachmentbutwere
inactivatingthebacteriathatdidadhereAnimmaturebiofilmofEMRSA‐16inPBSwas
grownonthesurfaceoftheAg‐TiO2thinfilmsandexposedtowhitelightfor24hours
a reduction in the viability of the attached bacterial cellswas observed Therewere
substantiallymore non‐viable cells on the Ag‐TiO2 thin films exposed towhite light
(Figure76)comparedthatobservedonthesurfaceoftheAg‐TiO2thinfilmsincubated
inthedark(Figure77)Thisdemonstratesthatwhite light irradiationoftheAg‐TiO2
thin films caused an increase in the permeability of the cell membrane to the
propidiumiodidestainandaccompanyingdamagetotheintegrityofthebacterialcell
membrane No antibacterial activity was observed in the absence of light which
suggests that the damage to the bacterial cell membranes was not caused by the
leakageofsilverionsfromthesurfaceofthethinfilm
246
Figure76ConfocalmicrographofEMRSA‐16inPBSontheAg‐TiO2thinfilmafter24hoursgrowthat37degC in thedarkand24hoursexposure towhite lightat22degC (xyprojection 300 x 300 μm) Viable bacterial cells are stained green and non‐viablecellsarestainedredThedepthofthebacterialgrowthisdisplayedunderneaththemainimage(xzprojection)
247
Figure77ConfocalmicrographofEMRSA‐16inPBSontheAg‐TiO2thinfilmafter24hours growth at 37degC in the dark and 24 hours incubation at 22degC in the dark (xyprojection 300 x 300 μm) Viable bacterial cells are stained green and non‐viablecellsarestainedredThedepthofthebacterialgrowthisdisplayedunderneaththemainimage(xzprojection)