Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Influence of ozone on shelf-life and quality of silvergrey rockfish (Sebastes brevispinis) Prahst, Armin-Wilhelm 1994

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1994-0599.pdf [ 2.63MB ]
JSON: 831-1.0087394.json
JSON-LD: 831-1.0087394-ld.json
RDF/XML (Pretty): 831-1.0087394-rdf.xml
RDF/JSON: 831-1.0087394-rdf.json
Turtle: 831-1.0087394-turtle.txt
N-Triples: 831-1.0087394-rdf-ntriples.txt
Original Record: 831-1.0087394-source.json
Full Text

Full Text

INFLUENCE OF OZONE ON SHELF-LIFE AND QUALITYOF SILVERGREY ROCKFISH (Sebastes brevispinis)BYARMIN-WILHELM PRAHSTB.Sc., University of Guelph, 1992A THESIS SUBMITTED IN PARTIAL FULFILMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIESDEPARTMENT OF FOOD SCIENCEWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIASeptember 1994© Armin-Wilhelm Prahst, 1994In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives, It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)Department ofThe University of British ColumbiaVancouver, CanadaDate 0 & I IDE-6 (2188)iiABSTRACTOzone treatments of the chilled water surrounding fishduring storage affects microbial and sensory properties of thestored product. In the first part of the study a betterunderstanding of the effect of ozone on the microbial and sensoryproperties of freshly caught Silvergrey Rockfish (SebasLesbrevispinis) was obtained. In the second part of the study, amethod to quantify the inactivation of various typical fish-spoilage microorganisms by ozone was developed.A survey examining the unloading process of fishing boatsindicated that the unloading process did not contribute to anincrease in microorganisms on the skin of tested fish samples.Sensory properties of the fish were not significantly improved byozone treatment during an experiment simulating storageconditions on British Columbia fishing boats. The onlysignificant difference (p<O.O5) due to ozonation was thereduction of the trimethylamine (TMA) level in the fish meatwhich, however, did not manifest in the sensory properties of thefish meat. Ozonation also did not cause any apparent colourdifferences in the fish meat. Although the number of skinmicroorganisms was initially reduced, after day 3 the microbialpopulation on the ozone treated fish increased. Gill samples fromozone treated fish showed consistently higher, but notsignificantly different (except for one day), levels ofiiimicroorganisms than the control samples.A model disinfection study was adequate to estimate thesensitivity of various typical fish spoilage microorganisms toozone under strictly defined conditions. Of the bacterial strainstested Pseudomonas tragi was the most sensitive followed byAlteromonas putrefaci ens and Alteromonas punctata subsp.punctata. Staphylococcus aureus was the least sensitive toinactivation by ozone.ivTABLE OF CONTENTSABSTRACTTABLE OF CONTENTS ivLIST OF TABLES viiLIST OF FIGURES viiiNOMENCLATURE xiACKNOWLEDGEMENTS xiiI. INTRODUCTION 1II. LITERATURE REVIEW 4A. FUNDAMENTALS OF OZONE CHEMISTRY 41. Structure and reaction mechanisms of ozone 42. Specific compounds that are effected by ozone 52.1. Fatty acids 72.2. Amino acids (thiols) and proteins 93. Effects of ozone on microorganisms 11B. PRACTICAL APPLICATIONS OF OZONE IN THE FOOD INDUSTRY 121. Drinking water disinfection 122. Disinfection of the chilled water used forfish storage on fishing boats 153. Use of ozone during the storage of fish and seafood 164. Use of the C*t model to estimate bacterialsensitivities to ozone 185. Other food system applications 20III. USE OF OZONE TO EXTEND SHELF-LIFE AND QUALITY OFFRESH FISH DURING STORAGE ON B.C. FISHING VESSELS ... 24A. INTRODUCTION 24VB. MATERIALS AND METHODS 251. Experimental conditions and setup 252. The microbial development on the individual wholefish (skin and gills) 323. The microbiology of the chilled water at variousstages of storage 344. Colour changes of fish skin, gills and meatdue to ozonation 345. BOD, COD, nitrite, nitrate, and ammonia levelsof the chilled water 356. Ozone concentration, pH development, temperatureof the chilled water 367. Sensory panels for filleted fish at various stagesof storage 388. TMA values of fish fillets 389. Statistical evaluation of data 39C. Results and Discussion 401. General observations 402 The microbial development on the individualwhole fish (skin and gills) 413. The microbiology of the chilled water atvarious stages of storage 464. Colour changes of fish skin, gills and meatsamples due to ozonation 495. BOD, COD, nitrite, nitrate and ammonialevels of the chilled water 526. Ozone concentration, pH development,temperature of the chilled water 607. Sensory panels for filleted fish at variousstages of storage 618. TMA values of fish fillets 65D. CONCLUSION 69viIV. SURVEY OF THE MICROBIAL CONDITIONS OF FISH SURFACESAND “PROCESSING LIQUID” DURING THE UNLOADING OF FISHAT A FISH PROCESSING PLANT IN VANCOUVER 70A. INTRODUCTION 70B. MATERIALS AND METHODS 741. Microbiological tests of the skin of individualwhole fish 742. Microbiological tests of fish storage liquids 75C. RESULTS AND DISCUSSION 76D. CONCLUSION 78V. MODEL EXPERIMENT TO DETERMINE THE SENSITIVITY OFTYPICAL FISH SPOILAGE BACTERIA TO OZONE TREATMENTS 78A. INTRODUCTION 78B. MATERIALS AND METHODS 801. Experimental procedure 802. Bacterial isolates 813. Generation and measurement of ozone 834. Preparation of glassware 845. Preparation of demand free phosphate buffer 85C. RESULTS AND DISCUSSION 86D. CONCLUSION 95VI. FINAL CONCLUSION AND RECONNENDATIONS 97VII. REFERENCES 99VIII. APPENDIX A: PHOTOGRAPHS OF EXPERIMENTAL SETUP ANDRESULTS OF AN EXPERIMENT TO SIMULATE THE STORAGECONDITIONS ON B.C. FISHING BOATS IN A PILOT SETUP ... 109APPENDIX B: LISTING OF PRELIMINARY EXPERIMENTSCONDUCTED DURING THE DISINFECTION EXPERIMENT TODETERMINE THE DESIRED OZONE CONCENTRATION AT WHICHTHE BACTERIA SHOULD BE ADDED TO THE SYSTEM 113viiLIST OF TABLES2.1: Relative reactivity of several types of organicgroups, that are present in lipids, proteins andsugars, with ozone at room temperature 82.2: Rate Constants (W1s’) at 20°C for ozonation ofsaturated fatty acids and organic acids 92.3: Amino acids and some of their possible ozonationproducts 102.4: Treatment goals for ozonation of drinking water 132.5: Summary of literature data on the extension ofstorage life of various foods 21-223.1: Colour of gills, skin and flesh of SilvergreyRockfish during storage of whole fish in ozonatedand untreated water over nine days 503.2: Mean values for sensory panel scores for fish storedin ozonated and unozonated chilled water 633.3: Analysis of variance of sensory panel scores for 9day old Silvergrey Rockfish stored in ozonated andunozonated chilled water 643.4: Analysis of variance of sensory panel scores for 9day old Silvergrey Rockfish stored with and withoutozonation of the chilled water, z-score transformedsensory data 643.5: ANOVA statistics of TMA data from fish samples of day7 and 9 of storage in ozonated and unozonated chilledwater 665.1: Results of disinfection experiment to determinesensitivity of single strain microorganisms to ozone,using four different bacterial strains at temperaturesof 0°C and 22°C 90viiiLIST OF FIGtJRES2.1: Resonance structures of the ozone molecule 42.2: Cyclo addition (Criegee mechanism) of ozone onunsaturated bonds 63.1: Setup of fish tank experiment to simulate storageconditions that prevails on B.C. fishing boats 273.2: Principle of the “Sander Ozone Generator Model 301.”Pure oxygen from a pressure bottle flows close by ahigh voltage electrode (7kv) where some of theoxygen is converted to ozone 293.3: Schematic view of the bubble column for the ozonationof chilled water used in the fish tank experiment tosimulate fish storage conditions on B.C. fishingboats 303.4: Generalised view of a Silvergrey Rockfish tovisualize the sampling area for microbiological,sensory and colorimetric tests performed during thecourse of storage of the fish in chilled water, withand without ozonation, over a period of nine days . . . 333.5: Microbial development on the skin of fish (postmortem) during the course of storage over a periodin chilled water, with and without ozonation, of ninedays 423.6: Microbial development on gills of fish (post-mortem)taken during the course of storage in chilled waterwith and without ozonation over a period of ninedays 453.7: Microbial development of the chilled water usedfor storage of fish (post—mortem) over a periodof nine days with and without ozonation 473.8: Microbial quality of the ice added for coolingof the chilled water during the course of storageof fish over a period of nine days with andwithout ozonation 483.9: BOD development of the chilled water used forstorage of whole fish over a period of 9 days withand without ozonation 53ix3.10: COD development of the chilled water used forstorage of whole fish over a period of 9 days withand without ozonation 553.11: Ammonia, nitrite and nitrate development ofchilled water used for storage of whole fishduring a period of 9 days with and withoutozonation 573.12: Partial schematic of ozonation of primary amines . . . . 583.13: TMA development in Silvergrey Rockfish over a9 day storage period in ozonated and unozonatedchilled water 674.1: Schematic representation of one of the holding tanksand recirculation circuit of chilled water on thefishing boat “Arctic Ocean” 724.2: Flow chart of the unloading process of fish at afish processing plant in Vancouver, B.C 734.3: Microbiology of fish and chilled water samplestaken during the unloading process at afish processing plant in Vancouver, B.C 775.1: Treatment scheme for the recovery of frozen storedbacteria for use as the stock culture in disinfectionexperiment, yielding 100 mL of stock culture 785.2: Standard curves of natural decay of ozone in demandfree 0.01 M phosphate buffer system at 0, 10 and22°C 875.3: Ozone concentration changes and reduction ofbacterial populations after introduction of Alteromonasputrefaciens into ozone residual containing demandfree phosphate buffer (0.01 M) 915.4: Ozone concentration changes and reduction ofbacterial populations after introduction ofStaphylococcus aureus into ozone residual containingdemand free phosphate buffer (0.01 M) 925.5: Ozone concentration changes and reduction ofbacterial populations after introduction ofPseudomonas fragi into ozone residual containingdemand free phosphate buffer (0.01 M) 93x5.6: Ozone concentration changes and reduction ofbacterial populations after introduction ofAlteromonas punctata subsp. punctata) into ozoneresidual containing demand free phosphate buffer(0.01 M) 94xiNOMENCLATUREANOVA Analysis of varianceB.C. British ColumbiaBOD Biochemical Oxygen Demandcfu/mL colony forming units per millilitreCOD Chemical Oxygen Demanddf degrees of freedomGC/MS Gas chromatography/mass spectrometryN/A not availablen sample sizeTMA trimethylamineTMAO trimethylamine oxideTSA tryptic soy agarTSB tryptic soy brothUBC The University of British ColumbiaxiiACKNOWLEDGENTSFirst of all I would like to express gratitude to mysupervisor, Dr. B.J. Skura for his helpful guidance. Hisconstructive and passionate attitude made this research a veryenjoyable experience. I also want to acknowledge the members ofmy graduate committee, Dr. D. Kitts, Dr. T.D. Durance and Dr. E.Li—Chan for their valuable comments. Joyce and Charlene, oursecretaries, must be recognised for their patient help regardingall office matters.I would also like to show appreciation to GemiaRodriguez—Lopez for her willingness to share her extensiveknowledge in microbiology with me and all the help she providedespecially during the fish tank experiment. Special thanks toJohannes Kötters, from Kiel Germany, who helped me develop manynew ideas for the setup of the experiments. Much of the work donefor this research was accomplished with his help since he waspursuing his practical work in the area of ‘fish disinfectionusing ozone’ for his “Diplomarbeit” for the University of Kielhere at the Department of Food Science at UBC. He also was a goodcomrade outside the lab.More thanks are addressed to Sherman Yee and ValerieSkura whose practical help and advise facilitated me inovercoming any technical hurdles. Mawele Shamaila should bementioned for his help in statistical questions. To Caroline Leeand Angela Nunez I would like to say merci bien for all the helpduring the Ct experiment. Without their helpful hands microbialdata could not have been collected every 30 seconds. Also manythanks to Michel, Beverly, Angela, Joe and all the otherstudents and staff, they made my stay in the department and labmore comfortable.I would also like to thank my wife Indira for herpatience, encouragement and ongoing support.This study was supported financially by the fishingcompany J.S. McMillan Fisheries Ltd. in Vancouver B.C. Thanks totheir cooperation we also obtained the fresh fish for the fishtank experiment. More financial support was provided by theNatural Science and Engineering Research Council of Canada andthe British Columbia Ministry of Agriculture, Fisheries and Food.I. INTRODUCTION 1I. INTRODUCTIONFish are increasingly becoming important as a proteinsource today. There is also growing demand for fresh fish. Sincefish spoils rapidly, effective preservation methods have to beemployed on fishing boats. After the fish are caught, ice isfrequently used to keep the catch cool and to delay bacterialgrowth and endogenous enzyme activity, hence prolonging thefreshness of the fish. In regions, where restrictions apply tothe use of ice, such as developing countries and countries withtropical climates, the use of an alternative preservation agentmay also be necessary. Ozone is already frequently employed todayon British Columbia (B.C.) fishing boats, however, theeffectiveness of ozone in maintaining the quality and inextending shelf-life of fin fish is uncertain.Ozone has been shown to be a strong disinfectant againstbacteria (Blogoslawski et al., 1975) and viruses (Englebrecht andChian, 1985; Farooq and Akhaque, 1983). Ozone, when used fordisinfection purposes toward the same organisms, is moreeffective at lower concentration levels than those required fordisinfection with chlorine (Garey and Dellart, 1979). Evaluationof ozone as an effective anti-microbial agent in the fishingindustry could lead to more uses of ozone as a singleanti-spoilage agent.I. INTRODUCTION 2Rice et al. (1982) summarized various studies that usedozone to possibly extend the shelf-life of many perishable foodsby slowing down microbial spoilage (Rice et al., 1982) . Little isknown about the effects ozone has on the microbial and sensoryproperties of the final products. There is also controversy aboutwhether ozonation might actually increase the microbial load ofthe final products due to increased after—growth ofmicroorganisms caused by additional nutrient availability fromlysed cells (Scott and Lesher, 1963). Fish and other foodcommodities can also harbour pathogenic microorganisms such asListeria monocytogenes (Farber, 1991) which pose a health threatfor consumers. Ozone treatments may eliminate those pathogensfrom seafoods.Much research was done in the area of ozonatingindividual bacterial cultures in aqueous solution (Sobsey, 1989),but little was done to evaluate results of ozonation of wholefood systems, such as fish. The results of previous studies arealso hard to compare since the experimental conditions variedgreatly between the individual research groups performing thetests and many studies are outdated. For this reason, the mainobjective of the present study was to determine the effect ofozone treatments on the sensory properties of fish meat underexperimental conditions which closely simulated situations knownto occur on a “typical’ B.C. fishing boat. To meet theI. INTRODUCTION 3objectives, a wide array of parameters that play a role in thesensory properties of fresh fish meat were monitored. Storageconditions for freshly caught fish, which prevail on B.C. fishingboats, were simulated with and without ozonation to demonstratethe effect of ozone on the individual parameters.A survey performed during the unloading of fish from afishing boat into the storage tank of a processing plant wasperformed to evaluate the effect of the unloading process on themicrobial quality of the fish.The other goal of the research was to develop a method toquantify the sensitivity of individual strains of microorganismsto ozone. An experimental setup was developed to quantify thesensitivity of typical fish spoilage microorganisms to ozone.II. LITERATURE REVIEW 4II. LITERATURE REVIEWA. FUNDAMENTALS OF OZONE CHEMISTRY1. Structure and reaction mechanisms of ozoneOzone (03) was discovered in 1840 by Schönbein (Schönbein,1840) and not much later its triatomic oxygen structure wasdescribed by Hunt (1848) . In aqueous solution ozone is present inthe form of two resonance structures (Figure 2.1) whichillustrates that ozone can act as a dipole, as an electrophilicagent and as a nucleophilic agent (Hoigné, 1988).I I /Q\Q. .Q: . .:Figure 2.1: Resonance structures of the ozone moleculeThe electrophilic reactivity (covalent bonds are formedby accepting an electron pair from a carbon molecule) of ozone isrestricted to certain aromatic compounds with molecular sitesbearing a strong electronic density (OH, NH2 and similarmolecules) . Final products of this reaction are quinoids and, dueto the opening of the aromatic ring structure, aliphatic productsII. LITERATURE REVIEW 5with carbonyl and carboxyl functions. The nucleophilic reaction(covalent bonds are formed by donating an electron pair to acarbon atom) may occur locally on electron deficit bearingmolecular sites (CH3, A1C13 and similar molecules) and, moreoften, on carbons carrying electron—withdrawing groups (Hoigné,1988)As a result of its dipolar structure, the ozone moleculemay lead to a 1-3 dipolar cyclo addition on unsaturated bonds(Figure 2.2), with the formation of a primary ozonide (I). If thereaction takes place in a protonic environment, such as water,this primary ozonide disintegrates into a carbonyl compound(aldehyde or ketone) and a zwitterion (II) that rapidly leads toa hydroxy-hydroperoxide stage (III) that in turn decomposes intoa carbonyl compound and hydrogen peroxide.2. SDecific comDounds that are affected by ozoneAs a powerful oxidant, ozone can react with virtuallyevery class of biological substances, including unsaturated fattyacids, amino acids, pyrimidine nucleotides, and flavins(Carmichael et al., 1982) . Since ozone also reacts very rapidlywith sulphur compounds, thiol-containing enzymes and peptides,such as glutathione are also primary targets (Pryor et al.,II. LITERATURE REVIEW 6/0 0/\00 00 00IR1R° PR—””3 0R2’C3/\H H HC0 R4HR 1G=0+H2OHOO---R3R4 HO7 R4Figure 2.2: Cyclo addition (Criegee mechanism) of ozone onunsaturated bonds (adapted from Bablon et al., 1991)(I = primary ozonide, U = zwitteriofl, III = hydroxy—hYdrOPerOXystage)II. LITERATURE REVIEW 71982). The reactivity with ethylene is highest and ozone reactswith this compound even at -193°C. It reacts with aliphatichydro—carbons at room temperature (Pryor et al., 1982) . On theother hand, ozone shows considerable selectivity. Reactions withalkenes even require activation; the activation energy forreaction with cis-3—hexene, for example, is about 11 kcal/mole(Nangia and Benson, 1980). Table 2.1 illustrates some relativerates for the reactions of ozone with several types of organicgroups that are present in lipids, proteins and sugars.The most important groups of compounds, from a FoodScientist’s perspective, are dealt with in more detail here:2.1 Fatty acidsOzone is highly reactive with tt fatty acids.Naturally occurring unsaturated fatty acids such as 9—hexadecenoic and 9,12—octadecadienoic acids are nearlycompletely consumed by the oxidation with ozone (Reynolds etal., 1989). The reactive site of the unsaturated fatty acids isthe double bond(s) with aldehydes and carboxylic acids formed asintermediate reaction products presumably through the Criegeemechanism (Glaze, 1988) . The reaction of ozone with 9-hexadecenoic acid, for example, results in the immediateformation of heptanal and heptanoic acid on one side of thedouble bond and 9—oxononanoic acid and nonanedioic acid on theother side (Reynolds et al., 1989).II. LITERATURE REVIEW 8Table 2.1: Relative reactivity of several types of organicgroups, that are present in lipids, proteins andsugars with ozone at room temperature.Alkane 0.01-1.0Benzene 0.1sec-Alcohol 7.0Hydroperoxide 180 .0Olefin 400,000.0(adapted from Kurz and Pryor, 1978)turt fatty acids react only slightly with ozone, asindicated by the kinetic constants obtained by Hoigné and Bader(1983) (Table 2.2). However, their reactivity rises withincreasing number of ethylenic bonds in the hydrocarbon chain.When this occurs, the main ozonation products are aldehydes,acids, and hydrogen peroxide. A common example is that of oleicacid ozonation, resulting in the formation of nonanal andhydrogen peroxide (Spanggorg and McClurg, 1978). The highermolecular-weight aldehydes produced by ozonation of fatty acidshave been linked to fruity, fragrant, and orange—like odours(Anselrne et al., 1988) which may affect the sensory propertiesof ozone treated food commodities.II. LITERATURE REVIEW 9Table 2.2: Rate Constants (M’s’) at 20°C for ozonation ofsaturated fatty acids* and organic acids.Compound Value of k03 at Value of k03 atpH=2 pH=8Acetic acid * i0Butyric acid * 10-2Oxalic acid * 10-2Malonic acid <4 7 ± 2Maleic acid 1 * l0 >5 * i0*Values are according to Hoigné and Bader, 1983.k03 = constant obtained by measuring the depletion of ozoneversus time in a batch reactor (k03 = k withstoichiometry = 1)2.2 )inino acids (thiols) and proteinsAmino acids, containing sulfhydryl groups are mostsusceptible to oxidation by ozone (Mudd et al., 1969). Ozone canreact with amino acids [R-CH(NH2)COOH] at two sites, the aminefunction and the R group. The reaction of the primary aminefunction with ozone is largely dependent on pH. The action ofozone on saturated aliphatic primary amines, apparently leads tothe formation of hydroxylamine and finally aldehydes, acids, andnitrate ions (Prengle and Mauk, 1978). Some of the possiblereaction products caused by the action of ozone on individualamino acids are summarized in Table 2.3.II. LITERATURE REVIEW 10Table 2.3: Amino acids and some of their possible ozonationproducts.Amino acid Ozonation productsGlycine Formic acid, nitrateAlanine Formaldehyde, acetic acid, ammonia, nitrateLeucine Formaldehyde, butanol, pyruvic acid, ammonia,nitrateCysteine Cysteic acidPhenylalanine Phenylacetaldehyde, phenylacetic acid, phenylacetamine, phenylhydroxylamine, hydroxy-2—benzoic acid, ammonia, nitrateTryptophan N—formyl—kynurenine, ammoniaTyrosine Di-hydroxyphenylalanine, hydroxy-tyrosinecondensation productsHistidine Ammonia(adapted from Bablon et al., 1991)II. LITERATURE REVIEW 113. Effects of ozone on microorcraiiismsOzone is known to have a strong disinfecting effect onmicroorganisms starting at concentration levels of 0.04 to 0.1ppm (Stockinger, 1959). The primary reaction mechanisms ofbacterial inactivation are thought to be of an oxidative naturewith the cell membrane being the first site of attack (Giese andChristensen, 1954). Glycoproteins and glycolipids (Scott andLesher, 1963) as well as certain amino acids, such as tryptophan(Goldstein and McDonagh, 1975), are the principal targets ofozone attack on the bacterial cell walls. The lethal effect ofozonation could be due to an increase in cellular permeability(Murray et al., 1965) followed by lysis of the cell or thedisruption of enzymatic activity of the cells. Enzymatic systemsthat are targets of ozone treatments include sulfhydryl enzymeswhich enable the bacteria to degrade sugars or produce gases(Vrochinskij, 1963). Besides cell wall constituents andenzymatic systems, ozone can also act on nuclear material, suchas purines and pyrimidines in nucleic acids (Scott and Lesher,1963; Christensen and Giese, 1954) . Pyrimidine bases inEscherichia coli were modified by ozone treatments during astudy performed by Prat et al. (1968) where thymine also wasmore sensitive to ozonation than cytosine and uracil. Theoverall detrimental effect of ozone on bacteria is thought to bea combination of the individual reactions occurring with eachII. LITERATURE REVIEW 12reaction enhancing the effect of the other.B. PPACTICAL APPLICATIONS OF OZONE IN THE FOOD INDUSTRY1. Drinkina water disinfectionThe disinfecting power of ozone was first recognized in1886 by de Meritens who used it to disinfect drinking water(Bablon et al., 1991). Since then, ozone became popular fordrinking water disinfection purposes especially in westernEuropean countries such as France, Germany and Switzerland. InNorth America, chlorination is still the most popular method fordisinfection of drinking water. However, concerns over thepotential hazard to human health posed by the chlorination ofdrinking water has led to the need to reevaluate the use ofalternative disinfectants (Johnson et al., 1982). Ozone isespecially useful in replacing chlorine as a disinfectant sinceit is more effective for disinfection purposes and also hasother beneficial effects when used in drinking waterapplications (Table 2.4). Chlorine is also known to producetoxic compounds such as trihalomethane, chloroform,chlorophenols, haloaceto-nitriles, haloacetic acids,haloketones, dichioroacetic acid and trichloroacetic acid. Thesecompounds often exert mutagenic and/or carcinogenic propertiesin mice (National Cancer Institute, 1976; Herren—Freud et al.,II. LITERATURE REVIEW 13Table 2.4: Treatment goals for ozonation of drinking water1. Disinfection and algae control2. Oxidation of inorganic pollutantsa. Iron and manganese3. Oxidation of organic micropollutantsa. Taste and odour compoundsb. Phenolic pollutantsc. Pesticides4. Oxidation of organic macropollutantsa. Bleaching of colourb. Increasing the biodegradability of organicsc. Destruction of trihalomethane formation potential, totalorganic halide formation potential, and chlorine demand5. Improvement of coagulation(adapted from Bablon et al., 1991)1987; Bull, 1985)Ozone used in water purification systems also produceschemical by-products which have to be evaluated with respect totheir potential health effects. In tests done by Cognet et al.(1986) , ozone appeared to either increase or decreasemutagenicity in bacteria depending on two field parameters:II. LITERATURE REVIEW 14treatment conditions and the type and quantity of organic matterpresent in the raw water. Measurable mutagenic activity inbacteria was shown to be significantly increased by ozone whenhigh levels of organics were present in the treated water(Cognet et al., 1986). After ozonation, oxygenated aldehydes andketones, which most likely have lower mutagenic power than theirprecursors, can be detected in the treated water (Zoeteman etal., 1982). Ozonation by-products include ketoacids formed fromaldehydes. Some ketoacids (eg. glyoxylic acid) are knownbacterial mutagens, therefore their formation may give rise tohealth concerns (Sayato et al., 1989). Ozone inactivatesmutagenic compounds such as pesticides, aflatoxin B1, somealkylating agents and aromatic amines whereas other chemicals,such as dimethyl—hydrazine, are converted into stable mutagens(Caulfield et al. 1979)Another possible disadvantage in using ozone fordrinking water purification purposes is that it does not producebactericidal residuals in the treated water. If however, therisk of post—treatment contamination can be eliminated, thisdisadvantage may turn into an advantage, since no presence of adisinfectant residual also means the absence of off—flavours.Off—flavours are often observed with chlorine treated water(Glaze et al., 1990)II. LITERATURE REVIEW 15The effect of even small concentrations of organic matterpresent in water to be disinfected, has been demonstrated byBroadwater et al. (1973) who found that the presence of trypticsoy broth increased the amounts of ozone needed to killbacteria. For this reason, ozone seems perfectly suited fordrinking water disinfection purposes where, typically, extremelylow amounts of organic materials are to be expected.2. Disinfection of the chilled water used for fish storaae onfishing boatsOzone is used today as a disinfectant on many fishingvessels, in B.C. and is thought to increase the shelf-life andmaintain the quality of fresh fish during storage. The chilledwater of the holding tanks of the boats is ozone treated toreduce its microbial load. Since the chilled water comes incontact with the fish, it obviously becomes highly saturatedwith organic compounds especially proteins and polysaccharidessteniming from the mucous layer, faeces and blood of the storedfish. Since the concentration of organic material in the chilledwater is high, the actual disinfection power of the ozoneapplied is likely to be reduced. Much of the ozone probablyreacts with the organics before even coming in contact with anymicroorganisms.II. LITERATURE REVIEW 16When seawater is used as the chilling water on fishingboats, the formation of bromine is also possible, since seawatercontains about 65 mg/L of bromide ions which in turn producebromine during ozonation (Haag and Hoigné, 1984). Severalauthors have observed the formation of bromoform through theozonation of humic substances in the presence of bromide ions ata neutral pH (Merlet et al., 1980; Haag and Hoigné, 1983; Legubeet al., 1989). By gas chromatography-mass spectroscopy (GC-MS)analysis, Croué (1987) identified different organobromideproducts after ozonation of a fulvic acid solution enriched withbromides. These products, primarily bromoform, bromoacetic anddibromoacetic acids, are known to exert toxic effects onbacteria and are possibly toxic to humans (Gibson et al., 1979).Whether these bromo—compounds produced during the ozonation ofseawater used for cooling purposes on fishing boats, leave anytoxic residual in the fish needs further investigation.3. Use of ozone during the storage of fish and seafoodSince fresh fish spoil rapidly, effective preservationmethods have to be employed during all stages of storage. Ozonehas been shown to be an effective viricidal and bactericidalagent with a higher disinfecting power than chlorine (Pavoni etal., 1972; Gomella, 1972; Fetner and Ingols, 1956; Scott andLesher, 1963)II. LITERATURE REVIEW 17The information available in studies undertaken to useozone for the enhancement of the quality of meat products isvery limited (Greer and Jones, 1989; Chen et al., 1992; Guyerand Jernmi, 1991; Sheldon and Brown, 1986) and only a smallnumber of research groups have tried to evaluate the use ofozone on fish products (Ravesi et al., 1987; Lee and Kramer,1984; Haraguchi et al., 1969; Nelson, 1982). It is also unclear,whether ozone can act as a bactericidal agent in various foodsystems. Research performed by Ravesi et al. (1987) did notdemonstrate beneficial effects from the use of ozonated water orozonated ice in shelf-life extension of Atlantic cod (Gadusmorhua) . A similar conclusion was drawn in a study whichcompared treatment of sockeye salmon, with chlorinated ice andozonated ice (Lee and Kramer, 1984) . In contrast to the previousstudies, a study using the black grouper (Mycroteropercabonaci), showed that continuous treatment of fish, with ozonatedwater and ozonated ice, from time of catch to retail, retardedmicrobial growth on whole fish and the resultant fillets (Riceet al., 1985). Nelson (1982), in a study of the use of ozonatedice containing 0.5 ppm ozone to store fresh Alaskan salmonfound, that the shelf-life could be extended by up to six days,with a significant reduction in bacterial counts anddecomposition products. Earlier this century, Salmon et al.(1937) successfully demonstrated the ozone treatment of seawaterfor oyster depuration. Ozonated ice was used to increase theII. LITERATURE REVIEW 18shelf-life of ice-stored shrimp aboard fishing vessels bypossibly 1-2 days (DeWitt et al., 1984). This general confusionabout the effects of ozone treatment of seafood productsindicates the need for a controlled study which shows the actualeffects of ozone on fish and seafood under conditions whichprevail in the fishing industry.4. The use of the Cxt model to estimate bacterial sensitivitiesto ozoneFish can harbour pathogenic microorganisms such asListeria monocytogenes, which can pose a serious health risk forconsumers (Farber, 1991) . For this reason, it is important todevelop a method to estimate the lethality of ozone on singlestrain microorganisms especially if ozone is to be used as adisinfectant for food commodities. In less recent studies, thefact that ozone dissipates rapidly in aqueous solution, whenorganic contamination is present, was mostly neglected (eg.Katzenelson et al., 1974) . This fact caused an under—estimationof the bactericidal effect of ozone. For this reason, most ofthe recent research undertaken has utilized the Cxt model(disinfectant concentration X contact time) for the estimationof bactericidal activity of ozone (Korich et al., 1990; Wolfe etal., 1989; Wickramanayake et al., 1984). With this model thedecrease in ozone concentration in the reaction vessel over timeII. LITERATURE REVIEW 19is taken into account. Wickramanayake et al. (1984) showed thatthe values generated from the survival curves for the cystsNaegleria gruberi were in close agreement with Watson’s lawwhich has the formCt’=kwhere C is the concentration, n is the coefficient of dilution,t is the exposure time and k is an empirical constant for givenpH and temperature. This shows that the CXt model seems to be auseful model for predicting the sensitivity of microorganisms toozonation.Research was undertaken in the area of ozonatingbacterial cultures in aqueous solution, but little was done toevaluate results of ozonation of whole fish. One also has todistinguish between attached and unattached microorganisms.Boyce et al. (1981) postulated that “adsorbed or embedded virusmay respond to a disinfectant differently from the virus in thefree state, thereby allowing the infectious agent to pass thedisinfection process in a viable state.” However microorganisms,such as Escherichia coli were not protected by the ozone demandfree bentonite clay indicating that a physical protection byclay was not the case (Boyce et al., 1981). In a study by Hoff(1978) , the differences between disinfection effects on virusesadsorbed onto inorganic and organic materials (such as fishII. LITERATURE REVIEW 20skin) were demonstrated. Organic turbidity represented by celldebris had a more pronounced protective effect on associatedviruses than inorganic turbidity (Hoff, 1978).5. Other food system aDDlicationsOzone has a variety of other possible uses, including theincreased shelf-life of eggs, vegetables and meat products, thedelay of the ripening of fruit and the reduction of surface moldgrowth on cheeses (Table 2.5) (Rice et al., 1982). Research donein these areas is very limited and mostly outdated. However, thechemical principles are similar to the disinfection of drinking,and cooling water with respect to the compounds affected byozone. Other areas, where the possible disinfecting effect ofozone may be applied, is for the depuration of mussels andshellfish (Salmon et al., 1937). Also, in a study on poultrycarcasses, ozonation of washing water resulted in a 99%reduction of microorganisms on the carcasses (Sheldon and Brown,1986)The successful evaluation of the effect of ozone as adisinfectant of food materials could lead to more uses of ozoneto serve as a single anti-spoilage agent. It has been reportedin various studies that ozone could extend the shelf-life ofmany perishable foods by slowing down microbial spoilage (RiceTable25:Summaryofliteraturedataontheextensionofstoragelifeofvariousfoods(Riceetal.,1982)FoodPeriodofStorageconditionsReferenceextensionoflifeozonesterilizedice[03]=?Blogoslawski,1982ozonesterilizedice,0.5ppmresidualozonejackmackerel&shimaajibeef(fresh)beef(frozen)1.2—1.6daysunstated30—40%soakin30%NaC1contg.0.6ppm0330-60mm.every2days0.6ppm03inair;0.3°C0.4°C;85—90%RH;0.01—0.02ppm03inair,iforiginalmicrobialcountis<103/cm2Haraguchietal.,1969KaessandWiedemann,1968KolodyaznayaandSuponma,1975poultrybananascranberriesstrawberries,raspberries,currants,grapes2.4days“substantial”damagedoubledsoakinicewaterwhilebubblingin03(3.88ppm)for20mm.afewppm03inair12°C,iffruitisnotwithinafewdaysofitsperiodofrapidripening60°F;0.60ppm03inair2-3ppm03inair,continuouslyforseveralhourseachdayYangandChen,1979Riceetal.,1982Nortonetal.,1968Riceetal.,1982salmonsalmon2-3days50%H HNelson,1982Riceetal..,1982Riceetal.,1982Baranovskayaetal.,1979Riceetal.,1982SheldonandBrown,1986Ravesietal.,1987GreerandJones,1989Gibsonetal.,1960FoodTable2.5(continued):Summaryofliteraturedataontheextensionofstoragelifeofvariousfoods(Riceetal.,1982)StorageconditionsReferenceH HPeriodofextensionoflifeapplesseveralweeks1.95ppm03inair,temp.=?orangesunstated40ppm03inair,temp.=?potatoes6months3ppm03inair;6-14°C;93-97%RHeggs8months0.6ppm03inair;31°F;90%RHpoultryunstated(butstoredfor25mm.in2.2-4.2ppmdecreasesin03watermicrobialnumberswereobserved)AtlanticCodnoextensionpartI:0.6ppm03inice;37°F;partII:03inchilledwater=?;temp.=37°Fbeefinsignificant0.03ppm;1.6±0.2°C;95%RH;0.5m/sairvelocitycheeses63days0.2-0.3ppm03inair,temp.=?t) 1%)II. LITERATURE REVIEW 23et al., 1982) . Despite the proven effectiveness of ozone as abactericidal agent little research was undertaken to developeffective methods for the use of these properties in the foodindustry.III. FISH TANK SIMULATION; Introduction 24III. USE OF OZONE TO EXTEND SHELF-LIFE AND OUALITY OF FRESHFISH DURING STORAGE ON B.C. FISHING VESSELSA. INTRODUCTIONOzone is frequently used on B.C. fishing vessels todecrease quality losses during the storage of freshly caughtground fish such as Silvergrey Rockfish (Sebastes brevispinis)and Pacific Ocean Perch (Sebastes alutus). After catching, thewhole fish are stored in holding tanks on board the boats for upto 8-10 days depending on the length of the fishing trip. Inorder to prevent the fish from spoiling during the storageperiod the temperature in the tanks is monitored and additionalflake ice is added to keep the fish at approximately 0°C.Rapidly growing microorganisms produce large amounts of heat(Atlas, 1988) . If this heat is not dissipated from the affectedareas elevated temperatures, “hot spots”, can develop in theseregions. The chilled water in the tanks is circulated twice aday to avoid hot spots from occurring in the tanks, that couldlead to increased spoilage. To possibly further reduce spoilageof the fish, ozone is injected into the circulating chilledwater by means of a venturi piece (Figure 4.1) . Also to decreasedamage to the fish tissue, caused by the weight of the fishexerted on the fish in lower layers, air is bubbled throughopenings in the bottom of the tank which loosens up any denseIII. FISH TANK SIMULATION; Materials and Methods 25spots in the tanks.Since no scientific evidence exists for the effectivenessof the ozone applied on the fishing boats the main goal of thisstudy was to determine the actual effect of ozone on the fish inthe holding tanks. Therefore, this part of the study was aimedto directly simulate the use of ozone for disinfection ofchilled water on fishing vessels, and to evaluate its effect onthe fish in a pilot plant setup. A study like this would nothave been feasible on a commercial fishing boat due to confinedspace and equipment restrictions. The pilot plant setup allowedfor a strict control of the storage conditions of the fish andreliable evaluation of the parameters of interest.B • MATERIALS AND METHODS1. Experimental conditions and setupFresh, unozonated whole fish (48 h post-mortem SilvergreyRockfish [Sebastes brevispinisj which had been layered in icesince capture) were stored in two stainless steel tanks (83cm X73cm X 61cm [length X width X depth]) with insulated walls tokeep temperature fluctuations at a minimum (Figure 3.1). A lidconsisting of a styrofoam plate covered with polyvinyl-chloridefoil minimized warming of the tank contents through the topIII. FISH PA11K SIMULATION; Materials and Methods 26opening (Appendix A, picture 1). Prior to the experiment, thetanks, pumps and lines were washed and sanitized with a 10,000ppm sodium hypochiorite solution (as is common practise on boats) and prechilled with flake ice. The fish wereimmersed in tap water with flake ice to keep the temperatureclose to the desired 0°C (as on fishing boats). For this purposethe fish were layered into the tanks. Each layer of fish wascovered with flake ice and another layer of fish placed on topof the ice layer until the tanks were approximately 90% full.Finally the tanks were filled with cold tap water. It was notpossible to determine the ratios of ice/fish/water due tofluctuating conditions in the tanks (thawing of ice, taking ofsamples and losses due to spills and leakages)Every day four pails of flake ice (two times 21 litres,after each ozonation run) were added to chill the fish,resulting in an addition of 42 litres of water (20 litres of iceflakes weighed 10 kg which is equal to 10 litre of water) to thetanks per day. The level of chilled water in the tanks was heldconstant during the whole experiment through the removal ofexcess water and fish samples. Similar to commercial fishingvessels, the chilled water (in which the fish were immersed) wascirculated for 40 minutes, by removing chilled water from thebottom of the tank and introducing it back to the top of thetank using a flexible hose (Figure 3.1). Circulation runs wereIII. FISH T?1’JK SIMULATION; Materials and Methods 2703Figure 3.1: Setup of fish tank experiment to simulate storageconditions that prevail on B.C. fishing boats.A = outlet stream from bottom of tanks = 2000 L/h;B = stream passing through bubble column = 400 L/h;C = total circulation stream = A + B = 2400 L/h.not ozonatedBIII. FISH TATTK SIMULATION; Materials and Methods 28performed two times a day (in the morning and afternoon) bypumping (-2000 L/h) chilled water out of an outlet at the bottomof the tank and introducing it to the top of the tank through aflexible hose. The chilled water of one tank received an ozonetreatment during each circulation run while the second tankserved as a control with no ozonation during circulation. Ozonewas generated using a Sander, model 301 ozone generator (Sander,Uetze—Eltze, FRG) from pure, research grade oxygen (Medigas,Vancouver, BC) containing less than 3 ppm of water. This type ofozone generator works by using the corona discharge principle[ozone is generated by electrical discharge in air or oxygen](Figure 3.2), which is presently the most widely used method forozone generation in water treatment. For ozonation of the fishholding tank a constant supply of ozonated chilled water (liquidB, Figure 3.1) was produced in a bubble column, by using asecondary circuit. The bubble column (Figure 3.3), specificallydesigned and built for the purpose of this experiment, consistedof a plexiglass column (height = 1.02 m, volume = 0.055 m3) intowhich chilled water was vigorously sprayed H400 L/h). In thesame chamber ozone was introduced and the spray drops andagitation action of the chilled water ensured that largequantities of ozone were dissolved in the chilled water. Thisozonated chilled water accumulated in the column until itreached the height of the outlet tube, which was connected tothe rest of the circulation stream. By moving the outlet tube inIII. FISH TA1TK SIMULATION; Materials and Methods 29System pressureozone out______cooling_waterlow meter /F].___________1< 0.C —0 —0 —__ __ __ __Needie valvewater outHigh voltage transformerFigure 3.2: Principle of the “Sander Ozone Generator Model 301.”Pure oxygen from a pressure bottle flows close by ahigh voltage electrode (7kV) where some of theoxygen is converted into ozone.III. FISH TAZVK SIMULATION; Materials and Methods 30I0D11 PI I I. ::mGIIIiIiIIIiIi\!IIiI111111111 I jII¼ilIii I I I’IVL 11111 jIILgj.j - --4.\\\\c2\rIIovSchematic view of the bubble column for the ozonization of chilled water used in a fish tank experimentto simulate fish storage conditions on B.C. fishingboats. Chamber data: total height = 1.03 m, Volume= 0.055 m3, inner diameter of inner cylinder = 0.19m. D = pressure release piece of inner cylinder, I= inlet of chilled water, 0 = inlet for ozone, P =pressure release piece of outer cylinder, G = ozonecontaining atmosphere, W = chilled water droplets,WL = ozonated chilled water level, OV = overflow ofozonated chilled water.Figure 3.3:III. FISH TAZ’sIK SIMULATION; Materials and Methods 31and out of the bubble column, the level of ozonated chilledwater could also be adjusted in the bubble column. The portionof chilled water which passed the bubble column (B) representedabout 20 percent of the stream from the outlet on the bottom ofthe tanks (A) . After each circulation/ozonation run the linesand bubble column were rinsed and sanitized using a 10,000 ppmsodium hypochiorite solution. The fish were held for 9 dayswhich simulated the time period of one fishing trip. The controltank was treated in the same manner except that the ozonator wasturned off and thus the chilled water was only exposed to oxygenin the bubble column.Three fish samples were taken after selected circulationruns (in the mornings of day 0, 1, 3, 5, 7 and 9) formicrobiological and colour evaluation. The remainder of thesampled fish after sample taking for microbial and colourevaluation were kept at -30°C for further chemical and sensoryevaluation. In addition to the fish samples, chilled watersamples were taken every single day for determination of pH,nitrate, nitrite, ammonia, biochemical oxygen demand (BOD) andchemical oxygen demand (COD) values. The direct effect ofozonation could be evaluated by comparing the results from theozonated and control samples.III. FISH TANK SIMULATION; Materials and Methods 322. Microbial develoDment on the individual whole fish (skinand gills)For microbial evaluation, a total of three fish sampleswere randomly removed from each (ozonated and control) tankevery second day. Surface and gill samples were taken from eachfish. Colony counts/cm2 were obtained from surface samples byremoving an area of 12 cm2 of skin (3 cm X 4 cm) with a sterilesurgical blade. The samples were taken from a location betweenthe lateral line and the dorsal fin (Figure 3.4). Gill sampleswere taken by removing one whole row of the gills yieldingcolony counts/g of gill samples.Each sample was weighed and diluted with the appropriatevolume of 0.1% (w/v) peptone water (DIFCO Laboratories,Detroit, MI) to obtain a one to ten dilution. The dilution mixwas then blended in a 11Stomacher” (Lab-blender 400) for 2minutes. Using the conventional serial dilution procedure,cfu/cm2 were obtained by plating 20p1 from each dilution onsolidified tryptic soy agar [TSA] (DIFCO Laboratories, Detroit,MI) by using the drop plate method. After incubation at 21°C for48 hours, the visible colonies were counted and the calculatedmicrobial load represented the psychrotrophic count.III. FISH TANK SIMULATION; Materials and MethodsSAMPLING AREAFOR SENSORY PANELS33Figure 3.4: Generalised view of a Silvergrey Rockfish tovisualize the sampling area for microbiological,sensory and colorimetric tests performed during thecourse of storage of the fish in chilled water, withand without ozonation, over a period of nine days.SAMPLING AREAFOR MICRO TESTS LATERAL LINE - -——-—--—-—-‘PECTORAL FINFINIII. FISH TAZ\TK SIMULATION; Materials and Methods 343. Microbioloay of the chilled water at various staaes ofstoraaeAfter circulation of the chilled water in each tank twowater samples were taken (about 50 mL) from the centre of eachtank every second day. Using the conventional serial dilutionprocedure (diluted up to 1/lO8th of the original sample) thepsychrotrophic plate counts (cfu/mL) were obtained by plating20pl (In triplicate) from each dilution on solidified TSA (DIFCOLaboratories, Detroit, MI) using the drop plate method. Afterincubation at 21°C for 48 hours, visible colonies were countedand the calculated microbial load represented the psychrotropiccount.4. Colour chances of fish skin, pills and meat due to ozonationA Hunterlab, model Labscan 2, reflectancespectrophotometer [aperture diameter = 1 cm] (Hunter AssociatesLaboratory Inc., Reston, VA) was used to objectively measureeffects of ozone treatment on the skin of individual fish andfillets. The samples used for this purpose were taken rightbeside the ones used for the microbial evaluation of the skin ofthe whole fish (Figure 3.4). One row of the gills was alsoremoved from each fish for colorimetric evaluation. Aftermeasuring the skin sample the skin was “peeled” off the sample,III. FISH TANK SIMULATION; Materials and Methods 35simulating a skinning machine in the industry, to determinewhether any bleaching effect was apparent on samples a consumeris expected to see when buying a skinned fillet. At the end ofthe experiment (day 9) ozonated and control fish were visuallycompared side by side (“informal visual observation”) to detectany obvious discolouration on skin, gills and meat (Appendix A,picture 4). In addition, photographs of the whole fish, fillets,gills, and intestinal cavity of the final fish were taken inorder to develop a permanent visual record of fish appearance asit was affected by the treatments used (Appendix A, Pictures 3and 4).5. BOD, COD, nitrite, nitrate and ammonia levels of the chilledwaterEvery day one water sample was taken (about 500 mL) fromthe outlet of the flexible hose into the tank. The sample wastaken at the beginning of circulation (1 minute after start ofcirculation) to obtain a representative sample of the tank andtested for 5-day BOD, COD, nitrite and ammonia. In case of theozonated tank the sample was taken immediately before theozonator was activated.Biochemical Oxygen Demand (5-day BOD) evaluation was doneaccording to standard analytical procedures (American PublicIII. FISH TAL\IK SIMULATION; Materials and Methods 36Health Association, 1975). By measuring the °2 concentration(mg/L) of the sample (stored in BOD bottles), before and aftera period of time, the amount of oxygen required for biochemicaldegradation (accomplished by the endogenous microorganisms ofthe water) of organic materials was estimated. BOD also includesthe oxygen used to oxidize inorganic materials such as suif idesand ferrous iron as well as the oxygen utilised to oxidizereduced forms of nitrogen (eg. ammonia into nitrite and nitrate)(American Public Health Association, 1975).Chemical Oxygen Demand (COD) was measured by using a“Hach” test kit #249399 (Hach, Loveland, CO) which works underprincipals stated in the standard method (American Public HealthAssociation, 1975) . COD estimates the sum of all the chemicallyoxidizable substances present in the liquid.Nitrite, nitrate and ammonia tests were performed withtest kits for the measurements of the water quality of aquariums(Red Sea Fish pHarm Ltd. Free Trade Industrial Zone, Eilat,Israel).6. Ozone concentration. H develovment. temDerature of thechilled waterThe storage conditions in the holding tanks were closelyIII. FISH TAI\IK SIMULATION; Materials and Methods 37monitored. Ozone concentrations were determined by using the“DP]Y’ method (Grunwell et al., 1983) which determines the amountof total oxidants present. This method was favoured over anyother method since it is reliable even when measuring turbidsamples (as was the case in this experiment) . The concentrationof total oxidants was measured at the outlet of the reactionchamber (liquid C in Figure 3.1).Temperature measurements of the chilled water in eachtank as well as the ambient air were taken hourly by usingconstantan/copper wire thermocouples hooked up to a Doric, modelMinitrend 25, temperature recorder (Doric Scientific Division,Emerson Electric Company, San Diego, CA) . Thermocouples forchilled water measurements were installed in a central positionin the tanks to ensure a representative sample of the coretemperature of the tanks. Calibration of the temperaturerecorder was done with 0°C ice—water, indicating temperaturefluctuations of ±1°C.The pH of the chilled water was measured on a daily basisusing a Corning, model 220, pH meter (Corning Science Products,New York, NY). A two point calibration was performed with the pHmeter once a day by using two reference buffers (pH 2 and pH 10)and the electrode was stored in a pH 7 storage buffer.III. FISH TANK SIMULATION; Materials and Methods 387. Sensory Danels for filleted fish at various stages of storageFor sensory analysis, each fish was filleted and skinned.Sensory analysis was done on pieces of fillets cut from the sameareas of the fish, on which previous test were performed (Figure3.4). Cooked fish was prepared by wrapping pieces of fishfillets (2 cm x 2 cm) in aluminum foil followed by baking in a195°C oven for 10 minutes. A 10—member trained panel evaluatedthe samples (under red light) on a scale from 0 (unacceptable)to 10 (excellent) using the following attributes: A) aroma, B)flavour, C) texture, and D) overall impression. Samples of day0 and 1 were tested in duplicate and the most important sample(day 9) in triplicate. To reduce judge effects, data of eachattribute from each panelist were z—score standardized to forceall judge scores into the same scale, as suggested by Reid andDurance (1992) . For this purpose the data were z—scaletransformed according to the formula Z = (Y-M) /sd; where Z = zscore of individual data point; Y = individual data point; M =mean of all data points from on attribute of one judge; sd =standard deviation of all data points from the mean (M) of alldata points from one attribute of one judge.8. TMA values of fish filletsFish fillets were also analyzed for trimethylamine (TMA)III. FISH TANK SIMULATION; Materials and Methods 39by using the picric acid method (Woyewoda et al., 1986). Forthe purpose of this test a 7.5% trichioroacetic acid extract ofthe fish samples is reacted with picric acid. TMA is quantifiedby measuring the absorbance at 410 nm. Samples for TMA analysiswere taken from the same area of the whole fish where previoussamples (for microbial evaluation and sensory panel) were takenfrom (Figure 3.4).9. Statistical evaluation of dataAll ANOVA statistical tests performed employed SYSTAT(Systat Inc., Evanstan, IL) statistical software (Addison-WesleyPublishing Company, Reading, MA) except for evaluation ofmicrobial data (paired t-test) which utilised the statisticalanalysis tools of Excel 4.0 (Microsoft Corp., Redmond, WA).III. FISH TANK SIMULATION; Results and Discussion 40C. RESULTS AND DISCUSSION1. General observationsDuring the whole experiment the chilled water of thecontrol tank had a more dirty appearance, resulting from bloodand faeces of the fish. In the ozonated tank those compoundsseemed to have become decolorized by the action of the ozone.Foam occurring on the surface of the chilled water of thecontrol tank was yellowish grey whereas the foam from theozonated tank had a fresh white colour (Appendix A, picture 1).The foam of the ozonated tank also contained flaky particles(presumably coagulated blood and other aggregated organiccompounds). In the late stages of the experiment (after day 7)the control tank started to smell quite fishy” whereas theozonated tank had a less fishy smell but more of a sour note.Due to large amounts of foam development in the bubble columnduring the ozonation, the level of chilled water in the bubblecolumn was lowered and reached almost the bottom of the chamber.Also the flow rate of chilled water through the column had to belowered, to reduce excessive foam formation, resulting in lessthan optimum ozone/chilled water contacting conditions insidethe bubble column. This may have been a reason, why ozoneconcentrations of chilled water leaving the bubble column didnot reach levels higher than 0.4 ppm.III. FISH TAI\IK SIMULATION; Results and Discussion 412. The microbial develoDment on the individual whole fish (skinand gills)The results for the microbial load on the skin (Figure3.5) indicated that ozone had no significant (p>O.O5; paired ttest for individual days) disinfecting effect on immobilizedmicroorganisms on the skin. Control samples even showedsignificantly lower (p<O.O5) amounts of microorganisms on dayfive. ANOVA statistics indicated a significant difference(p<O.O5) between the means of the counts (during the course ofthe 9 days of the experiment) of both treatments (ozonated andunozonated) with higher counts for the samples from the ozonatedchilled water. However posthoc comparison (Tukey HSD multiplecomparison, using SYSTAT software) of the data of individualdays did not show significant (p>O.O5) differences for any ofthe days of the experiment. It appears that microbial growth inthe early stages of storage was slowed down by ozonation,however in the later stages (after day five) the psychrotrophicpopulation increased to a greater extent on the fish stored inthe ozonated water. Since ozonation loosened the slime from thefish surface it is possible that the initial reduction inmicroorganisms on the skin of the fish was due to physicalremoval. This finding is supported by the higher microbialcounts of the ozonated chilled water in the early stages of theexperiment (especially day three) possible due toIII. FISH TANK SIMULATION; Results and Discussioncfu/cm21e81e7421e61e51e40 2 3 4 5 6 7 8days of storageFigure 3.5: Microbial development on the skin of fish (postmortem) during the course of storage over a periodof nine days with and without ozonation.-... -....-..-.....:..-....-....-. .- ---.- --.-.-...-.-.- ..-- .---- .. —— -...:::..::::.::..:.::.:: .::::.:.:::: .::z::::.::::::_ ::E:: ;.:.: ::::::::::I:::.:::.::.: :::.:::::::T— ::, .._=f71--.... .‘ -ozonatedI I III I III I I I I I I I I I It I I III I III! I III Iii t1 9III. FISH TAZVK SIMULATION; Results and Discussion 43the removal of the bacteria from the slime layer of the skininto the chilled water. Haraguchi et al. (1969) experienced alarge initial drop in microorganisms on the skin of jackmackerel (Trachurus trachurus) and striped mackerel (Caranxmertensi) in the early stages of their experiment butexperienced difficulties to decrease the number ofmicroorganisms on the skin after repeated ozone treatments. Ina study performed by Ravesi et al. (1987) ozone treatmentsresulted in no delay of an increase in bacterial counts on thefillets of gutted cod. However their sampling technique toobtain aerobic plate counts is unknown therefore it is difficultto compare their results with the data obtained through thepresent study. Nelson (1982) indicated, in a report on the useof “ozone ice” on salmon, that ozone treatments reducedmicrobial counts on the skin of salmon by 97%.One of the possible reasons for the increase of microbialcounts on the surface of samples from ozonated chilled water maybe the possible increase of organic nutrients due to theoxidation and subsequent breakdown of large organic moleculesinto smaller units produced by ozonation. Another explanationmay be the leakage of nutrients into the chilled water fromnonviable cells caused by ozonation as suggested by Scott andLesher (1963) . The release of essential nutrients from lysedcells may expand the log period of a certain microflora in theIII. FISH TAZVK SIMULATION; Results and Discussion 44medium. If the released nutrients were required for the growthof a broad spectrum of microorganisms then this may have causeda delay of the onset of the lag (stationary) phase or even thedeath phase of the total microflora.The effect of ozonation on the bacterial flora of a foodsystem seems to be by no means predictable and is also dependenton several parameters such as the endogenous microflora presentas well as the organic load of the system. This would explain,why so many researchers obtained different results in theirstudies on the use of ozone on food materials.There was no significant (p>O.O5; paired t-test)difference of microbial counts on the gills (Figure 3.6) exceptfor day seven (counts for sample from ozonated chilled water washigher). ANOVA statistics indicated a significant difference(p<O.O5) between the means of the counts (during the course ofthe 9 days of the experiment) of both treatments (ozonated andunozonated) with higher counts for the samples from the ozonatedchilled water. However posthoc comparison (Tukey HSD multiplecomparison using SYSTAT software) of the data of individual daysdid not show significant (p>O.O5) differences for any of thedays of the experiment. This may be attributed to the fact thatthe gills were mostly protected by the gill covers so that ozonehad little chance to reach the gills during the circulationIII. FISH TANK SIMULATION; Results and Discussion 45cfu/g1e81e71e61e51e4Figure 3.6: Microbial development on gills of fish (post-mortem)taken during the course of storage in chilled waterwith and without ozonation over a period of ninedays.0 1 2 3 4 5 6 7 8 9days of storageIII. FISH TANK SIMULATION; Results and Discussion 46runs. Starting from day three the psychrotrophic counts wereconsistently higher on gill samples taken from fish stored inozonated water. A possible reason for this may be the increasedamount of nutrients available in the ozonated chilled water ofthe ozonated tank due to oxidation of organic compounds such asproteins and polysaccharides (Bablon et al., 1991). Nocomparable data from other research groups was available.3. The microbioloav of the chilled water at various staaes ofstorageThe results from the microbial evaluation of ozonated andunozonated chilled water, showed high fluctuations but nosignificant differences (p>O.05) between the ozonated and thecontrol sample (Figure 3.7). There was an initial large increasein the bacterial counts, in both tanks, after day one which mayhave been due to the sloughing off of bacteria into the liquidfrom the skin surface of the fish. A substantial drop in countsafter day three for the ozonated batch may indicate a goodantimicrobial action of ozone on the liquid. It appears thatozone had a larger impact on the microflora suspended in thechilled water than on the bacteria adherent on the fishsurfaces. This finding was expected since attachedmicroorganisms are generally more resistant to disinfectants(Hoff, 1978). Also the microorganisms in the chilled water hadIII. FISH TAI\JK SIMULATION; Results and Discussioncfu/mLIe8 rn::-. :.:rr.:.r..un;... :.:.nrr.r:..::: ::. :::::_:: :: rr::.:r. ::: : .n::n:471e7Ie61e51e41e31e2IeI— —- ------ — - —-— —° -cOiturn...- - --- - ----- ----------------- ----- ----,-‘ w‘‘I I JIh%J /z:z :zz.z::.::.::::::::z:jz..zzEzr:::..::::zz:::::z:z::.:.:::::.::.zz.:::: .-___________S::EEE:E:_.-. EEEEEE-EEEz=zz::::.::::::::::i [ I 11111111 J I I I I- I: ;:.:V.= -,—,zJ_ -a-- - - -, --/,7.z.:: :.: z-z:- ----1 —--- —------- ---‘ -‘\ H -J,:z:vz::rzz::-------- z- -v ---- --------/,—10 I 2 3 4 5 6 -7 8 9days of storageFigure 3.7: Microbial development in the chilled water used forstorage of fish (post—mortem) over a period of ninedays with and without ozonation.III. FISH TA2VK SIMULATION; Results and Discussioncfu/mL1e81e7leG1e5481e41e31e2lel0 1 2 3 4 5 6 7 8 9days of storage- Microbial quality of ice added to tanksFigure 3.8: Microbial quality of the ice added for cooling ofthe chilled water during the course of storage offish over a period of nine days with and withoutozonation.(The ice was added after each ozonation run)I\—..w-—--- -—.-. .........--.-.......--...I ‘—...---——..—... —---1•---—-- —- -—--.---.-....-....-...—...—- -..II\ I:: ;::::::::.:::::::::::::::::x::::::::z:::: ::::.::z::::: ::::::::::::::::::::::::::::::::::r:L:r:z-z:n-z::zzz:zzzz:: :.:::::z:::r::::: z-..-z:r:::::::. z:z;z:-;t::::.-:xz.. .. .. -.-.- -..-‘‘I- I .. -.----\ I’ : .: ::::::::::::.:.:::\ I‘I‘s.,-..- ..---....-.::.:r.:::/:::.:::::::::::::iE:: .:EE::E:EEEEi....-‘--. - .-.-.:‘V............... .... ..::z.:.:zz::: .::::.z::: :.z:::z:I I—-III. FISH TANK SIMULATION; Results and Discussion 49more direct exposure to ozone since they had to pass through thebubble column in which a high ozone concentration atmosphere wasprevalent. It must also be noted here that the introduction ofthe ice had a diluting effect (unknown proportions) on thechilled water. Since the microbial load of the chilled water isrelated to the amount of microorganisms in the ice added it canalso be assumed that the ice was the cause for some of thecontamination of the chilled water. However, since ice was addedafter chilled water samples were taken for microbial evaluation,the contamination must have occurred at a previous point intime. Unfortunately samples could not be taken on a daily basisdue to time constraints. If ozone would have been used for thepre—treatment of the ice, as previously done (Salmon and LeGall,1936; Nelson, 1982; Rice et al., 1985), the microbial quality ofthe ice may have been improved eliminating uncertainties aboutpossible contamination of the chilled water due to contaminatedice.4. Colour chances of fish skin, gills and meat due to ozonationInformal visual observations as well as colour valuesobtained using the L, a, b scale indicated that the ozonetreatment caused little or no bleaching to fish skin, gills andmeat (Table 3.1; Appendix A, picture 4). Bleaching effects wouldhave been manifest as an increase in L—values (Table 3.1). TheIII. FISH TA?1K SIMULATION; Results and Discussion 50Table 3.1: Colour of gills, skin and flesh of SilvergreyRockfish during storage of whole fish in ozonated andunozonated water over nine days.GillsControl OzonatedDay L a b L a b0 N/A N/A N/A N/A N/A N/A1 22.75 5.62 4.12 26.27 5.34 6.113 32.09 2.55 5.42 29.58 2.44 5.045 32.66 3.58 7.13 33.36 3.11 7.787 29.99 0.86 4.95 32.44 1.97 6.829 33.73 1.50 7.85 34.32 0.97 6.31SkinControl Ozonatedy L a b L a b0 N/A N/A N/A N/A N/A N/A1 28.40 0.34 0.25 33.63 0.87 3.073 30.81 0.65 2.04 30.89 0.38 1.655 30.33 0.17 0.01 35.41 0.76 1.547 26.23 0.65 3.16 25.74 0.69 2.549 31.51 0.77 2.29 32.90 0.33 2.35MeatControl OzonatedDay L a b L a b0 N/A N/A N/A N/A N/A N/A1 33.25 —1.89 2.80 36.27 —0.41 4.123 41.67 —1.93 2.09 41.27 —2.02 0.605 36.67 0.30 3.20 43.27 —2.51 2.117 41.98 —2.59 1.44 40.66 -1.37 6.089 41.15 —1.69 3.35 42.67 —2.70 4.32N/A = not available, n = 3 for all samplesIII. FISH TANK SIMULATION; Results and Discussion 51high organic load of the chilled water probably prevented theozone from causing effects on the fish surfaces by reacting withthe ozone. Similar results were obtained by Sheldon and Brown(1986) who did not notice any bleaching of of chicken skin afterozone treatment except for a slight bleaching of the skin on thedrumsticks. Since no noticeable bleaching effect was present onthe skin of the whole fish it is almost certain that no colourfading could have happened on the skinned meat of the fishsince ozone would not likely have penetrated the fish skin.The only apparent trend in all the samples (control andozonated) was the loss of redness (decrease in a value) of thegill samples and an increase in lightness (increase in L value)of gill and meat samples (Table 3.1). Silvergrey Rockfishvisually observed on a B.C. fishing boat (during a trip,visiting a Vancouver fish processing plant) showed quiteextensive bleaching effects (of the skin) especially for fish onthe top of the tank where the ozonated chilling water wasintroduced. This was probably due to undissolved ozone whichcame in contact with the fish skin. Less efficient contactingbetween ozone and chilled water led to substantive amounts ofozone escaping through the man hole (Figure 4.1) . However, nofurther colorimetric tests were performed on the observed fish.III. FISH TANK SIMULATION; Results and Discussion 525. BOD. COD, nitrite, nitrate and aimnonia levels of the chilledwaterBOD values are frequently used to estimate waste loadingsto waste water treatment plants (Young, 1984). The test for BODis of limited value since the laboratory environment does notreproduce the conditions such as temperature, liquid movement,biological population and oxygen concentration (American PublicHealth Association, 1975). However BOD is the only parameteravailable to water chemists that provides an indication of theamount of biodegradable materials present (Young, 1984).Differences in BOD values for ozonated and unozonatedchilled water are extremely hard to interpret since BOD isaffected by the type and amount of microorganisms present aswell as the type of organic materials present (Waite, 1984)During the course of this experiment the BOD value showedsimilar trends for water samples taken from control and ozonatedtanks (Figure 3.9). The BOD value increased from an initiallylow value (45-55 mg/L) through a peak at day 4 approaching 240-260 mg/L in both tanks. This pattern was expected since in bothsystems microbial growth peaked at day 3 (Figure 3.7; BOD laggedone day after microbial growth values). High BOD (and COD)levels after this bacterial ‘bloom” may be caused by theremaining dead cells in the chilled water.III. FISH TANK SIMULATION; Results and Discussion 53350-300-250200-0)E II150 - II100- II50- •0- I I0 1 2 3 4 5 6 7 8 9days of storage in chilled water- -- BOD-ozonated water • BOD-control waterFigure 3.9: BOD development of the chilled water used forstorage of whole fish over a period of 9 days withand without ozonation.III. FISH TANK SIMULATION; Results and Discussion 54COD is a good, non-specific measure of the organiccontent of an aqueous system. High COD levels indicate a highorganic load, which may serve as substrate for ozone reactions.The relatively high total number of COD forming substances inthe chilled water was probably one of the main reasons why itwas difficult to attain ozone concentrations in the outlet ofthe bubble column above 0.4 ppm.As expected, the COD of the ozonated and unozonated waterwas higher than BOD the since microorganisms can not degradecertain organic compounds which are oxidized by strong chemicaloxidizing agents as employed in the COD test (Waite, 1984). CODvalues showed similar trends in both tanks (Figure 3.10). CODvalues for ozonated chilled water were lower than those for theunozonated chilled water. A lower COD for the ozone treatedwater was expected since ozone is able to break down most of theorganic substances (proteins, polysaccharides and other)representing the COD value (Bablon et al., 1991). However theoxidizing action of ozone did riot change the concentrations ofchemically oxidizable materials (as measured with COD tests)occurring in the chilled water of the ozonated tank as comparedto the control tank. The large fluctuations of COD values thatoccurred in the chilled water of both tanks between day four andday five are not easy to explain. They may be the result of thelarge oxidizing action (to biodegrade organic compounds) of theIII. FISH TAWK SIMUI1ATION; Results and Discussion 5516001400k.600 -400- I I I I I I I0 1 2 3 4 5 6 7 8 9days of storage in chilled water- -- COD-ozonated water • COD-control waterFigure 3.10: COD development of the chilled water used forstorage of whole fish over a period of 9 days withand without ozonation.III. FISH TANK SIMULATION; Results and Discussion 56increased amount of psychrotrophic microorganisms present in thechilled water of both tanks on day three (Figure 3.7). The sametheory would be true for the BOD values of both tanks.Ammonia occurs in aqueous solution as an ammoniumion/ammonia equilibrium (NH4/3) which is influenced by pH andtemperature (Emerson et al., 1975). At a temperature of 1°C anda pH level of 7.0 (which were the conditions prevalent in bothof the tanks during the experiment), only 0.0898% of the ammoniais present as NH3 (Emerson et al., 1975). Therefore it can beassumed that the concentration of ammonia (NH3) in the ozonatedchilled water did not exceed 0.006 mg/L until day six ofstorage. The maximum value for the total ammonia in both tankswas reached on day 8. However, ammonia levels stayed relativelylow in the control tank until day 6 after which they started toclimb (Figure 3.11) . The action of ozone on ammonia is slowespecially at pH levels below the PKa of 9.3, of the reaction(Bablon et al., 1991). For this reason, the direct influence ofozonation on ammonia levels should have been small (pH was 7.0in both tanks during the experiment).Ozone can react with the ‘2 group of amino acidsstarting with the electrophilic process on nitrogen (Figure3.12) and finally leads to decarboxylation plus a build-up ofacids, aldehydes, nitrate ions and ammonia (Bablon et al.,III. FISH TAWK SIMULATION; Results and Discussion 5710987 I’‘I /:.‘ \6 / — -— — Ammonia (ozonated)- - -- Nitrate (ozonated)5 - —— Nitrite (ozonated)Ammonia (control)• Nitrate (control)• Nitrite (control)3/,2Days of storageFigure 3.11: Ammonia, nitrite and nitrate development ofchilled water used for storage of whole fish duringa period of 9 days with and without ozonation.III. FISH TANK SIMULATION; Results and Discussion 58R-CH -NH2 2O Electrophilic AttackR-CH2-Nalpha-Carbon Pathwa/” (S 0%Nitogen PathwayHalpha-Carbon Pathway HR-CH-NH2or R-CH-N RCHN%mlnaUonOHOH0 R-CH•N-OH4’ + —R-C + NH + OH Oxidation of4 Double bond•0 - +R-C +N03HFigure 3.12: Partial schematic of ozonation of primary amines(adapted from Laplanche and Martin, 1982)III. FISH TAL\IK SIMULATION; Results and Discussion 591991) . Ozone can further produce ammonia and nitrate from the N-containing R—groups of certain amino acids, such as cysteine,histidine, tryptophan and tyrosine (Bablon et al., 1991). Thiscould be the reason for the early increase of total ammonia inthe ozonated tank, since amino acids derived from proteins(blood, faeces, slime etc.) were most likely present in theozonated chilled water. Nitrite (NO2), which is the first stageof oxidized arnmonium, was present in both tanks only in minuteamounts. This result was expected since nitrite is highlyreactive and is oxidized to nitrate or reacts with other organiccompounds such as amines, phenols and thiols (Haag et al.,1984). The nitrite concentration in the ozonated chilled waterreached a small plateau during the period of days 4 and 5,whereas in the control tank it was present in amounts below thedetection limit of the method employed. The reason for the smallplateau in nitrite concentration in the ozonated chilled watermay be the increased dissociation of amines from proteins(Laplanche and Martin, 1982) present in the chilled water (whichis also the possible reason for the increased ammonia levels inthe ozonated chilled water) which become partially oxidised tonitrite by ozone or the free radicals present.Nitrate (NO3) was considerably higher in the ozonatedwater during all phases of the experiment. This large differenceindicates that more of the ammonia was completely oxidized inIII. FISH TAJVK SIMULATION; Results and Discussion 60the ozonated chilled water. This oxidation phenomenon was notobserved for water in the control tank. Several Pseudomonasspecies can reduce nitrate into nitric oxide and ammonia (Payneet al., 1971). After day six, nitrate levels began to decline inthe ozonated chilled water which may be the result of theincreased occurrence of denitrifying bacteria which are capableof reducing nitrate to nitrite, oxides of nitrogen and ammonia(Young, 1984). Since nitrite readily reacts with other organiccompounds, during the denitrification process by microorganisms,it was not expected to be present in both tanks at substantiallevels. Addition of ice to the tanks probably accounts for someof the fluctuations that occurred in arnmonium, nitrite andnitrate levels simply by exerting a dilution effect on thesystems (Young, 1984).6. Ozone concentration. PH development, temperature of thechilled waterThe ozone concentration (total oxidants as measured withthe DPD method) in the outlet of the reaction chamber reached anequilibrium of less than or equal to 0.4 mg/L after 5 minutes ofstarting ozonation. Since the flow through the reaction chamberwas about 20% of the total circulation fluid, it is assumed thata total oxidant concentration of 0.08 mg/L was present in thefluid at the inlet into the fish tank (Figure 3.1).III. FISH TANK SIMULATION; Results and Discussion 61The pH levels of chilled water in both tanks were closeto neutrality throughout the whole experiment. Chilled watertemperature of both tanks reached a maximum of 4°C at day 4 butwas in the range between 0°C and 2°C for most of the experiment.Higher ambient temperatures at days 4 and 5 could have been thereason for the slightly higher temperature of the chilled waterin the tanks at those days. Most of the ice added after eachcirculation run melted but some ice was always present in thechilled water of both tanks.7. Taste Danels for filleted fish at various staaes of storageSensory panel data indicated no significant (p > 0.05)differences between the fish from ozonated and unozonatedchilled water after nine days of storage (Table 3.3). Howeverthe scores for the fish stored in ozonated chilled water wereconsistently slightly higher (except for flavour). The largest(but still not significant) difference was for the aroma datafor fish stored for 9 days which may be interpreted with theozonated fish being 15% better (Table 3.2). This could beattributed to the lower TMA values of the ozonated fish whichshould make the fish smell less fishy (Figure 3.13). After the9 day storage period the ozonated fish had a slightly fresher,less fishy smell which may also be indicative of a slightlyhigher fish quality.III. FISH TANK SIMULATION; Results and Discussion 62Analysis of variance indicated significance of judgeeffects (P<O.05) in all cases (Table 3.3), although ozonation Xjudge interaction was not significant (p>O.05). Examination ofthe individual scoring patterns of the judges, lead to theconclusion that despite training and reference samples, judgestended to use different scaling techniques. One judge, forexample, used the full scale while another judge only used theleft half of the scale. For this reason the data were z—scoretransformed to force all judge scores into the same scale (Reidand Durance, 1992). ANOVA statistics indicated that judgeeffects were not significant (p>O.O5) anymore after ztransformation of the data, however, no increase in the power ofthe analysis resulted (Table 3.4).Fillets from ozonated fish that a consumer may purchase,are not expected to have a better aroma, flavour or texturethan fillets from fish stored without ozone treatments.III. FISH TANK SIMULATION; Results and Discussion 63Table 3.2: Mean values for sensory panel scores for fish storedin ozonated and unozonated chilled water.Aroma FlavourControl Ozonated7.35 (±2.29) N/A6.30 (±2.74) 6.56 (±3.10)4.25 (±2.38) 4.17 (±2.71)Overall ImpressionControl Ozonated7.12 (±2.14) N/A5.82 (±2.62) 6.45 (±2.53)4.24 (±2.46) 4.60 (±2.69)N/A = not available)Sample Control OzonatedDay 0 6.44 (±2.61) N/ADay 1 5.95 (±2.67) 6.10 (±2.72)Day 9 4.55 (±2.84) 5.22 (±2.98)TextureSample Control OzonatedDay 0 7.32 (±2.14) N/ADay 1 5.73 (±2.80) 6.68 (±2.39)Day 9 6.49 (±2.48) 6.55 (±2.41)(mean values ± standard deviation,III. FISH TAIVK SIMULATION; Results and Discussion 64Table 3.3: Analysis of variance of sensory panel scores for 9day old Silvergrey Rockfish with and withoutozonation of the chilled water.Aroma Flavour Texture Overall dfImpressionTreatment F-value 1.30 0.03 0.01 0.431p—value 0.26 0.86 0.92 0.52judge F-value 4.98 8.07 3.20 5.089p—value <0.001 <0.001 0.005 <0.001TreatmentX judge F—value 1.37 1.63 0.67 0.609p—value 0.23 0.14 0.72 0.79(df = degrees of freedom, df for total = 9)Table 3.4: Analysis of variance of sensory panel scores for 9day old Silvergrey Rockfish with and withoutozonation of the chilled water; z—transformed sensorydata.Aroma Flavour Texture Overall dfImpressionTreatment F-value 0.96 0.53 0.03 0.301p—value 0.33 0.47 0.85 0.58TreatmentX judge F—value 1.08 0.69 0.68 0.519p—value 0.40 0.41 0.72 0.86(df = degrees of freedom, df for total = 59)III. FISH TANK SIMULATION; Results and Discussion 658. TMA values of fish filletsMost saltwater fish contain TMA-oxide (TMAO) forosmoregulation, with gadoid species (such as Rockfish andPacific Ocean Perch) containing particularly high levels (Hebardet al., 1982). The breakdown of TMAO into TMA has beenimplicated with the spoilage of fresh and frozen fish (Laycockand Regier, 1971). TMA is a good indicator for the microbialspoilage of fresh fish since many bacteria such as Pseudomonasputrefaci ens can reduce TMAO into TMA (Regenstein et al., 1982).Much of the fishy odour of spoiled fish can be attributedto TMA which is a volatile compound with a very low odourthreshold (Ikeda, 1979). For this reason, the TMA results wereexpected to correlate with the off—flavour results from thesensory panel. Since TMA values parallel bacterial growth (Wongand Gill, 1987) slightly higher TMA levels were expected at theend of the experiment (after day 5) in the fish stored inozonated chilled water than in fish stored in unozonated chilledwater (control fish) . However starting from day seven the TMAlevels were significantly lower (P<O.05) for the fish stored inozonated chilled water compared to the control fish (Figure3.13, Table 3.5). This finding may be explained by the fact thatozone may have acted on the TMA breaking it down into other non—odour producing compounds. However since ozone was unlikely toIII. FISH TANK SIMULATION; Results and Discussion 66Table 3.5: ANOVA statistics of TMA data from fish samples of day7*and 9 of storage in ozonated and unozonatedchilled water.Day7 Day9F—value 125.22 65.08p—value <0.05 0.001df 2 2(*significant at P<0.05, + significant at p<O.OOl, df = degreesof freedom)penetrate fish muscle tissue during the experiment TMA may onlyhave leached out of the tissue into the chilled water and wasbroken down there.A more possible explanation for the lower TMA levels inthe ozonated fish may be that the ozone treatment shifted themicroflora associated with those fish in a direction wherebacteria with less TMA producing ability were present comparedto the control fish. In a study performed by Grain (1992) onlynineteen strains out of 50 Pseudomonas species isolated fromfish produced strong, rotten of f-odours. Thirteen strainsproduced weak off- odours and the remaining 18 strains producedIII. FISH TANK SIMULATION; Results and DiscussionTMA-N (mg/bOg of fillet)1614121086420670 1 2 3 4 5 6 7 8 9days of storageFigure 3.13: TMA development in silvergrey rockfish over a 9day storage period in ozonated and unozonatedchilled water.1 IIII. FISH TAZVK SIMULATION; Results and Discussion 68only very weak or no of f-odours. This shows that not allbacterial species have the same capabilities to produce of f—odours such as those resulting from the presence of TMA. In anearlier experiment performed by Yang and Chen (1979), themicroflora of ozone treated poultry meat showed an increase ingram-positive cocci (from 39.6% to 52.7%) indicating thatvarious microorganisms differ in their sensitivity to ozonetreatments. These studies clearly indicate that a changingmicroflora will likely have different effects on the developmentof TMA depending on the spoilage potential as well as thesensitivity of the endogenous microorganisms to ozone.Fish with TMA levels above 10 rng/l00 g fish fillet aregenerally considered unacceptable (LeBlanc and LeBlanc, 1992).By only using TMA values as a measure for fish quality, thiswould mean that the ozonation treatment increased the shelf-lifeof the fish by approximately 36 hours. When samples of fish(after 9 days of storage) were given away to members of the FoodScience Department of U.B.C. for home-cooking purposes thecomments about the flavour of the samples were all positive (forboth treatments) indicating that even at the TMA levels abovethe critical point, fish were still quite acceptable for humanconsumption.III. FISH TANK SIMULATION; Conclusion 69D. CONCLUSIONFrom the results of the present study it can not beconcluded that ozone treatments of the chilled water used forstorage of Silvergrey-Rockfish prolong the shelf-life of thefish. Ozone had little effect on microbial and sensoryproperties of Silvergrey Rockfish. Since no significant sensoryor colorimetric advantages resulted from the ozone treatment,the use of ozone as a means of maintaining the quality of fishon fishing boats does not appear to be justified. The fact thatthe ozone treated fish had significantly lower TMA levels after9 days of storage did not manifest itself in the sensory testsas a better tasting fish. However, since the chilled water ofthe ozone treated tank smelled fresher and less offensivecompared to the control tanks chilled water, it is obvious whythe captains of fishing boats keep on using ozone during fishingtrips. Since a major detriment to the antimicrobial action ofozone is the high organic load in the chilled water it may bepossible to use a modified setup of equipment on the fishingboats. One of the main changes may include the addition of afoam skimmer to remove the excess organic material collected inthe foam accumulating during ozonation. Foam skimmers arerelatively cheap and easy to install. Applications of foamskimmers can be found already in the home aquaria, publicaquaria and for aquaculture purposes (Sander and Rosenthal, 1975)IV. MICROBIAL SURVEY; Introduction 70IV. SURVEY OF THE MICROBIAL CONDITIONS OF FISH AND “PROCESSINGLIOUID DURING THE UNLOADING OF FISH AT A FISH PROCESSINGPLANT IN VANCOUVER. B.C.A. INTRODUCTIONA typical fishing trip, in B.C. for groundfish, such asPacific Ocean Perch, may last up to 14 days. Freshly caught fishare manually transferred through the man hole, into holdingtanks by means of a gaff. Fish are hooked in the gill region,with the gaff, also causing occasional puncturing of the lateralmuscle region increasing the chance of contamination of theinitially sterile muscle tissue (Skura, 1993) . Since the boatsdo not have ice making or cooling facilities on board, theholding tanks are partially filled with flake ice, produced fromcity water at the processing plant, before leaving the harbour.Prior to loading the fish in a specific tank, it gets floodedwith seawater. Air is bubbled in from the bottom of the tanks tofacilitate the mixing of the ice with the seawater and to avoidclumping of the ice (champagne system). In order to keep thetemperatures in the tanks at low levels (between —1°C and 2°Cpreferably) more ice is added to the tanks during the remainderof the fishing trip (Kötters, 1993). To avoid hot spot formationin the tanks the chilled water is circulated once or twice aday. This is accomplished by removing the chilled water fromIV. MICROBIAL SURVEY; Introduction 71openings in the bottom of the tanks by means of strong pumps (upto 1620 L/min. pumping volume) and inserting it back into thetanks through the opening located at the highest position of thetank (the man hole, Figure 4.1) (Kötters, 1993).During the circulation run, ozone is injected into thechilled water stream through a venturi piece. For this purpose,ozone is produced on board by a “corona discharge” typegenerator (manufacturer unknown) employing a 10 kV voltage inits reaction chamber (Kötters, 1993). Air used for ozoneproduction is sucked in from the man hole which leads to themachine room (since the ozone generator is located here) . Thismeans that the air has a relatively high temperature (25°C -35°C) and may contain oil or solvent vapours with oxygen(Kötters, 1993) . Before reaching the ozone generator, the airused for ozone production is pushed through a drying unit whichalso splits nitrogen dioxide (NO2) to enrich the air stream(Kötters, 1993). The pipe used for circulation purposes (whichalso acts as a static mixer) is about 15 m long and has aninside diameter of 7.62 cm. Ozone is injected into thecirculation stream with a flow rate of 60 L/min. at aconcentration of 37 g ozone/m3 gas (Kötters, 1993)At the end of the fishing trip the fish are unloaded atthe fish processing plant by means of large suction pumps andIV. MICROBIAL SURVEY; Introduction 72A—60 /m4 man holes___v -0 T\max__\1620 L/miipost—mortem fish in chilled waterFigure 4.1: Schematic representation of one of the holding tanksand recirculation circuit of chilled water on thefishing boat “Arctic Ocean.” Ozone production iscarried out through the ozone generator (0) usingair (A) which was previously dried and oxygenenriched (S). T=holding tank for fish, P=chilledwater pump, V=venturi piece. (adapted from Kötters,1993)IV. MICROBIAL SURVEY;_Introduction 73conveyor belts (Figure 4.2). If the equipment used for unloadingis not in good sanitary condition it can pose a risk ofcontaminating the fish during the unloading process.QOflVYOFbelt holdingFIch pump____sortingcoor ls’e oor_________belt plant beltFigure 4.2: Flow chart of the unloading process of fish at afish processing plant of in Vancouver, B.C.The purpose of this part of the experiment was toevaluate the impact of the fish unloading process (from the boatat a processing plant) on the microbial conditions of the fishskin.IV. MICROBIAL SURVEY; Materials and Methods 74B. MATERIALS AND METHODS1. Microbiological tests of the skin of individual whole fishIn a survey at a fish processing plant, the microbialconditions during the unloading of ground-fish were evaluated byswabbing fish surfaces. Samples were taken before the unloadingprocess, during unloading, after the sorting table and in theholding tank inside the plant, shortly before filleting. For thepurpose of swabbing, templates were prepared by cutting squaredwindows with a side length of 7.1 cm from aluminum foil yieldingan open area of 50 cm2 in the foil. Swabbing was performed on theside of the fish in the region of the lateral line towards thepelvic fin (Figure 3.4) by rolling a sterile cotton swab threetimes over the area defined by the aluminum foil window. Thecotton swabs were previously moistened with 0.1% sterile(autoclaved at 121°C, 15 psi for 20 mm.) peptone water (DIFCOLaboratories, Detroit, MI) and immersed into 5 mL of 0.1%sterile peptone water, in a screw capped Pyrex® test tube, modelNo.9825 (Corning Science Products, New York, NY) and shakenvigorously. This solution represented the first dilution for thedilution series performed in the lab. Further dilutions wereperformed by suspending 2Opl of sample into l8OpL of sterile0.1% peptone water, representing a 1:10 dilution. Dilutions weremade up to 1:1,000 of the original first dilution. 2OpL of allIV. MICROBIAL SURVEY; Materials and Methods 75dilutions were plated onto TSA plates (DIFCO Laboratories,Detroit, MI) by using the drop plate method. The plates werethen incubated at 21°C for 48 h. For determination of thepsychrotrophic plate count plates from dilutions with colonycounts closest to twenty were used. Samples were taken intriplicates.2. Microbiological tests of fish storaae liguidsSamples of chilled water from the holding tanks of theboat and inside the processing plant were taken by immersing aplastic pail, which was previously sterilized with a 1% bleachsolution, into the liquid. 100 mL of the sample were stored in150 mL sterile whiripak plastic bags on ice for furtherinvestigations in the lab. From these 100 mL, 20 p1 weresuspended into 180 pL of sterile 0.1% peptone water,representing a 1:10 dilution. Dilutions were made up to 1:10,000of the original first dilution. Further steps for thedetermination of the psychrotrophic plate count employed thesame method as previously described for the skin samples of thewhole fish during the experiment to simulate storage conditionsas prevailing on B.C. fishing boats in a pilot plant setup(Section 3.2).IV. MICROBIAL SURVEY7 Results and Discussion 76C. RESULTS AND DISCUSSIONThe initial microbial load on the surface of the fish wasrelatively small, but comparatively large in all the chilledwater samples (Figure 4.3) . The further increase in microbialload of the chilled water during the pumping step indicated thatbacteria were most likely sloughed off the surface of the fishinto the liquid. No chilled water was present during theconveyor belt transfer of the fish, therefore fresh chilledwater and flake ice (unknown proportions) was used inside theplant’s holding tanks to keep the fish cold. Despite the factthat fresh chilled water was used inside the plant for coolingpurposes the bacterial load of the chilled water was stillrelatively high, indicating further sloughing off ofmicroorganisms from the skin of the fish. This is also suggestedfrom the bacterial counts of the skin of the fish during theunloading process. The microbial load on the fish surface stayedlow up to the filleting operation despite high microbial levelsin the chilled water.IV. MICROBIAL SURVEY; Results and DiscussionBoatduring unloadingafter sortingholding tank insideplant1 E÷1 1 E+2 1 E+3 1 Ei-4 IE+5 1E+6 1E÷7 1E+8cfu1 E+977LI Fish surface (cfu/cm2) LI processing liquid (cfu/mL)Figure 4.3: Microbiology of fish and chilled water samplestaken during the unloading process of fish at afish processing plant in Vancouver, B.C.2E÷89E+2 L_-J6E+8E+25E+26E+7HIV. MICROBIAL, SURVEY; Conclusion 78D. CONCLUSIONIn general it can be inferred that the unloading processat the fish processing plant did not contribute to an increasein the microbial load of the fish surface. This means that anincrease in contamination of the fish surface due to contactwith plant equipment or processing liquid is unlikely to occurup to the point of filleting. It is also interesting to notethat despite the high level of contamination of the pump fluid,as well as the cooling water inside the plant, no increase inthe contamination of the fish surfaces was observed.V. DISINFECTION EXPERIMENT; Introduction 79V. MODEL EXPERIMENT TO DETERMINE THE SENSITIVITY OFTYPICAL. FISH SPOILAGE BACTERIA TO OZONE TREATMENTSA. INTRODUCTIONIn order to determine the amount of ozone that should beapplied to a food system, to induce a certain antimicrobialeffect, studies have to be undertaken to evaluate thesensitivity of individual bacterial species to certainconcentrations of ozone. Since it is problematic to keep ozoneconcentrations constant in an aquatic medium, an attempt wasmade to use the CXt model to calculate the sensitivity ofbacteria to ozone treatments. In other CXt model studies (Wolfeet al., 1989; Wickramanayake and Sproul, 1988) microorganismswere subjected to a specified initial dose of ozone over acertain period of time. CT values were calculated by multiplyingthe initial ozone concentration (in milligrams/litre) X contacttime (in minutes) which resulted in either 90 or 99% reductionof microorganisms. Since contact times were between 6 and 12minutes and the experimental conditions were not strictlydefined (eg. : natural decay of ozone was neglected) the actualamount of ozone acting on the microorganisms was only estimated.The experiments discussed in the present study eliminated thestated problem by: a) strictly defining and monitoringV. DISINFECTION EXPERIMENT, Materials and Methods 80experimental conditions and, b) creating experimental conditionswhere the natural ozone decay can be neglected (since it isextremely small during the crucial experimental time period)Disinfection tests on single strain microorganisms wereperformed to evaluate the microbial effectiveness of ozone. Withthese methods employed in the present research, one should beable to calculate a value, equivalent to the CT value, for eachmicrobial strain indicating the sensitivity of the strain to acertain concentration of ozone over a certain period of time.B. MP.TERIALS AND METHODS1. Exoerimental DrocedurePhosphate buffers (at 22°C and 0°C) containing residualozone concentrations of 0.8-1.0 ppm were prepared (see SectionV.3) in a 2 litre Erlenmeyer flask and continuously stirred witha magnetic stirrer. Samples for determining the ozoneconcentration of the reaction mix and microbial load were takenat approximately 30 second intervals. At this stage of ozonedecay in the system, the rate of ozone decomposition was sosmall that the loss of ozone due to decay during the criticaltime of the experiment could be neglected. The ozoneconcentration was determined roughly every 2 minutes. When theV. DISINFECTION EXPERIMEL\IT, Materials and Methods 81concentration reached the desired level, the 24 mL of bacterialsuspension was added and a stop—watch was set. 1 mL samples forbacterial counts were taken every 30 seconds during the criticalfirst 10 minutes of the experiments and in larger time intervalsafter 10 minutes. These samples were diluted 10 times in 9 mL(0.1%) peptone (DIFCO Laboratories, Detroit, MI) watercontaining 15 mg/L sodium sulfite to neutralize any residualozone which might have been still present. Serial dilutions upto l0 were made and plated on TSA (DIFCO Laboratories, Detroit,MI) plates using the drop plate method. Inoculated TSA plateswere incubated at 25°C for 48 h.2. Bacterial isolatesTwo bacterial strains isolated from the surface of a fishtested in a previous experiment (Alteromonas putrefaciens andAlteromonas punctata subsp. punctata) as well as commerciallyobtained cultures (Staphylococcus aureus [American Type CultureCollection, Rockville, ML] and Pseudomonas fragi [American TypeCulture Collection, Rockville, ML]), were propagated. The singlestrain cultures were stored at -80°C in 10% glycerol solution.After thawing, 1 ml of the culture was transferred into a screwcapped Pyrex® test tube, model No.9825 (Corning ScienceProducts, New York, NY) containing 5 mL of TSB (Figure 5.1) andincubated at the desired temperature (5°C for cultures used forV. DISINFECTION EXPERIMENT, Materials and Methods 820°C experiment and 21°C for cultures used for 22°C experiment)for 24 h in a rotary-type shaking water bath, (LablineInstruments Inc., Meirose Park, IL) at 100 rpm. One loop full ofthis solution was streaked onto TSA and incubated for 48 h at25°C to produce individual colonies of the strain. From thisplate, one colony was transferred into 5 mL TSB and incubated aspreviously described, over night. From this suspension, 1 mL wastransferred again to 5 mL TSB and incubated overnight aslmL lloopf.I1 lcoloay lmL 1L SmL5.Lrn SnLTfl SaLTSI SLTSI i••nLTSJFigure 5.1: Treatment scheme for the recovery of frozen storedbacteria for the use as stock culture in disinfection experiment, yielding 100 mL of stock culture.V. DISINFECTION EXPERIMENT, Materials and Methods 83previously described. This step was repeated one more time. Thisstarter culture was then transferred into 250 mL Erlenmeyerflasks containing 100 mL TSB and incubated over night at 5°C forthe experiments performed at 0°C and at 22°C for the experimentsperformed at 22°C system temperatures. The 16 h old bacterialcultures were pelleted by means of centrifugation, in two 50 mLsteps at 1048 g for 30 minutes at a temperature of 20°C toobtain 2 pellets (from the 100 mL). Bacterial pellets were thenwashed twice with 50 mL of sterile, ozone demand free phosphatebuffer (pH 6.8) and centrifuged again. The final pellets werere—suspended in 25 mL of ozone demand free phosphate bufferwhich resulted in the inoculation medium to be added to theozonated phosphate buffer.3. Generation and measurement of ozoneA Sander, model 301, ozonizer (Sander, Uetze—Eltze, FRG)was used to prepare ozone from pure, research grade oxygencontaining less than 3 ppm of water (Medigas, Vancouver, BC)The ozone was distributed with teflon or tygon tubing and passedthrough a 0.45 pm hydrophillic, polyvinylidene membrane filter(Millipore, Bedford, MA) before contacting the buffer via apasteur pipette. Establishment of the desired ozoneconcentration was accomplished by bubbling the ozone in 2.0 L ofsterile, demand free phosphate buffer (22°C and 0°C) for 25V. DISINFECTION EXPERIMENT, Materials and Methods 84minutes at 10 g ozone/rn3 concentration in the gas stream and 6L/h flow rate to produce a residual ozone concentration of about2.5 ppm. After about 2 h of ozone decay the residual ozoneconcentration was close to the desired 1 ppm (Figure 5.2). Theozone concentration was measured using the indigo method(Grunwell et al., 1983, Bader and Hoigné 1982). This method wasdescribed by Gordon et al. (1988) as “the ideal method for ozonedetermination” since “all other potential interferences,decomposition products and samples originating from othersources are taken into account.” Another advantage of the indigomethod is the possibility of automating sample taking throughmeans of flow injection (Gordon et al. 1988).4. Prearation of alasswareAll glassware was washed with Palmolive laboratorydetergent (Colgate-Palmolive, Toronto, ON) and rinsed thoroughlywith tap-water and then distilled water. Volumetric glasswarewas filled with distilled, residual ozone (>0.5 ppm) containingwater for at least 1 h. All other glassware was soaked indistilled, residual ozone (>0.5 ppm) containing water for atleast 1 h. This procedure was followed a dry heat sterilizationfor at least 4 h at 180°C to oxidize any remaining organicmatter. According to Domingue et al. (1988), this procedureensured that no organic residual was present on the glasswareV. DISINFECTION EXPERIMENT, Materials and Methods 85which would otherwise have interfered with the determination ofozone concentration.5. PreDaration of demand free DhosDhate bufferA 0.01 M phosphate buffer was prepared using KH2PO4 andNa2HPO4 and set to a pH of 6.8 using 0.1 M HCL or NaOH.Distilled water was treated with ozone (>0.5 ppm) in 6-litrevolumes for 50 minutes to oxidize any substances that may havebeen present. This ozone-treated water was then boiled for 1 hto dissipate any residual ozone. Water treated in this mannerwhich was used to prepare the phosphate buffer, was renderedozone demand free. The buffer was finally sterilized beforeusage through a 0.22 pm cellulose ester membrane filter(Millipore Corp., Bedford, MA).V. DISINFECTION EXPERIMENT; Results and Discussion 86C. RESULTS AND DISCUSSIONFor the first part of the experiment the natural decay ofthe ozone in the phosphate buffer, at 0, 10 and 22°C, in thesystem without microorganisms, was determined (Figure 5.2)This step was necessary to verify that little ozone had“disappeared” due to natural decay during the time frame of theactual experiment (the time span after inoculation of the systemuntil the point where microbial counts did not change over timeor the residual ozone completely disappeared) . For this purpose,the system was initially set to a higher than desired ozoneconcentration level (usually around 3.0 ppm). After applyingozone to the buffer the ozone concentration in the reactionvessels decreased constantly (Figure 5.2) until it reached thedesired level (0.76-1.05 mg/L). Natural decay at the threeexperimental temperatures was estimated quite accurately byusing exponential models (note the high r2 values for the linearregressions of the log of the ozone concentration values, Figure5.2)The system was inoculated with the 25 mL of stock culturewhen the natural decay was very small, at the desired ozoneconcentration. Trial tests were performed to find out thedesired ozone concentration at which the bacteria should beadded to the system (see Appendix B). For this purpose the sameV. DISINFECTION EXPERIMENT; Results and Discussion 87time (mm.)Figure 5.2: Standard curves of natural decay of ozone in demandfree 0.01 M phosphate buffer system at 0, 10 and22°C.-0.1‘—-0.30)0-0.5-0.7V yOdegreeCrY’zk : •lOdegreeCt%’\ 22dêrEC‘\V\.‘. Y = -0.00314X + 0.356;, \ ‘\ rsquare=O.955\y.Y = -0.0157 + O.O5; V -O.00411X + 0.055;! r square = 0.963 \ r square = 0.993I0 50 100 150 200 250 300V. DISINFECTION EXPERIMENT; Results and Discussion 88amount of microorganisms (about 1*107 cfu/mL) of each singlestrain were added to the system at relatively high ozoneconcentrations (starting at around 2 ppm) . At these high ozonelevels no microorganisms could be detected in the system withthe method employed (drop-plate method), indicating an immediatetotal kill of all the introduced bacteria. The microorganismswere then added to increasingly lower ozone concentrations untila level at which surviving organisms could be detected. Thisozone concentration was then used for subsequent disinfectiontests for the individual strain of microorganism.After introduction of the microorganisms, ozonedisappeared rapidly in all trials. Time periods necessary forcomplete ozone depletion were between 2 and 10 minutes in allcases. The ability to take sequential samples for ozoneconcentration measurement was a limiting factor since the timeneeded to take a sample and measure its ozone concentration wasabout 2 minutes. Due to the sensitivity of the assay, sampleshad to be measured immediately since fading of the indigoreagent would have resulted in less accurate results. It isspeculated that ozone disappeared sooner in some trials than wasactually measured. For this reason no actual Ct values werecalculated since no definite contact times could be determined.However the calculations of the “sensitivity number” which wasobtained by dividing the decrease in bacterial population (inV. DISINFECTION EXPERIMENT; Results and Discussion 89log cycles) by the initial ozone concentration (equal to theamount of ozone used up by the bacteria) seemed to give a goodindication of the effectiveness of ozone on different types ofmicroorganisms.When lower bacterial counts or higher ozoneconcentrations were employed the microorganisms were almostinstantaneously (too fast to measure with the method employedfor this study) diminished to levels below the detection limit.Reduction of microorganisms showed similar patterns for allisolates (Figures 5.3 to 5.6). The sensitivity ofmicroorganisms, however, varied between individual strains whencalculated as the sensitivity number. PseucYomonas fragi seems tobe the most sensitive microorganism used in this experimentfollowed by Alteromonas punctata subsp. punctata and Alteromonasputrefaci ens with Staphylococcus aureus being the most resistantorganism tested in the present study (Table 5.1). A possibleexplanation for this phenomenon is, that gram—positivemicroorganisms are more resistant to ozone. Staphylococcusaureus was the only gram-positive microorganism tested duringthis part of the study and showed the highest resistance to theaction of ozone. This is possibly due to the thicker cell wallsof gram-positive bacteria as well as their higher glycopeptideand peptidoglycan content (Pelczar et al., 1977), which makes itmore difficult for ozone to oxidize the interior of the cell.V. DISINFECTION EXPERIMEZ’TT; Results and Discussion 90Table 5.1: Results of disinfection experiment to determinesensitivity of single strain microorganisms to ozone,using four different bacterial strains attemperatures of 0°C and 22°C.These findings are supported by earlier research studieswhich also indicated that gram-positive bacteria, such asBacillus species and the Mycobacteri urn sp. are amongst the mostozone resistant of the vegetative bacteria (Haufele andSprockhoff, 1973; Farooq and Akhaque, 1983). Amongst the gram-negative bacteria no large differences have been found in termsof sensitivity to ozone (Bablon et al., 1991).Even small amounts of impurities can cause largedifficulties for the success of this study. When one drop ofethanol (used for sterilization of tubing) came in contact withthe phosphate buffer before ozonation, it took more than 2 h toMicrobial isolate initialbacteriacount(cfu/m L)3.16*107final log cyclebacteria reductioncount(cfu/rn L)4.67*1 oinitial[03]ppmsensitivitylog red./[03]3.552.84 0.76systemtemperatu reoc0AlteromonasputrefaciensStaphylococcus aureus 3.30*107 1.53*104 3.18 1.05 2.86 22Pseudomonas fragi 2.04*107 2.29*1 0 3.98 0.88 4.45 22Alteromonas punctata 1 .13*1 6.30*1 3.48 0.91 3.85 22subsp. punctata(n = 3 for Alterornonasisolates)putrefaci ens and 1 for all otherV. DISINFECTION EXPERIMENT; Results and Discussion 910.8Alteromonas putrefaciens1 OOE+080.7— —— (03)mg/L______1.OOE+07A. Cfu/mL0.6_ ________0.51.OOE+06::0.21.OOE+040.10 I I 1.OOE+030 10 20 30 40 50 60time(min.)Figure 5.3: Ozone concentration changes and reduction ofbacterial population after introduction of singlestrain microorganisms (Alteromonas putrefaci ens)into ozone residual containing demand free phosphatebuffer (0.01 M).V. DISINFECTION EXPERIMENT; Results and Discussion 92Staphylococcus aureus1.2 • 1.OOEi-081 1.OOE+07- -- (03)mg/L£ Cfu/mL0.8______________1.OOE+06-I0)E 0.6 1.OOE+050.4 1.00E+040.2 : 1.OOE+03Ba‘BBi B0 I I I I I 1.OOE+020 10 20 30 40 50 60 70time (mm.)Figure 5.4: Ozone concentration changes and reduction ofbacterial population after introduction of singlestrain microorganisms (Staphylococcus aureus) intoozone residual containing demand free phosphatebuffer (0.01 M).V. DISINFECTION EXPERIMENT; Results and Discussion 93Pseudomonas fragi1 1.OOE+080.91.OOE+070.8— —— (03)mg/L0.7 “ Cfu/mL______________1.OOE+060.6-I0) IE 0.5 ‘ 1.OOE+050.4::0 I I I 1.OOE.i-020 10 20 30 40 50 60 70time (mm.)Figure 5.5: Ozone concentration changes and reduction ofbacterial population after introduction of singlestrain microorganisms (PseucYomonas Fragi) into ozoneresidual containing demand free phosphate buffer(0.01 M).V. DISINFECTION EXPERIMENT; Results and Discussion 94Alteromonas punctata subsp. punctata1 • 1.OOE+080.91.OOE+070.8- -- (03)mg/L0.7 L Cfu/mL______________1.OOE+06E 0.5 ‘ 1 .OOE+05 EI 3I0.4 :0 I . I 1.OOE+020 10 20 30 40 50 60 70time (mm.)Figure 5.6: Ozone concentration changes and reduction ofbacterial population after introduction of singlestrain microorganisms (Alterornonas punctata subsp.punctata) into ozone residual containing demand freephosphate buffer (0.01 M).V. DISINFECTION EXPERIMENT; Conclusion 95bring the ozone concentration in the solution up to about 1 ppm.This incident showed the difficulty in evaluating bacterialresistance to ozone when different kinds of contaminants arepresent (such as the high organic load during fish storage).However the results from the present study seem to provide agood indication of how individual bacterial strains areinactivated by ozone.E. CONCLUSIONThis disinfection model, as applied during the course ofthis study, seems to be an adequate system to measure thesensitivity of different strains of microorganisms to ozone.Various isolates varied in their resistance to the use of ozoneas a sanitizing agent. The gram-positive microorganism testeddid withstand exposure to ozone for a longer period of time thangram-negative organisms indicating that gram-positivemicroorganisms may be more resistant to ozone treatments.However more tests using other gram—positive and gram—negativebacteria are necessary to support this statement.The experimental setup described in the present study wasable to produce useful data for the determination of thesensitivity of single strain microorganisms to ozone. CombinedV. DISINFECTION EXPERIMENT; Conclusion 96with other techniques it should also be possible to determinesensitivities of microorganisms to ozone under variousenvironmental conditions, such as microorganisms attached tosurfaces, as long as the basic conditions are provided(especially an environment free of organic matter which mayinfluence the action of ozone on the microorganisms) . Thisexperimental setup proved to be especially successful since iteliminated the need to know exact contact times of bacteria withthe disinfectant.VI. FINAL CONCLUSION AND RECOMMENDATIONS 97VI • FINAL CONCLUSION AND RECONDATIONSSince ozone seems to facilitate the removal of the slimelayer from the fish, it may be useful to do a complete change ofthe chilled water after a couple of days. This would removeorganic materials (slime, blood, faeces etc.) from the system,as well as microorganisms adherent to these particles. Howeverlimitations of ice storage (larger quantities would be necessaryfor this purpose) exist on the fishing boats so that completewater changes would not be feasible. An alternative to waterchanges may be the installation of foam skimmers whichphysically remove foam from a system. The installation of foamskimmers would be advantageous since they are relatively cheapand easy to install. Removal of foam and other organic materialsfrom the system would be beneficial since this would lead to:a) physical removal of microorganisms (lower spoilagepotential of system)b) increased antimicrobial action of ozonec) increased efficiency of ozone transfer into thechilled water resulting in less ozone losses andhigher concentrations of ozone in the chilled waterd) increased oxidation of spoilage associated compounds(such as TMA)e) cleaner visual appearance of the chilled water.This research also suggested that there are some possibleVI. FINAL CONCLUSION AJ\ID RECOfrII’fENDATIONS 98problems associated with ozone treatments of food systems.Especially the difference in sensitivity of variousmicroorganisms could lead to consumer health risks. Ozone ismore active against gram-negative bacteria such as the typicalfish spoilers from the Pseudomonas family. Ozonation coulddecimate the natural spoilage flora which are in competition toother (eg. pathogenic bacteria). This may give rise to a changein the spoilage flora towards more ozone resistant gram—positivebacteria which in turn would possibly produce less of the“classic signs of fish spoilage: off—odours and off-flavours.Consumers would therefore be less likely to identify unsafefoods since the typical signs of spoilage would be missing. Allthese issues indicate the further need for detailed studies onthe use of ozone in food systems.Additional possible uses of ozone in food processinginclude the pre-ozonation of water used for flake ice production(to disinfect the water) at fish processing and food plants aswell as the treatment of the effluent waste of fish processingplants (which are discarded into the sea without any treatmentin the moment)VII. REFERENCES 99VII. REFERENCESAmerican Public Health Association, American Water WorksAssociation, and Water Pollution Control Federation. 1975.Standard methods for the examination of water and wastewater, 14th ed., p.543. Washington, DC.Anselme, C., Suffet, I.H., and J. Mallevialle. 1988. Effects ofozone on tastes and odours. J. Amer. Water Works Assoc.,80:45.Atlas, R.M. [ed.]. 1988. Industrial Microbiology, p.502. InMicrobiology - Fundamentals and Applications. MacMillanPublishing Co., New York, NY.Bablon, G., Bellamy, W.D., Bourbigot, M.-M., Daniel, F.B., Doré,M., Erb, F., Gordon, G., Langlais, B., Laplanche, A.,Legube, B., Martin, G., Masschelein, W.J., Pacey, G.,Reckhow, D.A. and C. Ventresque. 1991. Fundamental aspects,p.11. In B. Langlais, D.A. Reckhow and D.R. Brink [ed.]Ozone in Water Treatment-Application and Engineering. LewisPublishers, Chelsea, MI.Bader, H. and J. Hoigné. 1982. Determination of ozone in waterby the Indigo method; a submitted standard method. Ozone:Sci. Eng., 4:169.Baranovskaya, O.B., Zapol’skij, O.B., Ovrutskaya, I.Y.,Obodovskaya, N.N., Oshenichnaya, E.E. and 0.1. Yushkevich.1979. Use of ozone gas sterilization during storage ofpotatoes and vegetables. Kohservnaya ± Ovoshchesushhil’nayaProm., 4:10.Blogoslawski, W.J. 1982. Ozonized ice as a preservative.Ozonews, 10:9.Blogoslawski, W.J., Brown, C., Rhodes, E.W. and M. Broadhurst.1975. Ozone disinfection of a seawater system, p.674. In Proc.First Intl. Symp. on Ozone Technol. International OzoneAssociation, Norwalk, CT.Boyce, D.S., Sproul, O.J., and Buck, C.E. 1981. The effect ofbentonite clay on ozone disinfection of bacteria andviruses in water. Water Res., 15:759.VII. REFERENCES 100Broadwater W.T., Hoehn, R.C., and King, P.H. 1973. Sensitivityof three selected bacterial species to ozone. Appi.Microbiol., 26:391.Bull, R.J. 1985. Carcinogenic and mutagenic properties ofchemicals in drinking water, Sd. Total Environ., 47:385.Carmichael, N.G., Winder, C., Borges, S.H., Backhouse, B.L. andP.D. Lewis. 1982. The health implications of watertreatment with ozone. Life Sci., 30:117.Caulfield, M.J., Burleson, G.R. and M. Pollard. 1979. Ozonationof mutagenic and carcinogenic alkylating agents,pesticides, aflatoxin B, and benzidine in water. CancerRes., 39(6, pt.1):2155.Chen, H., Huang, S., Moody, M., and Jiang, S. 1992. Bactericidaland mutagenic effects of ozone on shrimp (Penaeus monoc7on)meat. J. Food. Sci., 57:923.Cognet, L., Courtois, Y. and J. Mallevialle. 1986. Mutagenicactivity of disinfection by-products. Environ. Health Persp.,69 :165.Christensen, E. and A.C. Giese. 1954. Changes in absorptionspectra of nucleic acids and their derivatives followingexposure to ozone and ultraviolet radiations. Arch.Biochem. Biophys., 51:208.Croué, J.P. 1987. Contribution a l’etude de l’oxidation par lechlore et l’ozone d’acides fulviques naturels extraitsd’eaux de surface. Thesis No. 89, University of Poitiers,Poitiers, France, l2Sp.DeWitt III, B.J., McCord, V., Holt, B.L., Ellis, D.K., Finne, G.and Nickelson, II, R. 1984. The potential use of ozonatedice for onboard storage of Gulf of Mexico shrimp. NinthAnn. Trop. Subtrop. Fish Technol. Conf. Am., Tex. A&M Univ.Sea Grant Rep. p.269.Domingue, E.L., Tyndall, R.L., Mayberry, W.R. and O.C. Pancorbo.1988. Effects of three oxidizing biocides on L,egionellapneumophila Serogroup 1. Appi. Environ. Microbiol., 54:741.Emerson, K., Russo, R.C., Lund, R.E. and R.V. Thurston. 1975.Aqueous ammonia equilibrium calculations: effect of pH andtemperature. J. Fish. Res. Board Can., 32:2379.VII. REFERENCES 101Englebrecht, D.R. and E.S.K. Chian. 1985. Comparativeinactivation of six enteroviruses by ozone, p.233. In Ozone inWater Treatment - Applications, Operations, and Technology.Amer. Water Works Assoc., Denver, CO.Farber, J.M. 1991. Symposium on microbiology update: Old friendsand new enemies: Listeria monocytogenes. J. Assoc. Of f.Anal. Chem., 74:701.Faroog, S. and S. Akhaque. 1983. Comparative response ofbacteria and virus to ozonation. Water Res., 17:809.Fetner, R.H., and Ingols, R.S. 1956. A comparison of thebactericidal activity of ozone and chlorine againstEscherichia coli at 1°C. J. Gen. Microbiol., 15:381.Garey, J.F. and R. DeHart. 1979. A comparison of effectivenessof ozone and chlorine in controlling biofouling withincondensers using freshwater as a coolant. Ozone: Sci. Eng.,1:201.Gibson, C.A., Elliott, J.A. and D.C. Beckett. 1960. Ozone forcontrolling mold on cheddar cheese. Can. Dairy & Ice creamJ., 39:24.Gibson, C.I., Tone, F.C., Wilkinson, P. and J.W. Blaylock. 1979.Toxicity and effects of bromoform on five marine species.Ozone: Sci. Eng., 1:47.Giese, A.C. and E. Christensen. 1954. Effects of ozone onorganisms. Physiol. Zool., 21:101.Glaze, W.H. 1988. Summary: Workshop on by-products of ozonation.Amer. Water Works Assoc. Research foundation, Denver, CO.Glaze, W.H., Schep, R., Chauncey, W., Ruth, E.C., Zarnoch, J.J.,Aieta, E.M., Tate, C.H. and M.J. McGuire. 1990. Evaluatingoxidants for the removal of model taste and odour compoundsfrom a municipal water supply. J. Amer. Water Works Asoc.,87 :79.Goldstein, B.D. and E.M. McDonagh. 1975. Effect of ozone on cellmembrane protein fluorescence I. In vitro studies utilizingthe red cell membrane. Environ. Res., 9:179.Gomella, C. 1972. Ozone practices in France. J. Amer. WaterWorks Assoc., 64:39.VII. REFERENCES 102Gordon, G., Pacey, G.E., Cooper, W.J. and R.G. Rice. 1988. Thechemical reactions of ozone and their role in developingimproved analytical methods. Ozone: Sci. Eng., 10:89.Gram, L. 1992. Spoilage of three Senegalese fish species storedin ice and at ambient temperatures, p.225. In Bligh,E.G., [ed.] Seafood Science and Technology, (Proceedings ofthe International Conference Seafood 2000, celebratingthe tenth anniversary of the Canadian Institute ofFisheries Technology of the Technical University of NovaScotia, 13-16 May, 1990, Halifax, Canada). Fishing NewsBooks Ltd., Don Mills, ON.Greer, G.G. and S.D.M. Jones. 1989. Effects of ozone on beefcarcass shrinkage, muscle quality and bacterial spoilage.Can. Inst. Food Sci. Technol. J., 22:156.Grunwell, J., Benga, J, Cohen, H. and G. Gordon. 1983. Adetailed comparison of analytical methods for residualozone measurement. Ozone: Sci. Eng., 5:203.Guyer, S. and T. Jernmi. 1991. Behaviour of Listeriarnonocytogenes during fabrication and storage ofexperimentally contaminated smoked salmon. Appl. Environ.Microbiol., 57 :1523.Haag, W.R. and J. Hoigné. 1983. Ozonation of bromide-containingwaters: Kinetics of formation of hypobromous acid andbromate. Environ. Sci. Techol., 17:261.Haag, W.R. and J. Hoigné. 1984. Kinetics of the reactions ofozone with various forms of chlorine and bromine in water.Ozone: Sci. Eng., 6:103.Haag, W.R., Hoigné, J. and H. Bader. 1984. Improved ammoniaoxidation by ozone in the presence of bromide ion duringwater treatment. Water. Res., 18:1125.Haraguchi, T., Slimidu, U., and K. Aiso. 1969. Preserving effectof ozone on fish. Bull. Japan. Soc. Sci. Fish., 35:915.Haufele, A. and H.V. Sprockhoff. 1973. Ozon alsDesinfektionsmittel gegen vegetative Bakterien,Bazillensporen, Pilze und Viren in Wasser. ParasitenkdeInfekt. Krankh. Hyg., 175:53.VII. REFERENCES 103Hebard, C.E., Flick, G.J. and R.E. Martin. 1982. Occurrence andsignificance of trimethylamine oxide and its derivatives infish and shellfish, p.149. In R.E. Martin, G.J. Flick, C.E.Hebard and D.R. Ward fed.] Chemistry and Biochemistry ofMarine Food Products. AVi Publishing Company, Westport, CT.Herren—Freud, S.L., Pereira, M.A., Khoury, M.D. and G. Olson.1987. The carcinogenicity of trichloroethylene and itsmetabolites, trichloroacetic acid and dichioroacetic acidin the mouse liver. Toxicol. Appi. Pharmacol., 90:183.Hoff, J.C., 1978. The relationship of turbidity to disinfectionof potable water, p.235. In Hendricks, C.W. fed.]Evaluation of the Microbiology Standards for DrinkingWater. EPA-570/9-78-006. U.S. Environmental ProtectionAgency. Office of Drinking Water, Washington, D.C.Hoigné, J. 1988. The chemistry of ozone in water, p.121. InStucki,S. fed.] Process Technologies for Water Treatment.Plenum Press, New York, NY.Hoigné, J. and H. Bader. 1983. Rate constants of reactions ofozone with organic and inorganic compounds in water. 2.Dissociating Organic Compounds. Water. Res., 17:185.Hunt, T.S. 1848. On the anomalies presented in the atomic volumeof sulfur and nitrogen. Amer. Jour. Sci., 6:717.Ikeda, S. 1979. Other organic components and inorganiccomponents, p.226. In J.J. Connell fed.] Advances in FishScience and Technology, (Jubilee Conf. Torry Res. Station,Aberdeen, Scotland, July 1979). Fishing News Books Ltd.,Farnham, Surrey, UK.Johnson, J.D., Christman, R.F., Norwood, D.L. and D.S.Millington. 1982. Reaction products of aquatic humicsubstances with chlorine. Environ. Health Persp., 46:63.Kaess, G. and J.F. Wiedemann. 1968. Ozone treatment of chilledbeef: I. Effect of low concentrations of ozone on microbialspoilage and surface colour of beef. J. Food Technol.,3:325.Katzenelson, E., Kietter, B. andH.I. Shuval. 1974. Inactivationof viruses and bacteria by use of ozone. J. Amer. Water WorksAssoc., 66:725.VII. REFERENCES 104Kötters, J. 1993. Versuche mit Ozon zur Disinfektion vonFischen und verschiedenen Zwischenprodukten in derFischverarbeitung. Diplomarbeit, 84p., ChristianAlbrechts University, Kiel, FRG.Kolodyaznaya, V.S. and T.A. Suponina. 1975. Storage of foodusing ozone. Kholodil’naya Tekhnika, 6:39.Korich, D.G., Mead, J.R., Madore, M.S., Sinclair, N.A. and C.R.Sterling. 1990. Effects of ozone, chlorine dioxide,chlorine, and monochloramine on Cryptosporidium parvumoocyst viability. Appl. Environ. Microbiol., 56:1423.Kurz, M.E. and W.A. Pryor. 1978. Radical production from theinteraction of closed shell molecules. 9. Reaction of ozonewith tert-butyl hydroperoxide. J. Am. Chem. Soc., 100:7953.Laplanche, A. and G. Martin. 1982. Action of ozone on variousorganic compounds, p.73. In W.J. Masschelein [ed.]Ozonation Manual for Water and Wastewater Treatment. JohnWiley & Sons, New York, NY.Laycock, R.A. and L.W. Regier. 1971. Trimethylamine producingbacteria on haddock (Melanogrammus aeglefinus) filletsduring refrigerated storage. J. Fish. Res. Bd. Can. 28:305.LeBlanc, R.J. and E.L. LeBlanc. 1992. The effect ofsuperchilling with CO2 snow on the quality of commerciallyprocessed haddock (Melanogrammus aeglefinus) fillets,p.247. In G.E. Bligh [ed.] Seafood Science and Technology,(Proceedings of the International Conference Seafood 2000,celebrating the tenth anniversary of the CanadianInstitute of Fisheries Technology of the TechnicalUniversity of Nova Scotia, 13-16 May, 1990, Halifax,Canada). Fishing News Books Ltd., Don Mills, ON.Lee, J.S., and Kramer, D.E. 1984. Effectiveness of ozone-treatedwash water and ice on the keeping quality and stability ofSockeye salmon. Report FOTC 84/T-l, Fishery IndustrialTechnology Centre, University of Alaska, Kodiak, AK.Legube, B., Croué, J.P., De Laat, J. and M. Doré. 1989.Ozonation of an extracted aquatic fulvic acid: Theoreticaland practical aspects. Ozone: Sd. Eng., 11:69.Merlet, N., De Laat, J., Brunet, R. and M. Doré. 1980. Role ofozone in trihalomethane formation. Environ. Technol.Lett., 1:384.VII. REFERENCES 105Mudd, J.B., Leavitt, R., Ongun, A. and T.T. McManus. 1969.Reaction of ozone with amino acids and proteins. Atmos.Environ., 3:669.Murray, R.G.E., Steed, P. and H.E. Elson. 1965. The location ofthe mucopeptide of sections of the cell wall of Escherichiacoli and other gram-negative bacteria. Can. J. Microbiol.,11: 547.Nangia, P.S. and S.W. Benson. 1980. Thermochemistry and kineticsof ozonation reactions. J. Am. Chem. Soc., 102:3105.National Cancer Institute Carcinogenesis Program, Division ofCancer Cause and Prevention, Bethesda, MD, March 1, 1976,Report on carcinogenesis bioassay of chloroform, 60p.Nelson, W. 1982. The use of ozonized ice to extend the shelf-life of fresh Alaskan fish, Rep. Subm. Alaska Dep. Commer.Econ. Dev., Of f. Commer. Fish. Dev., Anchorage, AK.Norton, J . S., Charig, A. J. and I.E. Denoranville. 1968. Theeffect of ozone on storage of cranberries. Proc. Am. Soc.Horticult. Sci., 93:792.Pavoni, J.L., Tittlebaum, M.E., Spencer, H.T., Fleischinan, M.,Nebel, C., and Gottschling, R. 1972. Virus removal fromwastewater using ozone. Water Sew. Works, 119:59.Payne, W.J., Riley, P.S. and C.D. Cox. 1971. Separate nitrite,nitric oxide and nitrous oxide reducing fractions fromPseudomonas perfectomarinus. J. Bacteriol., 106:356.Pelczar, M. J., Reid, R.D., and E . C.S. Chan [ed. 1. 1977. Themorphology and fine structure of bacteria, p.70. InMicrobiology, McGraw Hill Book Company, NY.Prat, R., Nofre, C. and A. Cler. 1968. Effets de’l hypochloridede sodium, de l’ozone et des radiations ionisantessur lesconstituants pyrimidiques d’Escherichia coli. Ann. Inst.Pasteur, 114:595.Prengle, H.W. and C.E. Mauk. 1978. Ozone methods and ozonechemistry of selected organics in water, p.302. In R.G.Rice and J.A. Cotruvo [ed.] Basic Chemistry inOzone/Chlorine Dioxide Oxidation Products of OrganicMaterials. bA, Cleveland, OH.VII. REFERENCES 106Pryor, W.A., Dooley, M.M. and Daniel F. Church. 1982. Mechanismsfor the reaction of ozone with biological molecules: Thesource of the toxic effects of ozone, p.7. In S.D. Lee,M.G. Mustafa and M.A. Mehlman [ed.] Advances in ModernEnvironmental Toxicology, Vol.V., Princeton ScientificPublishers Inc., Princeton, NJ.Ravesi, E.M., Licciardello, J.J., and Racicot, L.D. 1987. Ozonetreatment of fresh Atlantic cod Gadus morhua. Mar. Fish.Rev., 49:37.Regenstein, J.M., Schlosser, M.A., Samson, A., and Fey, M. 1982.Chemical changes of trimethylamine oxide during fresh andfrozen storage of fish, p.137. In Martin, R.E., G.J. Flick,C.E. Hebard and D.R. Ward led.] Chemistry and Biochemistry ofMarine Food Products. AVi Publishing Company, Westport, CT.Reid, R.A. and T.D. Durance. 1992. Textural changes of cannedchum salmon related to sexual maturity. J. Food Sci.,57:1340.Reynolds, G., Corless, C., Graham, N., Perry, R., Gibson, T.M.and J. Haley. 1989. Aqueous ozonation of fatty acids.Ozone: Sci. Eng., 11:143.Rice, R.G., Farquhar, J.W. and L.J. Bollyky. 1982. Review of theapplications of ozone for increasing storage times ofperishable foods. Ozone: Sci. Eng., 4:147.Rice, R.G., Otwell, W.S., Blake, N., Sweat, D.E., Marshalk,R.L., and Farquhar, J.W. 1985. Ozonized water and ice topreserve fresh fish, Grouper (Mycroteroperca bonaci),p.187. In Storage Lives of Chilled and Frozen Fish andFish Products, mt. Inst. Refrig. Comm. C2 & D3.Salmon, J. and J. LeGall, 1936. Application of ozone to maintainthe freshness and to prolong the durability ofconservation of fish. Rev. Gen. du Froid, 11:317.Salmon, J., LeGall, J. and A. Salmon. 1937. Purification ofedible mollusks by ozonized seawater. Ann. Hyg. Publ. md.Social, 15:44.Sander, E. and H. Rosenthal. 1975. Application of ozone in watertreatment for home aquaria, public aquaria and foraquaculture purposes.p.103-114. In W.J. Blogoslawski and R.G.Rice [ed.] mt. Ozone Inst. Workshop Series: AquaticApplications of Ozone. mt. Ozone Inst. , Syracuse, NY.VII. REFERENCES 107Sayato, Y., Nakamuro, K. and H. Ueno. 1989. Mutagenicity onchlorination of products formed by ozonation ofnaphtoresorcinol in water. Mutation Res., 226:151.Schönbein, C.F. 1840. Recherches sur la nature de l’Odeur qui semanifeste dans certaines actions chimiques. Compt. Rend.Dcad. Sci. Paris, 10:706.Scott, D.B.M., and Lesher, E.C. 1963. Effect of ozone onsurvival and permeability of Escherichia coli. J.Bacteriol., 85:567.Sheldon, B.W. and A.L. Brown. 1986. Efficacy of ozone as a.disinfectant for poultry carcasses and chill water. J. FoodSci., 51:305.Skura, B.J. 1993. Personal communication. University of BritishColumbia, Department of Food Science, Vancouver, BC.Sobsey, M.D. 1989. Inactivation of health related microorganismsin water by disinfection processes. Water Sci. Technol.,21:179.Spanggorg, R.J. and V.J. McClurg. 1978. Ozone methods and ozonechemistry of selected organics in water, p.115. In R.G.Rice and J.A. Cotruvo, [ed.] Basic Chemistry, inOzone/Chlorine Dioxide Oxidation Products of OrganicMaterials. IOA, Cleveland, OH.Stockinger, H.E. 1959. Factors modifying toxicity of ozone,p.360. In Ozone Chemistry and Technology. American ChemicalSociety, Washington, D.C.Vrochinskij, K.K. 1963. Experimental data on waterdecontamination with ozone. Hyg. Sanit., 28:3.Waite, T.D. 1984. Introduction to water quality concepts, p.21.In H.I. Shuval [ed.] Principles of Water Quality. AcademicPress, Inc., Orlando, FL.Wickramanayake, G.B., Rubin, A.J. and O.J. Sproul. 1984.Inactivation of Naegleria and Giardia cysts in water byozonation. J. Water Pollut. Control Fed., 56:983.Wickramanayake, G.B. and O.J. Sproul. 1988. Ozone concentrationand temperature effects on disinfection kinetics. Ozone:Sci. Eng., 10:123.VII. REFERENCES 108Wolfe, R.L., Steward, M.H., Scott, K.N. andM.J. McGuire. 1989.Inactivation of Giardia muris and indicator organismsseeded in surface water supplies by PEROXONE and ozone.Environ. Sci. Technol., 23:744.Wong, K. and T.A. Gill. 1987. Enzymatic determination oftrimethylamine and its relationship to fish quality. J.Food Sci., 52:1.Woyewoda, A.D., Shaw, S.J., Ke, P.J., and Burns, B.G. 1986.Recommended Laboratory Methods for Assessment of FishQuality. Canadian Technical Report of Fisheries and AquaticSciences. No.1448. p.35.Yang, P.P.W. and T.C. Chen. 1979. Effects of ozone treatment onmicroflora of poultry meat. J. Food Process. Preserv.,3:177—185.Young, J.C. 1984. Waste strength and water pollution parameters,p.2. In R.A. Minear and L.H. Keith [ed.] Water Analysis, Vol.III — Organic Species. Academic Press Inc., Orlando, FL.Zoeteman, B.C.J., Hrubec, J., de Greef, E. and H.J. Kool. 1982.Mutagenic activity associated with by-products of drinkingwater disinfection by chlorine, chlorine dioxide, ozone andUV-irradiation. Environ. Health Persp., 46:197-205.VIII. APPENDIX A; PICTURES OF FISH TANK EXPERIMENT 109VIII. APPENDIX A: PHOTOGRAPHS OF EXPERIMENTAL SETUP AND RESULTSOF AN EXPERIMENT TO SI!lIt.TLATE THE STORAGE CONDITIONS ONB.C. FISHING BOATS IN A PILOT-PLANT SETUPPicture 1: Experimental setup for experiment to simulate storageconditions of fish on B.C. fishing boats in a pilotplant setup: Ozonated fish storage tank, temperaturerecorder.VIII. APPENDIX A; PICTURES OF FISH TANK EXPERIMENT 110Picture 2: Ozonator and ozone measuring equipment for experimentto simulate storage of fish on B.C. fishing boats ina pilot plant setupVIIL APPENDIX A; PICTURES OF FISH TANK EXPERIMENT 111Picture 3: Fish samples after nine days of storage in chilledwater: ozonated samplesVIII. APPENDIX A; PICTURES OF FISH TANK EXPERIMENT 112Picture 4: Fish samples after nine days of storage in chilledwater: control samplesVIII. APPENDIX B; PRELIMINARY EXPERIMENTS FOR EXPERIMENT V 113APPENDIX B: LISTING OF PRELIMINARY EXPERIMENTS CONDUCTED DURINGTHE DISINFECTION EXPERIMENT TO DETERMINE THE DESIRED OZONECONCENTRATION AT WHICH THE BACTERIA SHOULD BE ADDED TO THESYSTEMbacteria initial final bact. initial final [03]bact. popn. [031 (ppm) (ppm)popn. (cfu/mL)(cfu/mL)Alteromonas 534*105 <50 1.27 0.51Pu trefaci ensAlteromonas 1.56*106 <50 1.42 0.37pu trefaci ensAlteromonas 2.40*106 <50 0.93 0pu trefaci ensAlteromonas 8.10*106 <50 1.33 0.33pu trefaci ensAlteromonas 8.70*106 <50 0.94 0pu trefaci ensPseudomonas 9.60*107 <50 0.98 0.08fi uorescens( ozone concentration after approximately 10 minutes afteraddition of bacteria to system; = detection limit of methodemployed was 50 cfu/mL)note: a. experiments using Staphylococcus aureus and Alteromonaspunctata subsp. punctata as inoculum did not requiretest trials.b. temperature for Alteromonas putrefaciens experimentswas 0°C and 22°C for all other bacterial strains


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items