1、Inactivation of Pseudomonas fluorescens and Streptococcusthermophilus in Trypticase Soy Broth and total bacteria in milk bycontinuous-flow ultrasonic treatment and conventional heatingMar Villamiel, Peter de Jong*Department of Process Innovation, NIZO Food Research, P.O. Box 20, 6710 BA Ede, Netherl
2、andsReceived 19 February 1999; accepted 6 March 2000AbstractA continuous-flow ultrasonic treatment apparatus has been set up in order to apply high-intensity ultrasound for the inactivationof bacteria important for the dairy industry. A comparative study has been carried out using a conventional tub
3、ular heat exchangerunder similar process conditions. Pseudomonas fluorescens and Streptococcus thermophilus, inoculated in Trypticase Soy Broth(TSB), were inactivated by ultrasound, the former (Gram-negative) being less resistant than the latter (Gram-positive). Although theeC128ectiveness of ultras
4、ound can be decreased by temperature increase during the treatment, an additive eC128ect was observed betweenheat and ultrasound. This result was indicated by computer simulations comparing ultrasound with conventional heating. Pre-liminary results obtained in milk demonstrated that continuous-flow
5、ultrasonic treatment could be a promising technique for milkprocessing. This method has several advantages such as homogenisation of milk and less energy consumption than batch sys-tems. 2000 Elsevier Science Ltd. All rights reserved.1. IntroductionDuring the last years, ultrasonic treatment and oth
6、eralternative food technologies have been studied due toconsumers continuing demand for products with im-proved quality and safety.High-intensity ultrasound (101000 W/cm2and 60.1MHz) (Povey McClements, 1995)has been used for some dairy process applications suchas cleaning (KivelC127a, 1996), inactiv
7、ation of bacteria(Ord onez, Aguilera, Garc a Garc a,Burgos, Sanz Wrigley Ord onez et al., 1987; Garc a et al.,1989) and static pressure (Raso, Pag an, Cond on Jong de, 1996).On the other hand, Pseudomonas fluorescens (Gram-negative) is one of the most important psychrotrophicbacteria responsible for
8、 undesirable flavours in milkand dairy products. The heat treatment applied duringconventional pasteurisation (7278 C, 1020 s) is suf-ficient to destroy large numbers of vegetative cells ofJournal of Food Engineering 45 (2000) author.0260-8774/00/$ - see front matter 2000 Elsevier Science Ltd. All
9、rights reserved.PII: S 0 2 6 0 - 8 7 7 4 ( 0 0 ) 0 0 059-5this microorganism (Nickerson 1 where P is the power (W); V the voltage 220 (V); Cthe current (A) and e is the e ciency 1.EvCap q DT; 2 where Evis the energy input per unit of volume (kJ/l);Cap the thermal capacity (water 4.18 kJ/kg/K); q the
10、density of the liquid (kg/l) and DT is the outlet tem-perature ) inlet temperature (K). The energy input canalso be calculatedEvP=u; 3 where P is the power input (kW) and u is the liquid flow(l/s).Two silicone tubes of 0.57 cm inside diameter wereconnected to the insulated ultrasonic cavity through
11、twoFig. 1. Scheme of the continuous-flow ultrasonic treatment system.172 M. Villamiel, P. de Jong / Journal of Food Engineering 45 (2000) 171179holes in the bottom and the left sides of the cavity.According to the cavity and tip dimensions, the totalvolume of the cavity was 18.76 ml. Inlet and outle
12、ttemperatures were continuously monitored using twothermocouples positioned outside the ultrasonic cavityand connected to a PM 8237 Multipoint Data Recorder(Philips). Samples were pumped through the systemusing a previously calibrated 7523-12 Easy-Load Mas-terflex Pump (Cole Parmer, Barnant Company)
13、. Theprocess conditions were achieved by varying the flowrate (1150 ml/min) and the output intensity level (5 and10). Taking into account the dimensions of the ultra-sonic cavity and the diC128erent flow rates (see Table 1), theliquid spent diC128erent theoretical residence times insidethis cavity.
14、Samples leaving the disruptor were cooled ina silicone coil (0:35 cm 200 cm) immersed in an ice/water bath. When the outlet temperatures were main-tained constant, samples of up to 100 ml were collectedin sterile containers. All experiments were performed induplicate.2.4. Conventional treatmentsThe
15、continuous-flow conventional heating systemwas similar to the ultrasonic one, except for the fact thatthe cavity was replaced by a thermostated water bath. Astainless steel tube with an internal volume of 18.76 ml(0.85 cm inside diameter) and a wall thickness of 0.55mm was immersed in the hot water
16、bath. Using the samemean residence time as those applied during the ultra-sonic treatment similar outlet temperatures were alsoachieved by adjusting the temperature of the water bath,as shown in Table 1. Experiments were carried out induplicate and the samples were collected as described forthe ultr
17、asonic treatment.2.5. Storage conditions of the treated milkMilk treated at temperatures around 62 C was col-lected in sterile containers and distributed under asepticconditions in a laminar flow cabinet (Interflow) intosterile 10 ml screw-capped tubes. These were refrigeratedat 5 C for 5 days and a
18、nalysed for total bacterial countand extent of proteolysis.2.6. Microbiological and analytical determinationsThe eC128ect of the treatments on the microorganismsinoculated in TSB was evaluated by determiningP. fluorescens and S. thermophilus counts and, in milk,total bacterial count and alkaline pho
19、sphatase activity.P. fluorescens suspensions were serially diluted in0.1% peptone physiological salt solution (PPS) (Oxoid)and plated on 2.35% plate count agar (Difco) with 1%skim milk (PCMA). The plates were then incubated for72 h at 30 C and the number of CFU/ml counted.S. thermophilus suspensions
20、 were serially diluted in0.1% PPS (Oxoid) and plated on 2.35% plate count agar(Difco) with 5% skim milk (PC5MA). The plates wereincubated for 48 h at 45 C and the number of CFU/mlcounted.The total bacterial counts in milk were determinedusing 0.1% PPS for the dilutions and 2.35% PCMA toplate on. The
21、 plates were incubated for 72 h at 30 Caspreviously described (IDF, 1991).The detection of alkaline phosphatase activity in milkwas performed by the method of IDF (1987).The extent of proteolysis during the storage period at5 C of milk treated at temperatures close to 62 C wasestimated by measuring
22、the increase in soluble nitrogenat pH 4.6 following the method of Koops, Klomp andElgersma (1975).The size of the fat globule in milk treated by thecontinuous-flow ultrasonic system at temperatures closeto 62 C was measured using a laser light-scattering in-strument (Malvern MasterSizer X, Malvern I
23、nstru-ments). Before the measurements were performed, themilk samples were gently shaken and diluted as requiredusing distilled water to eliminate multiple scatteringeC128ects.Table 1Maximum temperatures achieved ( C) during diC128erent conditions used in the continuous-flow ultrasonic and conventio
24、nal systemsFlow (ml/min)/tra(s)Ultrasonic treatment Conventional heatingOutlet temperatureb( C)Output intensity 10 Output intensity 5 Outlet temperatureb( C) Water bath temperature ( C)50/22.5 45.2 34.5 33/34.1 49.0 35.9 49.4 53.020/56.3 62.3 43.5 62.0 64.011/102.3 76.1 52.5 76.5 77.611/102.3 51.7 5
25、2.0atr: residence time inside the ultrasound cavity or inside the heat exchanger (conventional heating).bInlet temperature was 23:5 1:0 C.M. Villamiel, P. de Jong / Journal of Food Engineering 45 (2000) 171179 1732.7. Computer simulationsTo demonstrate the eC128ect of mixing in the ultrasoundcavity
26、on the thermal eC128ect, computer simulations werecarried by using a combined heat transfer and reactionmodel. The model is based on the assumptions that theultrasound cavity can be considered as a continuouslystirred tank reactor (CSTR) and the conventional sys-tem can be considered as a plug flow
27、reactor (PFR).To calculate the lethal eC128ect of both systems thetemperature/time profile should be known. For thecavity the temperature is constantTCSTR;tTCSTR;out:The temperature in the conventional system is given bydTPFRdtUpD;v/CpTwaterTPFR;where U is the overall-heat transfer coe cient (250W/m
28、2/K), D the tube diameter, v the liquid velocity, /the flow rate and Cpis the specific heat of the heatedproduct.From the experiments it turned out that the outlettemperature is correlated to the flow rateTPFR;out898:1 /0:03759:2;where / is the flow rate in ml/min and T is the producttemperature in
29、C. The bacteria concentration P. fluo-rescens in the cavity is calculated byCCSTR;outCCSTR;01 kTCSTR;outt;where t is the residence time of the liquid in the cavity.The outlet bacteria concentration in the tube is de-scribed bydCPFRdtkCPFR:The reaction rate constant k is dependent on the tem-perature
30、k k0exp Ea=RT ;where T is the absolute temperature, R the gas constant,k0the pre-exponential factor (7.6 1028/s) and Eais theactivation energy (184 kJ/mol).3. Results and discussion3.1. Set-up of the equipment: selection of the processingconditionsIn order to determine the equilibrium temperature at
31、several flow rates (from 11 to 174 ml/min), previousexperiments with distilled water were performed at themaximum output intensity level (10). After 20 min oftreatment the system was in equilibrium as indicated byconstant outlet temperatures. Once the steady state wasachieved, the outlet temperature
32、 was maintained withcoe cients of variations less than 1%. Flow rates from11 to 50 ml/min and output intensity values of 10 and 5provided a wide range of outlet temperatures (76.133.1 C).DiC128erent experiments using distilled water and a flowrate of 50 ml/min were performed at a sonicator outputint
33、ensity level of 110. By means of Eqs. (1)(3) (seeSection 2) the power delivered to the sonicator and thepower related to the increase of temperature DT werecalculated (Fig. 2). As shown, the former increased fromoutput intensity level 15 but it remained practicallyunchanged from 5 to 10 (150 W). The
34、 latter increasedwith the output intensity level, probably due to an in-crease in the wave amplitude and, therefore, a higherultrasound eC128ect.As can be observed in Fig. 2, the values of powercalculated from DT were lower (670 W) than thosedelivered to the sonicator (6150 W). This can be at-tribut
35、ed to the fact that part of the total energy trans-mitted to the sample is used in diC128erent mechanismssuch as disruption (cavitation) and turbulence or micro-streaming of the liquid. The rest of the energy is ab-sorbed for the heating of the sample (Floros Nick-erson Ahmed Berliner, 1984). The in
36、activation of bac-teria must progressively approach that produced bythermal treatments with the increase of temperatureuntil a temperature at which the lethality of ultrasoundis negligible.Table 2Inactivation of P. fluorescens during continuous-flow ultrasonic treatment and conventional heating of T
37、SB under diC128erent processing conditionsFlow (ml/min)/tra(s)Ultrasonic treatment Conventional heatingOutput intensity 10 Output intensity 5Toutletb( C)Reduction(log CFU/ml)cToutletb( C)Reduction(log CFU/ml)cToutletb( C) Reduction(log CFU/ml)c50/22.5 43.1 0:6 0:1 33.1 0:2 0:1 33/34.1 49.0 0:8 0:0 3
38、5.9 0:6 0:2 49.4 020/56.3 61.6 3:1 0:1 43.0 0:9 0:1 61.7 3:7 0:111/102.3 76.1 4:2 0:0 51.0 1:1 0:1 76.5 611/102.3 51.7 0:3 0:1atr: residence time inside the ultrasound cavity or inside the heat exchanger (conventional heating).bToutlet: outlet temperature of the sample (inlet temperature was 23:5 1:
39、0 C).cMean values standard deviation (initial bacterial count was 6.97.7 log CFU/ml).Table 3Inactivation of S. thermophilus during continuous-flow ultrasonic treatment and conventional heating of TSB under diC128erent processing conditionsFlow (ml/min)/tra(s)Ultrasonic treatment Conventional heating
40、Output intensity 10 Output intensity 5Toutletb( C) Reduction(log CFU/ml)cToutletb( C)Reduction(log CFU/ml)cToutletb( C)Reduction(log CFU/ml)c50/22.5 45.2 0:1 0:1 34.5 0 20/56.3 62.3 0:2 0:0 43.5 0 62.0 0:1 0:011/102.3 75.9 2:7 0:1 52.5 0 76.5 3:5 0:1atr: residence time inside the ultrasound cavity o
41、r inside the heat exchanger (conventional heating).bToutlet: outlet temperature of the sample (inlet temperature was 23:5 1:0 C).cMean values standard deviation (initial bacterial count was 6.87.3 log CFU/ml).M. Villamiel, P. de Jong / Journal of Food Engineering 45 (2000) 171179 175Other authors ha
42、ve previously reported this com-bined eC128ect of ultrasound and heat in the inactivation ofsome microorganisms subjected to batch ultrasonictreatments and heating at the same time (Ord onez et al.,1987; Garc a et al., 1989; Ciccolini, Taillandier, Wilhem,Delmas Sala, Burgos, Con-d on, Lop ez lnk066
43、:5) processing in the tubular heatexchanger and in the ultrasound cavity (temperature is determined bythe flow rate (see Table 2, output intensity 10) between 11 and 50 ml/min).176 M. Villamiel, P. de Jong / Journal of Food Engineering 45 (2000) 171179Gri ths, Phillips & Muir, 1986). This process ca
44、n bealso applied for extension of the safe storage life of milkto be used for powder manufacture (West, Gri ths,Phillips, Sweetsur & Muir, 1986).The shelf life of milk processed at temperatures of62 C by both systems was evaluated and the results areshown in Fig. 4(a) and (b). During the storage per
45、iod,the increase of bacterial counts was similar in bothtreated milks, as shown in Fig. 4(a). The number ofbacteria remained almost constant at 4 log CFU/mluntil the third day of storage and subsequently increasedto reach values of 5 log CFU/ml after 5 days at 5 C. Theextent of proteolysis, estimate
46、d as the increase in solublenitrogen at pH 4.6 (Fig. 4(b) remained unchangedduring the storage period in both milks studied. There-fore, no evidence of proteolytic activity was found underthe assayed conditions and continuous-flow ultrasonictreatment compares favourably with conventional heat-ing. H
47、su and Shipe (1986) observed an increase of pro-teolysis in whole milk subjected to long ultrasonic batchtreatment (3 h at 6.7 C) with respect to the untreatedmilk sample. They suggested that the increase might beattributed to dissociation of native milk proteinases frommilk fat globule membranes or
48、 casein micelles.The energy consumption is one of the most importantaspects to make new processes attractive for the indus-try. There are few data in the literature about theamount of energy needed for bacterial inactivation byultrasound (Villamiel, van Hamersveld & Jong de,1999). In a review about
49、the eC128ect of ultrasound pro-cessing on the quality of dairy products, these authorsnoted a high energy consumption (6751575 kJ/l), cal-culated from the data of Ord onez et al. (1987), for theinactivation of S. aureus by means of batch ultrasonictreatment.Using the experimental conditions in this work, theenergy input during continuous-flow treatment of milkwas 105220 kJ/l and 270830 kJ/l for conventionalheating and ultrasonic treatment, respectively (estimatedusing Eqs. (2) and (3), see Sec
