Innovative Food Science and Emerging Technologiesstatic.tongtianta.site/paper_pdf/fb885612-61d5-11e9-a14c... · 2019-04-18 · E. Bhuvaneswari, C. Anandharamakrishnan⁎ Food Engineering

Innovative Food Science and Emerging Technologies xxx (2014) xxx–xxx

INNFOO-01159; No of Pages 8

Contents lists available at ScienceDirect

Innovative Food Science and Emerging Technologies

j ourna l homepage: www.e lsev ie r .com/ locate / i fset

Heat transfer analysis of pasteurization of bottled beer in a tunnel pasteurizer usingcomputational fluid dynamics

E. Bhuvaneswari, C. Anandharamakrishnan ⁎Food Engineering Department, CSIR—Council of Scientific and Industrial Research, Mysore 570 020, India

⁎ Corresponding author. Tel.: +91 821 2513910; fax: +E-mail address: [email protected] (C. Anandhara

http://dx.doi.org/10.1016/j.ifset.2014.03.0041466-8564/© 2014 Elsevier Ltd. All rights reserved.

Please cite this article as: Bhuvaneswari, E., & Ausing computational fluid dynamics, Innovativ

a b s t r a c t
a r t i c l e i n f o
Article history:Received 30 June 2011Accepted 12 March 2014Available online xxxx

Editor Proof Receive Date 10 April 2014

Keywords:CFDBottled beerTunnel pasteurizerPasteurization units

Beer is one of the most widely consumed alcoholic beverages in the world. Pasteurization is an important unitoperation in beer processing that inactivates the spoilage microorganisms present in beer thereby extendingits shelf life. It is difficult to determine the temperature profile and slowest heating zone (i.e. minimum heatingregion) inside the bottle during industrial scale tunnel pasteurization. Computational fluid dynamics (CFD)modeling can be used as a tool to determine the temperature distribution pattern inside the bottled beer. Thisstudy is unique in terms of using the actual thermophysical properties of beer unlike in earlier publishedworks. Further, CFD simulation prediction of the temperature profile in the bottled beer was validated with ex-perimental measurements. The study was extended to investigate the effects of different zones' temperatureson inactivation of Saccharomyces cerevisiae, the most common beer spoilage organism. A tunnel pasteurizerwith seven zones was selected to study the temperature profile inside the bottled beer and found to be ingood agreement with the published temperature profiles in different zones. In addition, the effectiveness ofpasteurization was investigated in terms of pasteurization unit (PU) and the resultant PU value (15 to 30 PU)was adequate for achieving the maximum sterility of beer.Industrial relevance: In recent years, a rapid development in the application of CFD in food processing operationshas been witnessed. The main need for CFD analysis of pasteurization is to determine the uniform and effectiveheat distribution inside the bottled beer and to examine the position of slowest heating zone (SHZ). Relativelyfew works have been published related to applications of CFD during beer pasteurization. However, all the stud-ies were performedwithwater as amodel fluid. So far nowork has been published on the CFD simulation duringthe pasteurization process in bottled beer by using the thermophysical properties of beer. Hence, the presentstudy was aimed at investigating the temperature distribution inside the beer bottle during the pasteurizationprocess and in each zone of the industrial tunnel pasteurizer. This model can be used for the brewery industryto ensure that the required pasteurization temperature has been reached inside the beer bottle when conveyedthrough a tunnel pasteurizer. This is very essential to render a safe product with extended shelf life to theconsumers.

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Fermented beverages and foods have a unique place in most socie-ties because of their economical and cultural importance (Legras,Merdinoglu, Cornuet, & Karst, 2007). Beer is one such most consumedfermented beverage in the world and it continues to be a populardrink. The shelf-life of beer has become one of the most critical issuesas it is produced and marketed both nationally and internationally(Takashio & Shinots, 1998). The shelf-life of beer is mostly determinedby its microbiological, colloidal, foam, colour and flavour stabilities(Vanderhaegen, Neven, Verachtert, &Derdelinck, 2006). Heat treatmentis the widely applied and effective method for extending shelf-life ofbeverages packed in containers such as glass bottles (Horn, Franke,

91 821 2517233.makrishnan).

nandharamakrishnan, C., Heate Food Science and Emerging Te

Blakemore, & Stannek, 1997). The stability of the final product throughseveral months can be achieved if common contaminants of beer suchas Pediococcus spp., Lactobacillus spp., and wild yeasts like Saccharomycesspp.,Hansenula,Dekkera, Brettanomyces, Candida and Pichia are inactivated(Priest & Stewart, 2006). These organisms can be inactivated at a time–temperature combination of around 60 °C for 20 min (Briggs, Boulton,Brookes, & Stevens, 2004). Currently, the heat treatment of beer is eitherperformed by flash pasteurization (first pasteurized and then packagedaseptically) or by tunnel pasteurization (first filled into container (bottlesor cans) and then pasteurized through tunnel pasteurizers) or batchpasteurization (Buzrul, 2007). Tunnel pasteurization is an important unitoperation in the brewery packaging line. It deserves most attention interms of energy cost due to high steam consumption. A tunnel pasteurizercomprises a large metal-cased enclosure, divided into a series of zonesthroughwhich the bottles are passed by a conveying system. The continu-ous pasteurization of the product is achieved during the bottles’ transit

transfer analysis of pasteurization of bottled beer in a tunnel pasteurizerchnologies (2014), http://dx.doi.org/10.1016/j.ifset.2014.03.004

http://dx.doi.org/10.1016/j.ifset.2014.03.004

mailto:[email protected]


http://www.sciencedirect.com/science/journal/14668564


Head-space

Metal Cap

Beer

Glass

Fig. 1. Three-dimensional geometry of standard long neck beer bottle.

2 E. Bhuvaneswari, C. Anandharamakrishnan / Innovative Food Science and Emerging Technologies xxx (2014) xxx–xxx

through the tunnel, which consists of progressive hotter zones, holdingzone and progressively cooler zones (Briggs et al., 2004).

Most beers are pasteurized after filling to achieve microbiologicalstability and to inactivate molds and yeasts that might otherwise alterand deteriorate the product after processing (Reveron, Barreiro, &Sandoval, 2003). Beer spoilage organisms include wild yeasts, the pre-dominant of which belongs to the Saccharomyces species (Jespersen &Jakobsen, 1996). The inactivation of Saccharomyces cerevisiae in beer,based on PU values was studied by Zufall and Wackerbauer (2000)and Reveron et al. (2003). Reveron et al. (2003) studied the thermal re-sistance of S. cerevisiae and reported the decimal reduction times (DT °C)at various temperatures from 47 °C to 60 °C as follows: D47 °C =3.16 min, D48 °C = 2.65 min, D49 °C = 1.74 min, D50 °C = 0.68 min andD60 °C = 0.01 min.

The time–temperature combination for pasteurization of beer has amajor impact on product quality. Heat treatment affects the originalcharacteristics of beer in terms of flavour, colour, bitterness, chill hazeand protein sensitivity (Buzrul, Alpas, & Bozoglu, 2005). The inactiva-tion kinetics of beer spoilage micro-organisms forms the basis for PUcalculation (Zufall & Wackerbauer, 2000). Higher pasteurization tem-perature with shorter holding time favours the beer quality. Hence,the process temperature needs to be controlled by the water spray onthe beer bottles, during their residence inside the tunnel pasteurizer(Briggs et al., 2004).

In recent years, a rapid development in the application of CFD in foodprocessing operations such as drying, sterilization, mixing, refrigerationand storage has been witnessed (Anandharamakrishnan, 2003; Norton& Sun, 2006; Scott & Richardson, 1997). CFD is a simulation tool, whichuses powerful computers in combination with applied mathematics tomodel fluid flow situations and aid in the optimal design of equipmentsand industrial processes (Chhanwal, Anishaparvin, Indrani, Raghavarao,& Anandharamakrishnan, 2010; Kuriakose & Anandharamakrishnan,2010). The CFD simulations have been widely used to determinethe distribution of temperature and velocity profiles in the thermal pas-teurization and sterilization of canned food products (Anandpaul,Anishaparvin, & Anandharamakrishnan, 2011). It can be used to deter-mine the slowest heating zone (SHZ), which is defined as the locationthat receives theminimumheat (i.e. cold spot) during thermal process-ing and it can be tracked from temperature distribution predictions.

Relatively few works have been published on applications of CFD inthe pasteurization of beer. Augusto, Pinheiro, and Cristianini (2010)studied the effect of can orientation on beer pasteurization using CFD.They reported that, the orientation of beer cans did not result in any im-provement during beer pasteurization process. A major limitation oftheir study was that instead of using beer thermophysical properties,they used thermophysical properties of water in their simulation.Dilay, Vargas, Amico, and Ordonez (2006) optimized the pasteurizationtunnel configuration using CFD, with respect to minimum energy con-sumption. But, this model dealt only with the internal environment ofthe tunnel pasteurizer rather than the heat transfer inside the bottle.Hence, the present studywas aimed at developing a CFDmodel for pas-teurization of bottled beer, using thermophysical properties of beer toinvestigate the effect of thermal processing time on beer temperatureand location of SHZ. Further, the study was extended to investigatethe degree of inactivation of yeasts in the bottled beer during the tunnelpasteurization process. In this study, S. cerevisiae was selected as thetarget organism and PU value was used as the deciding parameter toascertain the influence of different tunnel pasteurization zone tempera-tures on the inactivation of the aforementioned target organism.

2. Materials and methods

2.1. Experimental methodology

Beer bottles of different shapes and sizes are available in themarket.Due to its extensive usage, a standard size long neck bottle (28.3 cm

Please cite this article as: Bhuvaneswari, E., & Anandharamakrishnan, C., Heatusing computational fluid dynamics, Innovative Food Science and Emerging Te

height and 7.3 cm width), containing beer (650 ml) with 7% alcoholcontent was selected (strong premium beer, United Brewery Pvt. Ltd.)for the experiment and CFD simulation studies. During experimenta-tion, a small hole was made at the centre of bottle's metal cap for inser-tion of the calibrated thermocouple. The hole was sealed to ensure thatair does not escape and the T-type thermocouple (sensitivity ± 0.5 °C)was held at the centre of the bottled beer (9.6 cm from the bottomof thebottle). The thermocouple was connected to data acquisition system(VR-18, Brainchild Electronics Co., Ltd.) for temperature and time acqui-sition. The water bath (pasteurizer) was allowed to reach a set pointtemperature of 60 ± 0.5 °C. As soon as the water bath attained thedesired temperature, the whole setup of the beer bottle (in a verticalposition) along with the probe was immersed into the water bath. Thetime–temperature readingwas recorded every 30 s till the beer reacheda stable temperature.

In this experimental setup, heat transfer occurred by natural convec-tion through water bath followed by conduction from glass bottle tobeer and inside the bottle, beer was heated by natural convection.Similar conditionswereused in this present study (boundary conditionsdescribed below) for experimental validation. While the beer bottleswere moved through different heating zones in a real case, differentheating zones were moved around the stationary beer bottle underthe simulated conditions. Hence, conditions used in this simulationstudy are expected to be similar to that prevalent in industrial scaletunnel pasteurization.

2.2. CFD simulation

Bottle geometry in three dimensionswas created using GAMBIT. Thegeometrical structure of beer bottle is depicted in Fig. 1. The geometryconsisted of three volumes, with the first volume representing the



3E. Bhuvaneswari, C. Anandharamakrishnan / Innovative Food Science and Emerging Technologies xxx (2014) xxx–xxx

outer glass bottle with a height of 28.3 cm and 3mm thickness. The sec-ond volume represented the beer in the bottom portion of bottle with aheight of 19.4 cm and finally, the third volume represented the head-space with a height of 8.9 cm. The surface and volume mesh were gen-erated and exported to CFD code (Ansys Fluent 12, 2009) for furtherprocessing. Tetrahedral grid was used to create a volume mesh for allthe three volumes and the total number of cells was 128370 (bottle —

34261, beer — 88219 and headspace — 5890 no. of tetrahedral cells).Coarse meshing was done for all 3 volumes, since it is normally usedto reduce the computational time without compromising the accuracy.Mesh (grid) independency study was carried out and it did not showany significant difference in predictions for the coarsemesh and normalmesh size simulations. In this study, a mesh was created with the totalnumber of cells 186,136 and exported to Ansys Fluent 12 (2009). Simu-lation prediction does not show much variation. Moreover, it requiresmore computation time and hard-disc space. The finite volumemethodwas employed to solve the partial differential forms of the continuityequations using the SIMPLE (semi-implicit pressure-linked equations)method for pressure–velocity coupling. A first-order implicit upwindscheme was used to interpolate the variables on the surface of controlvolume owing to its stability with respect to simulation time step size(set as 1 s) with 5 iteration/time step. The convergence was consideredto be attained at the root mean square value of less than 10−3 for pres-sure, density and velocity solution variables and 10−5 for temperature.A dell workstation (T5400), consisting of 2.00 GHz Intel Xenon proces-sor with 8 Gb RAMwas used for running the simulations and each sim-ulation took approximately 6 h.

2.3. Governing equations

The software Ansys Fluent 12 (2009) was used to solve the continu-ity, momentum and energy equations for the defined geometry and as-sociated boundary conditions (Ansys Fluent 12, 2009). The generalizedtransport equations solved are

(a) Conservation of mass equationEq. (1) is the unsteady, three-dimensional, mass conservation orcontinuity equation for the simplified case of a fluid flow.

∂ρ∂t þ∇: ρν!

� �¼ 0 ð1Þ

where, ν! is velocity vector (m/s), ρ is density (kg/m3).(b) Momentum equation

∂∂t ρν!

� �þ∇: ρν!ν!

� �¼ −∇pþ∇: μ ∇ν!þ ∇ν!ð Þþ

� �h iþ Sm ð2Þ

where, p is pressure (Pa), μ is dynamic viscosity (kg/m·s).Boussinesq approximation was applied for the density in thebody force term (Sm) and can be expressed as follows:

SM ¼ −ρrefβ T−Tref

� �g ð3Þ

where, ρref and Tref are the density (kg/m3) and temperaturerespectively at the boundary wall condition, and g is the gravita-tional force.

(c) Energy equation

∂∂t ρCpT

� �þ∇: ν! ρCpT

� �h i¼ ∇: k∇T½ � ð4Þ

where, k is thermal conductivity, Cp is the specific heat capacityand T is the temperature.


2.4. Boundary conditions

Thermo-physical properties of beer, headspace, glass bottle andmetal cap are listed in Table 1. Viscosity value of beer was adaptedfrom Romero et al. (2004). Transient 3D CFD simulation was carriedout. To account for conductive heat transfer through the bottle wall,its thickness was measured using a screw gauge and the same (3 mm)was used in the simulation. No-slip condition was applied at the bot-tle walls and beer surface. Beer inside the bottle was assumed to be atuniform initial temperature of 27 °C considering laminar flowthroughout the thermal processing. Beer was coupled with convec-tive heat transfer from headspace air. The hot water temperature(60 °C) was assumed to remain constant along the heating wallboundaries, during pasteurization. Hence, at the outer surface ofbottle wall, T water bath = T = T outer wall = 60 °C.

The published studies on modeling of beer pasteurization approx-imated the thermal properties of beer to be equivalent to that ofwater. A comparison graph for the thermal properties of water andbeer is presented in Fig. 2. From the figure, it can be inferred thatthe specific heat of beer is significantly different from water whilethe thermal conductivity values of beer are comparatively closer tothat of water. Hence, beer thermal properties were used in thisstudy.

The tunnel pasteurizer is vital equipment in the beer processing lineconsisting of different temperature zones. A typical temperature–timeprofile for a seven-zone tunnel pasteurizer is given in Table 2 (Priest &Stewart, 2006). Each zone is maintained at a different temperaturewith specific residence time for the bottles. The first two zones are thepreheat zones where the temperature of beer is gradually raised from2 °C to around 20 °C to avoid breakage of bottles due to pressure buildup. Zone 3 is the superheat zone, where the water spray temperatureis maintained at 65 °C for 14 mins and the temperature of beer is in-creased up to 60 °C. Followed by it, is the holding section, in whichthe temperature of both water spray and beer is maintained at 60 °Cfor 6 mins. Pasteurization of beer takes place in this zone 4 (holdingzone). The following three sections are cooling zones where the tem-perature of beer is cooled down to 28 °C. Out of the seven zones themost critical section is the superheat zone, where temperature mustbe very accurately controlled, otherwise over pasteurization of beermay take place, which results in flavour deterioration.

Owing to the difficulty in availability of computational resources, inthe present study, the temperature profile inside the bottlewas predict-ed by considering the bottles to be in stationary position. Hence, duringCFD simulation, the stationary bottle geometry was moved throughdifferent zones of the tunnel pasteurizer. The exit temperature of thebottled beer from the previous zone was set as its inlet temperature inthe following zone. The temperature of the bottle wall (T wall) in eachzone was set corresponding to the particular zone temperature. Thesimulated results were validated with the theoretical data from Priestand Stewart (2006).

2.5. Pasteurization unit

The application of heat during pasteurization causes microbial de-struction. The final quality of the product depends upon the amount ofheat it has received. The degree of heat treatment is represented by pas-teurization unit (PU). The number of PU per minute of holding time canbe calculated using the following equation (Zufall & Wackerbauer,2000):

PUmin

¼ 10T−Trefð Þ

z : ð5Þ

The desired yeast, S. cerevisiae was selected for the studying ofthe degree of inactivation. The z-value also called decimal reductiontemperature is used to express the temperature dependency of the



Table 1Thermophysical properties of beer, headspace, glass and metal used in CFD model.

Value Source

BeerSpecific heat capacity (kJ/kg °C) 2 * 10−5 × T2 − 2 * 10−3 × T + 4.118. Romero, Cabral, Telis, and Gabas (2004)Thermal conductivity (W/m °C) − 8.116 * 10−6 * T2 + 1.9 * 10−3 × T + 0.54611. Romero et al. (2004)

HeadspaceDensity (kg/m3) 1.067 * 10−5 × T2 − 4.4992 * 10−3 × T + 1.29008239. Erdogdu and Tutar (2011)Specific heat capacity (J/kg °C) 1007 Erdogdu and Tutar (2011)Thermal conductivity (W/m °C) 7.31 * 10−5 × T + 0.02367793 Erdogdu and Tutar (2011)Viscosity (kg/m s) 4.5 * 10−8 × T + 1.73568 * 10−5. Erdogdu and Tutar (2011)

Glass bottleDensity (kg/m3) 2500 Pinho and Marcelo (2005), Bailey and Elban (2008)Specific heat capacity (J/kg °C) 750 Pinho and Marcelo (2005), Bailey and Elban (2008)Thermal conductivity (W/m °C) 1.4 Pinho and Marcelo (2005), Bailey and Elban (2008) (add to reference)

Metal capDensity (kg/m3) 2702 Pinho and Marcelo (2005)Specific heat capacity (J/kg °C) 903 Pinho and Marcelo (2005)Thermal conductivity (W/m °C) 237 Pinho and Marcelo (2005)


inactivation rate. The z-value of S. cerevisiae is 6.94 °C at 60 °C (Zufall &Wackerbauer, 2000). Heating temperature of 60 °C was used as refer-ence (Tref). Eq. (5) was solved in the CFD simulation with respect totime and temperature of the process to evaluate the PU for each zoneof the tunnel pasteurizer.

3. Results and discussion

3.1. Experimental validation of the CFD simulation of beer in stationarybottle

A convective heatingmodewas selected for CFD simulation (station-ary beer bottle) and the surface heat transfer coefficient (h) was calcu-lated using combined CFD simulation and experimental temperatures,

Fig. 2. Comparison of thermal properties of beer and water (a) specific heat (b) thermalconductivity.


based on the method described by Denys, Pieters, and Dewettinck(2003) and Radhika, Malhotra, Anishaparvin, and Anandharamakrishnan(2010). The heat transfer coefficient values were selected in therange of 1500 to 500 W/m2 °C (including 1400, 1200, 1000, 800and 600 W/m2 °C). Heat transfer coefficient of 600 W/m2 °C wasfound to be in good agreement with the experimentally measuredtemperature (Fig. 3). It was observed that there was a curvilinear in-crease in temperature in the preheating zones and beer temperaturereached a steady state of 60 °C after 1200 s of processing time. Thisrapid increase in temperature may be due to the low viscous natureof the product (Anandpaul et al., 2011).

3.2. Temperature profiles of beer bottle during pasteurization in stationaryposition

Tracking the SHZ is important to assess the effectiveness of thepasteurization process. Fig. 4(a–g) shows temperature contours of thebeer bottle during pasteurization of beer at time intervals of 30 s, 90 s,120 s, 360 s, 420 s, 600 s, and 1140 s in the stationary position (waterbath experiment). During the initial stage of heating, SHZ was formedat the bottom of the bottle, due to changes in viscosity and density.However, the headspace region was heated faster due to the differencein thermophysical properties between the headspace air and beer. Inthe course of heating, SHZ was observed at the bottom of the bottle.Fig. 5 shows the CFD simulated temperature profiles extracted at threepoints, point 1 (top point 18.4 cm from the bottom of the bottle),point 2 (centre point 9.6 cm from bottom), and point 3 (bottom point2 cm from bottom) inside beer bottle during the pasteurization processat 60 °C. It is evident that point 1 reached a higher temperature, fasterthan other points (i.e. the top portion of the bottle heated much faster).The beer at point 2 (centre) was heated at a faster rate than point 3(bottom), which justifies the location of SHZ towards the bottom ofthe bottle.

Table 2Typical temperature–time profile for a seven-zone pasteurizer (Priest & Stewart, 2006).

Package temperature (°C)

Zone no. Water temp. (°C) Time (min) In Out

1. Preheat 22 6 2 92. Preheat 32 7 9 213. Superheat 65 14 21 604. Hold 60 6 60 605. Cooling 1 40 10 60 436. Cooling 2 32 7 43 367. Cooling 3 22 6 36 28



Fig. 3. Comparison of experimental measurement (centre point — 6.95 cm from thebottom) of temperature and CFD simulation results during pasteurization of bottledbeer at 60 °C.

Fig. 5. CFD simulated temperature profiles (at three point) for beer bottle at pasteur-ization temperature of 60 °C (point 1— top point 18.4 cm from bottom; point 2— centre;point 3 — bottom point 2 cm from bottom).


Erdogdu and Tutar (2011) studied the effect of temperature onheadspace in canned water and reported that headspace temperatureincreased rapidly due to the lower heat capacity and viscosity of theair. Hence, the insulation effect of headspace cannot be claimed andmoreover, the heating was even faster when compared to the situationwhere the headspace was assumed to be filled with the product itself.Augusto, Pinheiro, Tribst, and Cristianini (2009) studied the SHZ duringthe pasteurization of tomato concentrate in glass bottles and found thatSHZ was located at the bottom of the bottle. It was also observed thatthe funnel shaped geometry of the bottle neck pushed the cold spot tothe bottom of the bottle. Similar results were also obtained by Ghani,Farid, Chen, and Richards (1999a, 1999b) and the authors showed thatduring the thermal processing of liquid foods, the action of natural con-vection inside the can, forces the SHZ to migrate towards the bottomof the can. These observations agreed well with the present study,with the results (Fig. 4) clearly showing that the SHZ is present in thebottom region of the bottle. Kumar, Bhattacharya, and Blaylock(1990); Ghani, Farid, and Chen (2002) and Varma and Kannan (2006)also reported a similar type of results.

3.3. Temperature profile of the bottled beer in tunnel pasteurizer zones

An industrial scale tunnel pasteurizer is shown in Fig. 6. Tempera-ture contours of bottled beer for each zone of the tunnel pasteurizer isgiven in Figs. 7(a–m). Fig. 7a shows the temperature contour of bottle

Temperature (°C) a) 30s b) 90s c) 120s

Fig. 4. Temperature contours (a–g) of bottled beer during the


at 330 s in the first preheat zone, in which thewater spray temperaturewasmaintained at 22 °C. The beer temperature was 2 °C during the ini-tial time of preheating and it increased to 15.2 °C at the end of thepreheating time (figure is not shown). Fig. 7b shows the temperaturecontour of bottle at the 420th second of processing time in the secondpreheating zone (water spray at 32 °C with 15.2 °C and 27.3 °C beertemperature upon entering and the exit temperature of simulated bot-tled beer respectively). From Fig. 7a & b, it can be observed that the SHZwas located at the bottom of the bottle as discussed earlier (see Fig. 5).Moreover, the top portion of the bottle heated up faster than the bottomregion due to the natural convection and presence of headspace.

Due to natural convection, the temperature gradient created a den-sity difference and the hot beer began to flow upwards. This upwardmovement of the hot liquid also carried the colder fluid by viscousdrag. The fluid flowing upward was deflected by the top cone surfaceof the bottle, after which it started moving in radial direction and thendownwards. Thus, a recirculation phenomenon was observed till uni-form heating was observed in the fluid (Holdsworth & Simpson, 2007;Kiziltas, Erdogdu, & Palazoglu, 2010; Kumar et al., 1990). This observa-tion is similar to the results obtained by others on natural convectionheating (Ghani et al., 1999a; Kumar & Bhattacharya, 1991). Kumaret al. (1990) reported that the slowest heating point was located at

d) 360s e) 420s f) 600s g) 1140s

pasteurization (60 °C) process at different time intervals.



Fig. 6. Schematic diagram of tunnel pasteurizer.


the geometric centre of the can, during the conduction heating of acanned thick viscous liquid food from all sides (top, bottom and side-wall). In the case of pure convection heating of liquids, the SHZ waslocated at about 10–15% of the can height (Pflug, 1987).

Temperature (°C) Zone one (preheat 1) Zone two (preheat )2

Water spray temperature ( 22 °C)

Water spray temperatur

a) 330s b) 420s c) 51

Temperature

scale (° C)

Zone four ( holding) Zone five ( cooling 1)

Water spray

temperature (60°C)

Water spray

temperature (40°C)

h) 1980s i) 2010s j) 2

Fig. 7. Temperature profiles (a–m) of bottled beer in differen


The second preheat zone at processing time of 510 s (Fig. 7c) andsuper heat zone at 1140 s (with water spray temperature, 65 °C andinlet and exit temperature of beer respectively at 27.3 °C and 62.3 °C)(see Fig. 7f) showed that SHZ was located at the bottom of the bottle.

)taehrepus(eerhtenoZ

e (32°C) Water spray temperature ( 65°C)

0s d) 810s e) 960s f) 1140s

Zone six ( cooling 2) Zone seven ( cooling 3)

Water spray

temperature (32°C)

Water spray

temperature (22°C)

220s k) 2700s l) 3300 m) 3360s

g) 1320s

t zones of tunnel pasteurizer at different time intervals.




This correlated well with Fig. 5 (constant heating at 60 °C), where, SHZwas found at the bottle bottom due to convective heating and funnelshaped neck of the bottle. Fig. 7k shows that the temperature contourof bottle in the second cooling zone at 2700 s was maintained at32 °C, indicating that the bottom portion of the bottle cooled slowlywhen compared to the top portion. Hence, the rate of heat transferwas low in the bottom region when compared to other portions of thebottle during both heating and cooling processes. Moreover, rapidincrease in temperature increased the pressure inside the bottle whichin turn was dependent on the headspace of the bottle. Bottles withsmall headspace region experienced a steep rise in pressure, whichresulted in bursting of bottles. Hence, a low temperature, long timepasteurization profile is usually adapted for tunnel pasteurizationprocess (Priest & Stewart, 2006).

Fig. 8 compares the temperature profile of bottled beer in each zoneof the tunnel pasteurizer with the theoretical data. Theoretical data wasobtained from Priest and Stewart (2006) and wasmeasured using tem-perature probe in a beer bottle at each zone of the pasteurizer. It can beseen from Fig. 8 that, there was a gradual increase of temperature ofbeer in zone 2 (second preheat at 32 °C) due to gradual increase in tun-nel zone temperature (10 °C increase). A curvilinear increase in temper-ature at zone 3 (superheat at 65 °C) was observed and this may be dueto the sudden increase in zone temperature from32 °C to 65 °C. Further,a very gradual decrease in temperature with respect to tunnel zonetemperature was observed in cooling zones (40 °C, 32 °C and 22 °C).The exit temperature of beer bottle at the end of each of the 7 zoneswere 15.2 °C, 27.3 °C, 62.3 °C, 60.8 °C, 43.2 °C, 35.1 °C and 26.5 °C,respectively. These predictions are in reasonable agreement with thepublished data (Table 2).

However, some deviation in zones 1 and 2was observed. During theinitial period of heating (zone 1 and zone 2), a slight temperature differ-ence in beer, near bottlewall and core regionwas observed due to buoy-ancy force. Later, the temperature difference was reduced due to higherheat transfer promoted by the natural convection (Amoura, Zeraibi,Smati, & Gareche, 2006). In the simulation, buoyancy effect may pro-duce small-scale instabilities. The buoyancy-produced structuresmight have interacted directly with the existing local turbulence withthe strong coupling where laminar modeling could not predict thiswell. This could be the reason because of which an accurate predictionof the temperature gradient in the vertical direction was not possible(Erdogdu & Tutar, 2011). However, beer temperature predictions of re-maining zones from 3 to 7 were in good agreement with theoreticaldata obtained by Priest and Stewart (2006). Therefore, the model can

Fig. 8. Comparison of CFD simulation results with theoretical data for different zones oftunnel pasteurizer.


be extended to study the different zones of a tunnel pasteurizer withmore bottles.

3.4. Pasteurization unit

The degree of inactivation of the target microorganism S. cerevisiae(yeast) was estimated using Eq. (5) integrated into CFD simulation, forall the seven zones of the tunnel pasteurizer. The PU/min values werecalculated for every minute of the thermal process. The values wereextracted at three points (point 1, 2 & 3) inside the beer bottle duringthe pasteurization process for all seven zones. In Fig. 9, there exists asteep increase in PU values (around 2) in the superheat zone, wherethe zone temperature was maintained at 65 °C. A similar trend wasshownbyHough, Briggs, and Stevens (1971) in their experimentalmea-surement of PU during beer pasteurization in a tunnel pasteurizer. Fig. 9also showed that all the three points (points 1, 2 & 3) followed the sim-ilar trend of inactivation with slight variations due to natural convec-tion. The PU value for all 7 zones was calculated (using Eq. (5)), whichresulted in 19.89, 15.71 and 13.96 (only PU) for points 1, 2 and 3 respec-tively. Further, it could also be observed that PU value almost followedthe temperature predictions (see Fig. 8), where maximum temperature(62.3 °C)was attained at 1620 s. Similarly, higher PU valuewas predict-ed at 1620 s (Fig. 9). Beer pasteurizationwith 15–30 PU is considered tobe a safe practice Briggs et al. (2004). Hence, this study indicated thatthe obtained PU value of 17 is sufficient to achieve the requiredinactivation.

4. Conclusion

CFDmodel for thermal pasteurization of bottled beer was developedand validated with the experimental measurements of temperature.This study indicated that, using the actual thermophysical propertiesof beer was more appropriate than approximating the values withthat of water. The results of this work signified that, using the appropri-ate thermophysical values lead to close correlation between the simula-tion predictions and experimental measurements. It could also beinferred from the study that, the SHZ was located towards the bottomof the bottle due to natural convection effect and the location wasdependent on the processing time. Further, this study was extendedto investigate the temperature profile inside the bottle for an industrial

Fig. 9. Pasteurization unit (PU) calculated (at three points) inside the bottled beer forseven zone tunnel pasteurizer (point 1— top point 18.4 cm frombottom; point 2— centre;point 3 — bottom point 2 cm from bottom).




tunnel pasteurizer with seven zones, with each zone maintained at adifferent temperature. The temperature profile for each zone of thetunnel pasteurizer was predicted in stationary bottle position and vali-datedwith the available theoretical data. The time–temperature combi-nation for pasteurization of beer has a major impact on beer quality interms of flavour, colour, bitterness, chill haze and protein sensitivity.Hence, the progress of the pasteurization process was evaluated basedon PU values for all zones and obtained PU value of 17 was sufficientto achieve the required inactivation.

Acknowledgement

The authors wish to thank Prof. Ram Rajasekharan, Director,CSIR–CFTRI for his support. We also wish to thank Dr. N. NagendraGandhi, Professor, Department of Chemical Engineering, A.C. Tech,Anna University, Chennai for his help.

References

Amoura, M., Zeraibi, N., Smati, A., & Gareche, M. (2006). Finite element study of mixedconvection for non-Newtonian fluid between two coaxial rotating cylinders.International Communications in Heat and Mass Transfer, 33, 780–789.

Anandharamakrishnan, C. (2003). Computational fluid dynamics (CFD) — Applicationsfor the food industry. Indian Food Industry, 22(6), 62–68.

Anandpaul, D., Anishaparvin, A., & Anandharamakrishnan, C. (2011). Computational fluiddynamics studies on pasteurisation of canned milk. International Journal of DairyTechnology, 64, 305–313.

Ansys Fluent 12 (2009). Ansys fluent user's guide (version 12). Lebanon: Fluent Inc.Augusto, P. E. D., Pinheiro, T. F., & Cristianini, M. (2010). Using computational fluid

dynamics (CFD) for the evaluation of beer pasteurization: Effect of orientation ofcans. Ciência e Tecnologia de Alimentos, 30(4), 980–986.

Augusto, P. E. D., Pinheiro, T. F., Tribst, A. A. L., & Cristianini, M. (2009). Evaluation ofmeth-odologies for mathematical modeling of packaged conductive foods heat process.International Journal of Food Engineering, 5(4) (article 6).

Bailey, R. T., & Elban, W. L. (2008). Thermal performance of aluminum and glass beerbottles. Heat Transfer Engineering, 29(7), 643–650.

Briggs, E. D., Boulton, C. A., Brookes, P. A., & Stevens, R. (2004). Brewing Science and practice.Woodhead Publishing Limited and CRC Press.

Buzrul, S. (2007). A suitable model of microbial survival curves for beer pasteurization.LWT — Food Science and Technology, 40, 1330–1336.

Buzrul, S., Alpas, H., & Bozoglu, F. (2005). Effect of high hydrostatic pressure on qualityparameter of lager beer. Journal of the Science of Food and Agriculture, 85, 1672–1676.

Chhanwal, N., Anishaparvin, A., Indrani, D., Raghavarao, K. S. M. S., & Anandharamakrishnan,C. (2010). Computational fluid dynamics (CFD) modeling of an electrical heating ovenfor bread-baking process. Journal of Food Engineering, 100, 452–460.

Denys, S., Pieters, J. G., & Dewettinck, K. (2003). Combined CFD and experimentalapproach for determination of the surface heat transfer coefficient during thermalprocessing of eggs. Journal of Food Science, 68(3), 943–951.

Dilay, E., Vargas, J. V. C., Amico, S. C., & Ordonez, J. C. (2006). Modeling, simulation andoptimization of a beer pasteurization tunnel. Journal of Food Engineering, 77, 500–513.

Erdogdu, F., & Tutar, M. (2011). Velocity and temperature field characteristics of waterand air during natural convection heating in cans. Journal of Food Science, 76(1),E119–E129.

Ghani, A. G., Farid, M.M., & Chen, X. D. (2002). Theoretical and experimental investigationof the destruction of vitamin C in food pouches. Computers and Electronics inAgriculture, 34, 129–143.


Ghani, A. G., Farid, M. M., Chen, X. D., & Richards, P. (1999a). Numerical simulation ofnatural convection heating of canned food by computational fluid dynamics. Journalof Food Engineering, 41, 55–64.

Ghani, A. G., Farid, M. M., Chen, X. D., & Richards, P. (1999b). An investigation of deactiva-tion of bacteria in a canned liquid food during sterilization using computational fluiddynamics (CFD). Journal of Food Engineering, 42, 207–214.

Holdsworth, D., & Simpson, R. (Eds.). (2007). Thermal processing of packaged foods. Foodengineering series. Newyork, USA: Springer.

Horn, C. S., Franke, M., Blakemore, F. B., & Stannek,W. (1997). Modeling and simulation ofpasteurization and staling effects during tunnel pasteurization of bottled beer. Foodand Bioproducts Processing, 75(1), 23–33.

Hough, J. S., Briggs, D. E., & Stevens, R. (1971).Malting and brewing science. : Chapman andHall LTD.

Jespersen, L., & Jakobsen, M. (1996). Specific spoilage organisms in breweries and labora-tory media for their detection. International Journal of FoodMicrobiology, 33, 139–155.

Kiziltas, S., Erdogdu, F., & Palazoglu, T. K. (2010). Simulation of heat transfer for solid–liquidfood mixtures in cans and model validation under pasteurization conditions. Journal ofFood Engineering, 97, 449–456.

Kumar, A., & Bhattacharya, M. (1991). Transient temperature and velocity profiles in acanned non-Newtonian liquid food during sterilization in a still-cook retort.International Journal of Heat and Mass Transfer, 34(4–5), 1083–1096.

Kumar, A., Bhattacharya, M., & Blaylock, J. (1990). Numerical simulation of natural con-vection heating of canned thick viscous liquid food products. Journal of FoodScience, 55(5), 1403–1411.

Kuriakose, R., & Anandharamakrishnan, C. (2010). Computational fluid dynamics (CFD)applications in spray drying of food products. Trends in Food Science and Technology,21, 383–398.

Legras, J. L., Merdinoglu, D., Cornuet, J. M., & Karst, F. (2007). Bread, beer and wine:Saccharomyces cerevisiae diversity reflects human history. Molecular Ecology, 16,2091–2102.

Norton, T., & Sun, D. W. (2006). Computational fluid dynamics (CFD) an effective andefficient design and analysis tool for the food industry — A review. Trends in FoodScience and Technology, 7, 600–620.

Pflug, I. J. (Ed.). (1987). A textbook for introductory course in microbiology and engineeringof sterilization. Mineapolis, MN: Environmental Sterilization Lab.

Pinho, C., & Marcelo, C. (2005). Three-dimensional mathematical modeling of microbio-logical destruction of Bacillus stearothermophilus in conductive baby food packed inglass container. International Journal of Food Engineering, 1.

Priest, F. G., & Stewart, G. G. (2006). Handbook of Brewing. : CRC Taylor and Francis Group.Radhika, R., Malhotra, D., Anishaparvin, A., & Anandharamakrishnan, C. (2010). Computa-

tional fluid dynamics simulations studies on pasteurization of egg in stationary androtation modes. Innovative Food Science and Emerging Technology, 12, 38–44.

Reveron, I. M., Barreiro, J. A., & Sandoval, A. J. (2003). Thermal resistance of Saccharomycescerevisiae in Pilsen beer. Journal of the Institute of Brewing, 109(2), 120–122.

Romero, T., Cabral, R. A. F., Telis, V. R. N., & Gabas, A. L. (2004). Effect of corrugation onthe hydrodynamic and heat transfer behaviour of beer flowing through rectangularcorrugated channels. International Conference on Engineering and Food, 9.

Scott, G., & Richardson, P. (1997). The application of computational fluid dynamics in thefood industry. Trends in Food Science and Technology, 8, 119–124.

Takashio, M., & Shinots, K. (1998). Preventive production of beer against oxidation —Recent advances in brewing technology. Food Science and Technology International,4(3), 169–177.

Vanderhaegen, B., Neven, H., Verachtert, H., & Derdelinck, G. (2006). The chemistry ofbeer aging — A critical review. Food Chemistry, 95, 357–381.

Varma, M. N., & Kannan, A. (2006). CFD studies on natural convective heating of cannedfood in conical and cylindrical containers. Journal of Food Engineering, 77, 1024–1036.

Zufall, C., & Wackerbauer, K. (2000). The biological impact of flash pasteurization over awide temperature interval. Journal of the Institute of Brewing, 106(3), 163–168.


http://refhub.elsevier.com/S1466-8564(14)00046-0/rf0005































































































Documents

Innovative Food Science and Emerging Technologiesstatic.tongtianta.site/paper_pdf/fb885612-61d5-11e9-a14c... · 2019-04-18 · E. Bhuvaneswari, C. Anandharamakrishnan⁎ Food Engineering