Study a new atmospheric freeze drying system incorporating a vortex tube and multi mode heat input 6

6.1 Comparison Between Experimental and Simulation Results: Using Fix Bed Dryer and Multimode Heat Input 6.1.1 Drying kinetics Variation of moisture content with drying time for pur

CHAPTER 6 SIMUALTION RESULTS

This chapter provides a comparison between the predicted and experimental results of

a fix bed dryer on AFD process without using adsorbent Model is described in chapter

4 Three heat input schemes were compared: case1-pure convection, case2-two-stage

convection, case3-radiation-coupled convection Drying kinetics phenomena under

different ranges of operating conditions and effect of product thicknesses were

compared Numerical efforts were extended to predict the location and temperature of

the sublimation front layer during the course of drying Finally, moisture content and

temperature distribution inside the dry layer were predicted All results are discussed

and presented

6.1 Comparison Between Experimental and Simulation Results: Using Fix Bed

Dryer and Multimode Heat Input

6.1.1 Drying kinetics

Variation of moisture content with drying time for pure convection (case-1) at –11oC

and –6oC for potato pieces of both disc and rectangle shaped are shown in Figure 6.1

and Figure 6.2, respectively Higher drying rate was found at higher drying

temperature as expected Final dimensionless moisture content of the experimental and

predicted results at –11oC and –6oC was 0.40 and 0.36, and 0.13 and 0.12 for disc-

shape product, and 4.2 and 4.3, and 0.45 and 0.5 for cubical product, respectively, after

eight hours of drying time Good agreement between experiment and simulation results

was found in both cases Physical properties of the experimental dried products

(Rahman et al 2007) for the AFD under this condition proved a sublimation process

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Time, hr

Predicted-Single stage: -6C Predicted-Single stage: -11C Measured-Single stage: -6C Measured-Single stage: -11C

Figure 6.1 Variation of measured and predicted dimensionless moisture content

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

Time, hr

Measured-Single stage: -11C Predicted-Single stage: -11C Measured-Single stage: -6C Predicted-Single stage: -6C with time for disc shaped (16mm x 1mm) potato sample

Figure 6.2 Variation of measured and predicted dimensionless moisture content with

time for rectangle shaped (10mm x 5mm x 1mm) potato sample

from ice to vapor during drying Therefore, porous structure and thereby

non-shrinkable dried product due to absence of condensation inside the frozen product was

observed which was a key assumption in the model, results in a fairly good match with

experimental data

Figure 6.3 shows variations of the moisture content with time of disc shaped potato

samples for case-2 and case-3 In first-stage, drying process was conducted upto four

hours at –11oC for both cases through convection and then stepped up at –6oC for the

next four hours through convention heat input for the case-2 and radiation coupled

convection heat supply for case-3 In experiment, apparently a higher drying rate was

found after four hours of drying time in which drying temperature was stepped up at

–6oC from –11oC High-intensity drying conditions play an important role in

enhancing the sublimation rate, i.e higher drying rate In the case of multimode heat

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Time, hr

Predicted-Two stage: -11C & -6C

Predicted-Two stage: -11C & -6C ( Rad)

Measured-Two stage: -11C & -6C

Measured-Two stage: -11C & -6C ( Rad )

Figure 6.3 Variation of measured and predicted dimensionless moisture content

with time for disc shaped (16mm x 1mm) potato sample

input (case-3) shows further improvement in drying rate due to the additional

contribution of radiation heat input (Ratti and Mujumdar, 1995) These phenomena are

also captured well in the predicted results Final dimension less moisture content for

both experimental and simulation results for case-2 and case-3 after eight hours of

drying time were about 0.0775 and 0.049, and 0.058 and 0.0134, respectively Slightly

underprediction was observed in the predicted results just immediate after four hours

of drying time for case-3 This is probably due to a minor condensation in sublimation

layer during sublimation because of high intense drying condition at the beginning of

second-stage drying

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Time, hr

Predicted-Thickness-1mm Predicted-Thickness -2 mm Predicted-Thickness -3mm Predicted-Thickness -4mm Measured-Thickness-1mm Measured-Thickness- 2mm Measured-Thickness-3mm Measured-Thicnkness-4mm

Figure 6.4 Variation of measured and predicted dimensionless moisture content with

time for disc shaped (16mm x 1mm) potato samples of different thickness

Variation of product thickness of disc-shaped potato samples on the freeze-drying

kinetics for case-3 is shown in Figure 6.4 Final dimensionless moisture content from

experimental results was obtained about 0.07, 0.42, 0.61, and 0.68 for 1, 2, 3 and 4 mm

product thickness, respectively, after 8 hours of drying time while the corresponding

predicted results were 0.047, 0.49, 0.68, 0.74, respectively A good match was found

between the experiment and simulation for all thicknesses in terms of the final

moisture content; the curves also show similar behaviours Figure 6.4 show that drying

rate decreases with increase of product thickness The increases of sample thickness

implies an increase of dry layer thickness and increases in the water vapor diffusion

path, which decreases the rate of migration of sublimated vapor from inside to the

surface of the product Therefore, it can be argued that product thickness is one of the

key parameters in AFD process This result agrees with the previous work of Matteo et

al (2003) and Wolf and Gibert (1991)

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Time, hr

Measured-Single stage: -6C Measured-Single stage: -11C Predicted-Single stage: -6C Predicted-Single: -11C

Figure 6.5 Variation of measured and predicted dimensionless moisture content

with time for disc-shaped (16mm x 1mm) carrot

Comparison between simulation and experimental results on freeze-drying kinetics

was also carried out by using disc-shaped carrot samples for case-1, case-2 and case-3;

this is shown in Figure 6.5 and Figure 6.6 After 8 hours of drying time, the

dimensionless moisture content from the experiment and simulation results were

obtained 0.29 and 0.25, and 0.04 and 0.02 for case-1, respectively (Fig-6.5) The final

moisture contents for case-2 and case-3 were about 0.43 and 0.45, and 0.25 and 0.24,

0

1

2

3

4

5

6

7

8

9

10

11

Time, hr

Predicted-Two stage: -11C & -6C Predicted-Two stage: -11C & -6C (R) Measured-Two stage: -11C & -6C

Measured-Two stage: -11C & -6C ( R )

Figure 6.6 Variation of measured and predicted dimensionless moisture content

with time for disc shaped (16mm x 1mm) carrot sample

respectively (Fig-6.6) A similar trend of the drying kinetics curve between experiment

and simulation results as well as a close match between the final dimensionless

moisture content was found in both cases for carrot However, at the end of the drying

period, a slight discrepancy was observed between the predicted and simulation results

This is probably due to the melting of ice under such intense drying conditions during

experiment, which causes damage of internal structure due to surface tension effects

Hence it can cause flexible cell walls to collapse as the liquid in the pores is emptied

6.2 Predicted Parameters

6.2.1 Location and temperature of sublimation front

Besides these slight discrepancies between experiment and simulation results, fairly

good agreement was obtained to capture the drying phenomena of AFD system for

-20

-18

-16

-14

-12

-10

-8

-6

Time, hr

o C

Single stage: -6oC Single stage: -11oC

10

Single stage: -6C Single stage: -11C

Figure 6.7 Variation of the predicted sublimation front temperature with time for

potato for single stage drying process

various samples of different geometry and under drying conditions Therefore, the

present model can be used as a good tool to predict other important phenomena in an

AFD system, which usually not possible to measure experimentally

Figure 6.7 shows the variation of the predicted temperature with time at the interface

for case-1 As seen from this figure initially the temperature of the interface increases

rapidly from -17oC to –9oC within one hour of drying time at –6oC air and

subsequently becomes stable with time At the beginning of drying the interface ice

layer receives sensible heat to raise the temperature of water inside the product matrix,

which subsequently absorbs the latent heat for sublimation; this temperature is well

below then the triple point temperature of pure water At this point the sublimation

process begins After that the temperature of the evaporation front layer was

reasonably constant at the freezing temperature due to continuous sublimation of the

ice layer from the interface This result also proofs the frozen integrity of the product

during the entire AFD experiment

With air at –11oC, it takes two hours to approach a stable interface temperature of

about –13oC Due to the less intensity drying condition heat penetration rate decreases

from the carrier gas through the product to the sublimation layer This is because of

less temperature gradient between the carrier gas and the product temperature and

thereby takes more time to stabilize; consequently it decreases the drying rate

Figure 6.8 shows the predicted variation of the sublimation front temperature for

case-2 and case-3 In the first stage, for up to four hours of drying time, the two curves

overlaps in terms of the temperature distribution as well as the numerical value of the

final temperature, which is about -12oC This result is consistent as in both cases; the

drying condition was the same during this period In the second stage (4 hrs to 8 hrs), a

slightly higher temperature (0.16oC) was observed for case-3 than case-2 because of

the incorporation of the radiant heating, which provides uniform heating of the

-18.00

-16.00

-14.00

-12.00

-10.00

-8.00

-6.00

Ti

9

e, hr

Two stage: -11C & -6C

Two stage: -11C & -6C ( Rad )

m Figure 6.8 Variation of predicted sublimation front temperature with time for potato

for two stage drying process

0

0.0001

0.0002

0.0003

0.0004

0.0005

Time, hr

Stage: -11C Single stage: -6C Two stage: -11C & -6C Two stage: -11C & -6C ( Rad)

Figure 6.9 Predicted location of the sublimation front with time for potato

product

The computed location of the evaporation front under different drying conditions is

shown in Figure 6.9 Results show that at the beginning of drying, the location of the

evaporating front was at the surface of the product As drying progresses, the

evaporation front recedes inside the product due to sublimation of the interface ice

front, layer by layer The final location of evaporation front from the surface of the

product was about 0.000018 m, 0.000058 m, 0.0000028m, and 0.000007m,

respectively, for case-1, case-2 and case-3 At the higher intensity drying conditions,

particularly for case-3, a higher penetration rate of the evaporation front as the ice

layer deepens was observed

6.2.2 Temperature and moisture distributions inside the dry layer

Figure 6.10 shows the predicted distribution of the moisture mass fraction inside the

dry layer with distance for all four cases examined It was observed that the moisture

mass fraction increases from the surface towards the depth of the product as drying

progresses Water vapour formed by sublimation process at the interface layer results

in a higher partial pressure of the vapor near the evaporation surface Moisture then

travels from interior to the surface of the product due to a partial pressure gradient i.e

from higher concentration region to a lower conductive region During the flow of the

sublimed vapor higher accumulation of moisture takes place near the evaporation front

region due to the high concentration of water vapor and gradually decreases as

moisture travels towards the surface of the product Final moisture mass fractions for

case-1, case-2 and case-3 were computed to be 0.000308, 0.000611, 0.0008 and

0.000810, respectively, after 8 hours of drying The depth of the dry layers from the

0.0001

0.0002

0.0003

0.0004

0.0005

0.0006

0.0007

0.0008

0.0009

Distance, mm

Single stage: -11C Single stage: -6C Two stage: -11C & -6C Two stage: -11C & -6C ( Rad )

Figure 6.10 Predicted moisture distributions inside the dry layer with depth for potato

-7.70

-7.65

-7.60

-7.55

-7.50

-7.45

-7.40

Distance, mm

Single stage: -6C Two stage: -11C & -6C

Two stage: -11C & -6C ( Rad )

Figure 6.11 Predicted temperature distributions within the dry layer with depth for

potato

surface of the product for case-1, case-2 and case-3 were 0.182 mm, 0.058 mm, 0.028

mm, and 0.007 mm, respectively From this finding, it can be argued that accumulation

of moisture inside the dry layer is minimal This also implies that the major portion of

the sublimed water during the course of drying migrates to the carrier gas For the

two-stage process, combination of convection and radiation heat input (case-3) results in a

higher moisture distribution inside the dry layer in compared to other drying

conditions

The temperature distribution within the dry layer for different drying conditions with

distance is shown in Figure 6.11 After 8 hours of drying, the dry layer temperatures

for case-1, case-2 and case-3 were about -7.45oC, -7.64oC, and -7.51oC, respectively

The dry layer temperature increases with the increase of drying temperature, as

expected Higher temperature is observed near the surface of the product followed by a

reduction as dry layer depth increases Heat of sublimation penetrates through by

conduction from the surface of the product and gradually reaches the evaporation front

As s results, the dry layer temperature near the surface and subsequently towards

inside, come across more contact with inward heat flow and absorb more heat and

hence increases temperature distribution

6.3 Summary

Atmospheric freeze drying system using a vortex tube and multi-mode heat input was

studied experimentally and numerically Results showed good agreement between

simulation and experimental results which capture the drying phenomena of two

different products of different size and carrier gas temperatures under single and

multimode heat input Therefore, numerical efforts were extended and predicted well

the other important phenomena (sublimation front temperature, location of the

sublimation front and moisture content as well as temperature distribution inside the

dry layer) in AFD system, which usually not possible to measure experimentally

Results also illustrated that the process is recommended to work at the highest

possible temperature, which should, of course, be compatible with a high quality of

product conservation This simple model can be used as a tool to optimize the process

parameters