Study a new atmospheric freeze drying system incorporating a vortex tube and multi mode heat input 3

The test setup was designed and constructed to allow testing of multimode or intermittent heat input together with variable pressure of the compressed air supplied to the vortex tube to

CHAPTER 3 EXPERIMENTAL APPARATUS AND PROCEDURE

An experimental investigation was undertaken to evaluate the drying performance of a vibrated atmospheric freeze drying by varying over a range of possible parameter values The test setup was designed and constructed to allow testing of multimode or intermittent heat input together with variable pressure of the compressed air supplied

to the vortex tube to provide different subzero temperature inside the drying chamber The experimental program, including the design and construction of the test set-up, the instrumentation and the measurements is presented in this chapter Experiments were also conducted using a laboratory freeze dryer and a heat pump dryer in order to carry out a comparison with the newly developed proposed AFD system

3.1 Experimental Apparatus

A schematic of the atmospheric freeze drying system along with a photograph of the test rig is shown in Figures 3.1 and 3.2, respectively It consists of a vibrator with variable amplitude (1-5 mm) and frequency (1-25 Hz), a screw compressor, a vortex tube cooler, a noise muffler, a ceramic radiation heater assembly, a conduction plate,

an insulated dryer drum, freezer and insulated dryer exhaust Details of the major components of the set-up are described in the next section

Figure 3.1 Schematic layout of the atmospheric freeze drying system

Figure 3.2 Photograph of the experimental setup

3.1.1 Drying chamber

Figure 3.3 is a photograph of the AFD dryer The dryer was constructed of a horizontally oriented drum, made of acrylic and insulated with Armoflex Length, inner radius and wall thickness of the dryer were 300mm, 200mm and 5mm, respectively Two O-rings of different radious (115mm and 125mm) and equal thickness (3mm) were used with a flange type fitting on one side of the drum to seal it properly Exit air from the drying chamber passed through the open area of a 3mm circumferential gap between the acrylic drum and the flange The drying chamber was well insulated with 5mm thick Armoflux insulation material A rectangular tray made

of the aluminum sheet was placed approximately at the middle position inside the drum to hold the samples

3 mm gap

Figure 3.3 Photograph of the dryer

The length and width of the tray were about 300mm and 150mm, respectively The tray was seated on a strip made of an acrylic sheet of 5mm thickness and 280mm length The strip was fixed to the wall of the dryer on both sides using adhesive cement

3.1.2 Conduction and radiant heater

The drying samples received heat by conduction through the heated tray or by radiation A silicon rubber heater (300 W) was attached to the bottom of the tray to heat the plate by conduction The dimensions of the conduction heater were 245 mm x

147 mm

3.1.3 Vortex tube

A vortex tube cooler (Model 3240, EXRIR Corporation, Cincinnati, Ohio) with 0.82

kJ/s refrigeration capacity and 0.018876 m3/s air flow at STP was used to supply sub- zero temperature air to the dryer is shown in Figure 3.5 It is made of stainless steel

material The vortex tube is a device for producing hot and cold air when compressed air is made to flow tangentially into the vortex chamber through properly designed the

Figure 3.5 Photograph of the vortex tube used in experimental setup

Figure 3.6 Lateral section of vortex tube showing a graphical view of the cold and hot

air flows

inlet nozzles This causes a complex swirling motion within the vortex tube The cold air in the core region of the tube flows out through an orifice plate in the direction

opposite to that of the hot air near the shorter tube wall flowing out through the core A cross sectional view of the vortex tube is shown in Figure 3.6 Control of air temperature can be achieved by controlling supply the air pressure or by controlling the amount of cold air released

3.1.4 Muffler

Vortex tube generates a unacceptance level of noise A muffler was used at both ends

of the vortex tube to reduce the emitted noise to an exceptable limit It was made of an aluminum material A photograph of muffler is as shown in Figure 3.7

Figure 3.7 Photograph of muffler used to reduce noise level

3.1.5 Pressure regulator

A pressure regulator was used to control and measure the air pressure at the inlet of the vortex tube as well as inside the drying chamber It was used to control the carrier gas

temperature inside the drying chamber by controlling the inlet air pressure

3.1.6 Vibrator

The drying chamber was vibrated mechanically by using a magnetic coil vibrator (Model 406, Ling Dynamic, Germany) of variable frequency (3-30 Hz) and amplitude (0-5mm) was placed directly under the drying chamber Vibrator was fixed up in such way so that it only vibrates the tray to allow a gentle vibration of the samples in vertical directions Maximum load of the vibrator of 1 kg with a vibration factor (aω2 /g) in between 3-7

Detailed of the system component specifications are listed in Table 3.1

Table 3.1 Components specification and characteristics of the system parameters

1 Drying chamber

a Material : Acrylic sheet

b Shape : Drum type

c Dimension : Length: 300mm Inner diameter: 200mm Thickness: 5mm

d O-ring : 1) Inner radius: 115 mm Thickness: 3 mm

2) Inner radius: 125 mm Thickness: 3 mm

e Flange : Material – Acrylic sheet

Radius: 145 mm Thickness: 5mm

f Insulation material : Armoflax; Thickness: 5mm

2 Tray

a Material : Aluminum sheet

b Dimension : Length: 300mm Width: 150 mm

5 Vortex tube

b Flow rate : 0.0188 m3/sec

c Refrigeration capacity : 706 Kcal/hr

3.1.8 Heat pump dryer

In the heat pump drying runs the air temperature was set at 45°C with an air velocity and relative humidity (RH %) of about 1.45 m/sec and 18%, respectively Reductions

of mass of sample during the course of experiments were measured at a regular interval of 30 minutes A full description and experimental procedure for the heat pump dryer can be found in Lan et al (2005)

3.2 Experimental Procedure

3.2.1 Fixed bed couple with vortex tube and multimode heat input

Potato and carrot were used as the model drying materials as properties of this material are well documented Samples in the form of discs (16mm diameter x 1 mm thickness) were washed, peeled and cut for the AFD experiments To avoid enzymatic browning, the slices were immersed in a 5% sodium bicarbonate solution at a temperature of

96oC± 2o

C, for about two minutes and carefully wiped using tissue paper before start the drying run Potato samples were cooled down immediately to stop possible gelatinization Samples were weighed after blanching and placed on a wire mesh tray

in a freezer at -22°C for approximately 18 hours The product temperature was noted to

be about -20°C The samples were weighed again before placing them inside the drying chamber to measure the reduction of weight during freezing

Prior to start of the drying experiment the temperature of the chamber was kept below the freezing point of the samples to prevent melting which can cause damage to the structure of the product Four heat input schemes were compared experimentally: case1: single stage-pure convection, case2: two-stage-pure convection, case3: two-stage-radiation coupled with convection and case4: two stage-radiation-conduction coupled with convection Samples were then placed on a wire mesh tray seated on the heated conduction plate Weight of the samples was measured at a regular interval of 2 hours by taking them out of the drying chamber Time required to measure the samples weight at different stages of drying was about 45 seconds Table 3.2 shows details of the experiments performed in this work At the end of each experiment, the dried samples were placed in an oven at 105oC for 24 hours to measure their bone-dry

weight Drying kinetic data are presented in dimensionless moisture content verses drying time plots

Table 3.2 Schedule of experiments performed using fix bed dryer

Convective

heating (case-1)

Single stage: -11°C (0 to 8 convection)

hr-Convective

heating (case-2)

Two stage: -11°C (0 to 4 convection) and -6°C (after 4 to 8 hr-convection)

hr-Convective with

Radiation

heating (case-3)

Potato and carrot

Two stage: -11°C (0 to 4 convection) and -6°C (after 4 hr to

hr-8 hr – convection and radiation (12°C) heat input)

Disc shape:

16 mm (D)x1mm Rectangle shape: (10x 5x 1)mm Multi-mode

(Radiation and

Conduction)

heating (case-4)

Two stage: -11°C (0 to 4 convection) and -6°C (after 4 hr to

hr-8 hr – convection, radiation and conduction heat input )

gas inspite of clogging the passage through precipitation of the osmotic agents These phenomena will also contribute to enhance the drying rate in AFD process Therefore, osmotic dehydration as a pretreatment to AFD method for materials of biological origin (fruit in sucrose, fish and meat in salt) were undertaken to examine the effect of osmotic agents on AFD

Samples of the materials were cut to discs type of 26mm diameter and 1mm thickness

by using a special cutter and a knife Same initial weight 0.75 gm was taken for each material for all drying experiment 10 gm of sugar is mixed with 5 gm of water at room temperature (24oC) to obtain a concentrated sucrose solution while 5 gm of salt is mixed with 50 gm of water to obtain a concentrated salt solution for the osmotic pretreatment of the samples Banana, carrot and potato samples were immersed in sucrose solution, while beef liver, beef meat and cod fish samples were immersed in the salt solution for 30 minutes for the osmotic pretreatment The samples were then taken out and removed the surface moisture by using tissue paper The weight of the osmotic pretreated samples was then measured to determine the loss of moisture due to the osmosis process The samples were then kept in a refrigerator at

-22ºC for 18 hours on a wire mesh tray before being placed inside the atmospheric freeze dryer or the vacuum freeze dryer

3.2.2 Fibro-fluidized bed: Sample mixed with an adsorbent

Vibration was applied to the drying chamber using a magnetic coil vibrator of adjustable frequency (f) and amplitude (A) An analyzer and amplifier were used to measure the amplitude (3-4 mm) and frequency (17-22 Hz) values of vibration-parameter (3.5-6.4), respectively As noted earlier 2 mm cube of potato and carrot were

used as samples weighted about 1.3 gm Blanching of the product was carried out to avoid enzymatic browning Silica gel beads were used as adsorbent particles This material shows good characteristics of adsorption even at low air humidity Gel particles with an average diameter of 3 mm and an adsorbent-to-product mass ratio of 1:1 were used Experiments were conducted using both cooled and ambient temperature adsorbent Prior to start of the experiment, the temperature of the chamber was kept below the freezing point of the samples

Table 3.3 Schedule of experiments performed using the vibrating bed dryer

Vibration: f - 20 Hz,

A – 4mm and 3mm

• Multimode-Two stage: -11°C and -6°C (case-4)

Vibration: f-22 Hz, A - 4mm

• Multimode -Two stage: -11°C and -6°C (case-4) Vibration: f -17 Hz, A - 4mm

• Vibrating factor: 3.5, 4.8, 5.8 and 6.4

• Potato &

Carrot Cube: 2mm cubic

Silica gel Spherical shape: D-1mm

• Without Adsorbent

• With Adsorbent

• Adsorbent refreshing

Comparison

with

literature

results

• Literature data: Two stage:

-8°C and 20°C (convection heat input - using heat pump)

• Proposed AFD: Multimode heat input-Two stage: -8°C and 20°C (Heat input using convection-radiation coupled with conduction); Without vibration; With vibration (f-20 Hz, A-3mm) and adsorbent

• Potato &

Carrot Cube: 2mm cubic

• Cod: 5mm cubic

Silica gel Spherical shape: D-1mm

• Without Adsorbent

• With Adsorbent Adsorbent refreshing

Frozen samples were then mixed with the adsorbent particles and placed on a wire mesh tray seated on the hot plate At a regular interval of 2 hours during the course of drying, product samples were separated from adsorbent for weight measurement Experiments were also carried out by charging with new adsorbent Experiment with a fresh dry adsorbent at regular intervals (2 hours) is termed as “adsorbent refreshing” Table 3.3 shows the schedule of the experiments performed in this study The quality

of the dried products was characterized as indicated in the following section

3.3.2 Color

A Minolta spectral magic spectrophotometer (Model CM-3500d) was used to determine the colour change due to drying as shown in Figure 8 The following difference formula from CIELAB was used to quantify change of color from the original to the dried product

2 1 2 2

L

E

b b b a a

here L, a, and b denote reference color and L*, a* and b* denote target color,

respectively Smaller value of ∆E*ab, indicates closer color of the dried product to its original color

3.3.3 Scanning electron micrographs (SEM)

SEM tests were performed on a JSM5600 machine Samples were cut horizontally as well as vertically across the surface to visualize the structures on different sides of the product Due to non conductive properties of the samples, a thin film of gold was coated over the food samples using a coating machine A magnification power of 40 to

80 was used for observation of the sample microstructure

3.3.4 Freezing point depression

The freezing point depression was estimated using a method given by Strommen et al

2005 The Schwartzberg equation was modified to estimate the freezing temperature as

a function of the solid fraction as follows

+

=

T T k

E b

1)

RT

H M

= and solid fraction;

Weight Water Solid

Weight Solid

k E

and the y-intercept is (1+b) A plot of WS against T°C gives the estimated value of the freezing point depression of the product

3.4 Measuring Equipment

To evaluate the drying performance, it is necessary to measure the weight of the product, temperature of the drying chamber as well as product, to control the temperature of the conduction and radiant heater and to measure the amplitude and frequency of the vibrator The instruments used for this measurement are described below

3.4.1 Analytical balance

Weight of the drying product was measured with a high precision analytical balance (Model B-320C, Explorer OHAUS, and USA) to an accuracy of +0.0001 gm The weighing plate on which the product is placed to take the measurement was fully surrounded with transparent material to avoid the effect of natural airflow during measurement

3.4.2 Temperature measurements

T-type copper-constantan thermocouples (Omega, USA) were used to measure the temperatures of the carrier gas in the drying chamber, conduction and radiant heater temperature and product temperature Thermocouples were inserted in the middle of the product to measure the local temperature of the product All thermocouples were calibrated in a bath using a standard liquid (ethylene glycol) with an accuracy of

±0.050C