Fundamentals of spray drying

3.1.

Principle and terms

Spray drying is an industrial process for dehydration of a liquid feed containing dissolved and/or dispersed solids, by transforming that liquid into a spray of small droplets and exposing these droplets to a flow of hot air. The very large surface area of the spray droplets causes evaporation of the water to take place very quickly, converting the droplets into dry powder particles.

3.1.1.

Drying air characteristics

The drying medium used for drying of milk is atmospheric air, cleaned of dust by filtration and heated to provide the heat necessary for evaporation. Evaporation proceeds initially under adiabatic conditions. In such a system, all sensible heat from the drying air is utilized for evaporation of water, which becomes, as vapour, part of the drying air. The enthalpy of the air remains constant, supposing that the liquid entered the system with a temperature of 0°C (zero enthalpy) and absence of any heat loss. The various terms characterizing the drying air conditions are as follows:

Dry bulb temperature (td) is the temperature of air, which is not saturated with water vapour, as measured by an ordinary thermometer. In practice, the dry bulb is just referred as air temperature and is expressed either in °C (t) or as the absolute temperature in °K (T) whereby T = t + 273.15.

Wet bulb temperature (twb) (or more precisely Adiabatic saturation temperature) is a characteristic of moist air of a given dry bulb temperature, expressing the saturation temperature of that air with the same enthalpy, i.e. obtained by evaporation of 0°C water under adiabatic conditions. The difference between dry and wet bulb temperatures is a measure of drying capability (driving force). It is the temperature to which the air of dry bulb temperature (t) will drop, when evaporating water in an isolated air-water system until saturation condition occur (supposing that the temperature of water to be evaporated is 0°C). The enthalpy of the air during this evaporation remains unchanged, as the heat from the air is utilized for evaporation only. It can be also expressed as the temperature a droplet of water will obtain when exposed to a flow of air of temperature (t). Measuring wet bulb temperature is based on the same principle, i.e. the thermometer bulb is kept wet by a thin film of water and exposed to a flow of air. The relative humidity of the air at wet bulb temperature is 1.

Dew point temperature (tdp) is the temperature where condensation of vapour will commence, if the air is cooled down at constant absolute humidity. The relative humidity of the air at the dew point temperature is 1 and its enthalpy is lower than that of the same air at its dry bulb temperature and wet bulb temperature.

Air absolute humidity (y) is the ratio of the amount of water vapour (mv) to the amount of dry air (ma). Usually it is expressed in kg of water vapour per kg of dry air.

Thus:

[3,1]

…

[3,2]

…

Where:

p_v is the partial pressure of water vapour

p_t the total pressure and

0.622 the molecular weight ratio of the water vapour and of air, i.e. 18.015/28.954.

Air relative humidity (Φ or %RH) is the ratio of partial pressure of water vapour (pv) to the water vapour pressure at the saturation point (ps) at the same temperature.

[3,3]

…

The Extended Antoine Equation shows the relation between saturated water vapour pressure in Pa and temperature t in °C:

[3,4]

…

Saturation point is the air temperature at which any further temperature drop will result in condensation. Saturated air has equal dry bulb, wet bulb and dew point temperatures.

Drying air rate (Aa) is usually expressed as the mass flow of ambient air per hour (kg/h) and includes both the amount of dry air (Ad) and water vapour (Av) which can be calculated using equations:

[3,5]

…

[3,6]

…

[3,7]

…

Heat capacity is the amount of heat necessary to heat 1 kg of a substance by 1°C and is a function of the temperature.

Heat capacity of the air (c_a) is the amount of heat necessary to heat 1 kg of dry air by 1°C. It is expressed in J/kg/°C and is temperature dependent as shown below, where T is temperature in K:

[3,8]

…

To get c_a in kcal/kg/°C, equation [3,8] is divided by 4186.

For routine technical calculations a constant value 0.245 kcal/kg/°C or 1.026 kJ/kg/°C may be used.

The amount of heat (Q) necessary to heat a given amount of dry air (A_d) from t1 to t₂ °C is:

[3,9]

…

in which ca₁ and ca₂ are the values calculated from equation [3,8] corresponding to the temperatures t₁ and t₂.

Heat capacity of water (c_w) is approximately 1.0 kcal/kg/°C or 4.186 kJ/kg/°C.

Heat capacity of water vapour in J/kg/°C is:

[3,10]

…

To get c_v in kcal/kg/°C [3,10] is divided by 4186. For routine technical calculations of water vapour (c_v), 0.46 kcal/kg/°C or 1.926 kJ/kg/°C may be used.

Latent heat of evaporation (r) or vaporization is the amount of heat necessary to transform a liquid to vapour at constant temperature. The reverse process i.e. transforming a vapour to liquid requires a release of the same quantity of heat and is called heat of condensation. The latent heat of water is 597.3 kcal/kg or 2500 kJ/kg at the temperature 0°C and barometric pressure 760 mm Hg.

Enthalpy (h) of air is the thermal energy of that air expressed as sum of heat necessary to evaporate its moisture content at 0°C and to heat both the water vapour and dry air to its actual temperature, as expressed by the equation:

[3,11]

…

Density (p) of air is the weight per unit volume of air and it is a function of air temperature, moisture content and pressure. Usually it is expressed in kg/m3. The density of dry air p_d at the temperature of 80°C at the barometrical pressure 760 mm Hg is equal to 1, hence at the temperature of t°C it is:

[3,12]

…

The density of ambient (moist) air (p_a) is:

[3,13]

…

To calculate air density at the actual pressure B (in mm Hg), the above results must be multiplied by B/760.

3.1.2.

Terms and definitions

The following terms are used in spray drying technology:

Ambient air is the atmospheric air supplied to the system from the surroundings of ambient temperature (t_a) and ambient humidity (y_a)

Inlet air temperature (t₁) is the temperature of the air after heating or cooling at the inlet of a processing system having an inlet air absolute humidity (y₁)

Outlet air temperature (t₂) and outlet air humidity (y₂) express the same for air leaving the system

Water content of the feed (milk concentrate) or product (final powder) can be expressed in several ways:

a) Total solids content (TS) or Moisture content (W) expressed in weight percent.

b) Moisture content on dry basis (x) expresses the ratio of the quantity of moisture to the quantity of dry solids. The relationships between these expressions are:

[3,14]

…

[3,15]

…

[3,16]

…

Density or specific gravity of milk products, both concentrates and powders can be calculated using following formula:

[3,17]

…

Where: m₁, m₂ etc. are the contents of individual components in percent Q₁, Q₂ etc. their densities. The density is usually expressed in g/ml or kg/m3. The densities of some components of milk products are given in Table 3.1.

**Índice. 3.1.** Densities of some components of milk solids
Component	Density at 20°C,g/ml
Non-fat milk solids	1.52
Milk fat	0.94
Amorphous lactose	1.52
Alpha-lactose monohydrate	1.545
Whey solids	1.58
Milk proteins	1.39
Sugar (sucrose)	1.589
Water	1.00

Heat capacity of milk solids is also a function of temperature. However, for practical purposes it is sufficient to reckon with constants, as given in Table 3.2.

**Índice. 3.2.** Heat capacity of some milk components
Component	Heat capacity kcal/kg/°C
Non-fat milk solids	0.3
Milk fat	0.5
Water	1.0

Heat capacity of a product containing several components is calculated as a weight sum of heat capacity values of the individual components.

3.1.3.

Psychrometric chart

The conditions of drying air throughout the drying process are illustrated by the psychrometric chart (Mollier diagram or h-x, sometimes i-x diagram). The y-axis represents the temperature nd the x-axis absolute humidity. The psychrometric chart is constructed so that the isotherm corresponding to 0°C is horizontal. The isotherms for higher temperatures slope gradually more upwards. Lines representing enthalpy, saturation, constant relative humidity and vapour pressure are also shown. The saturation line divides the chart into the zone of unsaturated air and the zone of mist. The psychrometric chart illustrating all these air characteristics is given on fig. 3.1.

**Fig.3.1. The principle of h/x diagram.**

[The ambient air shown by point A has dry bulb temperature TG, enthalpy ha, absolute humidity xA, relative humidity RH, wet bulb temperature TWB and dew point temperature TDP].

3.2.

Drying of milk droplets

When spray drying milk, very high rates of heat and mass transfer take place in extremely short periods of time. Severe quality defects of the product can occur, if the factors, inducing degradation are permitted to dominate because of lack of knowledge or lack of operation control.

The milk concentrate leaves the atomizing device as a thin film at a velocity of 100 - 200 m/s, breaking up into droplets which immediately contact the hot drying air. Evaporation of most of the water in the droplets takes place during the time the droplets decelerate to reach the velocity of the surrounding gas. The smallest droplets lose about 90% of their moisture within a distance of 0.1 m from the atomizing device, whereas the largest droplets need about a 1m path. The rate of evaporation depends to a great extent on the total surface area of thedroplets, which is defined by the droplet size.

3.2.1.

Particle size distribution

Sprays of droplets as well as the produced powder particles are characterized by mean size and size distribution of the droplets and particles, respectively. The size distribution of a spray of droplets can be measured by laser light scattering techniques. Particle size distribution of a powder can be determined by the same method or by alternative methods such as microscopic counting, sifting or photographic methods with computer evaluation. The results of these methods express the frequency of droplets in a given size ranges or in cumulative numbers (smaller or larger than n microns). The results can also be expressed graphically by a histogram or by frequency. An example of expressing results in tabular form is given in Fig 3.2. Its graphical presentation is shown on Figs. 3.3 and 3.4.

3.2.2.

Mean particle size

The Mean particle size or droplet size can be expressed in several ways. The most common are:

Most frequent diameter, which can be seen directly from tabularized results or as the highest point of the frequency curve, possibly as an inflection point on the cumulative curve.

Arithmetic mean diameter, defined as the sum of the diameters of separate particles/droplets, divided by their number. This mean diameter is most significant when the size distribution is not overbalanced by either very large or very small elements.

Geometric mean diameter, defined as the n-th root of the product of the diameters of the n particles analysed. It has the highest frequency in the log-normal distribution.

Median diameter, which is the diameter corresponding to 50% of the number, weight or volume of the droplets /particles.

Apart from diameters based on frequency of size occurrence, there exist surface, volume and volume/surface mean diameters.

For characterization of the size distribution of a spray of droplets or dried powders, the most common is the geometric mean diameter. The volume/surface also called Sauter mean diameter is most suitable for spray drying operations as it expresses the same surface-tovolumeratio as the whole powder.

**Fig.3.2. Example of expressions for particle size distribution.**

**Fig.3.3. Cumulative curve of the example in Table 3.3.**

**Fig.3.4. Log-normal distribution curve of the example in Table 3.3.**

3.2.3.

Droplet temperature and rate of drying

The droplet and particle moisture profile during the whole process is often called the particle temperature history and it is of utmost importance not only for the structure of the particle and its surface, but also for potential product heat degradation.

The droplet/particle temperature during an ideal drying process is as follows:

The temperature of the droplet during the whole evaporation process lies between the temperature of the surrounding air and its wet bulb temperature. The droplet moisture determines the water activity of the droplet/particle. This, together with the relative humidity of the surrounding air decides where - between these two limiting points - is the actual droplet/particle temperature.
Droplets of water (having the water activity a_w = 1) will attain the wet bulb temperature regardless of the feed temperature once the first contact is made with the drying air. This temperature will be retained until evaporation is completed.
Droplets of milk concentrate at the beginning of the drying process will attain a temperature somewhat higher than the wet bulb temperature because the water activity of the concentrate is somewhat lower than 1 (about 0.85 - 0.90).
As water evaporates, the water activity (aw) gradually decreases. This results in a gradual rise of particle temperature towards the surrounding air temperature.
When equilibrium is achieved between the drying air and a particle, the particle water activity is equal to the relative humidity of the surrounding air and consequently the particle temperature is equal to the surrounding air temperature, i.e. aw =Φ and t_p = t2.

The evaporation of water from the surface of the milk droplets commences under so-called constant or first rate drying period conditions. It does not mean that the rate of drying is strictly constant because, as mentioned above, the water activity is decreasing. The droplet at this stage, however, is still a fluid in which the moisture can migrate easily from the droplet interior to the surface and keep it nearly saturated. The author of this book suggests the following relationship for the droplet temperature within this period:

[3,18]

…

At a later stage of the drying process the moisture content achieves a critical value at which the droplet loses the character of a fluid and becomes a wet solid. The critical moisture content of milk products is dependent on many factors and operating circumstances. However, it is in the range of 30 - 15%. It is characterized by a sudden occurrence of a moisture gradient across the droplet diameter. At this stage the factor controlling the rate of drying becomes the rate of diffusion of the moisture through the particle.

This period is known as falling or second drying rate period (Fig.3.5). The rate of heat transfer exceeds that of mass transfer and the particle begins to heat up faster, than indicated in equation [3,19]. There are both moisture and temperature gradients in the particle interior, and a hard crust forms on the surface.

3.2.4.

Particle volume and incorporation of air

During the evaporation of water the droplets decrease in size. The theoretical reduction of diameter, weight, volume and surface area when drying droplets under ideal conditions from 50 to 0% moisture, expressed in percent of initial values is graphically shown on fig. 3.6.

**Fig.3.6. Reduction of droplet size during evaporation**

During the early drying stage the droplet follows closely the ideal weight-volume-diameter relationship and retains its spherical shape. When hard crust formation on the particle surface occurs, the final size is more or less defined. The presence of air in atomized droplets has an important influence on the final shape and structure. There is always some air in the droplets depending on the aeration of the feed prior to spray drying, or during the atomization process. The composition and properties of the feed also play a role. The presence of air in of air inthe particles is usually undesirable and should be avoided (nevertheless it may be desirable for some special products and product characteristics). Depending on droplet size, the initial volume of incorporated air and its distribution (i.e. size of air bubbles) and particle temperature history, the air bubbles (and consequently also the particle) may expand, shrink, collapse, form balloons or even disintegrate. Air remaining in the droplets forms so called vacuoles in final dried particles. This is referred to as occluded, entrapped or void air. It increases the bulk volume i.e. decreases the bulk density, affects the reconstitution properties and makes packaging under inert gas more difficult.

To avoid heat degradation of the milk concentrate and expansion of air incorporated in the droplets, the constant rate of drying should be retained as long as possible with low surrounding air temperature until the critical point. The efforts to approach such conditions resulted in the development of the two-stage drying process, extended two-stage drying process (as accomplished in dryers with integrated fluid bed or belt) and three-stage drying (spray dryers with both integrated and external fluid beds) methods.

3.3.

Single-stage drying

In single-stage drying, the total removal of water takes place solely by spray drying, and the heat for evaporation is supplied by the drying air only. In other words, the milk droplets are in the dryer mixed with the hot drying air in such proportion as to achieve the required final moisture content just before particles and air are separated and leave the drying chamber. As discussed in the previous section the rate of drying, especially during the falling rate period, declines. The removal of the last portion of moisture at the end of the drying process proceeds slowly and is costly. For instance, drying of skim milk concentrate of 50% total solids, using air inlet temperature of 200°C to produce a powder with final moisture content 3.6% will in singlestage drying require an outlet temperature of 101°C whereas only 73°C, when drying to 7% moisture in the first stage of a two-stage dryer. The difference between drying to 3.6 and 7% respectively corresponds to 4.1% of total evaporation, however to achieve this evaporation in one stage 33% more heat is required.

The last phase of drying may also be harmful to powder quality due to the combination of high outlet temperature and low moisture content causing particles to be heated to relatively high levels. Therefore single-stage dryers must operate under conditions, which keep the feed concentrations have to be used, especially when drying heat sensitive and high quality products. This, of course, affects the drying economy. It is fair to mention, however, that in spite of the important advantages of two-stage drying when compared with single-stage, the application of the latter is sometimes unavoidable. Such is the case with certain thermoplastic and hygroscopic products which are too sticky at higher moisture content, thus making application of two-stage methods more difficult.

3.4.

Two-stage drying

Two-stage drying involves spray drying to a moisture content which is, for normal milk powders, about 2 - 5% higher than the required final moisture content. Subsequent fluid bed drying then removes the excess of moisture. The outlet temperature from the spray dryer is about 15 - 25°C lower than with a single-stage process. Consequently the surrounding air temperature at the critical drying stage and particle temperature are correspondingly lower as well. Therefore two-stage drying allows an increase of the inlet temperature and/or feed concentration above such values, which would simply be impossible in the single-stage process. This contributes further to economy improvement.

The completion of moisture removal is carried out by additional fluid bed drying. In this method, warm drying air is supplied gradually to meet the needs of the rate of diffusion to secure the completion of drying. The temperature of the powder, which in this case is anyhow relatively low, remains low and continues to decrease. It only begins to rise again when moisture content approaches its final value. However, no heat damage takes place under these conditions as the inlet air temperature to the fluid bed is too low to cause this.

The second drying stage conducted in the fluid bed requires of course energy input, but in spite of the specific heat consumption being relatively high, after-drying of powder by fluid bed requires only 30 - 50% of that energy, which would have been required if the same drying had been conducted in the first or spray drying stage. Thus, in comparison with single-stage drying, if all other parameters remain the same, the two-stage drying method requires at least 10% less heat. Under certain circumstances considerably more savings are possible by increasing the air inlet temperature and feed concentration. Apart from improved heat economy the plant capacity is also increased.

Two-stage drying has its limitations, but it can be applied to such products as skim milk, whole milk, pre-crystallized whey concentrates, caseinates, whey protein concentrates and similar powders. The level of moisture of the powder leaving the first drying stage is limited by the thermoplasticity of the wet powder, i.e. by its stickiness. With increasing moisture content the temperature at which the powder starts to be sticky (so-called sticking temperature) decreases. The sticking temperature is defined as the temperature at which the powder starts to stick to a warm metal surface and forms deposits and lumps. It depends on the powder composition. The components contributing to the stickiness and thereby to lowering of the sticking temperature are amorphous lactose, lactic acid, sucrose and other carbohydrates.

For skim milk and whole milk powder the moisture content of the powder leaving the spray dryer should be no higher than 7 - 8%. This is to ensure that the powder is continuously discharged under gravity into the fluid bed without lumps and that the chamber remains reasonably free of wall deposits. Any mechanical treatment of wet powder is undesirable as it will create hard lumps. Therefore the only type of drying chamber which is suitable for application of twostage drying techniques has a reasonably steep cone with a separate outlet for the drying air.

The two-stage drying techniques can be applied both for the production of non-agglomerated and agglomerated powders. Agglomeration requires special features which will be discussed later. However, even two-stage dried powder produced without these special means for agglomeration, is always slightly agglomerated and consequently has a lower bulk density. Nevertheless, agglomerates formed due to the high powder moisture content at the chamber outlet are very fragile and are broken down by pneumatic transport or by blow line transport to storage silos. After such treatment the bulk density is usually higher than that obtainable by single-stage drying.

Two-stage drying is very suitable for production of agglomerated powders by separating the non-agglomerated particles from the agglomerates (i.e. collecting the cyclone fractions and reintroducing these fine fractions, so-called fines, into the wet zone around the atomization device). The agglomeration is in this case much stronger since it takes place when the primary particles have much higher moisture content than they would under the same conditions in single-stage drying. For processing of whey the two-stage drying method is possible only with pre-crystallized whey concentrate. Pre-crystallisation transforms a great part of amorphous lactose (which is a component contributing to stickiness) into a-lactose mono-hydrate. Generally, products containing a high amount of amorphous lactose or other carbohydrates are difficult to treat by two-stage processing. It has to be decided on a case to case basis bytesting whether two- stage drying is feasible or not.

3.5.

Expansion of air bubbles during drying

Figures below shows the ideal reduction of the dimensions of an air-free droplet during the drying process. This condition, however, never occurs in practice. The presence of air in the feed and in the droplets together with the conditions of the drying process are then decisive as to (a) whether any reduction will take place at all, or (b) at what stage it will cease, or possibly (c) whether an expansion instead of shrinkage will take place.

Microphotos on Fig. 3.7 - 3.9., obtained by Scanning Electronic Microscopy (SEM) techniques show skim milk powder particles from various plants operating at various conditions, published by Písecký [51]. Fig. 3.7 illustrate particles from a high capacity single nozzle dryer operating in the single-stage drying mode using an air inlet temperature 195°C (see section 3.3). It can be seen that in spite of relatively low occluded air content in the droplets due to nozzle atomization, the droplets were exposed to expansion due to overheating. Fig. 3.7 (left) is a typical example of a blown-up particle (diameter approximately 100 μm). Expansion of some highly overheated air bubbles that are present close to the particle surface is accompanied by extensive sub-surface evaporation resulting in an explosion-type phenomena causing formation of a balloon of semi-plastic solids. Some of the small satellite particles seen in Fig. 3.7 are in fact such balloons.

Fig. 3.7 (right) shows such a balloon in higher magnification (diameter approximately 10 μm) having wall thickness of about 1 to a few microns. Such a particle will seldom survive further mechanical handling, and is thus broken down into small fragments, which, as fines, may not be collected in cyclones and therefore leave the dryer with the exhaust air. Sometimes the hard, but crispy crust cannot withstand the pressure and the particle fractures into two or more pieces as shown on Fig. 3.8. (left) Needless to say, the accompanying undesirable effect of overheating is the deterioration of solubility index and overall heat degradation. The microphotos on Fig. 3.8 (right) - 3.9 (right) show particles from a spray dryer with rotary wheel atomizer operating in the two-stage drying mode (see section 3.4). This process enables much lower surrounding air temperatures than is possible with single-stage drying.

Thus, during the critical stage, gentle drying is achieved that results in shrinkage of particles and protects solubility not only with inlet temperature 200°C (Fig. 3.8 (right)), but also with 250°C (Fig. 3.9 (left)). This effect is achieved in spite of the amount of the incorporated air in the droplets being higher from a wheel atomizer than from nozzles as shown in previous example. Fig. 3.9 (right) shows a particle from a single-stage dryer operating with steam-swept-wheel atomizer. In spite of single-stage drying, shrinkage was achieved also in this case due to the steamcreating air-free atomization environments. With no air present, no expansion takes place.

3.6.

Extended Two-stage drying

The advantages of the two-stage drying techniques regarding product quality and heat economy are obvious and therefore efforts have been made to overcome the limitations mentioned in the previous section. The critical phase of two-stage drying occurs when wet particles contact the surface of the equipment. Spray dryers with an integrated fluid bed as discussed in section 5.3 are better at handling this phase. The basic idea behind this dryer design was to operate the first drying stage at much higher moisture levels, than was previously possible with normal two-stage drying, and at the same time to avoid any contact of the wet powder with the chamber surface by introducing powder directly into the fluidized powder layer of a fluid bed placed at the bottom of the chamber.

The powder can be dried in this integrated fluid bed to the required final moisture content. In such a case the two-stage drying process is completed inside the chamber. Alternatively the product can be dried in the integrated fluid bed to a moisture content corresponding to the chamber outlet moisture of a normal two-stage drying process, and dried finally in an external fluid bed to the final moisture specification. Such a process is known as three-stage drying. The expression of extended two-stage drying was here used to emphasize that the process involves two-stage drying, but the feasible limit for moisture of the powder leaving the first stage has been extended or increased from 7 - 8% to 12 - 16% in the case of skim and whole milk powder.

Moreover, even products which are difficult to process by the normal two-stage drying techniques as baby food and high-fat products (including not only milk based, but also whey based fat-filled powders), maltodextrins etc. can be successfully produced by this method.

3.7.

Fluid bed drying

The conventional type of fluid bed used for final treatment of milk powders is a vibrating fluid bed. Dry milk products are so-called dead powders and are difficult to fluidize. Vibration is required to overcome this problem, i.e. to avoid channelling effects and to ensure true fluidization. At the beginning of the 1980’s a non-vibrating (so-called static) fluid bed was introduced for milk powder manufacture. This will be discussed later.

The first application of the vibrating fluid bed as a component in milk powder plants came about with the introduction of milk powders having high fat content (35 to 50 % and even higher). These powders were to be used as a component for dry mixing of milk replacers for feeding calves. Vibrating fluid beds overcame the difficulties experienced when trying to cool such products in pneumatic conveying systems. After introducing a fluid bed into the spray drying processing line, it was recognized that there was a distinct difference in the structure of fluid bed treated powder compared with powder from a pneumatic conveying system. The fluid bed treated powder was distinctly more coarse and free-flowing. The reason for better flowability was a partial agglomeration. This agglomeration, called primary agglomeration, is always taking place by collisions of droplets and particles of various moisture content in the atomization cloud. It is, however, a loose agglomeration which is easily broken down in an air transport system. The fluid bed treatment, however, is much gentler. The agglomerated structure is retained, resulting in better flowability and appearance.

Fluid beds exert a so-called classification effect by blowing off the smallest particles from the powder and collecting them in a cyclone. Recognizing this effect led to the second important application of vibrating fluid beds. This was the manufacture of agglomerated powders by the straight-through process whereby the cyclone fraction was recycled back to the spray dryer and introduced into the wet zone to increase the agglomeration. Initially such a process was introduced with the fluid bed acting as cooling and classifying bed only. In combination with the development of two-stage drying techniques, vibrating fluid beds were further applied for after-drying prior to cooling.

**Índice. 3.3.** Heat capacity of some milk components
Component	Heat capacity kcal/kg/°C
Non-fat milk solids	0.3
Milk fat	0.5
Water	1.0

When operating vibrating fluid beds one has to be aware that it is a not too flexible unit for treatment of powders with different properties. An important characteristic of vibrating fluid bed operation is the fluidizing velocity. This is the upwards velocity of the air calculated over the whole plate area. Fluidizing velocities in vibrating fluid bed for various products are given in Table 3.3. Velocities used in integrated (static) fluid beds range between 0.3 - 0.7 m/s for annular bed design and 0.5 - 1.5 m/s for circular design. Fluid beds for cooling operate mostly in two sections. The first section applies the ambient air and the second conditioned air i.e. air which has been cooled down to 11 - 5°C, first by condensation to remove the excess of moisture followed by reheating to reduce the relative humidity to 80% or lower depending on product properties. Vibrating fluid beds for drying operate at the temperature necessary for the required drying duty. The upper limit is normally approximately 110°C.

The drying efficiency of a vibrating fluid bed is also a function of the bed depth i.e. the height of the powder layer and product residence time. This is usually 50 - 300 mm. On the other hand effective cooling requires a low bed depth.

The operation of a fluid bed must be regularly checked to achieve the optimum performance. If the final moisture is controlled only at the discharge of the fluid bed it may happen that the powder is over-dried in the drying section, followed by picking-up of moisture in the cooling section. Therefore it is useful to check the moisture profile along the whole fluid bed length. As a routine control, it is helpful to check powder temperatures both in the drying and cooling section. This provides an indication of the moisture levels. There is temperature/humidity equilibrium between the air and powder. The water activity of common milk powders is in the range of 0.20 - 0.25. This means that these powders will begin to pick up the moisture from the cooling air when cooled below 30 - 34°C and 30 - 25°C when using cooling air of dew point 8°C and 5°C respectively.

Spray dryers with an integrated fluid bed operate with non-vibrating, so-called static fluid beds. This is because the method of drying, applied in these types of dryers, results in coarse powders of larger mean particle size. These powders are easier to fluidize. Static fluid beds operate therefore at much higher fluidizing velocities, e.g. 0.5 - 1.5 m/s and with higher bed depths, e.g. 0.3 to 1 m.

The duty of a fluid bed in both external and integrated mode is not only to evaporate the excess of moisture or to cool the powder, but also to classify the powder, i.e. to separate the small from the coarser particles. The usual aim is to blow-off fines from the agglomerates and re-agglomerate to achieve larger mean particle size and thereby better functional, mainly instant powder properties. The amount of particles, which can be separated by fluidization, depends on fluidizing velocity and particle size.

**Fig.3.10. Fall velocity of spherical particles**

Fig. 3.10 shows the free fall velocity which is the reciprocal value of fluidizing velocity or suspension velocity for particles sized between 0.01 and 10 mm (10 - 10000 μm) and having particle density 1400 and 1000 kg/m³. The air reference temperature is 80°C. This range represents roughly the particle density extremes for dried milk products. The calculations were done according to Schlünder [1], who has introduced dimensionless expressions for velocity v* and particle diameter d*, as follows:

[3,19]

…

[3,20]

…

[3,21]

…

where:

v_f fluidizing velocity m/s

d_p particle diameter m

d^* dimensionless particle diameter

ρ_a density of air kg/m3

ρ_p density of particle kg/m3

η_a air viscosity Pa.s

v^* dimensionless velocity

g gravity constant 9.81 m/s

{{result.chapterNumber}}. {{result.chapterTitle || result.title}}

{{result.abstract}}

Por favor, preencha os campos obrigatórios abaixo

Company

Job function

Country