One-stage drying is defined as the spray drying process where the product is dried to the final moisture content in the spray drying chamber, see Fig. 67. However, the fundamental theory about the droplet formation and the evaporation of the initial moisture is the same in this and the following processes and therefore discussed here.

Fig. 67 Conventional spray dryer with pneumatic conveying system (SDP)

The initial velocity of the droplets from the rotary atomizer is about 150 m/sec. Most of the drying takes place while the droplets are decelerated by their friction to the air. Droplets with a diameter of 100 microns have a deceleration path of less than 1 m, and for droplets with a 10 micron diameter it is only a few centimetres. The main temperature drop of the drying air, due to the evaporation of the water from the concentrate, takes place during this period. An enormous heat and mass transfer therefore takes place in the particles during an extremely short period of time, and the product quality may be seriously harmed, if the factors promoting degradation are not known, or are disregarded.

During the removal of water from the droplets a considerable reduction in weight, volume, and diameter of the particle takes place. Under ideal drying conditions the weight will decrease to about 50%, the volume to about 40%, and the diameter to about 75% of the created droplet from the atomizer. See Fig. 68.

Fig. 68 Weight, volume and diameter decrease of a droplet during ideal drying conditions

However, the ideal droplet creation and drying technique have not yet been developed. There will always be some incorporation of air in the concentrate during pumping from the evaporator, and especially when the concentrate is pumped into the feed tank due to splashing. But also during the atomization a lot of air is incorporated into the concentrate in the rotary atomizer, where the wheel besides atomizing the concentrate is acting as a fan sucking in air and whipping it into the concentrate. Specially designed wheels will, however, counteract the incorporation of air in the concentrate. In the curved vane wheel (the so-called high bulk density wheel), see Fig. 69, the air is partly separated from the concentrate again due to the centrifugal force, whereas in the steam-swept wheel, see Fig. 70, the problem is partly overcome by replacing the liquid/air interface with a liquid/steam interface. It was generally believed that the nozzles during the atomization incorporated no or very little air into the concentrate. However, it has been found that some air incorporation takes place during the very early stage of atomization, both outside and inside the spray cone due to the air friction prior to the droplet formation. The higher the capacity of the nozzle (kg/h) the more air will be whipped into the concentrate.

Fig. 69 Wheel with curved vanes for powder with high bulk desity

Fig. 70 Steam-swept wheel

The ability of a concentrate to incorporate air (or its foaming ability) is depending on the composition, temperature and the solids content. It has been found that concentrate with a low solids content has an extensive foaming ability which even increases with the temperature. Concentrate with a high solids content has considerably lower foaming ability which is even further reduced by increasing the temperature, see Fig. 71. Generally the foaming ability is less in whole milk concentrate than in skim milk concentrate. Determination of air in concentrate is described on page 189.

Fig. 71 Foaming ability af skim milk concentrate

The amount of air in the droplets (present in form of small air bubbles) is therefore one of the decisive factors as to how far the shrinkage will continue during the drying. Another factor even more important is the drying conditions, i.e. the surrounding air temperature. As mentioned, a lot of heat has to be transferred from the drying air to the droplets and much water vapour the other way. Therefore, there is a temperature and concentration gradient in the particle, and the whole process becomes very complex and not fully understood. Droplets of pure water (water activity 100%) will, when exposed to air at a higher temperature, evaporate keeping wet bulb temperature until completely evaporated, while solids containing products dried to the extreme (i.e. with a water activity approaching zero) are heated to the temperature of the surrounding air at the end of the drying, which in a spray dryer means the temperature of the outgoing air. See Fig. 72.

Fig. 72 Temperature Development

Not only from the centre to the surface is there a concentration gradient, but also from one point of the surface to another resulting in different water concentrations and thus different temperatures between different regions on the surface. The overall gradient intensity is bigger, the bigger the particle diameter, due to the smaller surface/mass ratio. Thus small particles dry in a more uniform way.

During the drying the solids content naturally increases due to the removal of water - and so does the viscosity and surface tension. This means that the diffusion coefficient, i.e. the water-vapour diffusion/time and area, becomes smaller and overheating occurs due to the slower evaporation rate. In extreme cases the so-called case hardening will take place, which is the formation of a hard crust on the surface through which the remaining water-vapour or occluded air will diffuse very slowly. If case hardening occurs, it is usually at a residual moisture content of 10-30% in the particle, at which stage the proteins, especially the caseins, are very sensitive to heat and easily denature resulting in a powder with poor solubility properties. Moreover, the amorphous lactose will become hard and almost impenetrable to water vapour, and the particle temperature increases further as the evaporation rate, i.e. diffusion coefficient, approaches zero.

As there will be more water vapour and air bubbles in the particle this will now get superheated, if the surrounding air temperature is high enough resulting in the vapour and air to expand. The pressure will increase, and the particle will blow up to a completely round ball with a smooth surface, see Fig. 73. The particle will have a lot of vacuoles inside, see Fig. 74. If the surrounding air temperature is high enough the particle may even explode, but even if it does not, the particle will have a very thin crust, about 1 micron, and it will not survive the mechanical treatment in the cyclones or in the conveying system and thus leave the dryer with the exhaust air. See Fig. 75.

Fig. 73 Typical paricle from one-stage drying

Fig. 74 Spray particle. One-stage drying

Fig. 75 overheaded particle. One-stage drying

If there is only a small content of air bubbles in the particle the expansion will, in spite of the overheating, not be too extensive. The overheating as a result of case hardening will, however, have a detrimental effect on the caseins resulting in bad solubility.

If the surrounding temperature, i.e. the outlet temperature, is kept low during the drying, the particle temperature will be equally low.

The outlet temperature is determined by many factors of which the most important ones are:

• Moisture content in the final powder
• Temperature and moisture content of the drying air
• Solids content in the concentrate
• Atomization
• Viscosity of the concentrate

Moisture Content in the Final Powder

The first and foremost factor is the moisture content in the final powder. The lower the residual moisture content wanted, the lower the relative humidity in the outlet air, and that means higher outlet temperature and with that higher particle temperature.

Temperature and Moisture Content of the Drying Air

As the moisture content is in direct relation to the relative humidity of the outgoing air, an increase of the inlet air will necessitate a slight increase in the outlet air due to the higher amount of moisture in the air resulting from the increased evaporation. Also the initial moisture content in the drying air plays a big role, and if that is high the outlet temperature has to be increased to compensate for the extra moisture.

Solids Content in the Concentrate

An increase in the solids content will require an increase in the outlet temperature, as the evaporation becomes slower (average diffusion coefficient smaller) and a bigger temperature difference (driving force) between the particle and surrounding air is necessary.

Atomization

Any attempt to improve the atomization and create a finer spray will result in a lower outlet temperature, as the specific surface/mass ratio of the particles becomes bigger. The evaporation will therefore be easier and a smaller driving force is required.

Viscosity of the Concentrate

The atomization is influenced by the viscosity. The viscosity increases with increased content of proteins, crystallized lactose and overall solids content. Heating the concentrate (beware of age-thickening) and increasing the atomizer speed or nozzle pressure can remedy the problem.

The overall drying efficiency is expressed in the following approximated formula:  

ξ =Ti - To / Ti - Ta    (17)

where: 
Ti =  air inlet temperature
To =  air outlet temperature
Ta = ambient temperature

It is thus obvious that the only possibility of increasing the efficiency of spray drying operation is by increasing ambient temperature by preheating (see page 169), f.inst. using condensate from the evaporator, or by increasing the inlet temperature or decreasing the outlet temperature.

The relation ξ is at the same time a good indication of the dryer performance, as the outlet temperature is determined by the residual moisture content which has to fulfil certain standards. A high outlet temperature will indicate that the drying air is not utilized in an optimal way due to various reasons such as bad atomization, bad air distribution, high viscosity, etc.

The ξ will in normal spray dryers operated on skim milk (Ti = 200ºC, To = 95ºC) be around 0.56.

The drying technology discussed so far has been related to a plant with pneumatic conveying and cooling system, where the powder when leaving the base of the chamber is dried to the wanted moisture content. The powder will at this stage be warm and consist of particles stuck together with very weak bindings in big loose agglomerates due to the primary agglomeration taking place in the atomizer cloud, where particles of different diameter will obtain different speed and therefore collide. However, when passed through the pneumatic conveying system, where the agglomerates are exposed to mechanical forces, the powder will break up in single particles. This kind of powder, see Fig. 76, can be characterized as follows:

• Single particles
• High bulk density
• Dusty, if it is skim milk powder
• Non-instant