Workshop Proceedings

This document can also be viewed in PDF form, Click Here

 

Contents

Workshop Programme

Day 1 Presentations

Day 2 Presentations

List of Participants

Workshop Programme

Day 1: Wed. 27th November

Morning Sessions
9.00-9.15 Introduction
Dr. John Fitzpatrick, University College, Cork, Ireland
Dr. Lilia Ahrné, Swedish Institute for Food and Biotechnology, Sweden

9.15-10.00 Dr.-Ing. Stefan Palzer, Nestlé, PTC Kemptthal, Switzerland
Problems while handling and processing powdered convenience foods
- What kind of further research activities are required to improve the manufacturing of culinary powders?

10.00-10.45 Dr. Peter Lillford, U. of York, formerly of Unilever, UK
The functional properties of food powders and particulates

10.45-11.00 Coffee/tea break

11.00-11.45 Dr. Gabrie Meesters, DSM Food Specialities, The Netherlands
Powder technology at DSM: Powders in applications

11.45-12.05 Dr. Elisabeth Pallai, University of Kaposvar, Kaposvar, Hungary
Food powder research in Hungary. part I - Drying of heat sensitive materials -pulps and suspensions- to produce powderlike dried food products

12.05-12.30 Professor Janos Gyenis, University of Kaposvar, Kaposvar, Hungary
Food powder research in Hungary. part II - Powder mixing, granulation and coating

12.30-14.00 Lunch

Afternoon Sessions

14.00-14.45 Professor Denis Poncelet, ENITIAA-Nantes, France
Microencapsulation of food ingredients

14.45-15.10 Ruud Verdurmen, NIZO, The Netherlands
Spray drying and particle engineering: optimisation and innnovation

15.10-15.35 Carl Hansen, Hamlet Protein, Denmark
An industrial perspective of bulk solids handling

15.35-15.50 Coffee/tea break

15.50-16.15 Dr. Paru Sellappan, Nestlé Research Center, Lausanne, Switzerland
Use of inverse gas chromatography (IGC) in food powders

16.15-16.40 Dr. Lilia Ahrné, SIK, Sweden
Work on food powders at SIK
Dr. Thomas Ohlsson, SIK, Sweden
Application of microwaves for measuring food powder properties
Professor Anne-Marie Hermansson, SIK, Sweden
Particle shape

16.40-16.45 Concluding remarks day 1

Day 2: Thurs 28th November

Presentation, Discussion and Brainstorming Sessions:
In each of these sessions, the session leader will start by giving a 30 - 35 minute presentation on the major issues highlighted so far in the project. The rest of the session will be opened up to discussion, comment and brainstorming by the all the participants.

Morning Sessions
8.30-8.45 Introduction
Dr. John Fitzpatrick, University College, Cork, Ireland
Dr. Lilia Ahrné, Swedish Institute for Food and Biotechnology, Sweden

8.45-10.00 Session 1: Quality and Safety
(leader: Dr. Peter Lillford, U. of York, formely of Unilever, UK)

10.00-10.15 Coffee/tea break

10.15-11.30 Session 2: Mixing and Agglomeration
(leader: Prof. Karl Sommer, Technical University, Muenchen, Germany)

11.30-12.45 Session 3: Added-value Technologies
(leader: Dr. Koen Dewettinck, University of Ghent, Belgium)

12.45-14.00 Lunch

Afternoon Sessions
14.00-15.15 Session 4: Storage and Transport
(leader: Richard Farnish, Wolfson Centre for Bulk Solids Handling, UK)

15.15-15.30 Coffee/tea break

15.30-16.45 Session 5: Food Powder Properties and Characterisation
(leader: Dr. Sivert Ose, POSTEC, Tel-Tek R&D Centre, Norway)

16.45-17.00 Conclusion and close of workshop

Top of Page

Presentations

Problems while handling and processing powdered convenience foods
- What kind of further research activities are required to improve the handling and processing of powder based foodstuff?

Dr.-Ing. Stefan Palzer
Nestlé PTC Kemptthal/ Zürich Switzerland

The aim of the presented key note was to point out where industry, and especially the food business is still requiring further basic research to understand processes or to solve apparent problems while handling powdered foodstuff. It was not intended to present solutions. Much more the discussed topics should be understood as input for the various institutions active in the field of powder research.
In this note especially while handling and processing culinary powders are tackled. In the dehydrated culinary business the following different product groups can be distinguished:

Powdered products:

Agglomerated and granulated products:

Tabletted and compacted powders:

The following problems can occur during powder handling and transport:
Mixes containing powder and coarse particles, which often have a different density and shape, tend to segregation. During transport and further handling the different components segregate and thus the homogeneity of the mix decreases. Due to this expensive separate filling of the sachets or tins is required and selling bigger containers for foodservice clients is difficult because the product segregates until it has been used.
The reason for this segregation is the different mobility of the particles in the bulk. However, there is still a standardised and accepted method missing to simulate this segregation behaviour depending on the real movement of the powder during production and transportation. Furthermore, a clear rule how to design the product to avoid this segregation is desired. How far can by changing of a single particle property differences in the particle mobility compensated? What role is playing a different particle density, particle size and particle shape for the segregation tendency? Is it feasible to give a quantitative rule concerning the impact of these features on the mobility of particles in the bulk? Solving these problems it would be possible to systematically design non-segregating products.
Another problem, which occurs quite often during handling of culinary powders, is linked with the Glass-Transition-Temperature. The Glass-Transition-Temperature TG is the temperature above which the powder tends (at a certain water activity aw) to get rubbery. While getting rubbery it shows a time dependent caking.

Fig. 1: Glass-Transition-Temperature in dependence of the water activity

Caking is well understood for single mixes with uniform particles. Storing these mixes at a lower temperature than TG prevents them from caking. Thus for single ingredients the TG concept is a very useful tool to avoid manufacturing problems due to caking and lumping. However, for powder mixtures containing particles with different properties it is much more difficult to apply this concept to predict the storage stability of the product. There are existing different formulas for calculating the TG of a mixture using the TGs of the single components and their quantities in the mix. However these calculations neglect the structure of the powder bulk. Important properties like the particle size distributions, the shape of the particles and the distribution of the particles in the mix are not considered. Thus these calculations are not suitable to predict the stability of powder mixtures. A suitable method for calculating the stability of powder mixes taking all parameters into account is still missing.
Some of the moisture sensitive ingredients even tend to get liquid exceeding the TG significantly in a moist atmosphere. Especially yeast and meat extracts are in these sense very sensitive ingredients. To prevent these ingredients from liquidification vapour tight and cheap coatings are required. The current coating processes are by far to expensive for standard raw materials used in the culinary industry and the coating is often not really vapour tight.
Concerning the flowability and the caking of powders an effective, food grade, water-soluble and non-E-number anti-caking agent is required. The commonly used effective silica acid powder is not water-soluble and it has to be labelled using an E?number, which is in general not well perceived by the consumer.
Beside these problems during storage and handling of powders several issues in the area of mixing and agglomeration are still not sufficiently investigated.
Problems sounding simple and trivial are still causing huge costs. For example crust formation in mixers while adding liquids and especially aqueous solutions to dry culinary mixes seems often to be quite problematic. Due to the exceeding of the Glass-Transition-Temperature moisture sensitive ingredients are getting sticky and thus they are forming a more or less stable crust on the inner side of the mixer. Since the mixers are normally cleaned dry to avoid contamination with Salmonella it is difficult to scrape the crust away. Coating the inner side of the mixer with a suitable food-grade material, which is also resistant to abrasion, could be one solution. Another approach is to design the mixing process in a special way to avoid the formation of such a crust.
A similar problem is linked to the placement of atomisers in the different available mixers. Where has the atomiser to be positioned to avoid a crust on the mixer walls and lump formation in the processed powder while injecting liquids? Such problems sound simple and in fact running various trials with each mixer type can solve them. However, practice shows that often the manufacturers of the mixers do not know where the best place for the atomiser in their apparatus is located. For food manufacturers it is difficult to test several atomiser positions because it has to be carried out in an industrial scale mixer, which thus need to be modified by cutting and welding. In mixers, running for the daily production, this is of course not feasible. Although this topic is not a very scientific one there is still a huge potential for optimisation.
The influence of the wetting of powder particles with liquids on processes like agglomeration and coating of powders is still not sufficiently investigated. The wetting process consists of two components: The spreading of single droplets on the solid surface of the particles and the penetration of droplets in inter- and intra-particle voids. Both processes seem to influence properties of the final product like the particle size, the strength of the agglomerates and the quality of a coating layer. For special material combinations a few quantitative results are published, but a general qualitative theory is still missing.
Another basic, complex and jet not solved problem is the measurement of the wetting properties of swelling and dissolving powders. Using the Washburn equation the wetting of inert particles can be described very well provided the particles are big enough.

Fig. 2: Time depending wetting of a silica acid powder with water

If the powder particles are tending to swell this effect can be considered by introducing a special swelling coefficient into the Washburn equation. Nevertheless, in most cases the powder particles are dissolving. This means the porosity of a porous particle package used for measuring the contact angle and thus the wettability of a powder is changing during the wetting process. In the mean time the composition of the liquid changes as well due to the dissolution of the powder material. Up to know there is no approach published to measure the wetting characteristics of such material systems.
The last topic linked to powder properties and their measurement is the assessment of the flow characteristics of powders under low normal stress. This product feature is one of the most relevant ones for the consumer. The flowability under low normal stress determines the dosing performed using the opened sachet, which contains the product. In addition the optical appearance of the powder which is quite important for the consumer is influenced by its flowability. Furthermore, knowing the flow characteristics of the powder under low normal stress suitable packaging solution, which would provide a satisfying dosing, could be developed. Up to know the majority of the research work has been focused on measuring the flow characteristics under high normal stress, which simulate the situation in silos or containers. First attempts to measure the flowability under low normal stress have been published recently. However, the link between these measurements and the calculation of a suitable container angle or the opening diameter of the packaging providing a proper dosing is still missing.
All these topics mentioned are certainly of interest for the food and specially the culinary industry. Finding low-cost solutions for the mentioned practical problems or delivering basic scientific knowledge concerning the measurement of powder characteristics would definitely lead to an improvement in the handling and processing of culinary powders.

Top of Page

Functional properties of food powders and particulates
Peter Lillford
University of York, UK

Issues concerning the food industry and food supply chain

The quality of powders relates to handling in the factory and in the home, and is measured either in the dry state, by its ease of use and stability; or in the manner in which it handles on rehydration and mixing.
As more people eat out of home, catering is a rapidly growing business. Powdered ingredients are convenient for storage and stability, and since the consumer never sees the food assembly process, any prejudices concerning the lower quality associated with dried ingredients is removed. Caterers therefore like big packs of powder that they often leave open, and this can give rise to problems of caking, oxidation etc.

Foods prepared from powdered ingredients are usually considered as lower quality (and therefore lower value) than fresh or frozen ingredients and products.Thus there is great difficulty in adding cost to powders, which has a huge inhibiting effect on innovation and problem solving in food powder production. Technological improvement has been limited because of cost constraints and as a result, innovation is slow or non existent . Many of the processes used today were designed for ingredients 10-20 years ago. There is a need for powder processors, ingredient people and marketing to identify routes to add more value to powders so as to overcome this resistance to innovation. Consumers may be willing to pay more for powders if they can perceive the high functionality and quality of a powder.

Food ingredient companies can tailor make their ingredients such that they can give a large variety of functionality, however these ingredients must be used exactly in the process for which they were designed and most applications have been developed purely empirically. This also tends to limit research and development since suppliers regard it as added cost, and there user customers are reluctant to change from established optimised processes even if they have only an empirical basi

Food ingredient powders can be broadly classified as follows:

Most of them will eventually be utilised in some sort of wet formulation, thus their functionality will depend on the powder particle and component interaction with water. The following sections outline some of the functions of food ingredient powders. Note that air is the cheapest food ingredient followed by water, and many ingredients are applied to entrap more air and water by utilising their aeration and gelation properties.

Functional properties of simple powders

Technical functions

Considering the multitude of functionalities, it is important to firstly know what functionality you require and then to target appropriate cost effective ingredients that will give this functionality. It is also important to make the right measurements of functionality and of factors that may affect functionality. For example, properties of simple protein powders which are often measured are the solubility and dispersibility, standardised as the NSI and PDI values, respectively. Figure 1 shows the effect on gel strength of two powders with the same PDI, but a difference of only 10% in NSI. The result shows a very significant effect on gel strength reduction by the insoluble material.

Figure 1. Effects of insoluble protein on gel strength - soya, pH 4.7, 3% NaCl.

Component stability is a major influence on functionality in applications, for example, will egg-white powder have the same functionality as the native fresh material? A lot of dehydration research has focussed on process engineering and drying efficiency, whilst a lot more work needs to focus on material science aspects and how material properties created during drying affect component stability during drying, storage/transport and final application,( usually as a wet formulation).

Physicochemical function
The factors which influence the final performance of simple powders are;

In particular, the ability of biopolymers to demix in solution,means that subsequent microstructures cannot be predicted by inspection of the formulation

Functional properties of encapsulated actives

Encapsulation allows the the stabilisation of actives (colours, flavour volatiles, acidulants,enzymes etc) during storage and transfer in the dried state, whilst allowing triggered release when required (in product manufacture or consumption). We have some understanding of the process and structural requirements of encapsulation. For example, Figure 2 shows that even with rapid formation of a surface "skin" during spray drying, the barrier can crack if outlet temperatures are too high.

Figure 2. Retention of volatile products in powder vs outlet air temperature (inlet air temperature = 350°C): (a) total; (b) surface. (From: Bhandari, B.R., et al. J. Food Sci., 57 (1), 212-221 (1992))

The chemical composition of the encapsulating layer determines efficiency of entrapment and also the dissolution and release rate when rehydration takes place (Figure 3.)

Figure 3. Effect of Surface Composition on Dissolution Rate of Spray-Dried Emulsions

So the phase behaviour of mixed solutes can either prove a problem for process control, or an opportunity for tailored release, if the microstructure of the surface layers are properly understood. The pharmaceutical industry also shares these interests and has developed sophisticated analytical techniques that the food industry could usefully "borrow". We will have even more in common as nutraceutical powders, containing encapsulated micronutrients are produced. These will require release to be triggered not in the mouth but after and during digestion.

Functional properties of particulates

The industry uses many dried components of sizes up to several millimetres.These are actually the highest added value components. Their functionality depends on visual appearance, texture, and flavour, which should be recognisable and represent the tissues from which they were derived (meat or vegetable pieces). This is extremely dependent on the drying process which itself creates the dry structure and determines the ability of the material to rehydrate and swell to its original form. Air drying normally collapses particulates irreversibly, producing very poor resultant properties. Careful freeze drying is preferred since the dried piece maintains its original volume, and the porous structure permits rapid and coplete rehydration by sequential uptake in open channels, followed by swelling of the hydrophilic matrix.
There is still scope to improve the properties of dried particulates. The microbiological stability that drying affords, can be combined with other ways of reducing water activity such as infiltration with small solutes. These combination drying processes offer considerable promise.

What next? -Learn from Nature

Some plants and even animals can dehydrate during a dry season and then rehydrate themselves into the fresh living entity when more favourable living conditions return. Whist they do this in a matter of hours rather than seconds, the advantage to the industry of stable storage and transport followed by complete recovery to the fresh state would be revolutionary. We do not yet know the "rules of the game" of anhydrobiosis, but selective gene expression is part of the key. Common phenomena in such organisms are

1. Membrane lipid composition
- no hexagonal phases must form

2. Lots of intracellular polyols
- plasticisers of H-bonds

3. Formation of expansins
- cell walls must be flexible

4. LEA Dehydrin Proteins
- unknown function, (stabilise other protein conformations?

Powder technology at DSM Food Specialties
Powders in applications

Gabrie M.H. Meesters

Introduction

Over 70% of all the products made are in a solid state. These solids, particles, powders or granules are giving a lot of problems during handling, manufacturing and in application. At DSM-Life Sciences, particle technology is performed at several R&D-locations. Specialist groups are established within the company to try to solve particle related problems. For the Food Specialties the particle related research is performed in the Netherlands, in Delft.
Typical dry product produced at large scale by DSM Food Specialties are food and feed additives like, enzymes, yeast extracts, yeasts, bacteria for dairy and meat applications, preservatives, vitamins and nutraceuticals.

Typically these food and feed products are mixtures of several components that are hygroscopic, hydrophobic, and unstable under ambient conditions, have a biological activity (micro-organisms and enzymes), oxidise easily, are a potential health hazard, etc. Therefore, these products need to be formulated in such a way that these undesired properties are masked, and formulated such that in application they do not give rise to any problems of the above-described nature.

The important parameters of our products are:
v Flow properties
v Caking properties
v Instant character
v Drying behaviour
v Low cost production
v Product stability
v Excellent Performance in Application
v Safety during production and application

Some typical areas where research is needed are:

Drying behaviour of food products
Many of the food powders are made from a liquid that is dried into a powder or agglomerate. Often salts, proteins, fats, nucleotides etc are present in these products. Due to these, drying behaviour is difficult to predict and often difficult to do in large dryers. Sticky behaviour during drying and hygroscopicity of these complex products are determined on a trial and error basis. Better understanding of the drying is needed.
Also the flow properties of these products cause many problems. Little is known about modelling and controlling these flow properties.

Safety during production and handling
Several of the food products (but the same goes for many pharmaceutical products) are potentially harmful for humans. For example enzymes, which are used in many food additives can case allergic reactions when dust is inhaled. Currently all the DSM products are granulated in a form such that dusting behaviour is low, guaranteeing safe handling. Possibly in the future stricter enzyme dust exposure limits may be imposed. This should need better analysis methods for determining enzyme dust formation of granules during handling, but also development of tools that can determine the strength of agglomerates and strength of coated particles. Table 1 shows a typical result of the spread in a measuring device used in many food industries. In this area DSM is active to promote the development of these tools through governments, but also formed alliances with University groups who develop new measuring tools. Still this needs more effort and money to be developed.

Table 1: dusting behaviour of a food powder

Coating optimization
Many of the food granules produced have an outer coating to prevent the deterioration of the product due to ambient influences like moisture and oxygen. Also these coatings need to prevent the manufacturers and end-users from exposure to these products.
These coatings are similar to the coatings used in the pharmaceutical business, as long as these are food grade. Contrary to the pharmaceutical products these coatings are often applied at minimum thickness, still ensuring safe and stable products. What we lack at this moment are testing devices that can help formulation development especially in relation to coating processes. Also quality control will benefit from these developments, since these testing devices will help to establish constancy of coating of particles and will help to check and find the of-spec. batches.
From the polymer and materials development it is possible to determine many physical properties of coatings. This should be extended to the coated particles. What are the coating physical properties when they are coated around a granule? How to determine the resistance to attrition, abrasion, fracture etc. of coated particles in relation to the physical properties of the coatings? These questions still need to be answered.
Figure 1 shows results of a newly developed machine (Repeated Impact Tester) developed together with the University of Delft, showing that different coating give different strength to the particles [1,2].

Figure 1: Attrition and breakage of particles coated with different types of polymers

Formulation of living organisms
Several of the food products contain living organisms, e.g. instant yeast (bread, wine and beer applications), lactic acid bacteria (yoghurt, cheese, meat and probiotic production). The organisms are dried in order to give them long stability during storage. The storage stability is important, since we cannot produce these cells all around the world close to the places of application. So we manufacture them at a few places dry them and transport them to many places around the world where they can be stored even longer (up to two years) before they are used.
Still a lot of research is needed in this area of drying of living microorganisms. The losses during drying, storage and hydration still need to be enhanced considerably. This will ask for still a lot of work.

Control of processes
We currently have few sensors available to monitor on-line or in-line how food processes run. By developing these sensors it will help the engineers to better steer and control the processes, giving better product quality and less recycle of product. Since many food powders are granulated, granulation prediction and modeling should be pushed to a higher level. Currently only the particle size distribution can be measured accurately on-/in-line, but parameters like density, shape, coating thickness, dustiness, strength are at least as important, but are not controlled on line yet. This is mainly due to lack of sensors [3].

Food Hygiene
In Europe the EU is sponsoring EHEDG (European Hygienic Engineering Design Group), a group developing procedures to develop equipment for use in the food area, which ensure hygienic production. Here the solids handling is taken as a special point of attention. Several food manufacturers, amongst others DSM, equipment producers and universities participate in this work. EHEDG publishes these documents produced by the groups [4].

Conclusions

We need to combine the research on food powders with application in mind. Many restrictions are caused by the application. The use of complex mixtures of food products makes it difficult to establish optimal drying, flowability prediction, control of dust, strength etc of the final products
Food powders are not just powders. They have activity. This can be microbial, enzymatic, flavour, taste etc. This means that these aspects need to be maintained as much as possible during manufacturing, creating extra problems during product development.
Due to this the products may contain actives that are potentially harmful. So proper understanding of dustiness and prediction is needed, not only during product development, but also during manufacturing and during quality control. In this area we need to acquire much knowledge still.
To ensure better production we need to build sensors that can monitor our processes, giving the engineers the possibility to control their processes better.
Food Hygiene for dry powders is given much attention, and will need this even more in the future, especially for dry product handling.

References

[1]: W.J. Beekman, G.M.H. Meesters, B. Scarlett, T. Becker, Measurement of granule attrition and fatigue in a vibrating box, Part Part. Syst. Characterisation 2002, 19, 1-7
[2]: Scarlett, B., Beekman, W.J., Meesters, G.M.H., Pitchumani, R., Particles-Their strength and weaknesses, Adv Materials Forum I Key Eng. Materials, 2002, 230(2), 203-212
[3]: P.A.L. Wauters, R.B. Jakobsen, J.D. Litster, G.M.H. Meesters, B. Scarlett, Liquid distribution as a means to describing the granule growth mechanisms, Powder Technology 2002, 123(2-3),166-177
[4]: G. Hauser, K. Mager, R.R. Maller, K. Masters, G.M.H. Meesters,W. Rumpf, G. Schleining, EHEDG document no 22, General Hygienic design criteria for the safe processing of dry particulate materials, Trends in Food Sci. Technol. 2002, 12, 296-301

Top of Page

Food Research in Hungary - Part 1

Drying of heat sensitive materials -
suspensions and pulps to produce powderlike dried food products

Elisabeth Pallai
University Kaposvár Research Institute of Chemical and Process Engineering
Veszprém, Hungary, H-8201 Veszprém, B.O.X.125,

In many food technologies, the final product must be recovered from solution or suspension. Consequently, in such cases the drying process is a substantial technological step to obtain dried powderlike products (e.g vegetables, fruits, herbs or other biologically active materials) of long shelf-life and of high quality.
The raw materials to be dried contain in many cases heat sensitive components, e.g proteins, vitamins or special active ingredients. To preseve these, the drying process should be performed at suitable low temperature. Low temperature drying of heat sensitive materials of high moisture content with acceptable drying efficiency is rather difficult.

Intensive, well controlled heat and mass transfer can be carried out in the so called Mechanically Spouted Bed (MSB) dryer, developed in the Research Institute of Chemical and Process Engineering in Veszprém,in Hungary. In this type of dryer the characteristic circulating motion of the particulate material is ensured by an inner vertical, housless conveyor screw /1/ (see Figure 1.).

Figure 1. Plot of the mechanically spouted bed dryer

The air flowing in through slots in the bottom of the dryer in tangential direction at high velocity causes intensive gas-solid contact. Due to the mechanical particle movement the circulation of the particles is independent of the flow rate of the drying air, thus, this latter can be set to a value which is optimal from point of view of drying.
By using inert charge, materials of high moisture content ( suspensions, sludges, etc.) can be advantageously dried in a single step process working continuously.

The wet material (suspension, pulps) is fed into the bed of inert particles, into the dense annular part sliding downward. The wet solid distributes on the large surface of inert particles, and forms an even, film-like layer (coating).
The drying of the filmlike coating happens in the zone characterised by turbulent particle flow in the vicinity of the gas inlet, in the bed height of a few (6…8) centimetres.
The dried coating wears off the surface as a consequence of the intensive friction in the rotation area of the inner screw, and leaves the dryer together with the air flow /2/.

In the followings several drying tasks and results will be demonstrated. Experiments were carried out in a laboratory scale MSB-dryer /3/.

Tomato powder occupies a significant place among powdered vegetables. Its raw material is usually tomato concentrate. The drying of tomato concentrate is critical because of its hygroscopic and thermoplastic characteristic. Namely food products of thermoplastic properties (e.g. tomato and apple powder, etc.), becomes deliquescent and sticky in a definite critical temperature-moisture content range.
The phenomenon of thermoplasticity of tomato is caused first of all as a consequence of its hygroscopic nature and of the "case hardening" process.
At adequate drying conditions in MSB-dryer ( optimum drying rate set by adequate coat thickness) the thermoplasticity could be avoided.
The drying curves at different coat thickness are shown in Figure 2.

Figure 2. Drying curves at different coat thickness for tomato concentrate

As it can be seen from the drying curves, that in case of film-like coating (d= 11-18µm) the drying process takes place with nearly constant rate in very short time (5-6s), giving chance to jump over the critical moisture content- temperature range.
In many cases microwave energy can be used successfully for well controlled drying of different food products, for example to avoid case hardening, and consequently the thermoplasticity phenomenon. In the Research Institute of Chemical and Process Engineering
a laboratory scale combined (spouted bed-microwave dryer) was developed. The picture of the dryer can be seen in Figure 3.

Figure 3. Combined (spouted bed-microwave dryer)

In this type of dryer heat sensitive food suspensions and pulps can be advantageously dried to produce powderlike products, working continuously. The heating process can be performed in different ways, that is by simultaneous convective and microwave heating, or successively.

On the basis of laboratory experiments industrial scale MSB-dryers were realized. For example for drying of bewery yeast suspension an MSB-dryer with a capacity of 100 kg water/h was put in operation. (Diameter of dryer was 1,0 m, bed height:1,0 m, initial moisture content: 5,0 kg water/kg db, final moisture content: 0,05 kg water/ kg db, residence time of wet material in the drying zone was 8-10 s, specific drying rate: 120-030 kg water/m2h, specific energy consumption was 3000-3500 kJ/kg water).The inactive brewery yeast suspension as the by-product of the beer production contains vitamin B and trace elements in relatively high concentration, therefore, the dried powder after tabletting can be used as roborant.

Improved construction of the MSB-dryer with inert particles

To improve the wearing process, that is to obtain finer grained powder-like dried product the spouted bed height (the effectual length of the inner screw) should be increased. However, parallel to the increase of the bed height also the pressure drop increases across the spouted bed affecting adversely the ventilation energy.
In order to increase the effectual length of the inner screw independently of the spouted bed height a modified construction was developed, that is a tube of changeable length was built in the dryer. This tube serves as a house for the elongated screw above the bed surface.
Applying this device the screw works above the bed surface as a closed conveyor improving the wearing, grinding effect. Furthermore, by the increase of the screw length above the bed surface parallel with the decrease of the spouted bed height, both improvement in wearing effect and decrease in pressure drop happen.
It could be stated that as a result of increase in wearing time (in the length of the screw above the bed surface) a decrease in particle size follows, the particles became more uniform /4/.

To recover from solution or suspension solid products having desired particle size and narrow size distribution, a fluidized-bed grinding dryer was developed.
In this continuously working equipment, particles of the final dried product are fluidized by preheated air and solution or suspension is sprayed directly onto their surfaces. Liquid evaporates, and the size of the fluidizing particles increases on effect of surface layering. To control the particle growth, roller grinders are placed at the bottom of the fluidized bed which peform selective desintegration. A controllable gap is formed between the rollers and the cylindrical wall (or air distributor).The rollers perform a circular motion and turn around their own shaft. Particles larger than the gap size break into several parts as an effect of compressing and shearing forces. Applying this dryer system, the minimum average particle size is about 200µm, but using special knife grinders, it is possible to produce less particle sizes too /5/.

Literature
/1/ T.szentmarjay, E.Pallai and A.Szalay: Drying process on inert particles in mechanically spouted
bed dryer. Drying Technology, 13(5-7), 1203-1219 (1995).
/2/ T.Szentmarjay, A.Szalay, E.Pallai, et al.: Control of drying process in mechanically spouted bed
dryer. Drying Technology, 14(3-4), 501-512 (1996).
/3/ T.Szentmarjay, E.Pallai, Zs.Regényi: Short-time drying of heat sensitive, biologically active
pulps and pastes. Drying Technology, 14(9), 2091-2115 (1996).
/4/ E.Pallai, T.Szentmarjay and E.Szijjártó: Effect of partial processes of drying in inert particles
on product quality . Drying Technology, 19(8), 2019-2032 (2001).
/5/ B.Dencs, Z.Ormós: Particle size control in fluidized bed spray-dryer during the recovery of
solids from liquids. Hung. J. of Industrial Chemistry Veszprém, Vol.21. pp. 225-231 (1993).

Top of Page

Food Research in Hungary - Part 2

Powder Mixing, Granulation and Coating

Janos Gyenis
University of Kaposvar, Research Institute of Chemical & Process Engineering, Egyetem u. 2, H-8200 Veszprem, Hungary,

Introduction

Mixing, granulation and coating are commonly used operations to produce, or to process food powders. These operations are also important for particulate solids other than foods, e.g. pharmaceuticals or chemical products. Their realization may be quite different, depending on the requirements and type of solids to be treated. Therefore, research and development in this field have crucial importance from respect of process improvement and quality enhancement. Hungary possesses very good conditions for agricultural and food production, and this was one of the main reasons that extensive research in this field was always in focal point at the Research Institute of Chemical & Process Engineering (RIChPE), Veszprem, Hungary.
Mixing studies
Studies carried out on mixing of particulate solids have been an important part of the research activity of RIChPE during the last dacades. The general aim of mixing is to achieve uniform distribution of constituents throughout the whole mass of the mix. The use of motionless or, in other words, static mixers had great potential to achieve this purpose.
Motionless mixers are widely known in other fields of material processing, especially to mix or agitate fluids, to improve direct and indirect heat and mass transfer, to enhance turbulence, dispersing or contacting materials in heterogeneous phase systems etc, but are not very often applied to mix solids. Essentially, these mixers are flow modifying devices inserted into a tube, duct or vessel, which do not move themselves. But, using pressure difference, the kinetic or potential energy of the treated materials, they can create predetermined flow patterns and/or random movements. Thus, velocity differences i.e. relative displacements of various parts of the moving material are generated, also in case of particulate solids. Splitting, shifting, shearing, rotating, accelerating, decelerating and recombining of different parts are common mechanisms in this process.
Research and development at RIChPE in this field took place in two direction: (i) Study of a special batch mixer, called Alternately Rotating Bulk Solids Mixer (ARBSM), and (ii) investigation of continuously operating gravity mixer tubes equipped with motionless mixer elements. The aims of studies were to clear up mixing mechanisms, kinetics, performance and other features of these mixers, as well as their utilization for practical tasks.
ARBM or "SysMix" Mixer shown in Figure 1 consits of two containers at both ends of a cylindrical mixer body and a mixing section in the middle, containing ordered motionless mixer grids. During operation, it is tumbling intermittently in alternating directions around a horizontal shaft, therefore food powders or other particulate materials flow down through the mixer grids with varying directions. Therefore, while mixing process is going on, segregation is hindered, due to the periodic variation of forces which otherwise might cause segregation. Operation principles and results are described in details elswhere [1-3].

Figure 1: ARBSM mixer with the enlarged view of the motionless mixer grids

Figure 2 shows typical examples for the high performance of this type of mixer: homogeneity is increasing very steeply, achieving a high equilibrium degree of mixedness without any sign of segregation, even for mixes composed of particles with very different sizes and densities [4]. Theoretical studies gave firm explanation of this excellent behaviour. Experiments were also carried out in commercial scale (from 20-30 to 2-300 kg charge).

Figure 3 shows gravity mixer tubes equipped with different types of motionless mixers.

Investigations carried out in continuously operating mixer tubes with helical mixer elements gave evidence that, in the contrary to a general belief, these mixers have excellent mixing performance for solids not only in radial but in longitudinal direction, too. Figure 4 shows residence time distributions of tracer particles measured in gravity mixer tubes with different helical mixer elements [5]. Quasi static mixers joined to each other through springs allow certain lateral and longitudinal movements during flow, preventing the solids from plugging. Such devices are well applicable for cohesive food powders, too. Capacity is changing from 0.5 to 60 tons per hour, depending on their diameter from 0.05 to 0.40 m I.D.

Power consumption of these gravity mixers are very low, about 0.1 kWh/ton. It means that apart from feeding, no energy is required if the components are available at the tube inlet.

Granulation and particle coating
Granulation is one of the most important research fields of RIChPE. Aims of studies cover process improvements and development of new products e.g. to achieve higher quality of food powders or granules. For this, needs of consumers have always guided our research activity, e.g. to produce non-perishable, free-flowing and easily soluble or dispersible instant granules with nice appearance.

Fluidized bed granulation
Food powders fluidized by hot gas (mostly air) exhibit very good heat and mass transfer between the gas and particles. It is important because this allows low temperature operation, avoiding any heat damage of food constituents. In batch granulators, very complex, highly intricated processes take place simultaneously and/or successively, often repeatedly. Homogenization, wetting of particles with binding solution sprayed onto their surfaces, agglomeration, fixation, drying (evaporation), maybe coating, and then cooling are the most important steps or processes. Material properties are generally very diverse, most of them highly influencing the whole operation. This complexness of the process needs sophisticated approach during research and designing equipment.
Characteristic mechanisms of fluidized bed granulation lead to produce loose agglomerats of primary particles with high porosity, which is favourable to accomplish instant feature.
Several new types of fluidized bed granulators have been developed at RIChPE during the last decades [6]. Special stirrers were constructed to improve particle bed motion and to diminish fluidization problems even for fine and cohesive food powders [7]. Two-phase spray nozzle (atomizer) ensures high efficiency in wetting particles uniformly with small dropplets.
Experiments generally start in a compact laboratory scale device composed of a fluidized bed granulator chamber of 0.2 m I.D., electrically heated air supply, and filter bag to prevent dust entrainment with the exhaust gas. On the basis of results and experiences obtained by an extensive research program, pilot scale tests are also carried out with bigger quantity of material in a similar granulator but having 0.4 m I.D. Experiences and measured data achieved in pilote scale equipment are then generally sufficient for safe design of process, resulting in a commercial scale equipment for the given purpose, usually with 1.2 m I.D.
The most interesting experiences during production of instantly soluble food products in fluidized bed spray granulation were related to the following tasks:
- Instant coffe granulation starting from spray dried powder as raw material. Floating up behaviour of the original powder particles (mostly tiny hollow spheres like small ping-pong balls) could be easily changed by granulation.
- Instantly dispersible and coated red pepper ("paprika") granules were produced from fine ground powder by granulation and then coating. Thin coating film well preserved the original colour, tast, flavor, preventing quality loss during storage.
As binding material, usually pure water, solution of the material same as the primary particles, or various cellulose derivatives are mainly used.

Spray granulation from liquid
At RIChPE, special equipment has been developed and patented to produce granules directly from solution or suspension [9-10]. The most important features are: (i) seeds of the material identical with the solids to be extracted from the liquid are fluidized by hot gas, (ii) solution or suspension is sprayed onto the surface of particles, (iii) growth takes place by surface layering mechanism, (iv) granule size is controlled by orbiting and rotating crushing rolls, (v) good heat and mass transfer between gas and particles allows to maintain low temperature to avoid heat damage of food components.

Roto-fluidized spray granulation and coating
To produce very compact (dense), spherical granules with smoth surface and uniform size distribution, a highly effective roto-fluid equipment has been developed at RIChPE. Main features: primary particles or seeds are forced to a circular (toroidal) path by a conical rotating air distributor plate. Particles are loosened up or fluidized by hot gas introduced through several circular gaps in the conical rotating plate. Binding or coating solution (or suspension) is sprayed onto the surface of seeds which are covered or enlarged by layering mechanism.
By this way, not only high density granules, but uniform and defect-fee coatings (film or thick layers) can be produced. Experiments start in this case directly in pilot scale with 0.4 m I.D. Several commercial scale equipment of 1.2 m I.D. are already in use by industry.

Coating of particles (granulates) in fluidized bed equipment
Coating of particles with thin film layer or size enlargement by covering them with given quantity of various materials necessary for appropriate thickness is usual operation in particle technology. Shape and size of particles, porosity, permeability, solubility, or other properties of the coating layer can be arbitrarily changed in order e.g. to achieve controlled release of kernel material. Effective methods have been developed at RIChPE to produce coatings of tailor-made properties of particles, seeds or granules. One of them is a fluidized bed spray coating with conical insert [10] seen in Figure 5. In this equipment, perfect (defect-free) film coating or controlled size enlargement can be realized by ensuring intensive particle movements with regular flow pattern, and good gas and droplet distribution.

A new research direction at RIChPE
Particles with nanostructured coating layer to create e.g. composite food powder components or ingredients will have increasing significance in food technologies too, similarly to certain pharmaceutical, cosmetic or fine-chemistry materials. From this consideration, research in this field has already started at RIChPE, cooperating with the National Institute of Resources and Environment, Tsukuba, Japan [11-12], and it seems to be a very promising new field of powder technology research at the Institute.

References
1. Gyenis, J., Arva, J., "Improvement of Mixing Rate of Solids by Motionless Mixer Grids in Alternately Revolving Mixer", Powder Handling & Processing, 1 (2), 165-171 (1989).
2. Gyenis, J., Arva, J., "Mixing Mechanism of Solids in Alternately Revolving Mixers - Part I. Change of Local Concentrations and Concentration Profiles", Powder Handling & Processing, 1 (3), 247-254 (1989).
3. Gyenis, J., Arva, J., "Mixing Mechanism of Solids in Alternately Revolving Mixers - Part II. Role of Convection and Diffusion Mechanisms", Powder Handling & Processing, 1 (4), 365-371 (1989).
4. Gyenis, J., Árva, J., Nemeth, J., "Mixing and Demixing of Non-Ideal Solid Particle Systems in Alternating Batch Mixer", Hung.J.Ind.Chem., 19 (1), 69-74 (1991).
5. Gyenis, J., "Motionless Mixers in Bulk Solids Treatments - A Review", KONA, xxx-xxx (2002). In print.
6. Ormos, Z., "Granulation and Coating. Chapter 11", in: Powder Technology and Pharmaceutical Processes. (Eds. D. Chulia, M. Deleuil, Y. Pourcelot) Elsevier, Amsterdam, 1994. pp. 359-376.
7. Ormós, Z., Pataki, K., Hajdu, R., "Granulation Process in Fluidized Bed Spray Granulator with Mechanical Stirrer", Proc. 5th Conference on Applied Chemistry, Unit Operations and Processes. Balatonfüred, September 3-7, 1989. Vol. 2. pp. 326-330.
8. Dencs, B., Ormos, Z., "Particle Formation from Solution in a Gas Fluidized Bed I-II", Powder Technology. 31, 85-91 and 93-99 (1982).
9. Hajdu, R., Ormos, Z., "Granulation in a Rotary Disk Fluidization Equipment". Proc. 5th Conference on Applied Chemistry, Unit Operations and Processes. Balatonfüred, September 3-7, 1989. Vol. 2. pp. 341-345.
10. Horvath, E., Ormos, Z., "Film Coating of Dragée Seeds by Fluidized Bed Spraying Method", Acta Pharmaceutica Technologica, 35, 95-105 (1989).
11. Endoh, S., Szepvolgyi, J. Jr., Tanimoto, T., Izumi, K., Naito, M., "Compression of the Particle Layer by High Speed Rotor in the Theta-Composer". Proc. 36th Summer Symposium on Powder Technology, Hayama, Japan, 31 July -2 August 2000, pp. 32-36.
12. Szepvolgyi, J. Jr., Endoh, S., Gyenis, J., Tardos, G.I., Dynamic Simulation of Particle motion and Surface Coating in a High Shear Mixer". Preprints of AIChE Annual Meeting, Dallas, 31 October - 5 November 1999, paper No.144c.

Top of Page

Bioencapsulation of nutraceutics for food applications

Denis Poncelet and Ernest Teunou
ENITIAA, BP 82225, 44322 Nantes cedex 3, France

In western countries, feeding habits have changed a lot during the last half-century. We have switched from home made to industrially processed food. Very often processing destroys some fragile elements. The abundance leads to a consumption of too rich foods, however generally not well equilibrated. On other hand, people are now considering foods as a health vector. Additives have to be added to the food to provide
additional nutritional values or to re-equilibrate the composition of the food. From this consideration, came the concept of nutraceutics or functional foods. A given food can be regarded as "functional" if it is satisfactorily demonstrated to affect beneficially one or more target functions in the body, beyond adequate nutritional effects in a way, which is relevant to either an improved state of health and well-being, or reduction of risk of disease. Nutraceutics include a large range of molecules types such as:
Ø inorganic and organic salts
Ø active molecules (vitamins, fatty acid ..)
Ø polymers (fibres, prebiotic, proteins)
Ø enzymes
Ø plants and yeast extracts
Ø biological cells (probiotics)

One could observe that it covers a very broad range of components. Their requirement varies as a function of the age, country, feeding and activity habits. They may interact with other food ingredients and need to be protected from them. This protection is also to avoid unwanted taste effects. In most cases, they are fragile and need protection during storage, processing and often during gastro-transit. They are added at very low concentration in food and required a form easily and reproducibly dispersible in a large system. They have to be release at the right place in body and even preformed to increase their absorption in the intestine.

One of the best solutions to these problems is microencapsulation. However, one must understand that encapsulation is one of the steps in the whole processing of a specific ingredient. All other steps will influence the microencapsulation efficiency and affect the integrity of the encapsulated materials but also the selection of the encapsulation method.

Let's take an example: the probiotics and even more specifically bifido bacteria which is recognized to have very important healthy effect such as cancer prevention, better food digestibility, protection against pathogens. To target such microorganism to the body, many steps are involved (Figure 1). The bifido species have to be selected not only for their nutritional quality but also for their good resistance to treatments and their capacity to be produced in large quantity at low cost. While produced, the bacteria may need to be placed in a suitable physiological stage to optimize their viability during drying, storage and regeneration. Futhermore to protect it and to provide adequate environment during reactivation of the cells, inoculum has to be formulated with different ingredients like cryo- or drying protectants, substrates or prebiotics (molecules that affect the capacity of colonization in the intestine). Then come a stage of drying either by spray drying or lyophilisation, and finally microencapsulation (these stages can sometimes be done in parallel). However, the story is not finished, even protected by encapsulation, the probiotics may be stored and processed in adequate conditions. One could not expect the encapsulated cells to be fully protected against high temperature or moisture. The stage of rehydratation is also critical. Most scientists agree now that this stage is as much important as dehydration step for cell viability. The technology used for encapsulation will affect strongly the rehydratation. Ingestion, gastro-transit, absorption and colonization are the final steps and are also strongly affected by the encapsulation procedure.

If you want to be successful in using an encapsulation process, you must take in consideration the whole "food chain". Even if looks simple for some ingredients (like salts), you may miss your objective if you dismiss this rule. On top, you will need to consider cost, technological and legal aspects. Despite all this difficulties, the increasing interest for and market development of encapsulated food ingredients prove that this technology is successful both technologically and financially to provide higher quality of functional foods to customers.

Figure 1 : Probiotic processing

There exists a wide range of microencapsulation methods but very few are really suitable for functional foods in regards to their biocompatibility, cost or protective actions. As functional foods are at the interface between pharmacy and food industries, the most usual technologies are often from one of these fields or a combination. Table 1 gives a summary of different methods.

Tabletting

Tabletting is a technology from the pharmaceutical field. It produces quite large capsules (a few mm) and, in many case, good protection. It is however difficult to figure out how to involve such technology in food applications. Sizes are obviously too large and this brings the question of the adequate microcapsule diameter for food applications. In fact, no real data exist about this and it may depend on the application stage. In dry food, the size must be related to powder grain size to avoid segregation. While suspended in liquid the size must be very small to avoid settling. However, density and particle surface tension play also an important role in reducing segregation and settling problem. Capsule detection in the mouth is obviously connected to their size but also probably to their surface roughness and "elasticity". On other hand, the lower is the capsule diameter, the more difficult is the control of the size distribution, the lower is the productivity as well as the protective effect, and finally the higher is the cost. Engineers are then advised to define the largest size acceptable for their application and work with this size.

Coacervation
Coacervation consists generally in dispersing an oil phase in a water polymer solution. The polymer is then demixed by addition of salt, acid or by changing the temperature. The resulting "coacervates" accumulate at the oil droplet interface and form a membrane. Capsules are then separated, cross-linked with glutaraldehyde and dried. Different types of polymers systems are proposed to form such capsules but the most usual is a mixture of gelatin and arabic gun. This technology allows to form capsules containing a liquid hydrophobic core and it is quite unique in this way as potential food grade process. However, the cost of the process, the use of gelatin and cross-linker have strongly reduced the interest for such process, especially in Europe. Much research work is being undertaking to use different wall materials (such as vegetable proteins) to reduce the cost. Coacervation may come back in the future as a useful method of encapsulation for nutraceutics.

Emulsions, double emulsions and multilamellar systems (liposomes)
While dispersing a hydrophobic material in food matrices (generally hydrophilic), one could consider emulsion as an encapsulation method if it is stable enough. By using double emulsions, even hydrophilic material could be entrapped in oily phase. If amphyphilic material is used, diverse configuration of multilamellar system could be realized where hydrophobic material is encapsulated in bilayer lamellas and the hydrophilic material in the intermediate layer. These technologies could be conducted in very soft conditions allowing very gentle encapsulation. However, most of these systems are not stable as true microcapsules and must be often considered as a vector for molecules than as a true encapsulation. To stabilize such system, emulsions and multilamellar systems could be included in hydrogel matrices for example. They will contribute to the controlled release of the active ingredient but also in many cases to its absorption. They could also contribute to maintain the integrity of enzymes and even enhance the activity of hydrophobic ones.

Spray cooling
An alternative to encapsulation by spray drying is to work by spraying a melted material and solidifying the droplets by cooling. The range of matrices available for such processes is more limited and consists often in hydrophobic material (such as fatty acid). Dry hydrophilic material offers generally a better protection to oxygen and is more compatible with many food powders (flour). Its protecting property against moisture may be a function of the relative humidity. At low moisture, hydrophilic material may be a good barrier but obviously at high moisture, hydrophobic barrier is needed. One great advantage of the encapsulation by spray cooling is the high productivity. One hundred percent of the sprayed solution form the particles, while with drying process it may represent from a few percents to maximum 40 %. Cooling process is then a lot faster than drying. Cooling reduces slightly the process cost as it requires limited amount of energy in regard to evaporation associated to higher volumetric productivity.

Film coating
While having a fine dry form (either the nutraceutics is initially solid, or has been previously encapsulated), it may be of interest to coat its surface. This coating could provide new surface properties, protection and controlled release. The coating could be realized in two main reactors: In pan coating, material is placed in a rotating drum ( or pan) and a coating solution is sprayed on the particles. To avoid agglomeration, the particle kinetic energy provided by the drum must be higher than the interparticle sticky energy. Such condition is easily performed only with large particles (larger than 1 mm). For small particles, fluidised bed could be used. Particles are placed in an upward air-flow reactor and then suspended in front of the spray nozzle. The most simple arrangement is a reactor with the spray nozzle placed on top of the reactor (top spray). It allows large particle loading but the coating is often not perfect and it exists some risk of agglomeration. To reduce these problems, a cylinder is placed at the center of the reactor, the air flow rate is higher in this zone promoting circulation of the particle (moving up in the central zone and downward in the outer zone). Spray nozzle is then introduced in the bottom of the reactor. This process is the so called Wurster process. The circulation of the particles ensures a better coating. Finally, a combination of the fluidisation and the rotating effects could be obtained by introducing a rotating disk at the bottom of the reactor, fluidising air is only provided at the periphery. Such system leads to a good circulation of the particles and to final spherical particles ( "spheronisation"). The spray is provided directly in the bed and promotes a very good coating properties. However, this process is incompatible with fragile particles.

Figure 2 : Film coating process (courtesy of Glatt Pharmatech, Germany)

Hydrogel bead entrapment
While looking simply for an immobilization method for cells or large molecules, hydrogel beads could constitute a cheap and gentle method. This could even be extended to hydrophobic molecules after its emulsification in the pre-gel solution. There exists a large range of polysaccharides or proteins, which can form a gel in the presence of ions (alginate), by decreasing (K-carrageenane) or increasing the temperature (konjac). The release could be obtained by sequestration (using some ions), or changing the temperature. Obviously, such encapsulation will not offer high potential of protection but could fit with some specific applications. To increase the retention and protection capacity, one could produce high dry matter beads by adding to pre-gel solution some filler such as inert powder or low viscous polymer (such as arabic gun). The resulting beads could even be dried at moderate temperature. This approach is not really developed today but could figure as a very gentle technique for encapsulation of very fragile material.

Conclusion and perspectives.
The present overview does not try to cover all the technologies available to encapsulate nutraceutics. There exists many variants to the presented ones. While selecting a method, the researcher or the engineer will have to take in account many aspects as reported in the beginning of this text.

To provide more information to the reader, you can visit the web site http://BRG.enitiaa-nantes.fr where you can find the slides associated to this presentation which are is available at:
(http://BRG.enitiaa-nantes.fr/Documents/neutraceutics) or contact the authors poncelet@enitiaa-nantes.fr
This work includes many links to web site related to the different technologies described here.

Top of Page

Spray drying and particle engineering:
optimisation and innovation

Ruud Verdurmen
NIZO food research, Ede, The Netherlands

NIZO food research
NIZO food research is an independent industrial research and knowledge organisation that carries out innovative contract research for companies active in the food and biotechnology industries. The aim is to develop applications in the fields of texture, flavour and health and to optimise and innovate production processes focussing on quality, safety and processing technology. To achieve this, NIZO food research is organised in multidisciplinary teams of highly educated and application driven professionals.

NIZO food research supplies solutions for companies based on a thorough knowledge of:
- Microbial systems
- Biopolymer systems
- Process systems
NIZO food research is part of various national and international research networks and participates in the Wageningen Centre for Food Sciences.

Main challenges in the production of powders
The following main challenges can be identified when looking to the production processes of food powders:

• The development of food powders with a higher added value, e.g. powders with a specific functionality (e.g. purity, nutritional value) or improved rehydration properties;
• Reduced time-to-market: in general the development time of new products/processes (the trajectory from idea to application) is urged to be shorter and shorter in order to be competitive. As a result the knowledge intensity of such trajectories needs to be higher and higher. Also having multi-product installations with a short change-over time can reduce the time-to-market;
• Reduction of processing costs. There usually still is adequate room for further optimisation of existing drying processes, for example by:
• Maximising the capacity of existing installations
• Reduction of fouling and thereby reducing the costs of product losses
• Reduction of energy consumption
• Using on-line product quality control systems.
  It is the experience of NIZO food research that predictive computer models are very effective tools to reduce time-to-market and to reduce processing costs.

General production process of powders by spray drying
A large portion of the food powders is produced by spray drying processes. The removal of water usually takes place in two stages, see Figure 1.

Figure 1: Schematic presentation of the production process of food powders

The first stage is concentration by vacuum evaporation and the second stage is drying; 90 % of the water is removed in the evaporator and only 9-10 % in the spray dryer when calculating the amount of water removal per dry mass. However, the energy required per kg water evaporated in the dryer is about 15 times the energy required per kg water removed in the evaporator, see also Table 1.

Table 1: Typical figures for the conversion from milk to milk powder.

Spray drying is a relatively gentle drying process that has replaced the cheaper but also the more product-denaturing drum dryers. Moreover spray drying makes it possible to manufacture powder qualities for different applications and quality standards.

In Figure 2 the scheme of a multi-stage-dryer is shown. In practice a spray dryer can consists of one, two or three stages. Multi-stage drying increases the thermal efficiency of the drying process, produces agglomerated powder with good rehydration properties and prevents overheating of powder particles. In the first stage, the preheated product (< 100 °C) is sprayed by atomisation into a chamber filled with circulating hot air. The inlet temperature of the air is normally 150-250 °C. By atomisation the concentrate is converted into droplets of 10-200 mm. In the industry two atomisation systems are used: stationary pressure nozzle and rotating atomisers. The droplets are flowing in the tower and adsorb heat necessary for evaporating of the moisture. The moisture is removed by the hot air. Depending on the dimensions of the tower the residence time of the powder particles is in the order of 5-30 seconds. The dried powder falls to the bottom of the dryer and is transported to the next drying stage or to a packaging system. The exhaust air is removed through an outlet duct and passes through cyclones and filters where small powder particles (fines) are removed. The fines can be recycled to the top of the dryer or to other drying stages. The result is an agglomerated powder.

Figure 2: Schematic representation of industrial configurations of spray-driers
(single-, two- and three-stage)

To obtain a high-quality powder, a constant dry matter content in the concentrate produced in the evaporator preceding the drying process is necessary. The occurrence of changes in dry matter content in the feed to the dryer is one of the major sources of disturbance in the drying process. It is also advantageous to remove as much water as possible at the evaporation stage from an energy-saving point of view. In practice however, due to variations that occur in dry matter content of the concentrate as a consequence of variations in feed and process variables, the set-point for this dry matter content is often lower than theoretically possible. This in order to reduce the risk too high a viscosity of the concentrate. Less variation in dry matter content of the concentrate enables a higher set-point and thus also improves the energy efficiency of the powder production process. When using conventional control technology, such as single-loop proportional-integral-derivative (PID) controllers, the long time delay from input (e.g. flow or dry matter content of milk fed to the evaporator) to output (e.g. total solids content of concentrate by controlling the steam supply) will result in a relatively long period of off-spec concentrate. Modern design methods for multivariable control make it possible to design compensators that reduce or eliminate the off-spec period. For the design of such a multivariable control system one should determine the dynamic behaviour of the evaporator involved. This can be done either by using a physical model simulating the dynamic behaviour or by carrying out step-response measurements on the actual evaporator and using system identification techniques to draw up a black-box model. The first approach is more flexible and robust for handling changes in the design and process operation. The advantage of the latter approach is that it requires less detailed knowledge about the design of the evaporator. Also in drying processes there is a trend to use more and more predictive models in the control strategy. The main issue for the automatic control of spray dryers is to achieve a reduced variation in the moisture content of the powder, enabling a higher setpoint for the moisture content, which strongly reduces the operating costs.
An industrial case is the design and implementation of a feed-forward control system for a four-stage falling-film evaporator with thermal vapour recompression. This control system contains a feed-forward compensation for dry matter content of the feed to the evaporator (e.g. by measuring the density of milk using an in-line sensor), a feed-forward compensation for flow to the evaporator and a feed-back control system using the measured density (in-line sensor) of the concentrate. The dry matter content of the concentrate is controlled by adjusting the steam supply to the evaporator. Based on step-response measurement the specific control algorithm is designed and implemented in the existing programmable logic controller (PLC) of the evaporator.
This new feed-forward control system has decreased the standard deviation in dry matter content of the concentrate from 0.31 % to 0.19 % (w/w%). Compared to a simple control system it is now possible to increase the set-point of the dry matter content of the concentrate by at least 0.7 %, resulting in an annual energy saving of Euro 10 000 based on a nominal capacity of 30 m3 milk per hour. The capacity of the evaporator can now easily be adjusted to the capacity of the spray dryer without large variations in dry matter content. It is estimated that as a result of this, the set-point for the moisture content of powder can also be increased by about 0.07 %, which will result in an annual energy saving of Euro 50 000.

Predictive models for spray drying
Two different predictive models for spray drying of dairy products have been developed, implemented and industrially validated by NIZO food research. These two models, DrySPEC2 and DrySim, will be described below. The development of predictive models is an ongoing process; at this moment a consortium of universities and companies (with NIZO food research as the co-ordinator) is developing a model to predict the agglomeration in spray drying installations.

DrySPEC2
The first drying model that was developed by NIZO food research is DrySPEC2 (DRYer System for Property and Energy Control). This computer model describes the relation between the processing conditions of the drying process, energy consumption and the properties of the powder produced for a two-stage dryer. The purpose of this model is to establish the process conditions that ensure optimal exploitation of the capabilities of existing drying installations with regard to energy consumption and the powder properties. This model assumes a near-equilibrium state of water vapour pressure between powder and outlet air, which eliminates the need for a detailed description of heat and mass transfer phenomena during the drying process. In Figure 3 a screenshot of DrySPEC2 is shown. The model is integrated in a user-friendly interface in which other software modules can also be accessed. The standard set-up of this model is for a two-stage dryer (spray chamber and fluid bed dryer), but it has also been adapted other types of dryers, e.g. containing an internal fluid bed.
DrySPEC2 has successfully been implemented for the production of dairy products such as skim milk, whole milk and whey permeate. The results obtained in increasing the earning capacity of industrial spray dryers are: up to 20 % increase of capacity, limiting the deviation in moisture content to lower than 0.05 % (for example by adjusting the process to variations in total solids content of the feed or moisture content of inlet air) and an annual energy reduction potential of about 250 000 m3 natural gas per installation.

DrySim
In order to simulate the drying process in more detail, it is necessary to gain insight into the flow pattern, local temperature and local moisture content of the air and the temperature-time history of drying particles . The flow pattern of air depends on the geometry of the dryer and the location and design of the air inlet and air outlet channels. The trajectories followed by the (drying) particles depend not only on the air-flow pattern but also on the position and method of atomisation. At NIZO food research the drying model DrySim was developed as a tailor-made simulation program for spray dryers, making use of computational fluid dynamics (CFD) techniques. DrySim is a two-dimensional simulation model of a spray dryer. It calculates the flow pattern, temperature and moisture content of air, the trajectories of the atomised particles and the drying behaviour of individual particles, see also Figure 4. Sub-models for the formation of insoluble material or for describing the stickiness of particles have been added to DrySim. DrySim has proven to be an effective tool in giving indications of how to adapt industrial dryers, for example to obtain a better product quality, a higher capacity or to reduce fouling.

Figure 3: Screenshot of NIZO DrySPEC 2 with user-friendly interface.

Figure 4: Computational simulation (NIZO Dry-Sim) of an industrial spray dryer.

Product and process innovation

The process optimisations as described above are mainly targeted at the supply-chain and the production of commodities. In contrast, innovation programmes focus at the development of new products and/or dedicated processes for:

• Food/feed specialities (products with high added value)
• Ingredient industry
• Biotech/pharmaceutical industry.

It turns out that product and process innovations:

• Are mainly driven by functionality
• Require a multidisciplinary approach (e.g. material science, food science, process technology)
• Require a wide variety of analysis equipment, process equipment and application tests
  For these reasons it is needed to co-operate (pooling of knowledge, know-how and tools, partnership) between universities, technology centres and producers to achieve real break-throughs.

Some examples of the type of work carried out by NIZO food research in this area are:

• Micro-encapsulation of flavours using food grade materials
• Mild spray drying of micro-organisms (e.g. probiotics)
• Fluid bed coating using food grade polymers to provide a tight barrier

The encountered knowledge barrier is the lack of fundamental knowledge on the:

• Behaviour of biopolymers during encapsulation and coating processes
• Behaviour of food grade materials for controlled release purposes
• Behaviour of micro-organisms during drying

Summarising remarks

Optimisation focuses mainly on the production of commodities. Main customers are production managers and technologists. There is still a long way to go for full implementation, although many solutions are already developed. Knowledge transfer and budget are the main barriers.

Innovation is mainly initiated by marketing and R&D. For real break-throughs co-operation between disciplines and parties (universities, technology centres and producers) is essential to reduce investments in equipment and personnel.

Process intensification has not been mentioned in the presentation, but is also an interesting and promising development. In comprises a smart combination of processes, also for the production of food powder. Examples are co-spray drying, total filters, steam-cooking nozzles and straight-through agglomeration.

Top of Page

An Industrial Perspective of Bulk Solids Handling
Carl Hansen
HAMLET PROTEIN A/S, Saturnvej 51, 8700 Horsens, Denmark

This presentation is based upon experiences from daily work and incidents in a factory. The production of this factory is characterized by:

1. Production of special soya protein,
2. Hygienic production,
3. The finished goods are two fine powders,
4. One product is a little cohesive,
5. The product can self-ignite,
6. Process industry with continuous production 24 hours a day all year round.

An important objective in a production facility is to optimise processes and get control over them, and some issues connect with that will be mentioned in the following.

Comminution
Comminution is a very common unit operation in many powder handling productions. The operation is an intense contact between product and comminution tool, and the result will eventually cause wear. There is a real need for materials for comminution tool, which can resist both erosive and abrasive wear. There already exist materials, which resist wear for an extensive period of time, but they are expensive.

The comminution process is not just breaking the particles into smaller particles, but it also creates functionality, which is connected with particle size, shape and surface.

Comminution is a process using much energy, and a lot of it is used to warm up the product in stead of breaking it.

Sampling
The importance of sampling cannot be overestimated. Poor sampling is a major reason for poor analysis. In order to meet its purpose the sample must first of all be representative, but before the sample can be analysed it must be divided correctly so that a very small sample will represent the whole lot. Such a lot can be as much as a shipload.

Especially in pneumatic conveying systems, the sampling is very difficult, but even in free falling systems it can be difficult.

Sampling is also important in the way that it can be decisive for price and even as evidence in a lawsuit.