STICKINESS DURING SPRAY DRYING

Dr Bhesh BhandariSPRAY DRYING RESEARCH GROUPSchool of Land and Food Sciences &

School of Engineering The University of Queensland

AUSTRALIA

Spray Drying Research Group

• Prediction of glass transition temperature of model mixtures- relevant to sugar-rich foods such as fruit juice, honey

• Design of static and dynamic stickiness testing devices for food powders

• In-situ stickiness measurement of droplets• Drying kinetics and dryer design for sticky

materials

Current research activities

My other research activities• Structural relaxation of dried food materials

• Application of ultrasound in food processing- meat tenderisation, homogenisation, encapsulation

• Development of microencapsulation process for food flavours, probiotics, vitamins…

• Water activity prediction (flavour powders, IMF)

• Extrusion and stability of microencapsulated flavours

• Ultrasound spectroscopy in non-invasive characterisation of food materials (gelation, composition, texture etc..)

Spray drying

• Most common process to convert liquid to solid

• Large throughput- capacity several tonnes per hour- (15 tonnes per hour- New Zealand)

• Produce free flowing, fine to granulated powders

• Low thermal effect on materials during drying

• Versatile in use- ceramic or milk

A typical two-stage

spray dryer

Source: Dairy Processing Handbook. Published by Tetra Pak Processing Systems AB, S-221 86 Lund, Sweden. pg. 369.

Source: Dairy Processing Handbook. Published by Tetra Pak Processing Systems AB, S-221 86 Lund, Sweden. pg. 370.

FILTERMAT DRYER

Stickiness issues during spray drying

• Stickiness on the drier wall (spray drying)

• Wet and plastic appearance

• Agglomeration and clumping in packing container

• Operational problems

• Losses

Stickyproduct

Hot air

Non-sticky product

Products exhibiting stickiness during drying

• Products with high amount of sugars or organic acids– Fruit juices/pieces/purees/leathers– Honey– Molasses– Whey (acid or sweet)– High DE maltodextrins (DE>30)– Pure sugars- glucose, sucrose, fructose– High acid foods

• High fat foods

Major factors causing stickiness

• High hygroscopicity

• High solubility

• Low melting point temperature

• Low glass transition temperature

(related to thermoplasticity)

Glass Transition Approach

•Recent approach to describe stickiness

•Applied to spray drying

Physical properties of sugars and stickiness

behaviour

Sugars Hygroscopicity Melting point Approx solubility

in H20

Tg Stickiness

(relative) (oC) 60oC (%,w/w) (oC) (relative)

Lactose + 223 35 101 +

Maltose ++ 165 52 87 ++

Sucrose +++ 186 71 62 +++

Glucose +++++ 146 72 31 +++++

Fructose ++++++ 105 89 5 ++++++

What is a glass transition?

– Amorphous• non-aligned molecular structure• very hygroscopic• go through glass transition• predominant in dried food

– Crystalline• aligned molecular structure• non hygroscopic• no glass transition

Physical states of dried solid materials

Semi-crystalline solid Liquid solution

GrindingExtrusion cookingThermal melting & cooling

Rapid waterremoval- drying

Rapid cooling below Tg

water <-135oChoney <-45oC

Amorphous solid(glass)

Crystalline solid

Solid Liquid

Glass transition

Stickiness

Property of an amorphous solid

_______________________________________________________Food materials Tg (

oC)abc

_______________________________________________________Fructose 14Glucose 31Galactose 32Sucrose 62Maltose 87Lactose 101Citric acid 6Tartaric acid 18Malic acid -21Lactic acid -60Maltodextrins

DEd 36 (MW=550) 100 DE 25 (MW=720) 121DE 20 (MW=900) 141DE 10 (MW=1800) 160DE 5 (MW=3600) 188

Starch 243e

Ice-cream f -34.3Honey g -42 to -51

Glass transition temperature of various food materials

General concepts• Product above glass transition temperature (Tg) exhibits

stickiness

• Shorter chain molecules- low glass transition temperature• Tg of monosaccharides<Tg of disaccharides

• Water depresses the Tg significantly• Tg of amorphous solid water is -135oC

• For a complex food system, Tg is a function of weight fraction of each component and their Tgs’- but the relationship is not linear

Spray drying of sticky product some guideline

• Drying below the glass transition temperature (often not feasible)

• Mild drying temperature conditions

• Increasing the Tg by adding high molecular weight materials (such as maltodextrins)- a predictive approach needed according to the composition

• Immediate cooling of the product below its Tg

• Appropriate drier design to suit the sticky product

Spray drying of honey

• Honey compositionGlucose

Fructose

(Sucrose, Maltose)

• Impossible to spray dry due to low Tg (<20oC)

Spray drying of honey

Tg curveStickiness curve

Moisture

Tg

Drying

10-20oC

Particle temperature

20oC

Stickiness curve

50oC

Tg curve after maltodextrin addition

Spray drying of whey

• Whey contains lactose

• Lactose Tg is sufficiently high (101oC)

• Not difficult to spray dry

• Hygroscopic- crystallisation- caking problem during storage

Spray drying of whey

Moisture

Tg

Drying

Particle temperature

Stickiness curve

101oC

Tg curve

Spray drying of acid and hydrolysed whey

• Presence of lactic acid

• Tg of lactic acid -60oC

• Dramatic reduction on Tg of whey

• Problem of stickiness

• Hydrolysed whey– Lactose glucose (Tg=31oC) + galactose (Tg=32oC) – Difficult to spray dry

Spray drying of hydrolised whey

Moisture

Tg

Particle temperature

101oC

Tg curve- lactose32oC

Tg curve- hydrolysed lactose

Hydrolysis

Empirical approach- Index method

• Index assigned for each components of food (Tin/Tout=160oC/60oC)

0

1

Difficult to dryEasy to dry

(+1)

Maltodextrin lactose maltose sucrose glucose fructose citric acid

Possible to dry

Overall index value to determine drying aid

Xi=fractional weight of a component i (eg maltodextrin sucrose, glucose..)

ai=index value assigned for that particular component and

Y= overall index

a X Yi

n

ii 1

Predicted and experimental determined

recoveries for model mixtures

Weight fraction

Source Sucrose Glucose Fructose Malto-

dextrin

Overall

index (Y)

Recovery

(%)

Experimental 0.20 0.2 0.2 0.4 0.97 28

0.183 0.183 0.183 0.450 1.02 56

Predicted 0.188 0.188 0.188 0.435 1.00 50

Experimental 0.34 0.34 0 0.32 0.98 25

0.315 0.315 0 0.370 1.02 51

Predicted 0.327 0.327 0 0.347 1.00 50

Weighted average drying index values for honey and pineapple juice

Honey Pineapple juice

Components* Drying index (ai)

Wt. fraction (Xi)

ai Xi Wt. fraction (Xi)

ai Xi

Fructose 0.27 0.553 0.149 0.210 0.057

Glucose 0.51 0.414 0.211 0.320 0.163

Maltose** 1.00 0.034 0.034 - -

Sucrose 0.85 0.002 0.002 0.440 0.374

Citric acid -0.40 - - 0.035 -0.014

Weighted average index (ai Xi) 0.396 0.580

Honey:Malto* Overall index

Recovery %

Pineapple: Malto*

Overall index

Recovery %

0.47:0.53 1.03 56.5 0.50:0.50 1.09 58.5

0.50:0.50 1.00 55.0 0.59:0.41 1.00 50.0**

0.53:0.47 0.96 48.0 0.60:0.40 0.99 53.0

0.55:0.45 0.94 20.3 0.75:0.25 0.84 45.0

Experimental recoveries during the spray drying of honey and pineapple juice at various proportions with maltodextrin

Conclusions• Stickiness is related to the material property

• It can be correlated to glass transition temperature

• An empirical approach can be used to optimise the processing condition- however the Tg concept can be more appropriate

• Drying parameters and drier design influence the stickiness property of droplet

• Further research is needed to correlate the stickiness property with the Tg, drying parameters, drying kinetics, evolution of surface property of droplets