Factors Which affect Nucleation, bubble growth and Expansion - PowerPoint PPT Presentation

Factors Which affect Nucleation, bubble growth and Expansion During Extrusion – impact on mechanical properties and crispness Jozef L.Kokini Department of Food Science Whistler Carbohydrate Center Purdue University

Outline of Presentation  Factors which affect air nucleation during extrusion Bubble growth and dispersion  Expansion during extrusion  Mechanical properties and crispness of food foams 

Nucleation and Expansion During Extrusion

Factors that influence nucleation and extrudate expansion MATERIAL PARAMETERS CHO, protein, Molecular Moisture H, OH, salt lipid interaction structure sugar, gums OPERATIONAL PARAMETERS Barrel & die Screw Screw Mechanical Die Air Temperature speed geometry energy input geometry incorporation TIME-TEMPERATURE-SHEAR HISTORY GELATINIZATION, DEXTRINIZATION, DENATURATION MELT RHEOLOGICAL PROPERTIES SHEAR RATE DIE L/D BUBBLE GROWTH AND COLLAPSE FINAL EXPANSION

Mechanism of food extrudate expansion 1. Order-disorder 3. Extrudate swell transformation and chemical 4. Bubble growth complexing (in the extruder) 5. Bubble collapse 2. Nucleation Screw Die Nucleation Expansion Contraction Bubble growth Bubble collapse

Air bubble nucleation Compaction of granular material Dissolved air (pressure, (BFL, pressure, barrel temperature) temperature, entrapped air) Extruder Bubble breakdown (“atomization”) Porosity of the raw material (screw speed, viscosity)  The total air volume entrapped (porosity) is determined by the barrel fill length, the barrel temperature profile and their interaction  The factors that affect bubble size are the original pore sizedistribution of the granular material and the bubble breakdown in the extruder barrel due to high shear conditions

Mechanism of bubble nucleation in starch extrudates A. Air pockets from interparticle, inter- and intragranular voids serve as water vapor nuclei B. Hydrophobic surfaces C. Mobility of water molecules decrease wetting increases with distance from polymer chains

Bubble size distribution - effect of screw configuration- Native amylopectin, rpm=150, 3.3g/s, m.c.=32%

Bubble size distribution - effect of screw speed and mass flow rate- Native amylopectin, 1.67g/s, m.c.=32%, Native amylopectin, rpm=150, m.c.=35%

Bubble size distribution - effect of type of starch and moisture content- rpm=150, 1.67g/s, m.c.=35% rpm=150, 1.67g/s

Air entrapment is determined by the barrel fill length 1. When EBFL is in the “cold” barrel section, the unexpanded extrudate has numerous air bubbles 2. When EBFL is in the “hot” zone, the unexpanded extrudate has few or no bubbles. feed zone 5 4 3 2 1 slit die 25ºC 25ºC 60ºC 123ºC 86ºC Filled section melt granular Many air bubbles in the extrudate Filled section Air melt granular No air bubbles in the extrudate

Porosity master curve  All data points generate a broad master curve  Two defined regions are observed: Porosity  at low EBFL, the porosity increases slightly  past a critical EBFL , the porosity increases rapidly and exponentially. Effective Barrel Fill Length, cm

Influence of the physical state of starch on air entrapment Native amylopectin Native, 150 rpm Pregelatinized amylopectin Native, 300 rpm Normal corn starch Native, 500 rpm Pregelatinized,150 rpm Bubble number density, g -1 Pregelatinized,300 rpm Pregelatinized,500 rpm Porosity Effective Barrel Fill Length, cm Effective Barrel Fill Length, cm  The physical form and type of starch do not influence the porosity  Pregelatinized amylopectin leads to much higher bubble number density values than native amylopectin

Effect of SME on air entrapment 38000 ° With slit die cooling Native, 150 rpm Native, 150 rpm Bubble number density, g -1 Native, 300 rpm Native, 300 rpm 36000 • Without slit die cooling Native, 500 rpm Native, 500 rpm Pregelatinized,150 rpm Pregelatinized, 150 rpm Pregelatinized,300 rpm Pregelatinized, 300 rpm Pregelatinized,500 rpm Pregelatinized, 500 rpm 12000 y (g ) Porosity 10000 8000 6000 4000 2000 0 0 200 400 600 800 1000 1200 1400 1600 SME, kJ/kg SME, kJ/kg  Porosity decreases with SME  Low SME (200-900kJ/kg) leads to high bubble number density, while high SME (900-1600kJ/kg) leads to low bubbl number density

Relationship between porosity and bubble number density Pregelatinized,150 rpm Pregelatinized,300 rpm Pregelatinized,500 rpm Native, 150 rpm Native, 300 rpm Bubble number density, g -1 Bubble number density, g -1 Native, 500 rpm Pregelatinized,150 rpm Pregelatinized,300 rpm Pregelatinized,500 rpm Native, 150 rpm Native, 300 rpm Native, 500 rpm Porosity Porosity For the same porosity, pregelatinized amylopectin gives higher bubble number densities than the native amylopectin  An increase in screw speed increases bubble breakdown (atomization)

Breakup of entrapped air during extrusion An increase in the shear field due to increase in screw speed caused breakup of entrapped air bubbles in the unexpanded extrudate 16 (Cisneros and Kokini, 2002)

Influence of polymeric melt viscosity Amioca (lower viscosity) Hylon 7 (higher viscosity) Amioca Hylon 7 Corn flour

Important phenomena in extrudate expansion Extrudate swell - the phenomenon that governs  the diametral expansion of extrudates in the absence of blowing agents. It is caused by elastic recovery. Bubble growth – in a high viscosity mass, it is  mainly determined by the driving force and the resistance to deformation. Bubble collapse – determines the final extrudate  expansion in high moisture and low viscosity materials.

Biaxail Bubble growth is controlled by the driving force and material viscosity  Vapor  •  σ   2 R − = + η σ P P 4   P L V L L R R     P V R Liquid • ∆  For very viscous materials: R P = η R 4 L •  For power law fluids: 1  ∆  R P n =   R  4 m 

The dependency is experimentally valid: Extrusion Extrusion data data Alveograph data

Influence of melt viscosity on extrudate expansion

First normal stress difference and recoverable shear strain Moisture increases Recoverable shear strain of First normal stress difference of amylopectin as function of C and T amylopectin as function of C and T

Conclusions on extrudate expansion Extrudate expansion is a result of structural order-  disorder transformations, nucleation, extrudate swell, bubble growth and bubble collapse  Air bubbles entrapped in the matrix can act as nuclei for further expansion of the matrix (by extrusion, microwave heating, etc.)  A mechanism for air bubble nucleation was proposed  Extrudate expansion is determined by the water vapor pressure and the rheological properties of the melt

Physical Properties of Foam like foods that affect textural properties • Pore size • Pore size distribution • Cell wall thickness • Strength of the cell wall • Porosity • Phase behaviour 24

Stages in the failure of a cellular material 1 3 0 0 0 1 3 0 0 0 1 3 0 0 0 1 2 5 0 0 1 2 5 0 0 1 2 5 0 0 1 2 0 0 0 1 2 0 0 0 1 2 0 0 0 Force (g) Force (g) Force (g) 1 1 5 0 0 1 1 5 0 0 1 1 5 0 0 1 1 0 0 0 1 1 0 0 0 1 1 0 0 0 1 0 5 0 0 1 0 5 0 0 1 0 5 0 0 1 0 0 0 0 3 , 2 3 , 7 4 , 2 4 , 7 5 , 2 5 , 7 2 6 , 7 6 , 1 0 0 0 0 3 , 2 3 , 7 4 , 2 7 4 , 5 , 2 5 , 7 6 , 2 6 , 7 Distance (mm) 1 0 0 0 0 Distance (mm) 3 , 2 3 , 7 4 , 2 7 4 , 2 5 , 5 , 7 2 6 , 6 , 7 Distance (mm) 25

Textural Analysis • Jaggedness of force deformation curve -Average # of peaks -Ratio of linear distance -Fractal dimensions • Average drop off • Average force • Maximum force 26

Measures of the jaggedness of force-deformation curve • Number of peaks Number of positive peaks greater than threshold force • Ratio of linear distance (RLD) At constant smoothening ratio, RLD is higher for more crisp products. •Fractal analysis Fractal dimensions were calculated according to Barrett and Peleg (1994). 27

Physicophysical model development • The psychophysical power law model has been used to develop a model for crispness. • According to this model the magnitude of crispness grows as a power function of the total number of force deformation peaks resulting from the fracture of cells. Crispness score= a (average number of peaks) b = a (Np) b • The value of exponent b determines the curvature of the power function. If b is close to 1.0 sensation varies directly with the intensity of stimulus. Stimuli that give a slope near 1.0 are those that are closer to identifying the real stimulus for the sensation. 28

Cell wall thickness to radius ratio in extruded snacks 16 R 2 = 0.63 Thin cell walls Thin cell walls Thin cell walls Thick cell walls Thick cell walls Thick cell walls Thick cell walls 12 Np 8 4 0 0.00 0.02 0.04 0.06 0.08 0.10 0.12 t/R Energy required to fracture big air cells surrounded with thin cell walls is lower than that is required for the cells having thick cell walls. Low t/R increases the fracturability of the solid foams and thus leads to a higher crispness sensation. 29