Food Physics Class 1- Inline measurement .pdf

Food Physics Assoc. Professor Enda Cummins Email: [email protected] Extn:7476 Process control  Need to control processing operations for maximum quality and safety  Ability to measure processing conditions and product quality parameters while processing is critical  Process control is one pa...

Food Physics Assoc. Professor Enda Cummins Email: [email protected] Extn:7476 Process control  Need to control processing operations for maximum quality and safety  Ability to measure processing conditions and product quality parameters while processing is critical  Process control is one part of quality assurance  Probes - Thermocouples, resistance thermometric devices (RTDs), pressure transducers, rheometers  Measure – texture, viscosity, moisture, temperature e.g. Sensors in food  Weighing – Belt weigher Sensors in food   Density sensors – Oscillating systems: U-shaped oscillating tubes through which liquid food pumped. Vibrates like a tuning fork that has a resonant frequency which is dependent on its mass Metal detectors Sensors in food   Flow sensors Refraction sensors Terminology  Process control involves monitoring a series of process variables or measurement of product characteristics (e.g. Rheology)  Rheology is the study of deformation and flow of foods  Rheometry: refers to techniques used to determine flow of matter under applied stress  Viscometers (often referred to as rheometers) measure several properties. Rheology  Important for.     Plant design, pumps and pipe sizing Quality control of raw materials and product Control of consumer determined quality attributes Assessment of food structure 1. In line measurement 1.1 Challenges in Process Rheometry    Rheometers - more effective (periodic samplinglab measurement) Provide immediate information subjected to a controlled deformation while the resulting force is monitored 1. In line measurement  Process rheometers can be classified into several different types  An "on-line” - sample stream, gear pump, pumped back   An "in-line” - process vessel (Adjust immediately) “Off-line” – lab (slow, cost prohibitive due to time delay from getting result and implementing a process change)  "in-line" measurements are the most desirable 1. In line measurement * 1.2 Needs of Process Rheometers 1.2.1 High Temperature Operation  Effects of high temp food processing 1.2.2 High Pressure Operation Leads to 2 problems: 1. dynamic seals for shafts 2. need to determine the torque on one of the fixtures in the case of a rotational viscometer  Solution 1: use magnetic-coupling device to transmit the torque,  Solution 2: measure the torque on the stationary member  Solution 3: Alternatively the measurement of torque can be avoided altogether by using a shear stress transducer 1.2.3 Sample Residence Time  minimise the time it takes for a rheometer to produce an output 1.2.4 Non -Newtonian Fluids  Bulk processed foods are usually non Newtonian often exhibiting viscoelasticity or yield stress. 1.2.5 Multiphase Systems  Many fluids involved in bulk food processing consist of two or more phases (solid particles/liquid/gas)  Rheometer measuring surfaces must be designed so not be damaged by particle abrasion 1.2.6 Sanitary Design and Cleanability  In food processing sanitary design and cleanability of process rheometers are critical compared to other industries (eg. Chemical) 1.2.7 Ideal Process Rheometer Requirements      designed so that they can be universally and readily installed operating at process pressure satisfactory sample renewal rate robust, measuring materials over the complete range of processes and conditions     suitable for use in hazardous environments. should be as free as possible from cleaning requirements. the process rheometers accuracy, repeatability and sensitivity should be better than what is demanded by the application. should be capable of modification 1.3 Benefits of Process Control Rheometers  Less wastage  Less "Off spec" product  Reduced manpower requirements  Less plant downtime  Reduced Energy consumption Stress and Strain! Yield stress behaviour The yield stress is the critical shear stress, applied to the material, which causes the onset of flow. The yield strain is the deformation, resulting from the applied stress, at which the flow starts. Ketchup, must flow out of a bottle when shaken or squeezed, but then solidify on the french fries. Shaking or squeezing the bottle stresses the ketchup so that it flows; after the ketchup settles on the fries, its structure rebuilds so that the ketchup “sits” in place rather than flowing off the fries like water. Stress and Strain! Stress and Strain! .  The tangential force F t divided by the area A is called the shear stress ,  Angle of deformation  is called shear deformation,  Tangent of that angle is shear strain, elongation is x.  The ratio of x/y is equal to tan  is the shear strain The change in shear angle () over time is called the shear rate (s-1) Stress and Strain! .  Newtonian fluids (ideal viscous) have the same viscosity at low shear rates as at high shear rates (e.g. water, milk, olive oil)  Non-Newtonian fluids have different viscosity at different shear rates (ketchup, starch suspensions, paint, blood, shampoo, mayonnaise, liquid chocolate) Stress and Strain! .  Newtonian fluid - relation between the shear stress and the strain rate is linear  Non-Newtonian fluid - relation between the shear stress and the strain rate is nonlinear, sometimes time-dependent 1. Newtonian 2. Pseudoplastic (shear thinning) 3. Dilatent (shear thickening) Stress and Strain! . Newtonian – viscosity remains constant Non-Newtonian –viscosity changes 1. Newtonian 2. Pseudoplastic (shear thinning) 3. Dilatent (shear thickening) 1.4 Review of Process Rheometers and Viscometers 1.4.1 Rotational  Commonly used for off-line viscosity measurements  process fluid is sheared between a stationary surface and a rotating surface  Flow must be high enough to give a fast response to process changes and to keep solids in suspension  Types of rotary viscometers  a) Concentric cylinder, b) Cone and plate, c) parallel plate  Rotary – (Concentric cylinder, flat, angled and recessed bottoms). Microprocessor. Either the inner, outer, or both cylinders may rotate, depending on instrument design. several configurations. (a) double gap, (b) cone and plate at the bottom, (c) hollow cavity SEARLE Vs COUETTE  SEARLE-type coaxial cylinder system - Cup fixed, spindle/bob rotates  COUETTE-type coaxial cylinder - Spindle/bob fixed, cup rotates Fixed Fixed Cone and plate  Cone and plate – cone of shallow angle 3-5 degree. Rotate cone, torque measured Cone and plate Cone & Plate - Almost touching tip, different angular velocities and torque applied through cone  Measure either transmitted torque on plate or torque required to rotate cone  Main advantage – works well with highly viscous materials (paste, whipped cream, emulsion) which would not flow into annular space of rotational viscometer  Parallel plate  Parallel plate – same as cone and plate. Also for highly viscous materials  But can vary gap. Important for course materials (e.g. 50 R microns, cone and plate too narrow) h 1.4.2 Vibrational  Consists of a probe vibrating at a known frequency and amplitude  Either the amplitude damping of the vibration at a constant frequency or the power required to maintain a constant amplitude of vibration at a known frequency is measured  Vibrational viscometers offer several advantages - simple - easy to clean in place and respond quickly to viscosity changes - The probes are stainless steel and can be immersed directly in a process tank 1.4.3 Capillary viscometers  Principle: when a fluid flows through a horizontal tube at a Reynolds number less than 2100 , the pressure drop is due to the viscous stress at the wall and is thus proportional to viscosity  Calibarion constant provided for each tube  Also used for: molten plastics and control of oil burner fluid viscosity µ = Kt Where µ = viscosity K = constant for each tube  = density of fluid t = time of flow END 1.4.4 Other Types of Viscometer • 'slide ring' viscosity sensor• Electromagnetic sensors • Time of travel • Advantages - small, simple to calibrate and relatively low cost Falling cylinder viscometers • gives discrete rather than continuous measurement of viscosity • A piston or cylinder falls through the process liquid within a tube and the time of fall is related to viscosity Falling Cylinder Falling ball in oil Falling Sphere Viscometer  Objective – measure terminal velocity of a falling sphere through liquid. Based on stokes law  Used for Newtonian fluids as no means of adjustment to change shear rate Falling Sphere Viscometer g.d ( k  F )  18.V 2 Where:  = dynamic viscosity of fluid (Pa.s) d = diameter of spherical particle (m) k = density of particle (kg. m-3) f = density of fluid (kg. m-3) V = velocity of particle (m. s-1) Funnel flow from beaker or cup • Beaker or cup with funnel outlet at bottom, filled to designate level, time required to drain recorded • Drain time relates to shear rate, elevation head in cup relates to shear stress • Drain time should be the same if Newtonian fluid Funnel flow from beaker or cup • Can convert to viscosity using chart, however in industry time just monitored • Simple, but as level decreases elevation head or hydrostatic pressure driving the flow decreases, therefore shear stress not the same throughout test Bostwick Consistometer • Primarily highly viscous liquids e.g. ketchup • Inclined plane and stop watch • Amount of time to flow to a designated endpoint recorded • Height of incline relates to shear stress (although usually not calculated) • Indicator of viscosity for quality control Bostwick Consistometer The DeZurik Blade Sensor Consistency transmitter Principle of sensing the fluid drag on the sensing blade to determine the consistency pressure required to maintain the sensor blade position varies in direct proportion to consistency changes Model functions  Reological model functions are mathematical equations derived to describe the various flow behaviour curves on shear stress-shear rate diagrams  Require 2-3 parameters specific to the material being tested  Models differ in “goodness of fit” for different materials Model functions  Useful for competitor-comparisons and marketbenchmarking  Important for understanding and controlling how a fluid product will perform in the "real world" of usage (subjected to different shear rates) Model functions Ostwald-de-Waele law (Power law) Common, uncomplicated law, extensively used to describe flow behaviour .   K ow  Where   shear stress in Pa n K ow  Consistency coefficien t in Pa .   shear rate s -1 n  Flow behaviour index Model functions n is measure of “non-Newtonian-ness”  n <1 pseudoplastic flow behaviour  n >1 dilatent flow behaviour  For Newtonian behaviour index n =1 and Kow =  hence: .    . Model functions  Three brands of cooking oil, premium, Economy and mid market Model functions Sample Consistency (Pa·s) Power Law (Flow) Index Premium 63.9 0.16 Mid-Market 34.1 0.27 Economy 24.7 0.15 • While the Premium brand has a higher viscosity than the Mid-Market brand over the measured range of shear rates it is also significantly more shear-thinning (lower Power Law Index) so the flow curves would be expected to cross over at higher shear rates. •Filling, pumping and spreading are all generally high shear rate, therefore products will differ in consistency when used Measurement - Cylinder Geometry  Relationship between parameters of the rotating cylinder and rheological quantities shear stress and shear rate necessary BOB and CUP A = 2..r.h where A = Area of the cylinder surface (m2) r = radius of the cylinder (m) h = height of the cylinder (m)  BOB and CUP   The torque experienced by the cylinder is a function of the radius and the tangential force Thus M = r.F where M = torque (N.m) r = radius of the cylinder (m) F = tangential force (N) BOB and CUP  Previously we noted shear stress () is equal to F/A F M   2 A 2. .r h  Thus the shear stress can be calculated as a function of the measured torque and bob geometry BOB and CUP  Shear rate can also be calculated as a function of cylinder geometry. .Ri   Ra  Ri . Where  = angular velocity of the inner cylinder in s-1 Ri = radius of inner cup Ra = radius of outer cup BOB and CUP  For Newtonian fluid M  1 1   2  2  4. .h.  Ri Ra    This is refereed to as MARGULE’s equation and applied to the flow of Newtonian fluids in a coaxial rotary viscometer. It provides the relationship between angular velocity () of the motor and the resulting torque M when a Newtonian fluid with viscosity  is in the annular space between Ri and Ra Cone and plate  For cone and plate geometry the shear rate is dependent on the velocity of the cylinder and the angle of the cone. Usually the angle of the cone is small (0-5O) in which case tan  can be approximated to  Cone and plate  Shear rate      tan   . Where  = angular velocity of the inner cylinder  = 2..n. n = speed of rotation (revolution per s)  = angle of the cone Cone and plate  Shear stress 3.M  3 2. .R   Where M = angular moment Viscosity   .  Cone and plate  Example Parallel plate  In a plate and plate geometry a round plate rotates above a fixed (non moving plate). This distance between the plates can be adjusted and provides space for the sample. Parallel plate  Shear stress 2.M  3  .R Where M is torque (N.m) R = radius of the cylinder (m) Parallel plate  Shear rate. R 2. .n.R   .  H H . Where  = angular velocity of the inner cylinder R = radius of the cylinder (m) H = gap (m) n = speed (RPM) Parallel plate  Example Model functions  BINGHAM model  Describing flow in puree, pastes .    o  .  Where 0 = BINGHAM yield stress (Pa)  = BINGHAM viscosity (Pa.s) Model functions  Casson model  Originally derived for characterising inks  Since 1973, the flow behavour of molten chocolate has been evaluated using Casson’s model  Particularly used for finding yield stress Model functions  Casson   0.5 model  0.5 0 .  (. ) 0.5 Where o = CASSON yield stress (pa)  = CASSON viscosity (Pa.s) Model functions  Herschel-Bulkley model - generalized model of a non-Newtonian fluid . n    O  K .   Where o = yield stress (pa) K = consistency coefficient in Pa.s n = flow behaviour index Note when 0 = 0 = Power law Model Functions Yield stress Graph 4 = BINGHAM (linear with yield stress) Graph 5 = HERSCHEL-BULKLEY (pseudoplastic with yield stress) Mayonnaise Non-Newtonian Flow Behaviour       A: viscosity decreases with increase shear rate B: viscosity increases with increase shear rate C: Has a yield stress (keeps deformation after shear stress e.g. butter) D: Following shear stress deforms but returns to initial value after shear stress released (e.g. marshmallow) E: Viscosity decreases over time at constant shear rate (yogurt) F: Viscosity increases over time at constant shear rate (e.g. whipped cream) 1.5 Process Rheometer Applications 1.5.1 Skim Milk Powder Manufacture  The viscosity of skim milk concentrate is influenced by five principal factors; temperature, total solids, protein content, pre-heat treatment and age-thickening.  Milk for manufacturing from spring calving dairy herds  Recommended viscosity for skim milk concentrate in falling film evaporators is generally stated to be 100 mPa/ s at 100/s shear rate  60% energy used in spray dryer to remove 10% water  Energy required per kg water removed in spray dryer is much higher than energy required per kg water removed in evaporator  Protein April/May higher due to quality of spring grass, drops off in June/July when less nutritious  Protein can vary by 3-4% (larger than other European countries) Skim milk production  Desirable    that continuous monitoring of skim milk concentrate viscosity is carried out (in line viscometer) Feed concentrate needs to be kept constant Energy requirement for dryer high therefore want as little water entering as possible However, trade-off with poor atomisation of skim concentrate & blocked pipelines 1.5.2 Cheese manufacture A number of devices have been employed to measure the firmness of the coagulum with the objective of predicting the optimal cut time Rennet/ microbial emzymes - Aggregation of micelles  Cheese cutting time Optimum cutting time subject of much research  Cutting when too soft decreases cheeses yield due to increased loss of fat and curd fines.  Cutting when too firm results in high moisture content in cheese Testing curd Testing curd Cutting curd Draining whey Blocks Milling Folding Blocks Salting Filling moulds Compressing cheese END  Mechanical force methods such as resistance to an oscillatory deformation and resistance to an oscillating probe have predominated  However all these methods physically disrupt the coagulum and are considered destructive tests  Ideal sensor would be non intrusive, having CIP (cleaning in place) sanitary standards and would not physically disrupt the coagulating milk.  Recently in Japan, a sensor which measures the rate of heat transfer from the surface of a thin platinum wire fixed in milk to be rennet treated, has been used for the measurement of fluid viscosity, which is related to the progress of milk curd formation.  The lower the viscosity of the fluid the more rapidly heat is transferred, so that the temperature of the hot wire in the equilibrium state will be correspondingly lower  This is because a lower viscosity fluid will possess a thinner boundary layer around the hot wire in the equilibrium state. Other applications +area of importance • Design - pumps, Pipe sizes, Mass and heat transfer, Mixing • Quality control – raw materials and finished product • Sensory attributes – consumer attitudes Rheology measurement 2 types– fundamental or emperical • Fundamental – exert stress and measure strain (or visa versa) viz viscometers (i.e. not derived from other measurement) • Emperical – non-homogenous foods where fundamental measurement not possible. E.g. cone pentrometers 1.6 Stokes Law and Terminal Velocity - Mixtures of solids and liquids - Basis of seperation and classification systems - Basis of the falling sphere viscometer, - Terminal velocity can be measured by the time it takes to pass two marks on the tube. - Knowing the terminal velocity, the size and density of the sphere, and the density of the liquid, Stokes' law can be used to calculate the viscosity of the fluid. Stokes Law and Terminal Velocity 1.6 Stokes Law and Terminal Velocity The forces acting on a spherical solid particle of diameter d and density s falling through a less dense fluid of density  are as follows: • Acting downwards will be a force due to the weight and given by  d3 s g/ 6 • Acting upwards will be a force due to the upthrust, and given by  d3  g/ 6 FR FR : retarding force Up : upward thrust W : weight Up W • There is also a force due to the viscous drag, i.e. the fluid exerts a retarding force on the particle. For a spherical particle, this retarding force FR is given by the expression (from dimensional analysis) FR  C d A p v 2 2 Where Cd : dimensionless drag coefficient Ap: projected area of the particle in a plane perpendicular to the plane of flow (for a sphere this equals (d2/4) ρ : density v : velocity • Therefore FR  C d  d 4 2  v 2 2 • The drag coefficient Cd is a function of the particle Reynolds number, which is defined as where Rep vt d    : density of the fluid : viscosity of the fluid Vt : terminal velocity d: particle diameter • Reynolds number, is a dimensionless number that gives a measure of the ratio of inertial forces to viscous forces and thus quantifies the relative importance of these two types of forces for given flow conditions. If the particle Reynolds number is less than 0.2, then the flow is streamline and the drag coefficient is given by 24 Cd  Rep 24 Cd  vt d Hence the force can be calculated as 2 2  v  d follows FR  C d   4 2 24  d 2 v 2 FR    v t d 4 2 FR = 3dvt  = viscosity, d = diameter of particle, Vt = terminal velocity This is known as Stoke’s law, after G.G. Stoke who discovered it. 1.6.1 Terminal velocity downward force of gravity (Fg)equals the upward force of drag (Fd). the net force on the object to be zero, resulting in an acceleration of zero. it plummets at a constant speed called terminal velocity (also called settling velocity). Terminal velocity varies directly with the ratio of drag to weight. More drag means a lower terminal velocity, while increased weight means a higher terminal velocity. 1.6.1 Terminal velocity • For a particle moving in a fluid when the upward forces equal the downward forces, the particle is either at rest or moving at its terminal velocity vt. • Under these conditions, Drag force + Upward thrust = Downward thrust due to weight F R + Up = W 3dv t  d 3 g 6  d 3  s g 6 This allows the terminal velocity to be calculated as follows 3dv t  d 3  s g 6  d 2g  s    3 v t  6 d 3 g 6 s : density of sphere  : density of liquid vt  d 2  s   g 18 This final equation is known as Stoke’s equation. It can be used for estimating terminal velocities or evaluating dynamic velocities of fluids. Application of Stokes Law and Terminal Velocity - Principle of separation by sedimentation and air classification. - E.g. oil seed processing, grain processing, air classification widely used to separate low density hulls, husks and shells from higher density seeds grains, seeds and kernels. Application of Stokes Law and Terminal Velocity A = (F - L)(dw/dt)/vr Where: A is the settling area in the tank. F is the mass ratio of liquid to solid in the feed, L is the mass ratio of liquid to solid in the underflow liquid, dw/dt is the mass rate of feed of the solids, V is velocity r is the density of the liquid Application of Stokes Law and Terminal Velocity A continuous separating tank is to be designed to follow after a water washing plant for liquid oil. Estimate the necessary area for the tank if the oil, on leaving the washer, is in the form of globules 5.1 x 10-5 m diameter, the feed concentration is 4 kg water to 1 kg oil, and the leaving water is effectively oil free. The feed rate is 1000 kg h-1, the density of the oil is 894 kg m-3 and the temperature of the oil and of the water is 38°C. Assume Stokes' Law. Viscosity of water = 0.7 x 10-3 N s m-2. Density of water = 1000 kg m-3. Diameter of globules = 5.1 x 10-5 m d 2  s   g vt  18 vt = (5.1 x 10-5)2 x 9.81 x (1000 - 894)/(18 x 0.7 x 10-3) = 2.15 x 10-4 m s-1 = 0.77 m h-1. and since F = 4 and L = 0, and dw/dt = flow of minor component = 1000/5 = 200 kg h-1, we have: A = 4 x 200/(0.77 x 1000) = 1.0 m2 i.e. 4+1 =1000kg/hr therfore 1kg oil = (1000/5)*1 = 200 kg/hr Learning outcomes  Explain the basic science governing selected physical properties of food  Fluid flow behavious (curves, flow diagram)  Methods of measurement  Behaviour described by model functions (power law)  Describe their importance to both consumer and food manufacturer  Needs of process rheometry  Benefits  Industry applications Learning outcomes  Perform simple calculations involving physical properties  Viscosity measurement  Application of stokes law  Determine the numerical values of physical properties.  Online practicals