A MIXER VISCOMETRY APPROACH FOR BLENDING DEVICES

The Matching Stress Method is a mixer viscometry technique used to estimate mixer constants and viscosity using systems of complex geometry. Three Newtonian standards and three hydroxy‐propyl‐methyl‐cellulose (HPMC) solutions were used as model systems. to determine the mixer constants, a counter‐rotating, twin‐rotor blender was modified with a torque transducer and speed sensor installed between a drive shaft and a 0.75 hp (559.3 W) electrical motor. Mixer constants were determined using a concentric cylinder approximation of shear stress in an effort to match this approximation with shear stress measurements determined from a bench top rheometer. Mixer constants for the model systems ranged between 2.01 and 2.87 rad^?1 with an average value of 2.57 rad^?1 for all fluids. the results demonstrate a successful technique for providing accurate viscometric data during mixing processes. Such efforts could be followed to implement real time, quality control strategies and power estimation in scale‐up procedures.     

EFFECTS OF SHEAR-THINNING BEHAVIOR ON MIXER VISCOMETRY TECHNIQUES3

Brookfield RVTD and HBTD viscometers equipped with a flag impeller, having an impeller-to-cup diameter ratio of 0.55, and non-Newtonian fluids 1.0% guar gum and 1.5, 2.0 and 2.5% carboxymethylcellulose were used to collect torque versus rotational speed 0.5–10.5 rad/s (0.5–100 rpm) data. Power Law consistency coefficient and shear thinning index were quantified using a Haake concentric cylinder viscometer. The proportionality constant which relates mixer shaft speed to effective shear rate (k₁) was very sensitive to the shear thinning index at low values of rotational velocity. Results indicate that it is not practical to use one constant value for k₁ when measuring properties of shear thinning fluids at low values of N or when using impeller-to-cup ratios significantly less than 1.0.     

EVALUATING SHEAR RATES FOR POWER LAW FLUIDS IN MIXER VISCOMETRY

Established mixer viscometry methods, the viscosity matching and the slope methods, were evaluated for determining average shear rates when agitating time-independent, non-Newtonian fluids. Results indicate that the use of a constant value of the mixer proportionally constant, k ', is not valid for all ranges of fluid rheological properties, system geometry (impeller and cup) and impeller rotational speed. A small impeller-to-cup diameter ratio and impeller rotational speeds greater than .33 rev/s (20 rpm) provide the optimum operating parameters for evaluation of k '.     

MODELING RHEOLOGICAL BEHAVIOR OF GELATINIZING STARCH SOLUTIONS USING MIXER VISCOMETRY DATA

A Brookfield RVTD mixer viscometer was used to gelatinize 5.5, 6.4, and 7.3% (d.b.) native corn starch solutions, and a 6% bean starch solution. Torque, the dependent variable, was used to estimate parameters in a generalized viscosity model. The independent variables were impeller speed, temperature (50–95°C), concentration, temperature-time history, and strain history. The starch dispersions thickened with decreasing temperature and had an Arrhenius activation energy between 6.4 and 12.7 kJ/mol. There was no evidence that retrogradation caused this effect. First-order reaction kinetics was accurate (9.8% standard deviation) in describing viscosity increase during gelatinization. After initial gelatinization, torque decayed exponentially with time. Predicted curves followed experimental pasting curves when interactions among the independent variables were accounted for in the model. The major variables controlling the pasting process were temperature-time history before peak torque, strain history between peak torque and cooling, and temperature during cooling.     

QUANTIFYING THIXOTROPY IN STARCH-THICKENED,STRAINED APRICOTS USING MIXER VISCOMETRY TECHNIQUES2

Starch-thickened, strained apricots were found to exhibit irreversible thixotropy. Shear stress and shear rate were estimated from mixer viscometer data. A structural parameter, yield stress, flow behavior index and a consistency coefficient were determined to quantify the rate of breakdown and equilibrium structure of the product. The techniques developed in this study are useful in describing the time-dependent behavior of suspensions because the slip at the wall is minimized.     

ADAPTATION OF A BROOKFIELD (HBTD) VISCOMETER FOR MIXER VISCOMETRY STUDIES

A low cost mixer viscometer system was developed using a Brookfield viscometer interfaced to a Hewlett Packard (HP) digital cassete drive. Tests were conducted in a Brookfield Small Sample Adapter with a specially designed flag type, mixer impeller. The system was tested for two power law type products (apricot puree and guar gum) with results being comparable to those obtained using a Rheometrics Fluids Spectrometer.     

SENSITIVITY ANALYSIS of ASEPTIC PROCESS SIMULATIONS FOR FOODS CONTAINING PARTICULATES1

A sensitivity analysis of a computer model, simulating heat transfer into particulate foods processed in a continuous aseptic system was conducted to determine how the model reacts to the variations in selected product and process input parameters depending on simulation types, namely, “Total”, “F₀ Hold” and “Hold Only”. Three major output variables, namely, holding tube required to destroy 6 D of Clostridium sporogenes (PA3679), destruction of thiamine and inactivation of peroxidase were selected for the sensitivity analysis. Theoretical results indicated that basically no difference between the “Total” system approach and the “F₀ Hold” approach was found on the model prediction. Particle size, particle thermal properties such as density and specific heat were the most sensitive parameters (within the range investigated) among product parameters which influence holding tube length required, thiamine and peroxidase retention while fluid properties such as fluid thermal conductivity and viscosity were less susceptible to affect the model prediction regardless of simulation types. Among process parameters, product flow rate and product initial temperature seemed to be the most critical parameters within the range covered in this study while rotor speed seemed to be one of the least critical parameters to influence the model prediction regardless of simulation types.     

SENSORY THRESHOLDS OF CONSISTENCY OF SEMILIQUID FOODS: EVALUATION BY SQUEEZING FLOW VISCOMETRY*

Identification of subtle textural differences or changes in foods in isolation has been a difficult task because they are usually accompanied by differences or changes in other sensory properties. Textural modification of semi liquid foods can easily be produced by mechanical means without affecting their taste and appearance. The changes can be then quantified using squeezing flow viscometry without introducing new uncontrolled modifications. Comparison of the measurements with sensory evaluations of the same samples can establish the threshold of sensory detection, at least in principle. To test this hypothesis, the consistency of practically intact mayonnaise, tomato paste, yogurt and strawberry jam of two commercial brands each were evaluated by squeezing flow viscometry using an Instron UTM. Other samples of the same brands were stirred mildly to modify their consistency and then compared with the undisturbed samples. The textural differences were expressed in terms of the apparent stress at three compressed specimen heights, and the residual stress after 60 and 120 s at the final height. Relatively large differences, as judged by these mechanical measures were also detected by an untrained sensory panel, but the ‘correct identifications’ never reached a 100%. In the case of the two different brands of yogurt, the panel failed to establish the existence of differences, which were clearly detected by the instrument. This suggests that not every difference, detectable instrumentally, has a corresponding sensory response. Although the number of results in this preliminary work was too small to establish the exact threshold for sensory detection, the methodology seems to be appropriate for such a purpose. This is because of the high reproducibility of the squeezing flow parameters and their sensitivity to even subtle difference in the tested specimens consistency.     

A GENERALIZED RHEOLOGICAL MODEL FOR INELASTIC FLUID FOODS1

A versatile four-parameter model is presented for the rheological characterization of inelastic fluid foods. The model has been shown to accurately represent shear stress versus shear rate, and apparent viscosity versus shear rate relationships for several food materials. By appropriate specification of the four parameters of the model, conventional rheological models used in fluid food analyses (power law, Bingham plastic, Herschel-Bulkley, Casson, Heinz-Casson, and Mizrahi-Berk) can be obtained from the generalized equation.     

ORAL PERCEPTION OF VISCOSITY IN FLUID FOODS AND MODEL SYSTEMS

The perceived in-mouth thickness ( T ) of a range of fluid foods and model systems was assessed by a trained sensory panel, using a ratio scaling technique, and correlated with objective measurements of viscosity (η_N). For Newtonian samples a simple linear correlation (r²= 0.98) is observed between log T and log η with exponent η= 0.22. The linear relationship between log T and log η for Newtonian materials was used to calculate the equivalent Newtonian viscosity (η_N) from subjective panel scores for the non-Newtonian samples studied. The shear rate at which observed viscosity is equal to η_N decreases with η_N, as reported previously by Shama and Sherman, but also decreases with increasing shear-rate dependence of viscosity. This implies oral perception of viscosity over a range of shear rates. Comparison of flow curves for samples assigned similar thickness scores, but showing very different shear thinning behaviour, suggests that the perceived thickness of extremely shear-thinning materials is dominated by their high viscosity at low rates of shear. As a simple practical index, viscosity at 10 s⁻¹ shows a better correlation (r²= 0.95) with the perceived thickness of the samples studied than values calculated by the methods suggested by Wood, or by Shama and Sherman (r²= 0.90 in both cases). The close agreement between panel scores for perceived thickness and perceived stickiness previously observed for concentrated solutions of random coil polysaccharides does not apply for several of the food systems studied, or for very concentrated solutions of xanthan.     

DEVELOPMENT OF A QUANTITATIVE VIDEO-BASED VISUALIZATION METHOD TO CHARACTERIZE THE FLOW BEHAVIOR OF FOOD PARTICULATES IN A MODEL CONTINUOUS ASEPTIC STERILIZER3

The objectives of this study were to develop a quantitative visualization tool for evaluating flow behavior of particles in a compartmented aseptic system or other flow systems. The three dimensional movement of polystyrene balls as influenced by ball diameter (0.95 and 1.90 cm), fluid flow rate (10, 20 and 30 L/min) and conveyor disk design (2 configurations) was recorded in a model heating apparatus and analyzed using motion analysis software. Ball speed and net-to-gross-displacement ratio (NGDR) values were calculated for ball movement in the x;y and x;z planes. As carrier liquid flow rate increased, there was an associated increase in both the mean and standard deviation speed and NGDR values. In general, larger ball sizes yielded lower speed and NGDR values (i.e., less movement). A concave (bowl-like) conveyor disk design as opposed to a 90° flat-edge disk design yielded greater speed and NGDR values when carrier velocity was greater than 20 L/min. Speed and NGDR values having higher standard deviations were interpreted as having more compartmental mixing. Furthermore, speed and NGDR mean and standard deviations were highly correlated. The results of this study demonstrated the potential utility of the flow visualization method for quantitating the flow behavior of particles through tubes. Furthermore, this method should be of value to food process engineers in developing continuous aseptic processes for particulate-containing foods.     

SHEAR RATES DURING ORAL AND NONORAL PERCEPTION OF VISCOSITY OF FLUID FOODS

Relationships were established for five sensory methods of oral and nonoral viscosity evaluation between viscosity scores and instrumentally measured dynamic viscosity for model and real Newtonian fluid foods. These relationships were then used to predict the effective shear rates under which the sensory tests were performed. The highest shear rates were predicted for viscosity perception by compression of samples between tongue and palate, and the lowest for pouring the fluid foods from a teaspoon. Mixing with a teaspoon, slurping and swallowing exhibited nearly the same dependencies of apparent shear rates on equivalent instrumental viscosity. All relations were of the hyperbolic type. The resulting relationships between the apprent shear rates and equivalent instrumental viscosity are in good agreement with a similar relationship predicted by Shama and Sherman (1973a) (see Cutler et al. 1983) for oral perception.     

DESKTOP COMPUTER BASED COLLECTION AND ANALYSIS OF CREEP-COMPLIANCE DATA ON FLUID FOODS

In a creep-compliance test, the strain due to an imposed stress is recorded as a function of time. Subsequently, the strain is converted to creep-compliance and the rheological parameters: compliances, retardation times, and viscosities are evaluated by a graphical method suggested by Inokuchi. A Macintosh Plus microcomputer was programmed in Basic language to collect time-strain data from a Deer Rheometer III. A Fortran language program was written first to compute creep-compliance from strain and subsequently to estimate rheological parameters using a nonlinear parameter estimation subroutine. Initial values for the nonlinear analysis were estimated by the computer program by means of a modified Inokuchi's method.     

MEASUREMENT OF FLOW PROPERTIES OF FOOD SUSPENSIONS WITH A MIXER

Abstract. Flow properties of selected food suspensions (tomato puree, apple sauce, and apple pulp) were studied with a mixer. Power law constants for the suspensions were determined by employing an approximate method based on power correlations for mixers in the laminar flow regime. The studied suspensions cannot be characterized with narrow gap viscometers because of the size and segregation of the particles (formation of a dilute phase next to the rotor).     

SOME EXPERIENCES WITH THE GENERAL FOODS TEXTUROMETER*

Abstract As part of a programme of investigation into the effects of processing on the texture of food the General Foods Texturometer was adopted, because it appeared to offer versatility, multiple-property determination and a fair degree of objectivity. Preliminary investigations with the instrument indicated a high degree of reproducibility when used on a ‘recoverable’ material (rubber). The results are highly dependent on the dimensions of the sample and reproducibility decreases somewhat with increasing hardness of material under test. The ‘representativeness’ of the results for a range of food materials varied widely, being influenced by the uniformity of the material and the accuracy with which samples can be prepared. Experiments to correlate the results from this instrument with those from the Instron Universal Testing Machine, the Kramer Shear Press and a subjective panel method indicate that the Texturometer ‘hardness’ might prove the most useful measure of the firmness/hardness of foods as judged subjectively. None of the other textural properties measured by the Texturometer in this work proved of value in indicating subjective textural characteristics and further study of such properties is currently in progress.     

PRESSURE DROP ESTIMATION IN TUBE FLOW of NON-NEWTONIAN FLUID FOODS BY NEURAL NETWORKS

A tube flow viscometer complete with data acquisition system was designed and developed for continuous measurement of pressure drop and flow velocity. Experiments were carried out with tomato ketchup, oyster sauce, mayonnaise, and 1% and 1.5% CMC solutions in the laminar flow region using stainless steel tubes of four diameters (0.00751–0.01636 m). the flow parameters determined with the tube viscometer after slip correction and a rotational viscometer were correlated for estimating the pressure drop indirectly. Finally, it was shown that neural networks could accurately predict the pressure drop in tube flow without making any correction for wall‐slip from the input data on tube diameter, fluid density, mass flow rate, and power‐law parameters determined with a rotational viscometer. Among three neural network architectures tested, the generalized regression neural networks were most easy to train and they predicted the pressure drop gradient in tube flow with greatest accuracy (4.7% average absolute error).