식품의 유변학적 특성을 레오미터로 확인하기

The use of modern rheometers  

in characterising the behaviour of foods  

Many foodstuffs are concentrated dispersions of solid particles in a liquid medium (e.g. tomato ketchup), or liquid droplets in a liquid medium (e.g. mayonnaise). The manufacture of these products has to be very reproducible in order to satisfy customer needs.  

 

The rheological properties of these materials give an indication of this reproducibility and can be correlated with, for example, storage stability, ease of pumping and sensory perception (i.e. mouthfeel). Often, ‘rheological analysis’ is performed using rotational viscometers and by measuring the viscosity at moderate to high shear rates. Under these circumstances, it is conceivable that this analysis would not show subtle differences in the properties, which may only be obvious at low shear rates.  

 

This may manifest itself in aesthetically displeasing syneresis (or sedimentation) or, even worse, a detection of a different ‘feel’ by the consumer. Modern rheometers offer an opportunity to characterise the rheological properties of these materials more precisely. This article discusses the increased range of measurements available and gives some specific examples of the use of these measurements. 

Introduction
 

 This note describes the use of a variety of rheological experiments to characterise foodstuffs. These methods have been chosen, and the experiments designed accordingly, to either predict the behaviour of the material, or to simulate a process to which the material may be subjected.
In all cases, the experiments were performed on a Bohlin rheometer. 

Rheological evaluation
 

   Example 1 – Stress viscometry 

 


In this example, a rheological experiment was used to predict the sedimentation behaviour of two samples of orange barley water. Sedimentation occurs due to the effect of gravitational forces on the suspended solid particles. 

 

The gravitational forces will have a larger effect on the larger particles, which exist due to either a fault in the process or small particles flocculating to form larger ‘particles’ during storage. The overall effect is to produce a ‘consumer unfriendly’ product. 

 

 In order to quality assure this product, it is necessary to evaluate the structure in the barley water at low (i.e. gravitational) stresses. The assumption is that the higher the structure, the less likely the particles are to sediment.* Using a Controlled Stress rheometer it is possible to measure the viscosity of fluids at low stresses, or more importantly at stresses generated by gravitational forces. In this type of experiment, a stress is applied to the material and the resultant equilibrium shear rate is measured.  

 

The stress is increased stepwise and the viscosity is calculated at each step. Figures 1 (a) (and 1 (b)) show the stress (and shear rate) viscometry curves for two samples of orange barley water.  

A double concentric cylinder geometry was used to enable access to the required low stresses.  

 

The graphs clearly show that sample a has a higher low shear stress viscosity than sample b and is, therefore, predicted to exhibit less sedimentation on storage. It is important to note that from Figure 1 (b), the viscosities are becoming similar at moderate to high shear rates and as a result, these subtle differences would not have been measured using conventional viscometry techniques. 

 * The problem is further complicated since the structure in the neat barley water can not be so high that problems are encountered by the consumer during dilution. 

  Example 2 – Yield stress 

 


Conventionally the ‘yield stress’ of a material has been calculated by generating a shear rate-shear stress curve (rheogram) on a viscometer and then extrapolating to zero shear. The point at which the curve intercepts the stress axis is taken as the yield stress.  

 

This value has been used to predict how the material may behave at low stresses, for example, in pumping the material during processing or application. However, this has been proven not to be entirely satisfactory and can lead to misleading interpretations on the data.  

 

Using a controlled stress rheometer, it is now possible to measure the yield stress directly. In fact, Bohlin rheometers offer an easy yield stress experiment. In this experiment, the stress is ramped from a low value and the resulting deformation is monitored.  

 

The ‘instantaneous’ viscosity is calculated and increases until the stress at which the material starts to flow is reached i.e. the yield stress. At this point there is a catastrophic decrease in viscosity.  

 

Figure 2 shows yield stress data obtained from samples of ordinary and low-cal mayonnaise. The major difference in the formulations is that the low-cal mayonnaise contains more water than in ordinary mayonnaise. Since mayonnaise is an oil-in-water emulsion, it consequently
contains more ‘stabilisers’.  

 

The rheological effect of this illustrated in Figure 2, since it shows that the low-cal mayonnaise requires a higher stress to force it to flow, indicating a more structured or
stable emulsion. 

 Example 3 – Oscillation experiment 

 


Dynamic experiments are used to generate data on the viscoelastic properties of materials. By subjecting the material to a sinusoidal stress and measuring the output sinusoidal strain and the phase-lag, while changing the frequency of the input signal, it is possible to obtain a mechanical
fingerprint for the material.  

 

The frequency-dependent reaction of the material can be translated into timedependency since low frequencies will correlate with long times and vice versa. Figure 3 shows a lot of the frequencydependent viscoelastic functions (G’, G”) of two margarine spread formulations.  

 

At low frequencies (long time scales), the data will give information on how the samples will behave during storage. To avoid syneresis (i.e. separation of the waterin-oil emulsion), it is required that the elastic component (G’) is of a similar magnitude to (or greater than) the viscous component (G”) at long times (low frequencies).  

 

Of the two samples in Figure 3, sample b is the one that exhibited slight syneresis, which is concordant with the previous hypothesis, since at low frequencies G”>>G’ – for sample a, G”~G’.  

 

The ease of spreadability of the two samples can be evaluated by comparing the viscoelastic functions at high frequencies (i.e. short times – consider spreading margarine on bread and the time scale of the shearing action between the knife and the bread).  

 

From the graph it can be seen that sample a has larger moduli values at higher frequencies and the elastic component dominates the viscous component (i.e G’ crosses over G”) at a lower frequency. Sample a would thus be less easily spreadable than sample b.