Technical University of Dortmund Facultyof Biochemical and Chemical Engineering

Report Lab Experiment SM 4 Viscometry Newtonian and Non-newtonian Fluids Chair of Fluid mechanics

Group Leandro Michieleto Mei Hai Tao

May 2012

1. Objectives Objective for this experiment is the introduction of the basics of measurements of viscous properties of Newtonian and Non-newtonian fluids and indirectly the introduction of the fluid mechanics theory approaching these topics, introducing and providing knowledge and first-hand experience of a real lab Haake viscometer.

2. Introduction 2.1 Viscosity Viscosity is the measure of the internal friction of a fluid. This friction becomes apparent when a layer of fluid is made to move in relation to another layer. The greater the friction, the greater the amount of force required to cause this movement, which is called shear. Shearing occurs whenever the fluid is physically moved or distributed. Highly viscous fluids, therefore, require more force to move than less viscous materials.

Isaac Newton defined viscosity by considering the model represented in the figure above. Two parallel planes of fluid of equal area A are separated by a distance dx and are moving in the same direction at different velocities V1 and V2. Newton assumed that the force required to maintain this difference in speed was proportional to the difference in speed through the liquid, or the velocity gradient. To express this, Newton wrote: where h is a constant for a given material and is called its viscosity. The velocity gradient, dv/dx, is a measure of the change in speed at which the intermediate layers move with respect to each other. It describes the shearing the liquid experiences and is thus called shear rate. This will be symbolized as S in subsequent discussions. Its unit of measure is called the reciprocal second (sec-1). The term F/A indicates the force per unit area required to produce the shearing action. It is referred to as shear stress and will be symbolized by F. 2.2. Types of Fluids and Viscosity Behavior The type of flow behavior Newton assumed for all fluids is called, not surprisingly, Newtonian. It is, however, only one of several types of flow behavior you may encounter. A Newtonian fluid is represented graphically in the figures below. Figure 1 shows that the relationship between shear stress (F) and shear rate (S) is a straight line. Figure 2 shows that the fluid's viscosity remains constant as the shear rate is varied. Typical Newtonian fluids include water and thin motor oils.

What this means in practice is that at a given temperature the viscosity of a Newtonian fluid will remain constant regardless of which Viscometer model, spindle or speed you use to measure it. A non-Newtonian fluid is broadly defined as one for which the relationship F/S is not a constant. In other words, when the shear rate is varied, the shear stress doesn't vary in the same proportion (or even necessarily in the same direction). The viscosity of such fluids will therefore change as the shear rate is varied. This measured viscosity is called the apparent viscosity of the fluid and is accurate only when explicit experimental parameters are furnished and adhered to. Non-Newtonian flow can be envisioned by thinking of any fluid as a mixture of molecules with different shapes and sizes. As they pass by each other, as happens during flow, their size, shape, and cohesiveness will determine how much force is required to move them. At each specific rate of shear, the alignment may be different and more or less force may be required to maintain motion. There are several types of non-Newtonian flow behavior, characterized by the way a fluid's viscosity changes in response to variations in shear rate. The most common types of non-Newtonian fluids it can be may encountered are: Psuedoplastic This type of fluid will display a decreasing viscosity with an increasing shear rate, as shown in the figure below. Probably the most common of the non-Newtonian fluids, pseudoplastics include paints, emulsions, and dispersions of many types. This type of flow behavior is sometimes called shear-thinning. Dilatant Increasing viscosity with an increase in shear rate characterizes the dilatant fluid; see the figure below. Although rarer than pseudoplasticity, dilatancy is frequently observed in fluids containing high levels of deflocculated solids, such as clay slurries, candy compounds, corn starch in water, and sand/water mixtures. Dilatancy is also referred to as shearthickening flow behavior. Plastic This type of fluid will behave as a solid under static conditions. A certain amount of force must be applied to the fluid before any flow is induced; this force is called the yield value. Tomato catsup is a good example of this type fluid; its yield value will often make it refuse to pour from the bottle until the bottle is shaken or struck, allowing the catsup to gush freely. Once the yield value is exceeded and flow begins, plastic fluids may display Newtonian, pseudoplastic, or dilatant flow characteristics. See the figure below.

2.3. Laminar and Turbulent Flow The very definition of viscosity implies the existence of what is called laminar flow: the movement of one layer of fluid past another with no transfer of matter from one to the other. Viscosity is the friction between these layers. Depending on a number of factors, there is a certain maximum speed at which one layer of fluid can move with relation to another, beyond which an actual transfer of mass occurs. This is called turbulence. Molecules or larger particles jump from one layer to another and dissipate a substantial amount of energy in the process. The net result is that a larger energy input is required to maintain this turbulent flow than a laminar flow at the same velocity. The increased energy input is manifested as an apparently greater shear stress than would be observed under laminar flow conditions at the same shear rate. This results in an erroneously high viscosity reading. The point at which laminar flow evolves into turbulent flow depends on other factors besides the velocity at which the layers move. A material's viscosity and specific gravity as well as the geometry of the Viscometer spindle and sample container all influence the point at which this transition occurs. Care should be taken to distinguish between turbulent flow conditions and dilatant flow behavior. In general, dilatant materials will show a steadily increasing viscosity with increasing shear rate; turbulent flow is characterized by a relatively sudden and substantial increase in viscosity above a certain shear rate. The material's flow behavior may be Newtonian or non-Newtonian below this point. 2.4. Factors that affect Rheological Properties Temperature - one of the most obvious factors that can have an effect on the rheological behavior of a material is temperature. Some materials are quite sensitive to temperature, and a relatively small variation will result in a significant change in viscosity. Others are relatively insensitive. Consideration of the effect of temperature on viscosity is essential in the evaluation of materials that will be subjected to temperature variations in use or processing. Shear Rate - non-Newtonian fluids tend to be the rule rather than the exception in the real world, making an appreciation of the effects of shear rate a necessity for anyone engaged in the practical application of rheological data. When a material is to be subjected

to a variety of shear rates in processing or use, viscosity measurements should then be made at shear rates as close as possible to the estimated values. 2.5. Experimental Apparatus Viscosity is measured with various types of viscometers and rheometers. A viscometer is used for those fluids which cannot be defined by a single value of viscosity and therefore require more parameters to be set and measured. A rotational speed is preset and the flow resistance of the sample is measured in other words, the torque maintaining the set speed is proportional to the viscosity. All final information on the viscosity, shear stress and the shear rate is calculated from the torque required, the set speed and the geometry factors of the applied sensor. They measure the torque required to rotate a disk or bob in a fluid at a known speed. 'Cup and bob' viscometers work by defining the exact volume of a sample which is to be sheared within a test cell. The torque required to achieve a certain rotational speed is measured and plotted. There are two classical geometries in "cup and bob" viscometers, known as the Couette systems - distinguished by whether the cup or bob rotates. In this apparatus, one of the cylinders is rotating at a fixed angular velocity.

Figure 3 Example of Rotational Viscometer and its chamber 3. Results and Discussion 3.1. Experiment 1: Choice of the most appropriate head In order to achieve good and trustworthy data from the equipment for the upcoming measurements, an experiment was carried to choose the best and most appropriate head of measurement for the applied conditions. Three heads were used, MVI, MVII, MVIII, for both liquids. A compete flow curve shear stress vs. Shear rate, achieved by the equipment by varying rotation speed, for each liquid was recorded and is presented below. The experiments were conducted at room

temperature according to the conditions described in the provided script, and also by varying the damping factor. The results presented are already with the damping factor set to zero.

1.2 1

Shear Stress (N/m2)

0.8 0.6 0.4 0.2 0 0 0.5 1 MV1 MV2 MV3

Shear Rate (1/s) Figure 4 Shear stress vs. Shear Rate for the different equipment heads liquid: GlycerolWater From the graph, we can conclude that, the head MVIII is more adequate for the upcoming measurements. As it can be seen in the picture, the values provided by the MVIII are smoother, and have a linear behaviour with the variation of shear stress. For the other 2 heads, there it is possible to visualize a region where the behaviour is not linear, or present some strong oscillations, what could cause errors and provide bad reproducibility of the data if those were used. As described, the same experiment was conducted for CMC-Solution and the data are presented in the graph below.

1.2

Shear Stress (N/m2)

1 0.8 MV1 0.6 0.4 0.2 0 0 0.5 1 MV2 MV3

Shear Rate (1/s) Figure 5 Shear stress vs. Shear Rate for the different equipment heads liquid: CMCSolution

Similar as for water-glycerol, we can conclude that, the head MVI is more adequate for the upcoming measurements. As it can be seen in the picture, the values provided by the MVIII are smoother, and have a linear behaviour with the variation of shear stress. For the other 2 heads, there it is possible to visualize a region where the behaviour is not linear, or present some strong oscillations, what could cause errors and provide bad reproducibility of the data if those were used. 3.2. Experiment 2 Dependence of Shear Stress and Shear Rate for different Temperatures The best suitable measurement head is used to measure the stress in dependence of the time which shall be plotted for the temperatures T = 30 C / 40 C / 50 C /60 C. TO this purpose, a curve shear stress vs. time (-t) was recorded for rotational frequencies 10, 20, 30, 40, 50, 60, 70, 80, 90 and 100% (Dmax=500 s-1), for each liquid and for the temperatures 30, 40, 50 and 60 C, in order to obtain data for the viscosity of each liquid with relation to temperature and shear stress, according to the settings described in the lab script provided. The results apre presented in the graphs and tables below.

3.2.1. Shear Rate and Shear Stress Dependency For the described experiment, the values of viscosity vs time for all temperatures and times were measured and the results are presented below. 3.2.1.1. CMC Solution Carboxymethylcellulose is a modified polymer derived from cellulose; which has a complex tridimensional polymeric structure. For such substances, a non-Newtonian behavior could be expected due to the complex intermolecular interactions in the bulk phase solutesolute and solute-solvent. As can be seen through the graph, the CMC-solution does not present a curvature in the relation shear rate vs. shear stress, neither shear-thickening nor shear-thinning behavior of non-Newtonian fluids, as it could be expected for this kind of solution. In contrast, it is possible so see n coefficients near to 1, as listed in the table below, what describe a linear behavior of the function and a constant viscosity.

140

shear stress[N/m^2]

120 100 80 60 40 20 0

y = 14.87x0.9524 y = 14.603x0.8961 y = 11.584x0.9638 y = 10.201x0.9384 T60 T50 T40 T30

Figure 6 Shear stress over shear rate, for different temperatures - CMC-solution

Besides that, it can be seen through the huge deviation of the values for the series 40C that the data does not represent the real behavior of the fluid, observed through strong oscillations almost between each point. For all other temperatures, the first 8 points have a relative exact relation between each other, and an exact linearity is observed. For all temperatures, the values for 90 and 100% present strong oscillations, which make the series lose its linearity. The threshold of 80% here can be easily observed and for any reason, due to any kind of error that cannot be investigated in the moment, the points after it could even be neglected from these experimental data. The experiment shows itself not capable of generating realistic data for these values.

Table1: Values for k and coefficient n for CMC solution T 30 40 50 60 K 14.87 14.603 10.722 10.201 n 0.9524 0.8961 0.9781 0.9384

3.2.1.2. Water-Glycerol Solution As can be seen from the picture below, the results from the solution Water- glycerol cannot be considered reasonable. From the picture, no particular behavior can be detected, neither Newtonian nor non-Newtonian. For the proposed system, it should be expected a

51

shear rate[1/s]
01 5

linear behavior for the relation shear stress/shear rate. The trend lines for linear regression were added for comparison, but the oscillations of the points of each series make the analyse and interpretation of the fluid flow profile difficult. No assumption can be made with relation to characteristics of the fluid.

90

shear stress[N/m^2]

80 70 60 50 40 30 20 10 0

y = 5.9795x + 23.576 y = 5.2872x - 11.333 y = 8.131x - 5.0367 y = 4.3641x - 9.7761 T30 T40 T50 T60

Figure 7 Shear stress over shear rate, for different temperatures Water-glycerol solution

Table2: Values for k and coefficient n for CMC solution

T 30 40 50 60

K 21.769 6.6606 0.9297 1.3218

n 0.576 1.0279 1.463 1.3934

3.2.2. Viscosity and Shear Rate Dependency 3.2.2.1. CMC Solution The values of viscosity with relation to the variation of shear rate for CMC-Solution are presented in the graph below.

21

01

shear rate[1/s]
8 6 4

140 130 120 viscosity[mPa/s] 110 100 90 80 70 60 50 shear rate[1/s] T4 0 T3 0

Figure 8 Values of viscosity in dependence of Shear rate applied for different temperatures CMC-solution. In contrast to the behavior presented in the last graph for this solution, it is clear to see that the carboymethylcellulose solution, as expected, shows a non-linear behavior of shear stress and shear rate, clearly seen though the non-constant behavior of the viscosity, as shown here. Although as well with high deviation from ideal, or expected, behavior, it is possible to identify a 2nd.order polynomial profile, characteristic of a shear-thinning fluid with n<1, which supports the assumption of a non-Newtonian fluid. 3.2.2.2. Water-Glycerol Solution Following the pattern of the previous graph, especially because these data are interdependent, the relation between viscosity and shear rate shows again an inconsistency with the literature for aqueous solution. The values of viscosity should stay constant along the shear rate variation, showing an Newtonian behavior for this relation. Nevertheless, no conclusion can be made for this experiment regarding the dependency of viscosity and shear stress for the data obtained and analysed for this experiment. It should also be mentioned that, in the analysis of the data coming from the curves of 50 and 60 C, if considered just the initial interval of 0 and 80%, a relative flat relation can be observed, reinforcing the theory of Newtonian fluids. Nevertheless, the experiment should be carried again and new data for the 30 and 40 C series should be obtained.

0001

005

200

viscosity[mpas/s]

180 160 140 120 100 80 60 40 20 0

T30

T40

T50

shear rate[1/s]

Figure 10 Values of viscosity in dependence of Shear rate applied for different temperatures Water glycerol solution.

3.2.3. Temperature Dependency of Viscosity The temperature dependency of viscosity is shown the graphs below. It is possible to see can see that for both Newtonian and Non-Newtonian fluids, the viscosity decreases with temperature, but the difference should be higher for lower temperatures, as observed in the graph viscosity vs. Shear rate.

3.2.3.1. CMC Solution It can be seen that the curves present the expected behavior, for which viscosity decreases with temperature for a constant shear rate. Although not all the curves are correctly related to each other in a decreasing order, as it should be expected, they do present a decreasing profile, when the trend-line is analysed. It cannot be seen from this picture the correct shear rate/viscosity behavior. It should be expected a decreasing viscosity with increasing shear rate, but, since the values of viscosity vary smoothly with shear rate, small experimental deviation of the values can lead to an overlap of the next point.

0051

0001

005

0.14

viscosity[Pa/s]

0.13 0.12 0.11 0.1 0.09 0.08 0.07 0.06

117 234 351 468 585 702 819 936

Temperature[C]

Figure 11 Temperature dependency of viscosity for constant shear rates CMC solution

3.2.3.2. Water Glycerol Solution Since all the values from this calculation sequence are linked, the values for temperature dependency of water also show an inconsistency, with strong oscillations inside each series and not following the normal pattern of decreasing viscosity with increasing temperature. Again, no conclusion can be taken from these data.

viscosity[mPa/s]

90 80 70 60 50 40 30 20 10 0

T[C]

Figure 12 Temperature dependency of viscosity for constant shear rates Water glycerol solution

56 56

55 55

54 54

53 53

52 52

1053 1170

117 234 351 468 585 702 819 936 1053

3.2.4. Activation Energy Calculation The activation energy is than calculated using the Arrhenius equation by a graphical method. Plotting ln() vs 1/T, the slope of the curve is numerically equal to -Ea/R and thus the activation energy can be calculated. 3.2.4.1. CMC Solution The values obtained for CMC Solution, for each shear rate are presented below, for the slope and the calculated activation energy.

4.9

ln(viscosity) [mPa/s]

4.8 4.7 4.6 4.5 4.4 4.3 4.2 0.0029 0.003 0.0031 0.0032 0.0033 0.0034

117 234 351 468 585 702 819 936 1053 1170

1/T [1/K] Figure 13 ln (viscosity) vs 1/T for different shear rates, for the determination of Ea CMC solution.

Table 2 Values for the slope of the curves and calculated Activation energy for shear rate / CMC Solution Shear Rate 117 234 351 468 585 702 819 936 1053 1170 Slope Ea/R 3245.57 3041.87 2790.88 2690.12 2542.1 2450.9 2305.6 2398.36 2008.53 2200.85 Ea (J/mol) 26983.669 25290.1072 23203.3763 22365.6577 21135.0194 20376.7826 19168.7584 19939.965 16698.9184 18297.8669

3.2.4.2. Water Glycerol solution The values obtained for CMC Solution, for each shear rate are presented below, for the slope and the calculated activation energy.
0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0 117 234 351 468 585 702 819 936 1053 1170

viscosity[Pa/s]

Temperature[C]

Figure 14 ln (viscosity) vs 1/T for different shear rates, for the determination of Ea Water Glycerol Solution solution. Since all the values obtained for water until now make no physical sense, it has no physical fundament the calculation of the values of activation Energy for this system.

4. Conclusion Unfortunately, not all investigations proposed in the script could be investigated with the data obtained from the actual experiment and expressive deviations from the reality were observed. Actually, we have not enough information to prove the origin of deviations between experimental and theoretical data appeared along the experiment and further investigations would be needed, in order to identify the real sources of error. Only new investigations could lead to consistent results, and maybe the identification of possible mistake sources that could be have been avoided. Though, it is expected in any case to lie basically on 3 factors: measurement errors of conduction in the experiment, exactness of the analytical method chosen and/or equipment errors. All the possible source of errors could normally be associated with didactical lab experiments conducted by students. Nevertheless, the experiment provides the students expressive didactical fundaments and is able to pass concise knowledge in the field of fluid mechanics.

56

55

54

53

52