Expert interview rheology: “It is always a matter of compromise.”
Thomas Mezger: Rheology is a branch of physics in which we look at materials under a certain load. This load is usually a force, but can also be a defined strain or velocity of deformation. Then we can observe that some materials are flowing while others just deform a little and more or less regain their shape when they are relieved. In the first case we talk about liquids, in the second case about solids and elasticity.
From a scientific point of view, there are two important basic laws. The first one defines viscosity and the second one is the law of elasticity. Classically, physics has distinguished between two areas, hydrodynamics for liquids and solid state physics.
We in the world of rheology now bring the two together and describe the deformation and flow behaviour of all materials. This applies to gases, low-viscosity and highly viscous liquids, semi-solid materials up to very solid materials such as reinforced plastics or steel.
Then let’s take a look at the coatings industry. Where can we find coatings in this approach?
Mezger: With coatings we may start with low-viscosity materials. For example, automotive coatings may show ideal-viscous or Newtonian flow behaviour. Once a coating is solidified, we are dealing with solid materials that can be measured, for example in the form of films. But with paints and coatings we usually talk about the processing state, i.e. the liquid or semisolid state.
What role do the forces acting on the material play here, as you mentioned at the beginning?
Mezger: Let’s take the example of an automotive coating again. It is delivered in a container and should then be pumpable for processing. Usually we find a certain structure in the coating at rest which is intended to prevent particles such as pigments from settling. This can be imagined as a very soft gel structure. We have to overcome this so that we can pump the coating. In this case, we are also talking about the yield point, which has to be exceeded. Therefore, it must not be too high. But it must not be too low either, otherwise the particles may settle.
Afterwards, when pumping through a pipe, the coating is in a fluid state. To determine the corresponding shear rate it is important to know the diameter of the pipe and the flow rate. Then we can measure the viscosity under these conditions.
From here we come now to the actual application onto the substrate. In our example this would be the car body. There, the coating should level well, of course, but also it should not sag off too much. Here we consider the so-called thixotropic behaviour. This means that immediately after the application where the material flows quickly, we finally have a structural recovery in the state quasi at rest. So there are typically three things that are important: Yield point, viscosity, thixotropy.
Setting all these properties correctly is certainly not easy when formulating a coating?
Mezger: Yes, the challenge is that the requirements are partially contradictory. For example, the coating should show low viscosity for levelling. On the other hand this can produce too strong sagging on a vertical surface, and therefore, there we need a not too low viscosity value. Rheology is like real life, it is always a matter of compromise. And there is a lot of experience behind it. This is something you cannot learn exclusively from textbooks. And lectures at universities usually lack practical relevance to this topic.
What strategies can be used to achieve the best possible compromise?
Mezger: There is no way around application testing. Rheology is a great help, but in the end there also has to follow always the practical test under varying conditions. It makes a difference whether I want to apply a coating outdoors in Sicily or in Iceland. The temperature, for example, plays a decisive role.
You then have to consult with the coatings laboratory and start “playing” with the additives. After all, there is a certain number of components in the coating and therefore we may find more or less complex interactions that can influence the rheological behaviour.
And additionally, it certainly makes a big difference whether I have a water-based or solvent-based coating?
Mezger: Yes, of course. Solvent-based coatings are generally easier to handle because typically they contain fewer components. Water-based coatings, for example, often need additives to keep them microbiologically stable. There is also another type of interaction, namely the presence of hydrophilic and hydrophobic groups. And each further component, whether intentionally or not, can influence this network via the pH-values or through unintentionally introduced surfactants.
After all, water-based coatings are used primarily for environmental protection reasons to reduce the emission of volatile solvents into the environment. The same applies to powder coatings. What does it mean for the rheology here?
Mezger: Here, the rheological behaviour is completely different, because there is no liquid in front of us at first. There are two possibilities to characterize samples in the form of powder. You can blow air into the powder to fluidise the mixture of powder and air, or you can compress the powder. There are special measuring methods for both types of samples.
When the powder is melted during the coating process, it finally becomes liquid. Here, of course, we are interested in the temperature at which it becomes liquid. At which temperature does it show the lowest viscosity? At the same time, we must be aware that a cross-linking reaction may already be underway. The viscosity must not be too low either, otherwise edges failure may occur, for example on the edge of a table or a similar substrate. These high-temperature measurements can be carried out either at a constant temperature or at rising temperatures.
You have spoken about measurements several times now. Which test methods are the most important from your point of view?
Mezger: Let’s start historically. From the mid-1980s onwards, measurement technology was completely repositioned when personal computers appeared. Before that there were basically only the simple rotational viscometers. They were operated with a spindle in a cup, pre-set was a rotational speed and measured was the resulting rotational torque, and finally from this the viscosity was calculated. This is still the case today for simple quality control. There are ISO and ASTM standards to describe this type of rotational measurements.
Then at some point it became clear that such simple viscosity measurements no longer met today’s requirements. As I mentioned earlier, when you want to pump a coating from a reservoir, it often doesn’t flow immediately because the yield point has to be overcome beforehand. In this case, viscosity as a parameter is not sufficient because it can only describe a substance in the flowing state. Oscillation experiments then became more and more popular because they enabled us to describe the entire viscoelastic properties also in a state of quasi at rest. With these tests we can see which of the two factors predominates, the elastic or the viscous.
This can be easily imagined using two examples from the bathroom. A hand cream, for example, is rather firm but can be easily made to flow. At rest, it is therefore a viscoelastic solid. I cannot completely characterise it with a simple viscometer and a rotational test, because the basic structure would already be destroyed at the first measuring point, and thus, the yield point would be exceeded.
The other example from the bathroom is the shampoo, which is liquid in its state-at-rest, but also has an elastic component and may pull strings if it is moved quickly. We are also familiar with this behaviour of coatings or adhesives. These are therefore viscoelastic liquids.
The oscillation test is particularly interesting if we want to know more than just for simple quality control; for example in research and development. Sometimes three or more measuring intervals are carried out one after the other combined in rotation and oscillation, as so-called step tests to determine the thixotropic behaviour.
What is of particular interest in measurements nowadays are the values of G’ and G”. G’ is the elastic shear modulus or storage modulus, which describes the elastic component of the viscoelastic behaviour. G” stands for the viscous shear modulus or loss modulus. Both values are displayed directly as curve functions in the oscillation test.
And what role do relaxation and creep tests play today?
Mezger: Both are tests that any modern rheometer can perform. In the past, these kinds of tests were carried out more often than today. You can imagine a creep test like this: A force is applied on a sample and then is observed how the sample is creeping, meaning how the deformation values develop over time. This typically results in a characteristic curve in the shape of an e-function. When the force is then removed, we can observe whether, or with which shape
of the measuring curve and how far the deformation values decrease again, the latter showing the degree of elasticity of the sample.
The disadvantage of the creep test is that you can only carry out a single test at a time and on a single load level. This means that it takes several tests to describe a sample, and each single creep test takes in minimum around ten minutes. An oscillatory test is much faster because it delivers a meaningful overall result in just 10 or 15 minutes. This is why nowadays creep tests are actually only used in basic research.
The situation is similar with the relaxation test. It is carried out in such a way that a material is suddenly deformed and then left to rest. Afterwards, one can see how the inner structure regenerates in the form of the shear stress values over time, i.e. how it relaxes. Also here the measurement curve usually shows the form of an e-function. Here too, measurements are only taken at a single load at a time, here in the form of a constant deformation, and several measurements would be needed to describe the behaviour of a test sample.
Are there actually typical measurement errors that you are repeatedly asked about?
Mezger: It is clear that a rheometer has more operating possibilities than a thermometer, for example. We should know what we want to achieve before we create the measurement program. Do we want to explore or to simulate the thixotropic behaviour or the pumping behaviour? Do we want to simulate the result if we apply a coating manually or with machine assistance? Possibly this is already a source of error, as we cannot measure anything relevant to practice if the specifications are not suitable.
That is what we are here for, that is our daily business. Here in the support department we help our customers to carry out the correct measurements. For example, we have already implemented a lot of helping tools in the software as templates.
But you should also remain pragmatic. For example, when I test a solvent coating in a plate-plate measuring geometry I find that the edges of the coating come into contact with air, so the coating can dry on, which naturally has a big impact on the measurement. The measuring temperature also plays a big role. If I measure for too long at very high shear rates, I generate additional heat in the test sample through internal friction between the molecules. With some experience, however, such errors can usually be recognised, e.g. via the course of the measurement curve.
You have written much of this expertise down in the rheology handbook. How did this come about?
Mezger: That goes back relatively far. Rheology initially had a bad reputation with me because the lectures at the university were quite dry and exclusively theoretical. In the 1980s, I started writing down typical questions and problem solutions from day-to-day practice and displayed them as examples using measurement curves. This resulted in the booklet “A Short Rheology Course” on about 100 typewritten pages. This was the basis for the first edition of the German version (Das Rheologie-Handbuch) which was published in 2000, followed by the English version of The Rheology Handbook in 2002.
The fifth edition has just been published, what has changed there?
Mezger: One example is the new chapter on “Shear tests with powders and bulk solids”. This is not only relevant for powder coatings, but also plays a role in the field of additives, for example, which are often supplied as powders or of powder-like materials which are used for additive manufacturing. The book also contains many new practical examples of typical daily questions, with the aim of presenting the subject of rheology in a comprehensible way.
It is also important for me to point out that the book not only contains my personal knowledge, but also that of other experts like my colleagues. For example, we are members of various standards committees (ISO, ASTM, DIN, etc.), we have daily international contact to the most diverse applications in practice through our customers, and we are constantly exchanging our experiences in a lot of discussions.