Strategies to avoid the pump-out effect
To minimize the pump-out effect, various strategies are required that take into account both the material properties of the thermal compound and the conditions under which it is used. A central starting point is the optimization of the matrix materials. The story of ever higher thermal conductivities has already been almost completely told, but there is still a real need for innovation in this area. Polymers with higher thermal stability and a lower coefficient of thermal expansion can reduce stresses, but are complex and expensive to develop. The DOWSIL pastes TC-5888 and the newer TC-5550 and 5960 are very good examples of this.
At the same time, chemical modifications, such as the introduction of cross-linking in the polymer structure, can increase the mechanical stability of the paste. Such cross-linking prevents the matrix from liquefying under high temperatures and being displaced laterally. We certainly remember my olfactory excursion with a Korean thermal paste. The choice of filler particles also plays an important role. Particles with a coefficient of thermal expansion closer to that of the matrix reduce the stresses between the two phases. The even distribution of the particles in the matrix also prevents the formation of weak points that could promote the pump-out effect. This also includes the particle sizes, which should not be too large.
Additives can further improve the properties of the paste. Thixotropic additives increase the viscosity of the matrix, especially at high temperatures, and make it more resistant to shear forces. Chemical adhesion promoters can improve the adhesion of the paste to the contact surfaces, making it more difficult to displace from the contact area. Such adhesion promoters ensure that the matrix adheres to the surfaces even at high temperatures and under mechanical stress.
Optimizing the layer geometry is another approach to minimize the pump-out effect. A uniform, thin layer not only reduces the thermal resistance, but also the mechanical stresses within the paste. Precise methods such as stencil printing can be used for application in order to precisely control the layer thickness. We recently discussed this very topic.
An alternative concept is the use of materials that eliminate the pump-out effect in principle or reduce it to a minimum. These include PTM (Phase Transistion Material), which I have already described often enough and which harden at operating temperatures and thus become mechanically more stable and, yes, are very promising. Such materials offer the advantages of a paste in adapting to surface irregularities, but retain their position after curing. Solid-state thermal interface materials, such as graphite foils or metallic pads, completely eliminate the displacement problem as they have no viscous matrix, but have other disadvantages such as very high interface resistances, which quickly puts the high nominal thermal conductivity into perspective.
Practical implementation and challenges
The implementation of the above strategies requires careful consideration of thermal conductivity, mechanical stability and practical applicability. Materials with improved mechanical properties could have disadvantages in thermal conductivity, as a higher particle density is required to reduce mechanical stresses. Likewise, additives or chemical modifications can influence the ageing properties of the paste, which could limit its service life. Another problem is the adaptation to real operating conditions. While laboratory tests focus on specific stresses such as temperature cycles or shear forces, in practice several factors often act simultaneously. Simulations and tests that take combined load scenarios into account are therefore essential to evaluate the actual performance of new paste materials.
Viscosity and particle dispersion – how not to do it
A less viscous silicone oil generally leads to faster solidification when filled with particles compared to a highly viscous oil such as the DOWSIL TC-5888. This effect can be explained physically and chemically and can even be illustrated with a practical example from the food industry: stirring oat flakes into milk or yoghurt. But I’ll come to that in a moment, because we’ll eat later. Let’s start with some theoretical basics first. You’ll have to get through that now.
Viscosity is a measure of the resistance of a liquid to flow. A highly viscous liquid flows slowly and creates great resistance to movement in its structure. A less viscous fluid has less resistance and flows more easily. In terms of silicone oil, a low viscosity means that the intermolecular forces are lower and particles in the liquid can be mixed more easily and quickly. This has advantages (lower costs, as it is quicker to produce), but also disadvantages. Which brings us back to my research, which confirms that KOLD-1 has a base matrix with a significantly lower viscosity.
When particles are added to a less viscous silicone oil, they can disperse relatively easily without much resistance to their movement. Due to the easier dispersion and the faster onset of the sedimentation process due to the lower viscosity, the particles adhere to each other more quickly and form more stable cross-links. This enables faster aggregation, which leads to rapid curing to the desired degree of strength. The disadvantage is that with the same degree of filling with thermally conductive particles and an otherwise identical mixing ratio, a significantly firmer, more viscous paste is produced.
In a highly viscous liquid such as the base matrix of the DOWSIL TC-5888 or the TC-5550, it takes longer for the particles to achieve a homogeneous distribution due to the greater friction and lower fluidity of the liquid. As a result, the particles need more time to arrange themselves in the spaces between the liquid and form stable structures. As a result of this process, the highly viscous silicone oil hardens more slowly and has a significantly longer processing time. This means that the strength of the material is built up more slowly, which is associated with higher costs due to a significantly slower manufacturing process, but also results in a better product in the end.
The right hydration process as a guarantee for quality and performance
Let’s illustrate this with oat flakes that are added to milk or yoghurt. Oat flakes that are added to a less viscous liquid such as milk absorb the liquid more quickly and therefore swell faster. Due to its low viscosity, milk spreads more quickly between the oats, which simplifies the interaction between the oats and the liquid and speeds up the absorption of the liquid. In a short time, the oat flakes become soft and the mixture thickens. The situation is different with oat flakes that are added to a highly viscous yoghurt. The higher viscosity of the yoghurt makes it more difficult to move and distribute the liquid around the oat flakes, which slows down the swelling process. The oat flakes need more time to be completely absorbed by the liquid and the mixture reaches its solid consistency more slowly.
This comparison shows that less viscous liquids allow faster penetration and arrangement of the particles due to their lower intermolecular resistances. For technical applications, this means that a less viscous silicone oil is advantageous when rapid curing and strength development is desired to save costs and time. High-viscosity silicone oils, on the other hand, are suitable when a longer processing time is acceptable, for example to achieve homogeneous distribution in more complex applications. These are also the better pastes, which of course cost more to produce, but also last significantly longer.

Oil binding capacity and silicone oil viscosity as an auxiliary tool for durability and performance
The right balance between the oil-binding capacity of a filler and the ratio to the total proportion of silicone oils used is crucial for the quality and properties of the end product. The oil-binding capacity of a filler is significantly influenced by its purity, surface structure and the particle size mentioned above. Pure and fine-grained fillers offer more surface contact and thus increase their ability to bind oil. Coarser particles, on the other hand, offer less contact surface and are often less effective at binding oil. Due to this correlation, the choice of filler directly influences the amount and viscosity of silicone oil required to achieve a stable and homogeneous mixture.
In practice, this means that the viscosity spectrum of the silicone oil used must be adjusted in order to optimally integrate the respective filler. A range of 1,000 to 250,000 cSt (centistokes) is common, depending on whether a higher or lower viscosity is required to achieve a stable and homogeneous dispersion. Fillers with a high oil-binding capacity may require lower viscosities as the particles tend to absorb and disperse the oil better, allowing for faster solidification. On the other hand, higher viscosity oils may be required if the oil binding capacity is lower or if a longer working time is desired, as these embed the particles more slowly and stabilize the mixture. Or you can do it like the manufacturer of the KOLD-01, who has optimized the particle sizes to the viscosity of the less viscous silicone oil. You have to come up with that first. Chapeau, and in all seriousness.
Normally, a mixture of different viscosities is often used instead of a fixed viscosity in order to achieve the desired physical and chemical properties of the material. The combination of different viscosities in the original makes it possible to control both the processing time and the curing speed of the material. The advantage of such a mixture is that a dynamic balance can be achieved between the rapid dispersion of the particles and the desired solidification. If you only have cheaper oils, you also have to control this effect via the particle sizes.
The precision in the coordination of oil binding capacity and silicone oil viscosity contributes significantly to the performance of the end product; this applies equally to both pastes, whether original or replica. Too much filler can lead to the oil binding capacity being exceeded and the mixture becoming unstable or not achieving the desired hardness. Too little filler, on the other hand, may cause the material to become too soft, as there are not enough particles to form a strong network with the silicone oil. A targeted adjustment of the mixing ratio between filler and oil, taking into account the optimum viscosity, ensures that the end product has the desired mechanical properties, such as strength, elasticity and stability, which are required for the respective application.
On the next and last page, we’ll explain how this works.
47 Antworten
Kommentar
Lade neue Kommentare
Mitglied
1
1
Mitglied
Urgestein
1
Mitglied
Veteran
1
Mitglied
1
Veteran
Urgestein
1
Mitglied
Urgestein
Veteran
Urgestein
Mitglied
Alle Kommentare lesen unter igor´sLAB Community →