CAYOM CA-4 vs. BOLIGO Z980 Thermal Pastes Review – More than just similar bargains?

1 - Introduction and overview 2 - Shear behavior, microscopy, and chemical composition 3 - Lab measurements, interface resistance and bulk thermal conductivity 4 - Real performance on CPUs, notebook chips, and GPUs 5 - Durability (pump-out), summary, and conclusion

Tear-off behavior and shear structure

In the macroscopic tear-off image, the CA-4 shows a fairly homogeneous, fine-grained filler image with well-distributed particles. The horizontal shear lines are clearly pronounced and evenly distributed over the entire surface, which indicates that the paste was continuously deformed during rolling. The grooves are narrow and regular, which is an indication of a rather soft matrix with a fine grain distribution. The tear-off edges appear frayed in places, but at the same time not abrupt. The transition from completely covered areas to torn segments is gradual. This indicates a moderate cohesive behavior of the paste. The material does not break off abruptly, but retracts in a kind of thread structure, which remains relatively short. In addition, small cohesive islands are visible, which have stuck to the opposite side during lifting. The tear-off pattern is largely structurally closed, without deep holes or larger break-outs.

In the detailed image, the surface of the CA-4 appears finely granulated with dense filler packing. The particles lie close together, but do not show any noticeable agglomerates. The distribution is relatively even, whereby the shear lines in detail result in a smooth, textured topography. The matrix shows no recognizable dry spots or separated phases.

The macroscopic tear-off image of the Z980 shows clearly more pronounced and irregular tear-off structures. The transitions between the covered surface and the bare substrate are sharper-edged and sometimes appear blotchy. The material detaches in larger clusters and leaves behind irregular islands that show neither a clear thread formation nor a uniform fracture edge. Although the shear lines are present, they appear coarser and less homogeneous than on the CA-4. The Z980 shows open areas in several places where the paste is completely missing, while neighboring segments show thicker material residues. This suggests that the cohesion within the paste body is less uniform than in the CA-4. The tearing of larger particle bundles is also visible, which do not detach smoothly from the surface.

The detailed image shows a somewhat coarser grain structure. The particles appear larger, unevenly distributed and sometimes standing together in groups. The topography is much more uneven. Small valleys appear between the grains, in which the matrix appears thinner. The overall surface effect is significantly more granular and less closed. Despite the differences, there are some clear similarities that can be derived directly from the images.

Both pastes have a silicon-free, oxide-ceramic filler base and show visually similar spectral reflection behavior. Both pastes develop a horizontal line structure in the shear phase, which indicates a comparable displacement behavior in the matrix. The flow direction is similar for both pastes and follows the rolling movement, which is clearly visible in the images. Both pastes show small coherent residues at the tear-off edges, which is an indication of the typical behavior of classic silicone greases with mineral filler.

The shear structure of CA-4 appears finer, more homogeneous and cohesively more stable. The tear-off edges are softer and show a more consistent material behavior over the entire surface. The Z980 shows a coarser grain distribution, more uneven tear-off edges and an overall less uniform distribution of the filler. The similarities lie primarily in the basic material class and the way in which both pastes align themselves during shearing, while the differences are particularly evident in the fine pattern and the tear-off pattern.

Particle morphology and dispersion

The first image of the CA-4 shows a comparatively fine and narrowly graded particle distribution. The measured particle sizes are predominantly between around 3 and 6 µm, with individual values reaching just under 8 µm. This range is relatively narrow, which indicates a well-sorted fractionation. The particles predominantly appear approximately round to slightly irregular, but without pronounced corners or elongated structures. It is also noticeable that hardly any large outliers are recognizable, which suggests a uniform mixing and dispersion quality.

The particles sit densely and relatively homogeneously in the matrix. There are only small areas between them where the binder phase is exposed. The surface appears closed, finely structured and has hardly any recognizable clusters. Although there are individual areas where particles are closer together, these appear more random than systematically clumped. Consequently, the paste shows an almost isotropic distribution without recognizable agglomeration nuclei. The morphology of the filler particles suggests a rather smooth, finely crystalline or ceramic-looking material. The reflection points are small, numerous and evenly distributed. This suggests that the particles are present in large numbers and with a very small average size, which in turn explains the low surface roughness of the paste. The dispersion appears to be careful, as the particle size distribution is close together and the matrix shows no visible cleavage zones or segregation.

In direct comparison, the Z980 shows a significantly broader and more uneven particle spectrum. The measured values are between around 4 and just under 10 µm, sometimes even higher. This range is noticeably wider than with the CA-4, and the particles appear coarser overall. In addition to smaller particles, larger grains regularly occur, which stand out clearly in the image. The morphology is more irregular, the particles appear less round and more polygonal to lumpy. In some cases, they appear in aggregated form, which indicates that the dispersion was not completely homogeneous.

Visible groupings can be seen in the matrix. Many particles are closer together than others, resulting in small-scale clusters. These can be recognized by areas with a high particle density and adjacent zones in which the binder phase is much more prominent. This indicates less uniform mechanical mixing. As a result, the surface appears more grainy and uneven overall, which also explains the coarser shear structure that was already visible in the tear-off behavior. The reflections are larger and more pronounced than in the CA-4, suggesting that the individual particles are not only larger but also have a more complex surface structure. The wider size distribution also leads to less dense packing. Small depressions open up between the particles, in which the binder phase is visible, making the surface appear more heterogeneous in comparison.

Both products use clearly recognizable irregularly shaped, hard-looking filler particles embedded in a viscous matrix. Both show a tendency towards isotropic distribution in the surface, whereby the CA-4 maintains this isotropy much more consistently. The basic morphology is comparable, which suggests the same general class of material, without being able to specify further. The optical reflection character of both shows a clear multiphase structure of typical ceramic or oxide systems. The CA-4 has a finer granular particle structure with a narrow size distribution and visible uniformity in the embedding. In contrast, the Z980 appears more heterogeneous, coarser and more aggregated. These differences have a plausible effect on the mechanical shear structure, the tear-off behavior and potentially also on the resulting BLT stability, as finer and more homogeneously dispersed particles reduce the risk of abrupt loss of cohesion.

LIBS analysis of the CA-4

In the LIBS analysis, the CA-4 shows a composition consisting of five clearly detected main elements. Aluminum is the highest with about 32.5 percent, closely followed by oxygen with about 31.2 percent. Zinc is also clearly represented with just under 19.4 percent, while silicon has a share of around 13.7 percent. Hydrogen is at around 3.2 percent. Carbon is not detected in the examined area, or not in relevant quantities. The element distribution is thus clearly oriented towards oxidic, inorganic-dominated filler systems, with aluminum and oxygen appearing as the strongest signals. The peaks in the spectrum are clearly separated from each other and show a calm, homogeneous image without broad mixed signatures. The intensities appear consistent, indicating a uniform matrix that shows no pronounced segregation in the measurement range. The lack of carbon signature is noticeable, but is within what can occur in highly pigmented pastes with a high proportion of heavy, inorganic particles when the binder phase is covered in the analysis point. The measuring points are densely placed in the image and show a largely constant element distribution.

LIBS analysis of the Z980

The Z980 shows a clearly different elemental pattern. Here, too, aluminum and oxygen are dominant components, although the aluminum content of around 28.2 percent is lower than in the CA-4, while oxygen remains at a similar level at around 29.5 percent. The significantly higher carbon content of around 17.7 percent, which was not noticeable in the CA-4, is remarkable. Zinc is slightly lower than in the CA-4 at 15.8 percent, while silicon is also lower at 5.2 percent. At around 3.6 percent, hydrogen is in the same range as in the CA-4. The spectrum shows more variation, particularly in the carbon lines. The particle structure is significantly more granular and heterogeneous in the optical image of the sample, which corresponds to the broader particle size distribution. The higher C signature clearly indicates that the surface of the Z980 in the measurement field is more strongly characterized by organic components, be it through a thicker binder phase or through carbon-containing surface residues between the mineral particles. The decisive factor is that this difference is clearly measurable here and does not emerge from the CA-4 data. The other elements are less evenly distributed than in the CA-4. The variations between the measurement points appear greater, which is not surprising due to the previously observed uneven dispersion of the fillers.

Comparison of both pastes

The two pastes differ visibly in their elemental composition, although they appear superficially similar at first glance. The CA-4 has a clearly inorganic-dominated profile with a high proportion of aluminum, oxygen and zinc, while silicon and hydrogen occur in moderate quantities and carbon is practically absent. The element distribution appears homogeneous and stable. This uniformity confirms the previously observed fine, dense and well-dispersed particle topography.

In contrast, the Z980 shows a much more hybrid composition. In addition to aluminum, oxygen and zinc, carbon appears in a relevant, measurable concentration. This point immediately distinguishes it from the CA-4. Silicon is also visibly lower, which is reflected in the somewhat less compact and coarser particle structure. The overall elemental picture appears more heterogeneous, which corresponds to the visually observed stronger segregation of the particles. Thus, the CA-4 is clearly an inorganic, highly filled paste with a very homogeneous distribution, while the Z980 shows a combination of mineral and carbonaceous components, which are also less evenly distributed. This difference is clearly visible in the LIBS analysis and is fully consistent with the previously observed differences in tearing behavior and particle morphology.