Linus Tech Tips PTM7950 Review – Original, OEM Product or Fake?

1 - Introduction and overview 2 - Microscopy, pad thickness and homogeneity 3 - Microscopy, particle analysis and chemical composition 4 - Burn-in and performance measurements 5 - Summary and conclusion

On the slightly shiny surface of the pad, a very uniform, densely packed filler structure can be seen under the microscope, as is familiar from filled phase change materials. The light reflective particles are embedded in the darker polymer matrix and are distributed largely homogeneously, without visible cavities or islands with a high binder content. If we take our previous experience with PTM7950 and comparable OEM pads as a reference and consider supplementary chemical analyses, there is much to suggest that these are essentially aluminum oxide particles supplemented by zinc oxide. I cannot verify this purely from the imaging, but the morphology and gloss level are consistent with this interpretation.

The measurements of the particle sizes in the undisturbed matrix predominantly show values between roughly 3 and 5 micrometers, with a few outliers upwards to around 7 micrometers. This indicates a relatively closely distributed, fine-grained primary particle size. When viewed from above, the particles usually appear almost round to slightly polygonal, without a pronounced rod or platelet structure. Aluminum oxide in classic filler quality is often present in this size range, and zinc oxide is similarly finely ground in many formulations. The isometric shape ensures that the particles behave isotropically under thermal and mechanical stress and do not force any preferred flow directions in the matrix, which benefits the stable transient behavior.

Things get interesting in the areas where the polymer has been slightly ablated with the laser. This creates a relief-like image in which the filler particles are virtually modeled out of the matrix. The local heating and material removal of the polymer causes the fillers to become more prominent, which also leads to more densely packed clusters and small agglomerates. In these laser-processed zones, the measured particle and agglomerate sizes increase to around 6 to 12 micrometers. This corresponds less to a genuine primary particle enlargement and more to the fact that several particles lie next to each other and are measured as a composite. Such agglomerates are common in all highly filled systems as long as their proportion remains limited and does not create macroscopic inhomogeneities.

Laser ablation provides an additional indication of the material quality. If the polymer softens locally and exposes the fillers in this form without entire areas segregating or breaking off over a large area, this indicates a stable bond between the matrix and filler. Particularly in the case of imitation or poorly formulated pads, you can see significantly larger cavities, flaking or clod-like filler packages. Such anomalies are absent in the images shown, which, together with the observed particle morphology and distribution, supports the impression that the structure and filler design of the investigated material fits very well with previous experience with PTM7950.

The comparison with known PTM7950 samples shows that the observed particle size distribution as well as the shape and packing density fit the picture very well. Typical Honeywell PTM7950 pads also show a dominant fraction of fine fillers in the range of 3 to 5 micrometers, supplemented by smaller agglomerates in the range of 8 to 12 micrometers. In combination with aluminum oxide and zinc oxide as the main components, this results in a filler architecture that enables sufficiently high thermal conductivity without making the material too brittle. The homogeneous distribution visible here, the relatively narrow particle sizes and the absence of larger “holes” or separated filler nests indicate that the dispersion of the powder in the polymer matrix was carried out cleanly and that the microstructure of the pad examined is very close to what is known from previous investigations of genuine PTM7950 batches.

Material composition (LIBS)

The evaluation shown here comes from my LIBS measurement, which was carried out directly in the Keyence VHX-7100 in combination with the AE-300. The VHX not only functions as a high-resolution digital microscope, but also as a positioning system for the LIBS unit. The first step is to focus optically on a defined surface of the pad, then a measuring grid is created and the AE-300 applies short, high-energy laser pulses to the marked points. These pulses locally ablate a tiny part of the surface and generate a short-lived plasma. The emitted light is spectrally resolved so that the elements present and their approximate mass fractions can be determined from the characteristic emission lines. The yellow crosses in the image mark the individual laser pulses, from whose sum spectrum the average composition was calculated.

As a result, aluminum, oxygen, zinc, carbon and hydrogen are detected, with approximately 37.9 percent aluminum, 26.3 percent oxygen, 18.3 percent zinc, 14.7 percent carbon and around 2.8 percent hydrogen by weight. Silicon is practically absent from the spectrum, i.e. it is either not present at all or only well below the detection limit. The high sum of aluminum, zinc and oxygen is a very strong indication of a filler combination of aluminum oxide and zinc oxide, as described for PTM7950. If the oxygen is added to the metal oxides, a typical, very highly filled ceramic matrix is obtained, while carbon and hydrogen originate from the organic polymer phase. The exact stoichiometric ratios can only be determined to a limited extent with LIBS, but the order of magnitude of the mass fractions fits well with a two-component oxide filler system with a high degree of filling.

The absence of silicon lines is particularly interesting, as PTM7950 is explicitly advertised as a silicone-free phase change material. A classic siloxane or PDMS matrix would almost inevitably generate clear silicon signals in such a measurement. The fact that only carbon and hydrogen appear here as the main organic components instead supports the assumption of a silicon-free, probably paraffin- or polyolefin-based polymer matrix. The element pattern corresponds to the expectation for a silicone-free PTM pad.

For PTM materials such as PTM7950, this silicon-free matrix is not just a marketing detail, but functionally crucial. Phase change pads should become soft or partially liquid at a clearly defined temperature, flow into the roughness of the cooler and chip under contact pressure, but then become sufficiently solid again so that they are not pumped out of the joint or displaced by gravity. Paraffinic or related polymer systems can be adjusted very precisely to such a melting or softening point and exhibit pronounced hysteresis behaviour, which gives the pad a certain degree of self-stabilization after burn-in. Classic silicone oils or siloxane-based binders would remain significantly more liquid in this temperature range, tend to pump out more and, in extreme cases, deposit on the surface of components or in connectors.

In addition, siloxanes are undesirable in certain applications because they can volatilize under vacuum or high thermal stress and then precipitate as contamination on optics, contacts or sensors. A silicone-free PTM pad reduces this risk and is more attractive for OEM designs in notebooks or graphics cards, where long-term stability and clean surfaces are required. If the LIBS results are now compared with known PTM7950 samples, it is noticeable that both the element distribution and the general pattern of Al, Zn, O plus a purely carbon-based polymer phase match the previously analyzed references very well. The “PTM Sheet” material library stored in the system is also recognized and displayed as the best-fitting reference, which further supports the findings. Although I cannot derive an official manufacturer’s confirmation from this, the combination of measured element composition, high ceramic filling of aluminum oxide and zinc oxide and completely silicon-free polymer matrix corresponds exactly to the profile expected of a genuine PTM7950-like phase change pad.