Nuomi Chemical SY-166 Thermal Paste Review – Cheap Poser or really cheap Alternative?

1 - Introduction, advertising claims, and technical specifications 2 - Fracture behavior, microscopy, and composition 3 - Performance comparisons and practical measurements 4 - Durability (pump-out), summary, and conclusion

Tear-off behavior and shear structure

We see here a rather low-viscosity paste with clear stringing during lifting and a clearly pronounced flow pattern under shear. A low yield point is evident during discharge and spreading, so little pressure is needed for the paste to start flowing. Under constant shear, it aligns itself into elongated flow paths, which can be recognized by the parallel grooves, while islands and tongues remain in zones with lower local shear stress. This pattern is typical for systems with moderate solids loading and a binder matrix that is hardly thixotropic. When the applicator is lifted, long, elastic-looking threads are formed which taper and only tear late. This indicates an increased extensional viscosity and a noticeable elastic portion of the matrix, the cohesion outweighs the adhesion to the substrate surface in the short term. Tearing is therefore predominantly cohesive, residual films and threads remain on both surfaces, and peaks and craters form at the tearing points, which can act as local air pockets or thick spots during subsequent assembly.

In the transition areas between high-shear and stationary zones, small breakouts and microscopic cavities, i.e. incipient dewetting points, can be seen, indicating local binder depletion. With increasing shear, the paste initially glides well, then shear islands appear due to particle contact and surface micro-roughness, which are pushed forward like small clods in the binder matrix film. This floe formation reduces the homogeneity of the layer thickness, burrs are formed at the edges of the floes, which only level out incompletely during subsequent pressing, which increases the thermal contact resistance locally. When the heat sink is placed on the surface, the thread drawing favors the entrainment of material over the edge, resulting in drawing flags that are deposited as uneven edges. In conjunction with the low yield point, this can lead to pump-out in the edge areas during temperature changes, as the paste tends to flow rather than remain structurally stable under cyclical expansion and compression.

During subsequent disassembly, predominantly cohesive failure is again to be expected, i.e. threads between the two surfaces and split residues instead of a clean, adhesion-dominated tear-off. In practical terms, this means that it is better to pre-spread the paste to create a thin, even layer without pulling threads and without forming islands; brief pre-compression followed by relaxation helps to allow trapped air to escape. Excess at the edges should be removed because these systems in particular tend to build up at the edges and pump away later. Overall, the shear and tear-off behavior is soft and elastic rather than stable, which makes processing easy, but can lead to fluctuations in layer thickness and thus to fluctuating contact resistances during load changes and long periods of operation.

Particle morphology and dispersion

The coarse fraction, which stands out in the image with average diameters between around 4 and 8 µm and occasionally forms agglomerates up to over 15 µm, most likely corresponds to aluminum oxide particles (Al₂O₃). Their morphology is typical of mechanically broken but subsequently rounded oxide fractions: irregular, sub-round and slightly platelet-shaped, with blunt edges and partially shiny facets. Such particles are usually produced by calcination and subsequent mechanical grinding with jet or ball mills of low energy, often combined with sieving or air classification. The lack of sharp-edged splinters indicates that no high-energy jet milling process (e.g. jet mill) was used here, but rather a softer, probably rotating grinding principle with subsequent segregation via sedimentation or air classification.

The fine fraction in the submicrometer to lower micrometer range, which appears in the image as a dense, uniform dot structure, most likely consists of zinc oxide particles (ZnO). Their almost spherical, isotropic shape is typical for low-temperature precipitation or spray drying from zinc salt solutions with subsequent calcination. These particles are much finer than the aluminum oxide, they fill the gaps very efficiently and thus increase the packing density and the proportion of interparticle binder films. In the microscopic view, they form a very homogeneous basic matrix that reflects evenly and visually embeds the Al₂O₃ filler.

The dispersion shows that the ZnO particles are well distributed but not completely de-agglomerated. The finest double or triple clusters with sizes of 4 to 7 µm can be recognized, which indicate incomplete disintegration of secondary aggregates. This indicates that although the mixer or dissolver has generated sufficient macro-mixing, the shear energy was not sufficient to completely overcome the adhesive bonds between the ZnO secondary particles. A well-adjusted three-roll mill would usually have dissolved these agglomerates, but a typical pattern of a planetary agitator or dissolver with a moderate speed difference can be seen here. Rheologically, this results in an interesting combination: the Al₂O₃ forms the framework for the viscosity, thus ensuring mechanical stability under compressive load, while the fine ZnO compacts the interstices and thus lowers the yield point. This explains the observed behavior during application: a flowable, slightly stringy paste whose binder matrix (presumably a soft polydimethylsiloxane) reacts relatively viscoelastically. The finely dispersed but only moderately bound ZnO particles increase the extensional viscosity, the cohesion remains high and elastic threads are formed on lifting.

The agglomerates of several Al₂O₃ grains, recognizable by the roundish caked islands over 10 µm, indicate that the aluminium oxide fraction was not pre-ground or sieved before incorporation. Such islands often form when granulated Al₂O₃ fillers are mixed directly with a low-viscosity silicone oil without first adding a dispersing additive (e.g. an amino or silanol-modified siloxane). As a result, individual particle clusters remain, which rheologically act as shear islands and locally increase the viscosity. All in all, this results in a two-stage particle system, i.e. coarse, mechanically rounded aluminum oxide grains as structural carriers and fine, chemically precipitated zinc oxide particles as space fillers and flow improvers. This explains the moderate but not perfect homogeneity and the observed shear behavior. Both fractions were subsequently only mixed but not deagglomerated together, which is why the characteristic double grains and agglomerate halos were retained.

The resulting rheological behavior is typical for this combination: flowing well under compression, but stringy and elastic under tension, with a tendency to cohesive tearing. Under cyclic thermal stress, a certain pump-out behavior can be expected because the binder phase remains soft and the particle matrix does not form a rigid structure. The microstructure is indicative of a low-cost, simply produced paste with a sufficient filler content, but not optimized in terms of process technology.

Chemical composition (LIBS analysis)

I analyzed the sample with my LIBS system, which is based on a short-pulsed Nd:YAG laser and is used in my laboratory for the elemental analysis of thermally conductive pastes. The surface of the sample is ablated with short, high-energy laser pulses, creating a plasma that emits characteristic spectral lines. These are recorded by an echelle spectrometer with high resolution and then spectrally analyzed. The method allows a spatially resolved, low-destruction analysis without chemical preparation of the sample and provides a reliable quantitative distribution of the elements. The method is ideal for determining the inorganic fillers and the composition of the binder matrix, whereby organic components are only recorded indirectly via silicon, carbon and hydrogen content.

The measurement protocol documented in the measurement shows a composition of 34.7 % by weight aluminum, 28.8 % oxygen, 14.0 % zinc, 9.7 % silicon, 9.7 % carbon and 3.1 % hydrogen. These values give a very consistent picture: the paste consists mainly of aluminum oxide (Al₂O₃) as a coarse-grained filler and zinc oxide (ZnO) as a finer additive. The ratio of aluminum to zinc is around 2.5 : 1, which corresponds exactly to the microscopically observed two-component structure, i.e. larger, roundish Al₂O₃ particles embedded in a finer ZnO matrix. The oxygen content fits stoichiometrically well with this combination.

The organic matrix accounts for the remaining 22 %. The ratio of silicon to carbon is around 1:1, which suggests a linear polydimethylsiloxane (PDMS) as the base oil. The moderate hydrogen content and the visually smooth, slightly shiny surface structure confirm this assumption. It is most likely an uncrosslinked, viscoelastic silicone oil of medium chain length with low thixotropy, which explains the observed soft rheology and stringing when the paste is lifted off.

Since neither nitrogen nor conspicuous foreign lines occur in the LIBS signature, I can assume that no amino, acetoxy or other functional siloxanes were added here. The binder is therefore purely inert and odorless, which is also consistent with practical observation. It is possible that hydrophobic silane admixtures are present in very low concentrations, which give the dispersion stability but leave no measurable organic signatures. We see a simple, technically clean formulation of PDMS with a two-stage oxide fill of aluminum oxide and zinc oxide. This combination provides acceptable thermal conductivity with low material cost and stable chemical inertness. The paste is homogeneous, chemically neutral and without volatile additives, which also explains why it remains completely odorless when heated.