THERMAL GREASE BASED ON HYPERBRANCHED OLEFINIC FLUID

Abstract

Disclosed is an effective thermal grease comprising a hyperbranched olefinic fluid and a thermally conductive filler. Property-modifying additives and fillers may also be included. The hyperbranched olefinic fluid is selected to have an average of at least 1.5 methine carbons per oligomer molecule and at least 40 methine carbons per one thousand total carbons. The thermal grease exhibits a flash point of 180° C. or higher, a pour point of 0° C. or lower, and a kinematic viscosity at 40° C. of no more than 200 cSt (0.0002 m 2/s). The composition may offer improved thermal conductivity, reduced tendency to migrate, and lower cost when compared with many other thermal greases, including silicone-based thermal greases.

Claims

1. A thermal grease composition comprising an admixture of (a) a hyperbranched, ethylene-based or ethylene- and propylene-based, olefinic fluid having an average of at least 1.5 methine carbons per oligomer molecule, and having at least 40 methine carbons per one-thousand total carbons, and wherein the average number of carbons per molecule is from 25 to 200; and (b) a thermally conductive filler.

2. The thermal grease composition of claim 1, wherein the hyperbranched, ethylene-based or ethylene- and-propylene-based olefinic fluid exhibits at least one of (a) a flash point of 180° C. or higher, as measured according to ASTM D-93; (b) a pour point of zero ° C. or lower, as measured according to ASTM D-97; (c) a kinematic viscosity at 40° C., as measured according to ASTM D-445, of no more than 0.0002 meter squared per second; and (d) a combination thereof.

3. The thermal grease composition of claim 1 wherein the ethylene-based or ethylene- and propylene-based fluid further comprises an alpha-olefin comonomer other than propylene.

4. The thermal grease composition of claim 1 wherein the thermally conductive filler is selected from beryllium oxide, aluminum nitride, boron nitride, aluminum oxide, zinc oxide, magnesium oxide, silicon carbide, silicon nitride, silicon dioxide, and zinc sulfide; solid metal particles, selected from silver, copper and aluminum; carbon materials, selected from diamond powder; carbon fibers, carbon nanotubes, carbon black, graphite, graphene and graphene oxide; liquid metals, selected from gallium-based alloys; and combinations thereof.

5. The thermal grease composition of claim 1 further comprising a phase segregator, a surfactant, a flame retardant, an antioxidant, a coupling agent, a bleed inhibiting agent, a rheology modifier, a filler, or a combination thereof.

6. The thermal grease composition of claim 5 wherein the phase segregator is a compound selected from polydimethylsiloxane; phenylmethyl polysiloxane; hydroxy-terminated polydimethyl-siloxane; polydimethyldiphenylsiloxane; polydiphenylsiloxane; methylalkyl polysiloxanes containing an alkyl group selected from a naphthyl group, an ethyl group, a propyl group, or an amyl group; and combinations thereof.

7. A process to prepare a thermal grease composition comprising (a) contacting together ethylene, and optionally propylene, and further optionally, an alpha-olefin, and at least one coordination-insertion catalyst, wherein the coordination-insertion catalyst is a metal-ligand complex wherein the metal is selected from zirconium, hafnium and titanium, and has an ethylene/octene reactivity ratio up to 20, and a kinetic chain length up to 20 monomer units; in a continuously-fed backmixed reactor zone under conditions such that a mixture of at least two products is formed, the mixture including (i) a hyperbranched oligomer having an average of at least 1.5 methine carbons per oligomer molecule, and having at least 40 methine carbons per one-thousand total carbons, and wherein at least 40 percent of the methine carbons is derived from the ethylene or, where the optional propylene is included, from the ethylene and the propylene, and wherein the average number of carbons per molecule is from 25 to 200; and (ii) at least one organic volatile product having an average number of carbons per molecule that is less than or equal to 14; (b) separating the hyperbranched oligomer from the organic volatile product; (c) recovering the hyperbranched oligomer; and (d) admixing the hyperbranched oligomer and a thermally conductive filler to form a thermal grease composition.

Description

EXAMPLE 1 AND COMPARATIVE EXAMPLE 1

[0077] Thermal grease samples are prepared, with a first series serving as inventive samples (Example 1a-1f), based on a hyperbranched olefinic fluid, and the second serving as comparative samples (Comparative Example 1a′-1f′) based on a silicone oil.

A. Preparing a Hyperbranched, Ethylene-Based Olefinic Fluid

[0078] In order to prepare a suitable hyperbranched, ethylene-based olefinic fluid, feeds comprising ethylene, ISOPAR-E™ as a solvent, and toluene (as a solvent to dissolve the catalyst) are passed through columns of activated alumina and Q-5 in order to first remove water and oxygen therefrom. These feeds are then introduced into an adiabatic, continuous stirred tank reactor (CSTR), with typical CSTR backmixing, with the solvent (toluene), catalyst (Formula V), and activator (ISOPAR-E™) being introduced into the reactor via stainless steel lines from syringe pumps located in a glovebox containing an atmosphere of nitrogen. The ethylene and the catalyst solution are introduced via independent dip tubes and metered with the aid of mass flow controllers. The reaction is allowed to proceed at a temperature of 60° C., with a residence time of 10 minutes, a C.sub.2 feed rate of 1.00 g/min, and a feed mass fraction of C.sub.2 monomer of 0.14 (C.sub.2 feed rate/total feed rate).

[0079] The vessel is heated by circulating hot silicone oil through the external jacket and cooled when required via an internal cooling coil with water. The reactor pressure is controlled with a GO REGULATOR™ BP-60 back pressure regulator. The system is run hydraulically filled with no head space and without a devolatilization unit. Polymer solutions are removed from the vessel for periodic sampling from an outlet on the reactor head fitted with an electrically heated stainless steel line. Solution olefin concentrations of the reactor effluent are then measured via a Fourier Transform Near Infrared (FT-NIR) spectrometer to determine the in-reactor concentration of ethylene. Further analyses of the product are carried out via .sup.13C NMR as described hereinbelow.

[0080] Once the desired reaction endpoint is reached, the hyperbranched olefinic fluid is treated, prior to collection, with a catalyst deactivator comprising 2-propanol with water and a stabilizer package containing IRGANOX™ 1010, pentaerythritol tetrakis(3,5-di-tert-butyl-4-hydroxyhydrocinnamate), and IRGAFOS™ 168, tris(2,4-di-tert-butylphenyl) phosphite from CIBA GEIGY CORPORATION™. Multiple runs are performed in the CSTR and the oligomer fractions are all combined. Oligomers are first rotary evaporated at 80° C./10 Torr to remove solvent, then passed through a wipe-film evaporator (WFE) set at 155° C./100 mTorr. The products from the WFE are collected and tested for viscosity. Among the products those designated as “lights” are generally residual solvent and light product molecules that tend to degrade the flash and fire points of the material, while the “heavies” are all other products that comprise the desirable hyperbranched ethylene-based olefinic fluid to be used in making a thermal grease. Kinematic viscosity (cSt, 40° C./100° C., according to ASTM D445) is 34.94/6.60. Samples are not hydrogenated, such as might be desirable on a commercial scale for product stability, and one olefinic unit remains for each oligomer chain.

[0081] Testing is carried out to determine flash point and pour point of the base, or matrix, hyperbranched olefinic fluid and also of a selected silicone-based matrix, as described hereinbelow. In general the physical properties such as fire point and pour point of the final thermal grease tend to be reasonably correlated to the same properties of the matrix fluid, so determination of such may be carried out on the fluid in the absence of fillers for convenience. It is also noted, however, that, while fire point is always higher than flash point, it is not always predictable how much higher it will be. However, because flash point determination generally requires a smaller sample, it is often used in lieu of fire point determination for experimental purposes.

[0082] Flash point: Measurements are made on an ERAFLASH™ instrument from ERA ANALYTICS™ with a high temperature attachment. This method follows ASTM D93 for a closed cup flash point measurement. In this protocol 2 mL of sample are added to the stainless steel sample cup via a micropipette and a stir bar is added. The sample cup and holder are placed into the sample chamber and the door is closed. Run parameters for the ERAFLASH™ include: stir rate=100 rpm, heat rate=10° C./min, Step=2° C. temperature range=70° C., ignition=2 milliseconds (ms), air volume=10 mt/min between 150° C. and 300° C. After each sample the chamber is cleaned and the electrodes are cleaned with a wire brush provided by the manufacturer.

[0083] Pour Point: In a 48 well plate with vials, 1 mL of sample is added to each vial, followed by one copper BB. This method follows ASTM D455. Measurements are run in triplicate and the measurement is based upon two agreements. A rubber mat is placed on top of the samples and the 48 well plate is placed into a temperature programmed freezer. After a minimum of 4 h, the samples are removed and flipped onto a scanner. The samples are allowed to stand for 1 min and then scanned, with the image being used to determine if the copper BB is apparent in the scanned image. This serves as a pass/fail test. The freezer temperature is then changed and the procedure repeated until the desired level of pour point resolution is attained.

[0084] Table 1 shows the experimental conditions that are used in synthesizing the hyperbranched, ethylene-based olefinic fluid for Example 1, for each of the runs. In this case the catalyst corresponds to Formula V.

TABLE-US-00001 TABLE 1 Typical experimental conditions for synthesis of the hyperbranched, ethylene-based olefinic fluid Total feed Catalyst FT-NIR C2 FT-NIR C2 FT-NIR Cx rate Total catalyst feed Rate Run Conv (%) (g/dL) (g/dL) (g/min) metal (ppm) (μmol/min*) 1 96.2 0.37 4.2 7.36 0.56 0.045 *μmol/min = micromoles per minute

[0085] For .sup.13C NMR confirmations, samples are dissolved in 10 millimeter (mm) NMR tubes in chloroform-d with 0.02 M chromium(III) acetyl acetonate (Cr(AcAc).sub.3, C.sub.15H.sub.21CrO.sub.6, tris(2-4-pentanediono)-chromium(III)) added. The typical concentration is 0.50 g/2.4 mL The tubes are then heated in a heating block set at 50° C. The sample tubes are repeatedly vortexed and heated to achieve a homogeneous flowing fluid. For samples with visible wax present, tetrachloroethane-d.sub.2 is used as the solvent instead of chloroform-d, and the sample preparation temperature is 90° C. .sup.13C NMR spectra are taken on a BRUKER™ AVANCE™ 400 megahertz (MHz) spectrometer equipped with a 10 mm cryoprobe. The following acquisition parameters are used: 5 seconds relaxation delay, 90 degree pulse of 13.1 microseconds, 256 scans. The spectra are centered at 80 ppm with a spectral width of 250 ppm. All measurements are taken without sample spinning at either 50° C. (for chloroform solutions) or 90° C. (for tetrachloroethane solutions). The .sup.13C NMR spectra are referenced to 77.3 ppm for chloroform or 74.5 ppm for tetrachloroethane. The analysis results from .sup.13C NMR spectra are given in Table 2.

TABLE-US-00002 TABLE 2 .sup.13C NMR analysis results of the hyperbranched olefinic fluid Degree of Branching Total Hexyl Branches Concentration of unsaturation Branches (per 1000 (per 1000 Vinyl % per Mn carbons) Butyl Ethyl Methyl carbons) Vinylene V1 V3 Vinylidene Vinyls molecule 528 39.0 22.1 64.9 0.7 126.5 4.25 5.57 11.75 6.54 61.6 4.78 Mn is number average molecular weight

[0086] Table 3 shows a comparison of the flash point and pour point of the Example 1 hyperbranched, ethylene-based olefinic fluid and, as Comparative Example 1, a selected silicone-based fluid.

TABLE-US-00003 TABLE 3 Comparison of flash point and pour point of hyperbranched olefinic fluid and a silicone-based fluid Flash point, ° C. Pour point, ° C. Fluid (ASTM D-93) (ASTM D-455) Hyperbranched olefinic fluid 229.8 −15 (ethylene-based) Silicone fluid* 293.7 Below −50 (freezer-limited) *A 50/50 vol % mixture of Dow Corning 510 ™ fluid (phenylmethyl polysiloxane) having a viscosity of 100 cSt and the same fluid having a viscosity of 50 cSt.

B. Preparing the Thermal Greases

[0087] The matrix hyperbranched olefinic fluid or selected silicone oil is first weighed into a ceramic cylinder cup. Surfactant, SPAN™ 85, used only with hyperbranched fluid to overcome polarity issues that tend to reduce grease homogeneity, is weighed and added into the matrix and the mixture is stirred with a metal spatula until the surfactant is fully mixed into the matrix. Thermally conductive fillers are weighed and pre-mixed by dramatic hand shaking. Thereafter the fillers are added into the ceramic cylinder cup and are dispersed into the matrix mixture by sufficient stirring and kneading with a metal spatula until the matrix appears visually to have fully wetted the filler surface and the composite appears as a smooth and uniform blend. The stirring is carried out at approximately 100 revolutions per minute (rpm) for three times, with each time being at least 10 min to ensure good visual homogeneity. The resultant thermal greases are transferred into and stored in a capped glass vials. The prepared greases have the constituencies given in Table 4.

TABLE-US-00004 TABLE 4 Constituents of thermal grease compositions for Examples 1a-1f and Comparative Examples 1a′-1f′ Component Specification Supplier Hyperbranched 72 cSt (25° C.)-ethylene based 650 cSt Synthesized as olefinic fluid* (25° C.)-ethylene and propylene based described Phenylmethyl 510 ™** fluid 100 cSt (50 vol %), 50 cSt (50 vol %) Dow Corning polysiloxane* Surfactant*** SPAN ™ 85, chemically pure reagent Sinopharm Chemical Reagent AlN WLS**** (D50 = 9.8 μm, D100 = 136 μm) Toyo Aluminum Alumina (Al.sub.2O.sub.3) AX3-75**** (D50 = 3 μm, D100 = 75 μm) Nippon Steel AX35-125**** (D50 = 35 μm, D100 = 125 μm) AX10-32**** (D50 = 10 μm, D100 = 32 μm) ZnO Nano-ZnO**** D50 = 0.35 μm Wuxi Zehui Chemical Alumina (Al.sub.2O.sub.3) ASFP-20**** (spherical alumina) Denka *These are matrix materials. Only one is used for any given formulation. **Dow Corning's 510 ™ fluid is polyphenylmethyldimethylsiloxane, which is a clear, heat stable silicone fluid. ***SPAN ™ 85 is a surfactant (sorbitane trioleate) that is used only with the hyperbranched olefinic fluid based thermal grease formulation. ****WLS, AX3-75, AX35-125, AX10-32, Nano-ZnO, and ASFT-20 are tradenames of the listed suppliers. D50 = MMD, mass-medium-diameter, average particle diameter by mass; D100 = maximum particle diameter by mass.

[0088] Following formation of greases following incorporation of all constituents shown in Table 5, testing is done to determine thermal conductivity (TC), shear viscosity, and thermal resistance. Testing is also done on a commercially available, silicone-based thermal grease denominated as Z9™, available from DEEPCOOL™. Testing methodology is described as follows.

[0089] Thermal conductivity (TC): Thermal conductivity (W/m.Math.K) of thermal grease samples is measured with a HOT DISK™ instrument (TPS 2500S, transient plane source), available from HOT DISK AB™, Sweden, conforming to the standard of ISO 22007-2:2008. For these grease samples, the measurement is done with a smaller HOT DISK™ sensor (3.2 millimeter, mm, radius) in a liquid cell. The experimental parameters used to collect the data are: Temperature 24° C., Power 0.2 watt (W), and Time 2 Sec.

[0090] Frequency sweep test for viscosity: Shear viscosity is measured at 25° C. on 25 mm steel parallel plates of an AR2000EX™ stress control rheometer, available from TA INSTRUMENTS™. The shear rate is set from 0.1/s to 5/s and the duration of the test is 10 min. The value at 1.1/s is recorded and used for comparison. Testing is carried out according to the protocol of modified ASTM D4440-08.

[0091] Thermal resistance: The thermal grease samples are evaluated for thermal resistance by Apparatus LW-9389™, available from LONG WIN SCIENCE AND TECHNOLOGY CORPORATION™, conforming to ASTM D5470-06 standard. The test conditions include: Constant T.sub.avg (average temperature of hot interface and cold interface between sample and thermo-sensor) 60° C., Contact Pressure 20, 40, 80 psi, Die Area 6.4516 cm.sup.2, Test Duration-30 min.

[0092] Table 5 shows formulations and testing results.

[0093] Example 1a to 1f and Comparative Example 1a′ to 1f′ shown in Table 5 illustrate differences in performance of thermal greases based upon hyperbranched ethylene-based fluids versus those based on 50/50 vol % mixture of phenylmethyl polysiloxane fluids having two different viscosities. The phenylmethyl polysiloxane mixture is selected in order to obtain a material having a viscosity comparable to that of the base hyperbranched olefinic fluid. The surfactant SPAN™ 85 is used for only Examples 1a to 1f and is intended to compensate for the higher polarity of the hyperbranched material, in order to assure comparable dispersion of the thermally conductive filler.

TABLE-US-00005 TABLE 5 Formulations and performance results of Examples 1a-1f and Comparative Examples 1a′-1f′. Example 1 Comparative Example 1 Volume % a b c d e f a′ b′ c′ d′ e′ f′ Z9 Matrix Hyperbranched 33 28.7 25 25.5 23.9 25.5 ethylene fluid, (1.0) (1.5) (1.0) (1.0) (1.0) (1.0) (SPAN ™ 85, wt % of fillers) Phenylmethyl 33 28.7 25 25.5 23.9 25.5 Silicone polysiloxane oil- based grease Filler AlN (WLS) 67 71.3 67 71.3 Spherical Al.sub.2O.sub.3 75 75 AX35-125/ AX3-75 = 7/3 Spherical Al.sub.2O.sub.3 59.7 60.9 67 59.7 60.9 67 AX10-32 Spherical Al.sub.2O.sub.3 7.5 7.5 ASFP-20 ZnO (D50 = 14.8 15.2 14.8 15.2 0.35 μm) Viscosity 165 424 135 1069 1919 1461 265 — 110 940 — 1461 531 (Pa .Math. s at shear rate of 1.1/s) Paste-like appearance & good good good fair good good good N/A good fair poor good fair dispensing property Thermal conductivity 3.13 3.84 3.50 3.14 3.22 2.69 2.93 — 3.17 2.93 — 2.37 2.87 (W/mK) — indicates data was not obtained N/A indicates grease could not be formed

[0094] As can be seen, inventive Example 1a has a higher thermal conductivity than Comparative Example 1a. In inventive Example 1b and Comparative Example 1b′, AlN loading is increased to 71.3 vol %, which leads to different performance results. Inventive Example 1b is described as a “good” thermal grease, having relatively low viscosity and higher thermal conductivity, while Comparative Example 1b′ cannot form a grease due to its much higher viscosity. This is characterized by the fact that the sample cannot be pasted at all. This is because, although the hyperbranched olefinic fluid and phenylmethyl polysiloxane base matrices have almost the same viscosity, the hyperbranched olefinic fluid system can accommodate more filler to achieve higher thermal conductivity.

[0095] It is also noteworthy that the inventive Example 1c and Comparative Example 1c′ matrices have similar appearance and viscosity adaptation for a spherical Al.sub.2O.sub.3 (35 μm:3 μm=7:3) system at the given loading level of 75 vol %. However, the thermal conductivity of inventive Example 1c is slightly higher than that of Comparative Example 1c′.

[0096] In order to simulate applications having thin gaps, i.e., small bond line thickness (BLT), inventive Examples 1d to 1f and Comparative Examples 1d′ to 1f′ use smaller cut point size (i.e., D100). The primary thermally conductive filler for each is spherical Al.sub.2O.sub.3(D50=10 μm, D100=32 μm). A submicron filler (ZnO and spherical Al.sub.2O.sub.3) is also used in combination. As in inventive Examples 1a to 1b and Comparative Examples 1a′ to 1b′, it is observed that the hyperbranched olefinic fluid matrix can incorporate more filler than the phenylmethyl polysiloxane matrix, resulting in a higher thermal conductivity for the hyperbranched olefinic fluid based thermal grease. It is also noted that both Inventive Example 1d and Comparative Example 1d′ exhibit good paste and tack properties, which would be helpful in practical application.

[0097] Thermal resistance results show that inventive Example 1d has a relatively low thermal resistance, of 0.028 degree Celsius-inch squared per watt (° C.-in.sup.2/W) at a relatively low contact pressure load of 20 pounds per square inch (psi, approximately 137.9 kilopascals, kPa). Thus, low mounting pressure is allowed for the use of such grease. For comparison, the product datasheet for Z9™ product, available from DEEPCOOL™ states that the silicone oil based commercial product has a thermal resistance of less than or equal to 0.058° C.-in.sup.2/W, which would be defined as an inferior performance result.

EXAMPLE 2 and COMPARATIVE EXAMPLE 2

[0098] Two sets of thermal greases are prepared, the first including a hyperbranched, ethylene- and propylene-based olefinic fluid as the matrix Component A, and the second including only a silicone fluid, phenylmethyl polysiloxane, as the matrix Component A.

[0099] A. Preparation of the hyperbranched, ethylene-and propylene-based olefinic fluid is conducted in a 2 L Parr™ batch reactor on a semi-batch basis. The reactor is heated by an electrical heating mantle, and is cooled by an internal serpentine cooling coil containing cooling water. Both the reactor and the heating/cooling system are controlled and monitored by a CAMILE™ TG process computer. The bottom of the reactor is fitted with a dump valve, which empties the reactor contents into a stainless steel (SS) dump pot, which is prefilled with a catalyst kill solution (typically 5 mL of a IRGAFOX™/IRGANOX™/toluene mixture). The dump pot is vented to a 30 gallon blowdown tank, with both the pot and the tank N.sub.2 purged. All chemicals used for polymerization or catalyst makeup are run through purification columns to remove any impurities that may affect polymerization. The propylene is passed through 2 columns, the first containing Al.sub.2O.sub.4 alumina, the second containing Q5 reactant to remove oxygen. The ethylene is also passed through 2 columns, the first containing Al.sub.2O.sub.4 alumina, and 4 Angstroms (Å) pore size molecular sieves, the second containing Q5 reactant. The N.sub.2, used for transfers, is passed through a single column containing Al.sub.2O.sub.4 alumina, 4 Å pore size molecular sieves and Q5 reactant.

[0100] The reactor is loaded first with toluene and then with propylene to the desired reactor load. After liquid feed addition, the reactor is heated up to the polymerization temperature set point. Where ethylene is used, it is added to the reactor when at reaction temperature to maintain reaction pressure set point. Ethylene addition amounts are monitored by a micro-motion flow meter.

[0101] The catalyst and activators are mixed with the appropriate amount of purified toluene to achieve a desired molarity solution. The catalyst and activators are handled in an inert glove box, drawn into a syringe and pressure transferred into the catalyst shot tank. This is followed by 3 rinses of toluene, 5 mL each.

[0102] Immediately after catalyst addition the run timer begins. Where ethylene is used, it is then added by the CAMILE™ to maintain reaction pressure set point in the reactor. These polymerizations are run for the desired amount of time, then the agitator is stopped and the bottom dump valve opened to empty reactor contents to the dump pot. The dump pot contents are poured into trays placed in a lab hood where the solvent is evaporated off overnight. The trays containing the remaining polymer are then transferred to a vacuum oven, where they are heated up to 140° C. under vacuum to remove any remaining solvent. After the trays cool to ambient temperature, the oligomers are weighed for yield/efficiencies, and submitted for testing.

TABLE-US-00006 TABLE 7 Characterizing data of the hyperbranched olefinic fluid Unsaturation Viscosity % % % Branches per Mn @ 40° C. @ 100° C. Vinyls Vinylidenes Vinylenes 1000 C's Mol % C3 (.sup.1H NMR) (cSt) (cSt) 31 68 1 195.04 48.6 708.05 109.5 16.0

TABLE-US-00007 TABLE 6 Reactor Parameters Batch Ethylene Batch Ethylene Run MMAO- Temp Toluene Ethylene Pressure Propylene g g time Catalyst RIBS-2* 3A** Exotherm ° C. g g psi) g initial added min Formula μmoles metal μmoles μmoles ° C. 120 300 17.1 359.7 140.5 17.1 10.1 3.8 X 2.5 Hf 3 10 1.9 *RIBS-2 co-catalyst: (CAS); Amines, bis(hydrogenated tallow alkyl)methyl, tetrakis(pentafluorophenyl)borate(1-) **MMAO-3A co-catalyst is a modified methyl aluminoxane

[0103] This is carried out as in Example 1, with the parameters shown in Table 6, except that the co-monomer propylene is included as a feed, with a C.sub.3 feed rate of 1.00 g/min, and a feed mass fraction of C.sub.3 monomer of 0.14 (C.sub.3 feed rate/total feed rate). The resulting hyperbranched olefinic fluid exhibits the characteristics shown in Table 7. The catalyst corresponds to Formula X.

[0104] Following preparation of the hyperbranched ethylene- and propylene-based olefinic fluid and selection of the silicone-based matrix for the Comparative Example 2 grease, the matrix fluids are tested to determine flash points and pour points, using the procedures described in Example 1 and Comparative Example 1. Results of this testing are shown in Table 8.

TABLE-US-00008 TABLE 8 Flash point and pour point for thermal greases Thermal Grease Matrix (Component A) Flash point, ° C. Pour point, ° C. Ethylene-propylene oligomer 182 0° C. Silicone fluid* 293.7 Below −50 (freezer-limited) *A 50/50 vol % mixture of Dow Corning 510 ™ fluid (phenylmethyl polysiloxane) having a kinematic viscosity of 100 cSt (0.0001 m.sup.2/s) and the same fluid having a kinematic viscosity of 50 cSt (0.00005 m.sup.2/s, = 5e−05 m.sup.2/s).
B. Example 2a to 2c and Comparative Example 2a′ to 2c′ are prepared, having the constituencies given in Table 4 and using the methodology described in Example 1 and Comparative Example 1. Table 9 shows formulations and testing results.

TABLE-US-00009 TABLE 9 Formulations and performance results of Examples 2a-2c and Comparative Examples 2a′-2c′. Example 2 Comparative example 2 Volume % a b c a′ b′ c′ Matrix Hyperbranched ethylene- 33 25 25.5 propylene fluid, (SPAN ™ (1.0) (1.0) (1.0) 85, wt % of fillers) Phenylmethyl 33 25 25.5 polysiloxane Filler AlN (WLS) 67 67 Spherical Al.sub.2O.sub.3 AX35- 75 75 125/AX3-75 = 7/3 Spherical Al.sub.2O.sub.3 AX10-32 59.7 59.7 ZnO (D50 = 0.35 μm) 14.8 14.8 Viscosity (Pa .Math. s) 455 728 1812 265 110 940 at shear rate of 1.1/s Paste-like appearance & dispensing good good good good good fair property Thermal conductivity (W/mK)* 3.30 3.36 3.08 2.93 3.17 2.93 *Watt/meter .Math. Kelvin

[0105] It will be seen in Table 9 that inventive Example 2a exhibits a higher thermal conductivity than Comparative Example 2a′ when using AlN as the thermally conductive filler. Based on other filler packages, a higher thermal conductivity is also achieved using hyperbranched ethylene-propylene fluid as the matrix than is achieved using phenylmethyl polysiloxane as the matrix. This is shown in Example 2b and 2c, which are compared with Comparative Example 2b′ and 2c′.