BACKGROUND OF INVENTION

This invention relates to thermally conductive compositions that have initial low viscosity and high bulk thermal conductivity. More particularly, the invention relates to compositions and methods of preparing compositions useful as thermal interface materials exhibiting high thermal conductivity and electrically insulating properties.

Thermal interface materials are particularly important in thermal management systems where a large amount of power is either generated or consumed. For instance, in the microelectronics industry, the drive for increasingly higher processing speed results in more heat generated per chip, and miniaturization results in a higher heat flux per unit area. The resulting high temperature often leads to mechanical stress, loss in performance and failure of electronic components due to CTE (coefficient of thermal expansion) mismatch. Most devices perform to rated specifications only within a narrow temperature range. Hence, efficient heat removal and heat transfer is a critical part in device design.

Thermal management is typically achieved by use of a heat dissipating component, such as a heat spreader, heat sink, lid, heat pipe, or any other designs and constructions known to those skilled in the art. Such heat dissipating components are used to conduct heat away from high temperature areas in an electrical system. A heat dissipating unit is a structure formed from a high thermal conductivity material (e.g. copper, aluminum, silicon carbide, metal alloys, polymer composites and ceramic composites) that is mechanically coupled to a heat-generating unit to aid in heat removal. In a relatively simple form, a dissipating unit can include a piece of metal (e.g. aluminum or copper) that is in contact with the heat-generating unit. Heat from the heat generating unit flows into the heat-dissipating unit through the mechanical interface between the units.

In a typical electronic package, a heat-dissipating unit is mechanically coupled to the heat-producing component during operation by positioning a flat surface of the heat- dissipating unit against a flat surface of the heat-generating component and holding the heat dissipating unit in place using some form of adhesive or fastener. However, when two solid surfaces are brought together, e.g. the back side of a flip chip and one surface of the heat spreader, rarely will the surfaces be perfectly planar or smooth, so air gaps will generally exist between the surfaces. As is generally known, the existence of air gaps between two opposing surfaces reduces the ability to transfer heat through the interface between the surfaces. Thus, these air gaps reduce the effectiveness and value of the heat- dissipating unit as a thermal management device. Direct surface-to-surface, or metal-to- metal contact without a thermal interface material leads to high thermal impedance and limited heat conduction capability.

To overcome this problem, a thermally conductive, mechanically compliant interface material is typically used to fill the gaps and to interconnect the two surfaces. Thermally conductive adhesives, gels, greases, phase change materials and pads or films carrying highly thermally conductive solid fillers have been devised for this purpose. For example, silver-filled silicones or epoxies are used as heat sink adhesives. Alumina or boron-nitride filled thermal interface materials are also known in the art.

Currently, one method for connecting heat dissipation devices such as heat sinks to heat generating devices such as semiconductor devices, is dispensing a filled matrix between the interconnecting surfaces and curing the composition in situ. This approach requires that the uncured material have a viscosity low enough for the material to be forced through an orifice for rapid manufacture. However, the effective thermal conductivity depends on the extent that the fillers are in contact with each other as well as with the connecting surfaces; high thermal conductivities are typically only achieved at high filler loadings. At this stage, the thermal interface material may be too viscous to process and dispense.

One attempt to increase the filler-filler and filler-surface contact has been directed to highly thermally conductive particulates, such as copper, coated with low-melting metal such as tin, dispersed in a thermoplastic matrix with a solvent, acid and fluxing agent per need. The resulting paste has been used to connect two thermally conductive surfaces e.g., a chip and a substrate pad. While a fusible solder approach provides better particulate-particulate and particulate-surface interactions, certain limitations emerge. For instances, re-solidified solder is prone to deformation and fatigue. The method employs a imx agent mat may raise environmental concerns. Further, the method is not particularly suited to wetting non-metallic surfaces.

A similar approach has been the use of liquid metal coated or bridged particle clusters. Liquid metal has also been used as thermal and electrical contacts for heat-generating semiconductor devices. While liquid metals mitigate mechanical stresses between the device and the adhered members and enhance theπnal conductivity, their tendency to form alloys or amalgams with other metals and their chemical reactivity with oxygen and moisture in air renders their long-term performance unacceptable. To alleviate the problem, liquid metals and their alloys or liquid-metal coated ceramic clusters have been dispersed in silicone oil to form an emulsion or a thermal paste. Thermal interface materials composed of curable or solidifiable compositions containing low melting solders or low melting solders in combination with solid particulates have been reported; however, such materials are electrically as well as thermally conductive, properties which are not desirable for many microelectronics applications. Methods have been described to circumvent the electrical conduction problem by hardening a polymer matrix prior to bringing two mating surfaces close together such that no continuous conductive bridges between the two surfaces are formed. While this provides electrical isolation between the two surfaces, the discontinuity also lowers the effectiveness of heat transfer.

Therefore, what is needed is a material composition having thermally conducting and electrically insulating properties, and when applied between heat generating and heat dissipating devices, such properties do not degrade over time. What is also needed is a low viscosity material composition to facilitate formation over various device geometries and architectures. Further, a method for making such material compositions and the optional curing and hardening of the material composition in situ is needed.

BRIEF DESCRIPTION OF THE INVENTION

The present invention meets these and other needs by providing a material composition that is thermally conducting and electrically insulating. By optionally curing or hardening the composition, the composition integrity as well as thermal, electrical and mechanical properties are retained over long periods of time. A method for making said composition is also provided in the present invention. Accordingly, one aspect of the invention is a composition comprising a liquid metal, an insulating solid filler comprising thermally conducting materials, and a resin. The liquid metal and the insulating solid filler are present in a volume ratio of from about 1 :1.1 to about 1 : 70 liquid metal to filler. A second aspect of the invention is to provide a method for preparing the composition described above. A third aspect of the invention is an electronic device, or component comprising the thermal interface composition described above. In a fourth aspect, the present invention provides a composition comprising a liquid metal, a boron nitride filler, and a resin, wherein said liquid metal and filler are present in a volume ratio of from about 1 :0.4 to about 1 :10 liquid metal to boron nitride filler.

These and other aspects, advantages, and salient features of the present invention will become apparent from the following detailed description, and the appended claims.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a schematic representation of an electrical component in accordance with the present invention.

Figure 2 is a cross-section view of a thermally conductive, electrically insulating thermal interface material.

DETAILED DESCRIPTION OF THE INVENTION

The present disclosure provides a liquid metal composition having thermally conductive and electrically insulating properties. The composition is thermally conducting and electrically insulating and is a dispensable blend at the time of application that can optionally harden on heating or irradiation. Or the composition may be a phase change film that becomes flow able upon heating and re-solidifies upon cooling. In some embodiments, the composition is an adhesive. In another embodiment, the composition is a grease. In yet another embodiment, the composition is a gel. The liquid metal composition comprises at least one liquid metal, an electrically insulating or semi¬ conducting solid filler comprising thermally conducting materials, and a resin that is optionally curable. The composition may further comprise an adhesion promoter and a catalyst. For a material composition that is thermally conducting and electrically insulating, the liquid metal acts as a bridge between two insulating filler particles and thermally conducts heat from particle to particle across the composition. Accordingly, heat transfer is facilitated. It is hence desired that the amount of liquid metal be sufficient to provide additional continuous heat transport pathways across the insulating filler particles and provide a rapid transport of heat. At the same time, the amount of liquid metal should not be excessive as to provide a continuous metallic contact across the composition that would make the composition electrically conducting. It is hence desirable to choose the amount of liquid metal such that a high thermal conductivity and a low electrical conductivity result. Typically, the composition includes, for each 100 parts by weight of resin, about 10 to about 1300 parts by weight of liquid metal, preferably about 100-1100 parts by weight, or any range or combination of ranges there between. The composition further includes a solid insulating filler in an amount of about 10 to about 1100 parts by weight, and preferably about 30-900 parts by weight, or any range or combination of ranges there between. The incorporation of liquid metals into the composition improves thermal performance, or the achievement of high thermal conductivity of final compositions while maintaining usable viscosity and processibility of the compositions.

By liquid metal is meant any metal (or meal alloy) that has a melting point that is less than 350C. Typically, the liquid metal (or meal alloy) has a melting point between about - 10°C and 35°. In one embodiment, the liquid metal(or meal alloy) employed has a melting point at about room temperature (about 18 to about 290C). In another embodiment, the liquid metal (or meal alloy) employed has a melting point below room temperature. Accordingly, in the present invention, any liquid metal that has a free flow may be used and includes, but is not limited to, low viscosity, freely flowing liquid metals and alloys such as gallium, indium, tin, mercury, metallic glasses, their alloys and combinations thereof. Preferably, the liquid metal is liquid gallium, its alloys, or combinations thereof. The liquid metal wets the particulate surface and provides a conduit for heat transfer from one insulating particle to another.

Since various properties of liquids and liquid metals depend upon their density and surface tension, it follows that dense liquid metals, such as mercury, would comprise a lower volume distribution as compared to liquid metals of lower density such as gallium and gallium alloys. In addition to its lower density, compared to mercury, gallium and gallium alloys have better wetting and surface tension characteristics that provide for easier distribution of the liquid metal phase within the composition.

The solid fillers, interchangeably referred to herein as "particulate fillers", are preferably thermally conductive but electrically insulating materials, and can be reinforcing or non- reinforcing. Further, the fillers can be micron-sized, sub-micron-sized, nano-sized, or a combination thereof.

The solid filler is preferably electrically insulating and comprises any insulator in particulate form, such as but not limited to, diamonds, graphite, carbon nanotubes, metal oxides (e.g., zinc oxide, aluminum oxide, magnesium oxide, titanium dioxide, zirconium oxide, chromium oxide, or iron oxide), metal hydroxides (e.g., aluminum hydroxides), metal oxy-hydroxides (e.g., bohemites), metal nitrides (e.g., boron nitride, aluminum nitride), metal nitrides with oxide coatings (e.g. silica coated aluminum nitride), metal particles or ceramic particles with an insulating coating (e.g., glass coated silver particles, alumina-coated silver particles, palmic acid coated aluminum particles, and aluminum particles with a natural oxide layer) and combinations and mixtures thereof. Glass coated silver particles include, but are not limited to silica coated silver particles, and borate coated silver particles. Additional fillers include fumed silica, fused silica, finely divided quartz powder, amorphous silicas, carbon black, silicone carbide, aluminum hydrates, and mixtures and combinations thereof.

The solid filler is typically present in an amount corresponding to between about 10 weight % and about 92 weight %, or any range, or set of ranges there between, based on the weight of the total final composition. In one embodiment, the filler is present in a range between about 13 weight % and about 90 weight %, based on the weight of the total final composition. In another embodiment, the filler is present in a range between about 15 weight % and about 80 weight %, based on the weight of the total final composition.

The selection of the filler size is established in order to achieve improved in-device thermal performance. The average particle size is typically within the range of about 0.01 to about 150 microns, in one embodiment about 0.01 to about 100 microns, and in another embodiment about 0.01 to about 75 microns. The maximum particle size in the formulation is typically between 0.1 - 1.0 times that of the desired bond line thickness, so that a balance can be achieved to minimize the resin-particle interfaces while still maintaining the desired bond line thick nesses. The desired bond line thickness is typically between about 0.01 mils and about 5 mils, with a range between about 0. 1 and 3 about mils being preferred in certain applications. In certain applications, such as thermal pads, the typical bondline thicknesses are in a range from about 8 mils to about 50 mils, and more typically between about 12 and about 25 mils.

The compositions of the present invention are characterized by a volume ratio of the liquid metal to particulate filler in a range from about 1 :1.1 to about 1:70 liquid metal to particulate filler. In one embodiment, the volume ratio of the liquid metal to particulate filler is in a range from about 1 :1.1 to about 1 :25 liquid metal to particulate filler. As is understood by those skilled in the art, the volume ratio and weight ratio of the various components of the compositions of the present invention are interrelated. Typically, in compositions wherein the volume ratio of liquid metal to particulate filler ranges from about 1 :1.1 to about 1:70 liquid metal to particulate filler, the weight ratio of liquid metal to particulate filler ranges from about 7:1 to about 1 :10 liquid metal to particulate filler. Thus, in one embodiment, the weight ratio of the liquid metal to particulate filler is in a range from about 7:1 to about 1 :10 liquid metal to particulate filler. In another embodiment, the weight ratio of the liquid metal to particulate filler is in a range from about 2.5:1 to about 1 :10 liquid metal to particulate filler. Table 1 illustrates how the weight ratios of the liquid metal to solid filler relate to the volume ratios as a function of the relative densities of the liquid metals and solid fillers. The combination of liquid metal and solid filler is present in the composition in an amount corresponding to about 20 to about 96 weight percent (or about 2 to about 90 volume percent), preferably about 30 to about 95 weight percent (or about 3 to about 88 volume percent). Tables 2 and 3 illustrate how the volume percentage of the total fillers relate to the weight percentage as a function of the densities of liquid metals, solid fillers and resins, as well as the weight ratio of liquid metal to solid fillers. The loadings of liquid metals and solid fillers may be varied depending on the application for which the composition is to be used in. For example, when the composition is to be used in managing heat dissipation in an electronic device, the various steps involved in device assembly; assembly conditions, resin curing conditions, and application pressure, as well as the identities of the liquid metal, the solid filler, and the resin employed, will determine the optimal concentration of each of the components of the composition. In one embodiment the present invention provides a composition comprising a liquid metal selected from the group consisting of gallium, gallium alloys, and mixtures thereof, a boron nitride particulate filler, and a silicone resin, wherein said liquid metal and particulate filler are present in a volume ratio of about 1 :0.4 to about 1 : 10.

In an another embodiment, the present invention provides a composition consisting essentially of a liquid metal selected from the group consisting of gallium, gallium alloys, and mixtures thereof, a boron nitride particulate filler, and a silicone resin, wherein said liquid metal and particulate filler are present in a volume ratio of about 1:0.4 to about 1 : 10.

In yet another embodiment, the present invention provides a composition comprising a liquid metal selected from the group consisting of gallium, gallium alloys, and. mixtures thereof, said liquid metal having a melting point in a range between about -10°C and about 35°C, a boron nitride particulate filler, and a silicone resin, wherein said liquid metal and particulate filler are present in a volume ratio of about 1 :0.4 to about 1: 10.

Typically, the composition is prepared by blending the liquid metal and solid fillers with a resin. The composition may also comprise, as optional components of the composition, an adhesion promoter and a catalyst. The presence of either or both of the adhesion promoter and a catalyst is optional, however. The resin employed is optionally curable or hardenable. Preferably, a high speeder mixer or a homogenizer is used during the mixing of the components to obtain a homogeneous mixture and to minimize "beading" of the liquid metal phase. "Beading" refers to the phase separation of the liquid metal into droplets within the composition. Additionally, the fillers may be further treated prior to or during mixing. Filler treatments include, but are not limited to, physical treatments such as ball-milling or jet-milling; chemical or physical coating or capping via procedures including chemical treatment such as treatments with silazanes, silanols, silane or siloxane compounds or polymers containing alkoxy, hydroxy or Si-H groups and any other known filler-treatment reagents, and any other procedures adopted by those skilled in the art. The final formulation can be hand-mixed or mixed by standard mixing equipment such as speedmixer, blender, dough mixers, charge can mixers, planetary mixers, twin screw extruders, two or three roll mills and the like. The blending of the formulations can be performed in batch, continuous, or semi-continuous mode by any means used Dy those skilled in the art. It is preferred that the blended formulation is a

homogenous mixture that does not bead.

Table 1

Weight Corresponding Ratio of DensityLiquid Metal Volume Ratio of Liquid Example Densitysoiid Fillers Liquid Metal to Metal to Solid Fillers Solid Fillers

7:1 7:1 to 1 :10 1:1 to 1:70 6:1 6:1 to 1 :10 1:1 to 1 :60 Hg and BN 5:1 5:1 to 1:10 1 :1 to 1 :50 Hg and Al 4:1 4:1 to 1 :10 1:1 to 1:40 Hg and γ- Al2O3 3:1 3:1 to 1 :10 1:1 to 1:30 Ga-In-Sn-Zn and BN 2.4:1 2.4:1 to 1:10 1:1 to 1:24 Ga-In-Sn-Zn and Al 2:1 2:1 to 1:10 1 :1 to 1 :20 Ga and AlN Ga-In-Sn-Zn and α- 1.7:1 1.7:1 to 1 :10 1:1 to 1 :17 Al2O3 1 :1 1 :1 to 1 :10 1 :1 to 1 :10 Ga and ZnO (1.07:1) Ga-In-Sn-Zn and borate 2:3 1.5:1 to 1 :10 1:1 to 1 : 6.67 glass coated silver particles 1:2 1 :2 to 1 :10 1:1 to 1 :5 Ga and coated Pd Ga-In-Sn-Zn and 1:3 1 :3 to 1:10 1:1 to 1 : 3.33 coated gold particles Table 2. The volume percentage of liquid metal and solid fillers as a function of the densities of the liquid metal, fillers and resins and the ratios of liquid metal and solid fillers. Total weight percentage of the fillers (liquid metal + solid filler) is 20 wt%. Examples Pliquid 3:1 liquid metal to 2:1 liquid metal to 1 : 1 liquid metal to 1 :2 liquid metal to 1 :10 liquid metal to metal-Pfiller- solid filler by weight solid filler by weight solid filler by weight solid filler by weight solid filler by weight Presin Vol% Vol% Vol% Vol% Vol% Vol% Vol% Vol% Vol% Vol% Vol% Vol% Vol% Vol% Vol% liquid total liquid total liquid total liquid total liquid filler total filler filler filler filler metal filler metal filler metal filler metal filler metal filler Ga-In- Sn-Zn 6.5 : and BN 2.7% 2.6% 5.4% 2.4% 3.5% 5.9% 1.8% 5.2% 7.0% 1.2% 6.8% 8.0% 0.3% 9.2% 9.5% 2.25: 1 in silicones Ga-In- Sn-Zn 6.5 : 2.7 4.9% 2.4% 2.9% 5.4% 1.8% 4.3% 6.2% 1.2% 5.7% 6.9% 0.3% 7.7% 8.1% and Al in :1 2.7% 2.2% silicones Ga-In- Sn-Zn 6.5 : 3.95 and 2.8% 1.5% 4.3% 2.5% 2.0% 4.5% 1.8% 3.0% 4.8% 1.2% 4.0% 5.2% 0.3% 5.4% 5.8% :1 Al2O3 in silicones Ga-In- Sn-Zn and 6.5 : 9.8 borate :1 2.8% 0.6% 3.4% 2.5% 0.8% 3.3% 1.9% 1.2% 3.1% 1.2% 1.7% 2.9% 0.3% 2.3% 2.6% coated Ag in silicones Table 3. The volume percentage of liquid metal and solid fillers as a function of the densities of the liquid metal, fillers and resins and the ratios of liquid metal and solid fillers. Total weight percentage of the fillers (liquid metal + solid filler) is 96 wt%.

Examples Pliquid 3:1 liquid metal to 2:1 liquid metal to 1 : 1 liquid metal to 1 :2 liquid metal to 1:10 liquid metal to metal-pfiller- solid filler by weight solid filler by weight solid filler by weight solid filler by weight solid filler by weight Presin Vol% Vol% Vol% Vol% Vol% Vol% Vol% Vol% Vol% Vol% VoP/ Vol% Vol% Vol% Vol% liquid total liquid total liquid total liquid total liquid filler total filler filler filler filler metal filler metal filler metal filler metal filler metal filler Ga-In- Sn-Zn 6.5 : and BN 43% 41% 23% 65% 2.25: 1 84% 35% 51% 86% 88% 13% 76% 89% 3.0% 88% 91% in silicones Ga-In- Sn-Zn 6.5 : 2.7 46% 37% 83% 38% 46% 84% 25% 61% 86% 15% 73% 88% 3.6% 86% 89% and Al in :1 silicones Ga-In- Sn-Zn 6.5 : 3.95 and 52% 29% 81% 45% 37% 82% 31% 52% 83% 20% 64% 84% 4.9% 81% 85% :1 Al2O3 in silicones Ga-In- Sn-Zn + 6.5 : 9.8 borate :1 63% 14% 77% 58% 19% 77% 45% 30% 75% 32% 42% 74% 9.4% 63% 72% coated Ag in silicones Using fillers in accordance with the present invention provides lower thermal resistance while maintaining sufficiently low viscosities to allow easy processing and which will flow as necessary for preparation of electronic devices, especially flip-chip devices.

The resin can be any polymeric material. Suitable organic resins include, but are not limited to, polydimethylsiloxane resins, epoxy resins, acrylate resins, other organo- functionalized polysiloxane resins, polyimide resins, fluorocarbon resins, benzocyclobutene resins, fluorinated polyallyl ethers, polyamide resins, polyimidoamide resins, phenol cresol resins, aromatic polyester resins, polyphenylene ether (PPE) resins, bismaleimide triazine resins, fluororesins, mixtures thereof and any other polymeric systems known to those skilled in the art. (For common polymers, see "Polymer Handbook", Brandrup, J.; Immergut, E.H; Grulke, Eric A; Wiley Interscience Publication, New York, 4th ed. (1999); "Polymer Data Handbook" Mark, James Oxford University Press, New York (1999)). Resins may also include hardenable thermoplastics.

The resin may be a curable or thermosetting resin, including but not limited to a silicone resin, an epoxy resin, an acryloxy resin, or any combination thereof. The composition is a dispensable formulation at mixing that hardens to an immobilized solid upon curing. The final properties of thermal conductivity and electrical insulation are obtained on curing. In some embodiments of the present invention, curing is facilitated at particular temperatures. Preferably, resins are chosen such that the curing temperature is about 10°C to about 250°C, still more preferably about 40°C to about 2000C.

Suitable curable thermoset matrices are acrylate resins, epoxy resins, polydimethyl siloxane resins, other organo-functionalized polysiloxane resins that can form cross- linking networks via free radical polymerization, atom transfer radical polymerization, nitroxide mediated radical polymerization, reversible addition- fragmentation transfer polymerization, ring-opening polymerization, ring-opening metathesis polymerization, anionic polymerization, cationic polymerization or any other method known to those skilled in the art, and mixtures thereof. Suitable curable silicone resins include, for example, the addition curable and condensation curable matrices as described in "Chemistry and Technology of Silicone", Noll, W.; Academic Press 1968.

In another embodiment, the resin can be an organic-inorganic hybrid matrix. Hybrid matrices include any polymers that contain chemically bound main group metal elements (e.g., aluminum, magnesium, gallium, indium), main group semi-metal elements (e.g. boron, germanium, arsenic, antimony), phosphorous, selenium, transition metal elements (e.g., platinum, palladium, gold, silver, copper, zinc, zirconium, titanium, ruthenium, lanthanum, etc.) or inorganic clusters (which include, but are not limited to, polyhedral oligomeric silsesquioxanes, nano metal oxides, nano silicon oxides, nano metal particles coated with metal oxides, and nano metal particles.) For typical examples and methods of forming inorganic-organic hybrids, see reviews such as "Hybrid Organic Inorganic Materials - in Search of Synergic Activity" by Pedro Gomez-Romero, Advanced Materials, 2001, Vol. 13, No. 3, pp. 163-174; "Inorganic Clusters in Organic Polymers and the Use of Polyfunctional Inorganic Compounds as Polymerization Initiators" by Guido Kickelbick and Ulrich Schubert, Monatshefte fur Chemie, 2001, Vol. 132, pp. 13-30; "Synthesis and Application of Inorganic/Organic Composite Materials", by Helmut Schmidt, Macromolecular Symposia, 1996, Vol. 101, pp. 333-342; and "Synthesis of Nanocomposite Organic/Inorganic Hybrid Materials Using Controlled/' Living' Radical Polymerization" by Jeffrey Pyun and Krzysztof Matyjaszewski, Chemistry of Materials, 2001, Vol. 13, pp. 3436-3448. As used herein, "chemically bound" refers to bonding through a covalent bond, an ionic interaction, an iono-covalent bond, a dative bond or a hydrogen bond. Organic-inorganic hybrid polymeric matrices may refer to, but are not limited to, co-polymerization products between organic monomers, oligomers or polymers that contain polymerizable groups such as alkenyl, allyl, Si-H, acrylate, methacrylate, styrenic, isocyanate, epoxide and other common groups known to those skilled in the art, and inorganic clusters or organometallic ■ compounds containing polymerizable groups. For example, the copolymerization product between an acrylate or a methacrylate and a metal acrylate or methacrylate compound is an organic-inorganic hybrid polymeric matrix. The copolymerization product between an epoxide and an epoxide-functionalized inorganic cluster is also considered an inorganic-organic hybrid polymer. The homo-polymerization products of organo-functionalized inorganic clusters or organometallic compounds, or the copolymerization products among different organo-functionalized inorganic clusters or organometallic compounds, are also considered organic-inorganic hybrid matrices. Organic-inorganic hybrid matrices also include cases where the inorganic cluster or organometallic compound has no polymerizable functional groups, but can become part of the polymer network through its surface OH or other functional groups.

In certain embodiments thermal interface composition can be formulated as a gel, grease or phase change material that can hold components together during fabrication and thermal transfer during operation of the invention. By phase change material is meant a material that has a melting or softening point above room temperature, at which point the material's viscosity decreases sufficiently to allow it to flow and wet the surfaces of a heat generating device and a heat dissipating device. The phase change materials may comprise wax compounds, polyalkylsiloxanes, siloxanes containing silicon-phenyl moieties, oligo- or low molecular weight polyolefins, C 12- Cl 6 alcohols, acids, esters, methyl triphenyl silanes, combinations thereof, and the like, but not limited thereto. When the polymer matrix is not curable or hardenable, common organic liquids such as ionic liquids can also be used as the resin material.

The composition of the present invention may further include an adhesion promoter. An adhesion promoter may not only facilitate improved chemical interaction between precursors within the composition such as an increased compatibility among the liquid metal-filler-curable resin and other additives, but also improve cured composition's adhesion to the substrate. The adhesion promoters are present in an amount of from about 0 weight percent and about 5 weight percent, preferably, from about 0.01 weight percent and about 5 weight percent, more preferably about 0.01 to about 2 weight percent of the total final formulation, or any range or combination of ranges there between.

Adhesion promoters that can be employed include alkoxy- or aryloxysilanes such as γ-aminopropyltrimethoxysilane, 3-glycidoxypropyltrimethoxysilane and bis(trimethoxysilylpropyl)fumarate, or alkoxy- or aryloxysiloxanes such as tetracyclosiloxanes modified with acryloxytrimethoxysilyl or methacryloxypropyltrimethoxysilyl functional groups. They may also include, but are not limited to, silanols, oligosiloxanes containing an alkoxy silyl functional group, oligosiloxanes containing an aryloxysilyl functional group, polysiloxanes containing an alkoxy silyl functional group, oligosiloxanes containing a hydroxyl functional group, polysiloxanes containing an aryloxysilyl functional group, polysiloxanes containing a hydroxyl functional group, cyclosiloxanes containing an alkoxy silyl functional group, cyclosiloxanes containing an aryloxysilyl functional group, cyclosiloxanes containing a hydroxyl functional group, titanates, trialkoxy aluminum, tetraalkoxysilanes, isocyanurates, and mixtures, and combinations thereof.

The composition may also further contain at least one catalyst. The catalyst is selected from any group of catalysts compatible with the curable resin utilized in the present invention. Where epoxy resins are utilized, hardeners such as carboxylic acid- anhydride curing agents and an organic compound containing hydroxyl moiety can be added as optional reagents with the curing catalyst. For epoxy resins, exemplary anhydride curing agents typically include methylhexahydrophthalic anhydride, 1,2- cyclohexanedicarboxylic anhydride, bicyclo[2.2.1] hept-5-ene-2,3-dicarboxylic anhydride, methylbicyclo [2.2.1 ] hept-5 -ene- 2,3-dicarboxylic anhydride, phthalic anhydride, pyromellitic dianhydride, hexahydrophthalic anhydride, dodecenylsuccinic anhydride, dichloromaleic anhydride, chlorendic anhydride, tetrachlorophthalic anhydride, and the like. Combinations comprising at least two anhydride curing agents may also be used. Illustrative examples are described in "Chemistry and Technology of the Epoxy Resins" B. Ellis (Ed.) Chapman Hall, New York, 1993 and in "Epoxy Resins Chemistry and Technology", edited by C. A. May, Marcel Dekker, New York, 2nd edition, 1988. Additional catalysts include amines, alkyl-substituted imidazole, imidazolium salts, phosphines, metal salts, triphenyl phosphine, alkyl- imidazole, and aluminum acetyl acetonate, iodonium. compounds, onium salts and combinations thereof. For epoxy resins, curing agents such as multi-functional amines or alcohols can be optionally incorporated as cross-linking agents. Exemplary amines may include, but are not limited to ethylene diamine, propylene diamine, 1 ,2- phenylenediamine, 1,3-phenylene diamine, 1 ,4-phenylene diamine, and any other compounds containing 2 or more amino groups. Exemplary alcohols may include, but are not limited to, pnenoiic resms, Novolak systems, bisphenols, and any other compounds containing two or more hydroxyl groups, or others known to one of ordinary skill in the art.

Where acrylates are used, curing catalysts can be selected from, but are not limited to, cationic curing initiators such as iodonium compounds or onium salts, or radical cuing initiators such as peroxides or azo-compounds, or others known to one of ordinary skill in the art.

Where condensation-cure siloxane resins are used, an optional Lewis-acidic catalyst such as an organometallic tin compound (e.g. Sn(acetate)2) can be used.

Additionally, for addition curable silicone resins, catalysts include compounds containing Group 8-10 transition metals (i.e. ruthenium, rhodium, platinum, palladium) complexes. Preferably the catalyst for an addition curable silicone resin is a platinum complex. Preferred platinum complexes include, but are not limited to, fine platinum powder, platinum black, platinum adsorbed on solid supports such as alumina, silica or activated carbon, choroplatinic acid, platinum tetrachloride, platinum compounds complexed with olefins or alkenyl siloxanes such as divinyltetramethyldisiloxanes and tetramethyltetravinylcyclotetrasiloxane, platinum compounds complexed with phosphine, phosphite, sulfide, amino or pyridinyl ligands, and combinations thereof.

Catalyst inhibitors can be added to modify the curing profile of addition curable silicone resins and to achieve desired shelf life for the composition. Suitable inhibitors include, but are not limited to, phosphine or phosphite compounds, sulfur compounds, amine compounds, isocyanurates, alkynyl alcohol, maleate and fumarate esters, and mixtures thereof, and other compounds known to those skilled in the art. Some representative examples of suitable inhibitors also include triallylisocyanurate, 2-methyl-3-butyn-2-ol, triphenylphosphine, tris(2,4-di-(tert)-butylphenyl)phosphite, diallyl maleate, diethyl sulfide and mixtures thereof.

Another aspect of the present invention includes methods for preparing the described composition. The method comprises the steps of: preparing a precursor composition, said precursor composition comprising at least one of a liquid metal, liquid metal alloy, or a combination thereof, a solid filler, a resin, and optionally an adhesion promoter and a catalyst; blending the mixture, and optionally curing the resultant composition.

To achieve homogeneity in the precursor composition, various mixing stratagems may be employed. It is sometimes preferable that the liquid metal be uniformly dispersed in the resin before the addition of solid particle filler; or a portion of the solid particle filler is added to the resin to form a flowable mixture, to which liquid metal is subsequently added and dispersed before adding the remaining filler; or a thorough mixing of liquid metal and filler is done before addition of the resin. Alternatively, resin and solid particulate filler may be pre-mixed prior to addition of liquid metal followed by homogenization with a high speed mixer, a homogenizer or any other types of mixer know to those skilled in the art. The order of addition, or mixing, of the liquid metal, filler and resin is not critical to this aspect of the invention and any combination of these steps will provide the composition of the present invention.

The liquid composition is thoroughly mixed and blended into a thixotropic paste. The thixotropic paste is applied on a variety of heat generating and heat dissipating substrates including glass, metal, plastic, ceramic, semiconductor, electronic devices and combinations thereof. The mixture may be applied between two surfaces and optionally cured or hardened in place to provide a thermal interface material. The preferred viscosity of the pre-cured composition is preferably less than about seven million cps and more preferably less than about five million cps at room temperature. In one embodiment of the claimed invention, the thixotropic paste is further degassed and cured at a temperature about 120-1500C. In addition to being applied as adhesive formulations, the present compositions may also be applied as grease, gel and phase change material formulations. Alternatively, the present compositions can be pre¬ formed into sheets or films and cut into any desired shape. In this embodiment, the compositions can advantageously be used to form thermal interface pads or films that are positioned between electronic components.

The curing process can be performed by any process known to those skilled in the art. Curing can be done by methods such as thermal cure, UV light cure, microwave cure, e-beam cure, free radical cure initiated with free radical initiators and combinations thereof. Typical free radical initiators may include, but are not limited to, organic peroxides (e.g., benzoyl peroxide), inorganic peroxides (e.g., hydrogen peroxide), organic or inorganic azo compounds (e.g., 2-2'-azo-bis-isobutyrylnitrile), nitroxides (e.g. TEMPO) or combinations thereof. Curing typically occurs at a temperature in a range between about 100C and about 25O0C, more typically in a range of about 4O0C and about 2000C. Curing typically occurs at a pressure in a range between about 1 atmosphere (atm) and about 5 tons pressure per square inch, more typically in a range between about 1 atm and about 100 pounds per square inch (psi). In addition, curing may typically occur over a period in a range between about 30 seconds and about 5 hours, and more typically in a range between about 90 seconds and about 120 minutes. Optionally, the cured composition can be post-cured at a temperature in a range between about 12O0C and about 1500C over a period of about 1 hour to about 4 hours. In one embodiment of the present invention, the composition after curing provides a volume resistivity of greater than about 108 ohm-cm and a dissipation factor of less than about 0.1, preferably less than about 0.01, and still more preferably, less than about 0.001.

Another aspect of the present invention is the use of the composition as a thermal interface material (TIM) in a wide variety of electrical devices. Electronic devices include, for example, electronic components found in computers, semiconductors, and electronic devices wherein heat transfer between components is needed. In one embodiment, the electronic component includes a semiconductor chip as a heat generating, or producing, component. In such a case, the heat producing component can be a chip carrier, an area array package, a chip scale package, or other semiconductor packaging structure. In other embodiments, the semiconductor chip itself is the heat producing component.

Application of the present thermal interface compositions may be achieved by any method known in the art. Conventional methods include screen printing, stencil printing, syringe dispensing, pick-and-place equipment and pre-application to either the heat generating or heat dissipating unit. wxiiie me present disclosure nas provided details on how the present compositions may be utilized as thermal interface material in electronic devices, the compositions of the present disclosure may be applied in any situation where heat is generated and needs to be removed. For example, the compositions of the present disclosure may be utilized to remove heat from a motor or engine, to act as underfill material in a flip- chip design, as die attachment means in an electronic device, and in any other applications where efficient heat-removal is desired.

As shown schematically in Figure 1, a thermal interface composition 20 can be inteiposed between a heat generating component 30 and a heat spreader (or heat dissipating) unit 10 to fill any air gaps and facilitate heat transfer, hi this embodiment, the same or different thermal interface composition is also interposed between the heat spreader 10 and a heat sink 40. This Figure is not intending to be limiting, but to show one embodiment of the present invention.

Figure 2 is a cross-section of a thermally conductive, electrically insulating TIM in accordance with the present invention. The TIM 100 is placed between an electronic device 50 and a heat sink/heat spreader 60. The TIM 100 is a polymeric resin 70, such as silicone based material, a liquid metal 80, such as gallium, and a particulate filler 90, such as aluminum oxide. The TIM fills any air gaps and facilitates heat transfer.

In one aspect, the present invention provides a method for improving heat transfer from a heat source to the environment. In one embodiment, this improved heat transfer takes the form of enhanced heat transfer rates from a heat source in contact with a composition of the present invention, said composition also being in contact with a heat dissipating unit. Methods for increasing heat transfer using the compositions of the present invention include positioning a heat producing component in contact with a thermally conductive, electrically non-conducting thermal interface composition of the present invention, said composition comprising a resin, a liquid metal and a particulate filler, and positioning a heat dissipating unit, such as a heat sink in contact with the thermal interface composition.

In one embodiment of the present invention, such as where the electronic component is a chip, the heat producing component may be placed in contact with a printed circuit board, and an electrical connection is formed between the component and at least one electrical contact of the printed circuit board. A thermal interface composition of the present invention, which includes a blend of a resin, liquid metal and particulate filler, is applied between the component and the printed circuit board so that the thermal interface composition encapsulates at least one electrical connection.

The following examples are included to illustrate the various features and advantages of the present invention, and are not intended to limit the invention.

Example 1

A commercial grade of addition curable polydimethylsiloxane, ECC 4865 (4.29 grams, GE Silicones) was used as the matrix material. About 18.98 grams of gallium (Aldrich, 9.9.999%) was melted in an oven at a temperature of about 500C and added to the silicone. After stirring and dispersing the gallium in silicone, about 3.72 grams of aluminum oxide (Sumitomo's AA04, average particle size 0.4μm) and a further 17.81 grams of aluminum oxide (Showa Denko's AS20, average particle size 21μm) were added in small portions with stirring to ensure proper mixing. In the final mixture, the resin to liquid metal to solid filler ratio was 1 : 4.42 : 5.02 by weight, and the liquid metal to filler volume ratio is 1 : 1.7 . The flowable gray mixture was poured into a 50 mm circular mold, degassed at 50°C for 1 hour and cured in a Carver press at 1500C, under a pressure of 5000 pounds retained for 45 minutes. The thickness of the resultant gray disc was measured and found to be about 2.50 mm in thickness, and the disc was determined to be electrically non-conductive using an Ohmmeter. Thermal conductivity was determined using a Holometrix TCA 300 instrument at 1000C. Viscosity was determined using a Brookfield cone and plate viscometer. The thermal conductivity was found to be 2.11 W/mK. The initial viscosity of the uncured formulation was 91,200 ± 2000 cps at 2.5 rpm at room temperature.

Example 2

The formulation of Example 1 was repeated but with a different ratio of components: 4.06 grams of ECC4865 were used as the matrix material. 25.57 grams of gallium were mixed with ECC4865 first. The liquid mixture was then mixed with 4.46 grams Of Al2O3 (Sumitomo's AA04, average particle size 0.4μm) and about 21.37 grams of Al2O3 (Showa Denko's AS20, average particle size 21 μm). In the final mixture, the resin to liquid metal to solid filler ratio was 1: 6.30 : 6.36 by weight, and gallium to alumina's volume ratio was 1: 1.5. A cured disc prepared as in Example 1 measured 1.61 mm in thickness, and was determined to be electrically non-conductive by an Ohmmeter. The thermal conductivity was outside the calibration range for the machine, but was estimated to be around 3.00 W/mK at 100°C. The initial viscosity of the uncured formulation was 208,000 ± 2000 cps at 2.5 rpm at room temperature.

Example 3

The formulation of Example 1 was repeated but with a different ratio of components: About 3.58 grams of ECC4865 were used as the matrix material. About 19.50 grams of gallium were prepared with about 3.95 grams of Al2O3 (Sumitomo's AA04, average particle size 0.4μm) and about 18.7 grams Of Al2O3 (Showa Denko's AS20, average particle size 21 μm). In this instance, gallium was added last, and beading of gallium was observed. In the final mixture, the resin to liquid metal, to solid filler ratio is 1 : 5.45 : 6.33 by weight, and gallium to alumina's volume ratio was 1: 1.7. A cured disc prepared as in Example 1 measured 2.58 mm in thickness, and was determined to be electrically non-conductive by an Ohmmeter. The sample underwent three thermal conductivity measurements at 1000C which yielded an average value of about 2.75 ± 0.01 W/mK. The initial viscosity of the uncured formulation was not measured.

Example 4

A base siloxane was prepared by mixing 20 grams of GE Silicones intermediate 81865, 5 grams of GE Silicones intermediate 88765, 1.7 g of GE Silicones intermediate 88104, 9.1 g of Gelest, Inc. intermediate DMSH03 and 0.85g of GE Silicones intermediate 89174. 6.65 g of the above mixture was mixed with 19.60 grams of gallium, 4.90 grams of Al2O3 (Sumitomo's AA04, average particle size 0.4μm), 27.60 grams of Al2O3 (Showa Denko's AS20, average particle size 21 μm) and 9.00 grams Of Al2O3 (Showa Denko's AS40, average particle size 10 μm). In the final mixture, the resin to liquid metal to solid filler ratio was 1: 2.95: 6.24 by weight, and gallium to alumina's volume ratio is 1 : 3.2. The final mixture was degassed at room temperature for 48 hours and cured in a Carver press at 150°C, 5000 pounds pressure for 40 minutes. The resultant cured disc measured 3.15 mm in thickness, and was somewhat uneven on one surface (Surface B). It was determined to be electrically non-conductive by an Ohmmeter. The sample underwent four thermal conductivity measurements at 100°C. The recorded thermal conductivity values were: 2.20 W/mK (Surface A facing up), 1.96 W/mK (Surface B facing up), 1.99 W/mK (surface B facing up) and 2.10 W/mK (surface A facing up). The viscosity of the uncured final formulation mixture was 330,600 cps at 2.5/s at room temperature.

Example 5

A base polymer matrix, C836-039-uv9380c, was prepared by mixing about 20 grams of methacryloxypropyltrimethoxysilane (MAPTMS), about 10 grams of acryloxy- capped polydimethylsiloxane polymer (DMSU22 obtained from Gelest) and about 0.63 grams of an iodonium cure catalyst (GE intermediate UV9380c). 5.50 grams of the above mixture was mixed first with 31.34 grams of Al2O3 (Showa Denko AS20, average particle size 21 μm), and then with 16.95 grams of gallium. Beading was observed. To this mixture, 0.16 g additional C836-039-uv9380c was added, and after proper mixing a smooth thixotropic mixture resulted. 2.50 grams of Al2O3 (Showa Denko AS20, 21 μm) was added to the final mixture to give a final formulation consisting of 5.66 grams of base polymer matrix, 16.95 grams of gallium and 33.84 grams Of Al2O3 in a wt. ratio of 1 : 2.99 : 5.98 (polymer : gallium : solid filler). The volume ratio between gallium and alumina was 1 : 3. The mixture was degassed at 40 0C for 1 hour and cured in a Carver press at 1500C, under a pressure of 5000 pounds for 45 minutes. The resultant cured disc measured 3.30 mm in thickness and was determined to be electrically non-conductive by an Ohmmeter. Two thermal conductivity measurements were completed on the sample at 100°C yielding an average value of 3.18 ± 0.05 W/mK. The initial viscosity of the uncured formulation was not measured.

Example 6 βu^δoo was mixeα wim appropriate amounts of gallium-indium-tin alloy (62 wt%

Ga: 25 wt%In : 13 wt%Sn, Indium Corporation of America) to form an emulsion.

Alumina (a combination of AS40 from Showa Denko and AA04 from Sumitomo) was

added to the mixture in small portions with stirring. The mixture was degassed at

room temperature for 3-12 hours and cured at 150 °C, under a pressure of 5000

pounds for 45-60 minutes. The results obtained using compositions of thermally

conductive adhesives containing a combination of these liquid metals and solid

particles are listed in Table 4. The viscosity was measured by a rheometer. When the

total filler volume was kept constant, the viscosity of the composition before cure,

decreased with increasing percentage of liquid metal alloys(6a - 6d; 6e - h; 6i - 61).

The thermal conductivity after cure, however, did not change significantly with a

varying liquid metal to alumina ratio.

Table 4 Compositions Tested Alumina (g) liquid metal : alumina (vol) Example ECC4865(g) Ga/ln/Sn (g) (x:y (total filler vol%) AS40: AA04) 6a 4.00 0 32.13 (4: 1) 0 : 2.03 (67%) 6b 4.00 12.28 24.62 (4:1) 1 : 3.4 (67%) 6c 4.00 16.44 22.35 (4: 1 ) 1 : 2.3 (67%) 6d 4.00 19.95 19.92 (4:1) 1 : 1.7 (67%) 6e 4.00 0 32.08 (1 :0) 0 : 2.03 (67%) 6f 4.03 12.35 24.60 (1 :0) 1 :3.4 (67%) 6g 4.04 16.71 22.26 (1 :0) 1 : 2.3 (67%) 6h 4.09 20.37 20.37 (1 :0) 1 : 1.7 (67%) 6i 4.00 0 32.07 (6:1) 0: 2.03 (67%) 6j 4.00 12.27 24.58 (6:1) 1 :3.4 (67%) 6k 4.00 16.42 22.1 1 (6:1) 1 :2.3 (67%) 61 4.00 19.96 19.94 (6:1) 1 : 1.7 (67%) 6m 4.01 12.83 25.32 (6:1) 1 :3.3 (67.6%) 6n 4.00 16.32 21.77 (6:1) 1 :2.2 (66.8%) 6o 4.00 19.10 19.10 (6:1) 1 : 1.7 (66.0%) Table 4 (Continued) lest Result Data

liquid metal : Thermal alumina (vol) conductivity Viscosity (cps, Dissipation factor Dielectric Constant Example (total filler (1000C, 2.5 /s, r.t.) (10kHz/100Hz) (10kHz/100Hz) vol%) W/mK) 6a 0 : 2.03 2.18 ± 0.05 88,400 0.0001 5.96/ (67%) /0.0030 5.98 6b 1 : 3.4 2.04 + 0.03 74,300 0.0012 9.31/ (67%) /0.0024 9.35 6c 1 : 2.3 2.27 ± 0.04 69,500 0.0008/ 10.95/ (67%) 0.0019 10.98 6d 1 : 1.7 2.21 ± 0.07 41,200a 0.0008/ 1 1.29/ (67%) 0.0021 1 1.29 6e 0 : 2.03 1.96 + 0.04 214,000 N/A N/A (67%) 6f 1 :3.4 2.01 ± 0.04 213,000 N/A N/A (67%) 6g 1 : 2.3 1.99 + 0.05 163,000 N/A N/A (67%) 6h 1 : 1.7 1.88 + 0.04 124, 000 N/A N/A (67%) 6i 0: 2.03 1.89 ± 0.04 163,000 N/A N/A (67%) 6j 1 :3.4 1.99 + 0.04 153,000 N/A N/A (67%) 6k 1 :2.3 1.85 + 0.04 98,500 N/A N/A (67%) 61 1 : 1.7 1.96 72,900 N/A N/A (67%) 6m 1 :3.3 2.08 130,000 N/A N/A (67.6%) 6n 1 :2.2 1.89 N/A N/A N/A (66.8%) 6o 1 :1.7 1.73 N/A N/A N/A (66.0%) a) Measured using a Brookfield viscometer at 2.5 rpm at room temperature.

Examples 7

Liquid metal (61% Ga, 25% In, 13% Sn and 1% Zn from Indium Corporation of

America) was added to commercial silicone adhesive TSE3281G from GE Toshiba a

weight ratio of 1 : 6.35 (7b, Liquid metal : TSE3281G) and 1 :2 (7c, Liquid metal :

TSE3281G). TSE3281G is a silicone adhesive containing ~ 84.6 wt% alumina. The

liquid metal was uniformly dispersed into the silicone adhesive first by hand mixing .

and then with a Speedmixer at 900 rpm for 5 seconds. Qualitatively, the final

mixtures had similar flowability as the control TSE3281G. In 7b, the ratio of the

polymer matrix to liquid metal to solid fillers was 1: 1.02: 5.49 by weight, and the

volume ratio between the liquid metal and alumina is 1 : 9.1. In 7c, the ratio of the poiymer matπx to liquid metal to solid fillers was 1: 3.25: 5.49 by weight, and the

volume ratio between the liquid metal and alumina was 1: 2.8. 3 -layer sandwiched

structures consisting of silicon-TIM-aluminum were built using these adhesives with

an assembly pressure of 10 or 50 psi. Force was applied for no longer than 3 seconds

at the prescribed pressure. The adhesives were cured between the silicon and

aluminum coupons at 150°C for 2 hours under no external pressure. Four sandwiched

structures were built for each adhesive under each prescribed pressure. The thermal

diffusivities of the sandwiched complexes were measured using Microflash 300

(Netzsch Instruments), and the in-situ thermal resistances were calculated using

manufacturer-provided macros. As shown in Table 5, lower thermal resistance and

thereby better in-situ thermal performance were obtained with formulations

containing liquid metal and assembly pressures of 50 psi. At 10 psi assembly

pressures, i. e. pressures that would be more practical in industry, formulations

containing sufficient amounts of liquid metal, 7b, also showed better in situ thermal

performance than 7a, formulations without liquid metal, while maintaining similar

flowability and processibility.

Table 5

Average Volume ratio of Bond Line Thermal Resistance Thermal Example Descripti on liquid metal : thickness (mil) Range (mm2- Resistance alumina K/W) (mm2- K/W) Remixed 2.68 ± 0.232 59 - 692 63 ± 42 7a TSE3281 N/A G 2.25± 0.453 55 - 613 57 ± 33 TSE3281 2.76 ± 0.442 56 - 732 66 ± 82 G + 7b liquid metal' 1: 9.1 2.03 ± 0.103 46 - 533 49 ± 33 (6.35:1 Wt) TSE3281 2.69 ± 0.482 53 - 562 54 ± 22 7c g + liquid 1: 2.8 metal1 2.17 ± 0.083 45 - 523 48 ± 33 (2:1 wt) ' Liquid metal is 61% Ga, 25% In, 13% Sn and 1% Zn from Indium Corporation of America; TSE3281g is a silicone adhesive containing ~ 84.6 wt% alumina. 2assembly pressure = 10 psi, sample size = 4 3assembly pressure = 50 psi, sample size =4 In the experimental descriptions which follow, the following components forming the compositions of the examples below were prepared or purchased. Component A A-I was a 400 cps vinyl-terminated polydimethylsiloxane fluid (SL6000, GE Silicones). A-2 was RTV615A from GE Silicones. A-3 was l-ethyl-3-methyl-imidazolium tetrafluoroborate. A-4 was a 1000 cps polydimethylsiloxane fluid (SF96-1000, GE Silicones). Component B B-I was a silicon hydride fluid with an average chain length of about 100, and a hydride content of about 0.72% to about 1%(88466, GE Silicones). B-2 was RTV615B from GE Silicones. B-3 was a silicon hydride fluid with an average chain length of about 21, and a hydride content of about 0.19% to about 0.25% (88405, GE Silicones). B-4 was a silicon-hydride-terminated polydimethylsiloxane fluid with an average molecular weight of about 6000, and an average hydride content of about 0.04% (DMS-H2l, Gelest). Component C Component C was a liquid metal (61% Ga, 25% In, 13% Sn and 1% Zn from Indium Corporation of America). Component D D-I was alumina powders of average size 5 microns (Denka). D-2 was alumina powders of average size 0.4 microns (Sumitomo). D-3 was Alcoa 9280, an aluminum powder (Aloca). JLJ-4 was (J(JόU04, plateiet-hke boron nitride powders from GE Advanced Ceramics.

D-5 was Teco 2003124 A, an experimental grade boron nitride powder from GE Advanced Ceramics, average particle size 60 - 70 microns.

D-6 was NXl, an experimental grade boron nitride powder obtained from GE Advanced Ceramics having an average particle size less than 1 micron.

D-7 was TECO 2004112-B, an experimental grade boron nitride powder from GE Advanced Ceramics having average particle size 25 - 30 microns.

D-8 was PTl 20, a platelet-like boron nitride powder from GE Advanced Ceramics.

Component E

Component E was a 75:8 (by weight) mixture of triallylisocyanurate (TAIC) and 2- methyl-3-butyn-2-ol (SURFINOL).

Component F

F-I was a stock solution of a tetramethyltetravinylcyclotetrasiloxane-complexed platinum catalyst (GE Silicones, 88346) in vinyl-terminated polydimethylsilxoane SL6000 ([Pt] = 570 ppm).

F-2 was a stock solution of a tetramethyltetravinylcyclotetrasiloxane-complexed platinum catalyst and IRGAFOS 168 in vinyl-terminated polydimethylsilxoane SL6000 ([Pt] = 480 ppm, molar ratio of Irgafos 168 : Pt = 2:1).

Component H

H-I was a 44:29 mixture (by weight) of A501S (GE Toshiba) and glycidoxypropyltrimethoxysilane (GLYMO).

H-2 was A580, an experimental grade silicone fluid from GE Toshiba.

Comparative Examples 1 - 2 & Example 8

Component A-I was blended with D-I and D-2 first by hand, then in a Waring pulverizer mounted on a Waring 2-speed blender base. The speed of the pulverizer was (jυnirυneα Dy com me Hi/UJ selection button on the blender base and the setting of the variac, to which the blender base was connected. The pulverizer was wrapped with a heating tape, which was connected to a second variac. The pulverizer was heated to 80 - 100° C, and the speed of mixing was set to LO, 15 - 20. The blender was periodically stopped, the sides and the bottom were scraped, and the blender was then re-started. This process was repeated over a period of 2 hrs. A portion of Component B-I was then added, and the mixture was blended for an additional 15 - 30 minutes. Afterwards, the mixture was transferred to a plastic container, and placed in a vacuum oven at 75° C. over night (- 12 hours). The vacuum gauge read between 25-27 inches Hg.

The mixture was allowed to cool to room temperature, and was then divided into four portions. To each portion, a predetermined amount of Component C (see Table 6 for amounts) was added, and the resultant composition was blended with a speedmixer (FlackTek Inc., Model # DAC400FV) until homogeneous mixture was obtained. Homogeneity was determined by visual inspection. Components E, F-I, H-I and the remainder of component B-I were then added to the mixture, and the composition was thoroughly mixed with a speedmixer (FlackTek Inc., Model # DAC400FV) at 1200 rpm for 5 seconds. The mixture was then degassed at room temperature for 15 — 24 hours (house vacuum). The compositions prepared are given in Table 6.

The mixture was then interposed between two metal coupons at a prescribed pressure for 1 second and cured under pressure (typically 10 or 20 psi) at 150°C over 2 hours. The diffusivity of the resulting sandwiched structure was then determined by laser flash (Netzsch Instrument, Microflash™ 300), and the in-situ thermal resistivity determined by a software macro provided with the Microflash™ instrument. No external pressure was exerted during the measurement for adhesives. 4 - 5 samples were prepared utilizing the formulation between the AlCr and Si coupons.

As seen from Table 6, when tested as adhesives, the addition of a small amount of liquid metal alloy (Example 8 relative to Comparative Example 1) resulted in a 35% reduction in thermal resistance over Comparative Example 1 at comparable bond line thicknesses. When a large quantity of liquid metal alloy was added (Comparative ϋxampie Z), a signmcant viscosity increase was observed, which resulted in reduced

thermal performance compared to Example 8.

Table 6

Examples Comparative Ex. 1 Example 8 Comparative Ex. 2 A-I 100 100 100 B-I 2.0 2.0 2.0 C 300 1200 D-I 480 480 480 D-2 120 120 120 E 0.58 0.62 0.61 F-I 1.0 1.0 1.0 H-I 5.1 5.2 5.1 (A+B+E+F+H): 1 : 0: 5.52 1 : 2.77: 5.55 1 : 11 : 5.5 C: D (wt.) C: D (vol.) 0: 1.4 1 : 3.4 1 : 0.84 Low viscosity, Flowable under Very thick, cream- Appearance flowable shear, thicker like, almost non- than control flowable Tested as Adhesives: Sandwich Al-TIM-Si Al-TIM-Si Al-TIM-Si Material Assembly/Cure 10/10 10/10 10/10 Pressure (psi) Bondline 1.2 ± 0.2 1.2 ± 0.2 5.6 ± 0.8 Thickness (1.1 - 1.6) (1.1 - 1.6) (4.2 - 6.2) (range, mils)* In-situ Thermal Resistivity 29 ± 3 19 ± 3 34 ± 6 @25°C (mm2- (26 - 34) (17 - 23) (28 - 40) KAV) * In-situ Thermal Conductivity 1.1 ± 0.2 1.6 ± 0.3 4.3 ± 0.6 @25°C (0.9 - 1.4) (1.4 - 2.0) (3.8 - 5.0) (W/mK)* *Numbers in brackets represent the range of values obtained; 4 —5 samples for each composition

Comparative Example 3 and Examples 9 - 11

Component A-I was first mixed with Components D-2 and D-3 on a speedmixer at

2000 rpm for 20 seconds, and at 2749 rpm for 2 X 20 seconds. The mixture was then divided into 4 roughly equal portions. To each portion, a predetermined amount of Component C was added, and the resultant mixture was blended with a speedmixer (FlackTek Inc., Model # DAC400FV) until a visibly homogeneous mixture was obtained. Components E and F-I were then added to the mixture, and the resultant mixture was first mixed with a spatula, and then mixed on the speedmixer (FlackTek Inc., Model # DAC400FV) at 2000 - 2745 rpm for 10 - 60 seconds. Components B-I and H-I were then added, and the mixture was blended with the speedmixer at 2000 rpm for 10 seconds. The product mixtures were then degassed at room temperature for 15 - 24 hours (house vacuum). The compositions of the product mixtures are listed in Table 7.

The individual product mixture was then interposed between two metal coupons under 10 psi pressure for 1 second and cured under a pressure of 10 psi at 150°C over 2 hours. The diffusivity of the resulting sandwiched structure was then determined by laser flash (Netzsch Instrument, Microflash™ 300), and the in-situ thermal resistivity determined by a software macro provided with the Microflash™ instrument. No external pressure was exerted during the measurement for adhesives. 5 samples were prepared utilizing the formulation between the AlCr and Si coupons.

As seen from Table 7, addition of liquid metal initially resulted in improved performance (Examples 9-11 vs. Comparative Example 3), but as more liquid metal was added, the in-situ thermal resistivity started to increase (Example 11). i aoie i

Examples Comparative Example 9 Example 10 Example 11 Ex. 3 A-I 100 100 100 100 B-I 2 2 2 2 C 0 225 450 900 D-2 120 120 120 120 D-3 328 328 328 328 E 0.6 0.60 0.6 0.6 F-I 1 1 1 1 H-I 5 5 5 5 (A+B+E+F+H): C: 1: 0: 4.1 1: 2.1 : 4.1 1 : 4.1: 4.1 1: 8.2: 4.1 D (weight) C: (D-2 + D-3) 0: 1.4 1 : 4.5 1 : 2.2 1: 1.1 (volume) Sandwich Material Al-TIM-Si Al-TIM-Si Al-TIM-Si Al-TIM-Si Assembly/Cure 10/10 pressure (psi) 10/10 10/10 10/10 Bondline Thickness (range, 0.7 ± 0.3 1.0 ± 0.2 1.3 ± 0.2 2.4 ± 0.2 (0.8 - 1.3) (1.1 - 1.5) mils)* (0.5 - 1.2) (2.1 - 2.6) In-situ Thermal 17 ± 2 Resistivity @25°C 20 ± 4 14 ± 2 16 ± 2 (15 - 24) (16 - 20) (mm2-K/W) * (12 - 17) (13 - 18) In-situ Thermal Conductivity 0.9 ± 0.2 1.5 ± 0.2 2.4 ± 0.1 3.8 ± 0.2 (0.8 - 1.3) (1.3 -1.7) (2.2 - 2.5) (3.6 - 4.1) @25°C (W/mK)* *Numbers in brackets represent the range of values obtained; 5 samples for each composition

Comparative Examples 4 — 5 and Examples 12 - 13

Component A-I was first mixed with Component D-4 on a speedmixer at 2749 rpm

for 60 seconds. The mixture was then divided into smaller portions. To each

portion, a predetermined amount of Component C was added, and the resultant

composition was blended with a speedmixer (FlackTek Inc., Model # DAC400FV)

until a visibly homogeneous mixture was obtained. Components B-I, E, F-I and H-I

were then added to the mixture, and the composition was first mixed with a spatula,

and then thoroughly mixed on the speedmixer (FlackTek Inc., Model # DAC400FV)

at 1000 rpm for 5 seconds. The mixture was then degassed at room temperature for 15 - 24 hours (house vacuum). The compositions of the mixtures are listed in Table 8.

The mixture was then interposed between two metal coupons with a prescribed pressure for 1 second and cured under a pressure of 20 psi at 1500C over 2 hours. The diffusivity of the resulting sandwiched structure was then determined by laser flash (Netzsch Instrument, Microflash™ 300), and the in-situ thermal resistivity determined by a software macro provided with the Microflash instrument. No external pressure was exerted during the measurement for adhesives. 5 samples were prepared utilizing the formulation between the AlCr and Si coupons.

Comparative Example 5 has a similar weight percentage loading of filler relative to Example 12 and similar volume percentage loading of filler relative to Example 13. However, because of the extra filler was added as a solid, the resulting viscosity was much higher than in either of Examples 12 or 13. As a result, its in-situ performance is much worse than either Example 12 or Example 13. Thus, addition of liquid metal alloy showed improvement in performance over not only compositions containing comparable amounts of solid fillers but no liquid metal alloys (e.g., Examples 12 and 13 vs. Comparative Example 4), but also compositions in which the amounts of solid fillers (weight or volume) are similar to the total amounts of solid fillers plus liquid metal alloys (e.g., Examples 12 and 13 versus Comparative Example 5). i aoie δ

Exam pies Comparative Example 12 Example 13 Comparative Ex. 4 Ex. 5 A-I 100 100 100 100 B-I 2 2 2 2 C 0 77 230 0 D-4 76 76 76 150 E 0.6 0.6 0.6 0.6 F-I 1.0 1.0 1.0 1.0 H-I 5.2 5.3 5.2 5.2 (A+B+E+F+H): C: 1: 0: 0.7 1: 0.7: 0.7 1 : 2.1: 0.7 1 : 0: 1.4 D (weight) (total (41%) (58%) (74 %) (58%) wt% filler) C: D(volume) (total 0 : 0.3 1 :3 1 :1 0 : 0.6 filler vol%) (24%) (29.5%) (38.7%) . (38.5 %) Non-flowable, Appearance Flowable Flowable Flowable very thick Dielectric constant - - 3.52, 3.46 2.30, 2.01 (@60Hz, @5MHz) Dissipation Factor - 0.0049**, 0.0858, - (@60Hz, @5MHz) 0.0024 0.0022 Sandwich Material Al-TIM-Si Al-TIM-Si Al-TIM-Si Al-TIM-Si Assembly/Cure 10/20 10/20 10/20 10/20 pressure (psi) Bondline Thickness 0.4 ± 0.1 0.5 ± 0.1 0.6 ± 0.1 2.1 ± 0.4 (range, mils)* (0.3 - 0.6) (0.4 - 0.6) (0.5 - 0.6) (1.6 - 2.7) In-situ Thermal 19 ± 2 15 ± 4 9 ± 1 57 ± 9 Resistivity @25°C (17 - 21) (12 - 22) (8 - 10) (48 - 69) (mm2-K/W) * In-situ Thermal 0.6 ± 0.1 0.8 ± 0.2 0.9 ± 0.1 Conductivity @25°C l.ό ± O.l (0.5 - 0.8) (0.7 -1.1) (1.5 - 1.8) (0.8 - 1.0) (W/mK)* *Numbers in brackets represent the range of values obtained; 5 samples for each composition ** Lower dissipation factor was observed at 60 Hz relative to Comparative Example 6.

A composition identical (within experimental error) to Example 13 was prepared.

Twenty-five AlCr-TIM-Si sandwiched structures were prepared under an assembly

pressure of 10 psi. Sixteen of these sandwiched structures sandwiched structures

were cured under a 10 psi pressure, and the remaining nine were cured under a 20 psi

pressure. The sandwiched structures cured under 10 psi pressure had an average bond

line thickness of 0.6 ± 0.2 mils (range: 0.3 - 0.9 mils), an average in-situ thermal resistivity ot 11 ± I mπT-K/W (range: 8 - 16 πW-K/W) and an average in-situ

thermal conductivity of 1.4 ± 0.2 W/mK(range: 0.9 - 1.6 W/mK). The sandwiched

structures cured under 20 psi pressure had an average bond line thickness of 0.5 ± 0.1

mils (range: 0.3 - 0.7 mils), an average in-situ thermal resistivity of 9 ± 2 mm2-K/W (

range: 7 - 12 mm2-K/W) and an average in-situ thermal conductivity of 1.5 ± 0.3

W/mK (range: 1.0 - 1.9 W/mK). Twelve of the twenty-five sandwiched structure

samples were subjected to 280 hours of reliability testing at 85°C and 85% relative

humidity. The thermal resistances of the samples before and after the test were

summarized in the table below (Table 9). No performance degradation was observed.

Table 9 Reliability Testing of The Composition of Example 13 at 85C/85% Relative

Humidity

Curing Pressure Time Zero Post 280 hours 85C/85% Relative Humidity test Bondline Thickness 10 psi 0.6 ± 0.1 0.6 ± 0.1 (range, mils)* (0.3 - 0.7) (0.3 - 0.8) In-situ Thermal 11 ± 2 10 ± 2 Resistivity @25°C (9 - 14) (8 - 14) (mm2-K/W) * In-situ Thermal 1.4 ± 0.3 1.5 ± 0.3 Conductivity (0.9 - 1.6) (1.1 -1.9) @25°C (W/mK)* Bondline Thickness 20 psi 0.5 ± 0.2 0.6 ± 0.2 (range, mils)* (0.3 - 0.7) (0.3 - 0.8) In-situ Thermal 9 ± 2 9 ± 3 Resistivity @25°C (7 - 12) (6 - 12) (mm2-K/W) * In-situ Thermal 1.3 ± 0.2 1.6 ± 0.2 Conductivity (1.0 - 1.5) (1.4 -1.8) @25°C (W/mK)*

A composition identical (within experimental error) to Example 13, was prepared, and

3-layer structures were prepared for air-to-air thermal shock testing at between —50

and 150°C (10 minute dwell time at each temperature extreme). Slight degradation in

performance was observed, and the results are summarized in the table below (Table

10). Table IU Reliability Testing of The Composition of Example 12 Under Thermal

Shock Conditions

Curing Pressure Time Zero Post 500 cycles of thermal shock Bondline Thickness lO psi 0.4 ± 0.2 0.4 ± 0.2 (range, mils)* (0.2 - 0.7) (0.2 - 0.7) In-situ Thermal 10 ± 3 12 ± 3 Resistivity @25°C (8 - 16) (9 - 17) (mm2-K/W) * In-situ Thermal 1.0 ± 0.3 0.8 ± 0.3 Conductivity (0.5 - 1.3) (0.4 -1.1) @25°C (W/mK)* Bondline Thickness 20 psi 0.5 ± 0.2 0.5 ± 0.2 (range, mils)* (0.2 - 0.7) (0.2 - 0.7) In-situ Thermal 10 ± 2 Resistivity @25°C 11 ± 2 (7 - 12) (9 - 13) (mm2-K/W) * In-situ Thermal 1.2 ± 0.3 1.2 ± 0.3 Conductivity (0.8 - 1.5) (0.6 -1.4) @25°C (W/mK)*

Comparative Example 6 and Examples 14 - 16

Component A-2 was first mixed with Component D-5 and D-6 on a speedmixer at

2749 rpm for 60 seconds. The mixture was then split into smaller portions. To each

portion, a predetermined amount of Component C was added, and the resultant

composition was blended with a speedmixer (FlackTek Inc., Model # DAC400FV) till

a visibly homogeneous mixture was obtained. Components B-2, E and H-I were then

added to the mixture, and the composition was first mixed with a spatula, and then

thoroughly mixed on the speedmixer (FlackTek Inc., Model # DAC400FV) at 1000

rpm for 5 seconds. The mixture was then degassed at room temperature for 15 — 24

hours (house vacuum). The compositions of the mixtures are listed in Table 11.

The mixture was then interposed between two metal coupons with a prescribed

pressure for 1 second and cured under pressure (typically 10 or 20 psi) at 1500C over

2 hours. The diffusivity of the resulting sandwiched structure was then determined by

laser flash (Netzsch Instrument, Microflash™ 300), and the in-situ thermal resistivity

determined by a software macro provided with the Microflash™ instrument. No external pressure was exerted during the measurement for adhesives. 4 - 5 samples

were prepared utilizing the formulation between the AlCr and Si coupons.

These compositions were also tested as greases between aluminum coupons at 25°C.

The phrase "tested as greases" means that the compositions were tested without first

curing. The assembly pressure was 20 psi, and was applied for 1 second. The

samples were tested under ca 100 psi pressure. The laser flash instrument (Netzsch

Instrument) was used to determine the diffusivity, and the in-situ thermal resistances

and thermal conductivities were determined by a macro supplied by Netzsch

Instrument. A modified sample holder was used, so that torques could be applied to

four bolts located at the corners of the sample. Steel collars were used to more evenly

distribute the applied force through the sample. The correlation between torques and

pressures were determined by calibration. A representative design of the sample

holder can be found in the conference proceeding entitled "Utilization of Carbon

Fibers in Thermal Management of Microelectronics" (Zhong et al.) presented at

March 2005 's Advanced Packaging Material Symposium in Irvine, CA. The results

are listed in Table 11.

No significant reduction in thermal resistivity was observed when the compositions

were tested as adhesives, but when tested as greases, the thermal resistivity decreased

with increasing volume percentage of liquid metal. In addition, it was observed that

when D-5 was used as a component of solid filler in adhesive formulations, effects of

liquid metal incorporation on thermal resistances are much smaller than when other

solid fillers are used.

Table 11

Examples Comparative 14 15 16 Ex. 6 A-2 100 100 100 100 B-2 10 10 10 10 C 0 75.9 153 219 D-5 58.5 58.6 58.6 58.6 D-6 14.6 14.7 14.7 14.7 E 1.1 1.1 1.1 1.1 H-I 5.2 5.2 5.1 5.1 (A+B+E+F+H): 1: 0: 0.63 1: 0.65: 0.63 1: 1.3: 0.63 1: 1.9: 0.63 C: D (weight) C: D(volume) 0: 0.28 1 : 2.8 1 : 1.4 1 : 1 Tested as adhesives Sandwich Al-TIM-Si Al-TIM-Si Al-TIM-Si Al-TIM-Si Material Assembly/Cure 10/20 pressure (psi) 10/20 10/20 10/20 Bondline 3.0 ± 0.2 Thickness 2.1 ± 0.1 2.3 ± 0.3 2.3 ± 0.2 (2.8 - 3.2) (range, mils)* (2.1 - 2.3) (2.0 - 2.7) (2.1 - 2.5) In-situ Thermal Resistivity 41 ± 6 46 ± 7 40 ± 6 36 ± 6 @25°C (mm2- (35 - 49) (37 - 53) (36 - 49) (31 - 47) KAV) * In-situ Thermal Conductivity 1.9 ± 0.2 1.2 ± 0.2 1.5 ± 0.2 1.6 ± 0.3 @25°C (1.6 - 2.1) (1.0 -1.4) (1.2 - 1.7) (1.2 - 2.1) (W/mK)* Tested as greases Sandwich Al-TIM-Al Al-TIM-Al Al-TIM-Al Al-TIM-Al Material Test pressure 100 100 (psi) 100 100 Bondline 1.9 ± 0.2 Thickness 2.1 ± 0.2 2.0 ± 0.1 1.6 ± 0.1 (range, mils)* (1.7 - 2.2) (1.7 - 2.2) (1.9 - 2.1) (1.4 - 1.8) In-situ Thermal Resistivity 20 ± 2 16 ± 3 14 ± 2 l l ± l @25°C (mm2- (17 - 22) (11 - 21) (12 - 17) (9 - 12) K/W) * In-situ Thermal Conductivity 2.4 ± 0.3 3.4 ± 0.4 3.6 ± 0.4 3.6 ± 0.2 @25°C (2.1 - 2.8) (2.8 -3.8) (3.1 - 4.2) (3.4 - 3.8) (W/mK)* *Numbers in brackets represent the range of values obtained; 5 samples for each composition

Comparative Example 7 and Examples 17 - 19

A portion of component A-I (ca 67% of the total amounts) was first mixed with

Component D-5 on a speedmixer at 2749 rpm for 60 seconds. The mixture was then

split into smaller portions. To each portion, the remainder of A-I and a predetermined amount oi component C were then added, and the resultant composition was blended with a speedmixer (FlackTek Inc., Model # DAC400FV) until a visibly homogeneous mixture was obtained. Components B-I, E, F and H-I were then added to the mixture, and the composition was first mixed with a spatula, and then thoroughly mixed on the speedmixer (FlackTek Inc., Model # DAC400FV) at 1000 rpm for 5 seconds. The mixture was then degassed at room temperature for 15 - 24 hours (house vacuum). The compositions of the mixtures are listed in Table 12.

The mixture was then interposed between two metal coupons with a 10 psi pressure for 1 second and cured under a pressure of 20 psi at 150°C over 2 hours. The diffusivity of the resulting sandwiched structure was then determined by laser flash (Netzsch Instrument, Microflash™ 300), and the in-situ thermal resistivity determined by a software macro provided with the Microflash™ instrument. No external pressure was exerted during the measurement for adhesives. 4 - 5 samples were prepared utilizing the formulation between the AlCr and Si coupons.

Compositions of Comparative Example 7 and Example 19 were also tested as greases between aluminum coupons at 250C, that is, the compositions were tested without curing first. The assembly pressure was 20 psi, and was applied for 1 second. The samples were tested under c.a. 100 psi pressure. The test set up was similar to that described in Examples 14 - 16. The results are listed in Table 12.

No significant reduction in thermal resistances was observed when the compositions were tested as adhesives. Significant reduction in thermal resistance was observed when the liquid-metal-containing composition was tested as a grease (Example 19 versus Comparative Example 7). Ln addition, it was observed that when D-5 was used as a component of solid filler in adhesive formulations, the effects of liquid metal incorporation on thermal resistances are much smaller than when other solid fillers are used. I SLOiQ U.

Examples Comparative 17 18 19 Ex. 7 A-I 100 100 100 100 B-I 2 2 2 2 C 0 77 152 227 D-5 76 76 76 76 E 0.6 0.6 0.6 0.6 F 1 1 1 1 H-I 5 5 5 5 (A+B+E+F+H): C: 1: 0: 0.7 1 : 0.7: 0.7 1 : 1.4: 0.7 1: 2.1: 0.7 D (weight) C: D(volume) 0: 0.3 1 :3 1 : 1.5 1 :1 Tested as adhesives Sandwich Material Al-TIM-Si Al-TIM-Si Al-TIM-Si Al-TIM-Si Assembly/Cure 10/20 10/20 pressure (psi) 10/20 10/20 Bondline 2.8 ± 0.2 2.4 ± 0.1 2.4 ± 0.1 2.8 ± 0.3 Thickness (range, (2.5 - 3.0) (2.3 - 2.6) (2.3 - 2.5) (2.5 - 3.2) mils)* In-situ Thermal 34 ± 3 35 ± 5 Resistivity @25°C 30 ± 3 27 ± 2 (29 - 38) (31 - 43) (27 - 34) (25 - 30) (mm2-K/W) * In-situ Thermal 2.1 ± 0.1 1.8 ± 0.3 2.0 ± 0.2 2.6 ± 0.4 Conductivity (1.9 - 2.2) (1.3 - 2.0) (1.8 - 2.2) (2.2 - 3.2) @25°C (W/mK)* Tested as greases Sandwich Material Al-TIM-Al - - Al-TIM-Al Test pressure (psi) 100 - - 100 Bondline 2.2 ± 0.2 2.2 ± 0.4 Thickness (range, - - (1.9 - 2.4) (1.8 - 2.8) mils)* In-situ Thermal lO ± l 6.5 ± 0.4 Resistivity @25°C - - (9 - 11) (6 - 7) (mm2-K/W) * In-situ Thermal Conductivity 5.5 ± 0.7 - • 8.7 ± 1.5 @25°C (W/mK)* (4.8 - 6.3) (7.1 - 11) *Numbers in brackets represent the range of values obtained; 5 samples for each composition

Comparative Example 8 and Example 20 component A-I was πrst mixeα wim component D-7 on a speedmixer, and degassed overnight. The mixture was then split into smaller portions. To each portion, a predetermined amount of Component C was added, and the resultant composition was blended with a speedmixer (FlackTek Inc., Model # DAC400FV) until a visibly homogeneous mixture was obtained. Components B-I, E, F-I and H-2 were then added to the mixture, and the composition was first mixed with a spatula, and then thoroughly mixed on the speedmixer (FlackTek Inc., Model # DAC400FV) at 1000 rpm for 5 seconds. The mixture was then degassed at room temperature for 15 — 24 hours (house vacuum). The compositions prepared are listed in Table 13.

The mixture was then interposed between two metal coupons with a 10 psi pressure for 1 second and cured under a pressure of 10 psi at 150°C over 2 hours. The diffusivity of the resulting sandwiched structure was then determined by laser flash (Netzsch Instrument, Microflash™ 300), and the in-situ thermal resistivity determined by a software macro provided with the Microflash™ instrument. No external pressure was exerted during the measurement for adhesives. 4 - 5 samples were prepared utilizing the formulation between the AlCr and Si coupons.

Table 13

Examples Comparative Ex. 8 20 A-I 100 100 B-I 2 2 C 0 175 D-7 58 58 E 0.6 0.6 F-I 1 1 H-2 5 5 (A+B+E+F+H): C: D (weight) 1 : 0: 0.535 1 : 1.61 : 0.535 C: D(volume) 0: 0.24 1 : 1 Tested as adhesives Sandwich Material Al-TIM-Si Al-TIM-Si Assembly/Cure pressure (psi) 10/10 10/10 Bondline Thickness (range, 1.8 ± 0.2 1.7 ± 0.1 mils)* (1.5 - 2.1) (1.6 - 1.9) In-situ Thermal Resistivity 32 ± 3 24 ± 1 @25°C (mm2-K/W) * (27 - 34) (23 - 26) In-situ Thermal Conductivity 1.4 ± 0.3 1.8 ± 0.2 @25°C (W/mK)* (1.1 - 2.0) (1.6 - 2.0) *Numbers in brackets represent the range of values obtained; 5 samples for each composition

Example 21

A portion of component A-I (4.9933 g) was first mixed with Component C (10.4031 g) with a speedmixer at 1200 rpm for 20 seconds. Component D-5 (3.49 g) was then added to the mixture, and blended with the high-speed mixer until a homogeneous mixture was obtained. To this, additional A-I was added (~ 1.89g total) and blended on a high-speed mixer for 20 seconds at 1200 rpm. The formulation was observed to flow under shear. The formulation was transferred to a syringe, and degassed over night at room temperature. The grease composition was tested between aluminum coupons using procedures outlined herein. The test set up was similar to that described in Examples 14 — 16. The composition and thermal performance data are summarized in Table 14.

Comparative Example 9

Component A-I was mixed with Component D-5 in a 100:51 A-I : D-5 ratio by weight. The mixture was blended on a high-speed mixer until a homogeneous mixture was obtained (2000 rpm, 15 seconds for 3 - 4 times). The formulation was transferred to a syringe, and degassed overnight at room temperature. The grease composition was tested between aluminum coupons. The test set up was similar to that described in Examples 14 - 16. The composition and thermal performance data are summarized in Table 14.

Comparative Example 10 In preparing the composition of Comparative Example 10 component A-I was mixed with Component D-5 on a high-speed mixer in 1:1 A-I : D-5 ratio by weight. This formulation contained approximately the same volume percentage of fillers as in Example 21. Whereas the composition in Example 21 could flow under shear, the composition of Comparative Example 10 was powdery, with the resins barely holding the fillers together. The composition of Comparative Example 10 was evaluated as a grease composition and was tested between aluminum coupons. The test set up was similar to that described in Examples 14 - 16. The composition and thermal performance results are summarized in Table 14.

The composition of Example 21 not only had lower thermal resistivity than that of Comparative Example 9, which had similar solid filler loadings (thus demonstrating the benefits of liquid metal incorporation); but also had lower thermal resistivity than that of Comparative Example 10, in which the total filler loading was the same as that in Example 21. The lower viscosity of liquid-metal containing formulations probably gave better wetting and flowability than ones containing only solid fillers at comparable volume loading. The better wetting and flow characteristics are believed to translate into better thermal performance.

Table 14

Examples Comparative 21 Comparative Ex.10 Ex. 9 A-I 100 100 100 C 0 150 0 D-5 51 51 100 A: C: D (by weight) 1 :0: 0.5 1 : 1.5: 0.5 1 : 0: 1 C: D (by volume) 0: 0.22 1 :1 0: 0.44 (total filler vol%) (18.5%) (31%) (31%) Flowable Flowable Powdery, thick bread- Appearance under shear under shear dough like Sandwich Material Al-TIM-Al Al-TIM-Al Al-TIM-Al Test Pressure (psi) 100 100 100 Bondline Thickness 2.1 ± 0.1 1.5 ± 0.1 3.7 ± 0.2 (range, mils)* (2.0 - 2.2) (1.4 - 1.6) (3.5 - 3.8) In-situ Thermal 7 ± 1 5 ± 1 lO ± l Resistivity @25°C (W-K/w) * (6.0 - 7.8) (3.6 - 6.3) (9 -11) In-situ Thermal 8 ± 1 8 ± 2 9 ± 0.3 Conductivity @25°C (7.2 - 8.7) (6.0 - 10.3) (8.7 - 9.5) (W/mK)* ^Numbers in brackets represent the range of values obtained; 5 samples for each composition

Example 22

5.0005 g of component A-3 was first mixed with 10.4483 g of Component C with a speedmixer at 1200 rpm for 20 seconds. Component D-5 (3.49 g) was then added to the mixture, and blended with the high-speed mixer until a homogeneous mixture was obtained. Because of the high viscosity, additional A-3 (~ 1.89 g total) was added to the mixture, and blended on a high-speed mixer for 20 seconds at 1200 rpm. The formulation was transferred to a syringe, and degassed overnight at room temperature. The composition of Example 22 was evaluated as a grease composition and was tested between aluminum coupons. The test set up was similar to that described in Examples 14 - 16. Compositional and thermal performance data are gathered in Table 15.

Comparative Example 11

Component A-3 was mixed with Component D-5 (100: 51 A-3 : D-5 by weight) on a high-speed mixer until a homogeneous mixture was obtained (2000 rpm, 15 seconds for 3 - 4 times). The formulation was transferred to a syringe, and degassed overnight at room temperature. The grease composition was tested between aluminum coupons. The test set up was similar to that described in Examples 14 - 16. Compositional and thermal performance data are gathered in Table 15.

Table 15 Examples Comparative Ex. 11 22 A-3 100 100 C 0 150 D-5 51 51 A: C: D (by weight) 1 :0: 0.5 1 : 1.5: 0.5 C: D (by volume) 0: 0.285 1 :1 Sandwich Material Al-TIM-Al Al-TIM-Al Test Pressure (psi) 100 100 Bondline Thickness (range, 2.0 ± 0.1 1.5 ± 0.01 mils)* (1.8 - 2.1) (1.51 - 1.54) In-situ Thermal Resistivity 7 ± 1 5 ± 0.5 @25°C (mm2-K/W) * (5.8 - 8.6) (4.2 - 5.5) In-situ Thermal Conductivity 7 ± 1 8 ± 1 @25°C (W/mK)* (6.1 - 8.2) (7.1 - 9.3) *Numbers in brackets represent the range of values obtained; 5 samples for each composition

Example 22 demonstrates that the benefits of incorporating liquid metal alloys not

only apply in traditional resin systems such as silicones, epoxies, acrylates and the

like, but also in non-traditional carrier such as ionic liquids.

Comparative Example 12 and Example 23

Component A-I was mixed with Component D-3 in a plastic container on a high¬

speed mixer at 2000 rpm for 15 seconds to make up the composition for Comparative

Example 12. A portion of this mixture was then transferred to a second container, to

which Component C was added. The resulting mixture was first blended by hand

with a spatula, and then with a high-speed mixer at 2750 rpm for 5 seconds. This

resulted in the composition of Example 23. The formulations were transferred to

syringes, and were degassed overnight at room temperature. The grease composition

was tested between aluminum coupons. The test set up was similar to that described

in Examples 14 - 16. Compositional and thermal performance data are gathered in

Table 16. Table 16

Examples Comparative 12 23 A-I 100 100 C 0 800 D-3 400 400 A: C: D (by weight) 1 :0: 4 1 : 8: 4 C: D (by volume) 0: 1.5 1: 1.2 Sandwich Material Al-TIM-Al Al-TIM-Al Test Pressure (psi) 100 100 Bondline Thickness (range, 2.0 ± 0.2 1.6 ± 0.6 mils)* (1.8 - 2.2) (1.1 - 2.6) In-situ Thermal Resistivity 15 ± 1.3 7.5 ± 3.0 @25°C (mm2-K/W) * (13 - 16) (5.6 - 12) In-situ Thermal Conductivity 3.5 ± 0.1 5.5 ± 0.6 @25°C (W/mK)* (3.4 - 3.6) (4.8 - 5.9) ^Numbers in brackets represent the range of values obtained; 5 samples for Comparative Example 12, and 4 samples for Example 23.

Comparative Example 13 and Example 24

A stock solution, I, consisting of components A-I and B-3 was prepared (A-I : B-3 is

approximately 48.29 : 1 by weight). A portion of the stock solution, I, was then

combined with component D-5 on a high-speed mixer until a visually homogeneous

mixture was obtained. The mixture was degassed overnight in an oven connected to

house vacuum at 75°C. The mixture was then split into two portions, to one of which

Component C was added. The resulting mixture was then mixed on a high-speed

mixer till a visually homogeneous mixture was obtained.

A stock solution, II, consisting of Components B-4, H-2 and F-2 was prepared (B-4 :

H-2 : F-2 is approximately 8.86: 4.03 : 1 by weight). An appropriate amount of stock

solution II was added to each of the two portions mentioned above. Afterwards, the

mixtures were blended on a high-speed mixer for 5 seconds at 750 rpm. The

formulations were transferred to syringes, and degassed over the weekend at room

temperature. The resulting gel compositions were cured between an aluminum

coupon and a silicon coupon under a 10 psi pressure at 150°C for 2 hours. The

resulting sandwiched structures were tested under 30 psi pressure. The test set up was

similar to that described in Examples 14 - 16. Compositional information and the mermai pertormance data for Comparative Example 13 and Example 24 are gathered

in Table 17.

Table 17 Examples Comparative 13 24 A-I 100 100 B-3 2.1 2.1 B-4 11.5 11.4 C 0 204 D-5 68 68 F-2 1.3 1.3 H-2 5.2 5.2 (A+B+F+H): C: D (by weight) 1 : 0: 0.567 1 : 1.70: 0.567 C: D (by volume) . 0: 0.26 1 : 1 Sandwich Material Al-TIM-Si Al-TIM-Si Test Pressure (psi) 30 30 Bondline Thickness (range, 2.0 ± 0.3 1.9 ± 0.1 mils)* (1.5 - 2.2) (1.8 - 2.0) In-situ Thermal Resistivity 8.6 ± 2.0 6.2 ± 0.8 @25°C (mm2-K/W) * (6.2 - 11) (5.1 - 6.8) In-situ Thermal Conductivity 5.9 ± 0.6 8.0 ± 1.2 @25°C (WZmK)* (5.1 - 6.5) (7.0 - 9.7) *Numbers in brackets represent the range of values obtained; 4 samples for each composition

Comparative Example 14 and Example 25

Two gel compositions were prepared analogously to Comparative Example 13 and

Example 24, except that the filler D-7 was used. The formulations were transferred to

a syringe, and degassed over the weekend at room temperature. The resulting gel

compositions were cured between an aluminum coupon and a silicon coupon under a

10 psi pressure at 1500C for 2 hours. The resulting sandwiched structures were tested

under 30 psi pressure. The compositions and the thermal performance data of

Comparative Example 14 and Example 25 are listed in Table 18. The test set up is

similar to that described in Examples 14 — 16. In this case, no reduction in thermal

resistance was observed, indicating that incorporation of liquid metal alloys does not

necessarily lead to improved in-situ performance. Table 18 Examples Comparative Ex. 14 25 A-I 100 100 B-3 2.1 2.1 B-4 11.5 11.4 C 0 204 D-7 68 68 F-2 1.3 1.3 H-2 5.2 5.2 (A+B+F+H): C: D (by weight) 1: 0: 0.567 L 1: 1.70: 0.567 C: D (by volume) 0 : 0.26 1 : 1 Sandwich Material Al-TIM-Si Al-TIM-Si Test Pressure (psi) 30 30 Bondline Thickness (range, 1.2 ± 0.1 1.1 ± 0.2 mils)* (1.1 - 1.3) (0.9 - 1.3) In-situ Thermal Resistivity 7.4 ± 1.2 7.4 ± 2.0 @25°C (mm2-K/W) * (6.3 - 9.0) (5.1 - 9.4) In-situ Thermal Conductivity 4.3 ± 0.8 3.8 ± 1.0 @25°C (W/mK)* (3.6 - 5.1) (2.9 - 5.1) *Numbers in brackets represent the range of values obtained; 4 samples for each composition

Comparative Example 15 and Example 26

Two gel compositions were prepared analogously to Comparative Example 13 and

Example 24, except that the filler D-8 was used. The formulations were transferred to

syringes, and degassed over the weekend at room temperature. The resulting gel

compositions were cured between an aluminum coupon and a silicon coupon under a

10 psi pressure at 1500C for 2 hours. The resulting sandwiched structures were tested

under 30 psi pressure. The test set up was similar to that described in Examples 14 -

16. The compositions and the thermal performance data for Comparative Example 15

and Example 26 are listed in Table 19. Table 19 Examples Comparative 15 26 A-I 100 100 B-3 2.1 2.1 B-4 11.4 11.5 C 0 204 D-8 68 68 F-2 1.3 1.3 H-2 5.2 5.2 (A+B+F+H): C: D (by weight) 1: 0: 0.567 1: 1.70: 0.567 C: D (by volume) 0: 0.26 1 : 1 Sandwich Material Al-TIM-Si Al-TIM-Si Test Pressure (psi) 30 30 Bondline Thickness (range, 0.7 ± 0.05 0.7 ± 0.04 mils)* (0.6 - 0.7) (0.6 - 0.7) In-situ Thermal Resistivity 6.9 ± 0.4 4.2 ± 0.4 @25°C (mm2-K/W) * (6.5 - 7.3) (3.9 - 4.6) In-situ Thermal Conductivity 2.5 ± 0.1 4.1 ± 0.2 @25°C (W/mK)* (2.5 - 2.6) (3.9 - 4.2) ^Numbers in brackets represent the range of values obtained; 3 samples for each composition.

Examples 27 and 28

D-5 was mixed with A-4 in a plastic container on a high speed mixer at 2750 rpm for

20 seconds and the resultant mixture was added C in pre-determined amounts (See

Table 20 for amounts of Component C employed). The resulting mixture was first

blended by hand with a spatula, and then with a high-speed mixer at 2750 rpm for 20

seconds. These grease formulations were tested between aluminum coupons. The

test set up was similar to that described in Examples 14 - 16. The compositions

tested and test results are summarized in Table 20. The differences between in situ

thermal resistivities and thermal conductivities between formulations containing

different levels of liquid metal were observed to be much smaller at testing pressure

of 30 psi than at 100 psi. l aoie /υ

Examples 27 28 A-4 100 100 C 323 179 D-5 79 75 A: C: D (by weight) 1: 3.23 :0.79 1 : 1.79: 0.75 C: D (by volume) 1 : 0.7 1 : 1.2 Sandwich Material Al-TIM-Al Al-TIM-Al Test Pressure (psi) 100 100 Bondline Thickness (mils)* 2.5 ± 0.23 1.8 ± 0.15 In-situ Thermal Resistivity @25°C 3.6 ± 0.34 5.6 ± 0.8 (mm2-K/W) * In-situ Thermal Conductivity 17.7 ± 1.5 8.0 ± 1.5 @25°C (W/mK)* Test Pressure (psi) 30 30 Bondline Thickness (mils)* 2.02 1.85 In-situ Thermal Resistivity @25°C (W-K/W) * 5.32 6 In-situ Thermal Conductivity 9.75 7.85 @25°C (W/mK)* * 5 samples for each composition.

Example 29

The composition listed in Table 21 was prepared using techniques described herein.

The resulting grease formulation was tested between aluminum coupons. The test set

up was similar to that described in Examples 13 - 15. The results are summarized in

Table 21. 1 able 21 Example 28 A-4 100 C 112 D-2 64.5 D-5 49.5 A: C: D (by weight) 1 : 1.12:1.14 C: D-2: D-5 (by volume)/ C: D (by volume) 1 : 0.94 : 1.28/ l : 2.2 Sandwich Material Al-TIM-Al Test Pressure (psi) 100 Bondline Thickness (range, mils)* 3.2 ± 0.2 In-situ Thermal Resistivity @25°C (nW-K/W) * 6.5 ± 0.3 In-situ Thermal Conductivity @25°C (WAnK)* 12.5 ± 0.8 * 5 samples.

While typical embodiments have been set forth for the purpose of illustration, the

foregoing description should not be deemed to be a limitation on the scope of the

invention. Accordingly, various modifications, adaptations, and alternatives may

occur to one skilled in the art without departing from the spirit and scope of the

present invention.