WO2006035360A2

WO2006035360A2 - Method of forming a coating on a substrate, and a coating thus formed

Info

Publication number: WO2006035360A2
Application number: PCT/IB2005/053090
Authority: WO
Inventors: Danielle Beelen; Petra E. De Jongh
Original assignee: Koninklijke Philips Electronics N.V.
Priority date: 2004-09-30
Filing date: 2005-09-20
Publication date: 2006-04-06
Also published as: TW200626519A; WO2006035360A3; CN101031519A; JP2008514415A; EP1797016A2

Abstract

The present invention relates to a method of forming a protective coating on an electronic substrate such as an IC, the protective coating having a good mechanical, chemical and physical resistance and providing a suitable non-transparency. According to the present invention a porous matrix (preferably comprising Ti02) is formed, which is filled with a filler component which can absorb or scatter light, such as TiN particles. After a curing step, a reinforcing precursor component is added. In preparing the protective coating a predetermined amount of the filler component is used in order to obtain at least 40 vol.% of the filler component in the protective coating eventually obtained, based on the porous matrix obtained.

Description

Method of forming a coating on a substrate, and a coating thus formed

The invention relates to a method of forming a coating on a substrate, said method comprising the steps of forming a porous structure and at least partially filling the porous structure with a second solution.

The invention also relates to a composition for forming the porous structure. The invention further relates to a substrate provided with a coating as obtainable by the method according to the present invention.

Such a method and such a device are known from EP-A 560485. The known coating is a ceramic coating, for instance of a silica compound - e.g. a SiO₂-network -, that is filled with a silicone polymer. The silica compound was provided by placing t a hydrogen silsesquioxane resin in an oxygen plasma reactor and heating it to 250 ⁰C, while being treated with oxygen plasma for three hours. This process resulted in a coating with a thickness of 200 nm and a porosity of 25 to 30%, as indicated in the related patent application EP-A 775680 (column 4, line 48-51). A solution of polydimethylsiloxane

(PDMS) is subsequently infiltrated into the porous structure. The devices were placed under a vacuum of 84.66 kPa followed by atmospheric pressure for 3 cycles and the excess fluid was wiped off.

It is a disadvantage of the known method that the infiltration of the porous structure is a rather awkward and time-consuming process, with the cycles of low and atmospheric pressure. Such a process may lead to loss of yield. Table 1 in EP-A 560485 also leads to worries about the yield loss, in that the 250 hours test gave a 100% yield of functional devices only with one specific concentration of the second solution. Yield loss is particularly problematic in view of the intended use of the coating as a protective coating in electronic devices, and integrated circuits in particular. This protective coating is applied as a last step in the manufacturing process, and yield loss in this step implies a large loss of value. It is therefore the object of the present invention to provide an improved method which is simpler to carry out and hence yields better results.

This object is achieved in that the porous structure is obtained by providing a coating composition that comprises a matrix precursor component and a particulate filler component, applying the coating composition on the substrate, and curing the composition.

The method of the present invention provides a porous structure that is based on the particulate filler component. The matrix precursor is used both as an encapsulant and an adhesive, and not as the porous structure itself, as is the case in the prior art method. Consequently, the matrix precursor can be chosen more freely and more broadly. Its primary properties are: sufficient adhesion to the particulate filler, and sufficiently able to form a network.

The method of the invention further allows the porosity to be chosen , and thus also allows for relatively large pores. The desired porosity can be set by the average size and the size distribution of the particulate filler component. A high porosity is obtained at a large average size and a relatively narrow size distribution. Porosities of 40 to 90 vol.% are obtained. For protective coatings with a thickness of at least tenths of microns, and probably in the order of several microns, the average particle size can suitably be chosen in the range of 1 to 3 microns; if the coating is a functional layer, the average particle size may be chosen to be smaller. The improved porous structure obtained in the method of the invention allows simpler filling processes for this structure. Instead of the plasma process used in the prior art, a coating process is used in the method of the invention. Such a coating process is simpler and more direct, and less dependent on the control of pressure.

The resulting coating has at least two advantageous structural properties; firstly, it need not be completely filled, but may be left partially porous. Secondly, its top side is closed, so that further layers can be applied on top of the porous structure. Moreover, a suitable choice of the second solution enables the adhesion to the substrate, and to any layer on top of the porous structure, to be optimized.

In a suitable embodiment, the matrix precursor component comprises hydrolysable compounds bonded to a metal. Typical hydrolysable groups in the compound include, but are not limited to, alkoxy, such as methoxy, ethoxy, propoxy, butoxy and hexoxy, aceloxy, such as acetoxy, other organic groups bonded to the metal through an oxygen such as acetyl acetonate or amino groups. Such compounds are known per se from sol-gel processing and MOCVD processes. Good results have been obtained with the use of ethoxide, isopropoxide and n-butoxide compounds. Suitable metals include aluminum, titanium, zirconium, silicon, niobium, tantalum and the like. The number of hydrolysable groups is generally four, but components with less than four hydrolysable groups may be used, as long as the resulting network is sufficiently strong. The resulting network is for instance an oxide network or a carbide network.

It has been shown that advantageous results can be obtained if the matrix precursor component comprises a compound selected from the group consisting of tetraethyl orthotitanate (TEOTi), tetra-isopropyl titanate, niobium(V) ethoxide, tantalum(V) ethoxide, tantalum n-butoxide, zirconium n-butoxide or a mixture thereof. The resulting porous structure will then comprise a compound selected from the group consisting of TiO₂, Nb₂O₅, Ta₂O₅, ZrO₂ or a mixture thereof, and furthermore the particulate filler component. It is even more preferred that said compound is TiO₂, and that the filler component predominantly comprises TiN.

The particulate filler component preferably comprises crystalline particles, and they may be of any shape. Amorphous particles could be used additionally or alternatively. Examples include aluminum oxides, zirconium oxides, silicon oxide, titanium oxide, titanium nitride, titanium carbide, zinc oxide, silicon carbide. The particles need not be comprised of one material only. Good results have been obtained with TiN and TiO₂ particles, and with a mixture thereof. These results were optimized for thick, non-transparent coatings, and other materials may be applied as well. Materials that can be chemically or physically bonded to the matrix precursor component are preferred. In the present example, bonding is achieved, at least partially, by means of hydrogen bonding.

Most suitably, a predetermined amount of the filler component is used in order to obtain a porous structure with at least 40 vol.% of filler component. This amount of filler component, using particles with a diameter in the range of 1 to 3 micrometers, leads to a suitable porous structure. However, the relative volume of filler component appears to be dependent on the average size of the filler component, on the specific materials used, and on the viscosity of the matrix precursor component. The vol.% of filler component is preferably even higher than 40 vol.%, more preferably at least 60 vol.% and most preferably, for certain applications, even 80 vol.%. This high amount of filler component leads to coatings of a suitable thickness. Coatings having a thicknesses of more than 2 microns are suitable for use as security coatings, and preferably such coatings have a thickness in the range between 2 and 10 microns. The application of the coating composition on the substrate may be performed in various ways, as long as a suitable coating layer on the substrate is formed. Usually the coating composition is applied by dip coating, spin coating or spray coating, which techniques are well known in the art . The application of the second solution may also be performed using known techniques, including the above-mentioned dip coating, spin coating or spray coating. The curing of the coating composition, and optionally that of the second solution, is usually performed using UV- light or an increased temperature (preferably between 100 - 450°C), although other suitable methods may also be used.

In a further embodiment, the second solution comprises a component with hydrolysable groups bonded to a metal that is converted into a network. The result is an inorganic coating with a specific microstructure. Suitable second solutions comprise for instance a compound selected from the group consisting of tetraethyl orthotitanate (TEOTi), tetra-isopropyl titanate, niobium(V) ethoxide, tantalum(V) ethoxide, tantalum n-butoxide, zirconium n-butoxide or a mixture thereof. Such type of coatings are very suitable as protective 'security' coatings for integrated circuits and may be used to inhibit access to the integrated circuit by unauthorized persons. For this reason, the coatings must offer chemical, optical and physical protection. The optical protection is achieved with the filler component, which is able to absorb, scatter or reflect visible light. Particle mixtures of TiN and TiO₂ furthermore allow protection against UV, IR and electron radiation to a sufficient extent. The chemical protection is optimized through the choice of the matrix material and, in this case, the second solution. Use can be made of a monoaluminum phosphate matrix material, or a matrix material based on tetra-ethoxy-orthosilicate (TEOS). A protective non-porous coating with a monoaluminum phosphate matrix material and a filler component is known per se from US-A 6,198,155. Excellent results have been obtained with Ti-based matrix components. The hardness of the coating allows physical protection. Additionally, sensors can be provided below and/or on top of the coating, which enable the impedance to be measured. This allows the generation of an unpredictable and physically embodied identification code, and is known per se from WO-A 2003/046986. However, the coating of the invention is not limited to such a security coating.

In an alternative embodiment, the coating is used as a protective coating within a package of a semiconductor compound. Generally, molding compounds are used as protective coatings. Well-known examples are epoxies that are filled with glass or glass-like particles. However, there is a need for protective coatings that can withstand the relatively high temperatures needed for lead-free soldering, and still have an adequate coefficient of thermal expansion in comparison with the packaged component.

The present method is very suitable to provide such a protective coating. It is then preferred that an organic polymer, particularly a moulding compound, is chosen as the second solution that fills the porous structure at least partially. Examples include epoxies, polyimides, polystyrenes, polyterephtalates and other materials known in the art for transfer and injection moulding processes. This moulding compound may, but need not, be filled with particles. If it is filled with particles, these particles suitably have a smaller average diameter than the filler component that is part of the porous structure. The average particle sizes are suitably less than 0.5 microns, and most preferably less than 0.3 microns, if the filler component has an average particle diameter between 1 and 2 microns. A moulding compound is a suitable choice, as the resulting coating will then be chemically similar (although structurally different) to the presently used protective coatings.

Using the coating as a protective coating is advantageous in view of the proper chemical resistivity of the matrix material and the second solution. A further advantage is presented by the option that the porous structure is not completely filled. This incomplete filling may well be used as an inherent compensation for differences in thermal expansion. This material may also be used in combination with wire bonding, since its precursor component can be converted into the porous structure at a sufficiently low temperature.

Additionally, the coating has a good adhesion to underlying layers, such as a passivation layer. The porous structure, on the basis of a precursor material such as an alkoxide, has a good adhesion to oxidic and nitridic surfaces. The second solution, i.e. the organic polymer, has a good adhesion to organic and apolar surfaces. The adhesion may be further optimized by choosing the precursor material and the second solution, by choosing a second solution with specific adhesive properties and by variation of the volume ratio between the precursor material and the compound used to fill the porous structure.

In a specific embodiment of such a moulding compound for packaging, its constituents are chosen to be transparent. The stability of transparent epoxies in practice is an even larger problem than that of standard moulding compounds. Moreover, contemporary integrated circuits tend to produce a lot of heat, and the coating should be able to withstand this without any adhesion problems. As a consequence of the inorganic structure of the coating of the present invention, this is not problematic. In another specific embodiment, the coating is chosen to have sufficient flexibility. Flexible devices are known from for instance EP-A 1256983. They are suitably used in security applications, and data stored in the integrated circuits therein should be protected properly. A prior art security coating, as known from US-A 6,198,155, does not fulfill the condition of flexibility, but the porous coating of the invention may be optimized to have sufficient flexibility, and it is possible to apply it only on a limited number of areas of the semiconductor substrate - which limited provision may help to achieve the required flexibility. One example of a class of suitable materials is polyimides. These materials can be applied in soluble form as a polyamide, that is used as the second solution to fill the porous structure at least partially, and can be subsequently converted to polyimides. Alternative polymer or polymerisable materials could be applied as well.

In another embodiment, the coating may be provided with specific functional properties. Herein, the porous structure and the second solution that fills the porous structure may have a complementary functionality. If one component generates or transmits optical radiation, the other component can be used to inhibit transmission of the radiation in a certain direction, or to diminish the intensity thereof. If one component has magnetic properties, the other component may be used as a shield for the field, particularly in certain directions. Particularly the porous structure that is closed at its top side appears to be suitable for such a shielding or inhibiting function. Additionally, by means of the patterning method, that will be explained later on, the porous structure can be provided in a limited number of areas only. This may allow the transmission of for instance light in a limited number of areas only, whereas the light may be generated on the complete surface area.

In a further embodiment, the coating is used as a porous coating, on top of which subsequent layers are deposited. The nature of the coating, and its closed top side, make it very suitable for integration into a stack of layers. Porous layers may be applied for different purposes, as known to the skilled person. Using them as an intermetal dielectric may be a suitable application within the field of semiconductor devices.

In an advantageous embodiment of the method, the coating is patterned. Prior to application of the coating composition, a patterned structure is provided on the substrate. Then, a surface of the substrate and of the patterned structure are modified such that the surface of the substrate is relatively hydrophilic and the surface of the patterned structure is relatively hydrophobic. Consequently, on application of the coating composition, the patterned structure is kept free of coating composition. After application of the coating composition, the patterned structure is removed. This removal is preferably done before curing of the coating composition, for instance after a pre-bake step at about 100 ⁰C. The modification of the surface could be carried out by the application of certain modifying agents. Alternatively, use is made of plasma treatments.

Advantageously, the following process is used. First a patterned resist layer is applied to the substrate and patterned by photolithography, resulting in the above-mentioned patterned structure. The photoresist will be present in areas that are to be kept free of the coating composition. Then a fluorine plasma etch step is applied. Both the exposed surface of the substrate, particularly an oxide or nitride layer, and the resist are affected by this treatment. For silicon nitride, Si-OH groups present at the surface will be largely replaced by Si-F groups, rendering the nitride less hydrophilic. The resist will be affected differently, and more complex reactions will take place. This basically results in fluoridation, polymerization and some damage to the upper part of the resist. These effects provide a very water-repellant resist. After the plasma treatment, a mild oxidation is applied, for instance by a dip in a mixture of ammonia, hydrogen peroxide and water. In this step, the Si-F groups at the surface of the substrate are again replaced by Si-OH groups, resulting in a hydrophilic nitride surface. A large difference in wetting behaviour is achieved thereby. By the subsequent application of an aqueous solution, liquid is deposited only on those areas that are easily wettable. The resist is removed in an organic solvent, such as ethanol. This method allows patterning of very thick films, up to 30 microns or even more, using photoresist layers of a limited thickness only, such as 2 microns.

The patterning of the coating allows access to bond pads or other metallic areas and pads hidden under the coating. Alternatively, the patterning is used to limit the presence of the porous structure to predefined areas on the substrate surface. The method is furthermore advantageous, as the removal of a porous structure by means of a conventional lift-off technique is inherently problematic in view of its inhomogeneous nature and the deterioration of its stability. Moreover, patterning by application of a photoresist on top of the filled coating is problematic in view of the strength of the network, and the different materials that require specific etching processes. Also, the particulate character of the filler component inhibits the provision of well-defined holes. It is a second object of the invention to provide a coating composition with which an improved porous coating can be formed. This object is achieved in that the coating composition comprises a matrix precursor component and a filler component, which matrix precursor component is convertible into a matrix material in a heat treatment and which filler component is present in an amount of at least 30 vol.%, with the average particle size being at least 1 μm. The coating composition of the present invention may be provided after preparation in situ or after supply by another company. It is observed that there may be alternative compositions that may be used in the method of the invention to form a porous coating. The coating composition has however the advantage of good rheological properties and hence good processability.

It is observed that a coating composition comprising the precursor material monoaluminum phosphate and filler particles in the submicrometer range is known from US 6,759,736. However, the use of this coating composition does not lead to a porous coating. Preferably, the average particle size is in the range of 1 to 3 microns, more preferably between 1.2 and 2.8 microns, and most preferably between 1.5 and 2.5 microns.

The matrix precursor material is preferably chosen such that it can be converted to the matrix in a heat treatment at less than 500 ⁰C, more preferably less than 450 ⁰C or even less than 400 ⁰C. This allows the use thereof within the interconnect structure of semiconductor devices. Examples of such materials include for instance tetraethylene orthosilicate (TEOS), and monoaluminum phosphate (MAP), tetraethyl orthotitanate

(TEOTi), tetra-isopropyl titanate, niobium(V) ethoxide, tantalum(V) ethoxide, tantalum n- butoxide, zirconium n-butoxide or a mixture of one or more of the said precursor materials with each other or with further precursor materials. For application within semiconductor packaging, the heat treatment is most preferably carried out at a temperature of less than 300 ⁰C. Many precursor materials are capable of being converted at such a temperature. A suitable example is for instance TEOS, which can even be converted by a heat treatment at less than 100 ⁰C.

Most preferably, the filler component and the precursor matrix material are chosen such that the filler and the converted matrix are bonded to each other. A very strong bond is a chemical bond. Such a bond can be achieved because t the filler component comprises precursor groups such as alkoxides, at least at a portion of its surface. Alternatively, use can be made of bonds formed by condensation polymerization, for instance with acrylate groups. Alternatively, the filler and the converted matrix material are bonded to each other with physical bonds, including capillary forces, Van der Waals forces and hydrogen bonds. The latter mechanism is assumed to take place with TiN particles in a matrix that stills contains hydroxyl groups.

It is a third object of the invention to provide a substrate comprising a coating with a high yield and sufficient porosity. This object is achieved by means of a porous structure comprising filler particles that are encapsulated by a layer of a matrix component, and an anchoring component filling the porous structure at least partially.

In addition to the observations made with respect to the method, it is observed that the anchoring component provides the coating with strength. By the term 'anchoring component' is meant that the body filling the porous structure extends on different sides of the porous structure, such that a mechanical anchoring effect is obtained. The coating may have any of the specific applications described with reference to the method claims. It is particularly suitable for integration into an electronic device. A suitable example hereof is a semiconductor device such as an integrated circuit. However, application in neighbouring technical fields, such as (biomedical) sensors and optical devices, is certainly not excluded.

These and other aspects of the present invention will be apparent from and elucidated with reference to the non-limiting example described hereinafter. Example 1:

A porous protective coating according to the present invention was prepared using tetraethyl orthotitanate (TEOTi) as the matrix precursor component (resulting in a porous matrix of TiO₂), TiN particles as the filler component and again TEOTi (resulting in TiO₂) as the precursor reinforcing component.

To this end, 1 g TEOTi was added to a solution of 0.3 g HCl (6M) in 14.5 g ethanol. A quantity of 4.58 g TiN was added to this solution, resulting in a 4.4 vol.% solution (thereby obtaining 91 vol.% of the filler component TiN based on the porous matrix OfTiO₂ eventually obtained).

To obtain a substantially homogeneous coating liquid, it was milled with ZrO₂ milling beads (2 mm diameter) for 21 hours in a 60 ml PE-bottle, at about 120 rotations per minute (rpm). The amount of ZrO₂ beads added was such that the beads were just beneath the coating liquid surface.

The suspension obtained was subsequently deposited on a glass substrate by spin coating. An excess of liquid was applied while the substrate was rotating at 450 rpm, for 18 seconds. Thereafter the sample was spun at 620 rpm for 60 seconds. After spinning, the sample was kept on a hot plate at 100 ⁰C for 1 minute, followed by 200⁰C for 2 minutes. Finally the porous coating was cured at 400⁰C for 1 hour.

The porous coating contained 91 vol.% of the filler component TiN based on the porous matrix of TiO₂.

After curing the porous coating, it was filled with a reinforcing precursor component, in this example TiO₂, using a TEOTi-solution as a precursor. To this end, 30 g TEOTi was added to a solution of 30 g ethanol and 3.53 g acetic acid. The glass substrate, containing the porous TiN-TiO₂ coating layer, was immersed in the latter TEOTi-solution under a vacuum (300 mbar) for 10 minutes. To remove the excess TEOTi-solution, the sample was spun at 5400 rpm for 2 minutes. After spinning, the sample was kept on a hot plate at 100°C for 1 minute, followed by 200°C for 2 minutes. Finally the coating was cured at 400°C for 1 hour. The protective coating obtained had a thickness of 3 μm.

It has been shown that the coating could also be easily applied on an IC instead of on the glass substrate as shown above. The protective coating showed an excellent mechanical, physical and chemical resistance, as well as a suitable non- transparency. The person skilled in the art will understand that many modifications may be made without departing from the scope of the appended claims. As an example, the chemical resistance of the protective coating may be further improved by applying further coatings on the protective coating, if desired. For instance it may be additionally coated or impregnated with TiO₂, ZrO₂, Nb₂O₅ Ta₂O₅, etc. Furthermore, the coating described above may be patterned, if necessary.

Claims

CLAIMS:

1. A method of forming a coating on a substrate, the method at least comprising: providing a coating composition that comprises a matrix precursor component and a particulate filler component; applying the coating composition on a substrate, and curing the composition, thereby obtaining a porous structure; applying a second solution on the substrate, thereby at least partially filling the porous structure and obtaining the coating.

2. A method as claimed in claim 1, wherein the filler component comprises particles within a predefined size range.

3. A method as claimed in claim 1 or 2, wherein the matrix precursor component comprises hydro lysable groups bonded to a metal, which component is converted into a network in the curing step, which network substantially encapsulates the particulate filler component.

4. A method as claimed in claim 1, wherein the second solution comprises an alkoxide, that is converted into an oxide network in the curing step.

5. A method as claimed in claim 1, wherein the second solution comprises a polymer or a polymerisable compound.

6. A method as claimed in claim 1, wherein a predetermined amount of the filler component is used in order to obtain a porous structure with at least 40 vol.% of filler component.

7. The method of claim 3, wherein the matrix precursor component comprises a compound selected from the group consisting of tetraethyl orthotitanate (TEOTi), tetra- isopropyl titanate, niobium(V) ethoxide, tantalum(V) ethoxide, tantalum n-butoxide, zirconium n-butoxide, or a mixture thereof.

8. The method of claim 1 or 2, wherein the filler component comprises a compound selected from the group consisting of TiO₂, TiN or a mixture thereof.

9. The method of claim 1, wherein the porous structure has a porosity of 40 - 90 %, preferably > 50%.

10. The method of claim 1, wherein the coating is patterned in that prior to application of the coating composition, (a) a patterned structure is provided on the substrate and (b) a surface of the substrate and of the patterned structure is modified so that the surface of the substrate is relatively hydrophilic and the surface of the patterned structure is relatively hydrophobic, such that on application of the coating composition, the patterned structure is kept free of coating composition, and after application of the coating composition, the patterned structure is removed.

11. A coating composition for forming a porous coating on a substrate, comprising a matrix precursor component and a filler component, which matrix precursor component is convertible into a matrix material in a heat treatment, and which filler component is present in an amount of at least 40 vol.%.

12. The coating composition as claimed in claim 11, wherein the filler component has an average particle diameter between 1 and 3 μm.

13. The coating composition as claimed in claim 11, wherein the matrix precursor component comprises a metal alkoxide, that is converted into a metal oxide matrix in the heat treatment.

14. The coating composition as claimed in claim 11 or 13, wherein the filler component and the matrix material are bonded to each other.

15. A coating on a substrate comprising: a porous structure comprising filler particles that are encapsulated by a layer of a matrix component, and an anchoring component filling the porous structure at least partially.

16. An electronic device comprising a coating as claimed in claim 15.