DETAILED DESCRIPTION OF THE INVENTION

[0030] FIG. 1 (a) to FIG. 1 (c) are illustrations showing a preferable embodiment of the anisotropic conductive adhesive film according to the present invention, wherein FIG. 1 (a) shows the constitution thereof after heat compression bonding, FIG. 1 (b) is an enlarged view of the part surrounded by chain line A, and FIG. 1 (c) is an enlarged view of the part surrounded by chain line B.

[0031] FIG. 2 is an example of electron microscopic photographs (5,000×magnification) showing the structure of the conductive particles to be used in the present invention.

[0032] As FIG. 1 shows, the anisotropic conductive adhesive film 1 of the present invention, which is to be used in, for example, connecting electrodes 3 of a circuit board 2 to electrodes 5 of a glass panel 4, comprises conductive particles 7 dispersed in a filmy insulating binder resin (insulating binder) 6. The electrodes 3 of the circuit board 2 adjacent to each other have a pitch of about 60 μm.

[0033] In this embodiment, an oxide film 8 is formed on the surface of the electrode 3 of the circuit board 2, as shown in FIG. 1 (c).

[0034] In the present invention, use can be made, as the insulating binder resin 6, of those containing epoxy resins as the main component together with coupling agents, curing agents, etc.

[0035] This anisotropic conductive adhesive film 1 is formed on a separate film made of, for example, polyethylene terephthalate (PET) and the surface of the anisotropic conductive adhesive film 1 may be covered, if necessary, with a cover film, though not shown in the figure.

[0036] As the conductive particles 7, on the other hand, use is made of those obtained by coating a resin particle 71 with a thin metal film (a conductive thin film) 72 and forming projections 72a on the surface of the thin metal film 72, as will be described hereinafter.

[0037] As the resin particles 71, it is possible to use those made of resins such as epoxy resin, phenolic resin, acrylic resin, acrylonitrile/styrene (AS) resin, benzoguanamine resin, divinyl benzene resin, styrene resin, etc.

[0038] From the viewpoint of ensuring the adequate conductive reliability, the average particle diameter of the resin particles 71 preferably ranges from 2 to 10 μm, more preferably form 3 to 8 μm. To ensure the adequate conductive reliability too, it is preferable that the dispersion in the particle diameter of the resin particles 71 falls within a range of ±2 μm.

[0039] It is preferable that the resin particles 71 have such a hardness as giving a K value at 10% compressive deformation of from 100 to 2,000 kgf/mm² (1 kgf/mm²=9.80665 MPa), more preferably from 200 to 1,500 kgf/mm².

[0040] When the K value at 10% compressive deformation of the resin particles 71 is less than 100 kgf/mm², the resin particles 71 are largely deformed under pressure and thus the projections 72a of the thin metal film 72 cannot adequately break through the oxide film, thus resulting in a problem that the conductive resistance is elevated. When the K value thereof exceeds 2,000 kgf/mm², on the other hand, the resin particles 71 cannot be adequately deformed so as to fail to make up for the dispersion in the thickness of the electrode terminals. In this case, there arise additional problems such as a high pressure being required in the pressuring step, the enlarged elastic rebound among the conductive particles causing interfacial separation, etc.

[0041] Further, it is preferable that the resin particles 71 have a recovery ratio from 10% compressive deformation of 5% or above, more preferably 7% or above.

[0042] When the recovery ratio of the resin particles 71 is less than 5%, the resin particles 71 in the deformed state cannot exhibit a sufficient elastic rebound. Thus, these resin particles 71 cannot follow up the displacement of the insulating binder resin 6, etc., which results in a problem that the conductive resistance is elevated.

[0043] The K value at 10% compressive deformation of the resin particles 71 is determined from the relationship between the compressive displacement and the compressive load on the resin particles 71 at 10% compressive deformation among the relationships between the compressive displacement and the compressive load on the resin particles 71.

[0044] Namely, the K value can be expressed on following equation.

K value=2.8 P/πd²(kgf/mm²)

[0045] P: The load at 10% compressive deformation of the resin particles 71

[0046] d: The mean diameter of the resin particle 71

[0047] The recovery ratio from 10% compressive deformation of the resin particles 71 is defined as the ratio (percentage) of the displacement (L₁) of the resin particles 71 caused by compressing the resin particles 71 from the original load (0.1 gf) to the reverse load (1.0 gf) and the displacement(L₂) thereof caused by relieving the load from the reverse load (1.0 gf) to the original load (0.1 gf).

[0048] Namely, the recovery ratio(R) can be expressed in following equation.

R(recovery ratio)=(L₁/L₂)×100(%)

[0049] In the present invention, the thin metal film 72 is formed on the surface of the resin particles 71 by, for example, electrolessly plating. In this step, nickel, gold, etc. can be used as the material for forming the thin metal film 72.

[0050] It is preferable that the thin metal film 72 has a compressive elastic modulus of 1.5×10⁴ kgf/mm² or above, more preferably 2.0×10⁴ kgf/mm² or above on 10% compressive deformation.

[0051] When the compressive elastic modulus of the thin metal film 72 is less than 1.5×10⁴ kgf/mm², there arises a problem that projections 72a of the thin metal film 72 are deformed under pressure and thus cannot adequately break through the oxide film 8.

[0052] In the present embodiment of the invention, the projections 72a on the surface of the thin metal film 72 can be formed by, for example, changing the treatment temperature in the step of the electroless plating so as to change the reaction speed of nickel, etc.

[0053] The height of the projections 72a of the thin metal film 72 preferably ranges from 0.01 to 3 μm, more preferably from 0.1 to 1.1 μm.

[0054] When the height of the projections 72a of the thin metal film 72 is less than 0.01 μm, there arises a problem that the projections 72a of the thin metal film 72 cannot break through the oxide film 8 and thus fail to reach the electrodes 3 of the circuit board 2, thereby achieving only insufficient connection of the electrodes 3 and 5. When the height thereof exceeds 3 μm, on the other hand, the projections 72a can be formed only in a small number and thus the contact between the conductive particles 7 and the electrodes 3 becomes insufficient. In this case, only insufficient connection of the electrodes 3 and 5 can be achieved too.

[0055] It is preferable that 4 to 300, more preferably 4 to 200, on average, projections 72a are formed on the surface of the thin metal film 72.

[0056] When the number of the projections 72a of the thin metal film 72 is less than 4, the projections 72a break through the oxide film 8 and come in contact with the electrodes 3 only in a small contact area. As a result, there arises a problem that adequate connection between the electrodes 3 and 5 cannot be ensured. When 300 or more projections 72a are provided, on the other hand, the projections 72a of the thin metal film 72 condense together, which makes the thickness of the thin metal film 72 irregular.

[0057] The content of the conductive particles 7 having the above-described constitution preferably ranges from 1 to 15% by volume, more preferably from 2 to 15% by volume.

[0058] When the content of the conductive particles 7 is less than 1% by volume, there arises a problem that the connection between the electrodes 3 and 5 is not ensured and thus the conductive resistance is elevated. When the content thereof exceeds 15% by volume, on the other hand, there arises another problem that the conductive particles 7 condense together and thus the insulation resistance between the electrodes 3 and 5 adjacent to each other is lowered.

[0059] When the content of the conductive particles 7 is from 5 to 15% by volume, on the other hand, it is favorable to form an insulation layer on the surface of the thin metal film 72 of the conductive particles 7.

[0060] The anisotropic conductive adhesive film 1 according to the present invention is produced in the following manner. First, conductive particles 7 dispersed in a solvent are added to a solution in which a definite epoxy resin is dissolved and mixed to give a binder paste.

[0061] Next, this binder paste is applied onto a separate film such as a polyester film and dried. Then a cover film is laminated thereon to give the anisotropic conductive adhesive film 1.

[0062] The electrodes 3 and 5 can be connected to each other by using the anisotropic conductive adhesive film 1 of the present invention by, for example, the following method. The anisotropic conductive adhesive film 1 is adhered to the surface of the glass panel 4. And the glass panel 4 followed by registration and temporal tacking to the circuit board 2. Next, heat compression bonding is carried out at definite temperature and pressure as shown in FIG. 1 (a). Thus, the insulating binder resin 6 is hardened while electrically connecting the electrodes 5 of the glass panel 4 to the electrodes 3 of the circuit board 2.

[0063] In the present invention, the projections 72a of the thin metal film 72 break through the oxide film 8, due to the elastic rebound of the resin particles 71, and thus come into contact with the electrodes 3 of the circuit board 2 in the step of heat compression bonding, as FIG. 1 (c) shows. Thus, the electrodes 3 of the circuit board 2 are electrically connected to the electrodes 5 of the glass panel 4 via the thin metal film 72 of the conductive particles 7.

[0064] As a result, a high connective reliability on the electrodes 3 arranged with a fine pitch can be maintained even in the case where an oxide film 8 is formed on the electrodes 3.

[0065] Furthermore, the electrodes 3 of the circuit board 2 can be connected surely to the electrodes 5 of the glass panel 4 and thus a high connective reliability can be established by appropriately controlling the hardness of the resin particles 7, the compressive modules of the thin metal film 72, the number and height of the projections 72a formed on the surface of the thin metal film 72 and the content of the conductive particles.

[0066] When the conductive particles 7 are used in an amount of from 5 to 15% by volume, the thin metal film 72 may be coated with an insulation layer. Thus, it becomes possible to prevent electrical short circuit among the conductive particles 7, even though the conductive particles condense together, thereby maintaining a high connective reliability.

[0067] Now, the anisotropic conductive adhesive film according to the present invention will be described in greater detail by reference to the following Examples and Comparative Examples.

EXAMPLE 1

[0068] First, an insulating binder resin solution (solid content:50%) was prepared by dissolving 48% by weight of a solid bisphenol A-type epoxy resin (EP1009™ manufactured by Yuka-Shell), 50% by weight of an imidazole-based curing agent (HX3941HP™ manufactured by Asahi Chemical Industry Co., Ltd.) and 2.0% by weight of a silane coupling agent (A187™ manufactured by Nippon Unicar Co., Ltd.) in toluene employed as a solvent.

[0069] To the obtained binder solution were added nickel-gold-plated benzoguanamine particles as conductive particles to give a binder paste containing 8% by volume of the conductive particles.

[0070] These benzoguanamine particles had an average diameter of 5 μm and that the dispersion in the particle diameter fell within a range of ±1 μm. The benzoguanamine particles showed such a hardness as giving a K value at 10% compressive deformation of 800 kgf/mm² and a recovery ratio from 10% compressive deformation of 10%.

[0071] As FIG. 2 shows, the plated conductive particles were photographed under an electron microscope (5,000×magnification) and the projections formed on the surface of the thin metal film of each conductive particles were counted by using the photograph. In this method, the projections formed on a hemisphere were counted and twice as much the obtained value was referred to as the number of the projections. The projections on 5 conductive particles were counted and the average thereof was referred to as the observed value. In this Example, 19.6 projections were observed on average.

[0072] On the surface of the thin metal film of the conductive particles, an insulation layer (thickness: 0.1 to 1 μm) was formed by the known hybridization treatment with the use of acryl/styrene particles having an average particle diameter of 1 μm.

[0073] Then the binder paste as described above was applied onto a separate PET film so as to give a thickness of 25 μm after drying thereby giving an anisotropic conductive adhesive film. This anisotropic conductive adhesive film was cut into slits and used as the sample of Example 1.

EXAMPLE 2

[0074] An anisotropic conductive adhesive film sample was produced as in Example 1 but using conductive particles having 25.6 projections on average, adjusting the content thereof to 2.5% by volume and forming no insulation layer on the surface of the conductive particles.

EXAMPLE 3

[0075] An anisotropic conductive adhesive film sample was produced as in Example 2 but using conductive particles having 20 projections on average.

EXAMPLE 4

[0076] An anisotropic conductive adhesive film sample was produced as in Example 2 but using nickel-plated resin particles having 25.6 projections on average as the conductive particles.

EXAMPLE 5

[0077] An anisotropic conductive adhesive film sample was produced as in Example 2 but using conductive particles having acrylonitrile/styrene particles as the core, which showed a K value at 10% compressive deformation of 480 kgf/mm² and a recovery ratio from 10% compressive deformation of 30%, and having 26 projections on average.

EXAMPLE 6

[0078] An anisotropic conductive adhesive film sample was produced as in Example 2 but using conductive particles having resin particles as the core, the dispersion in the particle diameter of which fell within a range of ±2 μm, and having 26 projections on average.

EXAMPLE 7

[0079] An anisotropic conductive adhesive film sample was produced as in Example 2 but using conductive particles having an average diameter of 3 μm and having 20 projections on average.

EXAMPLE 8

[0080] An anisotropic conductive adhesive film sample was produced as in Example 2 but forming an insulation layer on the surface of the conductive particles by the same method as in Example 1 and employing the conductive particles and adjusting the content thereof to 8% by volume.

[0081] These conductive particles had 25.6 projections on average similar to Example 2.

EXAMPLE 9

[0082] An anisotropic conductive adhesive film sample was produced as in Example 8 but using conductive particles having 25.6 projections on average and adjusting the content thereof to 15% by volume.

[0083] These conductive particles had 25.6 projections on average similar to Example 2.

COMPARATIVE EXAMPLE 1

[0084] An anisotropic conductive adhesive film sample was produced as in Example 2 but using conductive particles having styrene particles (degree of crosslinking: 5%) as the core, which showed a K value at 10% compressive deformation of 400 kgf/mm² and a recovery ratio from 10% compressive deformation of 0%, and having 26 projections on average.

COMPARATIVE EXAMPLE 2

[0085] An anisotropic conductive adhesive film sample was produced as in Example 2 but using conductive particles having 0.8 projections on average.

COMPARATIVE EXAMPLE 3

[0086] An anisotropic conductive adhesive film sample was produced as in Example 2 but using conductive particles showing the dispersion in the particle diameter of ±3 μm and having 26 projections on average.

COMPARATIVE EXAMPLE 4

[0087] An anisotropic conductive adhesive film sample was produced as in Example 2 but using conductive particles having styrene particles (degree of crosslinking: 1%) as the core, which showed a K value at 10% compressive deformation of 80 kgf/mm² and a recovery ratio from 10% compressive deformation of 0%, and having 30 projections on average.

COMPARATIVE EXAMPLE 5

[0088] An anisotropic conductive adhesive film sample was produced as in Example 2 but using conductive particles having an average diameter of 2 μm and having 18 projections on average.

COMPARATIVE EXAMPLE 6

[0089] An anisotropic conductive adhesive film sample was produced as in Example 2 but using as the conductive particles gold-plated nickel particles having no projection on the surface.

COMPARATIVE EXAMPLE 7

[0090] An anisotropic conductive adhesive film sample was produced as in Example 1 but using conductive particles having no insulation layer on the surface and adjusting the content thereof to 0.5% by volume.

[0091] The conductive particles had 19.6 projections on average similar to Example 1.

COMPARATIVE EXAMPLE 8

[0092] An anisotropic conductive adhesive film sample was produced as in Example 1 but adjusting the content of the conductive particles to 20% by volume.

[0093] The conductive particles had 19.6 projections on average similar to Example 1.

COMPARATIVE EXAMPLE 9

[0094] An anisotropic conductive adhesive film sample was produced as in Example 1 but using conductive particles having no insulation layer on the surface.

[0095] The conductive particles had 19.6 projections on average similar to Example 1.

COMPARATIVE EXAMPLE 10

[0096] An anisotropic conductive adhesive film sample was produced as in Example 2 but using conductive particles having 8 projections on average.

[0097] Evaluation Data

[0098] Conductive resistance

[0099] By using each of the above samples, a circuit board was compression bonded to a glass board and the conductive resistance was evaluated. As the circuit board in this case, use was made of a TCP prepared by forming an electrode pattern with a pitch of 50 μm, wherein a copper foil of 15 μm in thickness was nickel/gold-plated, on exclusively one face of a base board of 75 μm in thickness made of polyimide (UPIREX™ manufactured by Ube Industries, Ltd.). The width of the contact part (top width) of each electrode pattern was adjusted to 13 μm.

[0100] As the glass board, on the other hand, use was made of a test element group (TEG) formed by metallizing aluminum electrodes (thickness: 0.5 μm) on the whole face of a glass plate of 0.7 mm in thickness.

[0101] The compression bonding was performed at 170° C. under 40 kgf/cm² for 10 seconds. By using the above-described sample, the circuit board and the glass board were compression bonded in a width of 1 mm and then the conductive resistance between patterns adjacent to each other was measured. Table 1 shows the results.

[0102] In this evaluation, a sample showing a conductive resistance less than 1 Ω was regarded as good (◯), one showing a conductive resistance of 1 to 2 Ω was regarded as somewhat poor (Δ), and one showing a conductive resistance more than 2 Ω was regarded as poor (X).

[0103] Insulation resistance

[0104] By using each of the above samples, a glass plate having a thickness of 0.7 mm and a surface insulation resistance of 1×10¹⁵ Ω or above was connected by compression bonding (connection width: 1 mm) to the TCP as described above at 170° C. under 40 kgf/cm² for 10 seconds. Then a potential of 25V was applied between electrode patterns adjacent to each other and the insulation resistance was measured. Table 1 shows the results.

[0105] In this evaluation, a sample showing an insulation resistance more than 1×10¹⁰ Ω was regarded as good (◯), one showing an insulation resistance of 1×10⁸ to 1×10¹⁰ Ω was regarded as somewhat poor (Δ), and one showing an insulation resistance less than 1×10⁸ Ω was regarded as poor (X).

[0106] Conductive reliability

[0107] The glass board and the TCP, which had been compression-bonded to each other, employed in the insulation resistance test were aged at 85° C. under a relative humidity of 85% for 1,000 hours. Then a potential of 25V was applied between electrode patterns adjacent to each other and the resistance was measured. Table 1 shows the results.

[0108] In this evaluation, a sample showing a resistance more than 1×10¹⁰ Ω was regarded as being good in conductive reliability (◯), one showing a resistance of 1×10⁸ to 1×10¹⁰ Ω was regarded as somewhat poor (Δ), and one showing a resistance less than 1×10⁸ Ω was regarded as poor (X).

Table 1

[0109] 1

[0110] As Table 1 shows, when the K value at 10% compressive deformation and the recovery ratio therefrom of resin particles were changed as in Examples 2 and 5 and Comparative Examples 1 and 4, the samples of Examples 2 and 5 showed favorable data in all of the items examined, while the sample of Comparative Example 4 having a low compressive elastic modulus (80 kgf/mm²) showed a somewhat poor conductive resistance and a somewhat poor conductive reliability after aging.

[0111] The sample of Comparative Example 1 having a low recovery ratio from 10% compressive deformation (0%) showed a somewhat poor conductive resistance and a poor conductive reliability after aging.

[0112] The sample of Comparative Example 6 with the use of nickel particles as the conductive particles (compressive elastic modulus 2.1×10⁴ kgf/mm, recovery ratio 0%) showed a poor conductive reliability after aging.

[0113] When the number of projections of conductive particles were changed as in Examples 1 to 7 and Comparative Examples 2 and 10, the samples having 8 or less projections on a conductive particles showed each a somewhat poor conductive resistance and a poor conductive reliability after aging (Comparative Examples 2 and 10).

[0114] When the contents of the conductive particles were varied in the presence or absence of the insulation layer as in Examples 1, 2, 8 and 9 and Comparative Examples 7, 8 and 9, the sample containing an excessively small amount (0.5% by volume) of the conductive particles showed a somewhat poor conductive resistance and a somewhat poor conductive reliability after aging (Comparative Example 7).

[0115] When the conductive particles were employed in an amount of 8% by volume, on the other hand, the samples provided with the insulation layer (Examples 1 and 8) showed favorable data in all of the items tested, while the sample having no insulation layer (Comparative Example 9) showed a lowered insulation resistance.

[0116] The sample of Example 9 containing 15% by volume of the conductive particles and provided with the insulation layer, favorable data were obtained in all of the items examined. When the content of the conductive particles was increased to 20% by volume (Comparative Example 8), the insulation resistance was lowered.

[0117] As described above, the present invention makes it possible to provide an anisotropic conductive adhesive film capable of maintaining a high connective reliability on connection electrodes with a fine pitch on which an oxide film is formed.