Nano copper paste and film for sintered die attach and similar applications

Abstract

A sintering powder comprising copper particles, wherein: the particles are at least partially coated with a capping agent, and the particles exhibit a D10 of greater than or equal to 100 nm and a D90 of less than or equal to 2000 nm.

Claims

1. A method of forming a joint between two or more work pieces, the method comprising: providing two or more work pieces to be joined, providing a sintering paste in the vicinity of the two or more work pieces, the sintering paste comprising a sintering powder comprising copper particles, wherein the particles are at least partially coated with a capping agent, the particles exhibit a D10 of greater than or equal to 100 nm and a D90 of less than or equal to 2000 nm, the sintering powder comprises up to 1 wt. % of the capping agent, and the capping agent comprises triethanolamine, drying the sintering paste, and heating the sintering paste to at least partially sinter the copper, wherein the sintering paste is screen-printed onto one or more of the work pieces to be joined to form a screen-printed pattern, and solder is applied to the screen-printed pattern.

2. The method of claim 1, wherein the sintering powder comprises from 0.1 to 0.5 wt. % of the capping agent.

3. The method of claim 1, wherein the particles exhibit a D90 of less than or equal to 1000 nm.

4. The method of claim 1, wherein the two or more work pieces comprise a die and a substrate.

5. The method of claim 1, wherein the drying is carried out at a temperature of from 30 to 100 C. from 1 to 60 minutes.

6. The method of claim 1, wherein a pressure of from 2 to 18 MPa is applied during the step of heating.

7. The method of claim 1, wherein: the screen-printed pattern is coated by an electroless nickel immersion gold (ENIG) technique prior to the solder being applied.

8. The method of claim 1, wherein the particles exhibit a D10 of greater than or equal to 125 nm and a D90 of less than or equal to 450 nm.

Description

(1) The invention will be described in relation to the following non-limiting drawings in which:

(2) FIG. 1 shows an FESEM image of a synthesised copper powder according to the invention (10,000).

(3) FIG. 2 shows an FESEM image of a synthesised copper powder according to the invention (30,000).

(4) FIG. 3 shows boxplots of die shear at 5 and 10 MPa pressure and the effect of post cure.

(5) FIG. 4 shows boxplots of die shear at sintering temperature 250 C., at different pressures.

(6) FIG. 5 shows SEM cross-section images after die attachment using pressure sintering, at 250 C. and different time and pressure sintering conditions.

(7) FIG. 6 shows SEM cross-section images after die attachment using pressure sintering, at 200 C. and different time and pressure sintering conditions.

(8) The invention will now be described in relation to the following non-limiting examples.

(9) The materials used in the following Examples were purchased as follows: copper (II) acetate monohydrate was purchased from Fischer Scientific; triethanolamine (85%) was purchased from ViVochem; sodium hydroxide, hydrazine (85% LR), methanol and acetone were purchased from SDFCL; and demineralised water was purchased from Spectrum Chemicals.

(10) The equipment used in the following Examples was as follows: die shear was performed on Dage4000 PXY; film was casted on Pro-cast tape caster (TC-71LC); and sintering was done using Carver press (3891CEB.4NE1001).

Example 1

Synthesis of Copper Nanoparticles

(11) Cupric acetate (100 g) was dissolved in demineralized water (1500 ml). To the aqueous solution of cupric acetate, triethanolamine (20 g) was added dropwise with vigorous stirring and the solution was then stirred for 30 minutes. Potassium hydroxide/sodium hydroxide (20 g potassium hydroxide in 100 ml water) was then added to make the solution basic (pH>7) and the solution was allowed to stir for a further 30 minutes. To this mixture, excess hydrazine hydrate solution (150 ml) was added via a dropping funnel. As the reaction is very exothermic and emits effervescence, precaution should be taken. Hence, the reaction was carried out in a water bath where the temperature of the reaction was maintained between 20 and 25 C. The resulting solution was then stirred for 6 hours. The colour of the reaction mixture changed from bluish green to orange and then finally to a light brown colour. The solution was then allowed to stand for 1 hour so that the synthesized copper particles were allowed to settle at the bottom and the resulting solution (mother liquor) was then decanted. The synthesized powder was then washed thoroughly with excess water to remove the unwanted reactants. The washed powder then washed with acetone. The powder was dried at 35 C.

(12) The synthesized powder was then characterised by powder scanning electron microscopy (SEM) and powder X-ray diffraction (XRD). The D10 by image analysis was greater than or equal to 100 nm and the D90 was less than or equal to 500 nm. FIGS. 1 and 2 show field emission scanning electron microscopy (FESEM) images of the synthesised particles. FESEM shows that the particles have a distinct shape and have a narrow particle size distribution, ranging from 200 to 400 nm. Not wishing to be bound by theory, it is considered that the larger size of the copper nanoparticles makes them more resistant towards oxidation in comparison to copper nanoparticles in conventional sintering powders.

(13) The particle sizes were also determined using a particle size analyser (PSA), which showed a very consistent and narrow size distribution ranging between 150 and 350 nm, in conformity with SEM.

Example 2

Preparation of Copper Nanopaste

Example 2.1

(14) The above-synthesized nano copper powder (24 g) was dispersed with epoxy methacrylate urethane (0.279 g) using a high-speed mixer (1000 rpm, 1 minute). To the dispersion, triethanolamine (85%, 0.279 g), BYK 163 (0.558 g), malonic acid (1.116 g) and terpineol (1.674 g) were added and blended using a high speed mixer at 1000 rpm for 1 minute. The resulting mixture was then milled thoroughly using the EXAKT three roll mill. The collected homogeneous printable paste was then stored under standard temperature and pressure (STP).

Example 2.2

(15) The above-synthesized nano copper powder (15 g) was dispersed with epoxy methacrylate urethane (0.70 g) using high-speed orbital mixer. To the dispersion, BYK (0.70 g) was added. A methanol solution of malonic acid (1.41 g of malonic acid in 1.41 g of methanol) was added to the dispersion followed by the addition of terpineol (1.5 g). The mixture was then put in high-speed orbital mixture and milled in a three-roll mill for few minutes to provide a homogenous paste.

Example 2.3

(16) The above-synthesized nano copper powder (15 g) was dispersed with epoxy methacrylate urethane (0.754 g) using high-speed orbital mixer. To the dispersion, BYK (0.754 g) was added. Malonic acid (1.51 g) was added to the dispersion followed by the addition terpineol (2.64 g). The mixture was then put in high-speed orbital mixture and milled in a three-roll mill for few minutes to provide a homogenous paste.

Example 2.4

(17) The above-synthesized nano copper powder (15 g) was dispersed with epoxy methacrylate urethane (0.754 g) using high-speed orbital mixer. To the dispersion, BYK (0.754 g) was added. Malonic acid (1.51 g) was added to the dispersion followed by the addition of formic acid (1.511 g) and terpineol (2.64 g). The mixture was then put in high-speed orbital mixture and milled in a three-roll mill for few minutes to provide a homogenous paste.

(18) The electrical conductivity of the above-synthesized copper pastes was about 1.810.sup.6 S/m.

Example 3

Process of Film Casting

(19) The copper paste was casted on polyethylene terephthalate (PET) film with the help of a tape caster. The thickness of the film was set at 75 m. The copper paste was passed through the tape caster at 100 C. The casting time of the film was around 25 minutes. The thickness of the casted film was around 50 to 60 m.

Example 4

Film Transfer on Die Process

(20) A film transfer on die (DTF) process was implemented using Datacon die bonder. The stamping conditions for 3 mm*3 mm gold coated silicon dies were as follows in Table 1:

(21) TABLE-US-00001 TABLE 1 Stamping pressure Stamping temperature Stamping time (MPa) ( C.) (s) 5 200-225 1-10

(22) The film was fully transferred onto the die side with no remains of the copper film onto the PET substrate. The copper-coated Si die was then attached to gold/copper coated direct bond copper (DBC) using a Tresky die bonder under the following conditions in Table 2:

(23) TABLE-US-00002 TABLE 2 Stamping pressure Stamping temperature Stamping time (MPa) ( C.) (s) 5-10 300 100

(24) Die shear obtained from the above conditions is shown in FIG. 3. In FIG. 3, the post curing conditions were at a temperature of 300 C. for 30 minutes. The die shears at 5 MPa sintering pressure, with and without post cure, were about 14 MPa and about 20 MPa, respectively. The die shears at 10 MPa sintering pressure, with and without post cure, were about 24 MPa and about 30 MPa, respectively. It can be clearly seen that there is a drastic increase in joint strength when the vehicle is post cured at 300 C. for 30 minutes. The failure mode was a bulk failure irrespective of the conditions being used.

Example 5

Wafer Lamination for 2 Inch Silicon Gold Plated Wafer Using Carver Press

(25) Both of the platens of the Carver press were kept at 175 C. Lamination of the silicon wafer was done by using 5 to 10 MPa pressure. Silicon rubber was used as a cushion effect for laminating. The dwell time for the lamination is around 3 minutes. The stamped portion of the film showed no remains of the film on the PET sheet. The laminated wafer was then mounted on UV tape and diced using a dicing machine.

(26) The diced 3 mm*3 mm die was then attached to Au/Cu coated DBC using Carver press. A joint strength of around 30 to 32 MPa was achieved when the sintering pressure was 5 MPa. Whereas, a joint strength of 40 MPa was achieved when the sintering pressure was around 10 MPa at 250 C. for 3 minutes dwell time.

(27) FIG. 4 shows boxplots of die shear at sintering temperature 250 C., at different pressures. In FIG. 4, the first temperature, pressure and time sintering conditions were 250 C., 5 MPa and 3 minutes, respectively. The second temperature, pressure and time sintering conditions were 250 C., 10 MPa and 3 minutes, respectively. The die shears at 5 MPa and 10 MPa sintering pressure, were about 31 MPa and about 40 MPa, respectively.

Example 6

Process for Free Standing Film

(28) Free standing copper film was made using a Carver press. The casted film was pressed over a silicon wafer at 5 to 10 MPa pressure at 200 C. The dwell time was around 2 minutes. The copper film took the shape of the silicon wafer and did not diffuse into the silicon which, in turn, results in the detachment of the film from the polymeric substrate, resulting in a free standing film of around 30 to 40 m thickness. The conductivity of the nanocopper film when a 9V battery was connect across the film resulted in the glow of an LED. The electrical resistivity of the film was found to be 210.sup.8 .Math.m.

Example 7

Nanocopper Paste Die Attachment Using Pressure Sintering

(29) The attachment of semiconductor or other die elements can be accomplished by printing the nanocopper paste onto the substrates, such as direct bond copper (DBC), direct plate copper (DPC), metal core printed circuit boards (MCPCB), FR4, copper lead-frames, flexible PCBs, followed by drying the printed area by heating the printed substrate at 60 C. for 20 minutes. The process is then followed by die placement via a die bonder or a pick and place machine, and sintering in Carver press using pressure sintering.

(30) The joint strength of the sintered copper joints was evaluated on DBC using a 3 mm*3 mm silicon gold coated die using a sintering pressure of 5 MPa and 10 MPa, for a sintering time of 3 minutes and 5 minutes and at 200 C. and 250 C. sintering temperatures, respectively. After the die placement, the entire assembly was then covered with an aluminium foil, which would prevent the oxidation of copper. Cushioning effect was provided using a graphite sheet of 0.5 mm thickness kept above the silicon die. It was observed that at a lower sintering pressure (5 MPa) and temperature (with a sintering time of 3 minutes), the joint strength was less as compared to higher sintering pressure and temperature.

(31) The effect of different sintering temperature and pressure on the joint strength was tested. The results are summarised below in Table 3. Table 3 shows the effect of different sintering temperature and pressure on the joint strength. It is noted that, at higher sintering temperature and time, most of the dies shatter which reveals the fact that the joint strength is too good to shear the die from the substrate. Moreover, it also reveals strong diffusion of copper on both the interfaces.

(32) TABLE-US-00003 TABLE 3 Sintering Sintering Sintering time temperature pressure Die shear (minutes) ( C.) (MPa) (MPa) 3 200 5 about 19 3 200 10 about 29 3 250 5 about 32 3 250 10 about 39 5 200 5 about 28 5 200 10 about 43 5 250 5 about 31 5 250 10 about 45

(33) The diffusion of copper on the interfaces is also confirmed by SEM cross section, which shows a very good densification of copper nanoparticle within the sintered layer and excellent diffusion of the nanoparticle on both the interfaces resulting in a bulk failure. It is clearly seen that the diffusion and densification of copper nanoparticle at 10 MPa is much better than 5 MPa sintering pressure. This can be seen in the SEM cross-sections in FIGS. 5 and 6. In FIG. 5, the top two rows of cross-section images show interface comparisons; the top row being a die-copper layer interface and the middle row being a copper layer-substrate interface. The bottom row of cross-section images shows a microstructurel comparison of the copper layer. Each column of cross-section images exemplifies a different set of time and pressure sintering conditions. The sintering pressure in the first and second columns, and in the third and fourth columns were 5 MPa and 10 MPa, respectively. The sintering time in the first and third columns, and in the second and fourth columns, were 3 and 5 minutes, respectively. In FIG. 6, the top two rows of cross-section images show interface comparisons; the top row being a die-copper layer interface and the middle row being a copper layer-substrate interface. The bottom row of cross-section images shows a microstructurel comparison of the copper layer. Each column of cross-section images exemplifies a different set of time and pressure sintering conditions. The sintering pressure in the first column, and in the second and third columns were 5 MPa and 10 MPa, respectively. The sintering time in the second column, and in the first and third columns, were 3 and 5 minutes, respectively.

(34) Attachment of such semiconductor and die elements can also be accomplished by DTF and lamination on the die backside made from the said material, followed by die placement and sintering using pressure.

(35) The main advantage of the wafer lamination is the elimination of screen printing which is an added advantage in terms of machineries and man power. The diced copper coated silicon wafer was then singulated using a dicing machine and were attached to DBC using Carver press at 10 and 5 MPa pressure for 3 minutes. It was observed that the die shear was around 30 to 40 MPa respectively.

(36) It was observed that at 10 MPa pressure, most of the silicon dies shatter which again proves excellent bond strength of the die to the substrate. Without being bound by theory, it is considered that with the increase in sintering pressure, the nanoparticles increasingly come into contact with each other which, in turn, increases the contact point of the nanoparticles resulting in better fusion of the particle giving excellent joint strength.

Example 8

Nanocopper Paste Adhesion on Ceramic, FR4 and PET Substrates

(37) The nanocopper paste was tested for adhesion on ceramic, FR4 and PET. The paste was printed on various substrates using DEK printer and then cured at 1700 C. under nitrogen in a box oven for 30 mins. The print was then tested by typical scratch-adhesive tape test method. The adhesion to ceramic and FR4 was classified as 5B whereas adhesion to PET was classified as 4B.

Example 9

Screen Print and Soldering

(38) The synthesized nanocopper paste also has the ability to be screen printed. The paste was screen printed with a mesh screen. Design patterns were made using a screen of mesh size of 70 in a DEK printer.

(39) The screen-printed nanocopper pattern was then electroless nickel immersion gold (ENIG) coated followed by soldering with standard solder paste. The soldering showed very good spread but mid chip solder balls were seen at some places.

(40) Furthermore, the possibility of soldering on copper print printed by the synthesized nano copper paste on bare FR4 coupon has been explored. The soldering on nanocopper was undertaken with a standard solder paste. Soldering without ENIG coating on the copper print also showed a good spread and good soldering features.

(41) The foregoing detailed description has been provided by way of explanation and illustration, and is not intended to limit the scope of the appended claims. Many variations in the presently preferred embodiments illustrated herein will be apparent to one of ordinary skill in the art and remain within the scope of the appended claims and their equivalents.