Buffer layer for sintering

Abstract

A layer of material having a low thermal conductivity is coated over a substrate. A film of conductive ink is then coated over the layer of material having the low thermal conductivity, and then sintered. The film of conductive ink does not absorb as much energy from the sintering as the film of conductive ink coated over the layer of material having the low thermal conductivity. The layer of material having the low thermal conductivity may be a polymer, such as polyimide.

Description

BACKGROUND INFORMATION

The microelectronics and semiconductor packaging industries have begun to shift to printable electronics. Electronics circuits comprise a variety of components that are electrically connected to each other. Such electrical connections between different components may be made of conductive metal traces that can be printed on substrates with conductive inks. The inks are processed and sintered after deposition on a substrate in order to become conductive. Thermal sintering uses a high temperature (e.g., ≧250° C. to fuse the nanoparticles in the inks. Photonic (photo) and laser sintering utilize a very high intensity lamp/laser to fuse the nanoparticles in a very short period of time (e.g., microseconds) with a low temperature and so as not to damage the underlying substrates. However, the photo/laser sintering process has limits that require low thermal conductivity material for substrates in order for the nanoparticles to effectively absorb energy and sinter before heat energy dissipates into the substrate. In other words, the substrates that can be used in these applications will be very limited for low thermal conductivity materials.

On the other hand, low thermal conductive substrates can be used for flexible printable electronics. Low temperature melting point materials such as polyethylene (PE), polyester (PET), etc., will prevent the nanoparticle inks from proper sintering, and the substrates will be damaged, with the result that the resistivity will be very high.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a digital photograph showing copper inks photosintered on four silicon wafers.

FIG. 2 is a digital photograph showing copper inks before being photosintered.

FIG. 3 is a digital photograph showing copper inks after being photosintered.

FIG. 4 is a digital photograph showing laser sintered lines on a Kapton substrate.

FIG. 5 is an enlarged digital photograph showing the laser sintered lines of FIG. 4.

FIG. 6 illustrates a graph showing that copper ink resistivity sintered by a laser is not only inversely proportional to laser power, but also inversely proportional to buffer layer thickness made of polyimide.

FIG. 7 illustrates a graph showing thicknesses of cured polyimide measured at various spin speeds.

FIG. 8 illustrates a graph showing that resistivity of sintered copper film is inversely proportional to polyimide thickness.

FIG. 9 illustrates a graph showing that adhesion of copper ink film to polyimide is proportional to polyimide thickness.

FIG. 10 illustrates a graph showing that laser writing line width is proportional to the laser power density.

FIGS. 11A-11F illustrate a process in accordance with embodiments of the present invention.

DETAILED DESCRIPTION

Embodiments of the present invention disclose a photosintering process to effectively sinter metallic nanoparticles on a polyimide substrate, thus causing the film to be very conductive near the bulk material. On other hand, the photosintering process does not perform well on nanoparticle inks coated on substrates possessing a high thermal conductivity, such as ceramics and silicon wafer. Table 1 shows the thermal conductivity for a variety of materials.

TABLE 1
				heat		melting
		density	heat capacity	conductivity	thermal	point, ° C.
item #	material	(g/cm3)	(J/g · K)	(W/m · K)	effusivity	degrees
1	air	0.0013	1	0.025	0.00	NA
2	paper	0.33	0.73	0.030	0.01	NA
3	Polyimide (kapton)	1.42	1.09	0.120	0.19	NA
4	PMMA (resist)	1.19	1.46	0.160	0.28	180
5	PET (Mylar)	1.23	1.13	0.176	0.24	150
6	LCP (liquid crystal	1.4	1.6	0.500	1.12	300
	polymer)
7	PE (polyethylene,	0.95	2.3	0.500	1.09	125
	high density)
8	water	1	4.2	0.600	2.52	0
9	glass	2.3	0.753	1.330	2.30	950
10	SiO2	2.2	0.75	1.380	2.28	1600
11	MgO	3.2	0.84	5.900	15.86	2852
12	carbon-amorphous	1.51	0.707	6.280	6.70	3600
13	Si3N4	2.8	0.69	7.950	15.36	1900
14	TiO2	4.25	0.69	9.000	26.39	1843
15	CuO/Cu2O	6.5	0.536	18.000	62.71	1235
16	Ti	4.5	0.523	21.000	49.42	1668
17	Al2O3 (ceramics)	2.5	0.81	30.000	60.75	2054
18	solder (60/40	8.5	0.197	50.210	84.08	185
	Sn/Pb)
19	Ni	8.9	0.444	88.000	347.74	1455
20	Mo	10.2	0.25	134.000	341.70	2623
21	Si	2.33	0.7	148.000	241.39	1414
22	carbon-graphite	2.25	0.707	167.360	266.23	3600
23	Al	2.7	0.88	209.000	496.58	660
24	Au	19.3	0.13	318.000	797.86	1086
25	Cu	8.9	0.385	398.000	1363.75	1064
26	carbAL	2.3	0.75	425.000	733.13	3600
27	Ag	10.5	0.24	427.000	1076.04	962
28	carbon-diamond	3.51	0.506	543.920	966.03	3800
29	carbon nanotubes			6000.000

Low conductivity materials, such as polyimide, can be used as a coating material onto other high thermal conductivity substrates, such as ceramics and silicon wafer, in order to isolate heat energy dissipation from nanoparticles during a photosintering process so that the nanoparticles are fused more effectively. How quickly heat dissipates depends on the thickness of the low thermal conductivity material (e.g., polyimide film).

The following experiment was conducted for showing how the present invention operates. Three wafers were spin coated with 1, 1.5, and 2.3 microns thick DuPont PI-2610 polyimide, respectively, and thermal cured at 350° C. for 30 minutes. One bare silicon wafer was used for a reference (wafer #1). All four wafers were coated with copper ink using a drawdown process. After a 60 minute drying process at 100° C., each wafer was divided into three zones that were individually sintered with three different energy levels. The resistance for each zone and each wafer was measured with a voltmeter, with the results shown in Table 2, which shows the electrical resistances of a copper film after photosintering with various coating thicknesses of polyimide on silicon wafers.

TABLE 2
			Zone 1:	Zone 2:	Zone 3:
	polyimide	copper	resistance	resistance	resistance with
wafer #	thickness (μm)	ink (μm)	with energy 1	with energy 2	energy 3
1	0	3.2	>20 MΩ	>20 MΩ	>20	MΩ
2	1	3.2	>20 MΩ	>20 MΩ	>20	MΩ
3	1.5	3.2	>20 MΩ	>20 MΩ	>20	MΩ
4	2.3	3.2	>20 MΩ	>20 MΩ	20	Ω
where
energy 1 = 3 sinter shots with 850/1050 V, 1000 μsec
energy 2 = 4 sinter shots with 850/1150 V, 1000 μsec
energy 3 = 5 sinter shots with 850/1250 V, 2000 μsec

Except for zone 3 of wafer 4, all zones from the four wafers did not experience a change in resistance after photosintering. Zone 3 of wafer 4 experienced a change in its metallic color at the highest energy level, as shown in FIG. 1. The area had a severe blow off. The surrounding area had copper debris left that was conductive. This is clear evidence that the polyimide material may be used as a thermal insulator. The thickness of polyimide may be more than 3 microns. The thermal conductivity is 0.12 and 148 W/m·K for polyimide and silicon, respectively. The heat dissipated into the silicon substrate (wafer #1) too quickly to sinter the copper nanoparticles since there was no polyimide material.

Wafers 1, 2, and 3 all had high resistance (greater than 20 mega-ohms). Wafer 4 at the center zone with 20 ohms resistance as shown in FIG. 1 appeared that the copper nanoparticies film started to be fused, sintered, and turned into a copper color. The thicker low thermal conductivity material can thus be used as a good thermal insulator.

In addition to the liquid polyimide disclosed above, a dry polyimide film was also utilized. The copper ink was coated on a 50 micron polyimide film (e.g., Kapton). The sample was placed on a silicon wafer and a carbAL high thermal conductive heat sink, as shown in FIG. 2. Silicon grease was coated in between the polyimide film and the silicon wafer and heat sink to ensure good thermal contact. The sample was photosintered simultaneously in a single shot. The copper was sintered very well and turned a shiny copper color, as shown in FIG. 3. It did not matter what materials the polyimide film was residing on. At least a 50 micron thick polyimide film is sufficiently thick to isolate and prevent heat energy dissipation for photosintering processes, though a thickness of less than 50 microns may be utilized for embodiments where less conductivity is desired of the conductive traces.

In addition, laser sintering was utilized on silicon wafers with the same setup as described above. The laser was a solid state diode with an 830 nm wavelength and an 800 mW power. The focus beam size was 15 microns in diameter and controlled by a collimator and an objective lens, as shown in FIGS. 4 and 5.

This laser had sufficient power to sinter and fuse the nanoparticies and turn the copper ink conductive. There were four silicon wafers coated with various polyimide thicknesses of 1, 1.5, 2, and 3 microns, respectively, along with a bare silicon wafer as a reference. The resistivity of each wafer is plotted with laser power in FIG. 6, which indicates that the copper film conductivity is proportional to the polyimide thickness, and the heat generated by the laser is transferred to the substrate less with polyimide present than the bare silicon wafer without polyimide. This is clear evidence that any material having a low thermal conductivity, such as a polyimide material, may be used as a thermal insulator and enhance the photograph and laser sintering processes.

Furthermore, a variety of polyimide thicknesses were coated on silicon wafers and cured at 350° C. for one hour. Then the standard copper ink was coated by drawdown, dried in an oven, and photo/laser sintered. Electrical measurements were performed and characterized the copper ink samples.

Three types of polyimide material (e.g. made by DuPont) were used to spin coat on silicon wafer at 1000, 2000, 3000, 4000, and 5000 rpm. FIG. 7 illustrates a graph showing thicknesses of cured polyimide measured at various spin speeds. The range was from 1 to 20 microns on each wafer, respectively.

After samples were prepared, both photograph and laser sintering were performed on the copper inks. Different types of sintering were compared versus resistivity and adhesion, as well as line width for laser sintering. Table 3 shows samples photosintered at the same energy level with various thicknesses of polyimide. Table 4 shows samples laser sintered at a fixed power level with various thicknesses of polyimide,

TABLE 3
sample	polyimide	Cu ink thickness	resistivity	adhesion
#	thickness (μm)	(μm)	(ohm-cm)	(1-10)
1	0	3	3.00 × 101	1
2	5	2	1.30 × 10−4	2
3	6.5	2	4.00 × 10−5	4
4	8.7	2	1.60 × 10−5	7
5	12.5	2	1.52 × 10−5	7
6	10	1.5	1.50 × 10−5	8
7	14	1.5	1.40 × 10−5	8
8	20	1.5	1.14 × 10−5	8

TABLE 4
	polyimide	resistivity		line width	line width
	thickness	(ohm-cm) at	adhesion	(μm) at	(μm) at
sample #	(μm)	840 mW	(1-10)	840 mW	409 mW
11	0	1.60 × 10−4	1	70	35
12	1	1.26 × 10−5	5	74	38
13	1.5	1.36 × 10−5	5	77	39
14	2	9.33 × 10−6	3	83	40
15	3	6.00 × 10−6	1	88	42
16	5	4.75 × 10−6	8	92	65
17	7	4.82 × 10−6	8	103	75
18	12	3.61 × 10−6	8	150	88
19	20	5.47 × 10−6	8	180	120

FIG. 8 illustrates a graph showing that resistivity of sintered copper film is inversely proportional to polyimide thickness. The saturated points for resistivity are approximately at 10 microns for photosintering and approximately at 5 microns for laser sintering. Power density of photosintering is much lower than that of laser sintering, providing a reason why its resistivity is higher.

FIG. 9 illustrates a graph showing that adhesion of copper ink film to polyimide is proportional to polyimide thickness. There are some noise points, but the trend is clear from the graph. The thicker the polyimide is, the better the adhesion is. Again, critical points of polyimide thickness for the good adhesion are approximately at 10 microns for photosintering and approximately at 5 microns for laser sintering.

FIG. 10 illustrates a graph showing that laser writing line width is proportional to the laser power density. With given laser power, the laser writing line width is also proportional to the polyimide film thickness, providing more evidence that polyimide is a good thermal, insulator for these processes. The laser energy and heat deposited on the copper ink surface could not spread any deeper vertically but laterally while the polyimide thickness increased.

Referring to FIGS. 11A-11F, a process for performing embodiments of the present invention is illustrated. A substrate 1101 is provided on which electronic circuitry is to be mounted. In FIG. 11B, traces of a metal material 1102 are deposited in a desired pattern on the substrate 1101, using a well-known manufacturing process. In FIG. 11C, a layer of low thermal conductivity material 1103, such as polyimide, is coated over the metal traces 1102 and substrate 1101. To create further patterns for the conductive traces to be deposited, vias 1104 are formed through the material 1103, exposing portions of the metal traces 1102. In FIG. 11E, an ink jet apparatus 1106 deposits a conductive ink 1105, such as copper nanoparticles, over the material 1103 and the metal traces 1102 exposed by the vias 1104. In FIG. 11F a photograph or laser sintering process is performed on the deposited conductive ink nanoparticles 1105 to sinter them into conductive traces 1107, as described herein. Depositing of the conductive inks and the sintering processes are described in U.S. Patent Publication No. 2008/0286488 A1, which is hereby incorporated by reference herein.

SUMMARY

1. The effectiveness of a photosintering process depends on not only metallic nanoparticle size, but also the type of substances.

2. Effective photosintering is achieved with nanoparticles below 300 nm.

3. The thermal conductivity of substrates will affect metallic ink photosintering. The lower the thermal conductivity of the substrate, the better the electrical conductivity of the nanoparticle film.

4. High thermal conductive substrates can be tailored and isolated by coating low thermal conductivity material, such as polyimide or polymer, for an effective photosintering process.

5. The thickness of coating of polyimide required to isolate thermal heat dissipation is approximately 1-50 microns.

6. The copper ink becoming a conductive film has been demonstrated on high thermal conductive material such as silicon wafer with both laser and photosintering.

7. Heat dissipation on high thermal conductive silicon wafers has been shown with a variety of polyimide thicknesses coated on a wafer. A low thermal conductive material can be used as a buffer layer to slow down heat dissipation and enhance the photograph or laser sintering.

8. Copper ink may be sintered well with polyimide coated on a silicon wafer with resistivity at 1×10⁻⁵ ohm-cm by photosintering and 4×10⁻⁶ ohm-cm by laser sintering.

9. The polyimide material may be not only utilized as a heat insulator on high thermal conductive substrates and enhance copper ink photograph and laser sintering effectiveness, but also applied to low melting temperature substrates as a heat insulator to protect from heat damage during a sintering process.

10. Polyimide layer and metal trace layer can be repeated several times as multilayer circuits.

11. Polyimide layer can be used as a dielectric material and incorporated as capacitors.

12. Nano-copper ink can be used at top layer conductor as a contact metal in two-dimensional and three-dimensional chip packaging applications.

Claims

1. A method comprising: coating a layer of material having a low thermal conductivity over a substrate;depositing a film of conductive ink over the layer of material having the low thermal conductivity; andsintering the film of conductive ink.

2. The method as recited in claim 1, further comprising: depositing a metal layer in a pattern on the substrate, wherein the layer of material having the low thermal conductivity is coated over the patterned metal layer and the substrate; andforming a via through the layer of material having the low thermal conductivity thereby exposing a portion of the patterned metal layer, wherein the depositing of the film of conductive ink includes depositing the film of conductive ink into the via to thereby coat the portion of the patterned metal layer with the film of conductive ink, wherein the film of conductive ink coating the portion of the patterned metal layer is also sintered.

3. The method as recited in claim 1, wherein the substrate has a thermal conductivity greater than the layer of material having the low thermal conductivity.

4. The method as recited in claim 2, wherein the film of conductive ink coated over the portion of the patterned metal layer does not dissipate as much energy from the sintering as the film of conductive ink coated over the layer of material having the low thermal conductivity.

5. The method as recited in claim 4, wherein the layer of material having the low thermal conductivity comprises a polymer.

6. The method as recited in claim 4, wherein the layer of material having the low thermal conductivity comprises polyimide.

7. The method as recited in claim 6, wherein the polyimide has a thickness of at least 50 microns.

8. The method as recited in claim 6, wherein the sintering is performed with a photosintering apparatus.

9. The method as recited in claim 6, wherein the sintering is performed with a laser sintering apparatus.

10. The method as recited in claim 6, wherein the substrate comprises silicon.

11. The method as recited in claim 6, wherein the substrate comprises a ceramic.

12. The method as recited in claim 6, wherein the film of conductive ink comprises copper nanoparticles.

13. The method as recited in claim 6, wherein the polyimide has a thickness of at least 5 microns.

14. The method as recited in claim 6, wherein the polyimide has a thickness of at least 2.3 microns.

15. The method as recited in claim 9, wherein the laser sintering apparatus comprises a solid state diode with an 830 nm wavelength and 800 mW power.

16. The method as recited in claim 15, wherein the solid state diode has a focus beam size of 15 microns in diameter.

17. A method comprising: coating a layer of material having a low thermal conductivity as a buffer layer over a substrate;depositing a film of conductive ink over the layer of material having the low thermal conductivity; andphotonic or laser sintering the film of conductive ink with a photo or laser sintering apparatus.

18. The method as recited in claim 17, further comprising: depositing a metal layer in a pattern on the substrate, wherein the layer of material having the low thermal conductivity is coated over the patterned metal layer and the substrate; andforming a via through the layer of material having the low thermal conductivity thereby exposing a portion of the patterned metal layer, wherein the depositing of the film of conductive ink includes depositing the film of conductive ink into the via to thereby coat the portion of the patterned metal layer with the film of conductive ink, wherein the film of conductive ink coating the portion of the patterned metal layer is also sintered.

19. The method as recited in claim 17, wherein the layer of material having the low thermal conductivity comprises a polymer or polyimide, wherein the substrate has a thermal conductivity greater than the layer of material having the low thermal conductivity.

20. The method as recited in claim 18, wherein the film of conductive ink coated over the portion of the patterned metal layer does not dissipate as much energy from the sintering as the film of conductive ink coated over the layer of material having the low thermal conductivity.