Copper Materials for Low Temperature Sintering

Abstract

In this review paper, recent studies on low-temperature sintering strategies of copper materials for conductive layer preparation have been summarized. Coinage metals, gold, silver, and copper have been used as materials for conductive inks and pastes for printed electronics. Copper is a highly electrically and thermally conductive material that can be used in electronic circuits and die-attach materials. Recently, copper-based inks/pastes have gained significant attention of researchers and industries as conductive materials. However, copper is readily oxidized under air, especially, at the nanoscale, and copper particles may catch fire because of the rapid oxidization. To overcome this issue, copper nanoparticles and fine particles are coated with organic molecules which act as insulators after sintering. Some interesting surface treatments or activation strategies have been investigated in this regard. In this paper, different perspectives on the applications of copper in conductive and die-attach materials have been presented.

1. Introduction

The production of electronic devices using the printing process which is called printed electronics has recently attracted attention due to low-cost and environment friendly products. Metal-conductive patterns are generally prepared using two methods.¹⁾ One is the traditional photoetching process, and the other is the printing process. A schematic of these methods is shown in Fig. 1, and a detailed schematic diagram of the photolithography etching process is shown in Fig. 2.²⁾ When using the traditional photoetching process, the unwanted parts of the metal are removed by etching. The disadvantages of this process include the use of harmful etching reagents, loss of metal raw materials, and long process time. In contrast, when using the printing process, metal inks containing metal micro/nanoparticles are printed on only the required parts. The printed metal inks are dried and sintered to produce conductive metal patterns. In this process, harmful etching reagents are not required, and the process time can be reduced. Therefore, printed electronics have recently attracted attention for the production of metal conductive patterns and various devices (Fig. 3).

Fig. 1

The schematic images of the preparation of metal conductive patterns.

Fig. 2

Schematic diagram of a photolithography technology process. (a) Apply photoresist on substrate surface; (b) prepare photoresist film via spin coating; (c) mask with transparent and opaque areas; (d) exposed photoresist film; (e) negative photoresist via negative tone development; (f) positive photoresist via positive tone development; (g) selective removal of substrate with the help of photoresist during etching process; (h) desired pattern on substrate after striping off photoresist. (Reproduced from Ref. 2. Copyright 2020. Published by The Royal Society of Chemistry)

Fig. 3

Applications of the printed electronics.

Ag nanoparticles (NPs) are widely used as conductive materials because of their low resistivity and high stability. Ag NPs are not easily oxidized in air; therefore, Ag NPs can be stored and sintered in air. The electrical and thermal conductivities of some metal materials are listed in Table 1.³⁾ Recently, because of the high cost and low electromigration resistance of Ag, other metal NPs have been investigated for the replacement of Ag NPs. Cu NPs are most actively investigated because of their low cost, high electromigration resistance, and high electrical and thermal conductivities. However, Cu NPs are not able to readily replace Ag NPs because of the low oxidation resistance of Cu NPs. Cu NPs are easily oxidized under ambient conditions, and the oxidation layers on the particle surface impede the sintering of Cu NPs to obtain high conductivity. Therefore, Cu NPs should be sintered under a reducing atmosphere such as hydrogen or in the presence of formic acid to remove the Cu oxide surface layers on the particles. Moreover, in order to prevent the oxidation of the Cu NPs during storage, the surface of the Cu NPs must be coated with organic compounds⁴⁾ or metal oxides.⁵⁾ Selection of the protective layers is also one of the crucial factors for the preparation of copper NPs. These protecting layers are not easily decomposed by heat treatment, therefore, the sintering temperature of Cu NPs usually increases. In order to overcome these disadvantages, many studies have been carried out so far for the development of the preparation and application of Cu NPs for printed electronics.

Table 1 Conductivities and thermal conductivities of various metals.³⁾

The required particle size and morphology are strongly depending on the application of the NPs. Therefore, many strategies have been proposed to prepare metal NPs of various sizes and morphologies. Preparation methods of metal NPs are divided into chemical and physical syntheses as shown in Fig. 4. Both processes are considered as “bottom-up” processes. Copper NPs have also been prepared by using these processes.

Fig. 4

Applications of the printed electronics.

Chemical processes include chemical reduction,^6–8) polyol synthesis,⁹⁾ microwave-assisted synthesis,^10–12) sonochemical synthesis,¹³⁾ and plasma-in-liquid processes,^14–16) etc. In chemical synthesis, metal precursors, such as metal salts, metal oxides or metal complexes, are reduced by reducing agents to produce metal atoms, and their aggregation is controlled to produce metal NPs. Physical processes include laser ablation,¹⁷⁾ matrix sputtering,¹⁸⁾ thermolysis,¹⁹⁾ and so on. Evaporation, sputtering or decomposition of bulk metals or metal precursors are used to produce metal atoms in the physical processes, and their aggregation is controlled to produce NPs. Generally speaking, physical processes have disadvantages such as the high cost of preparation instruments and high energy consumption. Therefore, chemical processes are widely utilized for the preparation of metal NPs.

In chemical reduction procedures, hydrazine is frequently used as the reducing reagent for Cu NPs.²⁰⁾ Ethanol, methanol or citric acid can be used for the preparation of noble metal nanoparticles, but they are too weak to reduce Cu ions to metallic Cu atoms. Instead of small alcohol compounds, polyol, which has a higher boiling temperature, can be used for Cu NP production. Sodium borohydrate (NaBH₄) is another potential candidate.

In this review, recent studies on the preparation and applications of copper NPs and fine particles for low-temperature sintering will be discussed.

2. Copper Nanoparticles and Fine Particles for Low Temperature Sintering

Copper nanomaterials such as NPs, nanowires (NWs), and nanocrystals as well as copper fine particles have been prepared by various chemical and physical procedures.^21–23) Among these nanomaterials, Cu NPs with particle size under 10 nm exhibit low melting temperatures, as shown in Fig. 5.²⁴⁾ Therefore, the use of such nanomaterials is considered to be effective for low-temperature sintering. Cu NPs,²⁵⁾ Cu NWs,²⁶⁾ Cu flakes,²⁷⁾ CuO NPs,^28,29) Cu complexes,^30,31) and their mixtures have been investigated as Cu precursors for low-temperature sintering.³²⁾

Fig. 5

Predicted melting temperature of Cu NPs with various particle sizes. (Reproduced from Ref. 24. Published in 2009 by Springer-Nature)

Mixing of different-sized particles is effective in increasing the packing density and decreasing the resistivity of the sintered films. Moreover, microparticles have smaller surface areas than the NPs. Therefore, the addition of microparticles increases the oxidation resistivity. CuO NPs and Cu complexes have also been investigated as Cu precursors instead of Cu metal NPs because they are stable in air, unlike Cu metal NPs. Though they have to be reduced by reducing agents or gases during sintering, it is an advantage that they do not need anti-oxidizing treatment.

The sintering properties of Cu NPs (30–65 nm) have been reported by Lee et al.²⁵⁾ Polyvinylpyrrolidone (PVP) is a very effective polymer for metal NPs because it can be uniformly dissolved in water and various organic solvents. Therefore, Cu NPs can be prepared in aqueous media and the obtained NPs can be re-dispersed in organic solvents without other additives. Cu nanoink containing 30% Cu NPs and 70% 2-(2-butoxyethoxy)ethanol as a disperse medium has been prepared. After printing, the sample was sintered at 200°C for 1 h under a reductive atmosphere (nitrogen gas bubbled through formic acid), and the resistivity of the sintered film was 3.6 × 10⁻⁶ Ω·cm, which was only 2.2 times higher than that of bulk copper.

Polyethyelene imine (PEI) is also used for protecting copper NPs, as proposed by Tenhu et al.³³⁾ In the presence of PEI, copper chloride was reduced by NaBH₄. The obtained particles were metallic copper, as shown in Fig. 6. Some oxidation to Cu₂O was observed after preparation. From the wide-angle X-ray scattering (WAXS) results it was confirmed that during sintering under helium, the grain size became considerably larger at temperatures higher than 200°C. During the heating experiment, the Cu₂O signal disappeared probably owing to the reductive ability of PEI. Crystal growth at high temperatures explains the increased conductivity after sintering. The resistivity obtained after sintering at 250°C was 1.31 × 10⁻² Ω·cm.

Fig. 6

Polyethyleneimine-coated Cu NPs and their SAED pattern corresponding to metallic copper. (Reproduced with permission from Ref. 33. Copyright 2009 by American Chemical Society)

Hexanoicacid-stabilized copper nanoparticles can also be used for low-temperature sintering. They were prepared from CuO by chemical reduction with hydrazine. The particle size was unique (average diameter: 82 nm) and they could be re-dispersed readily into dipropylene glycol (DPG). Bead milling can be applied to prepare copper pastes. Pre-addition of DPG for bead milling resulted in more stable Cu NP pastes (Fig. 7). The obtained Cu NP pastes milled with ceramic beads (30 µm) were very stable and could be printed to obtain a very thin particle layer for low-temperature sintering. Sintering of the printed Cu NP layer at 200°C for 1 h under a 3%H₂–N₂ gas flow gave a resistivity of 6.62 × 10⁻⁶ Ω·cm (Fig. 8).

Fig. 7

SEM images of the printed copper samples (a) without and (b) with pre-addition of DPG. (Reproduced with permission from Ref. 34. Copyright 2020 Elsevier)

Fig. 8

SEM images of (a) Cross-sectional and (b), (c) surface of the copper layer sintered at 200°C. (Reproduced with permission from Ref. 34. Copyright 2020 Elsevier.)

3. Light and Laser Sintering of Copper Particles

3.1 Flash-light or intense pulsed light (IPL) sintering

Flash-light sintering or intense pulsed light (IPL) sintering is effective for copper because the heating time is quite limited (milliseconds) (Fig. 9).^{22,24,26,27,35–38)} While using flash-light or IPL sintering, no inert gas or reductive gas is required. Moreover, the polymer substrate is not damaged. A mixed ink of two types of Cu nanomaterials with different morphologies was proposed by Joo, et al.²⁶⁾ The ink is consisted of a mixture of Cu wires with the diameter of 150 ± 50 nm and a length of 1–2 µm and Cu NPs with a diameter of 20–50 nm. The lowest resistivity of 22.77 × 10⁻⁶ Ω·cm was achieved with a Cu NW/NP mixed ink with 5 wt% Cu NWs by flash-light sintering on a PI film. The resistivity of this film prepared using Cu NW/NP mixed ink was less than one-fourth of that of Cu NP ink (9.401 × 10⁻⁵ Ω·cm) and the Cu NWs ink (1.0415 × 10⁻⁴ Ω·cm). Figure 10 shows SEM images of the sintered Cu NW/NP ink films obtained when (a) 0, (b) 1, (c) 3, (d) 5, and (e) 100 wt% of Cu NWs were used. Mixing Cu materials with different sizes and morphologies led to low porosity of the sintered films. The authors also found that the addition of Cu NWs not only resulted in the decrease of resistivity, but also ensured reliability under repeatable mechanical fatigue (Fig. 11).

Fig. 9

Schematic representation the IPL sintering of the ink films to produce conductive copper films. (Reproduced with permission from Ref. 36. Copyright 2013 by American Chemical Society)

Fig. 10

SEM images of flash light-sintered Cu NW/NP ink film at ca. (a) 0, (b) 1, (c) 3, (d) 5, and (e) 100 wt% of Cu NWs (irradiation energy: 12.5 J·cm⁻², pulse duration: 10 ms, pulse number: 1). (Reproduced with permission from Ref. 26. Copyright 2015 by American Chemical Society)

Fig. 11

Outer bending fatigue test results about bending radius of (a) r = 7 mm, (b) r = 10 mm, and (c) r = 15 mm. (insets) Photographs of the specimen for outer bending fatigue test with respect to the bending radius. SEM images of the fatigue-tested Cu NW/NP ink film: the 1000 cycles of outer bending fatigue test results of (d) 0, (e) 1, (f) 3, (g) 5, and (h) 100 wt% of Cu NWs. Higher concentration of Cu NWs gave better results of the bending fatigue test. (Reproduced with permission from Ref. 26. Copyright 2015 by American Chemical Society)

Tam et al. reported the sintering properties of the mixture of flat Cu flakes (9.3 µm in diameter, 1.2 µm in thickness) and Cu NPs (60.8 nm).²⁷⁾ The lowest resistivity (2.8 × 10⁻⁵ Ω·cm) was obtained when the Cu flake/NP paste with 20 wt% Cu flakes was sintered by intense-pulsed light for a few milliseconds on a PET film. The sintered film with 20 wt% Cu flakes showed the lowest porosity. As a result, the resistivity of the Cu film decreased.

Sintering of PVP-coated Cu NPs with a diameter of approximately 50 nm was also carried out under air.³⁷⁾ The absorption of Cu NPs on the substrate surface shows maxima at around 360 and 800 nm. Cu NPs absorb the IPL emission light, and the plasmon resonance after light absorption leads to heat generation. These Cu NPs were covered by a Cu₂O thin shell (Fig. 12) which was reduced by IPL in the presence of ethylene glycol (solvent-reductant)³⁸⁾ and fused to forma connection between the large copper particles. In order to avoid the formation of cracks during the IPL sintering process, a multiple-pulse sintering process was applied to reduce the thermal yield stress across the sintered copper film. A series of pulses were applied with a fixed off time of 2 s between individual pulses (Fig. 13). A series of pulses irradiation yielded higher conductivities.

Fig. 12

(a) UV-vis spectrum of the synthesized copper nanoparticles (CuNPs) and (b) the XRD patterns before and after intense pulsed light (IPL) sintering. An IPL energy density of 3.23 J/cm₂ was used in the sintering. (Reproduced from Ref. 37. Published in 2019 by MDPI)

Fig. 13

(a) Schematic diagram for multiple pulse IPL. The total area of pulses is defined as the total energy. (b) Resistivity of copper film sintered by multiple pulses. An off time of 2 s is used between IPL pulses. (Reproduced from Ref. 37. Published in 2019 by MDPI)

3.2 Laser sintering

The sintering properties of CuO NPs by laser irradiation was reported by Kang et al.²⁸⁾ CuO NPs are more stable in air and cheaper than Cu NPs. CuO NPs (200–300 nm) ink was prepared using ethylene glycol and PVP as the reducing agent and dispersant, respectively. Figure 14 shows a schematic illustration of the CuO NP sintering process. CuO NPs in liquid on the surface of the substrate were reduced to Cu NPs by laser irradiation, and which were sintered to generate Cu film or lines. PVP can be considered to control the formation of Cu NPs, and ethylene glycol can reduce Cu²⁺ to Cu at higher temperatures. Figures 15(a) and (b) show the photograph and microscopic image, and SEM image of the Cu pattern sintered on a PI film, respectively. Voids were detected in the Cu patterns; however the resistivity of the obtained Cu patterns was considerably low (3.1 × 10⁻⁵ Ω cm).

Fig. 14

(a) Schematics of the proposed process: Conversion of CuO NPs into Cu film by photochemical reduction and photothermal agglomeration. (Inset) Transmission electron microscope (TEM) image of CuO NP. (b) Schematics of the experimental setup. (Reproduced with permission from Ref. 28. Copyright 2011 American Chemical Society)

Fig. 15

(a) Photograph and microscopic image of copper electrodes on a polyimide film. (b) SEM image of the Cu pattern of approximately 11 µm using a 10 µm spot diameter on a polyimide substrate at a pulse energy of 0.3 µJ. (c) Temperature distribution image of the heater using an infrared camera at an electrical input of 30 V and 0.15 A. (Inset) Schematic illustration of the microheater. (Reproduced with permission from Ref. 28. Copyright 2011 American Chemical Society)

Laser sintering was also applied to polymer-stabilized metallic copper fine particles with a diameter of 100–200 nm (Fig. 16).³⁹⁾ In this case, the atmospheric condition could be controlled by injecting inert gas. When sintering was carried out in air, the obtained copper layer was not conductive. However, when laser sintering was carried out under 4%H₂–Ar gas flow or Ar gas flow, the resistivity decreased dramatically to 1.05 × 10⁻⁵ Ω·cm or 2.45 × 10⁻⁵ Ω·cm, respectively. On the other hand, thermal sintering under Ar gas flow at 300°C gave a resistivity of 7500 Ω·cm. The SEM image of the sintered surface layer (4%H₂–Ar flow) is collected in Fig. 17. Melting of the surface of the copper fine particles by laser irradiation can be observed. It has been reported that the melting temperature of the surface of a copper nanoparticle is remarkably lower than that of the core.⁴⁰⁾ These results indicate that instant local heating by laser irradiation may result in the decomposition of surface polymers and melting of the surface of the particles.

Fig. 16

Schematic illustration of experimental setup of laser sintering. (Reproduced with permission from Ref. 39. Copyright 2014 The Japan Society for Applied Physics)

Fig. 17

SEM images of copper fine particle coating film after laser sintering under a hydrogen/argon (4/96) gas mixture flow (10 mL/min). (Reproduced with permission from Ref. 39. Copyright 2014 The Japan Society for Applied Physics)

4. Application of Metal-Organic Decomposition (MOD) Copper Ink

Copper(II) formate (CuF)-isopopanol amine (IPA) complex (CuF-IPA) was used as a self-reducible copper ion MOD ink for conductive copper layers by thermal decomposition.⁴¹⁾ The chemical structure of CuF-IPA complex is shown in Fig. 18. The CuF-IPA complexes generated large bubbles during heating by the decomposition of CuF. Thus, the surface morphologies of the prepared Cu thin layers were widely disturbed. In order to control the decomposition condition of CuF-IPA, a long alkyl amine, octylamine (OA) was introduced into the MOD ink. OA addition reduced the surface tension and dewetting problems. The size and number of bubbles generated during thermal decomposition were greatly reduced when the OA/IPA ratio was greater than 0.5. Consequently, the obtained film became smooth and connected particles could be observed in the surface SEM image (Fig. 19). The addition of polyvinylpyrrolidone (PVP) to the ink introduced a much better adhesion on the glass against adhesion tape tests. The decomposition of this MOD ink can be carried out even at temperatures as low as 110°C. A low resistivity of 2 × 10⁻⁵ Ω·cm was achieved by heating the ink at 140°C for 5 min under nitrogen (Fig. 20).

Fig. 18

Molecular structure of copper(II) formate (CuF)-isopropanol amine (IPA) complex.

Fig. 19

Copper thin film formation by heating (a) CuF–IPA and (b) CuF–IPA-OA ink on glass at 120°C for 40 minutes. (c) and (d) show the SEM images of (a) and (b), respectively. In (a) and (b), both liquid films have the same total volume of 63 µL. (Reproduced from Ref. 41. Published in 2017 by The Royal Society of Chemistry)

Fig. 20

Variation of resistivity of copper thin films with calcination time at various calcination temperatures. (Reproduced from Ref. 41. Published in 2017 by The Royal Society of Chemistry)

Our group has also proposed use of copper decomposable complexes as materials for conductive layer formation. The sintering properties of a mixture of Cu sub-micron particles and CuF-alkanolamine complex were studied.³⁰⁾ The size of the submicron sized Cu particles was 0.8 µm. Figure 21(c) shows a cross-sectional SEM image of the Cu film sintered at 100°C for 1 h under a nitrogen atmosphere. The resistivity of the sintered Cu film was 9.0 × 10⁻⁴ Ω cm. CuF-IPA complexes were decomposed under 100°C to generate Cu NPs which then connected the Cu submicron particles. In addition, the packing density of the sintered film was significantly increased by adding Cu submicron particles to the MOD ink. Therefore, a low resistivity was achieved by the low-temperature sintering.

Fig. 21

(a), (b) Photographs of CuF–IPA–Cu ink after printing and after sintering at 100°C under nitrogen for 1 h, respectively. (c) Cross sectional SEM image of the copper film shown in (b). The CuF–IPA–Cu ink was prepared using 0.8 µm copper particles and a 1:1:6 molar ratio of CuF to IPA to copper particles. (Reproduced from Ref. 30. Published in 2016 by The Royal Society of Chemistry)

A mixture of a Cu acetate-based MOD ink and Cu NPs is also proposed by Kawasaki et al.⁴²⁾ Copper acetate was mixed with 1-amino-2-propanol (IPA) to form the MOD complex (Cu-IPA). Micron-sized Cu particles were introduced into the ink. Various mixed inks were prepared for low-temperature sintering. The lowest resistivity was archieved by sintering at 180°C for 1 h using a particle:MOD = 3:1 (w/w) mixed ink (Fig. 22). Higher temperature sintering gave more stable conductive copper layers.

Fig. 22

Resistivity of Cu layers prepared from a copper particle:MOD = 3:1 (w/w) mixed ink. (a) The dependence with sintering temperature (1 h sintering) and (b) with sintering time (180°C sintering). (Reproduced from Ref. 42. Published in 2016 by Lifescience Global)

5. Surface Treatments of Cu NPs

The surfaces of the Cu NPs are readily oxidized and convered to copper oxides such as Cu₂O and CuO which hinder the sintering of Cu NPs. Therefore, anti-oxidizing treatments of Cu NPs and reduction of Cu oxides are necessary to decrease the resistivity of sintered Cu films. When Cu NPs are coated with a thick layer, the antioxidation properties are improved; however, the coating layer inhibits the sintering of Cu NPs. In this case, Cu NPs need to be sintered at high temperatures to remove the coating layer. Surface treatments of Cu NPs have been extensively studied in recent years, however, it is difficult to achieve both high antioxidation properties of Cu NPs and low resistivity of Cu sintered film.⁴³⁾ For example, polymers,⁴⁴⁾ decomposable polymers,⁴⁵⁾ organic compounds,^46,47) graphene,⁴⁸⁾ and silver⁴⁹⁾ have been researched as surface treatment agents for Cu NPs.

5.1 Two step low temperature sintering of copper particles

Two-step sintering has also been proposed to obtain conductive copper layers from polymer-protected copper NPs and fine particles. Our group proposed gelatin-protected copper NPs and fine particles prepared by hydrazine reduction of micron-sized CuO powders. The obtained particles were highly stable and could be redispersed into various organic liquid media. Cu fine particles with an average particle size of 130 nm were prepared by chemical reduction in the presence of gelatin as a stabilizing agent. Two-step sintering containing oxidation and reduction step was carried out.⁴⁴⁾ The resistivity of the obtained sintered Cu film was 8.2 × 10⁻⁶ Ω·cm when oxidized at 200°C in air for 4 h and reduced at 200°C in 3% H₂ in N₂ gas for 3 h. Figure 23 shows the schematic image of the sintering process of the Cu fine particles. Cu₂O NPs generated by the oxidative annealing promoted the sintering of Cu fine particles in the reductive annealing step. A similar process could be performed even at 150°C to obtain conductive copper layer.⁵⁰⁾ These results strongly suggest that the lowest sintering temperature of polymer-coated Cu NPs is the glass transition temperature of the polymer on the Cu surface.

Fig. 23

Schematic of the necking and connection formation between the copper fine particles by our two-step annealing process. (Reproduced from Ref. 44. Published in 2015 by The Royal Society of Chemistry)

Two-step low-temperature sintering properties of octylamine-stabilized Cu fine particles were also reported.⁴⁶⁾ Figure 24 shows a schematic image for the preparation of octylamine-stabilized Cu fine particles. The average diameter of these Cu fine particles was 280 nm. The X-ray diffraction (XRD) patterns of the Cu fine particles showed no Cu oxide peak because of the anti-oxidation property of the octylamine layer. Two-step sintering was performed for the low temperature sintering (Fig. 25). The resistivity of the obtained sintered Cu film was 5.6 × 10⁻⁶ Ω·cm when oxidized at 300°C in air for 4 h and reduced at 300°C in 3% H₂ in N₂ gas for 3 h. In the oxidative preheating step, octylamine was removed, and convex surfaces, nanorods, and Cu oxide NPs were generated on the Cu fine particles.

Fig. 24

Schematic image of the synthesis procedure of octylamine stabilized Cu fine particles. (Reproduced with permission from Ref. 46. Copyright 2015 The Royal Society of Chemistry)

Fig. 25

Schematic illustration of the oxidation preheating process which can be used for generating convex surfaces, nanorods or nanoparticles and finally promote the sintering of particles. (Reproduced with permission from Ref. 46. Copyright 2015 The Royal Society of Chemistry)

In the reductive sintering step, these Cu oxides were reduced to Cu metal, and the sintering of Cu particles was promoted. During the reduction from Cu₂O to Cu, the crystal structure changed dramatically (from cuprite to fcc). An effective diffusion of Cu atoms occurs during the deformation of the crystal structure, which results in the formation of a wide necking structure.

5.2 Reductive molecule coating on Cu NPs

As discussed in the MOD ink section, IPA was used as the ligand. It works as the reducing reagent for Cu²⁺ at higher temperatures; therefore, the CuF-IPA MOD ink as material for low-temperature sintering of copper. With this molecule, sub-10 nm Cu NPs was proposed.⁴⁷⁾ Figure 26 shows the TEM image and XRD pattern of the obtained NPs. The particle size was ca. 3.5 nm. The ink was printed and sintered at 150°C under nitrogen atmosphere. The resistivity of the obtained Cu layer was as low as 3.0 × 10⁻⁵ Ω·cm.

Fig. 26

(a) TEM image of IPA-Cu NPs. The high-magnification image is shown as the inset and illustrates that the crystal lattice fringes are 0.2 nm apart, which agrees with the d value of the (111) planes of the metallic Cu crystal. (b) XRD pattern for purified IPA-Cu NPs. (Reproduced with permission from Ref. 47. Copyright 2015 American Chemical Society)

5.3 Decomposable polymer-coating for low temperature sintering

Decomposable polymers are also good coating and stabilizing molecules for the low-temperature sintering of Cu NPs and fine particles. Poly(propylenecarbonate) (PPC) can be decomposed into smaller molecules such as amines or amino alcohols by aminolysis. The reaction of PPC is illustrated in Fig. 27.⁴⁵⁾ The particle size of the PPC-coated Cu fine particles was ca. 150 nm. The CuF-IPA complex was added as an additive to the Cu fine particle paste. As a result, PPC was effectively decomposed by IPA, and the particles could be sintered at 100°C for 1 h under nitrogen gas flow to obtain a highly conductive copper layer (8.8 × 10⁻⁵ Ω·cm).

Fig. 27

The aminolysis of PPC with IPA. (Reproduced with permission from Ref. 45. Copyright 2016 The Royal Society of Chemistry)

Another decomposable polymer proposed for Cu fine particles was poly-1,4-butanediol-divinylether (BDVE). Particles were prepared by chemical reduction of CuO in the presence of BDVE.⁵¹⁾ This polymer can be employed for the anti-oxidation of Cu fine particles, and readily decomposed in the presence of protons. Therefore, the BDVE-stabilized Cu fine particles could be sintered at 150°C in the presence of protons (Fig. 28).

Fig. 28

Schematic illustration of sintering of BVDE-coated Cu fine particles in the presence of protons. (Reproduced from Ref. 51. Published in 2015 by The Royal Society of Chemistry)

5.4 Graphene stabilized Cu NPs

The sintering properties of graphene-stabilized Cu NPs prepared by reducing flame synthesis was reported by Luechinger et al.⁴⁸⁾ Cu NPs were coated by graphene bi- or tri-layers with a thickness of only 3 nm. The obtained Cu NPs did not oxidize up to 165°C in air, thus are more stable than polymer-stabilized Cu particles.⁵²⁾ However, the resistivity of the Cu film obtained by sintering at 120°C for 2 h in air was higher by five orders of magnitude than that of bulk copper. This suggests that the graphene layers on the particle surface strongly inhibit the sintering of the Cu NPs.

In-situ TEM observations of gelatin-stabilized copper particles also indicated similar situation. Hydrazine reduction of CuCl₂ in water in the presence of gelatin at room temperature.⁵³⁾ A large amount of gelatin was added into the preparation solution, and uniform secondary aggregated particles with a diameter of ca. 1 µm consisted of 5–7 nm sized NPs covered by a thin gelatin layer. When these secondary particles were heated under high vacuum, no sintering, fusion or deformation of the small nanoparticles was observed. Small NPs can be observed at temperatures above 400°C. Under high vacuum, the gelatin layer changed to a carbon layer and did not decompose even at high temperatures. Therefore, no deformation of the NPs occurred.

5.5 Reactive sintering with amine-coated Cu NPs

Amine-coated Cu NPs were also considered for low-temperature reactive sintering.⁵⁴⁾ Oleylamine (OAM)-coated Cu NPs were cleaned by washing with formic acid, which introduced the desorption of OAM (Fig. 29). Subsequently, further oxidation was observed, which strongly suggested the desorption of OAM during washing. Sintering was then carried out by immersing the washed films in a NaBH₄ alkaline solution at pH = 12. The resistivities obtained after NaBH₄ treatment are collected in Fig. 30. A longer immersion period in NaBH₄ resulted in smaller resistivity. These results indicate that the reduction of the Cu₂O surface layer on Cu NPs introduces the formation on neckings and finally, the Cu layers could be sintered at room temperature. This is similar to the two-step sintering, discussed above.

Fig. 29

Schematic illustration of the chemical sintering mechanism. (Reproduced from the Ref. 54. Copyright 2020 by American Chemical Society)

Fig. 30

Final resistivity of the metal layer treated by HCOOH solution, followed by 0.75 wt% NaBH₄ solution for different time periods. (a) 5 vol% HCOOH in ethanol, (b) 10 vol% HCOOH in ethanol, (c) 20 vol% HCOOH in ethanol, (d) 5 vol% HCOOH in methanol, and (e) 10 vol% HCOOH in methanol. (Reproduced from the Ref. 54. Copyright 2020 by American Chemical Society)

5.6 Noble metal coating of copper particles

Noble metal coating of copper NPs, that is, Cu-noble metal core-shell particles, have also been proposed for low-temperature sintering. Lee et al. reported the sintering property of Cu–Ag core–shell NPs.⁵⁵⁾ In the first step, Cu NPs with a particle size of 13.50 nm were prepared by the thermal decomposition method. In the second step, Ag shells were formed on the surface of the Cu NPs using the galvanic displacement method. The measured composition was Cu:Ag = 77.5:22.5 (mol/mol). The Cu–Ag core–shell NPs were not oxidized after 2 months of storage as dry powder under ambient conditions. The resistivity of the film was 5.7 × 10⁻⁵ Ω·cm when sintered at 200°C for 1 h under a nitrogen atmosphere.

Copper nanowires (Cu NWs) were proposed as a promising alternative to transparent oxide electrodes with high conductivity and transparency by the group of Li and Suganuma.⁵⁶⁾ A facile adsorption and decomposition process was developed for galvanic replacement-free and large-scale preparation procedures. Adsorption of the Ag-amine complex ([Ag(NH₂R)₂]⁺) as a silver source on the Cu NW surface generated a Cu@Ag-amine complex core–shell structure (Fig. 31). The Ag-amine complex could be easily decomposed into a pure Ag shell through thermal annealing. By adjusting the concentration of Ag-amine in Cu NW dispersions, Cu@Ag core–shell nanowires with different thicknesses of silver shells were easily obtained. The obtained core–shell nanowires exhibited high stability for at least 500 h at high temperature (140°C) and high humidity (85°C, 85% RH) because of the protection of the Ag shell. More importantly, the conductivity and transparency of Cu@Ag NW-based conductors were similar to those of pure Cu NWs. The large-scale and facile synthesis of Cu@Ag core–shell NWs provides a new method for preparing stable metallic core–shell nanowires.

Fig. 31

SEM images of (a) CuNWs, (b) Cu@Ag-amine core–shell structure, and (c) Cu@Ag core–shell nanowires. TEM image of the cross section for Cu@Ag core–shell nanowires (d), and the corresponding elemental mapping images (e)–(g). The samples all with 20 wt% of silver. (Reproduced from the Ref. 56. Copyright 2019 by American Chemical Society)

6. Conclusions

This review summarizes recent research progress on sintering and bonding techniques for copper nanoparticles and fine particles for electronic conductive layers and electronic packaging applications. The use of highly conductive Cu layers as bulk Cu by sintering Cu NPs is still very challenging. Various Cu NPs have been proposed, which are stabilized by organic molecules, such as polymers or surfactants. These particles can be sintered at low temperatures to obtain highly conductive copper layers. Two-step sintering, intense pulsed light irradiation, and degradable polymers for coating are promising techniques for low-temperature sintering. MOD is another candidate material for Cu-based ink/paste. Use of inks/pastes for low-temperature sintering is also suitable for die-attach materials, especially for high-power semiconductor devices. Further studies of copper-based ink/pastes is expected to be beneficial for the development of next-generation electronics.

Acknowledgments

This work was partially supported by Hokkaido University. TY thanks the financial supports from a Grant-in-Aid for Challenging Research (Exploratory) (19K22094 to TY), and the Fund for the Promotion of Joint International Research (Fostering Joint International Research (B)) (18KK0159 to TY and MTN) from JSPS, Japan. Authors also acknowledge that the Adaptable and Seamless Technology transfer Program through targetdriven R&D (A-STEP) was funded by the Japan Science and Technology Agency (JST). Partial support by the Management Expenses Grants for Network Joint Research Centre for Materials and Devices (20211067, 20211242, 20201248, and 20191253) is gratefully acknowledged. Authors thank Dr. Y. Ishida (Hokkaido University), Dr. T. Tokunaga (Nagoya University) and Mr. H. Tsukamoto (Hokkaido University) for their fruitful discussions and experimental assistance. Fruitful discussions with Prof. T. Sugahara (Osaka University) are also acknowledged. A part of this review was included in the PhD thesis submitted by Dr. M. Nishimoto to Hokkaido University in March 2019.

REFERENCES

• 1)   W.  Wu: Nanoscale 9 (2017) 7342–7372. doi:10.1039/C7NR01604B
• 2)   C.  Luo,  C.  Xu,  L.  Lv,  H.  Li,  X.  Huang and  W.  Liu: RSC Adv. 10 (2020) 8385–8395. doi:10.1039/C9RA08977B
• 3)   S.  Tanaka and  H.  Nishimura: Butsuri Kyoiku 50 (2002) 237–242. doi:10.20653/pesj.50.4_237
• 4)   V.  Engels,  F.  Benaskar,  D.A.  Jefferson,  B.F.J.  Johnson and  A.E.H.  Wheatley: Dalton Trans. 39 (2010) 6496–6502. doi:10.1039/c0dt00134a
• 5)   Y.  Kobayashi and  T.  Sakuraba: Colloids Surf. A 317 (2008) 756–759. doi:10.1016/j.colsurfa.2007.11.009
• 6)   N.T.K.  Thanh,  N.  Maclean and  S.  Mahiddine: Chem. Rev. 114 (2014) 7610–7630. doi:10.1021/cr400544s
• 7)   S.  Diodati,  P.  Dolcet,  M.  Casarin and  S.  Gross: Chem. Rev. 115 (2015) 11449–11502. doi:10.1021/acs.chemrev.5b00275
• 8)   J.A.  Dahl,  B.L.S.  Maddux and  J.E.  Hutchison: Chem. Rev. 107 (2007) 2228–2269. doi:10.1021/cr050943k
• 9)   F.  Fiévet,  S.  Ammar-Merah,  R.  Brayner,  F.  Chau,  M.  Giraud,  F.  Mammeri,  J.  Peron,  J.-Y.  Piquemal,  L.  Sicarda and  G.  Viau: Chem. Soc. Rev. 47 (2018) 5187–5233. doi:10.1039/C7CS00777A
• 10)   M.N.  Nadagouda,  T.F.  Speth and  R.S.  Varma: Acc. Chem. Res. 44 (2011) 469–478. doi:10.1021/ar1001457
• 11)   M.B.  Gawande,  S.N.  Shelke,  R.  Zboril and  R.S.  Varma: Acc. Chem. Res. 47 (2014) 1338–1348. doi:10.1021/ar400309b
• 12)   Y.-J.  Zhu and  F.  Chen: Chem. Rev. 114 (2014) 6462–6555. doi:10.1021/cr400366s
• 13)   C.-H.  Su,  P.-L.  Wu and  C.-S.  Yeh: J. Phys. Chem. B 107 (2003) 14240–14243. doi:10.1021/jp035451v
• 14)   G.  Saito and  T.  Akiyama: J. Nanomater. 2015 (2015) 123696. doi:10.1155/2015/123696
• 15)   Y.  Mizukoshi,  F.  Hori and  K.  Okitsu: Jpn. J. Appl. Phys. 57 (2018) 0102A5. doi:10.7567/JJAP.57.0102A5
• 16)   T.  Narushima,  T.  Yoshida,  H.  Miyazaki,  I.  Suga,  S.  Sato and  T.  Yonezawa: J. Jpn. Inst. Metal. Mater. 76 (2012) 229–233. doi:10.2320/jinstmet.76.229
• 17)   D.  Zhang,  B.  Gökce and  S.  Barcikowski: Chem. Rev. 117 (2017) 3990–4103. doi:10.1021/acs.chemrev.6b00468
• 18)   Y.  Ishida,  R.D.  Corpuz and  T.  Yonezawa: Acc. Chem. Res. 50 (2017) 2986–2995. doi:10.1021/acs.accounts.7b00470
• 19)   M.  Nakamoto,  M.  Yamamoto and  M.  Fukusumi: Chem. Commun. (2002) 1622–1623. doi:10.1039/b203736j
• 20)   M.  Tomonari,  K.  Ida,  H.  Yamashita and  T.  Yonezawa: J. Nanosci. Nanotechnol. 8 (2008) 2468–2471. doi:10.1166/jnn.2008.237
• 21)   M.B.  Gawande,  A.  Goswami,  F.-X.  Felpin,  T.  Asefa,  X.  Huang,  R.  Silva,  X.  Zou,  R.  Zboril and  R.S.  Varma: Chem. Rev. 116 (2016) 3722–3811. doi:10.1021/acs.chemrev.5b00482
• 22)   M.I.  Din and  R.  Rehan: Anal. Lett. 50 (2017) 50–62. doi:10.1080/00032719.2016.1172081
• 23)   K.W.  Shah and  Y.  Lu: Constr. Build. Mater. 180 (2018) 544–578. doi:10.1016/j.conbuildmat.2018.05.159
• 24)   H.-S.  Kim,  S.R.  Dhage,  D.-E.  Shim and  H.T.  Hahn: Appl. Phys. A 97 (2009) 791–798. doi:10.1007/s00339-009-5360-6
• 25)   Y.  Lee,  J.  Choi,  K.J.  Lee,  N.E.  Stott and  D.  Kim: Nanotechnology 19 (2008) 415604. doi:10.1088/0957-4484/19/41/415604
• 26)   S.-J.  Joo,  S.-H.  Park,  C.-J.  Moon and  H.-S.  Kim: ACS Appl. Mater. Interfaces 7 (2015) 5674–5684. doi:10.1021/am506765p
• 27)   S.K.  Tam,  K.Y.  Fung and  K.M.  Ng: J. Mater. Sci. 51 (2016) 1914–1922. doi:10.1007/s10853-015-9498-7
• 28)   B.  Kang,  S.  Han,  J.  Kim,  S.  Ko and  M.  Yang: J. Phys. Chem. C 115 (2011) 23664–23670. doi:10.1021/jp205281a
• 29)   M.K.  Rahman,  Z.  Lu and  K.-S.  Kwon: AIP Adv. 8 (2018) 095008. doi:10.1063/1.5047562
• 30)   T.  Yonezawa,  H.  Tsukamoto,  Y.  Yong,  M.T.  Nguyen and  M.  Matsubara: RSC Adv. 6 (2016) 12048–12052. doi:10.1039/C5RA25058G
• 31)   D.-H.  Shin,  S.  Woo,  H.  Yem,  M.  Cha,  S.  Cho,  M.  Kang,  S.  Jeong,  Y.  Kim,  K.  Kang and  Y.  Piao: ACS Appl. Mater. Interfaces 6 (2014) 3312–3319. doi:10.1021/am4036306
• 32)   K.V.  Abhinav,  R.V.  Krishna Rao,  P.S.  Karthik and  S.P.  Singh: RSC Adv. 5 (2015) 63985–64030. doi:10.1039/C5RA08205F
• 33)   P.  Pulkkinen,  J.  Shan,  K.  Leppänen,  A.  Känsäkoski,  A.  Laiho,  M.  Järn and  H.  Tenhu: ACS Appl. Mater. Interfaces 1 (2009) 519–525. doi:10.1021/am800177d
• 34)   S.  Liu,  R.  Tokura,  M.T.  Nguyen and  T.  Yonezawa: Adv. Powder Technol. 31 (2020) 4570–4575. doi:10.1016/j.apt.2020.10.004
• 35)   S.  Wünscher,  R.  Abbel,  J.  Perelaer and  U.S.  Schubert: J. Mater. Chem. C 2 (2014) 10232–10261. doi:10.1039/C4TC01820F
• 36)   R.  Dharmadasa,  M.  Jha,  D.A.  Amos and  T.  Druffel: ACS Appl. Mater. Interfaces 5 (2013) 13227–13234. doi:10.1021/am404226e
• 37)   W.-Y.  Chung,  Y.-C.  Lai,  T.  Yonezawa and  Y.-C.  Liao: Nanomaterials 9 (2019) 1071. doi:10.3390/nano9081071
• 38)   G.L.  Draper,  R.  Dharmadasa,  M.E.  Staats,  B.W.  Lavery and  T.  Druffel: ACS Appl. Mater. Interfaces 7 (2015) 16478–16485. doi:10.1021/acsami.5b03854
• 39)   G.  Qin,  A.  Watanabe,  H.  Tsukamoto and  T.  Yonezawa: Jpn. J. Appl. Phys. 53 (2014) 096501. doi:10.7567/JJAP.53.096501
• 40)   O.A.  Yeshchenko,  I.M.  Dmitruk,  A.A.  Alexeenko and  A.M.  Dmytruk: Phys. Rev. B 75 (2007) 085434. doi:10.1103/PhysRevB.75.085434
• 41)   K.-M.  Huang,  H.  Tsukamoto,  Y.  Yong,  H.-L.  Chiu,  M.T.  Nguyen,  T.  Yonezawa and  Y.-C.  Liao: RSC Adv. 7 (2017) 25095–25100. doi:10.1039/C7RA01005B
• 42)   Y.  Kawaguchi,  R.  Arakawa and  H.  Kawasaki: J. Coat. Sci. Technol. 3 (2016) 56–61. doi:10.6000/2369-3355.2016.03.02.2
• 43)   S.  Magdassi,  M.  Grouchko and  A.  Kamyshny: Materials 3 (2010) 4626–4638. doi:10.3390/ma3094626
• 44)   T.  Yonezawa,  H.  Tsukamoto and  M.  Matsubara: RSC Adv. 5 (2015) 61290–61297. doi:10.1039/C5RA06599B
• 45)   Y.  Yong,  M.T.  Nguyen,  T.  Yonezawa,  T.  Asano,  M.  Matsubara,  H.  Tsukamoto,  Y.-C.  Liao,  T.  Zhang,  S.  Isobe and  Y.  Nakagawa: J. Mater. Chem. C 5 (2017) 1033–1041. doi:10.1039/C6TC04360G
• 46)   Y.  Yong,  T.  Yonezawa,  M.  Matsubara and  H.  Tsukamoto: J. Mater. Chem. C 3 (2015) 5890–5895. doi:10.1039/C5TC00745C
• 47)   Y.  Hokita,  M.  Kanzaki,  T.  Sugiyama,  R.  Arakawa and  H.  Kawasaki: ACS Appl. Mater. Interfaces 7 (2015) 19382–19389. doi:10.1021/acsami.5b05542
• 48)   N.A.  Luechinger,  E.K.  Athanassiou and  W.J.  Stark: Nanotechnology 19 (2008) 445201. doi:10.1088/0957-4484/19/44/445201
• 49)   C.  Lee,  N.R.  Kim,  J.  Koo,  Y.J.  Lee and  H.M.  Lee: Nanotechnology 26 (2015) 455601. doi:10.1088/0957-4484/26/45/455601
• 50)   M.  Matsubara,  T.  Yonezawa and  H.  Tsukamoto: Bull. Chem. Soc. Jpn. 88 (2015) 1755–1759. doi:10.1246/bcsj.20150305
• 51)   M.  Matsubara,  T.  Yonezawa,  T.  Minoshima,  H.  Tsukamoto,  Y.  Yong,  Y.  Ishida,  M.T.  Nguyen,  H.  Tanaka,  K.  Okamoto and  T.  Osaka: RSC Adv. 5 (2015) 102904–102910. doi:10.1039/C5RA21402E
• 52)   T.  Narushima,  H.  Tsukamoto and  T.  Yonezawa: AIP Adv. 2 (2012) 042113. doi:10.1063/1.4759498
• 53)   T.  Narushima,  A.  Hyono,  N.  Nishida and  T.  Yonezawa: J. Nanosci. Nanotechnol. 12 (2012) 7764–7776. doi:10.1166/jnn.2012.6150
• 54)   X.  Dai,  T.  Zhang,  H.  Shi,  Y.  Zhang and  T.  Wang: ACS Omega 5 (2020) 13416–13423. doi:10.1021/acsomega.0c01678
• 55)   C.K.  Kim,  G.-J.  Lee,  M.K.  Lee and  C.K.  Lee: Powder Technol. 263 (2014) 1–6. doi:10.1016/j.powtec.2014.04.064
• 56)   B.  Zhang,  W.  Li,  J.  Jiu,  Y.  Yang,  J.  Jing,  K.  Suganuma and  C.-F.  Li: Inorg. Chem. 58 (2019) 3374–3381. doi:10.1021/acs.inorgchem.8b03460