We achieve robust bonding of Cu wires to Cu pads on polyimide with silver nanopaste cured at 373 K. The paste is prepared by simply condensing Ag nanoparticle (NP) solution via centrifuging. The bonding is formed by solid state sintering of Ag NPs through neck growth and direct metallic bonding between clean Ag–Cu interfaces. Both experiment and Monte Carlo simulation confirm that the melting point of joint clusters increases during sintering. This creates improved bonds for use at an elevated operating temperature using Ag NPs.

Silver nanoparticle (NP) paste was fabricated and used to bond copper wire to copper foil at low temperatures down to 160°C. The silver NP paste was developed by increasing the concentration of 50 nm silver NP sol from 0.001 vol.% to 0.1 vol.% by centrifugation. The 0.001 vol.% silver NP sol was fabricated in water by reducing silver nitrate (AgNO 3 ) using sodium citrate dihydrate (Na 3 C 6 H 5 O 7 AE2H 2 O). The bond was formed by solid-state sintering among the individual silver NPs and solid-state bonding of these silver NPs onto both copper wire and foil. Metallurgical bonds between silver NPs and copper were confirmed by transmission electron microscopy (TEM). The silver NPs were coated with an organic shell to prevent sintering at room temperature (RT). It was found that the organic shell decomposed at 160°C, the lowest temperature at which a bond could be formed. Shear tests showed that the joint strength increased as the bonding temperature increased, due to enhanced sintering of silver NPs at higher temperatures. Unlike low-temperature soldering techniques, bonds formed by our method have been proved to withstand temperatures above the bonding temperature.

A molecular dynamics simulation based on the embedded-atom method was conducted at different sizes of single-crystal Ag nanoparticles (NPs) with diameters of 4 to 20 nm to find complete melting and surface premelting points. Unlike the previous theoretical models, our model can predict both complete melting and surface premelting points for a wider size range of NPs. Programmed heating at an equal rate was applied to all sizes of NPs. Melting kinetics showed three different trends that are, respectively, associated with NPs in the size ranges of 4 to 7 nm, 8 to 10 nm, and 12 to 20 nm. NPs in the first range melted at a single temperature without passing through a surface premelting stage. Melting of the second range started by forming a quasi-liquid layer that expanded to the core, followed by the formation of a liquid layer of 1.8 nm thickness that also subsequently expanded to the core with increasing temperature and completed the melting process. For particles in the third range, the 1.8 nm liquid layer was formed once the thickness of the quasi-liquid layer reached 5 nm. The liquid layer expanded to the core and formed thicker stable liquid layers as the temperature increased toward the complete melting point. The ratio of the quasi-liquid layer thickness to the NP radius showed a linear relationship with temperature.

Nanojoining, a burgeoning research area, becomes a key manufacturing of complicated nanodevices with functional prefabricated components. In this work, various nanojoining methods are first reviewed. For nanojoining of Ag/Au nanoparticles, three methods are investigated comparatively. Thermal annealing shows a two-step solid state diffusion mechanism. Laser annealing by millisecond pulses displays the thermal activated solid state diffusion. Meanwhile, two effects have been identified in femtosecond laser irradiation with different laser intensities: photofragmentation at rather high intensity (∼1014 W/cm2) and nanojoining at low intensity (∼1010 W/cm2). The photofragmentation forms a large number of tiny nanoparticles with an average size of 10 nm. Control over irradiation conditions at intensities near 1010 W/cm2 results in nanojoining of most of the nanoparticles. This nanojoining is obtained through a nonthermal melting and a surface fusion welding. Joined Au nanoparticles are expected to have numerous applications, such as probes for surface enhance Raman spectroscopy.

Sintering of Ag nanoparticles (NPs) is increasingly being used as a driving mechanism for joining in the microelectronics industry. We therefore performed molecular dynamics simulations based on the embedded atom method (EAM) to study pressureless sintering kinetics of two Ag NPs in the size range of (4 to 20 nm), and sintering of three and four Ag NPs of 4 nm diameter. We found that the sintering process passed through three main stages. The first was the neck formation followed by a rapid increase of the neck radius at 50 K for 20 nm particles and at 10 K for smaller NPs. The second was characterized by a gradual linear increase of the neck radius to particle radius ratio as the temperature of the sintered structure was increased to the surface premelting point. Different than previous sintering studies, a twin boundary was formed during the second stage that relaxed the sintered structure and decreased the average potential energy (PE). The third stage of sintering was a rapid shrinkage during surface premelting of the sintered structure. Based on pore geometry, densification occurred during the first stage for three 4 nm particles and during the second stage for four 4 nm particles. Sintering rates obtained by our simulation were higher than those obtained by theoretical models generally used for predicting sintering rates of microparticles.

Solid state sintering of Ag nanoparticles was used to bond Cu wires to Cu foils at temperatures less than 250 o C. The Ag nanoparticles are coated with an organic shell to prevent sintering at room temperature. After annealing the nanoparticles at 200 o C, the decomposition of the organic shell was confirmed using TGA and Raman spectroscopy. The joint strength was measured by tensile shear tests, which shows that the joint strength increases as the bonding temperature increases. Metallic bond between Ag nanoparticles and Cu was achieved with no contamination. Bonds formed by our method, was confirmed to withstand temperatures higher than the bonding temperatures.

A simple spray method using a plain orifice atomizer has been developed for depositing γ-aminopropyltriethoxysilane (APS) from solutions in water and in methanol onto copper surfaces. The peel strengths between copper foil and epoxy resin were measured with and without APS deposition. In all cases, a higher concentration of APS gives higher peel strength. APS applied from 1 wt% solution in methanol resulted in higher peel strength than when applied from a 1 wt% aqueous solution; the opposite was true with 0.2 wt% APS solutions, indicating a trade-off between deposited APS film thickness and surface coverage. APS was very effective when chemisorption occurred at the surface but much less effective when there was only physisorption. A study of the fracture surfaces showed that the failure is cohesive, inside the epoxy layer, and that the deposited APS on the copper surfaces had a long-range effect which was seen deep in the epoxy layer, well away from the copper surface.

No abstract