Grantee Research Project Results
2017 Progress Report: Green Nanosolder Paste for Next Generation Electronics Assembly and Manufacturing
EPA Grant Number: SU835938Title: Green Nanosolder Paste for Next Generation Electronics Assembly and Manufacturing
Investigators: Gu, Zhiyong
Current Investigators: Gu, Zhiyong , Gao, Fan , Wernicki, Evan , Shu, Yang , Fratto, Edward , Wang, Jirui , Kepner, Robert , Essigmann, Mikayla
Institution: University of Massachusetts - Lowell
EPA Project Officer: Page, Angela
Phase: II
Project Period: October 1, 2015 through September 30, 2017 (Extended to August 31, 2018)
Project Period Covered by this Report: October 1, 2016 through September 30,2017
Project Amount: $75,000
RFA: P3 Awards: A National Student Design Competition for Sustainability Focusing on People, Prosperity and the Planet - Phase 2 (2015) Recipients Lists
Research Category: P3 Challenge Area - Chemical Safety , Sustainable and Healthy Communities , P3 Awards , Pollution Prevention/Sustainable Development
Objective:
The objective of this project is to develop a new type of environmentally friendly solder paste, green lead‐free and halogen‐free nanosolder pastes. These new nanosolder pastes have the potential to replace the conventional solder paste with micron sized solder balls, which contains toxic and hazardous materials, including lead or halogen or both in a typical solder paste formulation in current electronics industry. As electronics and devices are getting smaller, lighter and more powerful, new assembling materials and methods are required to accommodate all electronic parts in a much smaller single device. The nanosolder pastes, composed of solder nanoparticles, can be an enabling material for next generation electronics assembly and manufacturing.
Progress Summary:
Nanoparticle Form:
Lead-free alloy nanoparticles were synthesized using a surfactant-assisted chemical reduction method in aqueous solution [8]. All alloys produced were near-eutectic alloy compositions due to the desire of having a low melting temperature or temperature range. Resulting nanoparticle size, shape, and composition were examined using a JEOL JSM 7401F scanning electron microscope (SEM) equipped with X-Ray Energy Dispersive spectroscopy (EDS). For further size and morphology analysis, a Philips EM400T transmission electron microscope (TEM) was also used. Currently, the nanosolder alloys that have been synthesized are Sn/In, Sn/Ag, and Sn/Ag/Cu with the Sn/In and Sn/Ag systems being prioritized due to the simpler nature of a binary alloy synthesis system. The overall resulting size of the synthesized Sn/In nanoparticles averaged 60 nm in diameter with an Sn/In ratio of approximately (50/50). The other alloys were confirmed via EDS to have the compositions of Sn-4Ag and Sn-4Ag-0.5Cu, with both having average particle diameters between 20 and 25 nm from a typical synthesis. The similarity in size for the Sn/Ag and Sn/Ag/Cu solder alloys can be attributed to the very minor difference in synthesis materials, with the latter containing an additional salt precursor. Due to a large increase in surface area as the radius of nanoparticles are decreased to below 50 nm, the Sn/In nanoparticles contained less oxygen (~7 wt%) in the form of surface oxides compared to the other alloys (~13 wt%).
Nanowire Form:
Tin (Sn) nanowires were synthesized using a template-assisted electrodeposition method that utilized a polycarbonate (PC) template and commercial tin plating solution. Nanowire size and morphology were characterized by the FE-SEM. For the synthesis parameters use in this research, the final length of the synthesized nanowires was in the range of 4 to 6 μm and the measured diameters ranged from 100 - 150 nm. The variance in nanowire diameter can be attributed to the PC template pore size variation, where the electrodeposition occurred. Tin was selected as the best element to study, as many lead-free solder alloys are largely tin based. The melting temperature of tin (232 °C bulk) differs from the lead-free alloys by only 12 – 15 °C, so this was deemed to be the best choice for a single element soldering system.
Nanosolder Paste Preparation:
The synthesized nanoparticles were mixed with a flux to form the nanosolder pastes with similar rheological properties as the traditional type 4 solder pastes. The flux materials used are commercially available and used in current microsolder systems. Due to the desire for similar rheological properties as current pastes, the nanosolder loading is lower, with typical loading ranging from 20-60 wt% nanosolder particles. Commercial pastes are commonly in the range of 87-90 wt% with type 4 micropowders. A Sn/4Ag nanosolder paste containing ~45 wt% of nanosolders was printed though an uncoated, stainless-steel stencil. The stencil had a thickness of 6 mil (152 µm) with an aperture diameter of 28 mil (711 µm). The Sn/4Ag nanosolder paste released from the stencil walls and remained on the Cu substrate that was used.
This demonstrated that it is possible to screen print nanosolder pastes. Opportunities still remain to improve the printing performance through nanosolder dispersion and paste homogeneity. As previously mentioned, the aperture walls were completely uncoated, generally representing the most difficult scenario for paste release to occur. In recent years, various coatings have been studied that have helped improve paste release for various geometries, which may further help nanosolder paste printing in the future. Future efforts in screen-printing nanosolder pastes will focus only on uncoated stencils, as it is desired for the release to occur without the added help of a coating. In addition, decreasing the feature size of the printed feature is also desired to show the material can accommodate the continuing miniaturization trend.
Nanosolder Paste:
To melt the nanosolders, different reflow methods were employed to test various parameters: (1) ceramic hotplate, (2) tubular furnace with programmable temperature profile, and (3) benchtop surface mount reflow oven. For Sn/In samples, solder pastes were heated to a surface temperature of ~140 °C to allow for complete melting of the material. Sn/Ag based nanosolder systems are still currently being optimized and therefore reflow methods (2) and (3) are primarily used due to greater temperature control and the ability to reflow in inert environments (N2 or Ar). Typical processing temperatures of the discussed non-Sn/In alloys range up to ~265 °C due to the large difference in alloy melting temperatures. With the Sn/In system that has been previously reported, complete melting of the nanosolder pastes remains a consistent observation. In addition, this paste system continues to be able to form solder features of many different sizes. The scalability of this system was demonstrated via a Cu-Cu wire model system. As of the present, Cu wires can be joined using a developed Sn/In nanosolder paste with diameters as small as 25 µm, nearly the size of one type 4 micropowder. Since the last reporting, the focus with this system has been on improving nanosolder paste loading in addition to characterizing the joining with 25 µm Cu wire from an interfacial perspective. Melting with Sn/Ag nanosolder pastes has shown to be less consistent when compared to the Sn/In nanosolder system. This refers to the incomplete melting of the nanosolder paste when subjected to a reflow process under similar conditions (~30 °C above melting temperature). This is commonly observed when prepared within the same nanosolder loading range and flux as the Sn/In system. Evidence of melting can be observed from a visual perspective; however, it results in a large amount of residue remaining on the solder bump and Cu substrate after solidification.
It can be observed that there is evidence of Sn/4Ag nanosolder paste melting due to the luster, bump height, and inability to remove it from the Cu substrate. It should be noted that the sections where the melted nanosolder was noticeable were on different focal planes on top of the bump and therefore not Cu. Additionally, the surface does not match the Cu substrate polishing orientation, which can be viewed as aligned horizontally. The incomplete melting that is observed is believed to be due to the oxygen that is present on the surface of the nanosolders. As previously mentioned, the oxygen content is nearly doubled compared to the Sn/In nanosolder system which may be over the limit of what the flux used can remove within the reflow parameters used. The dark black color of the residue strongly indicates unmelted particles are within the residue, as it is the color of the solder paste itself before reflow. Future work regarding the Sn/Ag and Sn/Ag/Cu nanosolder paste systems aims to increase the consistency of melting followed by an increase in loading. Prior to melting, screen printing will be performed to view the homogeneity of the paste overall, as different bump sizes will result. As with the Sn/In system, the intermetallic compounds formed at the interface are of great interest and importance.
Melting of Nanowires via Infrared Heating
To prepare nanowire samples for infrared (IR) heating and melting, a suspension of Sn nanowires was prepared by dispersing the Sn nanowires in ethanol. The suspension was deposited on to a silicon (Si) wafer, dropwise, and allowed to evaporate inside a fume hood. When flux was used, it was placed on a separate Si wafer. Following the depositions, both Si wafers were placed together on the IR preheating station. A thermocouple was attached to the surface of the substrate with the nanowires to monitor the temperature during the IR heating process. To better isolate the heating environment, a polydimethylsiloxane (PDMS) cover was made to cover the sample and the flux substrates. This resulted in lower temperature gradients and allowed the heat and flux vapor to stay contained in a relatively closed system. Prior to IR heating, there was a 1-2 min preheat step that causes the flux to evaporate. Implementing this step decreased the heating time, from starting temperature to the desired peak heating temperature. The heating time recorded was when the temperature rose to above 230 °C, or around the melting temperature of the tin nanowires. In industrial reflow processes, this time is referred to as the time above liquidus (TAL), which can have a very strong influence on the resulting solder joint properties. The peak temperature for the melting process was ~280 °C, close to the temperature the IR heating gun was set to.
Cu-Cu Bonding & Electrical Resistance Measurements:
The Cu-Cu wire system is the model system used to demonstrate the scalability of the Sn/In nanosolder paste while allowing electrical measurements post reflow. Electrical measurements utilize a four-point method to produce the most accurate solder resistance values while neglecting contact resistance of the probes used. Measurements were taken five times for each solder joint and the resistance values were averaged for the 500, 250 and 100 µm diameter Cu wires. Previous results have shown that increased resistance values should be observed when decreasing Cu wire diameter like any other joining system. This has been observed for both the SAC solder joints and Sn/In solder joints. The results showed that the electrical resistance of Sn/In solder joints are very comparable to those of SAC solder joints.
Journal Articles:
No journal articles submitted with this report: View all 16 publications for this projectSupplemental Keywords:
Lead-free, nanosolder, Cu-Cu joining, solder paste, electronics assembly and packagingRelevant Websites:
Nano-Solder Research at UMass. Lowell Exit Exit
Progress and Final Reports:
Original AbstractP3 Phase I:
Green Nanosolder Paste for Next-Generation Electronics Assembly and Manufacturing | Final ReportThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.