Search

Ag sintering has been widely studied as a lead-free die attach solution for power electronics. A soak time of a dozen minutes at the sintering temperature is necessary to establish strong bond strength by the conventional heating method. Chemical bonding can be achieved by various parameters associated with the metallurgical bond of the adjacent surfaces. In this study, attaching technology between Au coated Si die and Au coated Si substrate was developed by applying Ag sintering materials which can work at temperature up to 350°C. It was concentrated on finding so called "pressure-less sintering and pressure sintering" procedure in air and N2 environment. The sintering was performed in air or N2 at temperature range from 180°C to 280°C for 10 minutes to 3 hours. Highest shear strength was obtained at 200°C of sintering temperature bonded for 3 hours.

Lead-free solder alloys for high temperature applications is required to meet increasing demands for reliable replacements for lead-based alloys. Especially, high temperature solder is required for bonding in power electronics packages where first bonding on a lead-frame is performed at a temperature of about 330 °C. In this study, the SiC based TO-220 power package was developed by applying high temperature endurable material sets which can endure over 220°C device junction temperature. The interfacial reaction and reliability of SiC die attachment on and Ni plated lead frames using Zn-Al based solder was investigated. Zn-Al-Ge solder alloy showed good temperature tolerance after temperature cycle (TC), high temperature storage test whereas Zn-Al-Ge-Mg solder alloy showed failures after TC test.

Au-Ge eutectic has been used as a high temperature electronics interconnection material up to 250°C due to its favorable characteristics. However, both nano-indentation and shear testing, confirmed the loss of strength of the Au-Ge eutectic at a high temperature of 300°C due to the growth of the (Ge) phase. This coarsening has also resulted in the weakening of the (Au) phase due to the deterioration of the precipitation hardening of the (Au) matrix by the (Ge) dispersed phase. In this paper, various techniques for averting the coarsening of the (Ge) phase have been explored. It has been determined that Sn can dissolve in the Au-Ge and segregate in the (Ge) phase, resulting in restraining the coarsening of the (Ge) phase. The composition of the ternary Au-Ge-Sn alloy has been designed by taking into account; the compliance with the solidification criterion and precipitation of phases in the bulk solder. It has been ensured that no brittle intermetallic compounds (IMCs) precipitate in the matrix of the Au-Ge eutectic bulk solder, as a result of micro-alloying with Sn.

Due to different demands, there are many types of die attach technologies in the market. In this work, die attach materials such as solder, silver filled epoxies and sintered silver are investigated for their mechanical performance under standard processing steps and under power cycling conditions for a quad-flat no-leads package (QFN). In addition, two molding compounds, one with thrice the thermal conductivity of the reference molding compound are analyzed. A combination of experimental and finite element analysis suggested that, sintered silver die attach material is the most susceptible to failure at the die attach layer while one of the epoxy filled die attach material investigated is the least susceptible to failure in a QFN. However, sintered silver seems to induce the least die stress at peak solder reflow temperature. Although sintered silver had superior thermal conductivity, the temperature within the package during power cycling is very similar to the other die attach candidates. It appears that higher thermal conductivity of the die attach material and molding compound do not lead to an increase in fatigue life of the solder joining the package onto the board.

In this paper, characterization of 3 types of underfill for high performance applications has been presented. Characterizations of underfill materials such as adhesion testing, contact angle measurement on different surfaces, filler size distribution were conducted to understand the underfill materials. The adhesion test results revealed that failure is mainly mixed mode, which is the failure between bulk underfill failure and interfacial failure. The thermal conductive underfill has a wider range of filler sizes and has the most hydrophilic behavior. From the glass chip flow test, the flow behavior of the underfill could be observed, with the most thermal conductive underfill having the slowest flow. Process optimization for thermal chip and daisy chain chip for void free underfill was also carried out based on dispensing temperature and patterns.

In this paper, die attach process characterization on two type of highly conductive silver paste die attach materials was discussed. The first silver paste die attach materials (DA1) was used as a reference which is silver-loaded epoxy adhesive with high thermal conductivity of 60W/mK and electrical conductivity of 16Ms/m. Second silver paste die attach material (DA2) can be sintered with low pressure or pressure-less at temperature of 220°C to 280°C. DA2 material acquires high thermal conductivity range of 100–170W/mK and electrical conductivity range of 12–15Ms/m. Process specifications were set at die tilt < 1%, average bond line thickness between 25μm to 50μm and full die attach materials coverage without overflow of materials on top of die's surface. Process was optimized with 70μm thin silicon daisy chain chip with die size of 5mm×5mm on Ag plated QFN lead frame for both silver paste materials and achieved the required process specifications. Process optimized on DA1 achieved average bond line thickness ranged from 24.5μm to 30.5μm with die tilt less than 0.24% and DA2 had average bond line thickness ranged from 32.6μm to 44.2 μm with die tilt less than 0.15%. There was further evaluation on die attach process with silver sintered paste for different die thickness (which 50μm, 70μm and 175μm were used) on a fixed die size of 5mm×5mm. Porosity after die attach cure is always a curial factor which affects the modulus and conductivity of the device. Investigation on porosity of cured die attached materials was carrying out on different die size range from 0.5mm × 0.5mm to 5mm × 5mm. This helped to understand the effect of die size on sintering process. Optimization of dispensing pattern and die attach process challenges of thin die attachment were discussed in details in this paper.

This work describes an investigation into the performance of packaging materials, developed with a specific aim in improving thermal performance, such as silver-sintered materials that perform both die-attach and thermal-interface-material (TIM) duties. The test vehicles under study consist of a flip-chip BGA package (die size: 10.2×10.2×0.3mm in a package of 21×21×1.5mm) and a molded QFN package (thin die size: 5×5×0.070mm in a package of 8×8×0.7mm) Parametric thermal modeling using Finite-Volume-Method Simulation thermal modeling software covers the effect on package thermal resistance (junction-to-case) due to package geometry parameters, interface thickness (bond-line thickness), void coverage, thermal conductivity and contact resistance. In addition to steady-state thermal resistance, transient analysis of the QFN structure under a high frequency operational power profile is also investigated. Finally, experimental thermal characterization of the device for thermal resistance (junction-to-case) is also conducted under time-zero conditions and post-reliability tests (such as Moisture Sensitivity (JESD22-A113-D MSL3 and MSL1), Temperature Cycling (JESD22-A104-G TC-condition B or C), High Temperature Storage (HTS, +150°C and 1000 hours) and Unbiased Highly-Accelerated Stress Test (HAST)).

This paper presents a new numerical model to characterize the die attach process in the advance packaging. With its successful application on a 5 mm by 5 mm with 70um thickness die attached to the substrate, final fillet shape of the die attach material is predicted for various process conditions. Focuses have been given on the die attach material over flow on the die top surface. The contamination on the die top surface may cause failures in the subsequent processes. The simulation results were compared with the experiment. Good match was obtained. Moreover, process window for a given amount die attach material was established through the simulation and the corresponding bonding force that will not cause die attach material over flow was predicted. Key advantage of this numerical study is to give the insights into process parameters and provide initial process window to prevent die attach over flow.