The application of an alternative method of bond monitoring during high-temperature aging is reported using a custom made test chip with piezoresistive integrated CMOS microsensors located around test bond pads. The sensor detects radial stresses originating from the bond pad and can resolve changes because of intermetallic compound (IMC) formation, voiding, or crack formation at the bond interface. Optimized Au ball bonds are aged for over 2000 h at 175°C. It is found that stress sensors next to the bonds are capable of showing the stages of IMC growth, consumption of pad Al layers, and monitoring the formation of low-density and Al-rich IMC (AuAl 2 ) which shows an advanced stage of aging. In particular, a first stress signal increase corresponds to the conversion of all Al above the diffusion barrier into IMCs. The second increase in stress signal after a period of stability corresponds to conversion of all Al below the barrier into IMCs. The IMC formation in these periods causes shear strength increase. After complete bond Al consumption, the bond, however, reaches maximum strength. As bond degradation starts, e.g., by lateral IMC formation, voiding, and oxide formation, as well as because of lateral pad Al transformation to IMC, the signal exhibits a strong decrease. The findings are corroborated by results obtained from classical methods such as interruptive or destructive testing including visual inspection, shear testing, cross sectioning, and by bond resistance monitoring.
A 0.3-lm-thick electrolytic Pd layer was plated on 1 lm of electroless Ni on 1 mm-thick polished and roughened Cu substrates with roughness values (R a ) of 0.08 lm and 0.5 lm, respectively. The rough substrates were produced with sand-blasting. Au wire bonding on the Ni/Pd surface was optimized, and the electrical reliability was investigated under a high temperature storage test (HTST) during 800 h at 250°C by measuring the ball bond contact resistance, R c . The average value of R c of optimized ball bonds on the rough substrate was 1.96 mX which was about 40.0% higher than that on the smooth substrate. The initial bondability increased for the rougher surface, so that only half of the original ultrasonic level was required, but the reliability was not affected by surface roughness. For both substrate types, HTST caused bond healing, reducing the average R c by about 21% and 27%, respectively. Au diffusion into the Pd layer was observed in scanning transmission electron microscopy/ energy dispersive spectroscopy (STEM-EDS) line-scan analysis after HTST. It is considered that diffusion of Au or interdiffusion between Au and Pd can provide chemically strong bonding during HTST. This is supported by the R c decrease measured as the aging time increased. Cu migration was indicated in the STEM-EDS analysis, but its effect on reliability can be ignored. Au and Pd tend to form a complete solid solution at the interface and can provide reliable interconnection for high temperature (250°C) applications.
The method is based on a microheater integrated next to a wire bonding pad (test pad) on a test chip. It is fabricated in CMOS technology without additional micromachining. The microheater consists of two polysilicon resistor elements, placed at opposite sides of the pad, operated in parallel using a constant voltage, each element extending over 30 × 70 μm with a resistance of ≈140 Ω at room temperature, and is operated based on Joule heating. The polysilicon is located at least 20 μm but not more than 50 μm from the pad aluminum. To characterize the microheater, Al serpentine resistors are placed on and between the heaters next to the pad, serving as resistive temperature detectors, having resistances of about 9.4 Ω at room temperature. With a constant operation voltage of 15 V, ≈140 mA of current and ≈2.1 W of heating power are generated, resulting in a heat flux of ≈500 MW/m2. The thermal resistance of the heater is 200 K/W (i.e., loss coefficient of 5 mW/K). The maximum temperature measured on one of the microheater resistors was above 396 °C and was reached using 18 V within less than 5 s of voltage application starting at room temperature. When heating from 101 °C to 138 °C, even faster heating is possible, allowing the performance of highly accelerated thermocycles. These cycles are applied to a ball bond on the test pad. Compared to the 20 min cycles used by a standard test, the new microheater device performed cycles lasting 10 ms (5 ms on, 5 ms off) which is 5 orders of magnitude faster. The released energy is typically 10 mJ per cycle. A 50 μm diameter ball was made using 25 μm diameter Au wire and bonded to the test pad. The effect of the microheater-cycling on the contact resistance values of ball bonds is described. Starting with typical contact resistance values around 2.5 mΩ, the increase observed is between 4% and 7% after 5 × 106 10 ms cycles (≈14 h).
To qualify interconnect technologies such as microelectronic fine wire bonds for mass production of integrated circuit (IC) packages, it is necessary to perform accelerated aging tests, e.g., to age a device at an elevated temperature or to subject the device to thermal cycling and measure the decrease of interconnect quality. There are downsides to using conventional ovens for this as they are relatively large and have relatively slow temperature change rates, and if electrical connections are required between monitoring equipment and the device being heated, they must be located inside the oven and may be aged by the high temperatures. Addressing these downsides, a miniaturized heating system (minioven) is presented, which can heat individual IC packages containing the interconnects to be tested. The core of this system is a piece of copper cut from a square shaped tube with high resistance heating wire looped around it. Ceramic dual in-line packages are clamped against either open end of the core. One package contains a Pt100 temperature sensor and the other package contains the device to be aged placed in symmetry to the temperature sensor. According to the temperature detected by the Pt100, a proportional-integral-derivative controller adjusts the power supplied to the heating wire. The system maintains a dynamic temperature balance with the core hot and the two symmetric sides with electrical connections to the device under test at a cooler temperature. Only the face of the package containing the device is heated, while the socket holding it remains below 75 degrees C when the oven operates at 200 degrees C. The minioven can heat packages from room temperature up to 200 degrees C in less than 5 min and maintain this temperature at 28 W power. During long term aging, a temperature of 200 degrees C was maintained for 1120 h with negligible resistance change of the heating wires after 900 h (heating wire resistance increased 0.2% over the final 220 h). The device is also subjected to 5700 thermal cycles between 55 and 195 degrees C, demonstrating reliability under thermal cycling.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.