|
Using Silicone Materials in Lead-Free CSP Processing
|
|
Water absorption is the most significant weakness of epoxy CSP materials in a lead-free reflow cycle, and creates a problem that becomes more acute at higher temperatures. In contrast, experience shows that silicones possess good resistance to moisture and can withstand high temperatures and thermal stress.
|
By Edward J. Benson, Dow Corning Corp., Midland, Mich.
The electronics industry, driven by initiatives in Western Europe and Japan, has been emphasizing the need for a shift to lead-free solders for several years. These initiatives mandate lead-free manufacturing of electronics by this year and 2004, respectively.
With some countries planning to prohibit importation of lead-containing products, it appears that manufacturers who can achieve lead-free status will not only be protecting the environment and its inhabitants from toxic lead, but will also gain a strategic marketing edge over the competition.
A large body of work has been developed over the last several years on the topic of lead-free solder alloys.
The environmental issues have been discussed in detail, and the leading alloy combinations being evaluated as replacement candidates have been well documented. In addition to these issues, the far-reaching consequences that lead-free processing will have on component reliability, surface compatibility and rework have become a topic of widespread interest.
Reflow Temperatures
A key factor in lead-free soldering involves reflow temperatures, which may have to be as high as 260°C. Many of the existing chip-scale designs are unable to withstand the greater heat, which exceeds the limits of epoxy materials commonly used to manufacture components and circuit boards. Resulting defects may include cracking, popcorning and impaired device functionality.
Water absorption is the most significant weakness of epoxy CSP materials in a lead-free reflow cycle, and a problem that becomes more acute at higher temperatures.
As greater heat expands the trapped moisture in these formulations, the potential for popcorning rises, along with the stress placed on the CSP by thermal movement. In addition, CTE mismatches are exacerbated by the higher temperatures, leading to warpage and further stress on CSP components.
Silicone Materials
In contrast, testing and field experience show that silicones inherently possess good resistance to moisture, and are able to withstand high temperatures and thermal stresses.
The excellent heat stability and moisture resistance demonstrated in silicone materials allow them to perform at high temperatures without degradation, and their gas permeability prevents moisture entrapment which causes popcorning in non-silicone materials.
The data generated reflects the physical property improvements in new materials, such as reduced viscosity for easier processing and further reduced CTE. Along with the extended shelf life of the new formulations, silicones offer a desirable physical property profile that also includes flexibility and stress relief, good electrical properties, low ionic impurity levels, low alpha particle emission and gas permeability. These attributes can bring significant enhancements in device reliability and longevity to CSP designs.
Figure 1. Moisture absorption at room temperature
Testing
Researchers conducted moisture testing on a number of epoxy and silicone CSP material formulations, providing data on silicone encapsulant performance in a chip-scale package. Moisture resistance is a key feature in preventing popcorning in any CSP manufacturing operation, but is even more critical when components and board are subjected to a 260°C reflow temperature.
Both perimeter- and center-bonded packages were tested with three encapsulant materials evaluated against a silicone encapsulant, to examine the moisture pickup after being submerged at room temperature and then again after 85/85 testing (85°C/85% relative humidity).
The following results reinforce observations of current silicone products in CSP applications: The potential for popcorning is reduced or eliminated, due to the material's low moisture absorption.
Moisture Level
In room temperature experiments, the epoxies continued to absorb moisture from the air after removal from the water bath. The silicone equilibrates very quickly in comparison. When the samples were placed in a dessicator, the silicone values dropped to nil, whereas the epoxy materials require up to 24 hours before showing a measurable decrease in moisture level.
In the 85/85 testing, this trend is further demonstrated, but at much higher percentages. The best epoxy material was three times worse than the closest silicone, and the worst epoxy material was two orders of magnitude behind the performance of the silicone.
Figure 2. Moisture absorption at 85°C/85% R.H.
Physically, these competitive encapsulants have attempted to become more compliant to improve their performance in chip-scale packages, which depend on that compliancy to protect the solder balls. However, without improvement in moisture resistance, these materials could pose a performance risk at lead-free reflow temperatures.
| Physical Properties of Tested Materials |
| |
Epoxy 1 |
Epoxy 2 |
Epoxy 3 |
Silicone 1 |
| Viscosity @ 25°C |
50 |
115 |
310 |
350 |
| Cure Time, minimum @ 150°C |
60 |
15 |
120 |
30 |
| Ionic Content, ppm (Na, K, Cl) |
N/A |
N/A |
<5 |
<2 |
| Modulus, MPa |
4300 |
1300 |
6200 |
2.6 |
Third-Party Testing
Tessera, San Jose, has conducted additional testing on a number of µBGA components. One test was performed on a group of 46-ball, daisy-chained packages with a microelectronics grade silicone encapsulant.
The peripheral-bonded 5.5 mm x 7.5 mm package used 0.35 mm diameter solder balls at a pitch of 0.75 mm x 0.75 mm. The 70 mm x 165 mm test board held ten devices for testing, and the PWB finish was immersion gold over electroless nickel.
Another group consisted of DRAM 62-ball, daisy-chained packages, constructed with a new, low-viscosity microelectronic encapsulant.
The 9.3 mm x 11.5 mm package employed 0.30 mm diameter solder balls at a pitch of 0.8 mm x 1.0 mm. The 34.93 mm x 133.35 mm test board held eight of the center-bonded packages. The finish, again, was immersion gold over electroless nickel.
The package level test regimen for both groups of these JEDEC Level 1 packages included more than 600 thermal cycles from -65°C to 150°C, 1000 hours at 85°C and 85% RH, autoclaving at 121°C/100% RH/15 psig for 168 hours and high temperature storage at 150°C for 1000 hours.
Thermal Cycles
Board-level testing included more than 1000 accelerated thermal cycles from -40°C to 125°C. Results showed that silicone materials meet all the requirements for lead-free solder processing, with no sacrifice of the physical properties demonstrated in current eutectic Sn/Pb solder operations.
Conclusions
Current silicone materials for CSP manufacturing have shown good compatibility with lead-free processing, the ability to withstand the higher temperature range and resist moisture, while providing reliability data that demonstrates a robust package.
The compliancy of these formulations remains a key to absorbing mismatched CTEs and preserving lead-free solder connections which have a tendency to be more brittle than traditional eutectic Sn/Pb solder joints.
Current CSP material formulations have been shown to work well in a lead-free soldering environment, without contributing to additional stress. Moreover, new product development includes work on materials that will further lower viscosity and facilitate easier processing, while providing even lower CTE for superior thermal stress relief.
 |
| Edward J. Benson |
Mr. Benson is a field applications engineer for Dow Corning. He earned a bachelor's degree in chemical engineering from The University of Michigan, Ann Arbor. He has done extensive field work assisting companies using silicones in device packaging and is now responsible for the western region of the U.S. [e.benson@dowcorning.com]
|
|