In almost all integrated circuit chips, the wires that connect the internal circuitry to the external packaging are attached by a process called wire bonding. In that technique, ultrasound is used in combination with heat and pressure to weld the tip of the wire, usually gold, to the surfaces to be connected. It's been known for 40 years that ultrasound can make metals easier to work, an effect called acoustic softening. But the process of working the metal can have its own impact on the metal's hardness. Thus it's been difficult until now to get a clear picture of what's going on, and wire bonding has remained a largely empirical process. By placing gold microballs under different levels of applied force and ultrasound and measuring their resulting deformation, a team of researchers from the University of Waterloo in Canada and Tsinghua University in Beijing has succeeded in separating the softening contributions of the ultrasound from the effects of the mechanical force. The researchers were also able to quantify the residual effects of ultrasound on gold, and they found residual softening that increased with greater ultrasound amplitude above a certain threshold. They attribute the residual effects to the net balance between ultrasound's dynamic annealing and its potential opposing effect on activating and multiplying dislocations. (I. Lum et al., J. Appl. Phys., in press.) — Richard J. Fitzgerald
Ultrasound's role in wire bonding
No TrackBacks
TrackBack URL: http://blogs.physicstoday.org/mt/mt-tb.cgi/2946
1 Comment
Search
Categories
- Acoustics (12)
- Astronomy and cosmology (40)
- Atomic physics (23)
- Biography and personalities
- Biological physics (43)
- Careers and employment
- Chemical physics and molecular physics (37)
- Classical mechanics and electromagnetism (5)
- Computational physics (15)
- Condensed matter (47)
- Crystallography (8)
- Earth sciences (37)
- Education (1)
- Energy policy and R&D (6)
- Engineering and technology (34)
- Facilities and laboratories (1)
- Fluids & rheology (29)
- Government agencies (2)
- History, sociology, and philosophy (1)
- Instrumentation (12)
- Materials science (31)
- Medical physics (4)
- Metrology and fundamental constants (4)
- Microscopy (13)
- Nanoscale science and technology (25)
- Nonlinear science and emergent phenomena (17)
- Nuclear and particle physics (17)
- Optics and photonics (36)
- Plasma physics (7)
- Quantum physics and information (23)
- Science policy and politics (4)
- Scientific societies and awards
- Statistical physics and thermodynamics (14)
- Theoretical physics (16)
Monthly Archives
- June 2011 (5)
- May 2011 (8)
- April 2011 (8)
- March 2011 (9)
- February 2011 (7)
- January 2011 (8)
- December 2010 (9)
- November 2010 (8)
- October 2010 (9)
- September 2010 (8)
- August 2010 (9)
- July 2010 (8)
- June 2010 (8)
- May 2010 (8)
- April 2010 (9)
- March 2010 (9)
- February 2010 (7)
- January 2010 (7)
- December 2009 (8)
- November 2009 (8)
- October 2009 (10)
- September 2009 (8)
- August 2009 (9)
- July 2009 (9)
- June 2009 (9)
- May 2009 (7)
- April 2009 (9)
- March 2009 (9)
- February 2009 (7)
- January 2009 (7)
- December 2008 (8)
- November 2008 (7)
- October 2008 (9)
- September 2008 (9)
- August 2008 (8)
- July 2008 (11)
- June 2008 (2)
I wonder what the effect of wire bonding has on complex chips, smaller and smaller feature size could lead to damage being caused by nearby wire bonding effort. Anyone see this effect?