Microscopy technique could help computer industry develop 3D components
30 June 2013
New application of a not-so-new techniques could provide the semiconductor industry with a crucial tool for improving chips for the next decade or more.
A technique developed several years ago at the US National Institute of Standards and Technology (NIST) for improving optical microscopes, has now been applied to monitoring the next generation of computer chip circuit components, potentially providing the semiconductor industry with a crucial tool for improving chips for the next decade or more.
The technique, called Through-Focus Scanning Optical Microscopy (TSOM), can detect tiny differences in the three-dimensional shapes of circuit components, which until very recently have been essentially two-dimensional objects.
TSOM is sensitive to features that are as small as 10nm across, offering the potential to solve some important near-term measurement challenges posed by manufacturing process control, and helping maintain the viability of optical microscopy in electronics manufacturing.
For decades, computer chips have resembled city maps in which components are essentially flat. But as designers strive to pack more components into these devices, they have reached the same conclusion as city planners - the only direction left to build is upwards.
New generations of chips feature 3D structures that stack components atop one another, but ensuring these components are all made to the right shapes and sizes requires a new approach to measurement.
Optical microscopes are normally limited to features larger than about half the wavelength of the light used to examine the structure of a chip — that's about 250nm for green light. So microscopists have worked around the issue by lining up identical components at regular distances apart and observing how light scatters off the group and then fitting the data with optical models to determine the dimensions. But these optical measurements, as currently used in manufacturing, have great difficulty measuring newer 3D structures.
Other non-optical methods of imaging such as scanning probe microscopy are expensive and slow, so the NIST team decided to test the abilities of TSOM. The method uses a conventional optical microscope, but rather than taking a single image, it collects 2D images at different focal positions forming a 3D data space.
A computer then extracts brightness profiles from these multiple out-of-focus images and uses the differences between them to construct the TSOM image. The TSOM images it provides are somewhat abstract, but the differences between them are still clear enough to infer minute shape differences in the measured structures, bypassing the use of optical models, which introduce complexities.
Simulation studies show that TSOM might measure features as small as 10nm or smaller, which would be enough for the semiconductor industry for another decade. Moreover, TSOM could become useful in any field of work where 3D shape analysis of tiny objects is needed.