Tearing Down the
Nearly Invisible

View larger

For many engineers, the first taste of their chosen profession comes when they trace the cables on a bicycle to see how the gears work, or open a CD player to inspect the mechanism that moves the discs in and out. Or they might try to repair an automobile transmission or fix an appliance. Although they endure years of classwork in theory and math, most find that their profession's secrets are revealed by doing.

Reverse engineering—tearing down mechanical devices—is an instinctive way for engineers to learn how things work. Engineers have long disassembled products to seek clues for ways to make them better or more cheaply, to identify a competitor's hidden strengths and limitations, or to uncover patent violations.

This, however, becomes more challenging as devices become increasingly complex and miniaturized. What if the details are too fine to discern with the naked eye? How is it possible to discover the secrets of an integrated circuit or a microelectromechanical system (MEMS)? Some features are so tiny and delicate they crumble at the slightest touch. Differences in materials, an important element in any silicon design, are impossible to discern at this scale without specialized instrumentation.

For example, the dimensions of semiconductors today are measured in tens of nanometers. Reverse engineering of these devices requires expensive, state-of-the-art equipment, highly trained staff, and analytic procedures that are effective in the realm of the near-invisible.

The massive proof mass of the ADSXL330 three-axis accelerator hangs suspended on four pedestals above the polysilicon substrate surrounded by system electronics.

Chipworks, a company in Ottawa, ON, Canada, is highly skilled in this kind of investigative work. Initially, the company utilized an automated microscope stage to carefully photograph the devices it was analyzing. It then progressed to digital cameras and developed its own advanced software to blend the digital photos into three-dimensional representations of chip topology.

A detail of its complex structure shows the air gaps and channels that enable it to respond more freely to changes in acceleration.

Probing Nanoscale Devices

Today, however, with devices being constructed at the nanoscale, the company relies on a scanning electron microscope. "When chip features started to drop below 180 nanometers, we needed to move out of optical imaging," says Julia Elvidge, president of Chipworks. "We were using blue light in our optical microscopes. Its wavelength was 250 nanometers, so when we tried to capture 180-nanometer feature sizes, we were pushing our limits."

It takes an electron microscope three days running day and night to image a typical six-layer memory chip. Older chips, even single layer, can still possess engineering design value. Although they are not as complex, they can still be challenging to reverse engineer because of their intricate designs.

For example, Analog Devices' ADXL330 chip has much larger features than modern integrated circuits and only a single layer of metal. However, all its moving parts and embedded electronics are contained in a 4-mm² package hermetically sealed with low-melting-point glass.

The first step in examining this type of MEMS was an X-ray analysis, which revealed a 1.75-mm² die containing the proof mass. A technician using a microscope then used a scalpel to pry off the glass cap, exposing the proof mass, a massive plate of interconnected zigzagging channels of polycrystalline silicon (polysilicon) that covers most of the die.

Cross-sectional slices of the MEMS were then carefully removed and shipped to labs that specialized in materials identification. The features of MEMS under the microscope were then compared with published papers to determine the sequence of processing steps. For the ADXL330, this phase of work discovered an important structure that was not mentioned in any technical papers: a second layer of polysilicon. The first layer of polysilicon was used to form transistors and the fixed plate of the MEMS capacitor structure to measure acceleration along the x and y axes; the second polysilicon layer senses motion only along the z axis.

On the underside of the proof mass, the team discovered several small bumps that dropped down from the mass above. "These are stiction bumps," notes team leader St. John Dixon-Warren. "If the whole proof mass touched the substrate, the van der Waals electrostatic forces would lock it there and it would stop moving. The stiction bumps keep that from happening by only letting a small part of the proof mass touch the substrate."

Learning these finer engineering details delights Dixon-Warren. "An enormous effort goes into designing this type of device," he says. "A company like Analog Devices might have 100 engineers working for 10 years to develop the ADXL330. We're just skimming the surface."

The ability of Chipworks and similar companies to probe the micro and nanoscale world of today's silicon technology provides a wealth of valuable insights, even from older devices. For people like Elvidge and Dixon-Warren, it also fulfills that lifelong craving to trace the cables, lift the lid, and undo those little screws on the back cover. After all, taking things apart to see how they work is the most natural form of engineering.

[Adapted from “Tearing Down the Nearly Invisible,” by Alan S. Brown, Associate Editor, for Mechanical Engineering, August 2006.]

It takes an electron microscope running for three days and nights to image a typical six-layer memory chip.

Tweet

Share

 Email

 Print