Manufacturing Bits: March 25

Proving the Big Bang theory; 3D metrology; sorting through cells.


Proving the Big Bang theory
A team of cosmologists using the BICEP2 telescope at the South Pole have discovered the first direct evidence of the Big Bang theory.

The team includes Harvard University, the University of Minnesota, the California Institute of Technology/Jet Propulsion Laboratory, Stanford University/SLAC and others. Using the BICEP2, researchers found direct evidence of a cosmic inflation resulting from the so-called Big Bang that occurred almost 14 billion years ago.

BICEP2 telescope focal plane array  using NIST SQUID chips. (Source: JPL, NIST) Credit: Anthony Turner/JPL

BICEP2 telescope focal plane array using NIST SQUID chips. (Source: JPL, NIST)
Credit: Anthony Turner/JPL

The data included images of gravitational waves or ripples in space-time. The patterns, or so-called “B-mode polarization,” are evidence that the universe expanded rapidly in the first fraction of a second after the Big Bang. “Detecting this signal is one of the most important goals in cosmology today,” said John Kovac of the Harvard-Smithsonian Center for Astrophysics, on Harvard’s Web site.

The BICEP2 telescope is a small aperture, refracting telescope. BICEP2 has 512 antenna-coupled TES bolometers in the focal plane. BICEP2 relies on the same principles as the older-generation BICEP1, but the new system has more detectors that allows for closer packing onto the focal plane.

BICEP2 has been in operation since 2010. The BICEP2 camera relies, in part, on chips developed by the National Institute of Standards and Technology (NIST). The chips are superconducting quantum interference devices (SQUIDs). SQUIDs are the world’s most sensitive magnetometers and powerful signal amplifiers, with broad applications ranging from medicine and mining to cosmology and materials analysis.

NIST’s custom superconducting chips amplify electrical signals generated by microwave detectors. The 16 chips on the BICEP2 system contain a total of more than 2,000 SQUIDs. NIST invented a method for wiring hundreds of SQUID signal amplifiers together to make large arrays of superconducting detectors practical.

3D metrology
Process control is becoming more difficult and costly at each node. Dealing with leading-edge planar devices is hard enough, but now, process control teams must contend with new chip architectures, such as 3D NAND and finFETs.
These devices require 3D metrology to measure the structures, but the production tools generally don’t exist today. So, chipmakers must use the current methods, which can be a cumbersome and expensive process.

In metrology, the workhorse tool is the scanning electron microscope (CD-SEM), which measures the critical dimensions in chip structures. The CD-SEM provides information in tiny structures, which is typically interpreted as 2D images.

Looking to give the CD-SEM a boost, NIST has devised a technology for determining 3D data at features as small as 10nm wide. The model-based method compares data from CD-SEM images with stored entries in a library of 3D shapes to find a match and to determine the shape of the sample.

An SEM image of an IC sample of 10 nm wide SiO2 lines, with the bottom and top edges marked red and green respectively (left). The area marked with a yellow frame rendered into a 3D plot (right). (Source: NIST)

An SEM image of an IC sample of 10 nm wide SiO2 lines, with the bottom and top edges marked red and green respectively (left). The area marked with a yellow frame rendered into a 3D plot (right). (Source: NIST)

With the technology, researchers hope to solve several challenges. First, the image and measurement quality of the CD-SEM is degraded by drift. Second, the interpretation of results requires an accurate, physics-based model.

To solve those issues, NIST has developed two software programs. The first program involves an image acquisition method, which is capable of compensating for drift. The software, called ACCORD, works with 2D Fourier-transforms to piece together many images.

The second program involves a Monte Carlo simulation-based method to interpret the 2D images in 3D. The modeling software, dubbed JMONSEL, is used to generate a library of CD-SEM waveforms for 3D structures.

“Currently, the bottleneck is the speed,” said András E. Vladár of PML’s Semiconductor and Dimensional Metrology Division, on NIST’s Web site. “Generating the modeled libraries can take a long time. The interpretation of the data—finding the best 3D match—is also slow currently.”

Researchers plan to explore the feasibility of using the technique to model feature sizes even smaller than 10nm. “We have high hopes that this method will work well in the 5 to 7nm realm,” Vladár said.

Cell sorting
In 2007, the Massachusetts Institute of Technology (MIT) devised a type of microscopy to inspect the interior of a living cell in 3D. Now, MIT has adapted that technology to image cells as they flow through a tiny microfluidic channel. This is a step towards cell-sorting systems that could separate stem cells, or to distinguish healthy cells from cancerous cells.

Instead of using fluorescent tags, the MIT method analyzes the cells’ index of refraction. The refractive index of biological specimens can be used without any concerns of photobleaching or other harmful effects.

MIT’s system is based on a focused laser beam. It can illuminate cells from many different angles. Researchers used line illumination and off-axis digital holography. This, in turn, enabled the system to record the angular spectra of light scattered from flowing samples at high speeds. Applying the scalar diffraction theory, researchers obtain accurate refractive-index maps of the samples.

MIT demonstrated label-free 3D imaging of live RKO human colon cancer cells and RPMI8226 multiple myeloma cells. The researchers hope to use the system to learn more about how cancer cells grow and respond to different drug treatments. “This label-free method can look at different states of the cell, whether they are healthy or whether they maybe have cancer or viral or bacterial infections,” said Peter So, an MIT professor of mechanical engineering and biological engineering. “We can use this technique to look at the pathological state of the cell, or cells under treatment of some drug, and follow the population over a period of time.”