Imagine a small notebook about ten pages thick. Each page is filled with some symbols. Is it possible to read the text in each page without actually opening the notebook? Yes, it is, says Eduard Rau, leading scientific associate at the joint Microscopy and Electronic Microtomography laboratory of the Physics Department, Moscow State University. Professor Rau led the development of an original method of noncontact nondestructive sample test. But in this case, however, the researchers do not deal with notebooks (used as an example for students for visualisation purposes) but with microelectronic devices and appliances.
Eduard I. Rau, head of the joint Microscopy and Electronic Microtomography laboratory, established at the Institute of Microelectronics Technology and High Purity Materials, Russian Academy of Sciences; and the Physics Department, Physical Electronics faculty, Moscow State University; Professor, Doctor of Physics and Mathematics.
The laboratory performs electron probe analysis of microelectronic products, materials and devices. In recent years, there has been an active shift from microelectronics to nanoelectronics, and therefore research is also targeted at the nano-field. Control and diagnostics of microchips are getting dramatically more complicated due to constant decreases in the size of their components. At some point, that was measured in microns, then in submicrons, and now tens of nanometers are involved. That is a thousand times smaller than the diameter of a human hair (50 micron). For example, the size of the crystal in large chips installed in computer processors is several millimetres, with possible billions of elements featured in one chip all within very slim spaces.
Eduard Rau led the development of an original method of noncontact
nondestructive sample test
Microchips are increasingly made multi-layered with a sandwich-like structure. If some of the layers faces a failure it is extremely difficult to identify the specific area as the structures are non-transparent. Optical microscopes do not allow for a clear view of every and each element. However, a scanning electron microscope is capable of fulfilling the task as it magnifies the size of the object under scrutiny by a hundred thousand times. This is the device the researchers use. The main purpose of microtomography is to provide a view beneath the surface of the sample (‘tomography’ stands for section imaging).
Previously, the issue was addressed by studying the surface using a scanning microscope, then the upper layer was “removed” (with the help of chemical etching or ion beams) in order to look at the second layer, then the third, and so on. That of course was a destructive method. For visualising purposes, Professor Rau offers a medical analogy: it is similar to cutting off one body part of a patient after another to find a tumour.
Microchips are increasingly made multi-layered with a sandwich-like structure. A scanning electron microscope is used to look at each small element.
Later on, a group of Japanese and American researchers suggested another method or, to be precise, recovered an old method that had been invented 40 years ago at the onset of the scanning microscopy era. In essence, it required accelerating electrons focused in a probe to high energies. “Indeed, the more energy the initial electrons have, the father they get under the surface and provide the information from under the optically non-transparent surface, — comments Eduard Rau. — The only problem is that all reflected electrons get detected as well. The initial electron gets on the surface, then goes deeper, is reflected on a certain depth, and emerges again while gathering data concerning not only the layer where is was reflected — say, the third one — but also on the first and second ones. They all carry depth-integral data too. This way, we do get to see the first or the second layer of the sample but that view comes against a blurred background of the upper or lower layers. As a result, we acquire a multi-layered, blurred and clouded shadow.”
Professor Rau’s laboratory suggested a new mycrotomographic method allowing to get better-quality images of single thin layers of microstructure. The method is based on detecting a part of back-scattered electrons filtered by energy.
“The electrons that were reflected on a particular depth under the surface have respective energy which is inversely proportional to the depth of the reflected layer, — says Professor Rau. — The more the depth of the microheterogeneity in the three-dimensional structure, the longer path the electron takes and the more energy it loses, respectively. Detecting electrons of particular energy allows to visualise the sample layer on the specified depth.”
An original spectrometer with toroidal electrodes was used to analyse the electrons. The researchers adapted it for the scanning microscope in order to provide quality images.
The project of Development of Nanotomography Method and Building the Equipment for Measuring Geometric Parameters and Topology of Under-Surface Nanostructures was being carried out since 2008 to 2010, with the support of the Federal Task Program “Research and Development on Priority Directions for Russian Scienti....” The project budget was RUR20 million.
Simultaneous detecting of the electronically induced potential in the sample is necessary in order to acquire additional information on the distribution of potential barriers in the structure being studied. In this mode, a metallic ring placed immediately between the spectrometer and the surface of the structure being tested serves as a signal sensor. The signal is transmitted onto the microscope display, creating the image of all electrically active fragments of the semiconductor crystal or the chip.
The suggested method of simultaneous detecting of two informative video signals in the scanning microscope also allows to perform visual layerwise monitoring of the microstructure topological build by depth, and of electrically active microchip elements. The researchers point out that this diagnostic method is nondestructive and does not require electromechanical contacts for access to any elements of the microchip. That makes it useful for quality testing and control on every technological stage of device manufacturing. In other words, the methods can be applied both for testing of volumetric (three-dimensional) construction of thin-film multilayered micro- and nanostructures, and for mapping all electrically active elements of the sample under scrutiny (local potential barriers, semiconducting crystal flaws, impurities distribution, and recombination centre accumulation).
“We are still improving the methods, in particular by developing new models for quantity research — that is, not only to get electronic tomographic images but also to interpret the chemical composition as well as the depth of nanostructure fragments, — says Eduard Rau. — We address a complex task of quantity diagnostics of three-dimensional micro- and nanostructures.”