How to Accurately and Efficiently Process Information from Scanning Probe Microscopy

Interview conducted by Jake WilkinsonJan 19 2016

Image Metrology are leading suppliers of SPM image processing software; which they have been supplier to scientists in both academia and industry over the last two decades. AZoNano spoke to their founder and CEO, Jan F. Jørgensen, about SPM, the information their SPIP™ software elucidates for researchers and the continued evolution of the semiconductor industry.

Please could you give our readers a brief introduction to Image Metrology and the services you provide?

Image Metrology was founded in 1998 with SPIP™ as the primary product. The development of SPIP™ started as part of my PhD project together with IBM and the Danish National Metrology Institute and was continued while I was a guest researcher at NIST in 1995.

Our focus has always been on providing intelligent image processing techniques that facilitate the measurement and analysis of SPM images in a way that is as accurate as possible whilst being user-friendly.

How does the technique of scanning probe microscopy (SPM) work and what information can it be used to determine?

SPM is a technique that uses a very sharp probe to scan over a surface in a raster pattern. When the probe is within atomic distance of the surface an AFM (Atomic Force Microscopy) probe can sense the repulsive and attractive forces from the surface. The height of the probe is controlled such that the force is kept constant meaning that also the distance to the surface is kept constant. Therefore, a topographic landscape image can be produced by recording the z position of the probe for the x, y positions.

Effectively the AFM probe traces over the surface of a substrate and records the surface topography as it does so.

STM (Scanning Tunneling Microscopy) works in a similar manner to AFM but uses a different sensing method. In STM there is a bias voltage set between the probe and the surface and when in atomic distance to the surface a tunneling current can be measured by the probe.

Because both of the techniques involve scanning very close to the surface it is possible to obtain images with atomic resolution.

The big advantage of SPM techniques compared to optical techniques is the ability to obtain height information and the unique capability of obtaining images at atomic resolution. SPM allows a lot of geometrical information to be extracted at a very detailed level.

To obtain geometrically correct images it is crucial that the movement of the probe relative to the surface can be controlled better than the desired resolution, which is a big challenge. It is almost impossible to create images where the pixels are acquired equidistantly and where there is no coupling between the axes. Even in the most perfect instrument problems with environmental noise, vibration and temperature changes will lead to imperfect images.

In SPIP™ we have implemented several methods to characterize imperfections and correct for them which allows extremely accurate measurements to be performed.

How is the atomic scale information from SPM processed so that it can be understood by researchers?

Atomic lattice structures are characterized by well known lattice constants which quantify the periodic distances between atoms. Therefore, SPM images of such surfaces are ideal for calibrating the microscopes and can be used to create correction models for other SPM images of unknown structures. We use dedicated Fourier analysis to determine periodic structures with sub-pixel precision.

What inspired you to make your SPIP™​ (Scanning Probe Image Processor) software?

The high sensitivity of SPM instruments, which enables nano-scale measurements is also what makes such instruments very sensible to various kind of distortions.

While working for one of the early SPM manufacturers (Danish Micro Engineering) in 1988, I realized that some problems could only be solved by applying dedicated algorithms to determine the nature of the distortions and to facilitate correction functions that could provide more “true” surface images. This became the main subject for my PhD and later on it became the foundation of Image Metrology.

During my PhD everything was developed for Unix Workstations because PC’s at that time were not powerful enough for complex image processing. But after the launch of Windows 95 and the evolution of more powerful PC’s it was natural to migrate the Unix SPIP™ to PCs so it could be available for a wider audience.

When SPIP™ for PCs was launched as a commercial product in 1999 it was satisfying a large need for the correction of SPM images and the calibration of AFM microscopes; at that time there were no alternative 3rd party image processing programs.

Today, though now microscopes are capable of producing more distortion free images, the unique calibration facilities in SPIP™ are still used to obtain the most accurate images. However, the focus for our users and our development is today more on ease-of-use with effective and automated processing and we are constantly developing new features to keep our position as the leading supplier of SPM software.

The SPIP™ software can be used to measure the shape of particles. What, in your opinion, are the most exciting applications for this feature?

For people who have spent countless days counting the particles in a vast number of images the most exciting feature is probably that this can now be achieved automatically within a few minutes whilst simultaneously getting additional detailed information about the particle shapes.

We use the term “particle” for any type of isolated feature on a surface that can be described by a shape. Examples of such features are semi-conductor structures, proteins, DNA strings, etc.  Scientifically, one exciting feature is the ability to detect a DNA strand and extract a cross-section profile along its length direction which enables  accurate distance measurements for bound proteins/enzymes.

Particle & pore analysis applied to various types of particles, pores and grains

How does your software ensure accurate results that are free from errors and measurement artefacts?

To get as close as possible to the “true” surface, well-known structures that will reveal instrument artifacts are needed. These can be manmade calibration samples or naturally occurring nano structures.

From images of such structures, we can create models describing the imperfections of the instrument and then apply the inverse model to correct the images so that the following geometrical measurements will be more accurate.

As scientific research continues to increasingly focus on nanoscale electronics where do you expect the field of SPM to advance in the next decade?

SPM is today standard equipment in any nano R&D lab and within the next ten years it will also become a necessary QA (quality assurance) equipment in the production of nano-structures where critical dimensions can no longer be determined by traditional techniques.

For semiconductor devices to work properly it is crucial that the geometry of the smallest building blocks are within certain tolerances and that possible defects are of such small sizes that they will not alter the semiconductor’s function. While the dimensions are getting smaller, so are the tolerances meaning SPM together with proper image processing will be playing a crucial role for QA and inspection in the semiconductor industry.

How do you see Image Metrology being part of this change?

Creating solutions for production lines require not only close cooperation with the end-users, but also the SPM instrument manufacturers and the providers of sample handling. Therefore, Image Metrology will be focusing more on closer partnership with instrument manufacturers and on the development of dedicated solutions for specific QA applications.

Using SPM in a production line is still a big challenge and there is a limitation on the speed. Therefore, we are currently working on a completely new exciting product based on scatterometry that can measure the geometrical dimension of repeated nano structures in milliseconds and therefore perfect for QA in production lines see infoscat.com

About Dr. Jan F. Jørgensen

Dr. Jan F. Jørgensen is an internationally recognized expert in the field of nano- and microscale image correction and analysis. Originally, with a degree in biomedical engineering, he worked as a software engineer for hi-tech analytical companies, whereafter he concluded his PhD project in cooperation with IBM and the Danish National Metrology Institute (DFM), where he went on to work as an expert with fundamental metrology, standards and technology.

In 1998, he founded the company Image Metrology to provide image processing software for microscopy used in industrial and academic research.

Today, Dr. Jan F. Jørgensen is the CEO of Image Metrology, where he continues to drive innovation for the key product SPIP™ as well as development of new products within in the field of image processing software and nano technology.