Depth Profiling with GD-OES

When I first explored GD-OES some years ago, I was impressed by the capability of the technique to perform depth profiling analysis on solid samples, determining nanometric layers and at the same time going down 150 microns into samples. My background in elemental analysis up until then was limited to inductively coupled plasma (ICP)-OES and atomic absorption, so GD-OES was able to reveal a new world to me. I learnt a lot using GD-OES – not only about the technique itself, but also the added value it lends to material science.

It is unfortunate that, despite many advantages, GD-OES remains a technique that is not well known outside of the material science world. Perhaps because there are more fashionable techniques, such as X-ray photoelectron spectroscopy, secondary ion mass spectrometry or even Auger electron spectroscopy? Or does GD-OES need to bring something over and above those techniques to find its rightful place?

One limitation of GD-OES is the necessity to use an external tool to measure the depth of samples. Having the capability to measure depth directly while performing the profile would definitely attract more users and be seen as a breakthrough for the technique. Indeed, having to use a profilometer to determine the depth looks like a real limitation. Essentially, we must either i) assume that the sputtering rate is constant over the various layers to shorten the total time for the measurement – and that means accepting some bias on depth, or ii) return to the profilometer each time to determine the sputtering rate for each layer, when measuring layer by layer on various locations on the sample. And it’s not an easy choice. Moreover, profilometers can be really limited when it comes to the determination of very thin layers.

Thanks to the continuous work of passionate scientists and engineers, the GD-OES technique has continued to improve over the year. At Horiba Scientific, several different product groups (especially ellipsometry) have been working together for some time – the result? A new invention called differential interferometry profiling (DiP). For me, this innovation breaks boundaries for GD-OES by allowing real-time measurement of depth during analysis.

DiP is based on the use of a laser source divided in two beams, one focused on the middle of the measured area and the second one directed to the intact surface. The interference between the two reflected beams is measured as the sample is sputtered, which gives a direct measurement of the crater depth. The solution seems simple once summarized in a few lines (don’t all good inventions?) but a great deal of development was needed to ensure that GD-OES performance was maintained, in terms of the total amount of light reaching the optics and, crucially, also in terms of ease-of-use. The challenge was to ensure the same accuracy as a profilometer but also provide nanometric layer measurement capability.

Since its development and its introduction, DiP has been successfully tested on many different samples, including Silicon wafer, TiN coating on WC, Au thin films, CIGS solar cells, LED and DLC on Cr steel. DiP performs well on non-transparent layers, being the perfect complementary technique to ellipsometry for thickness measurement, while providing information on the elemental composition. As for the future of GD-OES, we foresee that identifying all the applications of interest will keep us busy for some time...