Peeling Back the Layers

A plethora of innovative materials are produced by depositing thin and ultrathin coatings on different substrates. For instance, hard disks able to store terabytes of information are based on multiple magnetic and non-magnetic nano-layers. Similarly, the latest photovoltaic cells consist of nano and micro layers of different elements.

The mechanical, optical, magnetic, and/or thermal properties of these materials are directly related to their chemical composition, including the elements used, their distribution within the layers, and the presence of disruptive trace elements. To make sure the layers are as they should be, we need solid analytical techniques that are able to provide fast multi-elemental chemical analyses with high depth resolution (to monitor the different layers) and high sensitivity (to detect major, minor and trace elements).

Reference techniques, such as secondary ion mass spectrometry (SIMS), have high sensitivity and spatial resolution, but the sputtering process to remove and analyse the different layers is slow, and requires ultra-high vacuum conditions. Long analysis times and low sample throughput are the result, limiting its appeal for routine analysis. Plus, SIMS is based on a single atomization and ionization step, which results in matrix effects that significantly affect quantification. Conversion of qualitative depth profiles (ion signal versus sputtering time) into quantitative depth profiles (concentration versus depth) are tedious and require matrix-matched calibrating samples.

Glow discharges (GDs) appear to provide a faster option. GDs are low-pressure plasmas produced by a voltage difference between two electrodes immersed in an inert gas (generally Ar), where the sample forms the cathode. Sputtering is produced by the bombardment of the material with Ar ions and fast Ar atoms, resulting in a relatively high sputtering rate. Moreover, sputtered atoms from the sample are then diffused into the negative glow region of the GD plasma, where they suffer different excitation and ionization processes. Therefore, atomization and ionization/excitation are temporally and spatially separated, lowering the potential for matrix effects.

In particular, glow discharge optical emission spectroscopy (GD-OES) has proved very effective for the quantitative analysis of thin and ultra-thin layers of diverse nature; for example, conductive and non-conductive layers. Recent instrumental developments, such as differential interferometry profiling (DiP), allow online calculation of the sputtered depth, simplifying the quantification process (see theanalyticalscientist.com/issues/1216/depth-profiling-with-gd-oes).

However, GD-OES suffers from some limitations when it comes to trace elements. Converting emission signals into concentrations still requires the calculation of emission yields that could be affected by the presence of non-metals in the material (such as oxygen, nitrogen or hydrogen). These elements could produce significant quenching effects in the plasma so it’s important to apply the appropriate correction in the quantification methods.

Would coupling glow discharge with mass spectrometry (GD-MS) overcome these problems? For the analysis of multi-elemental thin layers, very fast simultaneous or quasi-simultaneous mass spectrometers are required. In particular, GD has been coupled to time-of-flight (TOF) mass spectrometers, which can take a complete mass spectra every 30 µs. The operation of GD in a pulsed mode, using either direct current or radiofrequency, reduces sputtering and allows each ion signal to be integrated in the most appropriate temporal domain, which reduces the presence of polyatomic interference and allows the highest sensitivity.

Analysis time with pulsed-GD-TOFMS is almost two orders of magnitude lower than with SIMS, with similar depth resolution. And sample throughput is significantly improved, as ultra-high vacuum conditions are not required and the samples can be easily exchanged. A relatively accurate quantification can be achieved using matrix and non-matrix matched calibrating samples (1)(2)(3)(4)(5)(6).

But before pulsed-GD-TOFMS can fulfil its potential, we must improve on the reduction of background signals from non-metals, increase sensitivity, gain a full understanding of the physical and chemical plasma processes of the pulsed GD plasma, and develop more accurate quantification methods based on online determination of sputtered depth. If we can achieve all this, pulsed-GD-TOFMS will become a powerful method for fast multi-elemental analysis of thin and ultra-thin layers.