Sputter Coating in Argon Vs Air: Does it Matter?

One of the most frequently asked questions I receive is whether using air instead of argon as the process gas for sputter coating will affect the quality of coating. The short answer to this question is: Yes. However, whether this ‘matters’ has a slightly more nuanced answer.  

The main advantage to using argon for sputtering is that it is chemically inert. This property helps to reduce contamination on samples, as well as prevent oxidation of oxidisable targets such as chromium. This has been demonstrated by Foroughi-Abari et al., where XPS analysis demonstrated that exposure to air lead to formation of chromium(III) oxide within sputtered layers.1 Metal oxides will reduce the conductivity of the sputtered layer, thereby diminishing charge dissipation effects. Therefore, when using a target that is prone to oxidation, using argon as the process gas is essential.  

But what about noble metals? Common ‘noble metals’ used for coating for electron microscopy include gold (Au), gold/palladium (Au/Pd), platinum (Pt), and silver (Ag). Of these metals, Au is the most chemically inert and therefore we observe the least variation between sputtering in-argon vs in-air. Figure 1 shows a comparison of coating of polystyrene microspheres with Au in-argon (left) and in-air (right). Slight contamination effects can be seen within the in-air sputtered image.

Figure 1. Comparison of in-argon Au sputtering (left) and in-air Au sputtering (right) on polystyrene microspheres.

Metals such as platinum, palladium and silver are more reactive than gold and consequently exhibit more pronounced contamination effects. Typically these metals will not be forming metal oxides in-air, but instead weakly binding to oxygen, water or hydrocarbons from the air.  These residual hydrocarbons can then polymerise under the electron beam.2 Figure 2 shows a comparison of sputter coating  of polystyrene microspheres with Pt in-argon (left) and in-air (right). The image produced with in-air platinum sputter coating shows electron-beam induced contamination, likely resulting from contamination introduced during the sputter coating process.

Figure 2. Comparison of in-argon Pt sputtering (left) and in-air Pt sputtering (right) on polystyrene microspheres.

Contamination visibility will also depend on the electron source within the microscope. FEG-SEMs have smaller probe sizes and significantly higher brightness when compared to thermionic tungsten SEMs. This results in higher local current density on the sample and therefore, stronger beam-induced contamination effects.2 Therefore, the effects of using air as the process gas during the sputter coating will be more apparent when imaging using a FEG-SEM. 

Argon isn’t only used because of its chemically inert nature, argon also increases the rate of sputtering compared to in-air. This is due to argon having a relatively high atomic weight, and therefore ion bombardment of the target is more effective. Figure 3 shows a comparison of in-air and in-argon sputter rate in a TurboQ S for Pt.

Figure 3. Comparison of sputter rate of Pt in TurboQ ES when using air as the process gas (pink) and argon as the process gas (purple).

In response to the question, “Does it matter?”, the answer is yes. The use of argon reduces sample contamination, prevents formation of metal oxides, and increases control and rate of sputter coating. However, in-air sputter coating can be acceptable when coating with Au and analysing samples at low-medium magnifications (<50 Kx) using a tungsten SEM.

References

1. Foroughi-Abari, A., Xu, C., & Cadien, K.C. (2012). The effect of argon pressure, residual oxygen and exposure to air on the electrical and microstructural properties of sputtered chromium thin films. Thin Solid Films, 520, 1762-1767.
2. Goldstein, J. I., Newbury, D. E., Michael, J. R., Ritchie, N. W., Scott, J. H. J., & Joy, D. C. (2017). Scanning electron microscopy and X-ray microanalysis. springer.