Identifying particulate contaminants in metal powders using XRF: Stainless Steel contamination in Titanium alloy Ti-6Al-4V

Introduction

Contamination of metal powders used for additive manufacturing provides a real risk to product quality and safety, since the presence of contaminant particles with different melting characteristics or mechanical properties than the bulk powder can lead to localized stress points in the printed part. This could lead to premature or catastrophic failure and is especially important in risk-averse sectors such as aerospace, medical, and oil and gas. 

Contamination can originate at various stages in the value chain, including powder production, powder handling, and powder recycling at the Additive Manufacturing (AM) facility. Identifying small levels of contamination can be difficult though, and visual discrimination may be impossible since AM powders can look very much alike. 

In this study, we show the potential of X-ray fluorescence as a tool for identifying contaminants in metal powders down to ppm level, with specific emphasis on Stainless Steel SS 316L contamination in Titanium alloy Ti-6Al-4V.

Instrumentation

Measurements are performed using a Malvern Panalytical Epsilon 4 spectrometer, equipped with a 15W, 50 kV silver (Ag) anode X-ray tube, 6 software-selectable filters and a high-resolution SDD30 silicon drift detector and a 10-position sample carrousel. 

Flexible voltage settings from 4.0 to 50 kV and a maximum current setting up to 3.0 mA can be used to define application-specific excitation conditions.

Sample preparation

Aliquots of powders are poured directly into disposable analysis cups and assembled with a supporting (bottom) polymer foil. All samples are irradiated from below. 

Two main categories of AM powders were investigated: (a) virgin feedstocks - from different commercial sources, including the IARM-certified referenced powders and (b) deliberately contaminated test samples with customized contaminant concentrations. These special mixtures have been prepared at MTC – Manufacturing Technology Center, Coventry, UK. To simulate real-life situations, the levels of added contaminants were kept very low, starting from 100 ppmv (parts per million in volume).

Composition signatures

Table 1 shows the certified elemental composition for two CRM metal powders, namely Ti-6Al-4V and SS 316L. As some of the elements (i.e., Fe, Ni, Cr and Mo) are present in high concentration in the composition of stainless steel and only expected in trace levels in Ti-6Al-4V, these specific elements can be used as identifiers for SS316L contamination in Ti-6Al-4V powders. 

Figure 1: Typical XRF signals measured on Ti 6-4 powders 

Measurement procedure

When using optimal excitation conditions for the elements known to be present in the composition of a metallic alloy, the XRF signals for the major and minor elements will be easily identifiable in the spectra (see for example Figure 1). Quite often, the certificate received with a powder is not complete. If the Ti 6-4 alloy contains traces, usually for a concentration > 100 ppm, the spectra will show additional peaks.

Figure 2: Spectra of virgin Ti 6-4 powders from one single producer. Besides the major elements, the alloy has already traces of iron and nickel: (blue): Fe = 300 ppm, Ni =300 ppm and (red) : Fe = 900 ppm, Ni = 2000 ppm

Measuring multiple aliquots from a powder can indicate if the traces are distributed relatively uniformly in the composition of the alloy.

Due to the random arrangements of the particles in every analysis cup, one would expect some small variations of the XRF signals. 

Identify and measure contamination 

Compared to a virgin feedstock, a powder that was contaminated - during handling or usage in the printing equipment - will have either additional peaks or a significant change in the intensities of the elements characteristic for the contaminant. To confirm contamination beyond any doubt, especially for low levels of contaminant, we advise investigating multiple takes from the same powder. To indicate the statistical variations encountered when measuring powders with energy-dispersive XRF, Table 2 shows the data measured on 10 separate aliquots of pure and contaminated powders. 

The excellent deconvolution algorithm used in the Epsilon 4 software can calculate the individual contributions of signals that are very close in energy, for example, V KB (5.43 keV) and Cr KA (5.41 keV).

As the level of contamination with SS 316 L goes up, there is a gradual and obvious increase in the average signals specific to steel: see Fe, Cr, Ni and Mo In Table 2. The signals measured for two of the major compounds in the virgin alloy: Ti and V, do not show a significant change in intensity. However, there is a noticeable increase in the data spread (hence higher standard deviations compared to the pure Ti-6Al-4V). 

Figure 3: Superposition of spectra for a Ti 6-4 powder in the virgin state and (green) after contamination with 10000 ppmv of SS 316L powder. 

Conclusion

The data presented in this application datasheet show that XRF is a suitable technology to analyze possible contamination in metal powders used for additive manufacturing. Two major advantages of using EDXRF are: (i) the easy sample preparation and (ii) the possibility to reintroduce the investigated powder into the printing process or performing further analysis with other analytical techniques. 

Acknowledgements

We acknowledge the collaboration with MTC – Manufacturing Technology Center, Coventry, UK, which provided the contaminated samples.