Ken Turner
Ken Turner Consulting
E-Mail: ken.elaine@bigpond.com
Many laboratories spend a considerable amount of time and money setting up a XRF laboratory but sometimes fail to realise the full benefits. This is often due to poor sample preparation maintenance practices. Sample preparation equipment tends to be purchased, put into service and then forgotten about. Somehow it is expected to run forever. It is often not until a mill stops running, fused beads shatter repeatedly or the QC samples go out of control that action is taken. I can not emphasise too strongly that the results obtained from an XRF spectrometer are only as good as the sample presented to it. In this article I would like to look at a number steps in the sample preparation process and suggest some simple maintenance actions that will greatly assist the quality of results.
The greatest potential threat to analysis quality at this stage of the operation is sample contamination. This is especially so in automated grinding systems. A strict protocol needs to be in place to ensure that mills and sample dividers are cleaned between samples. For automated systems a maintenance regime that includes ensuring that drains are not blocked, cleaning reservoirs are full and that the cleaning cycle operates correctly are required. Mills and grinding elements should be replaced or refurbished if they become badly pitted (i.e. can retain traces of the previous sample).
To obtain good XRF results using the pressed powder technique, control of particle size is absolutely critical. Particle size can be controlled by closely controlling grinding time and the mass of sample introduced to the mill. Initially grinding times are determined by experimentally constructing grinding curves. These are plots of XRF intensity versus grinding time. XRF intensity initially rises sharply with grinding time and then tends to plateau. The chosen grinding time should be in the “plateau” region. Grinding time should be set by means of a timer connected to the mill power supply. Control of mass can often be achieved by using scoops of a suitable volume.
In production or when rapid analysis results are needed there is often a temptation to shorten the grinding time. Operators need to be given an understanding of why grinding times must not be altered.
A similar set of operational and maintenance protocols are required as for Primary Crushing to minimise the risk of sample contamination.
XRF bead quality is critical for good XRF results and yet many operators will tolerate poor quality (chipped, scratched, not flat) or cracked beads. Factors that effect bead quality include:
Assuming that we are dealing with an established fusion method (i.e. sample and flux type and ratio have already been established) then it is essential that the first five items above are maintained. Fusion equipment, especially furnace types, are often treated as “set and forget”. This is a dangerous practice. A regular checking and maintenance program is needed.
Adherence to the above practices will help ensure that your XRF laboratory operates smoothly and produces quality results. A regular maintenance program is preferable to a sample preparation crisis.