Lithium borate fusion is a sample preparation method that allows samples to be presented to an x-ray spectrometer (XRF) for accurate analysis. Industries such as glass, steel, cement, nickel, copper and iron ore often use this method for sample preparation. Companies use the analysis process to monitor any issues and control the quality of their production.
The following items are required to perform sample preparation:
| Item |
Conditions required |
The sample to be analyzed:
Cement, aluminum, iron ore, etc. |
- First, the sample must be crushed and pulverized into a powder consistency
- The sample must be very dry to ensure that errors are not introduced from moisture
|
Flux:
Lithium Tetraborate, or mixtures of Lithium Tetraborate or Lithium Metaborate such as 50:50 |
- The flux must be high quality – low in impurities
- The flux must be pre fused to make sure that it is dense and not a fluffy powder, and manufactured in a furnace at high temperatures above 900 °C
- To ensure good mixing, the flux must have a physical structure that is similar to the sample
- The flux must also be dry so that any moisture adhering to the surface is removed, to ensure no errors are introduced from moisture
|
Platinum Labware:
Crucible and Moulds (casting dishes) |
- The platinum moulds must be very flat (less than 10 micron flatness). If the glass bead is curved, it will unnecessarily introduce errors when analyzed by the XRF machine
- The crucibles must be in good condition so that no leakage occurs when inside the Fusion Machine
- To prevent cross contamination, the Platinum Labware must be cleaned for every new sample
|
Fusion Machine (gas or electric):
A high temperature furnace that can reach temperatures above 1000 °C |
- The Fusion Machine must enable precise temperature control, to allow conditions to be repeated for every single sample
|
An Analytical Weighing Balance:
With a precision of 0.1 mg and a weighing range of about 200 g |
- Must be calibrated on a regular basis to ensure correct readings
|
A Spatula:
For measuring out the Sample and Flux |
- Normally made from stainless steel, so that it can be easily cleaned between each sample and thus prevent cross contamination
|
Sample Preparation
Method of Preparing Samples for XRF Analysis
- The platinum crucible is placed onto the analytical weighing balance
- The sample and flux are weighed in an exact ratio into the platinum crucible, for instance 1.0000 grams of iron ore and 10.0000 grams of lithium tetraborate. Both weights are recorded so that they can be entered into the XRF spectrometer
- Using either a vortex mixer or stirring rod, the flux and sample are mixed inside the platinum crucible
- The platinum crucible is then placed inside the fusion machine with the platinum mould
- The fusion machine heats the platinum crucible to the required temperature, so that the mixture of flux and sample completely melts together into a single homogenous hot liquid
- After the sample has been fully dissolved into the flux, the fusion machine automatically pours the mixture into the platinum mould from the platinum crucible
- The mixture is then cooled in the platinum mould until it sets into a solid glass disc
- The glass disc is finally removed from the mould, ready to be placed into the XRF spectrometer but before doing this, the glass disc should be examined to make sure that it is fully transparent, and that no undissolved particles of sample are present
Method of Analysis of the Glass Disc Using an XRF Spectrometer
- The fused glass disc (sample dissolved into flux) is transferred into a sample holder in the XRF spectrometer
- To determine the type of chemical elements included, the XRF machine measures the x-rays from a high-energy x-ray source which are bounced off the sample
- Since the sample has been diluted by the flux, the XRF machine makes an adjustment by using the previously recorded weights of the flux and sample.To put it simply, if the flux to sample ratio is 10:1 (10 grams of flux and 1 gram of iron ore), the XRF machine will multiply all the results by a factor of 10 to obtain the actual result. For instance, if the XRF measures 30,000 parts per million (ppm) of iron (Fe) in the fused glass disc, then the reported result is in fact 300,000 ppm (30%), because the flux has diluted the iron content. Whenever the XRF machine makes this adjustment, it does so for all elements. Therefore, any contaminants that are included in the flux (calcium, sodium, phosphorous etc.) are also multiplied, which in this case, is a factor of 10. Hence, if the phosphorous contamination in the flux is 10 ppm, this will incorrectly add an amount of 100 ppm to the phosphorus result in the final sample analysis.
It is a standard process to examine a flux blank without any sample, especially when the contaminant concentrations in the flux have to be determined to correct or factor these out of the calculations. Yet, this is only possible when there is a low level of contaminants in the flux and when the contaminants are well below the levels that the lab is attempting to analyze for.
Preferably, the measured levels of contaminants in the flux must be at least an order of magnitude lower than the predicted levels in the sample. For instance, if a laboratory is attempting to study a sample with a predicted phosphorus level of 70 ppm while the contamination level of phosphorus in the flux is 1 ppm (or 10 ppm once multiplied by the dilution factor), then the final result can be determined by subtracting the contamination figure. Conversely, if the level of contamination in the flux is, for example, 10 ppm (100 ppm once multiplied by the dilution factor), the contamination level of phosphorous in the sample cannot be sufficiently reported by the XRF machine as it is likely to calculate a negative or erroneous result.
There are several reasons why high levels of contaminants in flux can considerably affect XRF results:
- The level of contamination in the flux is significantly high when compared to the sample, causing errors in corrections and calculations.
- The level of contamination in individual bottles or batches of flux is not necessarily homogeneous.
- The flux used for the original calibration of the XRF Spectrometer is different from the present flux being used which again leads to errors in precision and accuracy.
- The detection limit for some low level elements is compromised.
Conclusion
There are a number of reasons why high levels of contaminants in flux cannot be easily factored out by the XRF spectrometer. Not only is this not permitted under International Standards, but the levels of containments may vary in each bottle of flux, based on how well the batch (which is made up of numerous bottles) has been mixed by the manufacturer. It is also tricky when the levels and types of contaminates differ between batches of flux, as this hinders with the total calibration of the XRF spectrometer, that is only setup once and then checked periodicaly for accuracy.
This information has been sourced, reviewed and adapted from materials provided by XRF Scientific.
For more information on this source, please visit XRF Scientific.