What are the advantages of using fusion beads for XRF analysis instead of pressed powder pellets?

Fusion eliminates analytical errors associated with differences in particle size and mineralogy because the sample is converted into a homogeneous glass. Dilution with flux also reduces matrix effects, and standards can be prepared by mixing suitable reagents rather than finding reference materials closely matching the sample. The tradeoffs are reduced X-ray intensity because of dilution and a longer preparation time, typically about 15–30 minutes including weighing.

How is a fusion bead prepared for XRF analysis?

A dried, finely pulverized sample is accurately weighed with flux and any required oxidizing or releasing agents. The components are thoroughly mixed and, when necessary, oxidized or calcined before fusion. The mixture is typically heated to approximately 1000–1200°C until completely molten, then agitated to homogenize the melt and remove bubbles. The molten material is cast or retained in a combined crucible/mold and carefully cooled to form a homogeneous glass bead.

How should powder samples be prepared before making XRF fusion beads?

The sample should generally be pulverized to a particle size smaller than 106 µm (140 mesh). It is then dried for more than two hours at 110 ± 5°C, cooled, and stored in a desiccator. Fine pulverization also helps prevent unfused material because coarse particles require longer fusion times.

How is the appropriate flux selected for XRF fusion bead preparation?

Flux selection depends largely on the acid-base chemistry of the sample. Lithium tetraborate is relatively acidic and is suitable for samples containing basic oxides, such as limestone and cement. Lithium metaborate is more basic and is appropriate for materials containing acidic oxides, including silicates, rocks, and refractories. Mixtures of lithium tetraborate and lithium metaborate allow acidity/basicity and melting temperature to be adjusted. Sodium tetraborate melts at a relatively low temperature but is highly deliquescent and cannot be used when sodium must be determined.

Why are samples sometimes oxidized or calcined before fusion?

Metals, carbides, sulfides, carbon-containing materials, and similar samples can react with platinum vessels during fusion and cause irreversible damage, so oxidation is used to convert reactive components before high-temperature fusion. Calcination is also useful for carbonate-rich materials such as limestone and cement raw meal. Heating these materials before fusion releases carbon dioxide more gradually, reducing severe foaming, sample overflow, and trapped gas in the finished bead.

What causes fusion beads to crack or crystallize during cooling?

The cooling rate determines whether the molten sample remains a uniform glass. Cooling too rapidly can produce thermal stresses that crack the bead, while cooling too slowly can allow crystallization. A useful approach is relatively slow passive cooling initially, followed by active cooling. The optimum cooling time varies with the sample composition, flux, and dilution ratio.

What is the purpose of a releasing agent in XRF fusion bead preparation?

Iodides and bromides can be added to help the cooled bead separate from the mold. They alter surface behavior during fusion, reducing contact between the melt and vessel and also helping bubbles escape. Too little releasing agent can make the bead difficult to remove, while excessive amounts can cause the melt to form a crescent or ball rather than fully covering the vessel bottom. Residual bromine or iodine can also interfere with certain XRF spectral lines, so both type and quantity must be controlled.

How can trapped air and unfused material in XRF fusion beads be prevented?

Trapped air is especially problematic in calcium-rich carbonate materials because carbon dioxide is generated during heating. Calcination before fusion can substantially reduce this problem. Unfused material is more likely in high-silica or quartz-rich samples and can be minimized through fine pulverization and thorough mixing of the sample with the flux. If residue remains, changing the flux to better match the sample's acid-base characteristics or repeating the fusion may be necessary.

Read Article

Sample Preparation for X-ray Fluorescence Analysis IV.

Fusion bead method―part 1 basic principals

Mitsuru Watanabe

Summer 2015 Volume 31, No. 2 , 12-17

For the analysis of powders by XRF, sample inhomogeneity due to segregation, grain size and mineralogical effects influence X-ray intensity and can cause analysis errors. It is therefore recommended to analyze powder samples after fine pulverizing as described in “Sample Preparation for X-ray Fluorescence Analysis II. Pulverizing methods of powder sample.” However, when inhomogeneity can not be sufficiently removed by pulverization and more accurate analysis is required, fusion bead method is advisable.

The fusion bead method first established in the 1950s, has since progressed such that it is not only applicable to powders as oxides but also to non-oxides such as metals, carbides, sulfides which had previously been considered to be difficult. Characteristics of the fusion bead method are that (a) analysis error due to grain size and mineralogical effects can be removed, (b) matrix effect is reduced due to dilution, (c) standard samples can be prepared by mixing of reagents. Characteristics of fusion bead and pressed powder methods are compared.

In this article, general preparation methods, equipment, reagents and other important considerations for powders with typical grain size and drying conditions are described. If analysis must be performed in accordance to a specific standard test method, adhere to its prescribed conditions. In a following issue on fusion beads, various applications such as for ferroalloy, silicon carbide and copper concentrate samples will be reported.

Highlights

Fusion bead preparation minimizes grain-size and mineralogical effects in XRF and reduces matrix effects through dilution, making it useful when pulverization alone cannot provide sufficient sample homogeneity.
Successful preparation requires careful control of sample and flux composition, fusion temperature, cooling rate, and reagents to produce homogeneous, flat, glassy beads without crystallization, cracking, trapped air, or unfused material.
Fluxes, releasing agents, and oxidizing agents must be selected according to sample chemistry; proper reagent choice also protects platinum vessels and helps minimize analytical errors and spectral interferences.

Summary

Fusion bead preparation is an effective method for preparing powder samples for accurate X-ray fluorescence analysis, particularly when segregation, particle size, or mineralogical differences cannot be adequately eliminated by pulverization. Fusion removes grain-size and mineralogical effects and reduces matrix effects through dilution. It also allows synthetic standards to be prepared by mixing reagents, although dilution reduces measured X-ray intensity and preparation generally takes longer than the pressed powder method.

The general procedure begins with weighing dried sample, flux, and any necessary oxidizing agent, followed by thorough mixing. Samples containing metals, carbides, or sulfides may require oxidation at approximately 600–800°C before fusion. Materials containing significant carbonate, such as limestone and cement raw meal, can also benefit from calcination before fusion to reduce foaming, sample overflow, and trapped gas.

Fusion typically occurs at 1000–1200°C. Once the sample and flux have melted, agitation helps homogenize the melt and remove bubbles. Higher temperatures improve fluidity but can increase volatilization of both flux and analytes and shorten crucible life. After fusion, cooling must be controlled carefully: cooling too rapidly can crack the bead, while excessively slow cooling can cause crystallization. Initial passive cooling followed by active cooling provides a useful balance.

Flux selection depends strongly on sample chemistry. Lithium tetraborate is relatively acidic and is commonly used for materials containing basic oxides, including limestone and cement. Lithium metaborate is more basic and is suitable for materials rich in acidic oxides, such as silicates, rocks, and refractories, but is generally combined with lithium tetraborate because of its tendency to crystallize during cooling. Mixed lithium tetraborate/lithium metaborate fluxes provide adjustable acid-base characteristics and lower melting temperatures. Sodium tetraborate has an especially low melting point but is highly deliquescent and prevents determination of sodium in the sample.

Releasing agents such as iodides and bromides facilitate removal of the bead from the mold and can help eliminate bubbles. Their quantity must be carefully controlled because too little makes bead release difficult, while too much can prevent the melt from completely covering the vessel bottom. Residual halogens can also interfere with specific XRF analysis lines.

Oxidizing agents are particularly important for samples containing metals, carbon, or sulfur because these materials can react with platinum vessels and cause permanent damage. Nitrates are commonly used, while carbonates may be appropriate for ferroalloys and metals. Other additives can provide additional benefits; lithium fluoride lowers viscosity and melting temperature, while heavy-element oxides can further reduce matrix effects in samples covering wide elemental concentration ranges.

Good vessel condition is also essential. Residues, microcracks, and deformation can promote cracking, trapped air, poor bead release, and analytical error. Regular cleaning, polishing, and eventually recasting of platinum-alloy vessels help maintain consistent bead quality.

Frequently asked questions

: Fusion eliminates analytical errors associated with differences in particle size and mineralogy because the sample is converted into a homogeneous glass. Dilution with flux also reduces matrix effects, and standards can be prepared by mixing suitable reagents rather than finding reference materials closely matching the sample. The tradeoffs are reduced X-ray intensity because of dilution and a longer preparation time, typically about 15–30 minutes including weighing.
: A dried, finely pulverized sample is accurately weighed with flux and any required oxidizing or releasing agents. The components are thoroughly mixed and, when necessary, oxidized or calcined before fusion. The mixture is typically heated to approximately 1000–1200°C until completely molten, then agitated to homogenize the melt and remove bubbles. The molten material is cast or retained in a combined crucible/mold and carefully cooled to form a homogeneous glass bead.
: The sample should generally be pulverized to a particle size smaller than 106 µm (140 mesh). It is then dried for more than two hours at 110 ± 5°C, cooled, and stored in a desiccator. Fine pulverization also helps prevent unfused material because coarse particles require longer fusion times.
: Flux selection depends largely on the acid-base chemistry of the sample. Lithium tetraborate is relatively acidic and is suitable for samples containing basic oxides, such as limestone and cement. Lithium metaborate is more basic and is appropriate for materials containing acidic oxides, including silicates, rocks, and refractories. Mixtures of lithium tetraborate and lithium metaborate allow acidity/basicity and melting temperature to be adjusted. Sodium tetraborate melts at a relatively low temperature but is highly deliquescent and cannot be used when sodium must be determined.
: Metals, carbides, sulfides, carbon-containing materials, and similar samples can react with platinum vessels during fusion and cause irreversible damage, so oxidation is used to convert reactive components before high-temperature fusion. Calcination is also useful for carbonate-rich materials such as limestone and cement raw meal. Heating these materials before fusion releases carbon dioxide more gradually, reducing severe foaming, sample overflow, and trapped gas in the finished bead.
: The cooling rate determines whether the molten sample remains a uniform glass. Cooling too rapidly can produce thermal stresses that crack the bead, while cooling too slowly can allow crystallization. A useful approach is relatively slow passive cooling initially, followed by active cooling. The optimum cooling time varies with the sample composition, flux, and dilution ratio.
: Iodides and bromides can be added to help the cooled bead separate from the mold. They alter surface behavior during fusion, reducing contact between the melt and vessel and also helping bubbles escape. Too little releasing agent can make the bead difficult to remove, while excessive amounts can cause the melt to form a crescent or ball rather than fully covering the vessel bottom. Residual bromine or iodine can also interfere with certain XRF spectral lines, so both type and quantity must be controlled.
: Trapped air is especially problematic in calcium-rich carbonate materials because carbon dioxide is generated during heating. Calcination before fusion can substantially reduce this problem. Unfused material is more likely in high-silica or quartz-rich samples and can be minimized through fine pulverization and thorough mixing of the sample with the flux. If residue remains, changing the flux to better match the sample's acid-base characteristics or repeating the fusion may be necessary.

Contact Us

Whether you're interested in getting a quote, want a demo, need technical support, or simply have a question, we're here to help.

Contact Sales

Contact Service

Office Locations