XRF Analysis of Gold and Precious Metals: Fineness, Limits and Fire Assay

XRF analysis is today the fastest way to estimate the fineness of gold and precious alloys without destroying the item. In a few seconds it returns the elemental composition and the millesimal value. Yet its very speed often makes people forget its most important physical limit: XRF measures the surface only. Understanding where the reliability of the instrument ends and the territory of fire assay begins is what separates a serious inspection from a reading accepted without reservation.

The principle: X-ray fluorescence on precious alloys

In XRF (X-Ray Fluorescence) an X-ray source excites the atoms of the material, which re-emit characteristic X-rays at energies specific to each element. The detector measures the energy and intensity of these lines and the software derives the composition. For gold the reference lines are typically the L series (Au-Lα around 9.7 keV), because the K lines of gold are at high energy and harder to excite with the compact tubes of handheld instruments.

The practical consequence is decisive: the characteristic X-rays leaving the piece come from a very thin layer. In gold, which is dense and high atomic number (Z=79), the depth from which the useful signal originates is on the order of a few microns to a few tens of microns. Everything below that is invisible to the instrument. The X-ray mass attenuation coefficient tables published by NIST explain precisely why absorption in heavy metals confines the measurement to the surface.

Karats and millesimal fineness: reading the value correctly

The fineness of gold is expressed in two equivalent ways. The karat indicates the gold fraction in twenty-fourths: 24K is pure gold, while the most common commercial alloys are summarised below. Millesimal fineness (or legal fineness) expresses the same information as parts of fine gold per thousand. XRF provides the value directly in percentage or millesimal, from which the karat is derived. Karat definitions are also documented by the World Gold Council.

Caution: the XRF reading in millesimal is the composition of the analysed layer, not necessarily the legal fineness of the whole item. The distinction matters greatly when the surface does not represent the bulk.

XRF and cupellation (fire assay): two different roles

The reference method for determining gold fineness remains cupellation, or fire assay, standardised by ISO 11426 (determination of gold in gold jewellery alloys by the cupellation method). It is a destructive technique that melts a sample taken from the item, separates the base metals and weighs the residual gold: it provides the bulk fineness with very low uncertainty, on the order of fractions of a millesimal, and is the reference for official certification. Comparing the two approaches clarifies their respective scopes.

In practice XRF is the tool for daily screening and sorting; cupellation is the arbiter when a legal value is required or when XRF flags an anomaly. The two methods do not compete: they complement each other.

The key limit: it is a surface measurement

Because the signal comes from the first few microns, some situations lead XRF to overestimate the real fineness:

• Plating and electrolytic gilding: a gold layer of a few microns on a base-metal core can be enough to make the instrument read a high fineness.
• Rhodium plating: white gold is often rhodium-plated at the surface; excess rhodium in the reading signals that the coating is being measured, not the underlying alloy.
• Surface enrichment (depletion gilding): treatments that dissolve base metals at the surface leave a gold-richer layer that is not representative of the bulk.
• Solder joints and junctions: the solder point of a chain or a ring may use a different alloy (often lower fineness) than the body of the jewel.

Detectable elements and the role of detector and collimation

An XRF analyzer for precious metals typically detects gold (Au), silver (Ag), copper (Cu), palladium (Pd), platinum (Pt), rhodium (Rh), zinc (Zn), nickel (Ni), iridium and other alloying elements. The quality of the measurement depends on two often-underestimated factors:

• SDD (Silicon Drift Detector): high energy resolution is needed to separate close lines, for example those of gold, platinum and palladium, preventing one from being mistaken for another and skewing the fineness calculation.
• Beam collimation: on small items such as rings, chains or settings, a well-collimated beam (small spot) avoids measuring the holder, the air around the piece or solder joints, which would otherwise distort the result.

Reference standards and hallmarking

Beyond ISO 11426 for gold fire assay, international standardisation covers the determination of precious metals with reference methods: ISO 15093 covers the determination of gold, platinum and palladium in high-fineness alloys, while ISO 15096 concerns the determination of high-fineness silver. On the legal side, hallmarking follows in many countries the Vienna Convention on the Control and Marking of Articles of Precious Metals: for the official punching of fineness the reference remains the standardised analytical methods, not XRF screening. In the investment bullion market, refining and minimum-fineness requirements are defined for example by the LBMA Good Delivery standard (London Bullion Market Association).

Typical applications

• Gold-buying shops: fast, non-destructive screening of incoming jewellery and scrap, with verification of plating before valuation.
• Jewellers and counters: confirmation of declared fineness, control of batches and returns, distinction between solid and plated gold.
• Pawnshops: objective and repeatable valuation at the moment of appraisal, with traceability of the measurement.
• Refineries and toll refining: control of incoming and outgoing material, verification of fineness before and after processing, reduction of disputes.
• Precious-metal recycling: alloy sorting and separation of materials by fineness.

Common mistakes to avoid

• trusting the XRF reading on plated or rhodium-plated items without a depth check;
• measuring on solder joints or settings instead of on the body of the alloy;
• confusing surface fineness with legal bulk fineness;
• using too large a spot on small items, including holder or air in the measurement;
• treating XRF as a substitute for cupellation for official certifications;
• not cleaning the surface of dirt, oils or residues that alter the reading of light elements.

To explore the instrument and available configurations, see the PITECH XRF/PMI material analysis page and the Elvatech XRF/PMI analyzers product page. A related topic, but on the steel side, is covered in the article LIBS vs XRF for carbon measurement in PMI. PITECH supports a neutral technical-commercial assessment, comparing application, elements, volumes and the need to distinguish solid from plated before proposing the most coherent XRF configuration.