Precious sample analysis using a releasing-agent-free technique

Introduction 

Precious samples (e.g., extraterrestrial materials and archaeological artifacts), because of their extremely high value, irreplaceability, and limited sample mass, have driven continuous advances in X-ray fluorescence (XRF) technology, particularly in non-destructive or micro-destructive analytical capability, sensitivity, and spatial resolution. Consequently, analytical objectives have progressively expanded from major-element determination to the accurate quantification of minor and trace elements.

Figure 1. Schematic illustration of the proposed swirling procedure.

For precious extraterrestrial samples, major elements are used to determine mineral composition, lithology, and chemical classification (e.g., meteorite classification), whereas trace elements provide critical insights into planetary evolution, parent-body differentiation, and shock metamorphism, enabling the extraction of high-value scientific information under non-destructive or micro-destructive conditions.

For archaeological ceramic samples, major elements are used to identify raw material provenance, firing technology, and glaze composition, while trace elements serve as “chemical fingerprints” for tracing clay sources, reconstructing trade and circulation pathways, restoring the original appearance of artifacts, and assisting in authenticity assessment.

Challenges

XRF analysis nevertheless offers significant advantages, including rapid analysis, non-destructive or micro-destructive measurement, simultaneous multi-element determination, minimal sample preparation, and suitability for micro-sampling and high-precision analysis of precious samples and archaeological ceramics.

Advantages

Despite these advantages, several analytical challenges remain, including the extremely limited sample mass, low detection sensitivity for light elements, matrix effects and mineral heterogeneity that compromise analytical accuracy, and difficulties in quantification due to the lack of matrix-matched standard reference materials.

Instrumentation and experimental conditions

Lithium borate glass discs were prepared using an M4 propane-fired automatic fluxer (Claisse, Quebec, Canada). The resulting fusion beads were analyzed using a wavelength-dispersive XRF spectrometer (AXIOS Minerals WD-XRF; Malvern Panalytical, Almelo, The Netherlands)(Table 1).

Element	Line	Crystal	Collimator (μm)	Detector	Tube filter	X-ray tube		Angle (2θ)			Counting Time (s)	PHD*
						kV	mA	Line	Bg1**	Bg2		LL^	UL†
Major
SiO2	Si Kα	PE 002	150	Flow	None	30	120	109.09	2.308	--	40	31	72
TiO2	Ti Kα	LiF 200	300	Flow	None	40	90	86.12	–0.898	--	50	36	64
Cr2O3 (Minor)	Cr Kα	LiF 200	300	Flow	None	40	90	69.34	–0.967	0.938	60	13	60
Al2O3	Al Kα	PE 002	300	Flow	None	30	120	144.78	2.943	--	40	31	69
TFeO	Fe Kα	LiF 200	300	Flow	Al (200 μm)	60	60	57.51	–0.991	1.752	40	18	61
MnO	Mn Kα	LiF 200	300	Flow	None	60	60	62.96	–0.906	0.755	50	15	55
MgO	Mg Kα	PX1	300	Flow	None	30	120	23.23	–2.330	2.420	60	28	73
CaO	Ca Kα	LiF 200	150	Flow	None	30	120	113.07	1.672	--	50	34	65
Na2O	Na Kα	PX1	700	Flow	None	30	120	28.07	–2.047	2.208	60	32	68
K2O	K Kα	LiF 200	300	Flow	None	30	120	136.66	–1.045	2.143	50	34	66
P2O5	P Kα	Ge 111	300	Flow	None	30	120	141.14	–1.910	2.726	60	33	67
Trace
Ni	Ni Kα	LiF 200	150	Flow	Al (200 μm)	60	60	48.66	–0.938	0.928	60	20	60
Zr	Zr Kα	LiF 200	150	Scint.	Al (200 μm)	60	60	22.50	–0.870	0.936	50	32	67
*PHD = pulse height distribution; **Bg = background; ^LL = lower limit; † UL = upper limit; Flow = gas flow proportional detector; Scint. = scintillation detector.

Table 1. Instrumental conditions for wavelength dispersive X-ray fluorescence spectrometry

Fusion bead preparation without a releasing agent

Eliminating the releasing agent represents not only a chemical optimization but also a precise regulation of the physical behavior of the molten fusion mixture. The research team recognized that the addition of a releasing agent excessively increased the fluidity of the lithium borate melt, causing uncontrolled spreading across the platinum mold and making bead formation difficult. By removing the releasing agent, an optimal balance between viscosity and fluidity was achieved.

Xue et al. further developed an improved operational procedure involving rapid manual rotation combined with in-situ cooling. This approach effectively resolved long-standing problems such as uneven bead thickness and poor bead formation, while also preventing thermal cracking associated with removing the mold from the fusion apparatus. Through meticulous control of these physical parameters, flat, homogeneous, and high-quality glass beads representative of the bulk-rock composition could still be successfully prepared from extremely small samples (~30 mg) at a sample-to-flux ratio of 1:100 (Figure 1).

Resolving spectral interference and improving analytical accuracy

In conventional fusion preparation methods, releasing agents such as NH₄Br are commonly added to facilitate bead release. However, this practice introduces bromine, whose Kα emission line significantly overlaps with the Kα line of Mn, thereby compromising the accurate determination of Mn concentrations.

To overcome this challenge, the research team adopted a releasing-agent-free approach, eliminating the source of spectral interference entirely and significantly improving Mn analytical accuracy (Figure 2).

Figure 2. XRF pattern of Br Kα partially overlapping with that of Mn Kα. The blank glass sample and Mica-Fe glass sample with releasing agent, and the Mica-Fe glass sample without releasing agent were scanned by XRF between 61.0° and 63.7° for 188 s with a voltage and current of 60 kV and 60 mA, respectively.

Conclusion

For the invaluable lunar regolith samples returned by the Chang'e 5 mission and Chang'e 6 mission missions, the fundamental principle of analytical science is to achieve “minimal consumption with maximum information output.” By combining a releasing-agent-free fusion strategy with a high dilution ratio, this method enables efficient and synergistic acquisition of multiple categories of geochemical information from a single precious sample (Table 2).

This environmentally friendly, minimally destructive, and streamlined sample-preparation strategy not only provides reliable analytical data for current lunar research but also establishes a valuable “gold-standard” protocol for future deep-space exploration missions involving even more precious returned materials, such as Mars and asteroid samples.

	CE6-C0000YJFM00102		CE6-C0000YJFM00103
Element	Mean	U**	Mean	U
XRF Major (m/m %)
SiO2	45.74	0.28	45.83	0.29
TiO2	2.70	0.12	2.70	0.12
Al2O3	14.32	0.38	14.37	0.39
Cr2O3 (minor element)	0.25	0.04	0.25	0.04
TFeO	17.24	0.54	17.14	0.54
MnO	0.222	0.010	0.217	0.008
MgO	7.08	0.31	7.00	0.30
CaO	11.84	0.37	11.95	0.37
Na2O	0.28	0.03	0.23	0.03
K2O	0.073	0.011	0.079	0.013
P2O5	0.072	0.013	0.068	0.013
Total	99.82	--	99.84	--
XRF Trace (µg/g)
Ni	260	34	261	32
Zr	117	19	115	14
*n is the number of replicates; **U is the expanded uncertainty calculated with an expansion factor of 2

Table 2. Elemental composition of the lunar soil samples CE6 (n=3*)

Related resources

Download the resources below to learn more:

1. Tandem Elemental Analysis of Chang’e‑6 Lunar Soil Samples Using X‑ray Fluorescence and Femtosecond Laser Ablation Inductively Coupled Plasma Mass Spectrometry
2. Quantitative verification of 1:100 diluted fused glass beads for X-ray fluorescence analysis of geological specimens