Development of gold reference materials for in situ Ag isotopic analysis

Abstract

Microanalysis of silver isotopes in gold (Au) by laser ablation multi-collector inductively coupled plasma mass spectrometry (LA-MC-ICP-MS) has been an important tool to reveal the source and enrichment processes of Au. However, preparing an Au-matrix reference material with homogeneous Ag isotope compositions at a micron scale remains a major challenge. This study presents a novel method to synthesize a homogeneous gold reference material, named NWU-Au, via flux-free fusion. Moreover, a natural gold reference material, DD440, is prepared for in situ Ag isotope measurement by LA-MC-ICP-MS. Elemental mapping obtained by electron probe microanalysis indicates that both gold reference materials have homogeneous Au and Ag contents, while LA-MC-ICP-MS measurements further demonstrate their homogeneous distribution of Ag isotopes. The intermediate precision of δ¹⁰⁹Ag was better than 0.06‰ (2s) for NWU-Au and DD440, which were suitable to serve as matrix-matched calibration materials. Silver isotopic compositions of six gold samples were also determined using LA-MC-ICP-MS with matrix-matched calibration, and the δ¹⁰⁹Ag_NIST978a values were consistent with those obtained by SN-MC-ICP-MS, suggesting that LA-MC-ICP-MS can accurately determine Ag isotopic ratios in gold samples. Thus, this synthesis method, along with an analytical approach, enables potential determination of spatially resolved Ag isotope compositions at the mineral scale to reveal the source and enrichment processes of Au in the Au deposit.

---

1. Introduction

Gold (Au) only has one naturally occurring isotope (¹⁹⁷Au), making it unsuitable as an isotopic tracer for chemical reactions or source regions.¹ Gold and silver can form a continuous alloy series due to their same atomic radii. Gold typically contains 5–20% Ag, with some exceeding 50% in some deposits.² Previous studies have shown that the Ag isotope composition of gold grains can reveal the source and enrichment processes of Au.^1,3,4 The Ag isotope ratios are highly heterogeneous in Au grains, with δ¹⁰⁹Ag ranging from −0.83 to +0.83‰ relative to the NIST SRM978a silver reference material.^1,3–6 However, such subtle Ag isotope fractionation cannot be distinguished using thermal ionisation mass spectrometry due to the limited analytical precision (1–1.5‰). Over the past two decades, accurate and precise Ag isotope measurements have been achieved by multi-collector inductively coupled plasma mass spectrometry (MC-ICP-MS) in the form of pure solution with column chromatography.^5,7–11 Although the precision can be better than 0.05‰ (2s), this solution nebulizer (SN)-MC-ICP-MS approach requires complex, time-consuming sample preparation and high Ag yield (>99%) and cannot resolve the Ag isotopic variations on a grain/sub-grain scale. To overcome these limitations, an in situ microbeam measurement technique, namely femtosecond laser ablation (fsLA)-MC-ICP-MS, offers high sensitivity and spatial resolution, enabling Ag isotope analysis directly at the grain or sub-grain scale.

fsLA-MC-ICP-MS has been widely used for in situ determination of non-traditional stable isotope ratios in geological samples, such as Li,^12,13 Mg,^14,15 Fe,¹⁶ Ti,^17–19 Cu,^20,21 Zn,²² Ni,²³ Mo,²⁴ Ag,²⁵ Sb,^26,27 and Sn.^28,29 However, the fsLA-MC-ICP-MS analysis of stable isotope ratios requires calibration against the equivalent analysis of matrix-matched reference materials to correct the instrument-induced mass discrimination. Therefore, a high-quality gold reference material is essential to obtain accurate and precise δ¹⁰⁹Ag ratios of a gold sample. Recently, synthetic processes involving fast hot-pressing (FHP) sintering and high-temperature and high-pressure (HTHP) techniques have been developed for the preparation of matrix-matched in-house reference materials for in situ isotope analysis.^28,30–35 Zhang et al. (2023)²⁸ combined ultrafine powder preparation and HTHP sintering techniques to synthesize a matrix-matched cassiterite reference material for in situ Sn isotope analysis. Nevertheless, the amount of HTHP synthetic reference material is limited. Comparatively, the FHP technique can produce large quantities of matrix-matched reference materials, in which the sintering necks between ultrafine grains can provide sufficient mechanical stability.³⁶ Tian et al. (2024a)³⁵ used the FHP sintering technique to synthesize molybdenite reference materials for Mo isotope analysis. This synthetic molybdenite has homogeneous Mo isotopic compositions and a compact structure, making it a suitable and accessible reference material for LA-MC-ICP-MS Mo isotope analysis. Moreover, combining ultrafine powder preparation and flux-free fusion has also been used to synthesize reference materials for microanalysis.^14,37–39 For instance, with flux-free fusion, Yang et al. (2023)³⁸ prepared six synthetic clinopyroxene reference materials exhibiting homogeneous Li isotope compositions. In summary, ultrafine initial powder combined with specific temperature and pressure has become the method of choice for preparing various reference materials.

To the best of our knowledge, there is no well-characterised and high-quality gold reference material for in situ Ag isotope measurement. In this study, a new gold material, named NWU-Au, synthesized by flux-free fusion, and a natural gold sample, named DD440, were prepared for in situ Ag isotope measurement. The major elemental, trace elemental, and Ag isotopic homogeneity were tested by electron probe microanalysis (EPMA), LA-ICP-MS, and fsLA-MC-ICP-MS, respectively. The results suggest that NWU-Au and DD440 are sufficiently texturally and chemically homogeneous to be used as Au reference materials for in situ Ag isotope analysis. The Au-matrix reference materials were used to determine the Ag isotopic compositions of six Au samples. The matrix-matched calibration results from fsLA-MC-ICP-MS were highly consistent with those obtained by SN-MC-ICP-MS analysis.

2. Preparation of candidate gold reference materials

Two types of samples were investigated in this study: a synthetic gold-matrix sample (NWU–Au) and a natural gold sample (DD440).

2.1 Synthetic gold-matrix sample (NWU–Au)

The synthetic gold sample NWU-Au was prepared in the State Key Laboratory of Continental Evolution and Early Life (SKLCEEL), Northwest University, China. Ultrafine Au powder and ultrafine Ag powder were purchased from Zhongkianuo (Beijing) New Material Technology Co., LTD. The original Au and Ag particle sizes were d₅₀ = 1.5 µm and d₅₀ = 0.5 µm, respectively, sufficiently small for in situ isotope analysis. The Au powders doped with some Ag powder were then manually milled for 30 min using an agate mortar. Approximately 1.5 g of mixed powders was weighed and transferred into a ceramic crucible. The ceramic crucible was then placed into a high-temperature furnace (Thermo Fisher Scientific™), set at 1090 °C, and was then heated for 3 min to melt the Au powder. Then, the ceramic crucible was removed from the furnace using tongs and rapidly immersed in deionized water to separate the fused Au nugget (Fig. 1b). Finally, the obtained Au nugget was made into an epoxy mount for EPMA and laser ablation analysis.

Fig. 1 Photograph of natural gold DD440 (a) and synthesized gold NWU-Au (b).

2.2 Natural gold sample (DD440)

This natural gold sample was collected from the Qianhe gold deposit, a typical tectonic-altered gold deposit in Qinling, Central China. Sample DD440 was collected from the second hydrothermal stage, which mainly consists of ore minerals including pyrite, sphalerite, chalcopyrite, galena and native gold. Common gangue minerals in this stage include quartz, sericite, barite, and chlorite. Gold occurs primarily as native gold and argillite within the porous and disseminated pyrite grains or interstices of this stage. Sample DD440 (Fig. 1a) was further crushed, and 0.3 g of gold grains (100–1000 µm) were picked. Randomly selected gold grains were cast into one epoxy mount and then polished three times with gradually finer diamond paste. The mount was washed first in deionized water and then placed on an electric plate and heated at 40 °C for 2 h. The gold mount was then carbon-coated before electron probe microanalysis (EPMA).

3. Analytical method and instrumentation

All high-purity acids used in this study were distilled twice using a Savillex TM DST-100 sub-boiling still (Minnetonka, MN, USA) and then diluted using ultrapure water (resistivity: 18.2 MΩ cm). The SRM 978a silver and SRM 3138 Pd reference materials were purchased from the National Institute of Standards and Technology (NIST).

3.1 Electron probe microanalysis and LA-ICP-MS analysis

Major element analyses of gold were performed using EPMA (JEOL JXA-8230) at the SKLCEEL. The EPMA instrument was equipped with five spectrometers, and the analyses were performed with an accelerating voltage of 15 kV, a beam current of 10 nA, and a beam diameter of 1 µm. Simple substances (Au and Ag) were used as reference materials for quantifying Au and Ag signals. High-resolution 2-D elemental X-ray mapping was applied for one random gold grain and one random area using the same EPMA with an accelerating voltage of 15 kV, a beam current of 20 nA, and a beam diameter of 2 µm. The dwell time was set to 10 ms for each point. Gold Mα and silver Lα were analyzed using a PETJ crystal.

The trace element contents of gold were determined using a 193 nm ArF excimer laser system (RESOLution S155-LR, ASI) coupled with an Agilent 7900 ICP-MS at the SKLCEEL. The analytical approach was similar to that outlined in Liang et al. (2023)⁴⁰ with a laser spot of 43 µm, a repetition rate of 4 Hz, and an energy density of 6.0 J cm⁻². Helium was used as a carrier gas for the laser ablation process, and it was mixed with argon before entering the squid signal smoothing device and the ICP-MS. NIST SRM 610 reference glass was used as the calibration material. NIST SRM 612, BHVO-2G, and BCR-2G were analyzed for data quality control. Offline selection and integration of background and ablation signals, and time-drift correction and quantitative calibration were performed using ICPMSDataCal.⁴¹ Detailed instrumental parameters of LA-ICP-MS are summarised in Table 1.

Table 1 Operational parameters for ICP-MS and MC-ICP-MS instruments and nanosecond and femtosecond laser ablation systems

Femtosecond laser ablation system		Nanosecond laser ablation system
Laser type	Light conversions, Pharos HE	ArF excimer laser
Wavelength	266 nm	193 nm
Pulse duration	<130 fs	15 ns
Energy density	1.0 J cm^−2	6 J cm^−2
Carrier gas	He, 0.4 L min^−1	He, 0.28 L min^−1
Ablation mode	Line scan	Single spot
Laser beam	9–20 um	44 µm
Pulse repetition rate	8 Hz	5 Hz
Measuring time	58 s	45 s
Neptune Plus™ MC-ICP-MS		Agilent 7900 ICP-MS
Cooling gas	15.0 L min^−1	RF power	1350 W
Auxiliary gas	0.85 L min^−1	Plasma gas	13.0 L min^−1
RF power	1200 W	Auxiliary gas	0.8 L min^−1
RF reflected	1 W	Make-up gas	∼1.1 L min^−1
Interface cones	Nickel standard sampler cones and ‘H’ skimmer cones	Sampler cone	Ni orifice, 1.0 mm
Resolution mode	Low resolution	Skimmer cone	Ni orifice, 0.4 mm
Integration times	0.524 s for LA-MC-ICP-MS analysis, 4.194 s for LA-MC-ICP-MS analysis	Dwell time	15 ns

---

3.2 Bulk solution nebulizer MC-ICP-MS

The bulk SN-MC-ICP-MS analyses were conducted on the Neptune Plus MC-ICP-MS at the SKLCEEL. The sample digestion and column chromatography followed the procedure established by Brügmann et al. (2019)⁵ and are briefly summarised here. Approximately 1 mg of gold samples was weighed in 15 mL pre-cleaned PFA vials and dissolved in 2 mL of aqua regia at 80 °C for 2 days in the capped vial. The samples were dried down and converted to chloride form by adding 1 mL of concentrated HCl. A small split of this solution, containing 30 micrograms of Ag, was diluted to 6 mol L⁻¹ HCl for the subsequent chromatographic procedures. Purification of Ag from Au was performed on 0.2 mL AG1-X8 anion exchange resin. Two millilitres of 6 mol L⁻¹ HCl were added to the column for conditioning. Then, the dissolved sample containing about 100 µg of Ag was loaded into the column, followed by loading 4 mL of 6 mol L⁻¹ HCl to collect Ag. Finally, the collected Ag fractions were evaporated to dryness and dissolved with 2% v/v HNO₃ for subsequent mass spectrometry.

The δ¹⁰⁹Ag ratios were determined by using a double-focusing Thermal Scientific™ Neptune Plus MC-ICP-MS at the SKLCEEL. L3, L2, L1, C, H1, H2 and H3 Faraday cups were used to collect ¹⁰⁴Pd, ¹⁰⁵Pd, ¹⁰⁶Pd, ¹⁰⁷Ag, ¹⁰⁸Pd, ¹⁰⁹Ag, and ¹¹⁰Pd, respectively. A wet plasma setup with a standard sample cone, an H skimmer cone, a cyclone/double-pass spray chamber and a PFA nebulizer was used to determine Ag isotopic ratios. The standard-sample bracketing (SSB) method, combined with NIST SRM 3138 Pd as an internal reference material, was selected to correct the instrumental mass bias. The internal standard solution NIST SRM 987a served as a bracketing calibrator. Finally, the measured Ag isotopic ratio is expressed relative to the reference material NIST 978a. Detailed instrumental parameters of the Neptune Plus are summarised in Table 1.

3.3 In situ measurement by femtosecond laser ablation MC-ICP-MS

The fsLA-MC-ICP-MS analyses were performed on a Neptune Plus MC-ICP-MS combined with NWRFemtoUC Dualwave femtosecond laser ablation (ESI, USA) at the SKLCEEL. The femtosecond laser ablation system consisted of a Light Conversion Pharos HE laser, laser transmission system, gas flow control system, sample station, microscope system and software control system. All measurements were performed in line-scan mode with a diameter of 15 µm, 10 Hz laser repetition rate, 1 µm s⁻¹ scan speed and 1.0 J cm⁻² laser energy. Ultrahigh-pure helium was used as a carrier gas for the laser ablation process. The flow rates of argon and ultra-pure helium were 1000 and 450 mL min⁻¹, respectively.

For in situ Ag isotope measurements, Ni cones were employed, and the mass spectrometer was operated in the low mass resolution mode. Center and H2 Faraday cups were used to collect ¹⁰⁷Ag and ¹⁰⁹Ag, respectively. Each Ag isotope measurement was performed in time-resolved mode, acquiring 100 cycles of each 0.524 s. The washout time between two Ag isotope measurements was 60 s. The background obtained on ¹⁰⁷Ag and ¹⁰⁹Ag was less than 2 mV. The SSB method was selected to correct the instrumental mass bias. Finally, Ag isotope ratios were firstly expressed as a per mil deviation relative to the bracketing standard and then converted into values relative to the reference material NIST SRM 978a. Detailed instrumental parameters of NWRFemtoUC Dualwave femtosecond laser ablation are summarized in Table 1.

4. Results

4.1 Electron probe microanalysis and LA-ICP-MS results

EPMA element distribution mapping was performed on a random area (300 µm × 300 µm) on NWU-Au and DD440 (Fig. 2). The elemental maps did not display sector zoning of Au and Ag. A total of 70 EPMA measurements were conducted on NWU-Au and DD440 at the SKLCEEL, including profile analyses on five DD440 Au grains and random spot analyses on NWU-Au, which were used to check the homogeneity (Table 2). Each profile analysis included seven analytical spots and was conducted on five gold grains. The Au and Ag contents of these five grains are consistent, suggesting fair homogeneity in major element abundances within DD440. Quantitative compositional analyses in DD440 Au grains show that the major element abundances are homogeneous. The results show that DD440 is composed of 86.68 ± 0.76 wt% Au and 13.49 ± 0.85 wt% Ag. Moreover, 35 random analyses on NWU-Au indicated that the Au and Ag contents in NWU-Au are homogeneous, with the Ag contents ranging from 34.67 to 36.62 wt% and the Au contents varying from 62.75 to 64.84 wt%. Overall, NWU-Au is composed of 63.55 ± 0.54 wt% Au and 36.13 ± 0.38 wt% Ag, respectively.

Fig. 2 Elemental maps generated by EPMA showing the distribution of Au and Ag in one random area (300 µm × 300 µm). (a) Ag of DD440; (b) Au of DD440; (c) Ag of NWU-Au; (d) Au of NWU-Au.

Table 2 EPMA measurement results for major elements (g 100 g⁻¹) in NWU-Au and DD440

Grain ID	Comment	Au	Ag	Total
NWU-Au	Spot 01	64.21	35.82	100.03
NWU-Au	Spot 02	63.29	35.68	98.96
NWU-Au	Spot 03	64.69	35.93	100.62
NWU-Au	Spot 04	63.70	36.49	100.19
NWU-Au	Spot 05	63.53	36.22	99.75
NWU-Au	Spot 06	63.23	36.61	99.84
NWU-Au	Spot 07	63.89	36.10	99.99
NWU-Au	Spot 08	63.39	35.69	99.08
NWU-Au	Spot 09	63.77	36.51	100.29
NWU-Au	Spot 10	63.43	36.37	99.81
NWU-Au	Spot 11	63.07	36.03	99.10
NWU-Au	Spot 12	62.88	36.28	99.16
NWU-Au	Spot 13	62.71	36.62	99.34
NWU-Au	Spot 14	62.99	36.26	99.25
NWU-Au	Spot 15	63.43	35.56	98.99
NWU-Au	Spot 16	63.21	35.89	99.10
NWU-Au	Spot 17	63.32	36.45	99.77
NWU-Au	Spot 18	63.62	36.34	99.96
NWU-Au	Spot 19	63.09	36.06	99.14
NWU-Au	Spot 20	63.46	36.03	99.50
NWU-Au	Spot 21	64.48	36.45	100.93
NWU-Au	Spot 22	63.85	36.24	100.09
NWU-Au	Spot 23	63.92	36.28	100.20
NWU-Au	Spot 24	63.20	36.33	99.53
NWU-Au	Spot 25	63.73	36.62	100.35
NWU-Au	Spot 26	63.89	35.83	99.71
NWU-Au	Spot 27	63.15	36.12	99.27
NWU-Au	Spot 28	63.98	35.78	99.76
NWU-Au	Spot 29	63.62	36.27	99.89
NWU-Au	Spot 30	62.99	36.36	99.35
NWU-Au	Spot 31	62.75	36.28	99.04
NWU-Au	Spot 32	64.36	36.16	100.52
NWU-Au	Spot 33	63.79	35.87	99.66
NWU-Au	Spot 34	62.86	36.25	99.11
NWU-Au	Spot 35	64.84	34.67	99.50
	Mean	63.55	36.13
	1s	0.54	0.38
DD440 grain01	Spot 36	86.78	13.50	100.28
	Spot 37	86.45	13.44	99.89
	Spot 38	86.30	13.45	99.76
	Spot 39	86.13	13.62	99.75
	Spot 40	86.57	13.37	99.94
	Spot 41	87.01	13.61	100.62
	Spot 42	86.77	13.64	100.41
DD440 grain02	Spot 43	86.92	13.82	100.74
	Spot 44	86.77	13.69	100.46
	Spot 45	86.15	13.53	99.68
	Spot 46	88.45	11.48	99.93
	Spot 47	88.56	11.34	99.90
	Spot 48	88.05	11.22	99.27
	Spot 49	86.93	13.84	100.77
DD440 grain03	Spot 50	86.46	13.54	100.00
	Spot 51	86.36	13.74	100.09
	Spot 52	86.53	13.51	100.03
	Spot 53	86.85	13.41	100.26
	Spot 54	86.87	13.50	100.37
	Spot 55	86.52	13.79	100.31
	Spot 56	86.13	13.49	99.62
DD440 grain04	Spot 57	85.60	14.01	99.61
	Spot 58	85.29	15.15	100.44
	Spot 59	87.17	13.60	100.76
	Spot 60	87.30	13.38	100.68
	Spot 61	87.47	13.27	100.74
	Spot 62	87.12	13.35	100.47
	Spot 63	86.51	13.53	100.04
DD440 grain05	Spot 64	86.78	13.34	100.11
	Spot 65	85.19	15.27	100.46
	Spot 66	85.45	15.06	100.51
	Spot 67	87.37	12.64	100.01
	Spot 68	86.53	13.11	99.63
	Spot 69	85.97	14.08	100.04
	Spot 70	86.45	13.84	100.30
	Mean	86.68	13.49
	1s	0.76	0.85

---

The without-chromatographic purification procedure prior to in situ Ag isotopic analysis by fsLA-MC-ICP-MS requires checking the effects of isobaric interferences on the accuracy of the determined Ag isotopic ratios. For Ag isotope ratio measurements in gold samples, the major potential spectral interferences arise from polyatomic ions (⁶⁷Zn⁴⁰Ar⁺, ⁷¹Ga³⁶Ar⁺, ⁹¹Zr¹⁶O⁺, ⁸⁹Y¹⁸O⁺, ⁶⁹Ga⁴⁰Ar⁺, ⁹³Nb¹⁶O⁺, and ⁹¹Zr¹⁸O⁺). The LA-ICP-MS results indicate that the occurrence of Zn, Ga, Zr, Y, and Nd in Au is rare and the mass fractions are up to 3.7 µg g⁻¹. The mean contents of Zn, Ga, Zr, Y, and Nd were less than 1 µg g⁻¹ in NWU-Au and DD440 (Table S1). Therefore, their influence on the in situ Ag isotope measurement is considered negligible.

4.2 Homogeneity of NWU-Au and DD440

Due to the lack of a gold reference material, a self-calibrated method using the SSB approach was selected to correct the instrument mass bias for NWU-Au and DD440 using fsLA-MC-ICP-MS. All Ag isotope ratios measured in random areas over five days are plotted in Fig. 3. For the synthesized NWU-Au, day 1 consisted of 45 Ag isotopic measurements, yielding δ¹⁰⁹Ag values from −0.05‰ to 0.06‰, with a mean value of 0.00 ± 0.04‰ (2s, n = 45). Likewise, in situ Ag isotope measurements on day 2 yielded δ¹⁰⁹Ag values from −0.08‰ to 0.06‰, with a mean value of 0.00 ± 0.06‰ (2s, n = 39). In situ Ag isotope measurements on day 3, day 4, and day 5 yielded δ¹⁰⁹Ag values from −0.05‰ to 0.05‰, with a mean value of 0.00 ± 0.04‰ (2s, n = 54), 0.00 ± 0.04‰ (2s, n = 68), and 0.00 ± 0.04‰ (2s, n = 82), respectively. The δ¹⁰⁹Ag precision expressed as 2σ standard deviation in five days, determined using fsLA-MC-ICP-MS, was better than 0.06‰ without systematic variation, indicating good silver isotopic homogeneity of NWU-Au. For natural gold DD440, all in situ Ag isotope ratios from sixty grains in five days are plotted in Fig. 4. Thirty-seven Ag isotopic measurements on day 1 yielded δ¹⁰⁹Ag values from −0.03‰ to 0.03‰, with a mean value of 0.00 ± 0.03‰ (2s, n = 37). Likewise, the mean δ¹⁰⁹Ag values on day 2, day 3, day 4, and day 5 are 0.00 ± 0.04‰ (2s, n = 45), 0.00 ± 0.05‰ (2s, n = 94), −0.01 ± 0.06‰ (2s, n = 109), and −0.02 ± 0.06‰ (2s, n = 35), respectively. The δ¹⁰⁹Ag precision for DD440 was better than 0.06‰ without systematic variation, indicating good silver isotopic homogeneity of DD440.

Fig. 3 δ ¹⁰⁹Ag values of NWU-Au on five different days obtained using fsLA-MC-ICP-MS analyses. The gray areas represent the intermediate measurement precision (2s) of Ag isotope compositions of each day.

Fig. 4 δ ¹⁰⁹Ag values of DD440 on five different days obtained using fsLA-MC-ICP-MS analyses. The gray areas represent the intermediate measurement precision (2s) of Ag isotope compositions of each day.

4.3 SN-MC-ICP-MS measurement results

Eight microdrilled fractions of NWU-Au and five randomly selected DD440 Au grains were analyzed for Ag isotope ratios by SN-MC-ICP-MS, and the results of bulk Ag isotope measurements are listed in Table 3. Repeated measurements were conducted on GSB-Ag solution standards to check the instrumental stability before SN-MC-ICP-MS analysis. The mean δ¹⁰⁹Ag_NIST978a ratios of GSB-Ag solution are −0.03 ± 0.02‰ (2s, n = 15), which are consistent with the published values.²⁵ All the δ¹⁰⁹Ag_NIST978a ratios of NWU-Au ranged from −0.20‰ to −0.25‰, with a mean δ¹⁰⁹Ag_NIST978a of −0.23 ± 0.04‰ (2s, n = 8). A total of five single-grain SN-MC-ICP-MS measurements were used to check the homogeneity of the Ag isotopic compositions of DD440. The mean δ¹⁰⁹Ag_NIST978a value was 0.42 ± 0.03‰ (2s, n = 5). All δ¹⁰⁹Ag ratios ranged from 0.40‰ to 0.44‰, which agrees well with the bulk Ag isotope analyses within reported precision. The good intermediate measurement precision of the δ¹⁰⁹Ag_NIST978a ratio by single-grain analysis shows that there is no Ag isotopic variability among DD440 single grains.

Table 3 Sliver isotope composition of NWU-Au and DD440 determined by SN-MC-ICP-MS

Sample	δ ^109AgNIST978a	2s
NWU-Au-01	−0.20	0.04
NWU-Au-02	−0.25	0.02
NWU-Au-03	−0.23	0.03
NWU-Au-04	−0.25	0.02
NWU-Au-05	−0.23	0.03
NWU-Au-06	−0.21	0.02
NWU-Au-07	−0.20	0.02
NWU-Au-08	−0.24	0.03
Mean	−0.23	0.04
DD440-1	0.42	0.02
DD440-2	0.40	0.04
DD440-3	0.44	0.02
DD440-4	0.41	0.03
DD440-5	0.40	0.02
Mean	0.42	0.03

---

4.4 Accuracy and intermediate precision

Six additional gold samples were randomly selected from the Qianhe gold deposit, Gongyu gold deposit, Qinnan gold deposit and Qiyugou gold deposit in Henan Province. The six studied gold samples recorded a large variety in Ag contents ranging from 1.98 to 8.49 wt% and were further analyzed by fsLA-MC-ICP-MS over a 5 month period (Fig. 5), where the millimetre grade profiles of three gold samples (Au1, Au2, and Au3, ∼3 mm) were also measured in different measurement sessions (Fig. 5a–c). In order to avoid Ag isotope mass fractionation due to the effect of concentration mismatch, Ag signals of the samples and bracketing standard were always adjusted to within 20% of each other by changing the laser diameter. A correction strategy with the matrix-matched reference materials NWU-Au and DD440 as the bracketing calibrators has been applied for the characterization, and their δ¹⁰⁹Ag_NIST978a values of Au1 (0.52 ± 0.02‰, n = 4), Au2 (0.39 ± 0.02‰, n = 4), Au3 (0.48 ± 0.03‰, n = 4), Au4 (−0.02 ± 0.03‰, n = 4), Au5 (0.18 ± 0.03‰, n = 4), and Au6 (0.52 ± 0.03‰, n = 4) obtained by SN-MC-ICP-MS were adopted to normalize the Ag isotopic results. Overall, the δ¹⁰⁹Ag_NIST978a values from fsLA-MC-ICP-MS are identical to the results obtained from SN-MC-ICP-MS within uncertainties (Fig. 6), confirming that the proposed fsLA-MC-ICP-MS technique can accurately and precisely determine the Ag isotopic compositions of gold. The intermediate precision (2s) of LA-MC-ICP-MS is mainly related to the instrumental stability and the homogeneity of the bracketing calibrator. With one exception (i.e., ±0.22‰ for Au6), the intermediate precisions in δ¹⁰⁹Ag_NIST978a for these five gold samples are in the range of ±0.05‰ to ±0.12‰ by fsLA-MC-ICP-MS in different sessions (Fig. 5a–e), which is comparable to the intermediate precision for the Ag isotope in the published study (±0.05–0.08‰ for δ¹⁰⁹Ag_NIST978a).²⁵ All measurements of the five gold fragments and profiles can be used to evaluate the homogeneity of their Ag isotopic compositions. In particular, the intermediate precisions of δ¹⁰⁹Ag_NIST978a for the five gold samples range from ±0.05‰ to ±0.12‰ as determined by fsLA-MC-ICP-MS (Fig. 5a–e). All data were carefully checked against our analytical precision of ±0.10‰ to confirm homogeneity. In general, Au6 showed greater heterogeneity (i.e., 2sd > ±0.20‰) in two measurement sessions (Fig. 5f) and is thus not recommended as a quality control sample during LA-MC-ICP-MS analysis. In contrast, the intermediate precisions of δ¹⁰⁹Ag_NIST978a in Au1, Au2, Au3, Au4 and Au5 were better than ±0.12‰ in each session, indicating that these gold samples have homogeneous Ag isotopic compositions and are suitable for use as quality control samples. The good precision for NWU-Au and DD440 (>±0.06‰) and accurate δ¹⁰⁹Ag_NIST978a determined with the matrix-matched calibration indicate that both DD440 and NWU-Au are fairly homogeneous and suitable as potential Ag isotopic reference materials. The accurate δ¹⁰⁹Ag_NIST978a values of gold samples with the Ag contents ranging from 1.98 to 8.49 wt% can be determined with matrix-matched calibration. The material will be provided free of charge for academic research and can be obtained by contacting the corresponding author at E-mail: pengliu@nwu.edu.cn.

Fig. 5 Compilation of δ¹⁰⁹Ag_NIST978a values of six gold samples determined by repeated measurements in different measurement sessions over 5 months. Range bars represent 2SE (the within-run precision). The mean values from SN-MC-ICP-MS are plotted as red diamonds. Circles represent external calibration using DD440 as a matrix-matched standard and squares represent external calibration using NWU-Au as a matrix-matched standard. Different colors represent results from different batches.

Fig. 6 Comparison of the δ¹⁰⁹Ag_NIST978a values of six gold samples obtained by fsLA-MC-ICP-MS and SN-MC-ICP-MS. The range bars represent 2s. (a) DD440 served as the matrix-matched bracketing reference material; (b) NWU-Au served as the matrix-matched bracketing reference material.

5. Conclusion

A synthetic gold material, named NWU-Au, was produced from ultrafine Au and Ag powder under conditions of 1090 °C for 3 min. Moreover, a natural gold reference material, DD440, was prepared for in situ Ag isotope measurement by LA-MC-ICP-MS. The silver isotope ratios of NWU-Au and DD440 determined by fsLA-MC-ICP-MS confirm its individual homogeneity within the analytical uncertainty of 0.06‰ for δ¹⁰⁹Ag_NIST978a (2s), suggesting that NWU-Au and DD440 can serve as bracketing reference materials for microbeam Ag isotope analysis. The δ¹⁰⁹Ag_NIST978a values of six gold samples determined with matrix-matched calibration were in agreement with those obtained by SN-MC-ICP-MS analysis, validating the accuracy of LA-MC-ICP-MS for Ag isotopic ratio measurements in gold samples. We recommend the δ¹⁰⁹Ag_NIST978a values of −0.23 ± 0.04‰ (2s, n = 8) and 0.42 ± 0.03‰ (2s, n = 5) determined by SN-MC-ICP-MS for NWU-Au and DD440, respectively.

Author contributions

The manuscript was written with contributions from all authors. All authors have given approval to the final version of the manuscript. Z. Bao and P. Liu designed the study. Z. Bao and D. Peng performed the experimental work. Z. Bao, P. Liu, Y. Tian, C. Zong and H. Yuan contributed to the experiments, data analysis and interpretation.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The authors declare that the data supporting the findings of this study are available within the paper and its supplementary information (SI) files. Should any raw data files be needed in another format, they are available from the corresponding author upon reasonable request. Source data are provided with this paper. Supplementary information: the trace elements contents for NWU-Au and DD440, as measured by LA-ICP-MS. See DOI: https://doi.org/10.1039/d5ja00375j.

Acknowledgements

This study was financially supported by Deep Earth Probe and Mineral Resources Exploration – National Science and Technology Major Project (2025ZD1006500 and 2025ZD1009200), the National Natural Science Foundation of China (Grants 42130102 and 42173033), and the MOST Research Foundation from the State Key Laboratory of Continental Evolution and Early Life, Department of Geology, Northwest University, China.

References

1. J. Tepsell, Y. Lahaye, F. Molnár, O. T. Rämö and N. Cook, Miner. Deposita, 2024, 59, 931–947.
2. R. M. Hough, C. R. M. Butt and J. r. Fischer-Bühner, Elements, 2009, 5, 297–302.
3. C. R. Voisey, R. Maas, A. G. Tomkins, M. Brauns and G. Brügmann, Econ. Geol., 2019, 114, 233–242.
4. W. Argapadmi, E. R. Toth, M. A. Fehr, M. Schönbächler and C. A. Heinrich, Econ. Geol., 2018, 113, 1553–1570.
5. G. Brügmann, M. Brauns and R. Maas, Chem. Geol., 2019, 516, 59–67.
6. A. V. Chugaev and I. V. Chernyshev, Geochem. Int., 2012, 50, 899–910.
7. Y. Fang, Q.-Y. Wen, Z.-C. Xiao, H.-M. Yu and F. Huang, J. Anal. At. Spectrom., 2024, 39, 1750–1758.
8. Y. Luo, V. Celo, E. Dabek-Zlotorzynska and L. Yang, J. Anal. At. Spectrom., 2012, 27, 299–304.
9. H. Gong, Y. Qi, T. Lan, H. Fan, Y. Chen, T. Zhou and R. Hu, J. Anal. At. Spectrom., 2024, 39, 1980–1989.
10. L. Yang, E. Dabek-Zlotorzynska and V. Celo, J. Anal. At. Spectrom., 2009, 24, 1564–1569.
11. M. Schönbächler, R. W. Carlson, M. F. Horan, T. D. Mock and E. H. Hauri, Int. J. Mass Spectrom., 2007, 261, 183–191.
12. J. Lin, Y. Liu, Z. Hu, W. Chen, C. Zhang, K. Zhao and X. Jin, J. Anal. At. Spectrom., 2019, 34, 1145–1153.
13. Y. Zhang, N. Lv, Y. Zhang, X. Nie and H. Yuan, At. Spectrosc., 2025, 46, 150–160.
14. X. Zeng, M. Li, W. Zhang, H. Jin, Y. Liu, Z. Hu, T. Luo, S. Yang, Z. Liu and J. Wang, J. Anal. At. Spectrom., 2025, 40, 888–900.
15. J. Lin, Y. Liu, A. Yang, W. Chen, L. Zhu and Z. Hu, J. Anal. At. Spectrom., 2022, 37, 592–602.
16. L. Xu, W. Zhang, T. Luo, J.-H. Yang and Z. Hu, J. Anal. At. Spectrom., 2021, 36, 747–757.
17. C. Huang, H. Wang, L. Xie, N. X. Nie, Y. Yang, X. Zhao, J. Li, H.-C. Tian, S. Wu, L. Xu and J. Yang, J. Anal. At. Spectrom., 2022, 37, 2165–2175.
18. D. Peng, Z. Bao, L. Kang, X. Nie, K. Chen, Y. Zhang and H. Yuan, Microchem. J., 2025, 212, 113238.
19. H. Liu, Z. Deng, Z. Hu, W. Zhang, M. Schiller, M. Bizzarro, Y. Liu, Z. Wang, L. Feng and M. Li, Geostand. Geoanal. Res., 2024, 48, 269–287.
20. Z. Bao, N. Lv, K. Chen, Y. Luan, X. Sun, C. Zong and H. Yuan, Geostand. Geoanal. Res., 2021, 45, 401–418.
21. N. Lv, Z. Bao, K. Chen, K. Wu and H. Yuan, Sci. China:Earth Sci., 2022, 65, 2005–2017.
22. Z. Bao, X. Nie, K. Chen, C. Zong, N. Lv and H. Yuan, Geostand. Geoanal. Res., 2022, 46, 433–449.
23. M. Weyrauch, M. Oeser, A. Brüske and S. Weyer, J. Anal. At. Spectrom., 2017, 32, 1312–1319.
24. J. Tian, Z. Bao, K. Chen, C. Zong, Y. Zhang, D. Peng and H. Yuan, J. Anal. At. Spectrom., 2024, 39, 2021–2029.
25. D. Peng, Z. Bao, P. Liu, Y. Tian, K. Chen and H. Yuan, Anal. Chim. Acta, 2025, 1377, 344657.
26. A. B. Kaufmann, M. Lazarov, S. Kiefer, J. Majzlan and S. Weyer, J. Anal. At. Spectrom., 2021, 36, 1554–1567.
27. W. Wang, C. Li, B. Zu, J. Zhang, P. Yu, H. Ren, X. Li, L. Zhou and W. Qu, At. Spectrosc., 2024, 45, 33–43.
28. D. Zhang, Z. Bao, P. Liu, G. Brügmann, W. Yang, K. Chen, P. Liang and H. Yuan, J. Anal. At. Spectrom., 2023, 38, 204–211.
29. P. Liu, J. Mao, B. Lehmann, S. Weyer, I. Horn, R. Mathur, F. Wang and Z. Zhou, Am. Mineral., 2021, 106, 1980–1986.
30. N. Lv, Z. Bao, K. Chen, X. Nie, Y. Zhang, D. Peng and H. Yuan, Anal. Chem., 2025, 97, 56–59.
31. D. Peng, Z. Bao, L. Kang, P. Liu, C. Zong, K. Chen and H. Yuan, At. Spectrosc., 2024, 45, 498–507.
32. Z. Bao, K. Chen, L. Kang, C. Zong, X. Nie, N. Lv, P. Liang and H. Yuan, J. Anal. At. Spectrom., 2023, 38, 2313–2323.
33. X.-W. Huang, S. Lin, K. Chen, J.-T. Kang, Z.-H. Li, Y.-M. Meng and L. Qi, J. Anal. At. Spectrom., 2024, 39, 1043–1049.
34. X. Nie, Z. Bao, C. Zong, N. Lv, K. Chen and H. Yuan, J. Anal. At. Spectrom., 2023, 38, 1065–1075.
35. J. Tian, Z. Bao, K. Chen, N. Lv, W. Yang, X. Nie and H. Yuan, J. Anal. At. Spectrom., 2024, 39, 1503–1513.
36. N. Lv, Z. Bao, X. Nie, K. Chen, Y. Zhang and H. Yuan, Geostand. Geoanal. Res., 2024, 48, 411–421.
37. X. Zeng, M. Li, H. Jin, Z. Cao, D. He, Z. Hu, Y. Liu and X. Chai, At. Spectrosc., 2024, 45, 532–540.
38. A. Yang, J. Lin, Y. Liu, Z. Liu, R. Lin, K. Deng and Z. Hu, Geostand. Geoanal. Res., 2023, 47, 535–546.
39. S. Wu, Y. Yang, K. P. Jochum, R. L. Romer, J. Glodny, I. P. Savov, S. Agostini, J. C. M. De Hoog, S. T. M. Peters, A. Kronz, C. Zhang, Z. Bao, X. Wang, Y. Li, G. Tang, L. Feng, H. Yu, Z. Li, L. Zhang, J. Lin, Y. Zeng, C. Xu, Y. Wang, Z. Cui, L. Deng, J. Xiao, Y. Liu, D. Xue, D. Zhang, L. Jia, H. Wang, L. Xu, C. Huang, L. Xie, A. Pack, G. Wörner, M. He, C. Li, H. Yuan, F. Huang, Q. Li, J. Yang, X. Li and F. Wu, Geostand. Geoanal. Res., 2021, 45, 719–745.
40. P. Liang, Z. Bao, W. Yang, C. Zong, K. Chen, Y. Zhang and H. Yuan, J. Anal. At. Spectrom., 2023, 38, 414–421.
41. Y. Liu, Z. Hu, S. Gao, D. Günther, J. Xu, C. Gao and H. Chen, Chem. Geol., 2008, 257, 34–43.

---