A Review of the IAG GeoPT™ and Certification Programmes,their Protocols,and the Differences between the Original and Potential Modifications to the Certification Protocol

The IAG conducts two programmes, the GeoPT ^? proficiency test and a certification programme that are closely interconnected. Both support the quality control/quality assurance activities of geochemical laboratories. Each derives an estimate of ‘true value’ for a number of samples, but arrives at that estimate, and its uncertainty, differently. This review discusses the history of the two programmes and compares the ‘true values’ and their uncertainties obtained through each. It then considers ‘fitness‐for‐purpose’ issues related to both GeoPT ^? and certification uncertainties. Issues related to potential modification of the IAG protocol for certification are also considered.     

An Evaluation of Methods for Assessing the Competence of Laboratories Based on Performance in the GeoPT Proficiency Testing Scheme

The original certification protocol, published by the International Association of Geoanalysts in 2003, specified that the competence of laboratories selected as competent to contribute certification measurements should be evaluated from their performance in the GeoPT proficiency testing programme. Round 39 of the IAG GeoPT proficiency testing programme provided an opportunity to examine four methods of evaluating laboratory competence based largely on the use of proficiency testing z‐scores as performance indicators. This opportunity arose because two test materials were co‐analysed by participating laboratories in this round: a syenite, SyMP‐1, supplied by the USGS, and an established CRM, the nepheline syenite, CGL 006. The performance of laboratories was assessed in four ways; in each case, consensus values and their uncertainties as derived from selective data sets of competent laboratories were compared with results derived from the routine GeoPT data assessment, involving all submitted measurements. An overall comparison of results showed no significant statistical differences in either consensus values or uncertainties between these data sets. This conclusion was unexpected and calls into question the widely held assumption that ‘better’ consensus data would be obtained from a subset of laboratories judged to be competent on the basis of proficiency testing performance indicators.     

The Reliability of Assigned Values from the GeoPT Proficiency Testing Programme from an Evaluation of Data for Six Test Materials that have been Characterised as Certified Reference Materials

Assigned values derived from the GeoPT proficiency testing programme were compared with certified values for six certified reference materials that have been used as test materials in the GeoPT programme. Statistical analysis showed that there were few significant differences between these sets of data and that these differences had no significant impact on the GeoPT assessment when fitness‐for‐purpose criteria were taken into account.     

The GeoPT Proficiency Testing Scheme for Laboratories Routinely Analysing Silicate Rocks: A Review of the Operating Protocol and Proposals for its Modification

The GeoPT proficiency testing programme for geochemical laboratories has been operating for nearly 20 years. This review, based on an analysis of accumulated data and extensive experience, examines whether the protocol followed during this period continues to be appropriate. Possible changes that might be considered for future implementation are examined, in particular the procedures for calculating assigned values, target values, z‐scores and the applicability of dual fitness‐for‐purpose criteria. We conclude that the existing protocol remains largely appropriate and effective, but propose that a single fitness‐for‐purpose criterion be adopted, with an option for laboratories to recalculate their z‐scores if the criterion is inappropriate for their requirements.     

The Long‐Term Robustness and Stability of Consensus Values as Composition Location Estimators for a Typical Geochemical Test Material in the GeoPT Proficiency Testing Programme

Re‐issuing the same test material, its true identity unknown to participants, in two rounds of the GeoPT proficiency testing programme 18 years apart has demonstrated remarkable similarity, and therefore stability, of consensus values independently estimated for over fifty measurands in each round. A comparison of the two data sets, GeoPT5 (AMH‐1) from 1999 and GeoPT41 (ORA‐1) from 2017, shows that corresponding consensus values, rigorously derived by current procedures for complete compatibility, are for the most part statistically indistinguishable when account is taken of their associated uncertainties. Inferences that may be drawn from this exceptional agreement include: (a) the consensus estimation procedure was robust and gave consistent results; (b) overall, the balance of the contributed data was unaffected by significant changes in: (i) the populations of laboratories participating, (ii) the proportions of laboratories producing results by different techniques, (iii) the manner in which systems were calibrated, or (iv) the aspirations of laboratories to attain higher quality results over time; and (c) unsurprisingly perhaps, that the test material itself had remained stable and unchanged during that time. Additionally, a statistical comparison of the original values for AMH‐1 with consensus values derived by current procedures demonstrates that they are also effectively indistinguishable.     

Determination of Te,As, Bi,Sb and Se (TABS) in Geological Reference Materials and GeoPT Proficiency Test Materials by Hydride Generation‐Atomic Fluorescence Spectrometry (HG‐AFS)

The study of Te, As, Bi, Sb and Se (TABS) has increased over the past years due to their use in the development of low‐carbon energy technologies. However, there is a scarcity of mass fraction values of TABS in geological reference materials. This underlines the difficulty in undertaking routine determinations of these elements. The mass fractions of TABS were determined in geological reference materials using hydride generation‐atomic fluorescence spectrometry (HG‐AFS), calibrated with standard solutions. Comparisons with literature values were used to validate the method. Samples from the GeoPT proficiency test were also analysed. For most elements, there are no assigned or even provisional values for many of the GeoPT and reference materials because of the wide range of results reported. For mass fractions above the quantification limit of the method, our results are in good agreement with the median of GeoPT results. Thus, we propose GeoPT median values as informational values for these elements. In contrast, at mass fractions < 0.5 µg g^?1 median values of Se from GeoPT are systematically higher than our results. Our Se results are in agreement with the reference materials down to 0.02 µg g^?1, which suggest that many of the results for Se reported in GeoPT testing are too high.     

Bias in the Determination of Zr,Y and Rare Earth Element Concentrations in Selected Silicate Rocks by ICP‐MS when Using Some Routine Acid Dissolution Procedures: Evidence from the GeoPT Proficiency Testing Programme

The analytical results for the modified river sediment, SdAR‐1 circulated in Round 31 of the GeoPT proficiency testing programme, revealed unexpected discrepancies for Zr, Y and most higher atomic number rare earth elements, in determinations made by ICP‐MS using acid attack digestion methods. This investigation compares these ICP‐MS results with those obtained for SdAR‐1 by three other methods: (a) ICP‐MS using fusion/sintering for sample digestion, (b) XRF analysis and (c) INAA. The distribution of results for the elements Rb, Sr, Ce, Y, Yb and Zr is examined and compared with those of the test material for Round 25 of GeoPT, Paraná basalt, HTB‐1. A substantial proportion (though not all) of Y, Yb and Zr determinations in SdAR‐1 by ICP‐MS/acid attack was variably low (sometimes very low) compared with other methods. A detailed evaluation of the procedures used to determine these results indicated that successful determinations by ICP‐MS/acid attack could be made if digestions were made at 180 °C for 48 hr using at least 2 ml HF per 100 mg of sample. We suggest that the more benign conditions (used by many laboratories) resulted in incomplete digestion of resistant minerals, such as zircon.     

An Evaluation of the Inter‐Method Discrepancies in Ferromanganese Nodule Proficiency Test GeoPT 23A

Round 23 of the GeoPT international proficiency testing scheme included the ferromanganese nodule powder FeMn‐1 which was distributed as an additional sample (23A). The aim of this initiative was to assess overall analytical performance for such a challenging oxide matrix with a view to the possible certification of such a material in accordance with ISO Guide requirements. To investigate inter‐method discrepancies, precision data and the method means for the most frequently used analytical methods (XRF, ICP‐MS and ICP‐AES) and sample preparation techniques were calculated and then compared using statistical tests of equivalence. For most major elements, XRF and ICP‐AES data dominated and these were found to give equivalent results. In contrast, for most trace elements significant discrepancies were detected between data obtained by different analytical methods. Possible causes are discussed with a view to attributing their origin to calibration strategy, sensitivity or interferences. It is assumed that the unusual oxide matrix generated unexpected interferences and thus method bias. Discrepancies observed between data from different analytical methods provide valuable information for the participating analysts, helping them to avoid systematic errors and thus minimising bias. They also suggest actions necessary to improve results for any future certification of such a material.     

The GeoPT Proficiency Testing Programme as a Scheme for the Certification of Geological Reference Materials

ISO Guide 35:2017 provides, for the first time, an alternative way of characterising certified reference materials using proficiency testing. In this paper, the properties of assigned values derived from the well‐established GeoPT proficiency testing scheme are examined. This scheme, designed for laboratories that undertake the routine analysis of silicate rocks and related materials, routinely has over 100 participants contributing results. Following a detailed assessment of the metrological properties of GeoPT assigned values in relation to Guide 35 recommendations, it is demonstrated that these values may be regarded as certified values, provided a number of criteria are met. These criteria include the following: a demonstration of sufficient homogeneity of the candidate CRM; circulation, when judged to be appropriate, of an established matrix‐matched CRM for co‐analysis in that round; the robust statistical analysis of data sets using GeoPT established criteria; a decision whether assigning a value is justified (including a requirement of a minimum of fifteen valid results); and an expert group to manage the certification and the maintenance of appropriate records. In summary, the GeoPT proficiency testing scheme, subject to the arrangements summarised above, is considered to be competent for the certification of geological reference materials.     

Preparation of Four Chromium Ore Reference Materials

The preparation and characterisation of four chromium ore reference materials are described in this paper. The sample material for GCr‐1, GCr‐3 and GCr‐4 was collected from chromite deposits in Tibet, Qinghai province and Inner Mongolia. GCr‐2 is a composite sample from GCr‐1 and GCr‐4. Sample homogeneity was tested by WD‐XRF and the relative standard deviations were < 1.0%. An F‐test showed that all four materials were homogeneous. Thirteen laboratories involved in the inter‐laboratory programme provided 672 determinations (eighteen oxides and elements). Sixteen components were characterised as certified values, of which Cr₂O₃ ranged from 17.59 to 57.80% m/m. The contents of FeO and CO₂ were taken as reference values.     

The I.A.G.C. interlaboratory water analysis comparison programme

The I.A.G.C. international programme of interlaboratory comparison of water analysis quality involved seven waters, including surface waters, a ground water, an alkaline spring water, a sample ‘spiked’ with heavy metals, and a geothermal water. Forty-eight laboratories in eighteen countries participated.Sample stability was little problem for most major solutes and for minor elements at concentrations greater than tens of ppb. Iron analyses were erratic for some samples, and the stability of nitrogen species was suspect in unsterilized samples. For major constituents in surface waters the results were of reasonable quality but for waters of unusual composition e.g. the geothermal and alkaline spring water, the results were far from satisfactory. Of the common constituents, SO₄, NO₃, F, and SiO₂ showed the greatest range of results. The most consistent results for trace elements at the 10 ppb level were for Be, Cd, Cu, Hg, and Li. Lower quality results (coefficient of variation over 30) were obtained for Fe, As, Mn, Co, Ni, Pb and Zn.     

X射线荧光光谱法分析不同类型岩石中10种主量元素的测试能力验证

以28个不同类型岩石标样作外标,用玻璃熔片法制样,采用经验系数法进行基体校正,建立了X射线荧光光谱法(XRF)测定不同类型岩石中硅铝钙镁等10种主量元素的分析方法。该方法被用于参与GeoPT分析测试能力验证,15年来分析了23个不同类型的岩石验证样品。经验证该方法对SiO_2、Al_2O_3、K_2O和CaO具有非常好的分析效果(所有测定结果的|Z|2);对于MgO、MnO、TiO_2和Na_2O四个元素,尽管各有一个样品的分析结果的|Z|2,这主要是样品与标样的基体差异较大造成了低含量样品测定结果超差,但依然可获得好的分析效果;当TFe_2O_31.0%,P_2O_5含量在0.08%~0.80%区间时,分析结果的|Z|2,满足分析期望。总体上所建立的XRF分析方法稳定,可满足应用地球化学实验室测试不同类型岩石中10种主量元素的分析期望。     

Geochemical and statistical characterization of the phosphate facies of the Farim‐Saliquinhé phosphate mineralization (Guinea‐Bissau)

The Eocene Farim‐Saliquinhé phosphate mineralization (FSPM) is a buried sedimentary deposit located in the northern part of Guinea‐Bissau, discovered and spatially constrained during the 1980s by the French Geological Survey (BRGM). In the present work some of the data collected at that time are reworked in the framework of the development of a 3D geological and geochemical model of this mineralization. We show the usefulness of two classical multivariate statistical methods – principal component analysis (PCA) and cluster analysis (CA) – applied to eight geochemical variables (P₂O₅, SiO₂, Al₂O₃, Fe₂O₃, CaO, MgO, F and CO₂) measured in 247 samples from phosphate facies, for differentiation of samples taken from the different phosphate facies recognized in the FSPM, namely the main ore FPA (30 to 35% high grade phosphate), the calcareous low grade phosphate FPB (10 to 20% P₂O₅) (both Lutetian), and a mineralized Upper Eocene to Lower Oligocene cover (mainly including the FPO level and a silico‐aluminous phosphate bed), overlying the FPA, which can also be considered a third set of phosphate facies associated with the FSPM.     

矽线石成分分析标准物质研制

随着矽线石应用领域的逐步拓展,英国、南非和日本等国家已研制了4种矽线石标准物质,而我国仅有一种矽线石国家二级标准物质,无论从组分的浓度梯度范围还是定值指标等方面,均难以满足我国研究需求。本文针对我国矽线石的分布情况,在黑龙江林口县和河南内乡县采集典型矽线石原矿2种,在黑龙江林口县采集矽线石精矿1种,按照国家一级标准物质研制标准和规范要求,研制了3种矽线石成分分析国家一级标准物质(批准编号为GBW07843、GBW07844、GBW07845)。均匀性检验结果表明,除个别指标(Y-1的TFe_2O_3、Cu和J-1的Mn O等)外,3种标准物质检测指标的F值均小于临界值F0.05(24,25)=1.96,组内和组间无明显差异;Y-1的TFe_2O_3、Cu和J-1的MnO等指标的组内和组间差异主要来源于分析方法误差,由此表明此批标准物质均匀性良好。在14个月考察期内,3种标准物质计算得到的拟合直线斜率b1均不显著,表明3种标准物质有较好的稳定性。经我国10家实验室使用多种分析方法对矿石中的主量元素、痕量元素和矽线石含量(硅铝,SAl_2O_3)等共计39种组分联合定值,各组分的相对扩展不确定度处于0.60%~29.9%区间,3种矽线石标准物质主量成分Al_2O_3的含量分别为25.85%、28.16%和55.06%。该系列矽线石标准物质可满足地质、环境等研究领域相关样品分析质量监控工作的需求。     

氢化物发生-电感耦合等离子体发射光谱法测定铀矿地质样品中痕量硒

采用氢化物发生技术测定地质样品中的硒时,需要考虑样品的溶解、Se价态的预还原以及抑制共存离子的干扰。本文采用硝酸-盐酸-氢氟酸-高氯酸体系快速溶解样品,直接加入浓盐酸煮沸将六价硒还原为四价硒,将氢化物发生器与电感耦合等离子体发射光谱仪联用测定了铀矿地质样品中的痕量硒。样品中除了Cu2+其他离子的含量均不干扰硒的测定,通过在试液中加入铁盐溶液或在硼氢化钠还原剂中加入铁氰化钾抑制了Cu2+的干扰。方法检出限为0.12μg/L,精密度(RSD)小于5%。与前人报道的方法相比,本方法检出限较低,操作简单快速,冲洗30 s可消除记忆效应,适合批量铀矿地质样品中痕量硒的测定。     

Determination of Rare Earth Elements and Thorium at Nanogram Levels in Ultramafic Samples by Inductively Coupled Plasma‐Mass Spectrometry Combined with Chemical Separation and Pre‐concentration

A method was developed for the determination of low‐level rare earth elements (REEs) and thorium in ultramafic samples by inductively coupled plasma‐mass spectrometry. The conventional method for the digestion of ultramafic rocks using HNO₃ and HF results in considerable amounts of insoluble fluorides because of the high contents of Mg (generally up to 24% m/m) in these rocks. In this study, we used H₃BO₃ as a complexing agent to break down the insoluble fluorides, and then separated the REEs from Fe and Mg major and Ba, Ca, Cr minor matrices by anion exchange and co‐precipitation, respectively. The whole procedural blanks estimated from sample‐free analyses ranged from 0.232 ng for Ce to 0.006 ng for Tm and Lu. Limits of detection for this method, defined as three times the standard deviation of these blank analyses, varied from 0.51 ng g^?1 for Ce to 0.03 ng g^?1 for Lu. The recovery of REEs using this technique, as determined using the standard addition method, ranged from 92.9% for Y to 102.0% for Er with 3% (RSD) variation. The method was validated using GAS (GeoPT‐12), JP‐1 and PCC‐1, and the results were comparable to literature values, elucidating the applicability to the determination of ultra trace REEs in ultramafic rocks.     

New Chinese National Reference Materials for Hydrogen and Oxygen Isotopes in Water

The Institute of Hydrogeology and Environmental Geology, Chinese Academy of Geological Sciences recently prepared four certified reference materials for hydrogen and oxygen stable isotopes in water, which are called ‘China Standard Water' (CSW)‐HO1–HO4 (hereafter referred to as HO1–HO4). These reference materials are intended for calibration purposes and provide reference values of their relative difference in ²H/¹H and ¹⁸O/¹⁶O isotope‐amount ratios expressed in delta notation, normalised to the VSMOW–SLAP scale. The certified values of the reference materials were determined by an interlaboratory comparison of results from eleven participating laboratories. This paper describes in detail the production and certification procedure of the four reference materials. The first analytical data for the reference materials are also provided using a variety of analytical techniques, namely CO₂–H₂O equilibration and laser spectroscopy for δ¹⁸O and Cr reduction, as well as H₂–H₂O equilibration, laser spectroscopy, and high‐temperature conversion for δ²H. The reference values for materials HO1–HO4 and their associated uncertainties are assigned.     

Elemental Concentrations of Stream Sediments in Southern China

Over 3 million samples were collected from southern China at a density of 1–2 stream sediment samples per square kilometre as part of the Regional Geochemistry National Reconnaissance (RGNR) programme of China initiated in the late 1970s. Approximately 5244 composite samples of stream sediments from twelve provinces of southern China were prepared from the original RGNR samples collected from a territory with an area of 2300000 km² at a density of one composite sample per 1:50000 map sheet (about 400 km²). Seventy‐six elements were determined by the geoanalysis laboratory of Henan Province (platinum‐group elements) and by the Institute of Geophysical and Geochemical Exploration (IGGE) laboratory for the remainder. Internal quality and external quality control methods were applied to ensure that the analytical data were comparable. Statistics were used to derive the mean and background values and indicate the average concentrations of the seventy‐six elements. Comparisons were made against the mean values obtained from stream sediments, floodplain sediments, overall sediments and overall soils of the entire territory of China. The concentrations of Hg, Cd, rare earth elements and 24 other elements were higher than their background for the whole of China. In contrast, the concentration of Na₂O, CaO, Ba and Sr were lower in the stream sediments in southern China than their whole China background.     

High‐Precision Iron Isotope Analysis of Geological Reference Materials by High‐Resolution MC‐ICP‐MS

We report high‐precision iron isotopic data for twenty‐two commercially available geological reference materials, including silicates, carbonatite, shale, carbonate and clay. Accuracy was checked by analyses of synthetic solutions with known Fe isotopic compositions but different matrices ranging from felsic to ultramafic igneous rocks, high Ca and low Fe limestone, to samples enriched in transition group elements (e.g., Cu, Co and Ni). Analyses over a 2‐year period of these synthetic samples and pure Fe solutions that were processed through the whole chemistry procedure yielded an average δ⁵⁶Fe value of ?0.001 ± 0.025‰ (2s, n = 74), identical to the expected true value of 0. This demonstrates a long‐term reproducibility and accuracy of < 0.03‰ for determination of ⁵⁶Fe/⁵⁴Fe ratios. Reproducibility and accuracy were further confirmed by replicate measurements of the twenty‐two RMs, which yielded results that perfectly match the mean values of published data within quoted uncertainties. New recommended values and associated uncertainties are presented for interlaboratory calibration in the future.     

Simultaneous Determination of Major and Trace Elements in Fused Volcanic Rock Powders Using a Hermetic Vessel Heater and LA‐ICP‐MS

Fused glass prepared without the addition of a flux is generally more homogeneous than a pressed powder pellet and thus ideal for analysis of bulk samples by LA‐ICP‐MS. In this work, a new glass‐making method using a boron nitride crucible was developed to prepare homogenous glass samples from silicate rock powder. The apparatus consisted of a small boron nitride vessel with net volume of about 34 mm³ and two molybdenum strips. Applying the summed metal oxide normalisation technique, both major and trace element contents in the fused glass were measured by LA‐ICP‐MS. Analyses of five geochemical reference materials (spanning the compositional range basalt–andesite–rhyolite) indicated that the measured SiO₂, Al₂O₃ and P₂O₅ contents matched the preferred values to within 5%, and the other major elements generally matched the preferred values to within 8%. Except for the transition metals, the measured trace element contents generally matched the preferred values to within 10%. Compared with the iridium heater method developed by Stoll et al. (2008), element volatilisation during high‐temperature melting was effectively suppressed in our method, but metal segregation caused by reduction of BN may cause loss of Cr, Ni and Cu. Although analysis with a large spot size has the advantage of improving counting statistics, matrix effects induced by mass loading of the ICP may hamper the accurate determination of some elements.