基体效应是影响LA-ICP-MS分析结果准确性的主要因素之一, 但目前却没有一种量化基体效应研究的方法。提出了一种以分析元素与内标元素的强度比(Ii/Iis)为纵坐标、浓度比(ci/cis)为横坐标绘制Ii/Iis-ci/cis图, 以元素对Ii/Iis-ci/cis图的线性相关系数ｒ量化基体效应的思路。以Fe为内标, 考察了13个常用玻璃标准物质与2个硫化物标准及多个硫化物矿物中6个元素对的基体效应, 结果显示Cu/Fe和Zn/Fe的线性相关系数r都小于0.99, 而痕量元素对Mn/Fe, Co/Fe, Ga/Fe, Pb/Fe的线性相关系数r都大于0.999;以S为内标, 考察了2个硫化物标准与多个硫化物矿物中三个主量元素对Fe/S, Cu/S和Zn/S的基体效应, 结果显示其线性相关系数r都小于0.999。无论是以Fe为内标结合玻璃标准为外标, 还是以S为内标结合硫化物标准为外标分析硫化物矿物, 主量元素大多数分析结果的误差大于10%;而以Fe为内标时, 绝大多数玻璃标准获得的痕量元素分析结果与MASS-1较为一致, 误差小于15%。研究表明, 玻璃标准及硫化物矿物标准均与硫化物矿物存在一定的基体效应差异, 而采用元素对Ii/Iis-ci/cis图的线性相关系数r量化基体效应具有一定的合理性与实用性。研究也表明了以Fe为内标, 采用非基体匹配的玻璃标准可用于定量分析硫化物矿物中的痕量元素, 尤其是具有较高痕量元素含量的NIST610。

Matrix effect between reference materials and samples is one of the major factors affecting the accuracy of analytical results by laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS). However, there is no method or calculation formula to quantify matrix effect between standards and samples up to date. In this paper, the linear correlation coefficient r of the Ii/Iis-ci/cis graphs of element pairs were used to characterize the matrix effect, which took the ratios of concentrations(ci/cis) and intensities(Ii/Iis) of the analytical element and internal standard element as x-axis and y-axis, respectively. Matrix effects of 6 element pairs in 13 glass reference materials, 2 sulfide reference materials and 2 sulfide minerals using Fe as internal standard was studied, with the linear correlation coefficient r of Fe—Cu, Fe—Zn element pairs both less than 0.999 and trace Fe—Mn, Fe—Co, Fe—Ga, Fe—Pb element pairs all better than 0.999. Matrix effects of 3 major element pairs in 2 sulfide reference materials and 6 sulfide minerals using S as internal standard was also studied, with the linear correlation coefficient r of S—Fe, S—Cu, S—Zn all less than 0.999. The great majority of relative errors of EMPA analytical results for major elements in sulfide minerals were greater than 10%, whether analyzed using Fe as internal standard with glass reference materials as external standard, or S as internal standard with sulfide reference materials MASS-1, IMER-1 as external standard, respectively. But the most analytical results for trace elements calibrated by glass reference materials using Fe as internal standard were well agreed with sulfide standard MASS-1, with the relative errors less than 15%. The results showed that matrix effects existed in glass reference materials, sulfide reference materials and sulfide minerals, and it also proved a certain rationality and practicability for quantification of matrix effect using the linear correlation coefficient r of the Ii/Iis-ci/cis graphs by this element pair method. This study also indicated that trace elements in sulfide minerals could be calibrated using Fe as internal standard with non-matrix matched glass reference materials as external standard, especially for NIST610 contained nearly all the trace elements in sulfide minerals and with relative high concentrations of each element, which obtained analytical results agreed well with sulfide standard MASS-1.