Abstract A reliable classification of meteorites is of great importance in order to study the meteorite history. Secondary ion mass spectrometry is applied to collect a database of spectra for four meteorite classes, H chondrite, IAB iron, L chondrite, and achondrite and based on convolutional neural network to determine the meteorite class of un unknown sample. The focus of the presented analysis is on the Kosice meteorite which is identified with following elements Li, B, Na, Mg, Al, Si, P, K, Ca, Ti, Cr, Mn, Fe, and Ni. Due to the high mass resolution of 7000, several isotopes of these elements are clearly identified. The Secondary ion mass spectrometry images provide also mineralogical analysis by revealing distributions of chemical composition. The minerals of albite, troilite, augite, and diopside are recognized. Using 88 reference spectra, the unknown sample is identified as the H chondrite based on calculated probability. Secondary ion mass spectrometry technique coupled with convolutional neural network proves to be the advanced tool to analyze meteorites, to classify them, and to determine a meteorite of unknown origin.

中文翻译：

---

二次离子质谱作为陨石分类的高级工具

摘要 对陨石进行可靠的分类对研究陨石历史具有重要意义。应用二次离子质谱采集H球粒陨石、IAB铁、L球粒陨石、无球粒陨石4类陨石谱图数据库，基于卷积神经网络确定未知样品的陨石类别。所呈现分析的重点是科希策陨石，该陨石被鉴定为具有以下元素 Li、B、Na、Mg、Al、Si、P、K、Ca、Ti、Cr、Mn、Fe 和 Ni。由于 7000 的高质量分辨率，这些元素的几种同位素被清楚地识别。二次离子质谱图像还通过揭示化学成分的分布提供矿物学分析。钠长石、陨硫铁、辉石和透辉石的矿物是公认的。使用 88 个参考光谱，根据计算出的概率，未知样品被识别为 H 球粒陨石。二次离子质谱技术与卷积神经网络相结合，被证明是分析陨石、对其进行分类和确定不明来源陨石的先进工具。