Collaborative Research: Catalytic: Improving Accuracy and Efficiency of Multicollector Mass Spectrometry

在地球物质属性中，同位素的相对丰度——具有相似化学性质但质量不同的元素——为理解地球的历史、结构和演化提供了关键线索。质谱仪是一种灵敏的仪器，它通过分离原子的质量来测量不同同位素的相对丰度。热电离质谱仪（TIMS）和多收集器电感耦合等离子体质谱仪（MC-ICPMS）可以非常精确地测量各种元素的相对丰度。该合同将支持质谱仪制造商和同位素地球化学家开发软件，帮助用户可视化、解释和统计评估MC-ICPMS和TIMS的质谱仪数据。由此产生的数据集将更加准确，开发和实施该软件为学习基本的开放科学和质谱原理提供了机会。将通过社区讲习班和其他外联活动与用户社区进行广泛接触，这些活动强调来自历史上在地球科学领域代表性不足的群体的早期职业研究人员和科学家的参与。该奖项支持扩展同位素地球化学家使用的统计工具箱，用于解释高精度质谱仪数据，然后可以输出到任何同位素系统特定的数据还原协议。这包括整合现有的统计技术，如对数比数据分析，同位素地球化学的新算法，如处理拒绝数据的期望最大化（E-M）算法，以及贝叶斯推理的方法，如可逆跳跃马尔可夫链和自适应蒙特卡罗方法。总之，这些可以为提高质谱仪的精度和准确度提供下一步，而无需昂贵的仪器开发。将这些算法封装到新的健壮的开源软件产品中是至关重要的，这些软件产品具有支持新的和现有的科学工作流程的动态用户界面。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Among Earth material properties, the relative abundance of isotopes – elements with similar chemical properties but different masses – provides critical clues for understanding the history, structure, and evolution of the Earth. Mass spectrometers are sensitive instruments that measure the relative abundances of different isotopes by separating atoms based on their mass. Thermal ionization mass spectrometers (TIMS) and multi-collector inductively-coupled plasma mass spectrometers (MC-ICPMS) can measure very precise relative abundances for a wide variety of elements. This award supports the creation of software, developed with input from mass spectrometer manufacturers and isotope geochemists, that will help users visualize, interpret, and statistically evaluate mass spectrometer data from MC-ICPMS and TIMS. Resulting datasets will be more accurate, and developing and implementing this software provides opportunities for learning fundamental open science and mass spectrometry principles. Broad engagement with user communities will be pursued through community workshops and other outreach activities that emphasize participation of early career researchers and scientists from groups historically underrepresented in the geosciences. This award supports the expansion of a statistical toolbox used by isotope geochemists to interpret high precision mass spectrometer data, which can then be output to any isotope-system-specific data reduction protocol. This includes integration of existing statistical techniques, like log-ratio data analysis, new algorithms for isotope geochemistry, like the expectation-maximization (E-M) algorithm for handling rejected data, and methods for Bayesian inference, like reversible-jump Markov Chain and adaptive Monte Carlo methods. Together, these can provide the next step in improving mass spectrometer precision and accuracy without costly instrument development. It is critical to encapsulate these algorithms into new robust open-source software products with dynamic user interfaces that support new and existing scientific workflows.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.