Measurement Bias in spICP-TOFMS: Insights from Monte Carlo Simulations
Raven L. Buckman and Alexander Gundlach-Graham
Analytical Methods, 2024
DOI: 10.1039/d4ay00859f
This recent publication in Analytical Methods, led by researchers from Iowa State University, utilized the icpTOF S2 to confirm Monte Carlo simulations employed to reduce biases and inaccuracies due to measurement noise and threshold effects in nanoparticle analysis using single particle mass spectrometry. The simulations allow for better interpretation of particle data by comparing simulated results with experimental measurements, improving accuracy in size, composition, and concentration determinations.
Single-particle inductively coupled plasma mass spectrometry (sp-ICP-MS), particularly with time-of-flight mass spectrometers (sp-ICP-TOFMS), is a prominent technique for analyzing the size, number concentration, and composition of inorganic nanoparticles. However, the technique can introduce measurement biases due to various noise sources, including Poisson noise from ion detection. These biases can distort particle size distributions, element mass ratios, and particle number concentrations (PNCs), especially for low-abundance elements.
In this study, Monte Carlo simulations were used to investigate these biases by modeling sp-ICP-TOFMS signals under different conditions, such as particle size distribution, multi-element composition, and sensitivity variations. They demonstrated the simulations’ accuracy by comparing them with experimental data from various nanoparticles, showing that they can effectively predict and understand the biases affecting sp-ICP-TOFMS measurements. The effectiveness of Monte Carlo simulations was shown by modeling sp-ICP-TOFMS signals, accounting for factors like particle size distribution, multi-element composition, sensitivities, and critical values. The simulations accurately reflect sp-ICP-TOFMS analysis, as confirmed by their match with experimental data on CeO2, ferrocerium, and bastnaesite particles using the TOFWERK icpTOF S2. A case study showed that when particles are near or below critical levels, measured distributions differ significantly from true distributions, leading to overestimated metrics like median particle diameters. Monte Carlo simulations allowed for rapid, detailed exploration of particle distributions and instrument conditions, providing insights into how noise and thresholding affect measurement outcomes.
These simulations offer a valuable tool for understanding detected particle signals, potentially aiding in isotopic analyses, machine learning, and measurement validation in sp-ICP-TOFMS research. These Monte Carlo simulations enable the comparison of known and experimental data, helping to understand and correct biases, advancing nanoparticle analysis using sp-ICP-TOFMS with our icpTOF S2.
