The technique in question is the Mobility Particle Size Spectrometer, often known by its commercial name, the scanning mobility particle sizer or SMPS. These instruments are the workhorses of ultrafine particle research, capable of counting and sizing particles down to a few nanometers across. They work by first giving airborne particles a well-defined distribution of electrical charges, typically by passing them through a bipolar diffusion charger, sometimes called a neutralizer. Inside this device, radioactive sources or corona discharges generate swarms of positive and negative ions that collide with the particles until a steady-state charge equilibrium is reached. The particles are then sorted by their electrical mobility, which depends on both their size and how much charge they carry. Crucially, the instrument’s data inversion assumes it knows exactly what fraction of particles of any given size carries zero, one, or more charges.

That assumption is where the trouble begins. The steady-state charge distribution depends on the properties of the ions inside the charger, including their mobility and mass, and those properties in turn depend on the composition of the carrier gas. For decades, most researchers have relied on a standard approximation published by Wiedensohler in 1988, which is even codified in the international standard ISO 15900, to predict these charge fractions. The approximation assumes fixed values for the mean mobilities of positive and negative charger ions. But the new measurements show that this convenient picture can be upset by trace gas contamination at concentrations so low they would normally escape attention entirely.

The research team, led by corresponding author Fabian Schmidt-Ott of the Institute for Atmospheric and Earth System Research at the University of Helsinki, together with Robert Nishida, Sebastian Schmitt, Jason Olfert, George Biskos, and Juha Kangasluoma, did something that had rarely been done before: they measured the ion properties inside the charger and the resulting particle charge fractions simultaneously, rather than treating one as an assumed input. Their particles, all smaller than 20 nanometers, were charged under carefully controlled conditions while the positive and negative ion populations were characterized in parallel. This closure between measurement and model allowed them to test whether classical charging theory, specifically the framework developed by Hoppel and Frick, could actually predict the charge fractions of particles in this challenging size range.

The answer, under steady-state conditions, was a resounding yes. When the actual ion properties were fed into the Hoppel and Frick theory, the predicted charge fractions agreed with the measurements remarkably well, with relative differences of up to only minus 6.1 percent for positive charge fractions and 0.6 percent for negative charge fractions. This is a meaningful validation, because sub-20 nanometer particles sit at the edge of where classical charging theory has been trusted, and they are precisely the particles most relevant to new particle formation in the atmosphere and to emerging ultrafine particle regulations. For these tiny particles, the fraction carrying two or more charges is negligible, which simplifies the analysis but makes the single-charge fraction all the more critical to get right.

The surprise came when the team introduced a seemingly innocuous change to their experimental setup: the use of conductive silicone tubing, a staple of aerosol laboratories prized for its electrical conductivity and chemical inertness. It turns out that this tubing is not as inert as assumed. It slowly outgasses volatile methyl siloxanes, or VMS, a class of silicon-based organic compounds ubiquitous in consumer products and increasingly scrutinized as atmospheric contaminants. These trace emissions changed the mean positive ion mobility by 20 percent, and the resulting fraction of singly charged particles deviated by up to 25 percent from measurements made without the tubing. In other words, a laboratory plumbing choice was quietly rewriting the charge statistics that the instrument’s data inversion depends on.

Yet the siloxane contamination also revealed something unexpectedly useful. Once the silicone tubing was in place, the mean positive ion mobility stabilized at a highly repeatable value of 1.05 plus or minus 0.1 square centimeters per volt-second, apparently pinned by the VMS chemistry. The negative ions, by contrast, remained highly sensitive to the surrounding gas composition, shifting with water vapor, ethanol, and other trace species. This asymmetry suggested a practical path forward: because positive ion properties can be readily stabilized and negative ion properties can be approximated from a simple bipolar mobility measurement, researchers may be able to characterize their own charger environment routinely and feed environment-specific ion properties into the charging theory, rather than trusting a one-size-fits-all approximation.

The stakes are higher than they might appear. Reviewer Christof Asbach, commenting on the preprint, noted that the European Union’s new air quality directive, published in late 2024, obliges all member states to measure ultrafine particle number concentrations and number size distributions. That regulatory push will multiply the number of mobility particle size spectrometers in operation and demand a much better understanding of their measurement uncertainties. The referee report highlighted that deviations between measured and conventionally assumed charge fractions reached as much as 65.1 percent for positive and minus 38.1 percent for negative charge fractions in some trace gas conditions, with only low concentrations of ethanol coming close to reproducing the standard Wiedensohler values. If charge fractions are wrong, inferred particle size distributions are wrong, and comparisons between instruments, laboratories, and countries become unreliable.

The study has not been without controversy. A second anonymous referee argued that the novelty of the work is limited, pointing to earlier studies since 2015 showing that ion populations depend on carrier gas composition and tubing material, and criticizing the use of a 1971 mass-mobility correlation and the neglect of more recent charging theories. The referee also noted that bipolar diffusion charging is only weakly sensitive to ion mobility, meaning reasonable mobility values yield similar charge distributions within roughly 10 to 15 percent. These critiques, part of the open peer discussion now underway, underscore that the scientific community is actively debating how best to characterize charger ions, whether through single mean mobilities or full mass-mobility distributions, and how much classical flux-matching theories can be trusted at the smallest particle sizes.

What remains clear, however, is the practical message for anyone measuring nanoparticles. The charge equilibrium that underpins mobility-based sizing is not a universal constant but a fingerprint of the local chemical environment, down to parts-per-trillion trace gases leaking from laboratory tubing. As ultrafine particle monitoring becomes a legal obligation across Europe and a growing priority worldwide, the study suggests that instrument operators should pay attention to the materials in their sampling lines, consider documenting their charger ion properties, and treat the standard charge distribution approximations as what they are: approximations whose accuracy depends on gases most of us never think to measure. In the delicate business of counting particles a thousand times thinner than a human hair, it turns out that the invisible chemistry of the lab itself can tip the scales.

Subject of Research: Measurement-model closure of sub-20 nm particle charge fractions and the effect of trace gas composition on bipolar diffusion charging

Article Title: Measurement-Model Closure of Sub-20 nm Particle Charge Fractions under Varying Trace Gas Composition

Article References: Schmidt-Ott, F., Nishida, R., Schmitt, S., Olfert, J., Biskos, G., & Kangasluoma, J. (2026). Measurement-Model Closure of Sub-20 nm Particle Charge Fractions under Varying Trace Gas Composition. https://doi.org/10.5194/ar-2026-29

Image Credits: AI Generated

DOI: 10.5194/ar-2026-29

Keywords: aerosol science, bipolar diffusion charging, ion mobility, volatile methyl siloxanes, ultrafine particles, mobility particle size spectrometry, Hoppel and Frick theory, Wiedensohler approximation, trace gases, silicone tubing, air quality monitoring, nanoparticle measurement