Simulator-Designed Aerosol Sensor Tracks Particle Charges Every 20 Seconds

Atmospheric particles smaller than a micrometre are rarely electrically neutral. As they drift through the air, they collide with positive and negative ions, picking up charges that determine how long they survive in the atmosphere, how readily they are scavenged by clouds and rain, and how deeply they penetrate the human respiratory tract. A team of Japanese researchers has now rebuilt an instrument that measures these charges far faster and more precisely than before, using computer simulations to eliminate a design flaw that had quietly limited the accuracy of earlier measurements.

The device, called the Keio Measurement System of Aerosol Charging State, or K-MACS, works with optical particle counters to sort submicron particles measuring 0.3 to 0.5 micrometres into three streams: positively charged, negatively charged and uncharged. In the version reported in the Asian Journal of Atmospheric Environment, the system delivers a new measurement every 20 seconds, a dramatic improvement over the 15-minute averaging interval of the previous instrument. That speed matters because particle charge distributions can shift rapidly during atmospheric electrical events such as lightning and strong winds, and a 15-minute average simply blurs those changes together.

The improvement began with a diagnosis. Earlier evaluations of the K-MACS/OPC system had shown that measurements of the average particle charge number and the spread of the charge distribution carried uncertainties of roughly 15 percent, but nobody fully understood why. To find out, the researchers turned to computational fluid dynamics coupled with a discrete phase model, simulating the motion of air and particles inside the old apparatus in fine detail. The simulation revealed a subtle but consequential problem: as the sample air turned a 90-degree bend between the conditioning chamber and the classification section, particle inertia caused them to drift off the fluid streamlines, biasing the particle population toward the negatively charged electrode plate.

That bias mattered because the earlier analysis had assumed particles entered the classification section uniformly distributed across its width. When the researchers re-ran the numbers using the simulated, non-uniform particle distribution as a weighting function, the agreement between experimental and theoretical charge distributions improved markedly, with uncertainties falling to roughly 6 percent. The exercise confirmed that particle dispersion before the classification section was a key source of measurement error, and it pointed the way to a better design.

The new K-MACS incorporates three major modifications. A single sample inlet was replaced by four inlets, which lowered the incoming air velocity and distributed the flow more evenly, suppressing local flow irregularities near the entrance. The chamber size and internal geometry were redesigned to avoid abrupt curvature of the airflow, producing a more compact device with a straighter flow path. Finally, the sheath airflow rate, the particle-free air that sandwiches the sample between two parallel electrode plates, was optimised to allow the flow to fully develop before it reaches the region where charged particles are separated.

The simulations showed why the sheath flow rate is so critical. At a low sheath flow of 0.5 litres per minute, two recirculation zones formed inside the classification section, deflecting the sample flow toward the lower wall and scattering particles across all three outlets. As the sheath flow increased, the sample flow became progressively confined and focused. At 2.8 litres per minute, roughly 89 percent of particles exited through the middle outlet in experiments, with side-outlet transmission dropping below 3 percent. At 3.6 litres per minute, the flow became highly stable and symmetric, side-outlet transmission approached zero and more than 90 percent of particles passed through the middle outlet. Experimental measurements of particle transmission efficiency at all four tested flow rates agreed with the simulations within a few percent, validating the numerical model.

With the geometry settled, the team built the new device from aluminium, making it more compact and better suited to long-term operation than the previous acrylic version. Experiments confirmed the basic principle of the instrument: when no voltage was applied, almost all particles travelled to the middle outlet, but upon applying 300 volts, positively and negatively charged particles were clearly pulled toward the oppositely charged electrode plates, with concentrations at the electrode-side outlets rising sharply within seconds. The researchers also found that roughly 46 to 49 percent of artificially multiply charged particles deposited on the electrode plates under those conditions, underlining the importance of matching the applied voltage to the charge distribution being measured.

To find the optimal operating voltage for real atmospheric particles, the team generated polystyrene latex spheres of 0.37 micrometres, passed them through a neutraliser to establish a stationary charge distribution, and applied voltages ranging from 3,000 to 6,000 volts. As the voltage rose from 3,000 to 4,700 volts, the uncharged particle fraction increased and the width of the charge distribution narrowed, indicating sharper classification. Above 5,000 volts the distribution stabilised. Comparing experiment with theory, the researchers identified approximately 5,400 volts as the optimum, where the uncharged particle fractions agreed within 5.0 percent and the distribution widths agreed within 1.0 percent, a substantial reduction from the roughly 15 percent uncertainty of the previous system.

Not everything matched theory perfectly. The experimentally measured average particle charge number was about five times smaller in magnitude than the theoretical prediction, a discrepancy the researchers attribute to the relatively long residence time of particles in the tubing and within the instrument, which allows the charge distribution to shift slightly positive as positive ions accumulate downstream of the neutraliser. Small excesses in the fractions of highly charged particles also remained, possibly reflecting particle dispersion and minor flow fluctuations inside the device. The team notes that the ion counter used in their field test detects ions only above a fixed mobility threshold, so further work with full ion mobility distributions is needed for a complete evaluation.

As a demonstration, the researchers deployed the system in indoor air at Keio University’s Yagami Campus in Yokohama from 4 to 5 September 2025, measuring the charge distribution of 0.3 to 0.5 micrometre particles every 20 seconds alongside ion concentrations. Ion concentrations and charged particle numbers varied substantially over the observation period, and the ratio of positively to negatively charged particles, as well as the average particle charge number, correlated strongly with the ratio of positive to negative ion concentrations, with coefficients of determination of 0.95 and 0.93 respectively. The pattern indicates that ambient ions were depleted through collisions with aerosol particles, forming charged particles in the process, and that the electrical balance between positive and negative ions shapes the charging state of submicron particles. Because humidity-related parameters stayed relatively constant during the observation, the researchers conclude that moisture variations did not substantially drive the observed changes. The high temporal resolution of the improved system, the team argues, makes it a powerful tool for studying rapid atmospheric electrical processes and their influence on the deposition of charged particles in the human respiratory system.

Subject of Research: An improved simulation-based instrument for measuring the electrical charge distribution of submicron atmospheric particles at high temporal resolution

Article Title: Numerical-simulation-based improved measurement of submicron-sized particle charge distribution with high temporal resolution

Article References: Mori, T., Nor, M. A. B. M., Iwata, A., Watanabe, K., Andrade, A. X., Lin, T., Hu, S.-C., & Okuda, T. (2026). Numerical-simulation-based improved measurement of submicron-sized particle charge distribution with high temporal resolution. Asian Journal of Atmospheric Environment, 20(1), Article 18. https://doi.org/10.1007/s44273-026-00098-z

Image Credits: AI Generated

DOI: 10.1007/s44273-026-00098-z

Keywords: aerosol particles, particle charge distribution, atmospheric electricity, numerical simulation, computational fluid dynamics, optical particle counter, ion–aerosol interactions, submicron particles, K-MACS, indoor air measurement, lightning, respiratory deposition