Spark discharge aerosol generator for field calibration of absorption photometers: Aerosol properties and stability

Atmospheric aerosol particles can significantly impact the atmospheric radiative balance by scattering and absorbing incoming solar radiation. Additionally, they are known to strongly affect human health. Given the strong variations in geographical distribution of atmospheric aerosols there is a high need for ubiquitous measurements, while the variable chemical composition generates several technical and metrological challenges. Absorption photometers are commonly used to measure the atmospheric mass concentration of light‑absorbing particles like equivalent black carbon (BC_e), which is the most important radiative forcer among aerosol particles due to its strong infrared to visible spectral absorption. Although BC measurements have been done since decades, a reliable metrological aerosol absorption standard to ensure traceable calibration has not been established so far. Due to the wide field implementation there is strong need for a portable, metrological generator of BC-like aerosol particles for in‑field calibration of aerosol absorption photometers. Spark discharge volatilization is an interesting candidate for a BC particle generator, given its robust operation principle and the reduced media requirements.

The spark discharge aerosol generator (SDAG) produces graphitic, BC-like aerosol particles. Most important is the lack of any organic coatings, known from spray or combustion generators, which usually alter the optical properties of the BC particles. The SDAG consists of a chamber purged with an inert gas (usually nitrogen or argon), which houses two graphite electrodes, which are connected to a pulsed high-voltage source with variable pulse frequency and amplitude. In this study, a PALAS DNP 3000 (PALAS GmbH, Germany) has been used for generating graphitic particles and measure their optical properties, aerosol number size distribution and particle morphology by scanning electron microscopy in transmission mode (TSEM). The SDAG was operated by using inert N₂ only (avoiding dilution air), in order to facilitate the transportability and in-field operation. The N₂ flow rate was fixed to 10 l/min. The spark discharge frequency spanned 60 to 600 Hz. The voltage was varied from 2500 to 5000 V.

The mobility count mean diameter (CMD) of the particles produced could be varied from 28 to 80 nm, using the different set points described above. The single scattering albedo of the aerosol particles was almost constant for all operation modes with an average 0.11 ± 0.03. A repeatability analysis over 9 days was done using a single setting mode (140 Hz, 3500 V), which produced particles of 45 nm CMD with a variability of 6 nm CMD (2σ). The total particle number concentration ranged from 8 to 11 x 10⁶ cm^-3 and varied within 8% (1σ) within the different days. Hence, the SDAG is a promising source of stable, nascent and uncoated, graphitic BC particles and thus has good potential to improve field BC calibration.

This research is part of an international project that aims to establish a BC reference material and calibration procedure (EMPIR 16ENV02 ”Black Carbon”, http://www.empirblackcarbon.com/).