The influence of particle composition upon the evolution of urban ultrafine diesel particles on the neighbourhood scale

Abstract

Abstract. A recent study demonstrated that diesel particles in urban air undergo evaporative shrinkage when advected to a cleaner atmosphere (Harrison et al., 2016). We explore, in a structured and systematic way, the sensitivity of nucleation-mode diesel particles (diameter < 30 nm) to changes in particle composition, saturation vapour pressure, and the mass accommodation coefficient. We use a multicomponent aerosol microphysics model based on surrogate molecule (C16−C32 n-alkane) volatilities. For standard atmospheric conditions (298 K, 1013.25 hPa), and over timescales (ca. 100 s) relevant for dispersion on the neighbourhood scale (up to 1 km), the choice of a particular vapour pressure dataset changes the range of compounds that are appreciably volatile by two to six carbon numbers. The nucleation-mode peak diameter, after 100 s of model runtime, is sensitive to the vapour pressure parameterisations for particles with compositions centred on surrogate molecules between C22H46 and C24H50. The vapour pressure range, derived from published data, is between 9.23 × 10−3 and 8.94 × 10−6 Pa for C22H46 and between 2.26 × 10−3 and 2.46 × 10−7 Pa for C24H50. Therefore, the vapour pressures of components in this range are critical for the modelling of nucleation-mode aerosol dynamics on the neighbourhood scale and need to be better constrained. Laboratory studies have shown this carbon number fraction to derive predominantly from engine lubricating oil. The accuracy of vapour pressure data for other (more and less volatile) components from laboratory experiments is less critical. The influence of a core of non-volatile material is also considered; non-volatile core fractions of more than 5 % are inconsistent with the field measurements that we test the model against. We consider mass accommodation coefficient values less than unity and find that model runs with more volatile vapour pressure parameterisations and lower accommodation coefficients are similar to runs with less volatile vapour pressure parameterisations and higher accommodation coefficients. The new findings of this study may also be used to identify semi-volatile organic compound (SVOC) compositions that play dominating roles in the evaporative shrinkage of the nucleation mode observed in field measurements (Dall'Osto et al., 2011).

Experimental vapour pressures of eight n-alkanes (C17, C18, C20, C22, C24, C26, C28 and C31) measured at ambient temperatures

“…The disagreement between estimated vapour pressures of the same compound at high molecular weight (over 9 orders of magnitude) illustrates the need to directly measure vapour pressures at the conditions of interest, and show the difficulties in accurately extrapolating and predicting data at ambient temperatures. This is further discussed inNikolova et al (2018).…”

“…For atmospheric aerosol modelling and interpretation of environmental measurements we recommend direct use of our data, interpolated in carbon number where necessary, and extrapolated over temperatures relevant to the atmospheric boundary layer using equation 2, above. Of the existing parameterisations, we have used Compernolle et al (2011) as base case in a recent modelling study (Nikolova et al, 2018) because it follows our direct measurements reasonably well.…”

“…Accumulating evidence indicates that UFP are toxic and have potentially harmful effects on human health (Atkinson et al, 2010). In order to understand and model the atmospheric behaviour of SVOC, it is necessary to specify their chemical composition, their phase partitioning and the size distribution of the particulate fraction, and the vapour pressures of the constituent molecules (Harrison et al, 2019;Nikolova et al, 2018). Diesel exhaust vapour and particulate phases consist of SVOC in the range C 12 -C 34 (Alam et al, 2016).…”

“…Very few studies have measured vapour pressures of n-alkanes at 298 K (Chickos and Hanshaw, 2004a;Chickos and Hanshaw, 2004b) and more recently extended measurements (and estimations) for n-alkanes up to n-C 92 (Chickos et al, 2009). Although variability amongst reported vapour pressures is relatively small for the more volatile compounds, literature values for the lower volatility compounds can vary by more than five orders of magnitude for a single compound (see Figure 3 of Nikolova et al, 2018 and cf. Figure 1, below).…”

“…Figure 1, below). With such a large range of extrapolated vapour pressures for a given compound, the behaviour of atmospheric models will depend greatly on which set of vapour pressures is adopted (Nikolova et al, 2018).…”