Chemically speciated mass size distribution , particle effective density and origin of 1 non-refractory PM 1 measured at a rural background site in Central 2

18 The seasonal variability of non-refractory PM1 (NR-PM1) was studied at a rural background 19 site (National Atmospheric Observatory Košetice – NAOK) in the Czech Republic to examine 20 the impact of atmospheric regional and long-range transport in Central Europe. NR-PM1 21 measurements were performed by compact time-of-flight aerosol mass spectrometry (C-ToF22 AMS), and the chemically speciated mass size distributions, effective density, and origin were 23 discussed. The average PM1 concentrations, calculated as the sum of the NR-PM1 (after 24 collection efficiency corrections – CE corrections of 0.4 and 0.33 in summer and winter, 25 respectively) and the equivalent black carbon (eBC) concentrations measured by an 26 aethalometer (AE), were 8.58 ± 3.70 g m in summer and 10.08±8.04 g m in winter. 27 Organics dominated during both campaigns (summer/winter: 4.97 ± 2.92/4.55± 4.40 g m), 28 followed by SO4 in summer (1.68 ± 0.81/1.36± 1.38 g m) and NO3 − in winter (0.67 ± 29 0.38/2.03 ± 1.71 g m). The accumulation mode dominated the average mass size distribution 30 during both seasons, with larger particles of all species measured in winter (mode diameters: 31 Org: 334/413 nm, NO3 : 377/501 nm, SO4 : 400/547 nm, and NH4 : 489/515 nm) pointing to 32 regional and long-range transport. However, since the winter aerosols were less oxidized than 33 the summer aerosols (comparing fragments f44 and f43), the importance of local sources in the 34 cold part of the year was not negligible. The average PM1 particle effective density, defined as 35 the ratio of the mass to the volume of a particle, corresponded to higher inorganic contents 36 during both seasons (summer: ∼1.30 g cm and winter: ∼1.40 g cm). However, the effective 37 densities during episodes of higher mass concentrations calculated based on the particle number 38 (mobility diameter) and mass size distribution (vacuum aerodynamic diameter) were even 39 higher, ranging from 1.40 – 1.60 g cm in summer and from 1.40 – 1.75 g cm in winter. 40 Although aged continental air masses from the SE were rare in summer (7%), they were 41 https://doi.org/10.5194/acp-2021-516 Preprint. Discussion started: 14 July 2021 c © Author(s) 2021. CC BY 4.0 License.


47
Studies on airborne particulate matter (PM) are needed to better understand its temporal and 48 spatial variations, atmospheric processing, long-term trends, adverse health and environmental 49 consequences, and pollution sources (Putaud, et al., 2004;Tørseth et al., 2012;Belis et al., 50 2013; EEA 2019). Aerosol particles can be characterized by many different properties such 51 number concentration, mass concentration, particle size, mass, volume, density, etc. Particle 52 density is an important physical property of atmospheric particles and is linked to particle 53 emission sources and atmospheric physical and chemical ageing processes. The effective 54 density, which is defined as the ratio of the mass of the particle to its apparent volume, assuming 55 a spherical particle, and can be estimated by comparing the size distributions of the 56 aerodynamic and mobility diameters, is a quantity reflecting the physiochemical properties of 57 aerosol particles (e.g., DeCarlo 2004; Pitz et al., 2003Pitz et al., , 2008Hu et al., 2012;Qiao et al., 2018). 58 Over the last decades, a growing number of scientific studies have investigated the detailed 59 chemical composition of PM with variable temporal resolutions (1, 12, and 24 hours or higher) 60 using offline filter analyses (Putaud et al., 2010;Watson and Chow, 2011). Nowadays, online 61 methods with high temporal resolutions (30 min and less) are available, as aerosol mass 62 spectrometers (AMSs) are utilized that quantitatively measure chemical composition as well as 63 the chemically resolved size distributions of submicron non-refractory PM (NR-PM1) (Jayne et 64 al., 2000;Jimenez et al., 2003). Although measuring the seasonal variability of NR-PM1 is 134 To determine the collection efficiency (CE; Drewnick et al., 2005)   Therefore, CE correction was applied to the AMS data for both seasons to maintain consistency 141 in the data corrections. Similarly, using the same methodology, seasonal CE corrections 142 (summer CE = 0.29 and winter CE = 0.35) were also successfully applied to AMS data 143 measured at a suburban site in Prague (Kubelová et al., 2015).  145 Two approaches were employed to calculate the particle effective density. In the first approach, 146 AMS data representing the mass size distributions based on the vacuum aerodynamic diameter 147 ( ) in the size range from 10 to 7000 nm (calculated in Squirrelu software; 50 -800 nm in 148 reality) and MPSS data representing the dN/dlog Dp in the size range from 11.3 to 987 nm were 149 utilized. In the MPSS data, he were recalculated using the mobility diameters with a density 150 of 1.5 g cm -3 , and the were then recalculated back to mobility diameters with the 151 assumption of spherical particles as in DeCarlo et al. (2004): (2) 164 The densities were assumed to be approximately 1.75 g cm -3 for ammonium nitrate, ammonium 165 sulphate, and ammonium bisulphate (Lide, 1991); 1.52 g cm −3 for ammonium chloride (Lide,166 https://doi.org/10.5194/acp-2021-516 Preprint. Discussion started: 14 July 2021 c Author(s) 2021. CC BY 4.0 License. 1991); 1.20 g cm -3 for organics (Turpin and Lim, 2001); and 1.77 g cm -3 for black carbon (Park 167 et al., 2004).

179
To determine episodes of high particle number and mass concentrations, two approaches were 180 utilized: i) the application of positive matrix factorization (PMF) to PNSDs and ii) the depiction 181 of the mass size distribution of NR-PM1 species. The episodes were studied in detail from the 182 particle effective density and mass size distribution perspectives. to the receptor. Episodes in which the factor contributions to the total particle number 187 concentrations were higher than 80 % were chosen for the subsequent particle effective density 188 calculations.

189
The input data were prepared by merging three consecutive bins to reduce the noise in the raw 190 data, decrease the number of variables, and reduce the number of zeroes in the raw data (Leoni 191 et al., 2018). The uncertainties were calculated according to Vu et al. (2015). The total variables 192 were calculated by summing all the bins (N10 -800). PMF was conducted using different 193 uncertainty input matrices and different C3 (Vu et al., 2015) to obtain the Qtrue closest to 194 Qexpected; different modelling uncertainties and different numbers of factors were also applied.

195
A C3 of 0.8 was chosen.  ± 4.7 °C in summer and 1.4 ± 3.9 °C in winter, and negligible precipitation. The average PM2.5 206 was 10.9 ± 5.9 g m -3 in summer and 11.8 ± 9.9 g m -3 in winter (2019 average annual (2013). The average PNCs recorded during the two studied seasons were lower than the annual 217 mean total concentration (6.6 × 10 3 cm -3 , Zíková and Ždímal, 2013 8.04 g m -3 (filter-based 12-hour PM1 11.05 ± 7.22 g m -3 ), respectively. Since the PNSD (10-227 to 800-nm mobility diameter) was measured continuously in parallel with the eBC and NR-228 PM1 mass, mass closure of the 10-min averages was performed. To do so, two approaches were   Table 2. Organics dominated during both campaigns, followed by 4 2− in summer formation (Mbengue et al., 2018(Mbengue et al., , 2020. Another explanation could be the increased boundary 283 layer height, which enables mixing from higher altitudes and therefore the entrainment of aged, 284 and thus more oxidized, aerosols from long-range transport (Querol et al., 1998). On the other 285 hand, the winter season is characterized by fresh emissions of hydrocarbons owing to the 286 lowered boundary layer height in winter, which does not support the transport of oxidized 287 pollutants within the mixing layer (Schwarz et al., 2008).   (Zíková and Ždímal, 2013;Holubová Šmejkalová et al., 2021); however, accumulation-mode 296 particles were prominent in volume and species mass size distributions. The accumulation mode 297 of 4 2− does not show a large amount of variation, indicating a regional origin. In contrast, Based on the mass size distributions of the species (Fig. 2), ten summer (S1 -10) and 13 winter 308 (W1 -13) high-concentration episodes were selected (Table A1). The organic mass dominated 309 in summer; however, distinct episodes of high 4 2− concentrations (S2, S8, S9, S10) linked to 310 continental air masses from the NW and S-SE were also recorded (Fig. A6). In winter, episodes  freezing temperatures, which probably arose due to inversion conditions in Central Europe.

334
Organic aerosol ageing was examined on the f44 and f43 fragments (Fig. 3). Winter aerosols were 335 less oxidized than summer aerosols, pointing to the importance of local sources during the cold 336 part of the year. In summer, the oxidation rate of organic aerosols within the episodes does not 337 differ greatly, and most of the episodes revealed more oxidized organic aerosols (MOOAs) or 338 less volatile organic aerosols (LV-OOAs). Within the summer campaign, the most oxidized 339 aerosols were detected during afternoon episode S2 (Fig. 3), at which time the highest global 340 radiation was also measured (Table A1.). In contrast, S4, S6 and S7 represent night-time and 341 early morning episodes, and S5 represents a night-time and morning episode, and thus less 342 oxidized aerosols (Fig. 3). In winter, the difference between the episodes is more obvious, 343 mainly due to the higher variability in the local sources that influence the receptor site. The W7,

370
In winter, slow continental air masses from the SW cluster #1 (44%) prevailed. The air masses

387
The average mass size distributions of the main NR-PM1 species (except chloride) during the 388 entire summer and winter campaign are presented in Table 3. To determine the mode diameters 389 and the widths of the size distributions, the mass distributions were fitted with log-normal 390 modes using the Igor MultiPeak Package as follows:

392
where M is the amplitude, x0 is the peak position in nm, and width denotes the peak width. For 393 each season, the mean spectra were fitted separately with one peak, and fitting was also 394 performed for episodes S1-10 and W1-13. Due to the long duration of episode W6, the episode 395 was split into two sections: W6a (67 hours) and W6b (25.5 hours).

396
The accumulation mode dominated the average mass size distributions during both campaigns,

397
with larger particles of all species observed in winter (Table 3). Shifts towards larger 4 2− , except for during two episodes (W7 and W9) with regional transport (Table 3). An accumulation 403 mode of 4 2− with regional origin was even detected during a Mexico City Metropolitan Area  influenced by continental air masses of regional origin during the S2 episode (from the N-NE-432 E, Fig. A6). In contrast, the largest mode diameters (Org: 466 nm, continental origin and were also probably influenced by inversion conditions (Fig. A7).

439
Additionally, as expected, the Org particle size showed growth, and the increasing mode 440 diameter was more significant in the winter season, with the ageing of aerosols resulting in 441 oxygenated organic aerosols (Fig. 5).

449
The particle effective density was calculated for each episode of high particle numbers and 450 mass concentrations. The episodes were determined as follows: i) PMF application to PNSDs 451 and ii) depiction of mass size distributions of NR-PM1 species in a 3D plot (Fig. 2).

452
The PMF model was run several times until the most physically meaningful results and the best 453 diagnostics were obtained. The two-sided size bins containing variables (9.7 nm, 11.5 nm, 557. analysis, indicating that the solution was stable (Table A2). The non-normalized PNSD (N 460 cm −3 ) was analysed using the model.  The average summer density did not show a diurnal trend compared to the winter density ( Fig.   487 A13), followed by a diurnal trend (inverse dependence) observed for organics (Fig. A5). The 488 summer diurnal variation in the concentrations of organics was flatter than that in winter and 489 was not sufficient to significantly affect the diurnal density trend. In summer, we observed the 490 most significant diurnal trend for nitrate, but the absolute concentrations of nitrate were low, 491 and this variation therefore did not significantly affect the summer diurnal density trend (Fig.   492 A5).

493
In summer, with a higher ratio of ammonium sulphate, the density increased. In winter, the 494 density was influenced by the inorganic content (ammonium nitrate and sulphate). In both 495 seasons, the density increased with a decrease in the organic ratio and vice versa. This relation 496 evidently arises from the parameters in Eq.
(2) is linked to the density of organics, which was set 498 to 1.2 g cm -3 . The density applied for the organic fraction refers to the urban and urban 499 background stations (Turpin and Lim, 2001), and the organics density of a rural background 500 site is expected to be higher than that of an urban site due to organic aerosol ageing. However, 501 a density of 1.2 g cm -3 was also utilized in a study conducted by Freney et al. (2011) at a mid-502 altitude Puy-de-Dôme site and in a study conducted by Poulain et al. (2020) at a rural 503 background site in Melpitz. In this study, as the mass fraction of organics in the aerosols 504 increased, the density calculated using Eq. (2) converged to a value of 1.2 g cm -3 (Fig. A14).  (Salcedo et al., 2006).

511
The differences between the densities obtained using the two approaches (spectra fitting -Eq. eBC. In winter, the differences were larger, and both negative (compounds with lower densities 520 and/or particle physical characteristics) and positive (compounds with higher densities) 521 differences were obtained. However, the larger differences in winter could be strongly 522 influenced by the considerable CE correction applied to the AMS data.  Kubelová et al., 2015. 544 The accumulation mode dominated the average mass size distributions during both seasons,

545
with larger particles of all species in winter linked to seasonally differentiated regional and 546 long-range origins as well as to the variability in the local sources primarily observed in winter.

547
Although summer-aged continental air masses from the SE were rare (7%), they were connected 548 to the highest concentrations of all NR-PM1 species. In winter, the slow continental air masses 549 from the SW (44 %) linked to inversion conditions over Central Europe were associated with 550 the highest concentrations of organics, sulphate, nitrate, and ammonium.

551
The application of PMF on the PNSD enabled us to distinguish eight episodes of high particle 552 contributions to N10-800 to calculate the particle effective density based on the particle number 553 and mass size distributions. Additionally, a comparison of spectra fitting and chemical-based 554 calculations for determining the particle effective density during episodes of high mass 555 concentrations revealed differences in these two approaches due to the presence of compounds 556 that were not taken into consideration by the density calculations, such as particle physical 557 characteristics and calculation uncertainties.   fitting log-normal function to the AMS size distributions for the selected episodes in summer (S1 -S10) and 830 winter (W1 -13) along with meteorology recorded during the episodes (relative humidity -RH, global 831 radiation -GR, temperature -T, wind speed -WS and wind direction -WD)  Figure A6. Backward air mass trajectories calculated by HYSPLIT for corresponding summer 835 episodes (S1 -S10) of high concentration of species size distributions.