Insights into the Growth of Newly Formed Particles in a 1 Subtropical Urban Environment 2

The role of different chemical compounds, particularly organics, involved in the new particle 13 formation (NPF) and its consequent growth are not fully understood. Therefore, this study 14 was conducted to investigate the chemical composition of aerosol particles during NPF events 15 in an urban subtropical environment. Aerosol chemical composition was measured along with 16 particle number size distribution (PNSD) and several other air quality parameters at five sites 17 across an urban subtropical environment. An Aerodyne compact Time-of-Flight Aerosol 18 Mass Spectrometer (c-TOF-AMS) and a TSI Scanning Mobility Particle Sizer (SMPS) 19 measured aerosol chemical composition (particles above 50 nm in vacuum aerodynamic 20 diameter) and PNSD (particles within 9-414 nm in mobility diameter), respectively. Five NPF 21 events, with growth rates in the range 3.3-4.6 nm, were detected at two of the sites. The NPF 22 events happened on relatively warmer days with lower condensation sink (CS). Temporal 23 percent fractions of organics increased after the particles grew enough to have a significant 24 contribution to particles volume, while the mass fraction of ammonium and sulphate 25 decreased. This uncovered the important role of organics in the growth of newly formed 26 particles. Three organic markers, factors f 43 , f 44 and f 57 , were calculated and the f 44 vs f 43 27

The chemical composition of UFPs during NPF events in urban environments have been 29 investigated in only a handful of studies as it was thought that these events tend to happen 30 more in pristine air. However, NPF events have been observed frequently in Brisbane, an 31 urban area in Eastern Australia, and were associated with precursors emitted from traffic and 32 deployed to monitor the chemical composition of aerosol particles in the PM 1 fraction in real 1 time. A detailed description of the sampling method and c-TOF-AMS operation can be found 2 in Crilley et al. (2013). Briefly, the c-TOF-AMS was housed in a vacant room at each site and 3 sampled for 2-3 weeks. The sampling interval was 5 minutes, alternating equally between 4 particle time of flight and mass spectrometer modes. The instrument was calibrated for both 5 ion efficiency and particle size at the beginning of measurements at each site, and ion 6 efficiency calibration was also performed in the middle and at the end of measurement 7 campaigns at each site. All calibrations were performed according to the standard procedures 8 Solar radiation and other meteorological parameters were measured by a Monitor Sensors 10 µSmart Series weather station, a TSI 3781 water-based CPC was used for particle number 11 concentration (PNC) measurements, and a TSI DustTrak measured particle mass 12 concentration (PM 2.5 and PM 10 ). 13

Data analysis 14
Characterisation of New Particle Formation Events: NPF events were identified and 15 classified using the procedure developed by Dal Maso et al (Dal Maso et al., 2005). PNSD 16 data were analysed to identify a distinct new mode of particles in the nucleation mode size 17 range (particles with diameter less than 30 nm). If the mode grew in size and was persistent 18 for more than one hour, it was assigned as a NPF event. Once identified, the growth rate of 19 the nucleation events were calculated following the procedure described in (Creamean et al., 20 2011). 21 Condensation Sink: The surface area of aerosol particles available for condensation can be 22 estimated by the condensation sink (CS) parameter which is calculated from the measured 23 PNSD. The CS determines the rate of condensation of gaseous molecules on pre-existing 24 particles and is a function of PNSD (Pirjola et al., 1999;Lehtinen et al., 2003). CS, with unit s -25 1 , was calculated using the following formula (Lehtinen et al., 2003): 26 where D is the diffusion coefficient, d p is the particle diameter, N i is the concentration of 28 particles and M  can be expressed as (Fuks et al., 1971): where  is the sticking coefficient and is assumed to be 1 (Clement et al., 1996)  Particle Size: Due to the difference in detection limit between the two instruments (D va = 50 10 nm for AMS and D m = 10 nm for SMPS), measurement of the mass distribution for the 11 smallest particles detected during each nucleation was delayed for the AMS in order for 12 particles to grow to the detactable size . Ratio of mobility diameter (D m ), measured by the 13 SMPS, to vacuum aerodynamic diameter (D va ) is a function of size, composition, shape and 14 relative humidity for ambient particles. This ratio can be simplified to be equal to particle 15 density, assuming that particles are spherical (Jimenez et al., 2003a). The newly formed 16 particles were assumed to have the density of around 1.8 g/cm 3 and therefore, D m /D va = 1. 8 17 for newly formed particles (Zhang et al., 2004). Therefore, the AMS's lowest detection limit 18 will be equivalent to a D m = 30 nm. 19 Particle Chemical Composition: Concentrations of sulphate, nitrate, organics and ammonium 20 were measured by a c-TOF-AMS, as these have been identified as the main contributing 21 species to the NPF events and subsequent particulate growth in literature (Zhang et al., 2004).

Identified New Particle Formations 26
PNSD data was available for 6, 18, 11, 13, and 8 days in s1, s4, s11, s12, and s25 27 respectively. After the analysis of PNSD measured at five sites, five NPF events were 28 detected, where three events were observed in s12 and two in s25, while no events were 29 observed at s1, s4 and s11. Therefore, we focused only on s12 and s25, for which the NPF 30 growth rates varied from 3.3 to 4.6 (nmh -1 ) ( Table 1). 31 Figure 1 summarises the average solar radiation, relative humidity, temperature and CS 1 calculated using LOESS method for days with NPF events (nucleation days), as well as for 2 the days where no NPF events were observed (non-nucleation days) at s12 and s25. 3 Nucleation days had higher solar radiation intensity compared to non-nucleation days at s25 4 while opposite trend was observed at s12. Nucleation days had slightly higher temperature 5 compared to non-nucleation days at both sites. Higher relative humidity was observed on 6 nucleation days compared to non-nucleation days at s12, whereas opposite trend was 7 observed at S25. The wind speed/direction have not been plotted as they did not show a 8 typical trend during the nucleation and non-nucleation days. CS was found to be lower on 9 nucleation days compared to non-nucleation days, about two hours before the start of However, mass concentration of organics and nitrate increased dramatically between 3-10 pm 20 on nucleation days, at higher rate compared to non-nucleation days at s12, and the same trend 21 was observed at s25, with an hours delay in the increase of the mass concentration of organics 22 and nitrate. At s25, Sulphate and ammonium mass concentration started to increase at 12pm, 23 reached a peak and decreased subsequently, however, a significant increase was observed 24 around 4pm. 25

Evolution of Chemical Composition of Newly Formed Particles 26
The analysis and interpretation of the chemical composition of the newly formed particles 27 focusses mostly on the particles in the growth phase due to the lowest detection limit of the C-28 TOF AMS which does not permit to see them in the formation phase. Particle volume 29 distribution (PVD) was calculated from the PNSD data as it is a better measure for 30 comparison with the mass. 31 It has been previously determined in the literature that the main contributing species to NPF 1 events and subsequent particulate growth are sulphate, nitrate, organics and ammonium 2 (Zhang et al., 2004). Therefore, the percent fractions of each of these chemical species were 3 calculated by dividing the mass concentration of each chemical species by the total (sulphate 4 + nitrate + ammonium + organics). 5 Figure 3 illustrates the evolution of PNSD, PVD, and mass concentration of chemical species 6 (organics, nitrate, sulphate and ammonium) during the three consecutive NPF events at s12. 7 In general, mass concentration of the aerosol species followed the evolution of particle 8 volume distribution as expected. Ammonium, sulphate and nitrate mass fractions peaked just 9 before the particles volume increase due to the growth of newly formed particles. However, 10 organics mass fraction followed the opposite trend, with a significant rise after the increase in 11 particles volume due to the growth of new particles. In other words, the fraction of organics 12 increased and the fraction of ammonium, sulphate and nitrate decreased when the newly 13 formed particles grew enough to dominate the particles volume. This shows the import 14 contribution of organics to the growth of newly formed particles. 15 Time series of PNSD and mass concentration of particle species during two NPF events 16 happening on two consecutive days at s25 are illustrated in Figure 4. The mass concentrations 17 of the chemical species and their fractions followed similar trends as the ones at s12. At s12, 18 the magnitude of mass fractions changed from almost 50, 30, 10, and 10% before the 19 nucleation to 70, 20, 6, and 4% after the event for organics, sulphate, ammonium and nitrate 20 and respectively. At s25, the changes in mass fractions were more dramatic as they changed 21 from 40, 25, 25, and 10 % before the event to 85, 5, 5, and 5% after the nucleation occurred. 22

Role of organics 23
As indicated in the previous section, five NPF events were observed at s12 and the 24 contribution of organics to the growth of newly formed particles was high, therefore, further 25 data analysis regarding the role of organics were performed. Firstly, f 43  to non-nucleation days, K-means clustering technique was applied on the f 43 and f 44 data. A 12 period between 3-5 pm was selected for this purpose as this was the initial stage where the 13 newly formed particles grew enough to have significant contribution to the total signal. In 14 order to find the optimum number of clusters, SSE was plotted against the sequential number 15 of clusters and five number of clusters was found to be appropriate as it was located at the 16 elbow in the plot (Figure 6). 17 The five identified clusters as well as their 95% confidence ellipse are illustrated in Figure  18 7.a, 93% of the data measured in nucleation days were in clusters 3-5 (54%,15%, and 24% in 19 clusters3,4,and 5 respectively) while clusters1 and 2 contained 77% of the data measured in 20 non-nucleation days (Figure 7). In addition, cluster 1 and 5 contained less than 1% of 21 nucleation and non-nucleation days respectively. These show a distinct clustering on f 44 vs. smoothed average showed that f 57 followed exactly the same trend on nucleation and non-28 nucleation days, with an increase during the morning and afternoon rush hours (Figure 8). All 29 non-nucleation days did not have exactly the same meteorological conditions but the effects 30 of their variation on f 57 were minimal. 31

Summary and Conclusions 1
In summary, PNSD, chemical composition and meteorological parameters were measured at 2 five sites across the Brisbane Metropolitan Area. Five NPF events, with growth rates ranging 3 from 3.3-4.6 nm.hr -1 , were observed at two of the five sites, and the NPF events happened on 4 days with lower CS and higher temperature than non-event days. Higher sulphate, nitrate, 5 ammonium and organics were observed on nucleation days compared with days when no 6 nucleation was observed. Percent fractions of nitrate, sulphate, ammonium and organic 7 chemical species were calculated and their diurnal trends were modelled using the LOESS. 8 Ammonium, sulphate and nitrate mass fractions increased before the newly formed particles 9 grew enough to have a significant contribution to the particles volume, peaked around that 10 time and decreased after that. Conversely, the organics percent fraction increased significantly