Measurements of sub-3nm particles using a particle size magnifier in different environments: from clean mountain top to polluted megacities

Abstract

The measurement of sub-3nm aerosol particles is technically challenging. Therefore, there is a lack of knowledge about the concentrations of atmospheric sub-3nm particles and their variation in different environments. In this study, the concentrations of ∼1-3nm particles measured with a particle size magnifier (PSM) were investigated at nine sites around the world. Sub-3nm particle concentrations were highest at the sites with strong anthropogenic influence. In boreal forest, measured particle concentrations were clearly higher in summer than in winter, suggesting the importance of biogenic precursor vapors in this environment. At all sites, sub-3nm particle concentrations had daytime maxima, which are likely linked to the photochemical production of precursor vapors and the emissions of precursor vapors or particles from different sources. When comparing ion concentrations to the total sub-3nm particle concentrations, electrically neutral particles were observed to dominate in polluted environments and in boreal forest during spring and summer. Generally, the concentrations of sub-3nm particles seem to be determined by the availability of precursor vapors rather than the level of the sink caused by preexisting aerosol particles. The results also indicate that the formation of the smallest particles and their subsequent growth to larger sizes are two separate processes, and therefore studying the concentration of sub-3nm particles separately in different size ranges is essential.

Atmos. Chem. Phys., 17, 21632187, 2017 www.atmos-chem-phys.net/17/2163/2017/ doi:10.5194/acp-17-2163-2017 Author(s) 2017. CC Attribution 3.0 License.

Jenni Kontkanen1, Katrianne Lehtipalo1,2, Lauri Ahonen1, Juha Kangasluoma1, Hanna E. Manninen1, Jani Hakala1, Clmence Rose3, Karine Sellegri3, Shan Xiao4, Lin Wang4, Ximeng Qi5, Wei Nie5, Aijun Ding5, Huan Yu6,Shanhu Lee7, Veli-Matti Kerminen1, Tuukka Petj1, and Markku Kulmala1

1Department of Physics, University of Helsinki, 00014 Helsinki, Finland

2Paul Scherrer Institute, 5232 Villigen PSI, Switzerland

3Laboratoire de Mtorologie Physique, UMR6016, CNRS/UBP, 63178 Aubire, France

4Shanghai Key Laboratory of Atmospheric Particle Pollution and Prevention (LAP3), Department of Environmental Science & Engineering, Fudan University, 200433 Shanghai, China

5Joint International Research Laboratory of Atmospheric and Earth System Sciences, School of Atmospheric Sciences, Nanjing University, 210023 Nanjing, China

6School of Environmental Science and Engineering, Nanjing University of Information Science and Technology, Nanjing, China

7Department of Atmospheric Science, University of Alabama in Huntsville, Huntsville, Alabama Correspondence to: Jenni Kontkanen (jenni.kontkanen@helsinki.)

Received: 23 September 2016 Discussion started: 4 October 2016

Revised: 11 January 2017 Accepted: 22 January 2017 Published: 13 February 2017

Abstract. The measurement of sub-3 nm aerosol particles is technically challenging. Therefore, there is a lack of knowledge about the concentrations of atmospheric sub-3 nm particles and their variation in different environments. In this study, the concentrations of 13 nm particles measured

with a particle size magnier (PSM) were investigated at nine sites around the world. Sub-3 nm particle concentrations were highest at the sites with strong anthropogenic inuence. In boreal forest, measured particle concentrations were clearly higher in summer than in winter, suggesting the importance of biogenic precursor vapors in this environment. At all sites, sub-3 nm particle concentrations had daytime maxima, which are likely linked to the photochemical production of precursor vapors and the emissions of precursor vapors or particles from different sources. When comparing ion concentrations to the total sub-3 nm particle concentrations, electrically neutral particles were observed to dominate in polluted environments and in boreal forest during spring and summer. Generally, the concentrations of sub-3 nm particles seem to be determined by the availability of precursor vapors

rather than the level of the sink caused by preexisting aerosol particles. The results also indicate that the formation of the smallest particles and their subsequent growth to larger sizes are two separate processes, and therefore studying the concentration of sub-3 nm particles separately in different size ranges is essential.

1 Introduction

The majority of atmospheric aerosol particles, in terms of their number, are formed via gas-to-particle conversion (Spracklen et al., 2006; Yu et al., 2010), often referred to as new particle formation (NPF). NPF contributes signicantly to the global cloud condensation nuclei (CCN) budget and thus affects the climate (Merikanto et al., 2009; Wang and Penner, 2009; Kazil et al., 2010; Makkonen et al., 2012). According to current knowledge, NPF proceeds via the formation of molecular clusters from atmospheric vapors and their subsequent growth to larger sizes (Kulmala and Kerminen,

Published by Copernicus Publications on behalf of the European Geosciences Union.

Measurements of sub-3 nm particles using a particle size magnier in different environments: from clean mountain top to polluted megacities

2164 J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier

2008; Zhang et al., 2012; Kulmala et al., 2013, 2014). However, the chemical and physical processes leading to cluster formation and growth are not well understood, except in controlled systems in the laboratory (e.g., Kirkby et al., 2011;Almeida et al., 2013; Schobesberger et al., 2013; Kirkby et al., 2016; Lehtipalo et al., 2016; Trstl et al., 2016). The continuous existence of ion clusters in the atmosphere has been known for decades (see Hirsikko et al., 2011, and references therein), while knowledge about the concentrations and dynamics of atmospheric neutral clusters is more limited. There is evidence supporting the existence of neutral sub-3 nm particles and their importance in NPF, for example, in boreal forest (Kulmala et al., 2007, 2013), whereas some modeling studies claim that ion-mediated mechanisms dominate NPF there (Yu and Turco, 2000, 2008). Part of this controversy has been due to the inability to directly detect neutral clusters and sub-3 nm particles.

In recent years, the number of studies about the concentrations of atmospheric sub-3 nm particles has been increasing (Lehtipalo et al., 2010; Jiang et a., 2011a; Kulmala et al., 2013; Yu et al., 2014, 2016; Rose et al., 2015; Xiao et al., 2015; Kontkanen et al., 2016). However, the existing literature mainly comprises concentration data from campaign measurements in specic environments. Also, the chemical composition and sources of sub-3 nm particles (natural or anthropogenic) and their precursors in different environments are still largely unknown. In addition, it is unclear if neutral sub-3 nm particles exist in all environments or if ions dominate the sub-3 nm particle population in some conditions. Reecting this, the terminology for sub-3 nm particles has also been variable: they have been called, e.g., nano-CN (nanometer-sized condensation nuclei), nano-particles, clusters, or seeds, depending on the reference (McMurry et al., 2011). For convenience, in this article we refer to all measured sub-3 nm particles as particles, even though some of them can be ions, clusters, or even big molecules that are activated by the supersaturated vapor in condensation particle counters (CPCs), similarly as they could be activated in the atmosphere.

Until recently, measurements of atmospheric aerosol particles have been limited to the sizes above 3 nm, which is the

lowest detection limit of conventional ultrane CPCs (Mc-Murry, 2000). Sporadic measurements at smaller sizes have been reported, often with custom-made or modied instruments (e.g., Mordas et al., 2008; Sipil et al., 2008, 2009;Lehtipalo et al., 2009, 2010, 2011), but the data sets are not directly comparable due to differences in techniques and measured size ranges. The development of the Neutral cluster and Air Ion Spectrometer (NAIS) opened up the possibility to conduct systematic studies of ion concentrations down to 0.8 nm in mobility diameter and of the total concentrations, including both charged and neutral particles, down to about 2 nm (Kulmala et al., 2007; Manninen et al., 2009;Mirme and Mirme, 2013). The lowest size limit of the total concentration measurement with the NAIS is limited by the

size distribution of the corona ions used to charge the sample (Manninen et al., 2011). The development of CPCs using diethylene glycol (DEG) as a condensing vapor, which started after the study by Iida et al. (2009), has pushed the cutoff size of CPCs down to about 1 nm (Vanhanen et al., 2011;Jiang et al., 2011b; Kuang et al., 2012; Wimmer et al., 2013).Technically, all these CPCs consist of two stages: in the rst stage DEG is used to pre-grow particles, and in the second stage particles are grown further and counted by a conventional CPC. This new technology has played a key role in lling the gap between aerosol and mass spectrometric measurements and increasing the understanding of NPF starting from the molecular level (e.g., Kirkby et al., 2011; Kulmala et al., 2012, 2013, 2014; Almeida et al., 2013; Schobesberger et al., 2013; Kirkby et al., 2016). Here, we focus on the measurements performed with the particle size magnier (PSM), which is the rst commercially available DEG-based CPC (Vanhanen et al., 2011).

In this article, we review atmospheric measurements of sub-3 nm particle concentrations conducted with the PSM by different research groups. The measurement sites cover a wide range of different environments from a clean mountain top in France to very polluted Chinese megacities. Most of the data sets were obtained from 12 month intensive measurement campaigns, but we also present two longer-period and previously unpublished data sets from a boreal forest site and an urban environment in southern Finland. These measurements allow us to investigate the seasonal variation in sub-3 nm particles. The objective of this study is to provide the rst comparison on the concentrations and dynamics of sub-3 nm particles in different environments. More specically, we aim to get insights into (i) the concentration levels of sub-3 nm particles in different environments and their variation on a daily and seasonal basis, (ii) the fraction of ions of all sub-3 nm particles, and (iii) the possible sources and sinks of sub-3 nm particles.

2 Methods

2.1 The PSM

The PSM was developed at the University of Helsinki and later commercialized by Airmodus Ltd. The prototype instrument (PSMproto) and the measurement principle were introduced by Vanhanen et al. (2011). The rst generation of the commercial instrument is called A09, while the second generation with developments in the ow control and in the measurement software is called A10. The PSM A10, together with the Airmodus CPC A20 and the controlling software, is called the nano condensation nucleus counter (nCNC) system (A11). For simplicity, we refer here to all instruments as PSMs, regardless of small differences in the design and the variability in the counter CPCs deployed at different measurement sites.

Atmos. Chem. Phys., 17, 21632187, 2017 www.atmos-chem-phys.net/17/2163/2017/

J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier 2165

The operation principle of the PSM is based on a mixing-type CPC. The heated saturated ow is mixed turbulently with the colder sample ow to create supersaturation in the mixing region. DEG vapor starts to condense on particles in the mixed ow, and the particles grow in the growth tube of the instrument until they reach diameters of about 90 nm. After that the particles are sampled into a regular CPC for the further growth by the condensation of another vapor (usually butanol), and they are nally counted by an optical detector. The advantage of a mixing-type design is that the mixing ratio of the saturated and sample ow can be quickly and accurately adjusted by changing the saturator ow rate. This affects the supersaturation level which the particles encounter in the instrument and thus the cutoff size of the instrument.Comparing concentrations measured with CPCs at different cutoff sizes has often been used as a method to estimate the concentration of particles in the size range between the cutoff sizes (Alam et al., 2003; Kulmala et al., 2007). The PSM can be operated in a so-called scanning mode, in which the saturator ow and therefore also the cutoff size is changed continuously, which enables the measurement of particle size distributions (Gamero-Gastno and Fernndez de la Mora, 2000;Vanhanen et al., 2011; Lehtipalo et al., 2014). However, accurate calibrations are required for getting size information from measurements in the scanning mode.

Calibration methods in the sub-3 nm size range have evolved in recent years (Ude and Fernndez de la Mora, 2005; Sipil et al., 2009; Kangasluoma et al., 2014, 2015;Wimmer et al., 2015). Even so, the accuracy of calibrations is still limited by two factors. First, most of the calibrations are performed using electrically charged particles, as the size selection of particles is done based on their electrical mobility and an electrometer is used as the reference instrument for concentration. However, neutral particles need higher supersaturation to be activated in a CPC than charged particles (Winkler et al., 2008; Kangasluoma et al., 2016a). Second, the chemical composition of particles signicantly affects their activation in DEG-based CPCs (Jiang et al., 2011b;Kangasluoma et al., 2014, 2016a). For more discussion on the uncertainties caused by the charging state and chemical composition of particles, see Sect. 2.2.

The PSMs used in this study have been calibrated either with tetra-alkyl ammonium halides used as mobility standards (Ude and Fernndez de la Mora, 2005), tungsten oxide particles (Vanhanen et al., 2011), or ammonium sulfate clusters (Wimmer et al., 2013; Kangasluoma et al., 2014).The operation temperatures of each instrument were adjusted during the initial calibration so that only a few counts from homogenous nucleation were allowed at the highest saturator ow rates. The PSMs used in Helsinki and in Hyytil in 20152016 had an automatic background measurement system (Kangasluoma et al., 2016b), and they were thus allowed to have a higher background to maximize the activation efciency for organic clusters. The background counts were subtracted from the data during the data analysis. The con-

centration range measurable with the PSM is mainly depen

dent on the counter CPC.

The data measured in the scanning mode of the PSM need to be inverted to get a size distribution. Two methods have been presented by Lehtipalo et al. (2014). The rst one assumes a step-function-like cutoff curve for each saturator ow rate of the PSM. The difference in concentration between two ow rates (which determine the size bin limits) is corrected only by the detection efciency at the mean size of the bin to obtain the concentration in that size bin. This method resembles estimating the concentration of small particles by comparing the readings of two CPCs at different cutoff sizes. The second method takes into account the measured activation curves for each saturator ow rate and uses a nonnegative matrix inversion routine to calculate the size distribution. This method tends to lead to slightly higher concentrations than the rst method, which is partly due to rejecting negative values resulting from uctuations in the total concentration before data inversion and partly due to more accurate corrections for the detection efciency, which can have a large effect at the smallest sizes. In addition, all the data sets have been corrected for particle losses in the sampling lines (Kulkarni et al., 2001).

The different PSMs used in this study had slightly different lowest and highest cutoff sizes, and different size bins were used during the data inversion. Due to this and the uncertainties in determining the exact size limits (see Sect. 2.2), we chose to use the maximum size range available from the PSMs, which was from 1 to 23 nm for the PSMs op

erated in the scanning mode. If the largest size was smaller than 3 nm or if the PSM had not been operated in the scanning mode, we obtained the 13 nm concentration from the

difference in the concentration measured with the PSM and another aerosol instrument with the cutoff size of 3 nm, i.e., a differential mobility particle sizer (DMPS) or a scanning mobility particle sizer (SMPS). This way we aimed to obtain sub-3 nm particle concentrations for different sites in as similar size ranges as possible. Finally, when comparing different data sets, we calculated 30 min medians of all data. The time resolution of the PSM is 4 min in the scanning mode when averaging over an upward and a downward scan. The time resolution of the DMPS is typically 10 min, while the SMPS has a time resolution of about 3 min.

2.2 Measurement uncertainties

When interpreting the results of this study, one should keep in mind that obtained sub-3 nm particle concentrations are subject to considerable uncertainties. One major source of uncertainty is the effect of charging state and chemical composition of particles on the cutoff size of the PSM. As explained in the previous section, PSMs are generally calibrated using electrically charged clusters of a varying chemical composition. Kangasluoma et al. (2016a) showed that the cutoff diameter of the PSM is about 0.20.5 nm lower for charged

www.atmos-chem-phys.net/17/2163/2017/ Atmos. Chem. Phys., 17, 21632187, 2017

2166 J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier

particles than it is for neutral ones. If the majority of atmospheric sub-3 nm particles are assumed to be electrically neutral, the effect of charge can be taken into account by shifting the calibration to larger sizes, in which case the uncertainty in the cutoff size due to the charging state is reduced to [notdef]0.15 nm. On the other hand, the chemical composition of

particles can affect the cutoff size of the PSM even more than the charging state. Inorganic particles are activated more efciently by DEG vapor than organic particles, and thus their cutoff size can even be 1 nm lower than that of organic particles (Kangasluoma et al., 2014, 2016a). For inorganic particles the changes in the composition lead to about [notdef]0.2 nm

uncertainty in the cutoff size (Kulmala et al., 2013). Recent chamber experiments, mimicking atmospheric particle formation, indicate that if inorganic and organic precursor vapors are simultaneously present, they both participate in particle formation (Schobesberger et al., 2013). Therefore, when performing ambient measurements with the PSM, the cutoff diameters are likely somewhere between low cutoff diameters detected for pure inorganic clusters in laboratory experiments and high cutoff diameters detected for pure organic clusters.

In addition to the properties of the particles, ambient conditions can also affect the cutoff size of the PSM. The detection efciency of the PSM has been observed to increase when the humidity of the sample ow is increased (Kangasluoma et al., 2013). Thus, the changes in the water content of ambient air can induce changes in the cutoff size of the PSM. However, the diurnal cycle of air humidity can be estimated to have only a negligible effect on the PSM cutoff size (less than [notdef]0.1 nm) when considering the results of lab

oratory experiments by Kangasluoma et al. (2013) and the typical diurnal variation in air humidity. On the other hand, the seasonal cycle of air water content can be expected to have a larger effect on the cutoff size. Thus, to keep the cutoff sizes of the PSM constant in long-term measurements, regular background measurements need to be performed.

In addition to PSM measurements, DMPS and SMPS measurements were utilized to obtain sub-3 nm particle concentrations, which introduces additional uncertainty into our results. According to Wiedensohler et al. (2012), uncertainties in the size distribution measurements performed with a DMPS or SMPS system can be higher than 10 % for particles smaller than 3 nm. The measurements with the NAIS were also utilized at some of the sites. By comparing ion concentrations measured with the NAIS to the concentrations measured with the PSM, the fraction of ions of all sub-3 nm particles was determined. This fraction is subject to uncertainties in ion measurements of the NAIS as well as the abovementioned uncertainties in PSM measurements. Wagner et al. (2016) investigated the accuracy of the ion measurements conducted with the NAIS and concluded that the size information given by the NAIS is very accurate down to about1.5 nm size, while the concentrations of the smallest (< 5 nm) ions are often underestimated.

In addition to instrumental uncertainties, one also needs to consider the limitations in the comparability of the data from different sites, caused by differences in the measurement details. Due to the different settings of each PSM, which determine the level of supersaturation inside the instrument, and different calibrations, different PSMs had different cutoff sizes. In addition, when inverting the data to obtain particle size distributions, different size bins were chosen for different data sets. For these reasons, the size ranges measured with the PSM were not exactly same at different measurement sites (see Table 1). The differences in the lowest limit of the size range can particularly affect the comparability of the data. Furthermore, at different measurement sites, inlet systems were different (see Sect. 2.3), which may also affect the obtained sub-3 nm particle concentrations if the sampling line losses are not well characterized.

2.3 Measurement sites and instrumentation

In this study, PSM measurements from nine sites around the world were analyzed (Fig. 1). Measurements at each site are described below. In addition, the instruments used at different sites, the measurement time periods, and the size ranges for particle measurements are summarized in Table 1.

2.3.1 Hyytil (HTL)

In Hyytil measurements were conducted at the SMEAR II station (Station for Measuring Forest Ecosystem-Atmosphere Relations) in southern Finland (61 5[prime] N, 24 17[prime] E; 181 m above sea level) (Hari and Kulmala, 2005).The station is located about 200 km north of Helsinki. The closest urban area is the city of Tampere, which is located about 50 km southwest of the station and has a population of about 200 000. The station is surrounded by a Scots pine (Pinus sylvestris) forest, and monoterpenes dominate the emissions of biogenic volatile organic compounds (BVOCs) at the site (Rantala et al., 2015). The PSM measurements were conducted between 2010 and 2016. In the years before 2015, the measurements were shorter, intensive campaigns, while in 20152016 the measurement period covered 1 year (see Table 1 for the exact measurement periods). During the rst measurement campaign in 2010, the prototype PSM was used; after that in 2011 and 2012, the PSM model A09, in 2013 the PSM A10, and in 2014 and 2015 the PSM A11 (which is similar to A10) were used. The size bins used in the inversion were slightly different for different measurement campaigns as different instruments were used: the lowest cutoff size varied between 1.0 and 1.3 nm and the highest cutoff size between 2.0 and 2.5 nm (see Table 1). In 20102014 the sampling of particles was done with a 40 cm long inlet tube (6 mm in diameter) with 2.5 lpm (liters per minute) ow rate.Starting from 2015, an inlet specially designed to minimize sampling losses was used (Kangasluoma et al., 2016b). In this inlet, a ow of 7.5 lpm was taken directly from outside

Atmos. Chem. Phys., 17, 21632187, 2017 www.atmos-chem-phys.net/17/2163/2017/

J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier 2167

Figure 1. A map showing the locations of the measurements sites.

Table 1. Overview of the measurements analyzed in this study.

Measurement site Instruments Time period Size range (nm)

Hyytil (HTL 10 aut) PSMproto , DMPS, NAIS 427 Aug 2010 1.33.0 Hyytil (HTL 11 spr) PSMA09, DMPS, NAIS 17 Mar1 Apr 2011 1.13.0

Hyytil (HTL 11 aut) PSMA09, DMPS, NAIS 23 Aug11 Sep 2011 1.13.0 Hyytil (HTL 12) PSMA09, DMPS, NAIS 19 Apr9 May 2012 1.33.0 Hyytil (HTL 13) PSMA10, DMPS, NAIS 1 May23 Jul 2013 1.33.0 Hyytil (HTL 14) PSMA11, DMPS, NAIS 3 Apr30 May 2014 1.03.0 Hyytil (HTL 15) PSMA11, DMPS, NAIS 8 May 201530 Apr 2016 1.13.0 San Pietro Capoume (SPC) PSMA09, DMPS, NAIS 9 Jun9 Jul 2012 1.53.0 Puy de Dme (PDD) PSMA09, SMPS, NAIS 16 Jan29 Feb 2012 1.32.5 Brookhaven (BRH) PSMA09 , SMPS 22 Jul14 Aug 2011 1.33.0

Kent (KNT) PSMA09 , SMPS 15 Dec 20116 Jan 2012 1.33.0 Centreville (CTR) PSMA09, SMPS 1 Jun15 Jul 2013 1.12.1

Shanghai (SH) PSMA11 25 Nov 201323 Jan 2014 1.33.0 Nanjing (NJ) PSMA11, NAIS 1 Dec 201431 Jan 2015 1.13.0 Helsinki (HEL) PSMA11, DMPS 8 Jan 201531 Dec 2015 1.13.0

The PSM was not operated in the scanning mode.

air through a 40 cm long tube, and the actual sample (2.5 lpm) was taken from the middle of the ow with a probe (core sampling). An automatic background measurement was performed every third hour using ltered ambient air.

In addition to the PSM, measurements with a twin-DMPS system (Aalto et al., 2001) were utilized. The DMPS system measured the particle size distribution between 3 and 1000 nm. Thus, by subtracting the total particle concentration measured with the DMPS from the concentration measured with the highest cutoff size of the PSM, the particle concentration in the size range of 23 nm was obtained.

Furthermore, the ion size distribution between 0.8 and 42 nm was measured with the NAIS (Manninen et al., 2016). From these measurements, the concentration of sub-3 nm ions was obtained. The results of PSM measurements conducted during spring 2011 in Hyytil have been published by Kulmala et al. (2013).

2.3.2 Helsinki (HEL)

In Helsinki, measurements were performed at the SMEAR III station (60 12[prime] N, 24 58[prime] E; 26 m above sea level) (Jrvi et al., 2009). The city of Helsinki is located on the southern coast of Finland. The Helsinki metropolitan area, consisting of Helsinki and the neighboring municipalities, has a population of about 1.4 million. The measurement station is situated on a hill next to the university campus, about 5 km north of Helsinki city center. The surroundings of the station are heterogeneous, including buildings, parking lots, roads, deciduous forests, and low vegetation (for a more detailed description, see Jrvi et al., 2009). The measurements with the PSM A11 took place in 2015, covering 1 year (see Table 1). For the data inversion, the size bins of 1.11.3, 1.31.5, and 1.52 nm were used, which were identical with the PSM operated in Hyytil in 20152016. The sampling of

www.atmos-chem-phys.net/17/2163/2017/ Atmos. Chem. Phys., 17, 21632187, 2017

2168 J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier

the PSM was done by using an inlet system identical to the inlet used in Hyytil after 2014, including a core sampling probe and automatic background measurements (Kangasluoma et al., 2016b). Similarly to the procedure in Hyytil, a twin-DMPS system (Aalto et al., 2001) was used to measure the particle size distribution in the size range from 3 to 800 nm. Therefore, the concentration between 2 and 3 nm was obtained by subtracting the total particle concentration measured with the DMPS from PSM measurements.

2.3.3 San Pietro Capoume (SPC)

The San Pietro Capoume meteorological station is situated in northern Italy (44 39[prime] N, 11 37[prime] E; 11 m above sea level).

The station is located in the Po Valley with high emissions of anthropogenic pollutants, about 30 km northeast of the city of Bologna. The surroundings of the site are at and homogenous, mainly consisting of harvested elds (Decesari et al., 2001). The measurements with the PSM A09 were conducted at the station in JuneJuly 2012 during the PEGASOS (Pan-European GasAerosolClimate Interaction Study) zeppelin campaign (see Table 1). The cutoff sizes of the PSM were1.5 and 1.8 nm. In addition to the PSM, a twin-DMPS system covering the size range of 3600 nm (Laaksonen et al., 2005) was operated. Thus, by combining PSM and DMPS measurements the particle concentration in the size range of 1.53 nm was obtained. In addition, the ion concentration for the same size range was obtained from NAIS measurements. The results of these measurements, focusing on the analysis of NPF events, have been presented by Kontkanen et al. (2016).

2.3.4 Puy de Dme (PDD)

The Puy de Dme measurements site is located at the top of the Puy de Dme mountain in central France (45 46[prime] N, 2 46[prime] E; 1465 m above sea level). The station is surrounded by elds and forests. The closest town, Clermont-Ferrand, is located about 16 km east of the mountain at 396 m above sea level and has a population of about 150 000. See Venzac et al. (2009) for a more detailed description of the measurement site. The PSM A09 was operated at the station in JanuaryFebruary 2012 (see Table 1). From the PSM measurements, the particle concentration between 1.0 and 2.5 nm was obtained. Furthermore, the ion concentration in the same size range was obtained from NAIS measurements. A custom-made SMPS measuring the particle size distribution between 10 and 420 nm was also operated at the site. The SMPS was based on a TSI differential mobility analyzer (DMA) and a CPC (TSI 3010), and the sheath ow rate was controlled using a blower in a closed-loop arrangement. The data measured when relative humidity (RH) exceeded 98 % were omitted from the analysis, as high values of RH indicate that the station was inside a cloud (Rose et al., 2015). Rose et al. (2015) have published the results of this measurement campaign, concentrating on NPF events in the free tropo-

sphere (FT) and at the interface between the boundary layer and the FT.

2.3.5 Kent (KNT)

The measurements in Kent, Ohio, were conducted at the Kent State Universitys campus (41 9[prime] N, 81 22[prime] W; 320 m above sea level). Kent is a small Midwestern town with about 30 000 inhabitants. The closest larger cities are Akron (30 km west of Kent), Cleveland (60 km northwest), and Pittsburgh (100 km east). Measurements with the PSM A09 were performed between December 2011 and January 2012 (see Table 1). The PSM was operated with the xed saturator ow corresponding to the cutoff size of 1 nm. During the mea

surements, the ambient air was drawn at a ow rate of 3000 lpm into an air duct (1.5 m long and 10 cm in diameter) to which the PSM was directly connected via a 4 cm tube(0.64 cm in diameter). In parallel to the PSM measurements, the concentrations of particles in the size range of 3478 nm were measured with a combination of a nano SMPS (consisting of a TSI DMA 3085 and a TSI CPC 3776) and a long SMPS (consisting of a TSI DMA 3081 and a TSI CPC 3772) (see Yu et al., 2014, for more details). Thus, the concentration of particles between 1 and 3 nm was obtained by subtracting the total particle concentrations measured with the SMPSs from concentrations measured with the PSM. The results of these measurements, together with the measurements from Brookhaven (see below), have been presented in Yu et al. (2014).

2.3.6 Brookhaven (BRH)

In Brookhaven, New York, measurements were performed at Brookhaven National Laboratory on Long Island (40 52[prime] N, 72 53[prime] W; 24 m above sea level). The site is located 80 km east of New York City. The Long Island Sound is 16 km north of the site and the coast of Atlantic Ocean 25 km south.The measurement site is located within an urban neighborhood and surrounded by a mixed deciduous forest. The measurements with the PSM A09 were conducted in JulyAugust 2011 (see Table 1). Similarly to the procedure in Kent, the PSM had a xed cutoff of about 1 nm, and the particle concentration between 1 and 3 nm was obtained by combining PSM measurements with the SMPS measurements.The SMPS system in Brookhaven was similar to that used in Kent. During the measurements, ambient air was drawn into the instrument container at a ow rate of 150 lpm using a2.1 m tube (5.08 cm in diameter), and the PSM sampled from a split ow of 30 lpm with a 30 cm tube (2.54 cm in diameter). See Yu et al. (2014) for a more detailed description of these measurements.

2.3.7 Centreville (CTR)

The Centreville measurement site is located in Brent, Alabama (32 54[prime] N, 87 15[prime] W; 139 m above sea level). The

Atmos. Chem. Phys., 17, 21632187, 2017 www.atmos-chem-phys.net/17/2163/2017/

J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier 2169

landscape surrounding the research site is a mix of agricultural lands and mixed deciduous forests. Isoprene is the dominant BVOC emitted from forests, while agricultural lands have low emission rates of isoprene and monoterpenes. The strongest pollutant emission sources of NOx (17 000 tons yr1) and SO2 (92 000 tons yr1) in the state of Alabama are located within 100 km of the site (You et al., 2014). The measurements with the PSM A09 were conducted at the site during the SOAS (Southern Oxidant and Aerosol Study) campaign in JuneJuly 2013 (see Table 1). From the PSM measurements the particle size distribution between 1.1 and 2.1 nm was obtained. In addition, two SMPSs (TSI 3936) were used to measure the particle size distribution in the combined size range from 3 to 740 nm. One SMPS had a Nano DMA (TSI 3085) and a TSI 3786 water CPC. The second SMPS had a long DMA (TSI 3081) coupled with a TSI 3772 butanol CPC.

2.3.8 Shanghai (SH)

The measurements in Shanghai were conducted on the campus of Fudan University (31 18[prime] N, 121 30[prime] E) at about 20 m height from the ground. The site is located northeast of the center of Shanghai, which is the largest city in China with about 24 million inhabitants. One of the citys main highways is located 100 m south of the measurement site. The measurements with the PSM A11 were performed between November 2013 and January 2014 (see Table 1). The PSM was operated in the scanning mode, and the particle concentration between 1 and 3 nm was obtained from the measurements. In addition, a combination of a nano SMPS (consisting of a TSI DMA 3085 and a TSI CPC 3776) and a long SMPS (consisting of a TSI DMA 3081 and a TSI CPC 3775) was used to measure the particle size distribution between 3 and 615 nm. During the measurements, ambient air was drawn into a 5.0 m manifold (10.16 cm in diameter) at a ow rate of 4332 lpm. From this manifold, air was drawn at a ow rate of 1.75 lpm through an 18 cm tube (0.64 cm in diameter) and diluted with zero air ow at a ratio of 1 : 1 before enter

ing the PSM. Xiao et al. (2015) have published the results of these measurements, discussing especially the formation and growth rates of particles.

2.3.9 Nanjing (NJ)

In Nanjing measurements were performed at the Station for Observing Regional Processes of the Earth System (SORPES), which is situated about 20 km east of suburban Nanjing (Ding et al., 2013). The site is located on top of a hill on the Xianlin campus of Nanjing University (118 57[prime] E, 32 07[prime]4 N; 40 m above sea level). The measurements with the PSM A11 were conducted between December 2014 and January 2015 (see Table 1). The PSM was operated in the scanning mode and ve size bins between 1 and 3 nm were used for the inversion. In addition, AIS (air ion spectrometer;

Mirme et al., 2007) measurements were conducted, providing ion concentrations in the same size range (Hermann et al., 2013). The particle size distribution between 6 and 800 nm was also measured with a DMPS (Qi et al., 2015).

2.3.10 Supporting data

In addition to the measurements of sub-3 nm particle and ion concentrations, other data recorded at the measurement sites were utilized in the analysis. These data included different meteorological variables (e.g., temperature, RH, and radiation) and trace gas concentrations (e.g., SO2 and NOx). Condensation sink (CS), which describes the loss rate of vapor due to condensation on preexisting aerosol particles (Kulmala et al., 2001), was calculated from particle size distribution data measured with the DMPS or the SMPS. In addition, the concentration of sulfuric acid was measured with a CIMS (chemical ionization mass spectrometer; Eisele and Tanner, 1993) in Kent and Brookhaven (Yu et al., 2014), and in Hyytil during spring 2011 (Kulmala et al., 2013). For other measurement campaigns, sulfuric acid concentration was estimated using a proxy. For Hyytil the proxy in Petj et al. (2009) was used, as it has been derived and validated with measurements from this specic site. For other measurement sites, the proxy presented in Mikkonen et al. (2011) was utilized as it has been developed based on data from several different measurements sites.

3 Results and discussion

3.1 Sub-3 nm particle concentrations and their variation at different sites

3.1.1 Sub-3 nm particle concentrations at different sites

The concentration of sub-3 nm particles was observed to vary signicantly at each measurement site and between different environments. The medians (and different percentile ranges) of sub-3 nm particle concentration at different measurements sites are shown in Fig. 2 (see also Table 2). The concentration was highest at the sites with strong anthropogenic inuence in Nanjing and Shanghai, China, and in San Pietro Capoume, Italy. The median sub-3 nm particle concentration was 1.7 [notdef] 104 cm3 in Nanjing and 8.5 [notdef] 103 cm3 in

Shanghai and San Pietro Capoume. High concentrations were also observed at the urban site in Helsinki, Finland, where the median concentration was 5.8 [notdef] 103 cm3. At the

Finnish boreal forest site, Hyytil, the median concentration, calculated from all the data measured in 20102016, was lower than in Helsinki (2.0 [notdef] 103 cm3). The lowest

sub-3 nm particle concentrations were observed at the French mountain site, Puy de Dme, with a median concentration of 5.0 [notdef] 102 cm3, and at the North American sites Kent,

Brookhaven, and Centreville, where the median concentra-

www.atmos-chem-phys.net/17/2163/2017/ Atmos. Chem. Phys., 17, 21632187, 2017

2170 J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier

Figure 2. The variation in sub-3 nm particle concentration at different measurements sites. The red lines show the medians, the blue boxes indicate the 25th and 75th percentiles, and the vertical bars show the 5th and 95th percentiles.

tions were 4.7 [notdef] 102, 8.0 [notdef] 102, and 5.9 [notdef] 102 cm3, respec

tively.

The observed differences in sub-3 nm particle concentrations indicate that their formation is generally favored in polluted environments (see Sect. 3.4.1, where sulfuric acid concentration and condensation sink at different sites are compared). This can be explained by the high concentrations of low-volatile precursor vapors, which originate from, e.g., fuel combustion and trafc and can form small particles in the atmosphere (e.g., Arnold et al., 2012; Karjalainen et al., 2015; Sarnela et al., 2015). Some of the trafc-related particles may also be primary and formed inside vehicle engines (Jayaratne et al., 2010; Karjalainen et al., 2014; Alanen et al., 2015). At sites with lower anthropogenic inuence, like Puy de Dme, lower sub-3 nm particle concentrations were observed, which is likely due to the lower concentrations of precursor vapors and the absence of primary particle sources. On the other hand, in pristine environments the emissions of organic vapors from vegetation may promote the formation of sub-3 nm particles (Ehn et al., 2014). Interestingly, sub-3 nm particle concentration was clearly higher in a Finnish boreal forest, where BVOC emissions are dominated by monoterpenes, than in Centreville, southeastern US, where isoprene emissions dominate (Xu et al., 2015). Earlier, Kanawade et al. (2011) observed that NPF events are less frequent in mixed deciduous forests than in boreal forests, which they attributed to high emissions of isoprene.

When comparing concentrations between different sites, the median particle concentrations observed in Brookhaven and Kent can be considered to be relatively low compared to other urban sites. The low concentrations may be due to, for example, the properties of particles or ambient conditions, which can affect the detection efciency of the PSM, or technical reasons, such as the settings of the PSM or losses in the sampling lines (see also Sect. 2.2). Furthermore, one should note that measurements at different sites were conducted at different times of the year. Therefore, a possible seasonal

variation in sub-3 nm particle concentration due to the variation in their sources and sinks may bias the comparison. For example, in Kent and Puy de Dme, where the median concentrations were lowest of all sites, the measurements were conducted in winter when the photochemical production of precursor vapors and the emissions of biogenic vapors can be expected to be lower than in summer. The boundary layer dynamics may also affect concentrations especially at the high-altitude Puy de Dme site: in winter the station is often above the boundary layer, which prevents the transport of precursor vapors from near-ground sources to the site (Venzac et al., 2009).

3.1.2 Interannual variability in sub-3 nm particle concentration in Hyytil

Figure 3 presents sub-3 nm particle concentrations in Hyytil during different measurement campaigns starting from the rst eld measurements performed with the PSM in 2010 (see also Table 3). The data from 20152016, covering 1 year, were divided into spring (MarchMay), summer (JuneAugust), autumn (SeptemberNovember), and winter (DecemberFebruary) to enable the comparison to other years shorter measurement periods. Sub-3 nm particle concentration seems to have a clear seasonal variation in Hyytil (see also Sect. 3.1.3 and 3.2.2). The median concentrations were higher during measurements performed in spring and summer (9.4 [notdef] 1025.4 [notdef] 103 cm3) than in au

tumn and winter (5.8 [notdef] 1021 [notdef] 103 cm3). The measure

ments from different years agree rather well, despite the differences in the instrument model and the exact settings of the PSM and the sampling lines, which can affect the cutoff size of the instrument and particle losses. It needs to be noted, though, that in spring 2016 sub-3 nm particle concentration was on average lower than in other springs. This may be related to untypical environmental conditions, as the frequency of NPF events was clearly lower in that spring compared to other years (Table 3). The connection between sub-3 nm particle concentrations and environmental conditions is further discussed in Sect. 3.4. Generally, the median value and the variation in sub-3 nm particle concentration observed in the spring campaigns compare well to the concentrations reported by Lehtipalo et al. (2009, 2010), who measured 1.5 3 nm particles with a pulse-height CPC in Hyytil during spring 2007 and 2008.

3.1.3 Particle concentrations in different size bins in Hyytil and Helsinki

In addition to the total sub-3 nm particle concentration, the concentrations in different sub-3 nm size bins were investigated for Helsinki and Hyytil. For this, the data sets from 2011 and 20152016 were used, as they had almost identical size bins: 1.11.3, 1.31.5, 1.52, and 23 nm in 2015 2016 and 1.11.3, 1.31.5, 1.52.1, and 2.13 nm in 2011. To

Atmos. Chem. Phys., 17, 21632187, 2017 www.atmos-chem-phys.net/17/2163/2017/

J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier 2171

Figure 3. Sub-3 nm particle concentrations in Hyytil during different measurement campaigns. The red lines show the medians, the blue boxes indicate the 25th and 75th percentiles, and the vertical bars show the 5th and 95th percentiles. Note that the data from 20152016 are divided into different seasons: summer (sum), autumn (aut), winter (wint), and spring (spr).

Table 2. Medians of sub-3 nm particle concentration, the ratio of ion concentration to the total sub-3 nm particle concentration, sulfuric acid concentration, condensation sink, and the frequency of new particle formation (NPF) events at different measurement sites. Sulfuric acid concentration is estimated from a proxy for all other campaigns except those that are marked with an asterisk ( ). The explanations for abbreviations, the measurement periods and the exact size ranges for particle measurements are shown in Table 1. Data from Hyytil (HTL)

and Helsinki (HEL) are divided into different seasons: spring (spr), summer (sum), autumn (aut), and winter (wint).

Measurement Sub-3 nm particle Ions to all Sulfuric acid Condensation NPF event site concentration particles ratio concentration sink frequency

(cm3) (cm3) (s1) ( %)

HTL spr 2.9E+03 0.16 1.0E+06 2.6E03 40

HTL sum 2.0E+03 0.33 2.4E+05 3.6E03 19

HTL aut 7.9E+02 0.83 2.6E+05 2.0E03 15

HTL wint 5.8E+02 0.71 6.9E+05 2.1E03 0

SPC 8.5E+03 0.004 1.0E+07 1.2E02 86

PDD 5.0E+02 0.60 3.8E+06 3.6E03 23

BRH 8.0E+02 3.3E+05 6.7E03 17

KNT 4.7E+02 9.4E+05* 6.7E03 22

CTR 5.9E+02 0.47 4.0E+04* 1.5E02 9

SH 8.5E+03 3.1E+07 7.6E02 21

NJ 1.7E+04 0.02 2.0E+07 2.7E02 20

HEL spr 7.8E+03 2.0E+06 4.1E03 13

HEL sum 5.1E+03 2.5E+06 5.3E03 4

HEL aut 4.1E+03 9.2E+05 4.3E03 12

HEL wint 6.9E+03 2.2E+05 3.6E03 8

Sulfuric acid concentration was measured.

investigate the seasonal variation in particle concentrations, the data sets were divided into spring (MarchMay), summer (JuneAugust), autumn (SeptemberNovember), and winter (DecemberFebruary). The particle concentrations in different size bins in these seasons are presented in Table 4. Note that, for clarity, only the size bin limits used in 20152016 are marked in the table.

A seasonal variation in particle concentrations was ob-served at both sites. In Hyytil, sub-3 nm particle concentration was higher in summer and spring than in winter and autumn. In the sub-2 nm size bins the concentrations were highest in summer; this was clear especially in the smallest

size bin (1.11.3 nm) where the median concentration was1.1 [notdef] 103 cm3 in summer and 2.4 [notdef] 102 cm3 in winter. In

the largest size bin (23 nm), the highest concentrations were detected in spring, with the median concentrations ranging from 1.1 [notdef] 102 to 3.0 [notdef] 102 cm3 in different seasons. The

summer-time maximum in the concentration of the smallest particles is likely related to the strong photochemical production of precursor vapors and the high emissions of organic vapors from vegetation at this time of the year. Seasonal differences were observed also in the ratio of 1.12 to 23 nm particle concentrations in Hyytil. In summer and autumn, the 12 nm particle concentration was 68 times higher than

www.atmos-chem-phys.net/17/2163/2017/ Atmos. Chem. Phys., 17, 21632187, 2017

2172 J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier

Table 3. Medians of sub-3 nm particle concentration, the ratio of ion concentration to the total sub-3 nm particle concentration, sulfuric acid concentration, condensation sink, and the frequency of new particle formation (NPF) events in Hyytil during different measurement campaigns. Sulfuric acid concentration is estimated from a proxy for all other campaigns except in spring 2011.

Measurement Sub-3 nm particle Ions to all Sulfuric acid Condensation NPF event campaign concentration particles ratio concentration sink frequency

(cm3) (cm3) (s1) ( %)

HTL 10 aut 6.7E+02 0.56 4.1E+05 3.1E03 8

HTL 11 spr 2.7E+03 0.17 1.1E+06 1.8E03 75

HTL 11 aut 7.2E+02 0.76 6.9E+05 2.8E03 10

HTL 12 spr 3.2E+03 0.10 4.1E+05 2.2E03 62

HTL 13 sum 4.5E+03 0.10 2.0E+05 3.8E03 37

HTL 14 spr 5.4E+03 0.09 3.8E+05 2.9E03 48

HTL 15 sum 2.1E+03 0.37 2.1E+05 3.6E03 12

HTL 15 aut 1.0E+03 0.83 2.2E+05 2.0E03 17

HTL 15 wint 5.8E+02 0.71 6.9E+05 2.1E03 0

HTL 16 spr 9.4E+02 0.57 4.0E+05 2.6E03 13

Sulfuric acid concentration was measured.

Table 4. Medians of the total particle concentration and the ratio of ion concentration to the total particle concentration in four size bins(1.11.3, 1.31.5, 1.52, and 23 nm) in Hyytil (HTL) in 2011 and 20152016 and in Helsinki (HEL) in 2015. Data are divided into different seasons: spring (spr), summer (sum), autumn (aut), and winter (wint).

Measurement site Particle concentration (cm3) Ions to all particles ratio

Size range (nm) 1.11.3 1.31.5 1.52.0 2.03.0 1.11.3 1.31.5 1.52.0 2.03.0

HTL spr 4.6E+02 1.6E+02 7.8E+01 3.0E+02 0.75 1.02 0.78 0.03

HTL sum 1.1E+03 2.2E+02 1.3E+02 1.9E+02 0.34 1.25 0.89 0.06

HTL aut 3.8E+02 1.5E+02 9.0E+01 1.1E+02 1.05 1.45 1.02 0.07

HTL wint 2.4E+02 6.8E+01 4.2E+01 1.7E+02 1.05 1.36 0.79 0.03

HEL spr 2.3E+03 6.7E+02 7.5E+02 2.2E+03

HEL sum 1.2E+03 3.9E+02 4.1E+02 1.9E+03

HEL aut 1.1E+03 3.7E+02 3.2E+02 1.9E+03

HEL wint 2.1E+03 8.6E+02 7.2E+02 2.1E+03

the concentration in the 23 nm size range, while in spring and winter the difference was only a factor of 22.5. This may indicate that in summer and autumn sub-2 nm particles are not able to grow to sizes larger than 2 nm efciently. In spring, environmental conditions in Hyytil are favorable for particle growth, as indicated by frequent NPF events (Table 2; Dal Maso et al., 2005), which probably explains why the difference between the size bins was then smaller. In winter, on the other hand, particle concentrations were low in all size bins.

In Helsinki differences in particle concentrations between different seasons were less distinct than in Hyytil. The highest concentrations were detected in spring and winter.In the smallest size bin (1.11.3 nm) the median concentration varied between 1.1 [notdef] 103 and 2.3 [notdef] 103 cm3 in dif

ferent seasons and in the largest size bin (23 nm) between1.9 [notdef] 103 and 2.2 [notdef] 103 cm3. The high wintertime concen

trations suggest that in Helsinki the formation of sub-3 nm particles is unlikely to be driven by the emissions of organic

compounds from biogenic sources. In Helsinki the ratio of1.12 to 23 nm particle concentrations varied between 1 and 2 in different seasons, and it was highest in winter and spring and lowest in summer and autumn. The lower value of this ratio in Helsinki compared to Hyytil indicates that particle growth may be favored in an urban environment with stronger anthropogenic inuence compared to clean boreal forest (Kulmala et al., 2005). It should be kept in mind, though, that the composition of particles can be different in different environments, which can affect their activation probability in the PSM (see Sect. 2.2).

3.2 Diurnal variation in sub-3 nm particle concentration

3.2.1 Diurnal variation at different sites

The median diurnal variation in sub-3 nm particle concentration at different measurement sites is presented in Fig. 4.The differences in concentrations between different sites

Atmos. Chem. Phys., 17, 21632187, 2017 www.atmos-chem-phys.net/17/2163/2017/

J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier 2173

Figure 4. The median diurnal variation in sub-3 nm particle concentration at different measurement sites.

are obvious also here: sub-3 nm particle concentration was high at sites with strong anthropogenic inuence and lower in cleaner environments. Generally, sub-3 nm particle concentrations were highest during the daytime and lowest at night. Still, at many sites moderate concentrations were ob-served also at night. The highest daytime and nighttime concentrations were detected in Nanjing, Shanghai, and San Pietro Capoume. The next highest concentrations were ob-served in Helsinki, where particle concentration was high from around 08:00 to 16:00 LT (local time). In Hyytil sub-3 nm particle concentration was lower than in these urban sites, and the median diurnal cycle of concentration was also clearly weaker. It needs to be noted, though, that in Hyytil the diurnal cycle had a strong seasonal variation, which is discussed in the next section (Sect. 3.2.2). In Kent, Brookhaven, and Puy de Dme, sub-3 nm particle concentrations were low but had a similar diurnal cycle to other sites: the concentrations were lowest early in the morning and highest around noon. In Brookhaven particle concentration additionally had a secondary maximum in the evening after 19:00 LT. In Centreville sub-3 nm particle concentration had a minimum early in the morning and two separate maxima; the rst maximum occurred before noon and the second in the evening around 20:00 LT.

The observed daytime maxima in sub-3 nm particle concentrations likely result from the photochemical production of low-volatile precursor vapors during the daytime and the emissions of precursor vapors, and possibly also primary particles, from different anthropogenic and biogenic sources. Daytime maxima in sub-3 nm particle concentrations have also been reported in previous studies (Kulmala et al., 2013; Yu et al., 2014, 2016; Xiao et al., 2015; Rose et al., 2015; Kontkanen et al., 2016), where they have often been linked to NPF events. On the other hand, the daytime increase in sub-3 nm particle concentration is not necessarily followed

by an NPF event where particles grow to large sizes (Yu et al., 2014, 2016; Xiao et al., 2015; see Sect. 3.5). At urban sites, the diurnal cycle of sub-3 nm particle concentrations can be affected by variation in trafc conditions and other anthropogenic activities, which could explain the wide maximum in particle concentration observed in Helsinki. Furthermore, in Puy de Dme the diurnal variation in sub-3 nm particle concentration can be inuenced by the diurnal cycle of the boundary layer height, affecting the transport of precursor vapors to the site (Venzac et al., 2009; Rose et al., 2015).

The relatively high particle concentrations observed at many sites at night suggest that the formation of sub-3 nm particles may also occur in the absence of solar radiation.This may imply the importance of low-volatile precursor vapors originating from the oxidation of, for example, organic compounds by ozone or nitrate radical (Ehn et al., 2014).Kirkby et al. (2016) observed in their chamber experiments that -pinene ozonolysis products can form new particles efciently even in the absence of sulfuric acid. In Hyytil high concentrations of sub-3 nm particles in the evening have been reported earlier (Lehtipalo et al., 2009), and they have been proposed to be related to the ozonolysis products of monoterpenes (Lehtipalo et al., 2011). Evening maxima are frequently observed also in sub-3 nm ion concentrations in Hyytil (Junninen et al., 2008; Buenrostro Mazon et al., 2016). In Brookhaven Yu et al. (2014) found that the elevated concentrations of sub-3 nm particles at night were linked to marine air masses, and they were probably not connected to the oxidation of monoterpenes. In Centreville, where the evening maximum in sub-3 nm particle concentration was most distinct, BVOC emissions are dominated by isoprene (Xu et al., 2015). In earlier measurements in an isoprene-rich deciduous forest, the concentrations of 310 nm particles were observed to increase in the evening when SO2 concentration was high (Kanawade et al., 2011).

3.2.2 Diurnal variation in Hyytil and Helsinki in different seasons

To study the diurnal variation in sub-3 nm particle concentration in different seasons, the data sets from Hyytil and Helsinki (from where longer times series were available) were divided into spring, summer, autumn, and winter. Figure 5 illustrates the median diurnal cycle of sub-3 nm particle concentration in Hyytil and Helsinki in these seasons.

In Hyytil, the diurnal cycle of sub-3 nm particle concentration was stronger in spring and summer than in autumn and winter. In spring, particle concentration started to increase from the nighttime level after 05:00 LT and reached a maximum around 14:00 LT. Sub-3 nm particle concentration had a daytime maximum also in summer, although it was not as distinct as during spring. In autumn and winter particle concentration stayed relatively low throughout the day. The more pronounced daytime maximum in spring and summer

www.atmos-chem-phys.net/17/2163/2017/ Atmos. Chem. Phys., 17, 21632187, 2017

2174 J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier

Figure 5. The median diurnal variation in sub-3 nm particle concentration in Hyytil (HTL; dashed lines) and in Helsinki (HEL; solid lines) in different seasons: spring (spr), summer (sum), autumn (aut), and winter (wint).

than in other seasons is likely related to the stronger production of low-volatile precursor vapors in these months.

In Helsinki the daytime maximum in sub-3 nm particle concentration was distinct in all seasons. Sub-3 nm particle concentration started to rise in the morning after 05:00 LT, was highest around midday, and started to decrease again after 16:00 LT. The highest concentrations were obtained in spring and lowest in autumn. The similarities in the diurnal cycle of sub-3 nm particle concentration in different seasons suggest that in Helsinki sub-3 nm particle concentrations are more affected by anthropogenic sources of precursor vapors, which typically are fairly constant throughout the year, than the emissions from biogenic sources. The fact that particle concentration started to rise in the morning at the same time in all seasons implies that the increase is not triggered by photochemistry. Instead, it may be related to the morning trafc on the nearby roads. This hypothesis is supported by the results of Jrvi et al. (2008), who found that the trafc rate on the road close to the SMEAR III station increased in the morning after 05:00 LT, which coincided with the rise in black carbon concentration at the station.

3.2.3 Diurnal variation in particle concentration in different size bins

The median diurnal cycle of particle concentration in Helsinki and Hyytil was also studied in different sub-3 nm size bins (Fig. 6). For this, only data from the years 2011 and 20152016 were utilized as the size bins were most comparable with each other in these years (see Sect. 3.1.3).

In Hyytil clear differences in the diurnal cycle of particle concentrations in different size bins were observed. In the size bin of 1.11.3 nm, the particle concentration had its strongest diurnal cycle in summer: the concentration had a

minimum in the early morning, after which it increased and stayed high between 10:00 and 21:00 LT. In other seasons, the 1.11.3 nm particle concentration stayed more stable. In the size bin of 1.31.5 nm, the particle concentration did not have a strong diurnal cycle in any season. However, in the size bins of 1.52 nm and 23 nm, the particle concentration increased during the daytime in spring. This increase is probably linked to NPF events, which are most frequent in Hyytil in spring (Table 2). On the other hand, the high daytime concentrations of 1.11.3 nm particles in summer likely result from the strong production of precursor vapors, originating from biogenic sources, at this time of the year.When comparing the diurnal cycles in different size bins to the diurnal cycle of the total sub-3 nm particle concentration in Hyytil (Fig. 5), one can see that in summer the daytime maximum in the total sub-3 nm particle concentration was caused by the maximum in the concentration of the smallest,1.11.3 nm particles, while in spring the daytime peak was mostly due to the increase in the concentration of the largest, 23 nm particles.

In Helsinki the diurnal cycles were quite similar in different size bins and during different seasons. In all size bins, particle concentrations were lowest in the early morning (around 04:00 LT), after which they started to increase reaching the maximum around midday, and started to decrease again after 16:00 LT. In spring the daytime peak values were higher than in other seasons; this was clear especially in the size bins of 1.11.3 and 1.52 nm. On the other hand, nighttime particle concentrations in sub-2 nm size bins were highest in winter. In the 23 nm size bin, the diurnal cycles in different seasons were almost identical.

3.3 Ratio of ions to total sub-3 nm particle concentrations

3.3.1 Ion ratio at different sites

To determine how large a fraction of measured sub-3 nm particles was electrically charged, we studied the ion ratio,i.e., the ratio of sub-3 nm ion concentration measured with the NAIS to the total sub-3 nm particle concentration measured with the PSM in the corresponding size range. The ion ratios at different sites (only those with ion measurements available) are depicted in Fig. 7 (see also Table 2).Generally, the ion ratios were mainly determined by the total sub-3 nm particle concentration, as the ratios were lowest at the sites with the highest total concentrations and highest at sites with the lowest total concentrations. This results from the smaller variation in ion concentrations between different environments than in the total particle concentrations: the median sub-3 nm ion concentrations (calculated for the same size ranges as the total particle concentration) were between 3 [notdef] 102 and 7 [notdef] 102 cm3 at different sites. The ob-

served smaller variation in ion concentrations is consistent with earlier observations by Manninen et al. (2010), and it

Atmos. Chem. Phys., 17, 21632187, 2017 www.atmos-chem-phys.net/17/2163/2017/

J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier 2175

Figure 6. The median diurnal variation in particle concentration in four size bins (1.11.3, 1.31.5, 1.52, and 23 nm) in Hyytil (HTL; dashed lines) in 2011 and 20152016 and in Helsinki (HEL; solid lines) in 2015. The data are divided into different seasons: spring (spr), summer (sum), autumn (aut), and winter (wint).

can be explained by the fact that ion concentrations at the ground level are generally limited by ion production rates (Hirsikko et al., 2011). In San Pietro Capoume, the median ion to the total particle concentration ratio was as low as 0.004, and in Nanjing it was 0.02. At other sites the ion ratios were higher. In Centreville the median ion ratio was0.5, and in Puy de Dme it was 0.6. On the other hand, Rose et al. (2015) showed that neutral particles dominate in Puy de Dme during NPF events. In Hyytil, the ion ratio exhibited a strong seasonal variation. The median ion ratio was rather low in spring and summer (0.2 in spring and 0.3 in summer), which is consistent with the high total sub-3 nm particle concentrations observed in these seasons. In autumn the median ion ratio was 0.8, and in winter it was 0.7, and the ratio often exceeded unity in these seasons (see the discussion below). Lehtipalo et al. (2009, 2010) observed the ion ratio to be only about 0.01 in springtime in Hyytil by comparing ion concentrations measured with a BSMA (balanced scanning mobility analyzer) to the total concentrations from

a pulse-height CPC, but their measurements only started at about 1.31.5 nm.

The ion ratios exceeding unity, observed in Hyytil, Puy de Dme, and Centreville, are not physical and thus indicate that the PSM is not able to detect all 13 nm ions and particles. This may be caused by uncertainties in the detection efciency of the PSM due to the properties of particles and changing environmental conditions (Kangasluoma et al., 2013, 2016a; see Sect. 2.2). Especially, it should be noted that charged particles have been observed to be activated in the PSM more efciently than neutral particles. Therefore, it is possible that the PSM detects charged particles of a certain size but not the neutral species of the same physical size. Further work is still needed to consider these issues when conducting eld measurements with the PSM. In addition, inaccuracies in ion concentrations measured with the NAIS may also cause uncertainties in ion ratios (Wagner et al., 2016; see Sect. 2.2).

Finally, it needs to be noted that the observed ion ratios depend strongly on the limits of the studied size range. This

www.atmos-chem-phys.net/17/2163/2017/ Atmos. Chem. Phys., 17, 21632187, 2017

2176 J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier

Figure 7. The ratio of sub-3 nm ion concentration to the total particle concentration at different measurement sites. The red lines show the medians, the blue boxes indicate the 25th and 75th percentiles, and the vertical bars show the 5th and 95th percentiles. The data from Hyytil (HTL) are divided into different seasons: spring (spr), summer (sum), autumn (aut), and winter (wint).

is due to the pool of small ions which is constantly present in the atmosphere because of ionization of air molecules (e.g., Hirsikko et al., 2011). For example, in Hyytil the median size of this ion pool is about 1.11.3 nm (Manninen et al., 2009). Thus, the observed differences in the ion ratio between different measurement sites and different measurement campaigns in Hyytil (see Tables 2 and 3) can partly be due to differences in the studied size ranges. In the next section the ion ratio in Hyytil is studied separately in different sub-3 nm size bins.

3.3.2 Ion ratio in Hyytil in different size bins

Table 4 shows the ratios of ion concentrations to the total particle concentrations separately in four sub-3 nm size bins in Hyytil (data only from the years 2011 and 20152016). The ratio was highest in the size bins below 2 nm, as anticipated due to the constant pool of small ions (Manninen et al., 2009; Hirsikko et al., 2011). In the smallest size bin,1.11.3 nm, the ratio was lowest in summer (median value0.3) when the total particle concentration in that size bin was high. In spring the median ratio in this size bin was 0.8, and in autumn and winter it was 1. In the next two size bins (1.31.5 and 1.52 nm) the ion ratio was high in all seasons, with the median ratios ranging between 0.8 and 1.5. This further demonstrates that the PSM does not detect all sub-2 nm particles, as discussed in the previous section (3.3.1). On the other hand, it should be noted that during spring 2016, sub-3 nm particle concentrations observed in Hyytil were lower than in other years (see Table 3). This can partly explain the rather high value obtained for the springtime ion ratio. In the largest size bin, 23 nm, the ion ratio was low in all seasons, with the median ratios varying between 0.03 and 0.07. This is expected, as at this size range most ions originate from diffusion charging of neutral aerosol particles by collisions with

the small ions or from ion-induced nucleation. In Hyytil 23 nm ions have been observed to exist almost only during NPF events (Leino et al., 2016).

3.3.3 Diurnal variation in the ion ratio

The median diurnal cycles of the ratio of ion concentration to the total sub-3 nm particle concentration at different sites are illustrated in Fig. 8. At all sites the ion ratio was lowest during the daytime and highest early in the morning, having the opposite diurnal cycle to that of the total sub-3 nm particle concentration (see Fig. 4). In San Pietro Capoume the ratio was very low throughout the day, reaching the highest value in the morning around 06:00 LT. Kontkanen et al. (2016) also observed the morning maximum in the ion ratio in San Pietro Capoume and proposed that it was caused by the earlier increase in ion concentration than in the concentration of neutral particles during NPF events. In Nanjing, Puy de Dme, and Centreville, the ion ratio was lowest around midday and highest at night or in the early morning. As discussed in the previous sections, in Hyytil the ion ratio was lowest in spring and summer and highest in autumn and winter. In spring and summer the ion ratio was lowest around noon and highest in the early morning. In autumn and winter the diurnal variation in the ratio was weaker and the ratio was high throughout the day.

Figure 9 presents the median diurnal cycles of the ratio of the ion concentration to the total particle concentration in different sub-3 nm size bins in Hyytil in different seasons. In the size bin of 1.11.3 nm the ion ratio had its strongest diurnal cycle in summer, when the ratio had a minimum during the daytime reecting the increase in the total particle concentration (see Fig. 6). In the next size bin, 1.31.5 nm, the ratio did not have a clear diurnal cycle in any seasons. However, in the two largest size bins, 1.52 and 23 nm, the

Atmos. Chem. Phys., 17, 21632187, 2017 www.atmos-chem-phys.net/17/2163/2017/

J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier 2177

Figure 8. The median diurnal variation in the ratio of sub-3 nm ion concentration to the total particle concentration during different measurement campaigns. The data from Hyytil (HTL) are divided into different seasons: spring (spr), summer (sum), autumn (aut), and winter (wint).

ion ratio decreased during the daytime in spring and winter. In spring this decrease may be related to the formation of particles in NPF events, which are frequent at that time of the year. In the size bin of 23 nm, the ion ratio also had an evening maximum in autumn, which may be linked to the formation of ion clusters observed frequently in the evening time in Hyytil (Junninen et al., 2008; Buenrostro Mazon et al., 2016).

3.4 Effects of environmental conditions on sub-3 nm particle concentrations

3.4.1 Sulfuric acid concentration and condensation sink at different sites

To understand the connection between environmental conditions and sub-3 nm particles, we investigated the relation between the daytime median values of sub-3 nm particle concentration and the medians of sulfuric acid concentration and condensation sink. Figure 10 shows the medians of sub-3 nm particle concentration as a function of sulfuric acid concentration (for most sites estimated from a proxy, see Sect. 2.3.10) for different measurement campaigns. At the sites where the median sub-3 nm particle concentrations were highest, i.e., in Nanjing, Shanghai, and San Pietro Capoume, the median sulfuric acid concentrations were also highest (daytime median values 1.53.6 [notdef] 107 cm3). Apart from

this, no clear relation between the medians of sub-3 nm particle concentration and sulfuric acid concentration can be ob-served. Thus, although the formation of sub-3 nm particles is likely favored in the conditions with high sulfuric acid concentrations, other factors seem to also affect sub-3 nm parti-

cle concentrations. The signicance of sulfuric acid for the formation of clusters and small particles has been observed earlier in several studies, including both eld measurements and laboratory studies (e.g., Weber et al., 1997; Kulmala et al., 2004; Erupe et al., 2010; Sipil et al., 2010). On the other hand, recent chamber experiments have shown that particle formation and growth processes are very sensitive to the trace level of base compounds, such as ammonia or amines (Kirkby et al., 2011; Almeida et al., 2013; Lehtipalo et al., 2016), which may explain some of the variability in our data.In addition, the uncertainties in the proxies used for estimating sulfuric acid concentrations may affect the results.

Figure 11 illustrates the relation between the daytime medians of sub-3 nm particle concentration and condensation sink in different measurement campaigns. In Nanjing, Shanghai, and San Pietro Capoume, where the median sub-3 nm particle concentrations (and sulfuric acid concentrations) were highest, high values of condensation sink were also detected (daytime medians 0.010.07 s1). On the other hand, in Centreville condensation sink was also high (daytime median 0.01 s1), but sub-3 nm particle concentration was low, possibly due to low sulfuric acid concentration (daytime median 9.6 [notdef] 104 cm3, see Fig. 10). Therefore, it

seems that the concentration of sub-3 nm particles is determined more by the availability of precursor vapors than by the level of condensation sink. This observation is in agreement with the results of previous studies on sub-3 nm particles (Yu et al., 2014; Rose et al., 2015; Kontkanen et al., 2016).

3.4.2 Correlation between sub-3 nm particle concentrations and environmental variables

The correlation between sub-3 nm particle concentration and different variables was also studied separately for each measurement site. The correlation coefcients obtained at different sites are shown in Table 5 together with their condence intervals at the 95 % condence level. The condence intervals were calculated using Fishers transformation. The autocorrelation, which reduces the effective data size, was also taken into account (Mudelsee, 2010). In addition, in Table 6 correlation coefcients are presented separately for the 1.12 and 23 nm size ranges for Helsinki and Hyytil. Note that for Hyytil only data from 20152016 were used for calculating the correlation coefcients.

Sulfuric acid concentration had a moderate positive correlation with sub-3 nm particle concentration at all sites (Table 5). The correlation coefcient varied between 0.16 and0.48, being lowest in Hyytil and highest in Nanjing. In Hyytil the correlation coefcient depended strongly on the studied size range: the particle concentration in the 1.12 nm size range did not correlate with sulfuric acid (R = 0.02),

but the particle concentration in the size range of 23 nm had a positive correlation (R = 0.38) (Table 6). In Helsinki

no similar difference between these two size ranges was ob-

www.atmos-chem-phys.net/17/2163/2017/ Atmos. Chem. Phys., 17, 21632187, 2017

2178 J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier

relation was observed in Puy de Dme (R = 0.26). Thus,

the relation between condensation sink and sub-3 nm particle concentration seems to vary between different environments. The positive correlation observed in Puy de Dme is likely due to the simultaneous transport of large particles and precursor vapors to the site. In previous studies at high-altitude sites, condensation sink has been observed to be usually higher on NPF event days than on non-event days for the same reason (Boulon et al., 2010; Manninen et al., 2010; Rose et al., 2015). When investigating the correlation with condensation sink separately for the 1.12 and 23 nm size ranges (Table 6), it can be observed that in Hyytil the negative correlation with condensation sink was stronger in the size range of 23 nm (R = 0.29) than in the smaller

size range (R = 0.12). This suggests that condensation sink

may limit the growth of sub-2 nm particles to larger sizes.

Interestingly, at some sites sub-3 nm particle concentration had a positive correlation with ambient temperature (Table 5). The correlation was clear especially in Hyytil (R = 0.54), San Pietro Capoume (R = 0.56), and Shanghai

(R = 0.44). The positive correlation with temperature im-

Atmos. Chem. Phys., 17, 21632187, 2017 www.atmos-chem-phys.net/17/2163/2017/

Figure 9. The median diurnal variation in the ratio of ion concentration to the total particle concentration in four size bins (1.11.3, 1.31.5,1.52, and 23 nm) in Hyytil in 2011 and 20152016. The data are divided into different seasons: spring (spr), summer (sum), autumn (aut), and winter (wint).

served. A moderate positive correlation between sub-3 nm particle concentration and sulfuric acid concentration at different measurement sites has been observed also in previous studies (Kulmala et al., 2013; Yu et al., 2014; Kontkanen et al., 2016). The correlation indicates that sulfuric acid may be one precursor of sub-3 nm particles, but they likely also have other precursors. Furthermore, the fact that in Hyytil particle concentrations in the sub-2 nm size range do not correlate with sulfuric acid concentration suggest that at least in Hyytil the smallest particles (or clusters) may be predominantly formed from vapors other than sulfuric acid. This is in line with the strong seasonal variation in sub-2 nm particles observed at this site, pointing towards the importance of biogenic sources. On the other hand, the uncertainties in the proxies used for estimating sulfuric acid concentrations may also deteriorate the correlations.

A correlation coefcient between sub-3 nm particle concentration and condensation sink was negative at some sites and positive at other sites (Table 5). The strongest negative correlation was observed in Hyytil, Brookhaven, and Centreville (R = 0.20 to 0.34), and the strongest positive cor-

J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier 2179

Table 5. Pearsons correlation coefcients between sub-3 nm particle concentration and other variables at different measurement sites. The condence intervals for the coefcients (the 95 % condence level) are shown in parentheses.

Measurement Sulfuric Condensation Temperature RH Radiation site acid conc. sink

HTL 0.16 0.22 0.54 0.48 0.43

(0.100.21) (0.29 to 0.16) (0.490.59) (0.53 to 0.42) (0.380.48)

SPC 0.43 0.05 0.56 0.51 0.54

(0.280.56) (0.080.18) (0.460.65) (0.60 to 0.40) (0.420.63)

BRH 0.44 0.34 0.29 0.41

(0.330.53) (0.46 to 0.20) (0.330.53) (0.53 to 0.29)

KNT 0.37 0 0.01 0.32 0.46

(0.270.50) (0.150.15) (0.170.15) (0.46 to 0.18) (0.330.58)

CTR 0.31 0.20 0.24 0.22

(0.240.39) (0.28 to 0.12) (0.150.32) (0.30 to 0.13)

PDD 0.37 0.26 0.12 0 0.41(0.180.54) (0.130.38) (0.020.25) (0.130.14) (0.290.51)

SH 0.27 0.03 0.44 0 0.34(0.050.47) (0.120.17) (0.300.56) (0.160.15) (0.210.46)

NJ 0.48 0.16 0.22 0.48 0.55

(0.340.61) (0.310) (0.060.38) (0.60 to 0.33) (0.420.65)

HEL 0.26 0.15 0.05 0.23 0.31

(0.200.31) (0.100.20) (0.110.01) (0.28 to 0.17) (0.260.36)

Table 6. Pearsons correlation coefcients between particle concentration in the size ranges of 1.12 and of 23 nm and other variables in Hyytil (HTL) and Helsinki (HEL) in 20152016. The condence intervals for the coefcients (the 95 % condence level) are shown in parentheses.

Site and Sulfuric Condensation Temperature RH Radiation NO conc. NOx conc. size range acid conc. sink

HTL 1.12 nm 0.02 0.12 0.61 0.40 0.37 0.14 0.45

(0.030.08) (0.18 to 0.06) (0.560.65) (0.45 to 0.34) (0.320.41) (0.17 to 0.11) (0.49 to 0.41) HTL 23 nm 0.38 0.29 0.05 0.44 0.33 0.13 0.05

(0.330.42) (0.33 to 0.25) (0.010.10) (0.48 to 0.41) (0.290.37) (0.100.16) (0.09 to 0.01) HEL 1.12 nm 0.24 0.10 0.09 0.22 0.29 0.51 0.39

(0.180.30) (0.050.15) (0.15 to 0.04) (0.27 to 0.16) (0.250.34) (0.470.54) (0.350.43) HEL 23 nm 0.25 0.03 0.02 0.24 0.26 0.34 0.31

(0.210.29) (0.060.01) (0.060.01) (0.27 to 0.21) (0.230.29) (0.310.37) (0.270.34)

plies that at these sites the formation of sub-3 nm particles may be related to biogenic organic compounds, as their emissions from vegetation usually depend strongly on temperature (Gnther et al., 2012). On the other hand, the positive correlation may also reect the correlation between sub-3 nm particles and solar radiation (discussed below), as temperature and solar radiation generally correlate with each other. A closer look at the correlations in different size ranges shows that in Hyytil the positive correlation existed only in the1.12 nm size range (R = 0.61), whereas the particle con

centration in the 23 nm range did not correlate with temper-

ature (R = 0.05) (Table 6). This indicates that in Hyytil the

smallest, sub-2 nm particles may be formed from organic vapors, which is also consistent with the fact that their concentration does not correlate with sulfuric acid. Recently, Kirkby et al. (2016) showed in their chamber study that aerosol particles can be formed from highly oxidized organic compounds in the absence of sulfuric acid. In addition, the condensation of oxidized organic compounds has been observed to dominate particle formation at a high-altitude Alpine site (Bianchi et al., 2016). On the other hand, in Hyytil sulfuric acid is likely needed for the growth of sub-2 nm particles to larger

www.atmos-chem-phys.net/17/2163/2017/ Atmos. Chem. Phys., 17, 21632187, 2017

2180 J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier

Figure 10. The daytime medians of sub-3 nm particle concentration as a function of sulfuric acid concentration in different measurement campaigns. Sulfuric acid concentration was calculated from a proxy for all other measurement campaigns except those in Kent, Brookhaven, and Centreville and in Hyytil during spring 2011.

sizes, as numerous eld measurements have proven the importance of sulfuric acid in particle formation in the boreal environment (e.g., Sihto et al., 2006; Nieminen et al., 2009; Petj et al., 2009; Kulmala et al., 2013). In Helsinki, sub-3 nm particle concentration did not correlate with air temperature in either of the two size ranges (R = 0.02 to 0.09),

which further strengthens the conclusion that biogenic precursors are likely less important for sub-3 nm particles in this urban environment.

At most sites there was a negative correlation between sub-3 nm particle concentration and RH (Table 5). This was clear in Hyytil (R = 0.48) and San Pietro Capoume

(R = 0.51), which is in agreement with the strong positive

correlation between particle concentration and temperature at these sites. A negative correlation existed also in Helsinki, Nanjing, and the US sites (R = 0.22 to 0.48). There were

no clear differences in correlation coefcients in the 1.12 and 23 nm size ranges in Hyytil and Helsinki (Table 6).

Sub-3 nm particle concentration had a positive correlation with global radiation at all sites. The correlation coefcient ranged from 0.31 obtained in Helsinki to 0.55 in Nanjing (Table 5). In Hyytil and Helsinki, these correlation coefcients did not greatly differ between the 1.12 nm and 23 nm size ranges (Table 6). The positive correlation with radiation suggests the importance of the photochemical production of precursor vapors, and it is consistent with the observed daytime maxima in sub-3 nm particle concentrations (Fig. 4).

Finally, we also investigated the correlation between sub-3 nm particle concentrations and nitrogen oxides (NO and NOx) in Hyytil and Helsinki. In Hyytil there was no clear relation between particle concentrations and NO, but a

Figure 11. The daytime medians of sub-3 nm particle concentration as a function of condensation sink during different measurement campaigns.

negative correlation with NOx was observed (R = 0.45 in

the 1.12.1 nm size range and R = 0.05 in the 23 nm size

range). The negative correlation is likely related to the fact that NOx concentration is high when there is little radiation, and thus oxidation by OH and photodissociation processes are slow (Lyuobuotseva et al., 2006). In addition, high NOx concentrations in Hyytil are often linked to anthropogenic pollution episodes. In contrast, in Helsinki sub-3 nm particle concentration had a positive correlation with NO and NOx.

The correlation was stronger in the size range of 1.12.0 nm (R = 0.51 for NO and R = 0.39 for NOx) than in the size

range of 23 nm (R = 0.34 for NO and R = 0.31 for NOx).

As nitrogen oxides are tracers for trafc emissions, this suggests that the formation of sub-3 nm particles in Helsinki may be linked to the emissions from engines of cars and buses driving near the measurement site. This conclusion is consistent with the observations made of the diurnal cycle of particle concentration in Helsinki (see Sect. 3.2.2). Likewise, it is probable that sub-3 nm particles are formed, at least partly, due to trafc emissions also at other urban sites.

3.5 Connection to NPF events

NPF events are characterized by the appearance of a new mode of small (< 25 nm) particles and their subsequent growth to larger sizes (Dal Maso et al., 2005). The frequency of NPF events observed at different sites is shown in Table 2.The event frequency was highest in San Pietro Capoume (86 % of days) and in Hyytil during spring (40 % of days).In Hyytil, the NPF event frequency was lower in other seasons (1519 % in summer and autumn, 0 % in winter). In Helsinki the event frequency was highest in spring (13 %) but lowest in summer (4 %). In Puy de Dme, Brookhaven, Kent,

Atmos. Chem. Phys., 17, 21632187, 2017 www.atmos-chem-phys.net/17/2163/2017/

J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier 2181

Shanghai, and Nanjing, the event frequency was between 17 and 23 %. In Centreville the event frequency was only 9 %.When studying particle concentrations in different sub-3 nm size bins, it can be observed that in Hyytil the concentration of 23 nm particles and its ratio to the concentration of1.12 nm particles was high in spring when NPF events were frequent (see Sect. 3.1.3). In Helsinki, the link between the concentration of 23 nm particles and the event frequency was not as clear. It should be noted that the lower NPF event frequency in Helsinki compared to Hyytil is likely due to the fact that in Helsinki only the strongest regional NPF events can be observed, due to preexisting aerosol particles, and therefore the majority of days are so-called undened days (Hussein et al., 2005).

All in all, the results indicate that the occurrence of NPF events does not depend solely on the concentration of sub-3 nm particles. This indicates that the formation of sub-3 nm particles and their subsequent growth to larger sizes are two separate processes, as suggested already by Kulmala et al. (2000), and the growth occurs only if conditions are favorable. The favorable conditions may be, for instance, high enough concentrations of condensable precursor vapors (e.g., sulfuric acid and low-volatile organic compounds) and low enough concentrations of preexisting aerosol particles which act as a sink for small particles. As the growth from sub-3 nm sizes to larger particles is generally not observed at night, the photochemical production of condensable vapors is likely needed for the initial growth of particles. In the earlier studies discussing sub-3 nm particle concentrations in Brookhaven and Kent (Yu et al., 2014), Shanghai (Xiao et al., 2015), and Nanjing (Yu et al., 2016), it has been concluded that the increase in sub-3 nm particle concentration observed in the daytime does not always lead to an NPF event. For example, Xiao et al. (2015) found that sub-3 nm particles were able to grow to larger sizes only when aerosol surface area was low and sulfuric acid concentration moderate. In other environments the conditions limiting the growth of particles may be different and determining them is not within the scope of this study. Still, studying the concentration of sub-3 nm particles separately in different size bins seems to be essential to understand the dynamics of sub-3 nm particles and their connection to NPF events.

4 Conclusions

In this study, the concentrations of sub-3 nm particles were investigated at nine sites around the world. The particle concentrations were measured with a PSM, together with a DMPS or SMPS. The concentration of sub-3 nm particles was observed to vary signicantly at each measurement site and between different environments. The highest sub-3 nm particle concentrations were detected at the sites with the strongest anthropogenic inuence, i.e., in Nanjing and Shanghai, China; San Pietro Capoume, Italy; and Helsinki, Finland. Sub-3 nm particle concentrations were lower at a bo-

real forest site in Hyytil, Finland, at a high-altitude site in Puy de Dme, France, and at three sites in the United States (Kent, Brookhaven, and Centreville). This indicates that the formation of sub-3 nm particles is favored in the conditions with high concentrations of low-volatile precursor vapors originating from anthropogenic pollution sources. When studying the diurnal variation in particle concentrations, sub-3 nm particle concentration was observed to be highest during the daytime at all sites. The daytime maxima are likely related to the photochemical production of low-volatile precursor vapors and the emissions of precursor vapors, and possibly also primary particles, from different sources. On the other hand, at most of the sites sub-3 nm particle concentration was relatively high at night, which suggests that sub-3 nm particles can be formed also in the absence of solar radiation.

In Hyytil and Helsinki, measurements allowed us to study sub-3 nm particle concentration separately in four size bins (1.11.3, 1.31.5, 1.52, and 23 nm) in different seasons. In Hyytil, sub-3 nm particle concentration exhibited a clear seasonal cycle with the highest concentrations in summer and spring, and the lowest in winter and autumn. In the smallest size bin the particle concentration was clearly highest in summer, which likely results from strong photochemical reactions and high concentrations of biogenic organic compounds at this time of the year. In the largest size bin the particle concentration was highest in spring, showing that at that time of the year the conditions in Hyytil are most favorable for the growth of particles. In Helsinki, the differences in sub-3 nm particle concentrations between different seasons were less obvious, and high concentrations were ob-served also in winter in all four size bins. Thus, in Helsinki the formation of sub-3 nm particles is likely connected to vapors originating from anthropogenic sources, whereas in Hyytil biogenic sources are probably more important.

In addition to PSM measurements, at some of the sites the measurements with a NAIS were conducted, which enabled us to study the ratio of ion concentration to the total sub-3 nm particle concentration. The ion ratios were observed to be low at sites where the total sub-3 nm particle concentrations were high. In Hyytil the ion ratio was relatively low in summer and spring but high in winter and autumn.In winter and autumn the ion ratio often exceeded unity, which shows that the PSM was not able to detect all sub-3 nm particles. When studying different sub-3 nm size bins in Hyytil, the ion ratio was observed to be highest in the sub-2 nm size bins. All in all, the results imply that neutral particles dominate sub-3 nm particle concentrations in polluted environments and in boreal forest during spring and summer. However, determining the ion ratios more reliably would require more knowledge about the properties of sub-3 nm particles and their activation in the PSM in different conditions. Also, more simultaneous measurements with the PSM and ion spectrometers should be performed in different environments.

www.atmos-chem-phys.net/17/2163/2017/ Atmos. Chem. Phys., 17, 21632187, 2017

2182 J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier

The effect of environmental conditions on sub-3 nm particle concentrations was also investigated. The concentration of sulfuric acid, estimated for most sites from a proxy, was observed to be highest at the sites with high sub-3 nm particle concentration. On the other hand, condensation sink was also highest at these sites, which indicates that the concentration of sub-3 nm particles is determined by the availability of precursor vapors rather than the value of the sink. When studying correlations between particle concentrations and different variables, sub-3 nm particle concentration was observed to have a positive correlation with sulfuric acid concentration and solar radiation. The correlation with condensation sink was positive at some measurement sites and negative at others. In addition, at some sites sub-3 nm particle concentration showed a positive correlation with temperature. This was clear particularly in Hyytil in the sub-2 nm size range, which further suggests the importance of biogenic sources of precursor vapors in boreal forest. On the other hand, in Helsinki sub-3 nm particle concentration correlated with nitrogen oxides, which indicates that sub-3 nm particles ob-served at this site may be linked to trafc emissions.

When studying the connection between sub-3 nm particle concentrations and NPF events, it was concluded that the occurrence of NPF events is not determined only by the concentration of sub-3 nm particles. Thus, the formation of particles and their further growth should be considered as two separate processes. Altogether, our results demonstrate that to better understand the dynamics of sub-3 nm particles, long-term measurements of sub-3 nm particle concentrations, preferably separately in different size bins, are needed. Such measurements should be conducted in different environments and ecosystems, also including the Southern Hemisphere and polar areas. Finally, instrumental development is essential to ensure the reliability of the measured concentrations, especially in the sub-2 nm size range, and to determine the composition of detected particles. In addition, to enable more accurate comparisons between different measurement sites, standardized procedures for the calibration and measurements of sub-3 nm particles should be established.

5 Data availability

Data are available on request from the authors.

Competing interests. The authors declare that they have no conict of interest.

Acknowledgements. This research has received funding from the Academy of Finland Centre of Excellence program (grant no. 1118615 and 272041), the European Research Council (ERC) project ATM-NUCLE (grant no. 227463), the European Unions Horizon 2020 research and innovation programme projects

ACTRIS-2 (grant no. 654109) and nano-CAVa (Marie Sklodowska Curie grant no. 656994), the European Commission under the Framework Programme 7 project PEGASOS (grant no. 265148), and the Maj and Tor Nessling foundation. Shanhu Lee acknowledges funding from National Science Foundation (AGS-1137821; AGS-1241498) and Vijay Kanawade, Janek Uin, and You Yi for the help in collecting the PSM data at the US sites. Huan Yu acknowledges funding from NSFC 41405116, the Jiangsu Province NSF BK20140989, and the Jiangsu Specially Appointed Professor grant. Wei Nie and Aijun Ding acknowledge funding from the National Natural Science Foundation of China (D0512/41675145) and the Jiangsu Collaborative Innovation Center for Climate Change. Lin Wang acknowledges funding from the National Natural Science Foundation of China (grant no. 21222703 and 21561130150) and the Royal Society-Newton Advanced Fellowship (NA140106).

Edited by: I. SalmaReviewed by: three anonymous referees

References

Aalto, P., Hmeri, K., Becker, E.,Weber, R., Salm, J.,Mkel, J.M., Hoell, C., ODowd, C. D., Karlsson, H., Hansson, H.-C., Vkev, M., Koponen, I. K., Buzorius, G., and Kulmala, M.: Physical characterization of aerosol particles during nucleation events, Tellus B, 53, 344358, 2001.

Alam, A., Shi, J. P., and Harrison, R. M.: Observations of new particle formation in urban air, J. Geophys. Res., 108, 4093, doi:http://dx.doi.org/10.1029/2001JD001417

Web End =10.1029/2001JD001417 http://dx.doi.org/10.1029/2001JD001417

Web End = , 2003.

Alanen, J., Saukko, E., Lehtoranta, K., Murtonen, T., Timonen, H.,

Hillamo, R., Karjalainen, P., Kuuluvainen, H., Harra, J., Keskinen, J., and Rnkk, T.: The formation and physical properties of the particle emissions from a natural gas engine, Fuel, 162, 155161, 2015.

Almeida, J., Schobesberger, S., Kurten, A., Ortega, I., Kupiainen-

Mtta, O., Praplan, A., Adamov, A., Amorim, A., Bianchi,F., Breitenlechner, M., David, A., Dommen, J., Donahue, N., Downard, A., Dunne, E., Duplissy, J., Ehrhart, S., Flagan, R., Franchin, A., Guida, R., Hakala, J., Hansel, A., Heinritzi, M., Henschel, H., Jokinen, T., Junninen, H., Kajos, M., Kangasluoma, J., Keskinen, H., Kupc, A., Kurten, T., Kvashin, A., Laaksonen, A., Lehtipalo, K., Leiminger, M., Lepp, J., Loukonen,V., Makhmutov, V., Mathot, S., McGrath, M., Nieminen, T., Olenius, T., Onnela, A., Petj, T., Riccobono, F., Riipinen, I., Rissanen, M., Rondo, L., Ruuskanen, T., Santos, F., Sarnela, N., Schallhart, S., Schnitzhofer, R., Seinfeld, J., Simon, M., Sipil,M., Stozhkov, Y., Stratmann, F., Tome, A., Trostl, J., Tsagkogeorgas, G., Vaattovaara, P., Viisanen, Y., Virtanen, A., Vrtala,A., Wagner, P., Weingartner, E., Wex, H., Williamson, C., Wimmer, D., Ye, P., Yli-Juuti, T., Carslaw, K., Kulmala, M., Curtius, J., Baltensperger, U., Worsnop, D., Vehkamki, H., and Kirkby, J.: Molecular understanding of sulphuric acid-amine particle nucleation in the atmosphere, Nature, 502, 359363, doi:http://dx.doi.org/10.1038/nature12663

Web End =10.1038/nature12663 http://dx.doi.org/10.1038/nature12663

Web End = , 2013.

Arnold, F., Pirjola, L., Rnkk, T., Reichl, U., Schlager, H., Lhde,T., Heikkil, J., and Keskinen, J.: First on-line measurements of sulphuric acid gas in modern heavy duty diesel engine exhaust:

Atmos. Chem. Phys., 17, 21632187, 2017 www.atmos-chem-phys.net/17/2163/2017/

J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier 2183

Implications for nanoparticle formation, Environ. Sci. Technol., 46, 1122711234, 2012.

Bianchi, F., Trstl, J., Junninen, H., Frege, C., Henne, S., Hoyle,C. R., Molteni, U., Herrmann, E., Adamov, A., Bukowiecki, N., Chen, X., Duplissy, J., Gysel, M., Hutterli, M., Kangasluoma, J., Kontkanen, J., Krten, A., Manninen, H. E., Mnch, S., Perkyl,O., Petj, T., Rondo, L., Williamson, C., Weingartner, E., Curtius, C., Worsnop, D. R., Kulmala, M., Dommen, J., and Baltensperger, U.: New particle formation in the free troposphere: A question of chemistry and timing, Science, 352, 11091112, 2016.

Boulon, J., Sellegri, K., Venzac, H., Picard, D., Weingartner, E.,

Wehrle, G., Collaud Coen, M., Btikofer, R., Flckiger, E., Baltensperger, U., and Laj, P.: New particle formation and ultrane charged aerosol climatology at a high altitude site in the Alps (Jungfraujoch, 3580 m a.s.l., Switzerland), Atmos. Chem. Phys., 10, 93339349, doi:http://dx.doi.org/10.5194/acp-10-9333-2010

Web End =10.5194/acp-10-9333-2010 http://dx.doi.org/10.5194/acp-10-9333-2010

Web End = , 2010.Buenrostro Mazon, S., Kontkanen, J., Manninen, H.E., Nieminen,T., Kerminen, V.-K., and Kulmala, M.: A longterm comparison of nighttime cluster events and daytime ion formation in a boreal forest, Boreal Env. Res., 21, 242261, 2016.

Dal Maso, M., Kulmala, M., Riipinen, I., Wagner, R., Hussein, T., Aalto, P. P., and Lehtinen, K. E. J.: Formation and growth of fresh atmospheric aerosols: Eight years of aerosol size distribution data from SMEAR II, Hyytil, Finland, Boreal Environ.Res., 10, 323336, 2005.

Decesari, S., Facchini, M. C., Matta, E., Lettini, F., Mircea, M., Fuzzi, S., Tagliavini, E., and Putaud, J. P.: Chemical features and seasonal variation of ne aerosol water-soluble organic compounds in the Po Valley, Italy, Atmos. Environ., 35, 36913699, 2001.

Ding, A. J., Fu, C. B., Yang, X. Q., Sun, J. N., Zheng, L. F., Xie,Y. N., Herrmann, E., Nie, W., Petj, T., Kerminen, V. M., and Kulmala, M.: Ozone and ne particle in the western Yangtze River Delta: an overview of 1 yr data at the SORPES station, Atmos. Chem. Phys., 13, 58135830, doi:http://dx.doi.org/10.5194/acp-13-5813-2013

Web End =10.5194/acp-13-5813- http://dx.doi.org/10.5194/acp-13-5813-2013

Web End =2013 , 2013.

Ehn, M., Thornton, J. A., Kleist, E., Sipil, M., Junninen, H., Pullinen, I., Springer, M., Rubach, F., Tillmann, R., Lee, B., Lopez-Hilker, F., Andres, S., Acir, I.-H., Rissanen, M., Jokinen, T., Schobesberger, S., Kangasluoma, J., Kontkanen, J., Nieminen,T., Kurtn, T., Nielsen, L. B., Jrgensen, S., Kjaergaard, H. G., Canagaratna, M., Dal Maso, M., Berndt, T., Petj, T., Wahner,A., Kerminen, V., Kulmala, M., Worsnop, D. R., Wildt, J., and Mentel, T. F.: A large source of low-volatility secondary organic aerosol, Nature, 506, 476479, doi:http://dx.doi.org/10.1038/nature13032

Web End =10.1038/nature13032 http://dx.doi.org/10.1038/nature13032

Web End = , 2014.Eisele, F. and Tanner, D.: Measurement of the gas phase concentration of H2SO4 and methane sulfonic acid and estimates of

H2SO4 production and loss in the atmosphere, J. Geophys. Res., 98, 90019010, 1993.

Erupe, M. E., Benson, D. R., Li, J., Young, L.-H., Verheggen,B., Al-Refai, M., Tahboub, O., Cunningham, V., Frimpong, F., Viggiano, A. A., and Lee, S.-H.: Correlation of Aerosol Nucleation Rate with Sulfuric Acid and Ammonia in Kent Ohio: An Atmospheric Observation, J. Geophys. Res., 115, D23216, doi:http://dx.doi.org/10.1029/2005GL023268

Web End =10.1029/2005GL023268 http://dx.doi.org/10.1029/2005GL023268

Web End = , 2010.

Gamero-Gastno, M., and Fernndez de la Mora, J.: A Condensation Nucleus Counter (CNC) sensitive to singly charged subnanometer particles, J. Aerosol Sci., 31, 757772, 2000.

Guenther, A. B., Jiang, X., Heald, C. L., Sakulyanontvittaya, T.,

Duhl, T., Emmons, L. K., and Wang, X.: The Model of Emissions of Gases and Aerosols from Nature version 2.1 (MEGAN2.1): an extended and updated framework for modeling biogenic emissions, Geosci. Model Dev., 5, 14711492, doi:http://dx.doi.org/10.5194/gmd-5-1471-2012

Web End =10.5194/gmd-5- http://dx.doi.org/10.5194/gmd-5-1471-2012

Web End =1471-2012 , 2012.

Hari, P. and Kulmala, M.: Station for measuring ecosystem-atmosphere relations (SMEAR II), Boreal Environ. Res., 10, 315322, 2005.

Herrmann, E., Ding, A. J., Kerminen, V.-M., Petj, T., Yang, X.Q., Sun, J. N., Qi, X. M., Manninen, H., Hakala, J., Nieminen,T., Aalto, P. P., Kulmala, M., and Fu, C. B.: Aerosols and nucleation in eastern China: rst insights from the new SORPES-NJU station, Atmos. Chem. Phys., 14, 21692183, doi:http://dx.doi.org/10.5194/acp-14-2169-2014

Web End =10.5194/acp- http://dx.doi.org/10.5194/acp-14-2169-2014

Web End =14-2169-2014 , 2014.

Hirsikko, A., Nieminen, T., Gagn, S., Lehtipalo, K., Manninen, H.E., Ehn, M., Hrrak, U., Kerminen, V.-M., Laakso, L., McMurry,P. H., Mirme, A., Mirme, S., Petj, T., Tammet, H., Vakkari,V., Vana, M., and Kulmala, M.: Atmospheric ions and nucleation: a review of observations, Atmos. Chem. Phys., 11, 767 798, doi:http://dx.doi.org/10.5194/acp-11-767-2011

Web End =10.5194/acp-11-767-2011 http://dx.doi.org/10.5194/acp-11-767-2011

Web End = , 2011.

Hussein, T., Martikainen, J., Junninen, H., Sogacheva, L., Wagner,R., Dal Maso, M., Riipinen, I., Aalto, P. P., and Kulmala, M.: Observation of regional new particle formation in the urban atmosphere, Tellus B, 60, 509521, 2008.

Iida, K., Stoltzenburg, M. R., and McMurry, P. H.: Effect of working uid on sub-2 nm particle detection with a laminar ow ultra-ne condensation particle counter, Aerosol Sci. Tech., 43, 8190, 2009.

Jayaratne, E. R., Ling, X., and Morawska, L.: Ions in motor vehicle exhaust and their dispersion near busy roads, Atmos. Environ., 44, 36443650, 2010.

Jiang, J., Zhao, J., Chen, M., Eisele, F. L., Scheckman, J., Williams,B. J., Kuang, C., and McMurry, P. H.: First measurements of neutral atmospheric cluster and 12 nm particle number size distributions during nucleation events, Aerosol Sci. Tech., 45, iiv, 2011a.

Jiang, J., Chen, M., Kuang, C., Attoui, M., and McMurry, P.H.: Electrical Mobility Spectrometer Using a Diethylene Glycol Condensation Particle Counter for Measurements of Aerosol Size Distributions Down to 1 nm, Aerosol Sci. Tech., 45, 510 521, 2011b.

Junninen, H., Hulkkonen, M., Riipinen, I., Nieminen, T., Hirsikko,A., Suni, T., Boy, M., Lee, S.-H., Vana, M., Tammet, H., Kerminen, V.-M., and Kulmala M.: Observations on nocturnal growth of atmospheric clusters, Tellus B, 60, 365371, 2008.

Jrvi, L., Junninen, H., Karppinen, A., Hillamo, R., Virkkula, A.,

Mkel, T., Pakkanen, T., and Kulmala, M.: Temporal variations in black carbon concentrations with different time scales in Helsinki during 19962005, Atmos. Chem. Phys., 8, 10171027, doi:http://dx.doi.org/10.5194/acp-8-1017-2008

Web End =10.5194/acp-8-1017-2008 http://dx.doi.org/10.5194/acp-8-1017-2008

Web End = , 2008.

Jrvi, L., Hannuniemi, H., Hussein, T., Junninen, H., Aalto, P. P.,

Hillamo, R., Mkel, T., Keronen, P., Siivola, E., Vesala, T., and Kulmala, M.: The urban measurement station SMEAR III: Continuous monitoring of air pollution and surface-atmosphere interactions in Helsinki, Finland, Boreal Environ. Res., 14, 86109, 2009.

Kangasluoma, J., Junninen, H., Lehtipalo, K., Mikkil, J., Vanhanen, J., Attoui, M., Sipil, M., Worsnop, D., Kulmala, M., and

www.atmos-chem-phys.net/17/2163/2017/ Atmos. Chem. Phys., 17, 21632187, 2017

2184 J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier

Petj, T.: Remarks on ion generation for CPC detection efciency studies in sub-3-nm size range, Aerosol. Sci. Tech., 47, 556563, doi:http://dx.doi.org/10.1080/02786826.2013.773393

Web End =10.1080/02786826.2013.773393 http://dx.doi.org/10.1080/02786826.2013.773393

Web End = , 2013.Kangasluoma, J., Kuang, C., Wimmer, D., Rissanen, M. P., Lehti-palo, K., Ehn, M., Worsnop, D. R., Wang, J., Kulmala, M., and Petj, T.: Sub-3 nm particle size and composition dependent response of a nano-CPC battery, Atmos. Meas. Tech., 7, 689700, doi:http://dx.doi.org/10.5194/amt-7-689-2014

Web End =10.5194/amt-7-689-2014 http://dx.doi.org/10.5194/amt-7-689-2014

Web End = , 2014.

Kangasluoma, J., Attoui, M., Junninen, H., Lehtipalo, K., Samodurov, A., Korhonen, F., Sarnela, N., Schmidt-Ott, A., Worsnop,D., Kulmala, M., and Petj, T.: Sizing of neutral sub 3 nm tungsten oxide clusters using Airmodus Particle Size Magnier, J.Aerosol Sci., 87, 5362, 2015.

Kangasluoma, J., Samodurov, A., Attoui, M., Franchin, A., Junninen, H., Korhonen, F., Kurtn, T., Vehkamki, H., Sipil, M., Lehtipalo, K., Worsnop, D. R, Petj, T., and Kulmala, M.: Heterogeneous nucleation onto ions and neutralized ions: insights into sign-preference, J. Phys. Chem. C, 120, 13, 74447450, 2016a.

Kangasluoma, J., Franchin, A., Duplissy, J., Ahonen, L., Korhonen,F., Attoui, M., Mikkil, J., Lehtipalo, K., Vanhanen, J., Kulmala,M., and Petj, T.: Operation of the Airmodus A11 nano Condensation Nucleus Counter at various inlet pressures and various operation temperatures, and design of a new inlet system, Atmos. Meas. Tech., 9, 29772988, doi:http://dx.doi.org/10.5194/amt-9-2977-2016

Web End =10.5194/amt-9-2977-2016 http://dx.doi.org/10.5194/amt-9-2977-2016

Web End = , 2016b.

Karjalainen, P., Pirjola, L., Heikkil, J., Lhde, T., Tzamkiozis,T., Ntziachristos, L., Keskinen, J., and Rnkk, T.: Exhaust particles of modern gasoline vehicles: a laboratory and an on-road study, Atmos. Environ., 97, 262270, doi:http://dx.doi.org/10.1016/j.atmosenv.2014.08.025

Web End =10.1016/j.atmosenv.2014.08.025 http://dx.doi.org/10.1016/j.atmosenv.2014.08.025

Web End = , 2014.

Karjalainen, P., Timonen, H., Saukko, E., Kuuluvainen, H., Saarikoski, S., Aakko-Saksa, P., Murtonen, T., Bloss, M., Dal Maso, M., Simonen, P., Ahlberg, E., Svenningsson, B., Brune,W. H., Hillamo, R., Keskinen, J., and Rnkk, T.: Time-resolved characterization of primary particle emissions and secondary particle formation from a modern gasoline passenger car, Atmos. Chem. Phys., 16, 8559-8570, doi:http://dx.doi.org/10.5194/acp-16-8559-2016

Web End =10.5194/acp-16-8559- http://dx.doi.org/10.5194/acp-16-8559-2016

Web End =2016 , 2016.

Kazil, J., Stier, P., Zhang, K., Quaas, J., Kinne, S., ODonnell, D., Rast, S., Esch, M., Ferrachat, S., Lohmann, U., and Feichter,J.: Aerosol nucleation and its role for clouds and Earths radiative forcing in the aerosol-climate model ECHAM5-HAM, Atmos. Chem. Phys., 10, 1073310752, doi:http://dx.doi.org/10.5194/acp-10-10733-2010

Web End =10.5194/acp-10- http://dx.doi.org/10.5194/acp-10-10733-2010

Web End =10733-2010 , 2010.

Kirkby, J., Curtius, J., Almeida, J., Dunne, E., Duplissy, J., Ehrhart,S., Franchin, A., Gagn, S., Ickes, L., Krten, A., Kupc, A., Metzger,A., Riccobono, F., Rondo, L., Schobesberger, S., Tsagkogeorgas, G., Wimmer, D., Amorim, A., Bianchi, F., Breitenlechner, M., David, A., Dommen, J., Downard, A., Ehn, M., Flagan,R. C., Haider, S., Hansel, A., Hauser, D., Jud, W., Junninen, H., Kreissl, F., Kvashin, A., Laaksonen, A., Lehtipalo, K., Lima, J., Lovejoy, E. R., Makhutov, V., Mathot, S., Mikkil, J., Minginette,P., Mogo, S., Nieminen, T., Onnela, A., Pereira, A., Petj, T., Schnitzhofer, R., Seinfeld, J. H., Sipil, M., Stozhkov, Y., Strat-mann, F., Tome, A., Vanhanen, J., Viisanen Y., Vrtala, A., Wagner, P. E., Walther, H., Weingartner, E., Wex, H., Winkler, P. M., Carslaw, K. S., Worsnop, D. R., Baltensperger, U., and Kulmala,

M.: The role of sulfuric acid, ammonia and galactic cosmic rays in atmospheric aerosol nucleation, Nature, 476, 429433, 2011.Kirkby, J., Duplissy, J., Sengupta, K., Frege, C., Gordon, H.,

Williamson, C., Heinritzi, M., Simon, M., Yan, C., Almeida, J. and Trstl, J., Nieminen, T., Ortega, I. K., Wagner, R., Adamov,A., Amorim, A., Bernhammer, A.K, Bianchi, F., Breitenlechner,M., Brilke, S., Chen X., Craven, J., Dias, A., Ehrhart, S., Flagan,R. C., Franchin, A., Fuchs, C., Guida, R., Hakala, J., Hoyle, C.

R, Jokinen, T., Junninen, H., Kangasluoma, J., Kim, J., Krapf,M., Krten, A., Laaksonen A., Lehtipalo, K., Makhmutov, V., Mathot, S., Molteni, U., Onnela, A., Perkyl, O., Piel, F., Petj,T., Praplan, A. P., Pringle, K., Rap, A., Richards, N. A. D., Riipinen, I., Rissanen, M. P., Rondo, L., Sarnela, N., Schobesberger,S., Scott, C. E., Seinfeld, J. H, Sipil, M., Steiner, G., Stozhkov,Y., Stratmann, F., Tom, A., Virtanen, A., Vogel, A. L., Wagner,A. C., P. E., Wagner, Weingartner, E., Wimmer, D., Winkler, P.M., Ye, P., Zhang, X., Hansel, A., Dommen, J., Donahue, N. M., Worsnop, D. R., Baltensperger, U., Kulmala, M., Carslaw, K. S., and Curtius, J.: Ion-induced nucleation of pure biogenic particles, Nature, 533, 521526, 2016.

Kontkanen, J., Jrvinen, E., Manninen, H. E., Lehtipalo, K., Kangasluoma, J., Decesari, S., Gobbi, G. P., Laaksonen, A., Petj,T., and Kulmala, M.: High concentrations of sub-3 nm clusters and frequent new particle formation observed in the Po Valley, Italy, during the PEGASOS 2012 campaign, Atmos. Chem.Phys., 16, 19191935, doi:http://dx.doi.org/10.5194/acp-16-1919-2016

Web End =10.5194/acp-16-1919-2016 http://dx.doi.org/10.5194/acp-16-1919-2016

Web End = , 2016.Kuang, C., Chen, M., McMurry, P. H., and Wang, J.: Modication of laminar ow ultrane condensation particle counters for the enhanced detection of 1 nm condensation nuclei, Aerosol Sci.Tech., 46, 309315, doi:http://dx.doi.org/10.1080/02786826.2011.626815

Web End =10.1080/02786826.2011.626815 http://dx.doi.org/10.1080/02786826.2011.626815

Web End = , 2012.Kulkarni, P., Baron, P. A., and Willeke, K.: Aerosol Measurement:

Principles, Techniques, and Applications, Third Edition, John Wiley & Sons, 2011.

Kulmala, M. and Kerminen, V.-M.: On the formation and growth of atmospheric nanoparticles, Atmos. Res., 90, 132150, doi:http://dx.doi.org/10.1016/j.atmosres.2008.01.005

Web End =10.1016/j.atmosres.2008.01.005 http://dx.doi.org/10.1016/j.atmosres.2008.01.005

Web End = , 2008.

Kulmala, M., Pirjola, L., and Mkel, J. M.: Stable sulphate clusters as a source of new atmospheric particles, Nature, 404, 6669, 2000.

Kulmala, M., Maso, M. D., Mkel, J. M., Pirjola, L., Vkev, M.,

Aalto, P., Miikkulainen, P., Hmeri, K., and ODowd, C. D.: On the formation, growth and composition of nucleation mode particles, Tellus B, 53, 479490, 2001.

Kulmala, M., Vehkamki, H., Petj, T., Dal Maso, M., Lauri, A.,

Kerminen, V.-M., Birmili, W., and McMurry P. H. Formation and growth rates of ultrane atmospheric particles: A review of observations, J. Aerosol Sci., 35, 143176, 2004.

Kulmala, M., Petj, T., Mnkknen, P., Koponen, I. K., Dal Maso,M., Aalto, P. P., Lehtinen, K. E. J., and Kerminen, V.-M.: On the growth of nucleation mode particles: source rates of condensable vapor in polluted and clean environments, Atmos. Chem. Phys., 5, 409416, doi:http://dx.doi.org/10.5194/acp-5-409-2005

Web End =10.5194/acp-5-409-2005 http://dx.doi.org/10.5194/acp-5-409-2005

Web End = , 2005.

Kulmala, M., Riipinen, I., Sipila, M., Manninen, H. E., Petaja, T.,

Junninen, H., dal Maso, M., Mordas, G., Mirme, A., Vana, M., Hirsikko, A., Laasko, L., Harrison, R. M., Hanson, I., Leung, C., Lehtinen, K. E. J., Kerminen, V.-M.: Toward direct measurement of atmospheric nucleation, Science, 318, 8992, 2007.Kulmala, M., Petj, T., Nieminen, T., Sipil, M., Manninen,H. E., Lehtipalo, K., Dal Maso, M., Aalto, P. P., Junninen,

Atmos. Chem. Phys., 17, 21632187, 2017 www.atmos-chem-phys.net/17/2163/2017/

J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier 2185

H., Paasonen, P., Riipinen, I., Lehtinen, K. E. J., Laaksonen, A., and Kerminen, V.-M.: Measurement of the nucleation of atmospheric aerosol particles, Nat. Protoc., 7, 16511667, doi:http://dx.doi.org/10.1038/nprot.2012.091

Web End =10.1038/nprot.2012.091 http://dx.doi.org/10.1038/nprot.2012.091

Web End = , 2012.

Kulmala, M., Kontkanen, J., Junninen, H., Lehtipalo, K., Manninen, H. E., Nieminen, T., Petj, T., Sipil, M., Schobesberger,S., Rantala, P., Franchin,A., Jokinen,T., Jrvinen,E., ijl ,M., Kangasluoma, J., Hakala, J., Aalto, P. P., Paasonen, P., Mikkil,J., Vanhanen, J., Aalto, J., Hakola, H., Makkonen, U., Ruuskanen, T., Mauldin, R. L., Duplissy, J., Vehkamki, H., Bck,J., Kortelainen, A., Riipinen, I., Kurtn, T., Johnston, M. V, Smith, J. N., Ehn, M., Mentel, T. F., Lehtinen, K. E. J., Laaksonen, A., Kerminen, V.-M., and Worsnop, D. R.: Direct observations of atmospheric aerosol nucleation, Science, 339, 943946, doi:http://dx.doi.org/10.1126/science.1227385

Web End =10.1126/science.1227385 http://dx.doi.org/10.1126/science.1227385

Web End = , 2013.

Kulmala, M., Petj, T., Ehn, M., Thornton, J., Sipil, M., Worsnop,D. R., and Kerminen, V.-M.: Chemistry of atmospheric nucleation: on the recent advances on precursor characterization and atmospheric cluster composition in connection with atmospheric new particle formation, Annu. Rev. Phys. Chem., 65, 2137, 2014.

Laaksonen, A., Hamed, A., Joutsensaari, J., Hiltunen, L., Cavalli,F., Junkermann, W., Asmi, A., Fuzzi, S., and Facchini, M. C.: Cloud condensation nucleus production from nucleation events at a highly polluted region, Geophys. Res. Lett., 32, L06812, doi:http://dx.doi.org/10.1029/2004GL022092

Web End =10.1029/2004GL022092 http://dx.doi.org/10.1029/2004GL022092

Web End = , 2005.

Lehtipalo, K., Sipil, M., Riipinen, I., Nieminen, T., and Kulmala,M.: Analysis of atmospheric neutral and charged molecular clusters in boreal forest using pulse-height CPC, Atmos. Chem.Phys., 9, 41774184, doi:http://dx.doi.org/10.5194/acp-9-4177-2009

Web End =10.5194/acp-9-4177-2009 http://dx.doi.org/10.5194/acp-9-4177-2009

Web End = , 2009.Lehtipalo, K., Kulmala, M., Sipil, M., Petj, T., Vana, M., Ceburnis, D., Dupuy, R., and ODowd, C.: Nanoparticles in boreal forest and coastal environment: a comparison of observations and implications of the nucleation mechanism, Atmos. Chem. Phys., 10, 70097016, doi:http://dx.doi.org/10.5194/acp-10-7009-2010

Web End =10.5194/acp-10-7009-2010 http://dx.doi.org/10.5194/acp-10-7009-2010

Web End = , 2010.Lehtipalo, K., Sipil, M., Junninen, H., Ehn, M., Berndt, T., Kajos,M. K., Worsnop, D. R., Petj, T., and Kulmala, M.: Observations of nano-CN in the nocturnal boreal forest, Aerosol Sci. Tech., 45, 499509, 2011.

Lehtipalo, K., Lepp, J., Kontkanen, J., Kangasluoma, J., Franchin,A., Wimmer, D., Schobesberger, S., Junninen, H., Petj, T., Sipil, M., Mikkil, J., Vanhanen, J., Worsnop, D. R., and Kulmala, M.: Methods for determining particle size distribution and growth rates between 13 nm using the Particle Size Magnier, Boreal Environ. Res., 19, 215236, 2014.

Lehtipalo, K., Rondo, L., Kontkanen, J., Schobesberger, S., Jokinen,T., Sarnela, N., Krten, A., Ehrhart, S., Franchin, A., Nieminen,T., Riccobono, F., Sipil, M., Yli-Juuti, T., Duplissy, J., Adamov,A., Ahlm, L., Almeida, J., Amorim, A., Bianchi, F., Breitenlechner, M., Dommen, J., Downard, A. J., Dunne, E. M., Flagan, R.C., Guida, R., Hakala, J., Hansel, A., Jud, W., Kangasluoma, J., Kerminen, V.-M., Keskinen, H., Kim, J., Kirkby, J., Kupc, A., Kupiainen-Mtt, O., Laaksonen, A., Lawler, M. J., Leiminger,M., Mathot, S., Olenius, T., Ortega, I. K., Onnela, A., Petj,T., Praplan, A., Rissanen, M. P., Ruuskanen, T., Santos, F. D., Schallhart, S., Schnitzhofer, R., Simon, M., Smith, J. N., Trstl,J., Tsagkogeorgas, G., Tom, A., Vaattovaara, P., Vehkamki,H., Vrtala, A. E., Wagner, P. E., Williamson, C., Wimmer, D., Winkler, P. M., Virtanen, A., Donahue, N. M., Carslaw, K. S.,

Baltensperger, U., Riipinen, I., Curtius, J., Worsnop, D. R., and Kulmala, M.: The effect of acid-base clustering and ions on the growth of atmospheric nano-particles, Nat. Commun., 7, 11594, doi:http://dx.doi.org/10.1038/ncomms11594

Web End =10.1038/ncomms11594 http://dx.doi.org/10.1038/ncomms11594

Web End = , 2016.

Leino, K., Nieminen, T., Manninen, H. E., Petj, T., Kerminen, V.-M., and Kulmala, M: Intermediate ions as a strong indicator for new particle formation bursts in boreal forest, Boreal Env. Res., 21, 274286, 2016.

Lyubovtseva, Y. S., Sogacheva, L., Dal Maso, M., Bonn, B., Keronen, P., and Kulmala, M.: Seasonal variations of trace gases, meteorological parameters, and formation of aerosols in boreal forest, Bor. Environ. Res., 10, 493510, 2005.

Makkonen, R., Asmi, A., Kerminen, V.-M., Boy, M., Arneth, A.,

Hari, P., and Kulmala, M.: Air pollution control and decreasing new particle formation lead to strong climate warming, Atmos. Chem. Phys., 12, 15151524, doi:http://dx.doi.org/10.5194/acp-12-1515-2012

Web End =10.5194/acp-12-1515- http://dx.doi.org/10.5194/acp-12-1515-2012

Web End =2012 , 2012.

Manninen, H. E., Petj, T., Asmi, E., Riipinen, I., Nieminen, T.,

Mikkil, J., Hrrak, U., Mirme, A., Mirme, S., Laakso, L., Kerminen, V.-M., and Kulmala, M.: Long-term eld measurements of charged and neutral clusters using Neutral cluster and Air Ion Spectrometer (NAIS), Boreal Env. Res., 14, 591605, 2009.Manninen, H. E., Nieminen, T., Asmi, E., Gagn, S., Hkkinen,S., Lehtipalo, K., Aalto, P., Vana, M., Mirme, A., Mirme, S., Hrrak, U., Plass-Dlmer, C., Stange, G., Kiss, G., Hoffer, A., Tro, N., Moerman, M., Henzing, B., de Leeuw, G., Brinkenberg, M., Kouvarakis, G. N., Bougiatioti, A., Mihalopoulos, N., ODowd, C., Ceburnis, D., Arneth, A., Svenningsson, B., Swietlicki, E., Tarozzi, L., Decesari, S., Facchini, M. C., Birmili, W., Sonntag, A., Wiedensohler, A., Boulon, J., Sellegri, K., Laj, P., Gysel, M., Bukowiecki, N., Weingartner, E., Wehrle, G., Laaksonen, A., Hamed, A., Joutsensaari, J., Petj, T., Kerminen, V.-M., and Kulmala, M.: EUCAARI ion spectrometer measurements at12 European sites analysis of new particle formation events, Atmos. Chem. Phys., 10, 79077927, doi:http://dx.doi.org/10.5194/acp-10-7907-2010

Web End =10.5194/acp-10-7907- http://dx.doi.org/10.5194/acp-10-7907-2010

Web End =2010 , 2010.

Manninen, H. E., Franchin, A., Schobesberger, S., Hirsikko, A.,

Hakala, J., Skromulis, A., Kangasluoma, J., Ehn, M., Junninen,H., Mirme, A., Mirme, S., Sipil, M., Petj, T., Worsnop, D.R., and Kulmala, M.: Characterisation of corona-generated ions used in a Neutral cluster and Air Ion Spectrometer (NAIS), Atmos. Meas. Tech., 4, 27672776, doi:http://dx.doi.org/10.5194/amt-4-2767-2011

Web End =10.5194/amt-4-2767-2011 http://dx.doi.org/10.5194/amt-4-2767-2011

Web End = , 2011.

Manninen, H. E., Mirme, S., Mirme, A., Petj, T., and Kulmala,M.: How to reliably detect molecular clusters and nucleation mode particles with Neutral cluster and Air Ion Spectrometer (NAIS), Atmos. Meas. Tech., 9, 35773605, doi:http://dx.doi.org/10.5194/amt-9-3577-2016

Web End =10.5194/amt- http://dx.doi.org/10.5194/amt-9-3577-2016

Web End =9-3577-2016 , 2016.

McMurry, P. H.: The history of CPCs, Aerosol Sci. Tech., 33, 297

322, 2000.

McMurry, P. H., Kulmala, M., Worsnop, D. R.: Special issue on

Aerosol Measurements in the 1 nm Range, Aerosol Sci. Tech., 45, p. i, 2011.

Merikanto, J., Spracklen, D. V., Mann, G. W., Pickering, S. J., and Carslaw, K. S.: Impact of nucleation on global CCN, Atmos. Chem. Phys., 9, 86018616, doi:http://dx.doi.org/10.5194/acp-9-8601-2009

Web End =10.5194/acp-9-8601-2009 http://dx.doi.org/10.5194/acp-9-8601-2009

Web End = , 2009.

Mikkonen, S., Romakkaniemi, S., Smith, J. N., Korhonen, H.,

Petj, T., Plass-Duelmer, C., Boy, M., McMurry, P. H., Lehti-

www.atmos-chem-phys.net/17/2163/2017/ Atmos. Chem. Phys., 17, 21632187, 2017

2186 J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier

nen, K. E. J., Joutsensaari, J., Hamed, A., Mauldin III, R. L., Birmili, W., Spindler, G., Arnold, F., Kulmala, M., and Laaksonen, A.: A statistical proxy for sulphuric acid concentration, Atmos. Chem. Phys., 11, 1131911334, doi:http://dx.doi.org/10.5194/acp-11-11319-2011

Web End =10.5194/acp-11- http://dx.doi.org/10.5194/acp-11-11319-2011

Web End =11319-2011 , 2011.

Mirme, A., Tamm, E., Mordas, G., Vana, M., Uin, J., Mirme, S., Bernotas, T., Laakso, L., Hirsikko, A., and Kulmala, M.: A widerange multi-channel Air Ion Spectrometer, Boreal Environ.Res., 12, 247264, 2007.

Mirme, S. and Mirme, A.: The mathematical principles and design of the NAIS a spectrometer for the measurement of cluster ion and nanometer aerosol size distributions, Atmos. Meas. Tech., 6, 10611071, doi:http://dx.doi.org/10.5194/amt-6-1061-2013

Web End =10.5194/amt-6-1061-2013 http://dx.doi.org/10.5194/amt-6-1061-2013

Web End = , 2013.

Mordas, G., Manninen, H. E., Petj T., Aalto, P. P., Hmeri, K., and Kulmala, M.: On operation of the ultra-ne water-based CPC TSI 3786 and comparison with other TSI models (TSI 3776, TSI 3772, TSI 3025, TSI 3010, TSI 3007), Aerosol Sci. Tech., 42, 152158, 2008.

Mudelsee, M: Climate Time Series Analysis: Classical Statistical and Bootstrap Methods, Springer, Dordrecht, the Netherlands, 2010.

Nieminen, T., Manninen, H. E., Sihto, S.-L., Yli-Juuti, T., Mauldin III, R. L., Petj, T., Riipinen, I., Kerminen, V.-M., and Kulmala,M.: Connection of sulfuric acid to atmospheric nucleation in bo-real forest, Environ. Sci. Technol., 43, 47154721, 2009.Petj, T., Mauldin, III, R. L., Kosciuch, E., McGrath, J., Nieminen, T., Paasonen, P., Boy, M., Adamov, A., Kotiaho, T., and Kulmala, M.: Sulfuric acid and OH concentrations in a boreal forest site, Atmos. Chem. Phys., 9, 74357448, doi:http://dx.doi.org/10.5194/acp-9-7435-2009

Web End =10.5194/acp- http://dx.doi.org/10.5194/acp-9-7435-2009

Web End =9-7435-2009 , 2009.

Qi, X. M., Ding, A. J., Nie, W., Petj, T., Kerminen, V.-M., Herrmann, E., Xie, Y. N., Zheng, L. F., Manninen, H., Aalto, P., Sun,J. N., Xu, Z. N., Chi, X. G., Huang, X., Boy, M., Virkkula, A., Yang, X. Q., Fu, C. B., and Kulmala, M.: Aerosol size distribution and new particle formation in the western Yangtze River Delta of China: 2 years of measurements at the SORPES station, Atmos. Chem. Phys., 15, 1244512464, doi:http://dx.doi.org/10.5194/acp-15-12445-2015

Web End =10.5194/acp- http://dx.doi.org/10.5194/acp-15-12445-2015

Web End =15-12445-2015 , 2015Rantala, P., Aalto, J., Taipale, R., Ruuskanen, T. M., and Rinne, J.:

Annual cycle of volatile organic compound exchange between a boreal pine forest and the atmosphere, Biogeosciences, 12, 57535770, doi:http://dx.doi.org/10.5194/bg-12-5753-2015

Web End =10.5194/bg-12-5753-2015 http://dx.doi.org/10.5194/bg-12-5753-2015

Web End = , 2015.

Rinne, J., Bck, J., and Hakola, H.: Biogenic volatile organic compound emissions from the Eurasian taiga: current knowledge and future directions, Boreal Environ. Res., 14, 807826, 2009.Rose, C., Sellegri, K., Asmi, E., Hervo, M., Freney, E., Colomb, A.,

Junninen, H., Duplissy, J., Sipil, M., Kontkanen, J., Lehti-palo, K., and Kulmala, M.: Major contribution of neutral clusters to new particle formation at the interface between the boundary layer and the free troposphere, Atmos. Chem. Phys., 15, 3413 3428, doi:http://dx.doi.org/10.5194/acp-15-3413-2015

Web End =10.5194/acp-15-3413-2015 http://dx.doi.org/10.5194/acp-15-3413-2015

Web End = , 2015.

Sarnela, N., Jokinen, T., Nieminen, T., Lehtipalo, K., Junninen, H., Kangasluoma, J., Hakala, J., Taipale R., Schobesberger, S., Sipil, M., Larnimaa, K., Westerholm, H., Heijari, J., Kerminen, V.-M., Petj, T., and Kulmala, M.: Sulphuric acid and aerosol particle production in the vicinity of an oil renery, Atmos. Environ., 119, 156166, 2015.

Schobesberger, S., Junninen, H., Bianchi, F., Lonn, G., Ehn, M., Lehtipalo, K., Dommen, J., Ehrhart, S., Ortega, I. K., Franchin,

A., Nieminen, T., Riccobono, F., Hutterli, M., Duplissy, J., Almeida, J., Amorim, A., Breitenlechner, M., Downard, A. J., Dunne, E. M., Flagan, R. C., Kajos, M., Keskinen, H., Kirkby,J., Kupc, A., Kuerten, A., Kurten, T., Laaksonen, A., Mathot, S., Onnela, A., Praplan, A. P., Rondo, L., Santos, F. D., Schallhart,S., Schnitzhofer, R., Sipila, M., Tome, A., Tsagkogeorgas, G., Vehkamaki, H., Wimmer, D., Baltensperger, U., Carslaw, K. S., Curtius, J., Hansel, A., Petaja, T., Kulmala, M., Donahue, N. M., and Worsnop, D. R.: Molecular understanding of atmospheric particle formation from sulfuric acid and large oxidized organic molecules, P. Natl. Acad. Sci. USA, 110, 1722317228, 2013. Sihto, S.-L., Kulmala, M., Kerminen, V.-M., Dal Maso, M., Petj,T., Riipinen, I., Korhonen, H., Arnold, F., Janson, R., Boy, M., Laaksonen, A., and Lehtinen, K. E. J.: Atmospheric sulphuric acid and aerosol formation: implications from atmospheric measurements for nucleation and early growth mechanisms, Atmos. Chem. Phys., 6, 40794091, doi:http://dx.doi.org/10.5194/acp-6-4079-2006

Web End =10.5194/acp-6-4079-2006 http://dx.doi.org/10.5194/acp-6-4079-2006

Web End = , 2006.

Sipil, M., Lehtipalo, K., Kulmala, M., Petj, T., Junninen, H., Aalto, P. P., Manninen, H. E., Kyr, E.-M., Asmi, E., Riipinen, I., Curtius, J., Krten, A., Borrmann, S., and ODowd,C. D.: Applicability of condensation particle counters to measure atmospheric clusters, Atmos. Chem. Phys., 8, 40494060, doi:http://dx.doi.org/10.5194/acp-8-4049-2008

Web End =10.5194/acp-8-4049-2008 http://dx.doi.org/10.5194/acp-8-4049-2008

Web End = , 2008.

Sipil, M., Lehtipalo, K., Attoui, M., Neitola, K., Petj, T., Aalto,P. P., ODowd, C. D., and Kulmala, M.: Laboratory verication of PH-CPCs ability to monitor atmospheric sub-3 nm clusters, Aerosol Sci. Tech., 43, 2, 126135, 2009.

Sipil, M., Berndt, T., Petj, T., Brus, D., Vanhanen, J., Stratmann,F., Patokoski, J., Mauldin, R. L., Hyvrinen, A.-P., Lihavainen,H., and Kulmala, M.: The role of sulfuric acid in atmospheric nucleation, Science, 327, 12431246, 2010.

Spracklen, D. V., Carslaw, K. S., Kulmala, M., Kerminen, V.-M., Mann, G. W., and Sihto, S.-L.: The contribution of boundary layer nucleation events to total particle concentrations on regional and global scales, Atmos. Chem. Phys., 6, 56315648, doi:http://dx.doi.org/10.5194/acp-6-5631-2006

Web End =10.5194/acp-6-5631-2006 http://dx.doi.org/10.5194/acp-6-5631-2006

Web End = , 2006.

Trstl, J., Chuang, W. K., Gordon, H., Heinritzi, M., Yan, C., Molteni, U., Ahlm, L., Frege, C., Bianchi, F., Wagner, R. and Simon, M., Lehtipalo, K., Williamson, C., Craven, J. S., Duplissy,J., Adamov, A., Almeida, J., Bernhammer, A.-K., Breitenlechner, M., Brilke, S., Dias, A., Ehrhart, S., Flagan, R. C., Franchin,A., Fuchs, C., Guida, R., Gysel, M., Hansel, A., Hoyle, C. R., Jokinen, T., Junninen, H., Kangasluoma, J., Keskinen, H., Kim,J., Krapf, M., Krten, A., Laaksonen, A., Lawler, M., Leiminger,M., Mathot, S., Mhler, O., Nieminen, T., Onnela, A., Petj,T., Piel, F. M., Miettinen, P., Rissanen, M. P., Rondo, L., Sarnela, N., Schobesberger, S., Sengupta, K., Sipil, M., Smith, J.N., Steiner, G., Tom, A., Virtanen, A., Wagner, A. C., Weingartner, E., Wimmer, D., Winkler, P. M., Ye, P., Carslaw, K. S., Curtius, J., Dommen, J., Kirkby, J., Kulmala, M., Riipinen, I., Worsnop, D. R., Donahue, N. M., and Baltensperger, U.: The role of low-volatility organic compounds in initial particle growth in the atmosphere, Nature, 533, 527531, 2016.

Ude, S., and Fernndez de la Mora, J.: Molecular monodisperse mobility and mass standards from electrosprays of tetra-alkyl ammonium halides, J. Aerosol Sci., 36, 12241237, 2005.

Atmos. Chem. Phys., 17, 21632187, 2017 www.atmos-chem-phys.net/17/2163/2017/

J. Kontkanen et al.: Measurements of sub-3 nm particles using a particle size magnier 2187

Wang, M. and Penner, J. E.: Aerosol indirect forcing in a global model with particle nucleation, Atmos. Chem. Phys., 9, 239260, doi:http://dx.doi.org/10.5194/acp-9-239-2009

Web End =10.5194/acp-9-239-2009 http://dx.doi.org/10.5194/acp-9-239-2009

Web End = , 2009.

Vanhanen, J., Mikkil, J., Lehtipalo, K., Sipil, M., Manninen, H.E., Siivola, E., Petj, T., and Kulmala, M.: Particle size magnier for nano-CN detection, Aerosol Sci. Tech., 45, 533542, 2011.

Venzac, H., Sellegri, K., Villani, P., Picard, D., and Laj, P.: Seasonal variation of aerosol size distributions in the free troposphere and residual layer at the Puy de Dme station, France, Atmos. Chem.Phys., 9, 14651478, doi:http://dx.doi.org/10.5194/acp-9-1465-2009

Web End =10.5194/acp-9-1465-2009 http://dx.doi.org/10.5194/acp-9-1465-2009

Web End = , 2009.Wagner, R., Manninen, H. E., Franchin, A., Lehtipalo, K., Mirme,S., Steiner, G., Petj, T., and Kulmala, M.: On the accuracy of ion measurements using a Neutral cluster and Air Ion Spectrometer, Boreal Env. Res. 21, 230241, 2016.

Weber, R. J., Marti, J. J., McMurry, P. H., Eisele, F. L., Tanner, D.J., and Jefferson, A.: Measurements of new particle formation and ultrane particle growth rates at a clean continental site, J.Geophys. Res., 102, 43754385, 1997.

Wiedensohler, A., Birmili, W., Nowak, A., Sonntag, A., Weinhold,K., Merkel, M., Wehner, B., Tuch, T., Pfeifer, S., Fiebig, M., Fjraa, A. M., Asmi, E., Sellegri, K., Depuy, R., Venzac, H., Villani, P., Laj, P., Aalto, P., Ogren, J. A., Swietlicki, E., Williams,P., Roldin, P., Quincey, P., Hglin, C., Fierz-Schmidhauser, R., Gysel, M., Weingartner, E., Riccobono, F., Santos, S., Grning,C., Faloon, K., Beddows, D., Harrison, R., Monahan, C., Jennings, S. G., ODowd, C. D., Marinoni, A., Horn, H. G., Keck,L., Jiang, J., Scheckman, J., McMurry, P. H., Deng, Z., Zhao, C.S., Moerman, M., Henzing, B., de Leeuw, G., Lschau, G., and Bastian, S.: Mobility particle size spectrometers: harmonization of technical standards and data structure to facilitate high quality long-term observations of atmospheric particle number size distributions, Atmos. Meas. Tech., 5, 657685, doi:http://dx.doi.org/10.5194/amt-5-657-2012

Web End =10.5194/amt- http://dx.doi.org/10.5194/amt-5-657-2012

Web End =5-657-2012 , 2012.

Wimmer, D., Lehtipalo, K., Franchin, A., Kangasluoma, J., Kreissl,F., Krten, A., Kupc, A., Metzger, A., Mikkil, J., Petj,T., Riccobono, F., Vanhanen, J., Kulmala, M., and Curtius,J.: Performance of diethylene glycol-based particle counters in the sub-3 nm size range, Atmos. Meas. Tech., 6, 17931804, doi:http://dx.doi.org/10.5194/amt-6-1793-2013

Web End =10.5194/amt-6-1793-2013 http://dx.doi.org/10.5194/amt-6-1793-2013

Web End = , 2013.

Wimmer, D., Lehtipalo, K., Nieminen, T., Duplissy, J., Ehrhart, S., Almeida, J., Rondo, L., Franchin, A., Kreissl, F., Bianchi, F., Manninen, H. E., Kulmala, M., Curtius, J., and Petj, T.: Technical Note: Using DEG-CPCs at upper tropospheric temperatures, Atmos. Chem. Phys., 15, 75477555, doi:http://dx.doi.org/10.5194/acp-15-7547-2015

Web End =10.5194/acp-15-7547- http://dx.doi.org/10.5194/acp-15-7547-2015

Web End =2015 , 2015.

Winkler, P. M., Steiner, G., Vrtala, A., Vehkamki, H., Noppel,M., Lehtinen, K. E. J., Reischl, G. P., Wagner, P. E., and Kulmala, M.: Heterogeneous nucleation experiments bridging the scale from molecular ion clusters to nanoparticles, Science, 319, 13741377, doi:http://dx.doi.org/10.1126/science.1149034

Web End =10.1126/science.1149034 http://dx.doi.org/10.1126/science.1149034

Web End = , 2008.

Xiao, S., Wang, M. Y., Yao, L., Kulmala, M., Zhou, B., Yang, X., Chen, J. M., Wang, D. F., Fu, Q. Y., Worsnop, D. R., and Wang, L.: Strong atmospheric new particle formation in winter in urban Shanghai, China, Atmos. Chem. Phys., 15, 17691781, doi:http://dx.doi.org/10.5194/acp-15-1769-2015

Web End =10.5194/acp-15-1769-2015 http://dx.doi.org/10.5194/acp-15-1769-2015

Web End = , 2015.

Xu, L., Guo, H., Boyd, C. M., Klein, M., Bougiatioti, A., Cerully,K. M., Hite, J. R., Isaacman-VanWertz, G., Kreisberg, N. M., Knote, C., Olson, K., Koss, A., Goldstein, A. H., Hering, S.V., de Gouw, J., Baumann, K., Lee, S., Nenes, A., Weber,R. J., and Ng, N. L.: Effects of anthropogenic emissions on aerosol formation from isoprene and monoterpenes in the southeastern United States, P. Natl. Acad. Sci. USA, 112, 3742, doi:http://dx.doi.org/10.1073/pnas.1417609112

Web End =10.1073/pnas.1417609112 http://dx.doi.org/10.1073/pnas.1417609112

Web End = , 2015.

You, Y., Kanawade, V. P., de Gouw, J. A., Guenther, A. B., Madronich, S., Sierra-Hernndez, M. R., Lawler, M., Smith, J.N., Takahama, S., Ruggeri, G., Koss, A., Olson, K., Baumann,K., Weber, R. J., Nenes, A., Guo, H., Edgerton, E. S., Porcelli,L., Brune, W. H., Goldstein, A. H., and Lee, S.-H.: Atmospheric amines and ammonia measured with a chemical ionization mass spectrometer (CIMS), Atmos. Chem. Phys., 14, 1218112194, doi:http://dx.doi.org/10.5194/acp-14-12181-2014

Web End =10.5194/acp-14-12181-2014 http://dx.doi.org/10.5194/acp-14-12181-2014

Web End = , 2014.

Yu, F. and Turco R.: Ultrane aerosol formation via ion-mediated nucleation, Geophys. Res. Lett., 27, 883886, 2000.

Yu, F. and Turco, R.: Case studies of particle formation events observed in boreal forests: implications for nucleation mechanisms, Atmos. Chem. Phys., 8, 60856102, doi:http://dx.doi.org/10.5194/acp-8-6085-2008

Web End =10.5194/acp-8- http://dx.doi.org/10.5194/acp-8-6085-2008

Web End =6085-2008 , 2008.

Yu, F., Luo, G., Bates, T. S., Anderson, B., Clarke, A., Kapustin, V., Yantosca, R. M., Wang, Y., and Wu, S.: Spatial distributions of particle number concentrations in the global troposphere: Simulations, observations, and implications for nucleation mechanisms, J. Geophys. Res., 115, D17205, doi:http://dx.doi.org/10.1029/2009JD013473

Web End =10.1029/2009JD013473 http://dx.doi.org/10.1029/2009JD013473

Web End = , 2010.

Yu, H., Kanawade, V. P., You, Y., Hallar, A. G., Mccubbin,I. B., Chirokova, G., Sedlacek, A. J., Springston, S. R., Wang, J., Mc-Graw, R. L., Mikkila, J., and Lee, S. H.: Sub-3 nm particles observed at the coastal and continental sites in the United States, J. Geophys. Res. Atmos., 119, 860879, doi:http://dx.doi.org/10.1002/2013JD020841

Web End =10.1002/2013JD020841 http://dx.doi.org/10.1002/2013JD020841

Web End = , 2014.

Yu, H., Zhou, L., Dai, L., Shen, W., Dai, W., Zheng, J., Ma, Y., and Chen, M.: Nucleation and growth of sub-3 nm particles in the polluted urban atmosphere of a megacity in China, Atmos. Chem. Phys., 16, 26412657, doi:http://dx.doi.org/10.5194/acp-16-2641-2016

Web End =10.5194/acp-16-2641-2016 http://dx.doi.org/10.5194/acp-16-2641-2016

Web End = , 2016.

Zhang, R., Khalizov, A., Wang, L., Hu, M., and Xu, W.: Nucleation and Growth of Nanoparticles in the Atmosphere, Chem. Rev., 112, 19572011, 2012.

www.atmos-chem-phys.net/17/2163/2017/ Atmos. Chem. Phys., 17, 21632187, 2017