The aerosol time-of-flight mass spectrometer (ATOFMS) provides size-resolved information on the chemical composition of single particles with high time resolution. Within SAPUSS (Solving Aerosol Problems by Using Synergistic Strategies), continuous ATOFMS measurements of ambient particles were made simultaneously at two urban locations: urban background (UB) site and roadside (RS) site in the city of Barcelona (Spain) from 17 September to 18 October 2010. Two different instrumental configurations were used: ATOFMS (TSI 3800) with a converging nozzle inlet (high efficiency at about 800–2000 nm) at the UB site and ATOFMS (TSI 3800-100) with an aerodynamic lens inlet (high efficiency at about 300–700 nm) at the RS site. This is the first time, to our knowledge, that two ATOFMS instruments have been deployed in the same field study. The different instrument configurations had an impact on the observed particle types at the two sites. Nevertheless, 10 particle types were detected at both locations, including local and regional elemental carbon (22.7–58.9 % of total particles), fresh and aged sea salt (1.0–14.6 %), local and regional nitrate-containing aerosols (3–11.6 %), local lead-containing metallic particles (0.1–0.2 %), and transported Fe-nitrate particles (0.8–2.5 %). The ATOFMS at the UB also characterized four particle types: calcium-containing dust (0.9 %), Saharan dust (1.3 %), vanadium-containing particles (0.9 %), and vegetative debris (1.7 %). By contrast, the high statistical counts of fine particles detected at the RS allowed identification of eight particle types. Four of these contained organic nitrogen of primary and secondary origin, which highlights the complex nature of the sources and processes that contribute to this aerosol chemical component. Aminium salts were found related to coarse sulfate-rich particle types, suggesting heterogeneous reaction mechanisms for their formation. The other four particle types mainly containing organic carbon were found spiking at different types of the day, also showing a complex single-particle mixing state relationship between organic carbon and nitrate. This ATOFMS study clearly shows that the composition of atmospheric fine particles in Barcelona, and likely other Mediterranean urban areas, is complex, with a wide range of local and regional sources combining with chemical processing to produce at least 22 different particle types exhibiting different temporal behaviour. The advantage of using two ATOFMS instruments is also demonstrated, with the nozzle-skimmer configuration enabling detection of coarse dust particles and the aerodynamic lens configuration allowing better identification of particles rich in organic carbon and amines. Overall, we find that organic nitrogen is a considerable fraction of the single particles detected, especially at the traffic-dominated RS site. Further studies are needed, especially at high time resolution, to better understand the sources and properties of particulate organic nitrogen.
A substantial number of studies have shown a relationship between measures of particulate air pollution and a variety of adverse health indicators (WHO, 2004). Formulation of cost-effective air pollution control policies depends upon a sound knowledge of source contributions to ambient concentrations. Only with such knowledge can realistic cost-benefit evaluations be conducted. The major sources of ambient particles in most urban areas are primary emissions from road traffic and other fuel combustion, secondary particles arising from condensation or chemical processing, and resuspension of soils and road dusts (AQEG, 2005; Harrison et al., 2012). Marine aerosol can also contribute in coastal locations and the interactions of anthropogenic trace gases with natural aerosol (i.e. dust, sea salt) can also have significant effects on aerosol composition (Abbatt et al., 2012).
Measurement of particle composition by online mass spectrometry has developed extensively over the last 2 decades and is currently the fastest growing area of atmospheric aerosol research (Laskin et al., 2012). The aerosol time-of-flight mass spectrometer (ATOFMS) has been used in many previous field studies to determine the chemical constituents of atmospheric aerosols (Pratt and Prather, 2012). It can identify both refractory and non-refractory species in single particles and can provide size-resolved information on particle sources and atmospheric processing at high time resolution (Prather and Pratt, 2012; Laskin et al., 2012). The ATOFMS has been used in a number of recent field studies in urban areas of Europe (Dall'Osto and Harrison, 2012; Healy et al., 2013) to identify and characterize particles from a diverse range of anthropogenic sources including traffic, solid fuel burning, industry, soil and road dust, marine aerosol, and secondary aerosol formation processes. However, it is worthy of mention that ATOFMS source apportionment capabilities are limited by the difficulties in quantification of its outputs (Reilly et al., 2000; Schoolcraft et al., 2001). Nevertheless, single-particle analysis is an important analytical tool that allows us to determine how the myriad chemical constituents are distributed between individual particles (mixing state; Pratt and Prather, 2012). The ATOFMS has often reported a number of particle types, which at times are difficult to associate with a specific aerosol source (Pastor et al., 2003; Dall'Osto and Harrison, 2006, 2012).
The objective of the present manuscript is to report a detailed analysis of the ATOFMS particle types detected during a field measurement campaign carried out in Barcelona, Spain, as part of the SAPUSS (Solving Aerosol Problems by Using Synergistic Strategies) project. The ATOFMS cannot provide quantitative aerosol mass loading concentrations, but its strength relies in the fact that it can monitor in real-time variations in the single-particle composition. In other words, small variations in the particle mixing state results in a single-particle mass spectra. As a result, a number of atmospheric processes and aerosol sources can be monitored in real time. In this paper we discuss not only information on the mass spectra but also diurnal trends persisting over 4 weeks. Further information on the intensive field campaign can be found in the overview paper by Dall'Osto et al. (2013a). Two different ATOFMS instruments were deployed during the 4-week measurement period – one at a roadside (RS) site and the other at an urban background (UB) site. This is the first time an ATOFMS has been deployed in Spain and, to the best of our knowledge, it is also the first time (worldwide) that two ATOFMS instruments have been deployed simultaneously in the same field campaign. The similarities and differences in particle types detected at both sites is described in detail and attributed to a range of local and regional sources as well as to different chemical and physical processes.
The SAPUSS field measurement campaign involved a large variety of
instrumentation deployed simultaneously at a number of monitoring sites in
Barcelona (Spain), between 17 September and 18 October 2010
(local time, UTC
RS site was situated in a car park next to a major road (Carrer Urgell). The road, which crosses the city from south-east to north-west, is a street canyon composed of a two-way cycling path and a one-way four lane vehicle road. Vehicle intensity for the month of measurements was about 17 000 vehicles per day.
UB site was situated in a small park at the north-western
periphery of the city centre. A main road (Avenida Diagonal, 127 000 vehicles day
The two sites were about 2 km from each other (Dall'Osto et al., 2013a). While the UB site was open to wind from all directions, the wind flow and turbulence at the RS site were partially affected by the nearby street canyons and vehicular traffic. Previous reports from the SAPUSS campaign, based on measurements of organic and elemental components of the aerosol, indicate that the particle composition and thus sources are similar at both sites (Dall'Osto et al., 2013b; Alier et al., 2013). Specifically, six organic aerosol (OA) components were identified at both sites: two of primary anthropogenic origin, three of secondary origin, and one whose source was not clearly defined (Alier et al., 2013). Elemental analysis provided by particle-induced X-ray emission (PIXE) enabled identification of nine different aerosol sources at both sites: three of regional origin, three types of dust aerosols, and three types of industrial aerosols (Dall'Osto et al., 2013b).
The mass spectrometers were housed in air-conditioned trailers at both
sites. Sampling was performed ca. 4 m above ground using a quarter-inch
internal diameter stainless steel tube fitted with a PM
The two different inlet configurations (aerodynamic lens and nozzle skimmer) strongly affect the size distributions of the detected particles and the overall aerosol population (Gard et al., 1997; Su et al., 2004). Hence, only a qualitative description of the detected particles is presented in this study. Furthermore, the ATOFMS mass spectrum is qualitative in that the intensities of the mass spectral peaks are not directly proportional to the component mass but are dependent on the particle matrix, the coupling between the laser and the particle, as well as the shot-to-shot variability of the laser (Dall'Osto and Harrison, 2012). Recent studies (Jeong et al., 2011) report excellent correlations for inorganic species (sulfate, nitrate, and ammonium) but weaker ones between total organic and elemental carbon (EC) detected with ATOFMS and other instruments (Jeong et al., 2011). However, the ATOFMS can provide quantitative information on particle number as a function of composition, providing a measure of all particle components, and can be used to assess mixing state.
The ATOFMS datasets were imported individually into YAADA (Yet Another
ATOFMS Data Analyzer) and single-particle mass spectra were grouped with
adaptive resonance theory neural network, ART-2a (Song et al., 1999). The
parameters used for ART-2a in this study were learning rate 0.05, vigilance
factor 0.85, and 20 iterations. Further details of the parameters can be
found elsewhere (Dall'Osto and Harrison 2006; Rebotier and Prather 2007).
An ART-2a area matrix (AM) of a particle cluster represents the average
intensity for each
The study area is affected by a convergence of air masses with different characteristics: the cold air coming down from medium and high latitudes and the warm air coming up from tropical and subtropical latitudes (Dall'Osto et al., 2013a). Five air mass meteorological regimes were classified during the SAPUSS field study, following the procedure described in Dall'Osto et al. (2013b): Atlantic (ATL), European-Mediterranean (EUR), North African east (NAF_E), North African west (NAF_W), and Regional (REG). Furthermore, meteorological variables (atmospheric pressure, wind speed, wind direction, solar radiation, temperature, and relative humidity) were also recorded at UB and RS SAPUSS monitoring sites during the whole field study. For further details, the reader is referred to the SAPUSS overview paper (Dall'Osto et al., 2013a).
ATOFMS particle clusters identified from the SAPUSS campaign.
Overall, 890 873 particle mass spectra were apportioned at the RS and
221 139 at the UB. This large difference in detected particle numbers is
likely a result of the combined effects of the location and detection
efficiencies of both instruments. As shown in Fig. 1, the number and size
distribution of the particles detected by the two mass spectrometers is
quite different and reflects their expected performance characteristics
(Gard et al., 1997; Su et al., 2004). The instrument with the aerodynamic
lens detected considerably more particles below 1
Size distributions of collected ATOFMS particles at the two SAPUSS monitoring sites.
Average mass spectra of the 10 single-particle types observed at both the RS and UB sites.
Two main EC particle types, representing together more than 50 % of
detected particles (58.9 % at the UB, 53.8 % at the RS), were
identified at both sites. For reasons outlined below, they are named
EC_Aged_R (regional) and EC_Aged_L (local). Both EC particle
types presented a fine aerosol size distribution mode at both sites (about
300–500 nm; see Fig. S2 in the Supplement). Figure 2a shows the positive
mass spectrum of particle type EC_Aged_R. It is dominated by EC peaks
at
Diurnal profiles of the common particle types detected at both the RS and UB sites. Differences between RS and UB sites were minor and average diurnal profiles are presented.
The application of the ART-2a neural network algorithm to the ATOFMS data
apportioned two main distinct nitrate particle types, already previously
reported (Dall'Osto et al., 2009; Harrison et al., 2012; Decesari et al.,
2014). The first (local nitrate (Loc-NIT); 4.2–7.3 % of particles by
number) appears to be locally produced in urban locations during night-time,
whilst the second (long-range transport – nitrate (LRT-NIT); 4.2–7.3 %
of particles by number) is regionally transported within the Iberian
Peninsula and the rest of Europe. Briefly, particle type Loc-NIT is
characteristic of nitrate aerosol in small particles (
The K-CN particle type was a minor one, representing only 1.3–2.4 % of
the total particles analysed. Figure 2f shows the average mass spectrum,
which features a strong peak at
Two sea salt particle types (fresh and aged) were detected at both sites,
accounting for 9.4 and 18.3 % of the total particles sampled at RS and
UB, respectively (Table 1). The higher percentage detected at the UB is
likely due to the instrument configuration given that the nozzle inlet
enables more efficient detection of coarser particles. The average mass
spectrum (Fig. 2g) for the particle type assigned to fresh sea salt (labelled
NaCl) shows peaks typical of sodium chloride clusters ([Na]
Two common ATOFMS particle types have average mass spectra that were
dominated by metals. The first was rich in iron (type Fe, 1.4 % of the
total particles) and has a spectrum (Fig. 2i) characterized by a strong
signal at
The average mass spectrum of the second metal-rich particle type is shown in
Fig. 2j. This particle type is labelled Pb since lead is one of the largest
contributors in the positive mode, occurring at
Average mass spectra of the four single-particle types only observed at the UB site.
The ATOFMS fitted with converging nozzle inlet located at the UB site detected four particle types that were not observed at the RS site. Each of the particle types make a minor contribution (0.9–1.7 %) and overall they represented less than 5 % of the total particles sampled at the UB site.
A particle type containing vanadium (
The average ATOFMS mass spectrum of a dust particle type rich in calcium is
shown in Fig. 4b, where peaks for calcium (
Average mass spectra of the four amine-containing single-particle types only observed at the RS site.
A rarely observed particle type with peaks due to titanium at
The ATOFMS has already proven to be a good tool for identifying and
separating dust (mainly Ca-rich or Al-Si-rich) and biological particles
(Fergenson et al., 2004). A particle type dominated by K (
The ATOFMS fitted with an aerodynamic focussing lens located at the RS site detected eight particle types that were not observed at the UB site.
Amines are ubiquitous in the atmospheric environment and have been detected
in marine, urban, and rural atmosphere in the gas and particle phases as well
as aqueous fog and rain water (Ge et al., 2011). The ATOFMS is a particularly
good aerosol instrument for studying amines because the LDI laser wavelength
(266 nm) ionizes them very efficiently (Angelino et al., 2001; Healy et al.,
2015). During the SAPUSS measurement campaign, four amine particle types were
detected at the RS. To the best of our knowledge, this is the first time such
a variety of organonitrogen particle types has been detected at the single-particle level in real time in urban air.
Diurnal trend of amines
Average mass spectra of the four OC-rich single-particle types only observed at the RS site.
Overall, the four amine-containing particle types were found distributed
mainly in the sub-micron mode. Amine (POA 58) and amine (ETS 84), which are
attributed to primary emissions from traffic and ETS respectively, have
similar size distributions which peak around 400–500 nm (Fig. S2). Amine
(SOA 59) and amine (SOA 114) were also found in the same size range, but
presented a broader shape, suggesting partial condensation of SOA material on
existing particles. Previous work of Angelino et al. (2001) used the peak at
The high efficiency of the ATOFMS equipped with the aerodynamic lens and
deployed at the RS allowed us to characterize four particle types rich in
organic compounds.
Org. ( Org. ( Org. (
Overall these four OC-rich particle types indicate that a number of processes
and sources are likely producing oxidized organic aerosols. There are at
least four main peaks during the day: a morning traffic rush hour
(09:00–10:00), an afternoon one during the hottest part of the day (15:00),
and two evening ones at 20:00 (sunset) and at about 22:00–23:00. This is
likely to be governed by a combination of emissions from local sources during
rush hour periods, as well as by meteorological parameters such as
atmospheric wind speed, wind direction, relative humidity, and temperature.
During the SAPUSS intensive field study two ATOFMS instruments were deployed simultaneously. The ATOFMS deployed at the RS site was equipped with an aerodynamic lens inlet system, allowing characterization of primary traffic aerosols as well as other primary and secondary aerosols affecting this heavily urbanized area of Barcelona. This type of ATOFMS (Su et al., 2004) has a very high efficiency in sampling aerosols (more than 1 000 000 single-particle mass spectra were collected at the RS), particularly for sub-micron particles in the size range 300–700 nm. The ATOFMS deployed at the UB site was equipped with a converging nozzle inlet system (Gard et al., 1997), which has a low aerosol collection efficiency (Dall'Osto et al., 2006), but it is particularly well suited for sampling coarser aerosols in the size range 800–2000 nm.
Overall, 10 particles types were detected at both sites (Table 1). Two of these particle types, composed of EC internally mixed with secondary inorganic species, described more than half of the classified single-particle mass spectra. EC_aged_R was found accumulating within stagnant air masses, with a flat diurnal profile and suggesting a certain physico-chemical stability. In contrast, a more local but processed form of EC (EC_aged_L) was found to possess a finer submicron mode and enhanced concentrations during afternoon periods.
Two different types of nitrate-dominated aerosols were observed, in line with
a previous ATOFMS study in London (Dall'Osto et al., 2009). The first
(LRT_NIT) was attributed to regional nitrate (accumulation mode, volatile,
more NH
Two types of sea salt particles were also identified at both sites. NaCl particles showed a peak in the diurnal profile at 15:00, related to the sea breeze and enhanced under NAF_E air masses. NaCl-NIT were mainly observed during more anthropogenically influenced air masses.
Two types of particles rich in metallic elements were found at both
monitoring sites. One, rich in iron and internally mixed with nitrate, was
found to be distributed in the fine accumulation mode at about 400 nm and
related to regional air masses. This observation supports previous findings
(Dall'Osto et al., 2010; Harrison et al., 2012) that showed fine
iron-containing aerosols are able to travel long distances and are thus
related to aged air masses. Other studies have reported anthropogenic
Fe-containing particles internally mixed with secondary species such as
sulfate (Furutani et al., 2011; Moffet et al., 2012) originating from coal
combustion in Asian continental outflows. This study shows that – within the
European continental outflow – Fe-containing particles are instead
internally mixed with nitrate. This observed difference is likely due to
emissions from coal combustion in Asia which are rich in SO
A second metallic particle type rich in lead and chloride was identified. This particle type was related to more local sources, presenting sharp spikes in concentration. It is interesting to note this particle type was found correlated with hourly elemental mass concentrations determined by PIXE analysis (Dall'Osto et al., 2013b), showing that this particle type can be a major source of submicron chloride in the urban area of Barcelona.
The ATOFMS equipped with the converging nozzle inlet at the UB site detected
four further different particle types. Two types of dust particles were
found, both occurring mainly in the coarse mode (
Eight particle types were detected by the ATOFMS with aerodynamic focussing lens at the RS site. Overall, these particle types described less than 10 % of the aerosol population, but their mass spectra, as well as their peculiar diurnal profiles, allow us to advance our understanding of the OC-ON-nitrate mixing state of urban aerosols.
Four particle types contain amines, which, in addition to ammonia, are
important atmospheric bases (Ge et al., 2011). Urban concentrations of
ammonia in Barcelona are higher than those reported in similar urban
background sites in Europe, especially in summer (Reche et al., 2012).
Conversely, in winter, levels of ammonia were higher at traffic-affected
sites, suggesting a contribution from vehicle emissions (Reche et al., 2012).
Emissions of ammonia from vehicular traffic have been widely reported and
they may increase in the future because it is not a regulated pollutant
(Suarez-Bertoa et al., 2015). In comparison, the sources, atmospheric
transformation, and sinks of amines are more poorly characterized. Overall,
both primary (amine POA 58 and amine ETS 84) and secondary (amine SOA 59
and amine SOA 114) sources of amine-containing particles were identified
during SAPUSS. The most abundant amine particle type (amine POA 58,
0.8 % of total particles at RS) was attributed to traffic activities and
the second-most abundant (amine ETS 84, 0.5 % of the total particles at
RS) was also associated with environmental tobacco smoke. Concurrent SAPUSS
measurements (Alier et al., 2013) of nicotine concentrations were much higher
at the RS site (58 ng m
Amines were also related to secondary aerosol production, although a very
complex dynamic was found associated with their occurrences. They were found
to peak during the warmest part of the day (15:00) and during evening times
(22:00–23:00). Amine (SOA 59) was found particularly enhanced in regional
air masses (13–17 October 2010) when nitrate concentrations were also high
(Dall'Osto et al., 2013b). By contrast, amine (SOA114) was more abundant in
NAF_E humid air masses (7–11 October 2010). This latter type of SOA
was internally mixed with nitrate, suggesting aminium salt formation under
such specific conditions. Previous ATOFMS studies reported that most of the
amines volatilized during cold seasons, whereas during summer most were in
the form of low-volatility aminium nitrate and sulfate salts when particle
acidity was higher (Pratt et al., 2009). This observation supports previous
laboratory studies which reported that aerosol-containing non-salt organic
amines are more stable and less volatile than nitrate salts (Murphy et al.,
2007). Overall, amines can undergo oxidation by OH, O
Finally, some consideration should be given to the four specific organic particle types detected at the RS. One (Org. (Lub. Oil)) was found to be related to primary lubricating oil traffic emissions. More difficult is the attribution of the remaining three, which each contain an internal mixture of OC and nitrate. This is not surprising given the fact that the urban atmosphere is heavily contaminated by traffic emissions, the main producers of the two chemical species (Dall'Osto et al., 2013a).
Two different types of organic carbon/nitrate particle types were found. One
(OC-NIT) was found spiking in the afternoon. By contrast, a nitrate with a
strong aromatic signature (OC-Aro-NIT) was found mainly during night-time
(80 % of the time) and showing a sharp concentration peak at
19:00–20:00.
Part of the OC-Aro-NIT could be associated with products from the reaction of
aromatic components with NO
The fourth organic particle type (OC-CHO) was found rich in oxidized organic
carbon and associated with nitrate from a traffic source. In a previous
ATOFMS study, considerable effort was made to apportion cooking-related
particle types (Dall'Osto and Harrison, 2012). However, only a particle type
exhibiting maximum frequency during the warmest part of the day and
associated with secondary aerosol production from traffic-related
semi-volatile aromatic compounds was found. During this study, we were again
not able to associate a specific particle type with cooking activities. In a
companion SAPUSS study, Alier et al. (2013) reported an aerosol source formed
mainly by C
In summary, the two ATOFMS instruments deployed during the SAPUSS field
measurement study showed that the urban atmosphere contains a complex
mixture of aerosol particles emitted from a variety of sources and formed
via numerous atmospheric processes. We have identified 22 different particle
types, characterized by specific single-particle mass spectra and temporal
trends. European Union abatement of traffic-related NO
Data are available by contacting the corresponding author.
Financial support for this study was provided by the Marie Curie FP7 SAPUSS (FP7-PEOPLE-2009-IEF, Project number 254773) and previously supported by research projects from the D. G. de Calidad y Evaluacion Ambiental (Spanish Ministry of the Environment) and the Plan Nacional de IyD (Spanish Ministry of Science and Innovation) CGL2010-19464-VAMOS, CTQ2009-11572 and CTQ2009-377-14777-C02-01-AERTRANS). The SAPUSS team is also acknowledged. Edited by: S. Decesari Reviewed by: two anonymous referees