Evaluation of the Submicron Particles Distribution Between Mountain and Urban Site: Contribution of the Transportation for Defining Environmental and Human Health Issues

Transportation is one of the main causes of atmospheric pollution, especially in downtown big cities. Researchers usually point their attention to gaseous and/or particulate matter pollutants. This paper investigated the role of submicron particles, particularly the fraction ranging between 5–560 nm, in aerosol chemistry for identifying the contribution of autovehicular traffic and investigating the doses deposited in the human respiratory tract. Measurements carried out by two Fast Mobility Particle Sizer (FMPS, TSI) analyzers were simultaneously performed at two different sampling sites (an urban and a mountain site) during workdays and weekends in July. The total particle number (2–2.5 times higher in the urban site), the aerosol size distribution (different modes during the day), and the ultrafine/non-ultrafine particle ratios (ranging between 2–4 times between two sites) were investigated and discussed in relationship to the high autovehicular traffic in Rome and the almost null anthropogenic emissions at the mountain site, as well as the differing contributions of both to the “fresh nucleation” and to “aged aerosol”. Furthermore, the regional cumulative number doses deposited in the human respiratory tract were studied for both sites: The difference between the urban/mountain site was very high (up to 15 fold), confirming the pollutant role of transportation.


Introduction
Although urban areas occupy only 2% of the planet's surface, they are primarily responsible for atmospheric pollutant emissions (e.g., about 80% of CO 2 worldwide) as over half of the human population lives in such areas [1]. This occurrence simultaneously means a strong driving force for economic development and an excessive exploitation of resources and the environment, degradation and congestion. In Italy, the demographic trend is similar to the global one; urban areas represent 3.3% of the country whereas 23.4% of the population reside in such areas and this concentration affects various factors, particularly those of public health [2]. In 2005, the Position Paper [3] on doses deposited in the human respiratory tract based on the measurements performed in the two sites will be estimated for evaluating the impact of anthropogenic sources on the human health.

Materials and Methods
The methodology used is based on performing simultaneous determinations in two different sites and relies on the aerosol aerodynamic-diameter analysis for evidencing the contribution of traffic emission to the submicron particle fraction.

Sampling Sites
Submicron particle measurements were carried out in downtown Rome and at Monte Terminillo. Figure 1 shows the locations of Rome and Monte Terminillo in the Latium region (Central Italy): Rome is 30 km from the sea, whereas Monte Terminillo is on the eastern part of the region. evaluation of the doses deposited in the human respiratory tract based on the measurements performed in the two sites will be estimated for evaluating the impact of anthropogenic sources on the human health.

Materials and Methods
The methodology used is based on performing simultaneous determinations in two different sites and relies on the aerosol aerodynamic-diameter analysis for evidencing the contribution of traffic emission to the submicron particle fraction.

Sampling Sites
Submicron particle measurements were carried out in downtown Rome and at Monte Terminillo. Figure 1 shows the locations of Rome and Monte Terminillo in the Latium region (Central Italy): Rome is 30 km from the sea, whereas Monte Terminillo is on the eastern part of the region. In Rome, the sampling site was located at INAIL's Pilot Station (41°53′46″ N, 12°29′46″ E), the sampling point was at 3 m height from the ground. The site is characterized by a high density (ratio of 706 vehicles per 1000 inhabitants, source Automobile Club d'Italia, ACI) of autovehicular traffic (e.g., cars, motorbikes, buses, pullmans), whereas the street could be assessed like a canyon (height/width ratio 3:1 between buildings and street). Furthermore, according to recent data on vehicle fuels (source ACI), almost 40% of the total park was diesel.
Monte Terminillo (42°28′23.77″ N, 12°59′50.35″ E), a massif in the Monti Reatini, Central Italy, is located some 20 km from Rieti and 100 km from Rome with a high altitude of 2217 meters (7274 ft), and the temperature ranged between 10 °C and 17 °C with a relative humidity of 70% during the entire investigated period. No rainy day was recorded during the campaign. There were two different ways to reach Monte Terminillo by car but access was limited and the anthropogenic emissions were basically absent. Monte Terminillo could be considered a remote site.

Measurement Equipment
Particle number size distributions at high time-resolution (1 s time) were carried out by means of a Fast Mobility Particle Sizer (FMPS, model 3091, TSI, Shoreview, MN, USA) with a range from 5.6 to 560 nm electrical mobility diameter. The instrument worked at 10 L·min −1 for minimizing the diffusion UFP losses and counts and classifies the particles in 32 size channels. The FMPS performance was investigated by comparing it with a Scanning Mobility Particle Sizer (SMPS) [14,15]. The measurement campaign was simultaneously performed in both sites (Rome and Monte Terminillo) and was three weeks long and carried out in July 2018. In Rome, the sampling site was located at INAIL's Pilot Station (41 • 53 46 N, 12 • 29 46 E), the sampling point was at 3 m height from the ground. The site is characterized by a high density (ratio of 706 vehicles per 1000 inhabitants, source Automobile Club d'Italia, ACI) of autovehicular traffic (e.g., cars, motorbikes, buses, pullmans), whereas the street could be assessed like a canyon (height/width ratio 3:1 between buildings and street). Furthermore, according to recent data on vehicle fuels (source ACI), almost 40% of the total park was diesel.
Monte Terminillo (42 • 28 23.77 N, 12 • 59 50.35 E), a massif in the Monti Reatini, Central Italy, is located some 20 km from Rieti and 100 km from Rome with a high altitude of 2217 m (7274 ft), and the temperature ranged between 10 • C and 17 • C with a relative humidity of 70% during the entire investigated period. No rainy day was recorded during the campaign. There were two different ways to reach Monte Terminillo by car but access was limited and the anthropogenic emissions were basically absent. Monte Terminillo could be considered a remote site.

Measurement Equipment
Particle number size distributions at high time-resolution (1 s time) were carried out by means of a Fast Mobility Particle Sizer (FMPS, model 3091, TSI, Shoreview, MN, USA) with a range from 5.6 to 560 nm electrical mobility diameter. The instrument worked at 10 L·min −1 for minimizing the diffusion UFP losses and counts and classifies the particles in 32 size channels. The FMPS performance was investigated by comparing it with a Scanning Mobility Particle Sizer (SMPS) [14,15]. The measurement campaign was simultaneously performed in both sites (Rome and Monte Terminillo) and was three weeks long and carried out in July 2018.

Dose Deposition Model
Particle deposition in the human respiratory system was evaluated using the Multiple-Path Particle Dosimetry model (MPPDv3.01, ARA 2015, ARA, Arlington, VA, USA) [16]. The 60th percentile human stochastic lung was considered along with the following settings: (i) A uniformly expanding flow, (ii) upright body orientation, and (iii) nasal breathing with a 0.5 inspiratory fraction and no pause fraction. Moreover, the following parameters were used for a Caucasian adult male under light work physical activity, based on the ICRP report [17][18][19][20]: (i) A functional residual capacity (FRC) of 3300 mL, (ii) an upper respiratory tract (URT) volume equal to 50 mL, (iii) a 20 min −1 breathing frequency, and (iv) an air volume inhaled during a single breath (tidal volume, V t ) of 1.25 L.
Since FMPS measures aerosol size number distribution as a function of the electrical mobility diameter (d), d values were transformed to aerodynamic diameter (d a ) according to Equation (1) [21]: where C c is the Cunningham slip factor for a given diameter, ρ is the particle density, and χ is the particle dynamic shape factor. A total of 1.5 g cm −3 particle density was assumed. χ as a function of d a was estimated by interpolating, through a cubic spline function [22] with the data reported by Hu et al. [23] in the range from 0.1 to 1.8 µm in Beijing. For d a below this range, the relevant lower bound χ values were adopted. For each respiratory act, the doses described below were calculated as regional dose size number distributions (D R ) as function of time (t) in the head (H), tracheobronchial (TB), and alveolar (Al) regions (R): where d i is the diameter of particles classified in the i th FMPS size channel, F R (d i ) is the relevant deposition fraction at a given R region, C(d i ,t) is the concentration of particles in the i th FMPS size channel as a function of time, and V t is the tidal volume. Average regional and total size number dose distributions over 1h time interval ∆t: where C(d i , ∆t) is the average concentration over the time interval ∆t of the particles in the i th FMPS size channel. Total regional number doses as functions of time: where 32 is the number of FMPS size classes. The total number dose in the respiratory system as functions of time: Cumulative regional number doses and cumulative total number dose in the respiratory system were calculated over a 1 h time interval ∆t, according to Equations (5) and (6), respectively:

Results
The number concentrations and the relative aerosol size distributions can show different values in different environmental conditions. The meteorology, the daytime period, and the possible presence of local emission sources close to the monitoring site are parameters significantly affecting aerosol size distribution. In this study, according to the description reported in the previous section, the sampling site in Rome allowed to minimize the influence of weather conditions. Furthermore, the measurements performed only in the summer period allowed the study to not consider the domestic heating sources, which are an extremely important contribution to the aerosol. Therefore, the only contribution to aerosol came from autovehicular traffic. This source was quite null in the Monte Terminillo site whereas it was strong in the Rome site.

Particle Number Concentration
Preliminary data evaluation is shown in Table 1 where the number particle concentration data (5.6-560 nm) obtained in the two sites are shown (to be noted that for workdays the authors mean the week period from Monday to Friday, whereas for weekends the authors mean Saturday and Sunday). Table 1. Typical average number concentration (# cm −3 ), standard deviation, and 95th percentile of total particles, ultrafine particles (UFPs), and non-ultrafine particles (non-UFPs) determined in the two sites during workdays and weekends.  Table 1 shows the total number concentration data along with a subdivision in ultrafine particles (UFPs, 5.6-99.3 nm) and in non-ultrafine particles (non-UFPs, 99.3-560 nm). Looking at the data, it can be seen that the total particles in the urban site averaged twice the level recorded at the mountain site during workdays and about 2.5 times on weekends. Similar ratios are observed for UFPs on workdays and weekends. On the other hand, non-UFPs show different relationships; during workdays the average ratio between the urban site and the mountain site was about 4, whereas during weekends it fell to 1.2. Thus, over the primary origin of such particles, it should also consider the physical-chemical processes leading to the novel particle formation starting from precursors in vapor or gas phase and from transformation processes: These phenomena are more incisive with high intensity of emission sources, i.e., autovehicular traffic. The measurements 95th percentile value confirms this occurrence.

Particles
Although the average values on workdays and weekends determined in the Rome site were quite similar, as well as the maximum total particle value being about 7 times higher on weekends than on workdays, 95% of workday measurements were less than 26,930 # cm −3 whereas only 5% of the weekend measures exceeded 18,615 # cm −3 (Figure 2). incisive with high intensity of emission sources, i.e., autovehicular traffic. The measurements 95th percentile value confirms this occurrence. Although the average values on workdays and weekends determined in the Rome site were quite similar, as well as the maximum total particle value being about 7 times higher on weekends than on workdays, 95% of workday measurements were less than 26,930 # cm −3 whereas only 5% of the weekend measures exceeded 18,615 # cm −3 (Figure 2).

Human Respiratory Doses
Figure 3 describes the cumulative number doses (Dc R (Δt)) deposited in the 00:00-01:00 and 07:00-08:00 time intervals, in the H, TB, and Al regions, as well as the relevant total doses (Dc Tot (Δt)), estimated for the Rome and Terminillo measuring sites during workdays and holidays. The two Δt time intervals were chosen in order to evidence the effects of the emission source intensities and of the planetary boundary layer (mixing height) on aerosol concentrations and hence on the relevant respiratory doses. In the 07:00-08:00 time interval, the contribution of vehicular traffic in Rome started increasing, whereas in the Terminillo area aerosol emissions, apart from biogenic ones, were almost negligible. From 00:00 to 01:00, Rome traffic intensity decreased and in the Terminillo area, the contribution of biomass burning ceased being active. In this time interval, the Planetary Boundary Layer (PBL) mixing height was shallower and the pollutants tended to concentrate.

Particle Size Distribution
A complete and scrupulous aerosol size distribution should be based on the attribution of a (channel) size to each particle but the result should be extremely uncomfortable and not very usable. Therefore, the entire particle range (<1000 nm) was divided into a limited size range for measuring the particle number concentration in each size channel [24][25][26][27][28], i.e., the aerosol particle diameters were distributed in a rather wide range and their number concentrations vary considerably

Particle Size Distribution
A complete and scrupulous aerosol size distribution should be based on the attribution of a (channel) size to each particle but the result should be extremely uncomfortable and not very usable. Therefore, the entire particle range (<1000 nm) was divided into a limited size range for measuring the particle number concentration in each size channel [24][25][26][27][28], i.e., the aerosol particle diameters were distributed in a rather wide range and their number concentrations vary considerably depending on the size. It should be noted that, in the aerosol size distribution studies performed in urban and suburban sites in the presence of traffic emissions, the particle classifications were observed even in more than three intervals [29]. For a better size distribution discussion, the authors considered the atmospheric submicron aerosol divided into three size ranges defined as follows: -Nucleation range (~3-25 nm); -Aitken nuclei range (~25-90 nm); -Accumulation mode (~90-1000 nm; this paper studied particles in the range~90-560 nm).
The nucleation range was defined below 25 nm, with the upper limit ranging between 20 and 30 nm [30][31][32]. The particles in the nucleation mode can derive both from direct emission and from formation processes, i.e., conversion reactions due to: -Rapid cooling and dilution of gases and/or vapors produced by emissions; -Chemical reactions involving precursors already present in the atmosphere.
In agreement to what has been reported in the literature [33], high particle concentrations in nucleation mode were shown in the early hours of the morning, i.e., in the presence of high emission sources (autovehicular traffic). These particles tended to rapidly decrease spatially and temporally. The average daily particle size distribution in the nucleation mode showed a peak centered between 7 and 16 nm during workdays and between 8 and 14 nm during weekends in downtown Rome ( Figure 5). Once formed and suspended in the atmosphere, the particles in the nucleation mode were characterized by Brownian motions: The collisions generated aggregation processes, increasing the particles in the nucleation mode up to the accumulation mode. It should be noted that the levels were different depending on the different intensity of the emission sources during the day.
On the other hand, aerosol in the range 25-90 nm (i.e., the Aitken nuclei range) comes from coagulation and condensation processes of particles in nucleation mode, but it can also be (produced and) directly emitted into the atmosphere from combustion sources. Particles in the accumulation mode are generally made up of carbon compounds, such as soot or dust. These can derive both from fuel engine combustion and from lubricating diesel or petrol oils, as well as from coagulation processes of particles in nucleation mode [34]. The "accumulation" aerosol (i.e., aerosol in accumulation mode) has a long lifetime in the atmosphere and can be transported over long distances as well. Figure 6a shows the average UFP percentage to be predominant with respect to the relative percentage of non-UFPs in both sites both on workdays and weekends. A more detailed subdivision in the aforementioned size ranges makes it possible to highlight further considerations (Figure 6b).
Particles in nucleation mode are predominant in the mountain site, whereas in downtown Rome their percentage depends on the emission source intensity: They represent~35% on workdays whereas they are~55% on weekends. On the other hand, the accumulation mode becomes relevant during workdays in the urban area. In fact, it was about twice recorded during weekends. The particles were continuously emitted into the atmosphere by autovehicular traffic but at the same time they aggregated and/or coagulated and increased both the Aitken mode (small contribution) and the accumulation mode. In the Monte Terminillo site, the average particle number percentages both in nucleation mode and in the Aitken nuclei remained quite constant, whereas there was a noticeable increase in accumulation mode. This increase was opposite to what happened in the urban site but leads to the same considerations. The explanation is that during weekends the Monte Terminillo site could be reached by tourist coaches: access was limited but emissions from bus traffic strongly affected the ratio. The daily particle number concentration in the three subdivisions shows more than 2-3 times the value in the urban site than in the mountain site (Figure 6b). On the other hand, aerosol in the range 25-90 nm (i.e., the Aitken nuclei range) comes from coagulation and condensation processes of particles in nucleation mode, but it can also be (produced and) directly emitted into the atmosphere from combustion sources. Particles in the accumulation mode are generally made up of carbon compounds, such as soot or dust. These can derive both from fuel engine combustion and from lubricating diesel or petrol oils, as well as from coagulation processes of particles in nucleation mode [34]. The "accumulation" aerosol (i.e., aerosol in accumulation mode) has a long lifetime in the atmosphere and can be transported over long distances as well. Figure 6a shows the average UFP percentage to be predominant with respect to the relative percentage of non-UFPs in both sites both on workdays and weekends. A more detailed subdivision in the aforementioned size ranges makes it possible to highlight further considerations (Figure 6b).    Figure 6. (a) UFPs and non-UFPs percentage ratio in the two sites during workdays and weekends; (b) particle number concentrations and average percentages of the submicron aerosol subdivided in three modes, i.e., nucleation mode (red bars), Aitken nuclei mode (yellow bars), and accumulation mode (green bars), in the two sites investigated.
The difference between the two sites is evident in the aerosol size distributions reported in Figure 7a,b. Figure 7a,b show typical size distributions in the workday morning (07:00-09:00) at the Monte Terminillo site and in downtown Rome, respectively. In addition to the differences in the particle number (full-scale 6 × 10 3 for the Monte Terminillo site, full-scale 1.4 × 10 5 for downtown Rome), the size distribution profiles were different: In the urban area the mode between 7 and 14 nm was prevalent, meaning fresh particles emitted by vehicular traffic ("fresh nucleation"). This occurrence confirmed what has been reported in the literature; in the atmosphere most of the nanometer-sized particles (d ≤ 20 nm) mainly derive from gas-particle conversion processes [35], among which homogeneous (nucleation in vapor phase) and heterogeneous (nucleation in different phases) nucleation phenomena. The typical distribution recorded at the Monte Terminillo site showed a mono-mode centered around 60 nm. Figure 7c,e show the typical particle size profiles recorded at the Monte Terminillo site during rush hours and in the evening on workdays: It can be noted how both profiles are characterized by a single mode centered around 80-100 nm (with the full scale at 3 × 10 3 particles cm −3 ). A different situation is evidenced in the Rome sampling site. Figure 7e shows the typical profile of the aerosol size distribution recorded during rush hours on workdays (full scale up to 10 × 10 4 particles cm −3 ), i.e., during maximum intensity of autovehicular traffic (between 17:00-19:00). It can be noted that two modes, the first centered around 10 nm and the second between 30 and 110 nm. This second mode is an index of aged aerosol ("aged nucleation") [29]. A similar consideration can be drawn for Figure 7f (full scale up to 3 × 10 4 particles cm −3 ), where a typical size distribution occurring on the weekend evening in Rome is reported.
particles were continuously emitted into the atmosphere by autovehicular traffic but at the same time they aggregated and/or coagulated and increased both the Aitken mode (small contribution) and the accumulation mode. In the Monte Terminillo site, the average particle number percentages both in nucleation mode and in the Aitken nuclei remained quite constant, whereas there was a noticeable increase in accumulation mode. This increase was opposite to what happened in the urban site but leads to the same considerations. The explanation is that during weekends the Monte Terminillo site could be reached by tourist coaches: access was limited but emissions from bus traffic strongly affected the ratio. The daily particle number concentration in the three subdivisions shows more than 2-3 times the value in the urban site than in the mountain site (Figure 6b).
The difference between the two sites is evident in the aerosol size distributions reported in Figure 7a and 7b. Figure 7a and 7b show typical size distributions in the workday morning (07:00-09:00) at the Monte Terminillo site and in downtown Rome, respectively. In addition to the differences in the particle number (full-scale 6 × 10 3 for the Monte Terminillo site, full-scale 1.4 × 10 5 for downtown Rome), the size distribution profiles were different: In the urban area the mode between 7 and 14 nm was prevalent, meaning fresh particles emitted by vehicular traffic ("fresh nucleation"). This occurrence confirmed what has been reported in the literature; in the atmosphere most of the nanometer-sized particles (d ≤ 20 nm) mainly derive from gas-particle conversion processes [35], among which homogeneous (nucleation in vapor phase) and heterogeneous (nucleation in different phases) nucleation phenomena. The typical distribution recorded at the Monte Terminillo site showed a mono-mode centered around 60 nm. Figure  7c and 7e show the typical particle size profiles recorded at the Monte Terminillo site during rush hours and in the evening on workdays: It can be noted how both profiles are characterized by a single mode centered around 80-100 nm (with the full scale at 3 × 10 3 particles cm −3 ). A different situation is evidenced in the Rome sampling site. Figure 7e shows the typical profile of the aerosol size distribution recorded during rush hours on workdays (full scale up to 10 × 10 4 particles cm −3 ), i.e., during maximum intensity of autovehicular traffic (between 17:00-19:00). It can be noted that two modes, the first centered around 10 nm and the second between 30 and 110 nm. This second mode is an index of aged aerosol ("aged nucleation") [29]. A similar consideration can be drawn for Figure 7f

Dose Evaluation
Cumulative regional number doses (DC R (Δt)) and cumulative total number doses (DC Tot (Δt)) were from about 8 to 15 fold higher in downtown Rome than in the Terminillo area (Figure 3). At both sites DC R (Δt) and DC Tot (Δt), doses were higher in the 00:00-01:00 interval than in the 07:00-08:00 1h time interval, suggesting that the lower nocturnal atmospheric mixing layer outweighed the traffic emission reduction in Rome. In the Terminillo area, the shallow nocturnal PBL mixing height

Dose Evaluation
Cumulative regional number doses (D C R (∆t)) and cumulative total number doses (D C Tot (∆t)) were from about 8 to 15 fold higher in downtown Rome than in the Terminillo area (Figure 3). At both sites D C R (∆t) and D C Tot (∆t), doses were higher in the 00:00-01:00 interval than in the 07:00-08:00 1 h time interval, suggesting that the lower nocturnal atmospheric mixing layer outweighed the traffic emission reduction in Rome. In the Terminillo area, the shallow nocturnal PBL mixing height was synergic with the biomass burning aerosol emissions that were active until 00:00. At both sites, the highest contribution derived from particles deposited in the Al region, followed by the TB and H doses, on average respectively about 56%, 29%, and 15% of the total doses deposited into the respiratory system at both sites. Figure 4 shows that at both sites D R (d i , ∆t) and D Tot (d i , ∆t) were almost in the UFP size region (<100 nm). The average contributions of nucleation, Aitken, and accumulation mode particles to D C Tot (∆t) in downtown Rome and in the Terminillo area were respectively 55%, 37%, 8%, and 11%, 69%, and 20%. Nucleation mode particles brought about a higher contribution to total particle doses in Rome, due to the remarkable influence of freshly emitted traffic aerosol, typical of an urban area [36]. Finally, taking into account the literature, some considerations about the chemical composition and emissions sources in these areas could be drawn by measuring particle number and size. In urban areas, nucleation mode particles are mainly made of hydrocarbons, sulphates, and water and derives from vehicular exhaust dilution and cooling, whereas Aitken mode particles are mainly diesel soot particles [37,38] and accumulation mode particles include urban background and aged aerosol, road dust resuspension, and brake wear [39]. In the Terminillo area, nucleation and Aitken particles may derive from biogenic Volatile Organic Compounds (VOCs) emissions as well as from biomass burning, when this emission source is active. Such particles may grow into larger accumulation mode particles [40].

Conclusions
This study reported a deep comparison of the submicron aerosol fractions collected in a downtown big urban area (Rome), characterized by anthropogenic emissions, and a site (Monte Terminillo), where no anthropogenic emissions were recorded. Relevant differences both in the number particle concentrations and in the size aerosol distributions were found. Such divergences were essentially due to the different anthropogenic emissions between the two sites, particularly with autovehicular traffic. The study evidenced the different contribution of transportation to the local atmospheric pollution. This was quite evident in analyzing the aerosol profiles during nocturnal and diurnal profiles: The first ones were quite similar, whereas the second were different depending on the emission sources present in the two sites. These differences greatly reflected on human health and particularly on the cumulative regional/total number doses. A future study (actually, in preparation) will regard the synergic effect of transportation and domestic heating on aerosol composition, especially submicron fraction and UFPs, both in this remote location and in downtown Rome with the possibility to discern the two different contributions in aerosol fractions.