Atmospheric Environment (v.40, #2)

The heterogeneous reactions of NO2 and OH radicals with polycyclic aromatic hydrocarbons (PAHs) naturally present in diesel particulate exhaust (NIST SRM 1650a) have been investigated. Using a fast flow reactor, relative degradation rate constants have been determined by following the decays of particulate-PAH concentrations vs. the reaction time. Quantitative analyses have been performed by gas chromatography coupled to mass spectrometry detection using internal standards. Relative rate constants show a significant structure reactivity effect for the reaction of PAHs with NO2. Pyrene and benzo(a)pyrene are the most reactive with NO2 whereas all PAHs studied present similar reactivities with OH within uncertainties. All PAHs appeared to be approximately four orders of magnitude more reactive with OH than with NO2. These rate constants confirm that the OH heterogeneous reaction will be the dominant atmospheric loss process of PAHs compared to that with NO2. Our data are compared to previous literature studies, concerning both the particulate and the gas phases. This work demonstrates that the reactivity of PAHs in the gas phase would be significantly larger than when associated with carbonaceous particulate substrates.
Keywords: PAHs; Diesel particles; Reactivity; Kinetics;

Development and evaluation of personal respirable particulate sampler (PRPS) by Seung Joo Lee; Philip Demokritou; Petros Koutrakis; Juana M. Delgado-Saborit (212-224).
This paper presents the development, laboratory evaluation, and field tests of a personal respirable particulate sampler (PRPS). The PRPS is designed as a personal sampling system to collect particulate matter (PM0.5, PM1.0, PM2.5, PM4.5, and PM10) and gaseous pollutants, including O3, SO2, and NO2. It operates at a flow rate of 5.0 LPM and consists of five selectable impaction stages (with cutpoints of 10, 4.5, 2.5, 1.0, and 0.5 μm), a backup filter, and two diffusion passive samplers. In each impaction stage, particles are collected onto a polyurethane foam (PUF) substrate. This substrate, using no adhesive, was shown to have minimum particle bounce and re-entrainment. A backup 37 mm Teflon membrane filter is used downstream to collect particles smaller than the cutoff diameter of the final impaction stage. The impaction stage cutpoints were characterized in the laboratory using artificially generated polydisperse aerosols. Particle losses for each stage were found to be acceptably low. The performance of the PRPS was also compared with that of a collocated micro-orifice cascade impactor (MOI) and real-time particle sizing instruments (SMPS/APS) in laboratory experiments using artificially generated particles. The size distributions measured by the PRPS were found to be much closer to those measured by the real-time particle sizing instruments than to those measured by the MOI. A field PM intercomparison study was also conducted using the PRPS and three reference samplers, the Harvard Impactor (HI), the USEPA PM2.5 Well Impactor Ninety Six (WINS), and the Harvard Personal Environmental Monitor (Harvard PEM) sampler. The PM10, PM2.5, and sulfate concentrations measured by PRPS were in a very good agreement with those obtained from the reference samplers.
Keywords: Personal sampling system; Particulate matter; Impaction; PUF;

Toxic trace elements and organic compounds in the ambient air of Kabul, Afghanistan by Luigi Antonello Di Lella; Stefano Loppi; Giuseppe Protano; Francesco Riccobono (225-237).
To assess the ambient air quality in Kabul, we measured the contents in tree bark samples of 17 chemical elements by ICP–MS, polychlorinated dibenzo-p-dioxins and furans (PCDD/Fs) by GC–MS, polychlorinated biphenyls (PCBs) by GC–ECD and polycyclic aromatic hydrocarbons (PAHs) by HPLC. While there were rather high levels of sulfur (up to 2277 μg g−1), the heavy element contents were rather low (i.e. Pb was in the range 3.12–5.00 μg g−1), even though a slight peak value was recorded in the area of most intense traffic (Pb up to 13.91 μg g−1). Slight traces of organohalogen compounds, i.e. PCBs and PCDD/Fs, were detected (ΣPCBs=1.184–6.318 ng g−1; ΣPCDD/Fs=1.42–2.38 pg g−1). Highly chlorinated congeners, i.e. OCDD/Fs and penta-, hexa- and hepta-CBs, dominated the profiles of these compounds. Only three- and four-ringed PAH compounds were detected, but at very low levels comparable with the natural background. The slightly anomalous trace element values and the profiles of PCDD/Fs and PAHs determined in the bark samples suggest a close relationship with emissions from automotive traffic and the domestic burning of wood and fossil fuels. The presence of distinct but unimportant PCB emission sources can also be inferred.
Keywords: Chlorinated organic compounds; Kabul; Polycyclic aromatic hydrocarbons; Heavy elements; Tree bark;

The regional distributions of, and inhalation cancer risks for, a group of volatile organic hydrocarbons (VOCs) consisting of aromatic hydrocarbons (AHCs), volatile organic halogenated compounds (VOHCs), and carbonyl compounds (CCs) were determined in an industrial harbor city (Shimizu, Japan) in the summer and winter of 2000. The concentrations of most of these VOCs tended to be higher in winter than in summer, but this trend was not observed for CCs. Significant correlations ( p < 0.05 ) were observed between the concentrations of most of the AHCs in both summer and winter, and between the concentrations of certain AHCs and CCs in summer, which suggests common emission sources and similar atmospheric behavior for these compounds. The results of principal-component analysis indicated that vehicle exhaust and industrial activity were the predominant emission sources of the VOCs. To determine the local spatial distributions and emission sources for the VOCs, we created contour maps for the monitored VOCs. These maps indicated that the local distributions varied dramatically depending on the characteristics of the VOCs, such as their emission sources and fates. The lifetime cancer risks (LCRs) calculated from the corresponding VOC concentrations were significantly higher ( p < 0.0001 ) in winter than in summer. Risk was derived mainly from benzene, carbon tetrachloride, formaldehyde, and acetaldehyde in both summer and winter; the spatial distribution of high LCR levels was similar to the distribution of high benzene, formaldehyde, and acetaldehyde concentrations.
Keywords: VOCs; Outdoor air; Regional distribution; Emission source; Cancer risk;

Ergosterol as a biomarker for the quantification of the fungal biomass in atmospheric aerosols by Arthur P.S. Lau; Alex K.Y. Lee; Chak K. Chan; Ming Fang (249-259).
The prevailing warm and humid climate in subtropical cities favors fungal growth in the ambient environment. Fungal growth has implications for public health because fungal cells, spores and their metabolites are allergenic and potential health hazards. In this regard, better and quicker alternatives than the available sampling and species identification methods are needed for quantifying fungal communities in atmospheric aerosols. In this study, the fungal membrane ergosterol was used as a biomarker for assessing the abundance and mass loading of fungi in atmospheric aerosols. Gas chromatography-mass spectrometry (GC-MS) was utilized for quantification of this biomarker in fine (PM2.5) and coarse (PM2.5–10) particulates collected by high volume samplers simultaneously at a rural site and an urban site in Hong Kong. The geometric means of the total ergosterol concentrations at the rural and urban sites were 120.2 and 93.9 pg m−3 in the PM10 (calculated as the sum of PM2.5 and PM2.5–10) particulates. The significantly higher ergosterol loading at the rural site was related to the vegetation coverage around the sampling site. Ergosterol loading was higher in the autumn at both sites, which correlated with seasonal drops in the relative humidity below 70%. Approximately 65–66% of the ergosterol in PM10 is associated with fine particulates, indicating that atmospheric fungi likely lead to chronic respiratory symptoms. The mass loading of the fungal spores on the dry mass was on the order of 10–102  ng m−3. The geometric mean concentration of the fungal spores was estimated as 46 and 36 spores m−3, which was one-sixth of the measured viable samplings of 292 and 247 CFU m−3 at the rural and urban sites, respectively. This underestimation leads to the need for establishing proper conversion factors from conditions identical to or simulating the study system of interest when markers are quantified and estimated for microbial mass loading in ambient aerosols. This study was the first to utilize the fungal biomarker ergosterol in monitoring and assessing ambient fungal loading and prevalence in atmospheric aerosols for a period of eight months. It confirms that biomarkers can provide quantitative information on microbial communities in atmospheric aerosols.
Keywords: Biomarker; Bioaerosols; Ergosterol; Fungal biomass; Airborne particles and size distribution;

The vertical distributions of three heavy metals: Hg, Pb and Cd were determined in 3 cores sampled from two ombrotrophic bogs in the west of Ireland, one at Knockroe Co. Mayo, and the second at Letterfrack National Park, Co. Galway. Core chronologies were established using 210Pb dating techniques and were checked with fallout radionuclides from weapons testing. Variations were found in metal concentrations and cumulative inventories of each of the metals within each site and between the two sites. Maximum accumulation rates of the anthropogenically derived elements Hg, Pb, and Cd, were found in peat sediments dated between 1950 and 1970s at both sites.Pb and Hg accumulation rates are slightly lower than those found in similar studies from remote sites in Europe. Hg accumulation rates are fairly similar to those found in peatlands in America. Unlike the Pb and Hg concentration profiles, the Cd concentration profiles at the Letterfrack site were dominated by a surface enrichment, thought to be due to biological cycling of Cd in the peat. However Cd accumulation rates calculated at the Knockroe site are lower than those observed in Eastern Europe. Local meteorological conditions at the sites chosen for this study may account for the lower concentrations profiles observed when compared with some of the European studies. The similarity between the timing of the increase in metal accumulation rates in peat bogs in Northern America and this study could indicate that long range transportation of trace metals from Northern America may be occurring. Lead accumulations in the surface peat sediments (1993–1996) were between 1.5–3.0 mg m−2  yr−1 and 4–5 mg m−2  yr−1 at Knockroe and Letterfrack, respectively. Mercury accumulation rates for the same period at Knockroe were found to be between 6–11 μg m−2  yr−1, and between 19–24 μg m−2  yr−1 at Letterfrack. A greater variation in surface Cd accumulation rates was observed at both sites, with surface layer accumulation rates varying from 25 to 50 μg m−2  yr−1 at Knockroe and between 166 and 405 μg m−2  yr−1 at Letterfrack. Higher metal concentrations were found at the Letterfrack site, which are most likely due to local sources and the history of the site.
Keywords: Lead; Mercury; Cadmium; Ombrotrophic peatlands; Western Europe;

Characterization of phosphorus in the aerosol of a coastal atmosphere: Using a sequential extraction method by Hung-Yu Chen; Tien-Hsi Fang; Martin R. Preston; Saulwood Lin (279-289).
Particulate aerosol was sequentially extracted by citrate/dithionite/bicarbonate and acetic acid/sodium acetate buffers so as to isolate phosphorus associated with oxides/hydroxides (Fe-P) and Acet-P, respectively. Total phosphorus (TP) and total inorganic phosphorus (TIP) contents were analyzed separately and the difference between them used to calculate the organic phosphorus (OP) content. The difference between TIP and the sum of Fe-P and Acet-P was assumed to be the detrital apatite (Det-P). The results of this study indicate that phosphorus species show a clear monthly variability that is related to the source strengths of the crustal weathering and biological blooming in winter and spring, respectively. There are highly significant covariations between the TP, TIP and Det-P, which suggests that these derive from similar sources and share transport mechanisms. The percentage concentrations of OP show that phosphorus tends to be associated with organic particles in warm periods when biological activity is greatest.
Keywords: Phosphorus; Inorganic phosphorus; Organic phosphorus; Aerosol; Sequential extraction method;

Spatial distribution of PM2.5 associated organic compounds in central California by Lynn R. Rinehart; Eric M. Fujita; Judith C. Chow; Karen Magliano; Barbara Zielinska (290-303).
As part of the California Regional PM2.5/PM10 Air Quality Study, annual average concentrations of PM2.5 associated organic compounds were measured using filter samples collected every sixth day for 24 h for 1 year. During this time many central California sites had 24-h PM2.5 mass concentrations exceeding the National Ambient Air Quality Standard of 65 μg m−3. The highest concentration of 175 μg m−3 was recorded in Fresno on January 1, 2001. Organic speciation of 20 sites within and near the central California valley provides a measure of the spatial differences of emission sources through the use of organic molecular markers. Additionally, it provides an opportunity evaluate their utility at ambient concentrations to measure the influence of emission sources at rural and urban locations. The most abundant particulate phase organic compounds identified were polar organic compounds. Sugar anhydrates, molecular markers of wood combustion, constituted the largest weight fraction of total carbon, followed by alkanoic acids, and alkanedioic acids. Local emission sources such as residential wood combustion, gasoline and diesel vehicles were distinguished by unique molecular markers and found to vary among the annual average sites. The annual average concentrations of individual organic species within compound classes were observed to be highly correlated for many of the central California sites. The two Fresno sites were highly correlated to each other especially with respect to polycyclic aromatic hydrocarbons, but not well correlated to other sites.
Keywords: Levoglucosan; Polycyclic aromatic hydrocarbons; Polar organic compounds; Organic speciation; Biomass combustion; Trimethylsilylation; Fresno; CRPAQS;

Weekly periodicity of environmental variables in Phoenix, Arizona by Shade T. Shutters; Robert C. Balling (304-310).
Though there is no known meteorological cause for weekly cycling of environmental variables, weekly cycles have been discovered at local to global scales, particularly in areas affected by human urbanization. To uncover such cycles in Phoenix, AZ, and to highlight possible mechanisms for their existence, data from several public domain sources were collected and analyzed for cycles in three categories of variables: meteorological, pollution, and human activity measured as vehicle traffic flows. Results indicated that many meteorological and pollution variables do exhibit weekly periodicity and that these cycles are likely due to the weekly pattern of human traffic flows. Atmospheric concentrations of O3 and NO X gases exhibit a high degree of negative correlation, supporting recent research that suggests anthropogenic NO X gases are effective scavengers of ozone in urban cores. Results further suggest that vehicle-generated NO X gases may be a significant generator of atmospheric nitrate particulates. Finally, both traffic flow and NO X gas concentrations display a strong correlation with wind speed in the urban core, though this study does not speculate on a mechanism for this relationship.
Keywords: Automobile emissions; NO X gas; Urban atmosphere; Nitrate aerosols; Ozone;

A paper that was previously published in Atmospheric Environment [Pohjola et al., 2003. Modelling of the influence of aerosol processes for the dispersion of vehicular exhaust plumes in street environment. Atmospheric Environment 37(3), 339–351.] contained a description of the properties of particulate matter in exhaust emissions and in urban background air. This data contained particle number concentrations and particle masses for four size classes. The mass and number concentrations of particulate matter exhaust emissions were computed correctly. However, a subsequent checking has shown that the values of particle masses of urban background air that were used in the numerical computations were one to two orders of magnitude too small, in relation to the corresponding particle number concentrations of urban background air. This has affected the numerical results on particle sizes, and produced an erroneous temporally decreasing particle size in some cases; this is notable particularly for the Aitken mode. The calculations have therefore been repeated, replacing the erroneous data with the correct data. However, the main conclusions drawn by Pohjola et al. (2003) are shown still to be valid, using the corrected input data.
Keywords: Particulate matter; Aerosol dynamics; Coagulation; Condensation; Dilution;

Removal of ozone on clean, dusty and sooty supply air filters by Marko Hyttinen; Pertti Pasanen; Pentti Kalliokoski (315-325).
The removal of ozone (O3) on supply air filters was studied. Especially, the effects of dust load, diesel soot, relative humidity (RH), and exposure time on the removal of O3 were investigated. Some loss of O3 was observed in all the filters, except in an unused G3 pre-filter made of polyester. Dust load and quality influenced the reduction of O3; especially, diesel soot removed O3 effectively. Increasing the RH resulted in a larger O3 removal. The removal of O3 was highest in the beginning of the test, but it declined within 2 h reaching almost a steady state as the exposure continued. However, the sooty filters continued to remove as much as 25–30% of O3. Up to 11% of O3 removed participated in the production of formaldehyde. Small amounts of other oxidation products were also detected.
Keywords: Filter; Ozone; Dust; Diesel soot;

The estimation of the ratio β = T L / T E between Lagrangian and Eulerian time scales is an important problem, both in basic turbulence theory and in atmospheric pollution dispersion studies. The present paper aims at contributing to this research field by examining planetary boundary layer (PBL) turbulence data generated by an LES model in strongly convective, buoyancy-dominated and neutral stability conditions, respectively. Standard deviations of wind velocity, integral time scales, autocorrelation functions and spectra, for the three wind components, both in the Lagrangian and Eulerian frames, have been calculated. The Hay and Pasquill hypothesis which says that, at a first approximation, Eulerian and Lagrangian autocorrelation functions and spectra are identical after rescaling of the time axis by a factor β , was verified for the lower time lags of autocorrelation functions, and for frequencies describing the energy-containing eddies. Also. the usual assumption that β is inversely proportional to the turbulence intensity i, i.e. β = γ / i , was verified. Other four relationships, all relating β and i were tested, finding that all could be confidently used in practical applications. It was found that γ is a function of the different turbulent regimes, being greater in the strongly convective PBL, than in the neutral one. Values found are in agreement with previous experimental results and theoretical evaluations.
Keywords: LES; Atmospheric Boundary layer; Lagrangian and Eulerian time scale ratio; Lagrangian and Eulerian statistics; Kolmogorov constant;

Winter comparison of TEOM, MiniVol and DustTrak PM10 monitors in a woodsmoke environment by Simon Kingham; Michael Durand; Teresa Aberkane; Justin Harrison; J. Gaines Wilson; Michael Epton (338-347).
This paper compares PM10 mass concentrations data generated by three co-located particulate monitors, the tapered element oscillating microbalance (TEOM40) (operating at 40 °C), MiniVol and DustTrak, operating outdoors during winter woodsmoke air pollution (Christchurch, New Zealand). TEOM40 data were adjusted to adjTEOM40 with an algorithm derived from a co-located TEOM40 and HiVol at a second field site. Corrected and corrected-loge-transformed data were analyzed with reduced major axis regression. Logged adjTEOM40 and DustTrak data correlated best ( r 2 = 0.8 1 ) but the DustTrak required a substantial correction factor to make data comparable in real terms. Consistent over-recording by the DustTrak and under-recording by the TEOM40 were consistent with previously published work. The MiniVol did not correlate well with other instruments ( r 2 = 0.6 2 and 0.53 against the adjTEOM40 and DustTrak, respectively) but reasons for this cannot be ascertained.
Keywords: Air pollution; Monitoring networks; Ambient monitoring; Particulate matter;

During a fixed-route including (i) surface walk on streets, (ii) underground spaces (corridors, vestibules and platforms) and (iii) underground trains of the Prague Metro Rail System (PMRS) a fast response portable photometer was used to measure concentrations of PM10 (particulate matter <10 μm). On the basis of more than 100 monitored trips the PMRS may be characterised as more polluted than the streets above. On average, the highest PM10 concentration was recorded inside the Metro trains (113.7 and 1.44), the second highest in the underground spaces of stations (102.7 and 1.29), followed by outdoor environment (74.3 and 0.85), expressed in μg m−3 and in dimensionless relative normalised units, respectively. In spite of the statistically significant difference ( p < 0.001 ) between ambient and underground PM10 levels recorded by the photometer, the ambient concentrations were tightly associated with those from the underground spaces ( r 2 = 0.820 ) as well as from the Metro trains ( r 2 = 0.774 ) . The correlation between concentrations from both underground microenvironments was even higher ( r 2 = 0.964 ) indicating a common source of aerosol inside the PMRS. Since also reasonable correlation was found between fixed site monitors (FSM) and both ambient and underground photometer PM10 levels a strong influence of surface traffic-related particulate pollution on the underground air quality is suggested.
Keywords: Air quality; Urban transport; PM10; Personal exposure; DustTrak;

A new drift–flux model for particle distribution and deposition in indoor environments is developed. Gravitational settling and deposition are examined thoroughly and the model is applied to simulate particle distribution and deposition in a ventilated model room. The turbulent airflow field is modeled with the renormalization group (RNG) kε turbulence model. For the particulate phase, the concentration field is divided into two regions, the core region and the concentration boundary layer. The concentration distribution in the core region is obtained by solving a three-dimensional particle transport equation. Deposition flux towards the wall is determined with a semi-empirical particle deposition model. With the feasibility of the proposed Eulerian model, the influences of the other deposition mechanisms can be captured easily into the model. The model is validated experimentally and a good agreement between numerical and experimental results is found. Inferring from the result, it shows that the well-mixed assumption cannot hold for coarse particles. The model presented in the current work provides an efficient and reliable tool for investigation of spatial and temporal particle concentration in enclosures and the result will be very useful for improving our current understanding of human exposure assessment.
Keywords: Computational fluid dynamics; Drift–flux model; Ventilation; Indoor particles; Mixing;

Odour in the surroundings of Copenhagen Airport by J. Fenger; P. Løfstrøm; M. Winther; U. Kousgaard; A. Oxbøl (368-374).
A method for determining odour from airports has been developed. It is based on sampling of emission from a typical engine run at different modes followed by olfactometric determination of odour by a panel of test persons. The determinations are transformed to odour emissions and extended to other engines by table values for hydrocarbon emissions. With activity accounts from the airport an emission inventory with a resolution of 50 m is set up. Finally the concentrations are calculated with dispersion modelling to give 99% percentiles of 1-h average values. These can approximately be transformed to 1-min values. The method is used at Copenhagen Airport in Kastrup.
Keywords: Odour; Airport; Olfactometry;

The correlation between photocatalytic oxidation performance and chemical/physical properties of indoor volatile organic compounds by Kuo-Pin Yu; Grace W.M. Lee; Wei-Ming Huang; Chihcheng Wu; Shinhao Yang (375-385).
In this study, six species of volatile organic compounds (VOCs), n-hexane, iso-butanol, toluene, p-xylene, m-xylene, and mesitylene, were selected as the target pollutants to investigate how the photocatalytic oxidation (PCO) performance is related to their physical/chemical properties. The PCO kinetics were well fit by a Langmuir–Hinshelwood (L–H) model for bimolecular surface reaction and competitive adsorption at gas flow rate above 1000 mL min−1 (reaction-controlling region), where the gas-phase mass transfer effect was negligible. The rate constants of PCO for toluene, p-xylene, m-xylene, and mesitylene ranged from 1.22 to 4.00 μmol m−2  s−1, and were proportional to VOC-hydroxyl radical rate constant (k OH). The Langmuir adsorption constants of the four aromatic VOCs investigated and water ranged from 0.95 to 1.35 ppm−1 and from 5.61×10−3 to 1.44×10−3  ppm−1, respectively. A strongly linear positive relationship was found between the reciprocal of the Langmuir adsorption constants of the four aromatic VOCs investigated and Henry's Law constants. Conversely, the reciprocal of Langmuir adsorption constants of water showed a strong negative relationship with Henry's Law constants (in units kPa m3  mol−1) of the four aromatic VOCs investigated. The relationships noted above were not found between different classes of VOCs (n-hexane, iso-butanol, and aromatic VOCs investigated). The percentage of residual intermediates (partially oxidized and incompletely mineralized organic compound from the primary VOCs) decreased as the inlet VOCs concentration decreased.
Keywords: Aromatic hydrocarbons; Langmuir–Hinshelwood model; Henry's Law constant; Hydroxyl radical; Photocatalytic rate constant; CO2 yield rate;

Exposure visualisation of ultrafine particle counts in a transport microenvironment by S. Kaur; R.D.R. Clark; P.T. Walsh; S.J. Arnold; R.N. Colvile; M.J. Nieuwenhuijsen (386-398).
An increasing number of studies indicate that short-term peak exposures, such as those seen in the transport microenvironment, pose particular health threats. Short-term exposure can only be sufficiently characterised using portable, fast-response monitoring instrumentation with detailed summaries of individual activity. In this paper, we present an exposure visualisation system that addresses this issue—it allows the simultaneous presentation of mobile video imagery synchronised with measured real-time ultrafine particle count exposure of an individual. The combined data can be examined in detail for the contribution of the surrounding environment and the individual's activities to their peak and overall exposure. The exposure visualisation system is demonstrated and evaluated around the DAPPLE study site in Central London using different modes of transport (walking, cycling, bus, car and taxi). The video images, synchronised with the exposure profile, highlight the extent to which ultrafine particle exposure is associated with traffic density and proximity to pollutant source. The extremely rapid decline in concentration with increasing distance away from the pollutant source, such as from the main street to the backstreets, is clearly evident. The visualisation technique allows these data to be presented to both technical audiences and laypersons thus making it an effective environmental risk communication tool. Some exposure peaks however are not obviously associated with any event recorded on video—in these cases it will be necessary to use advanced dispersion modelling techniques to investigate meteorological conditions and other variables influencing in-street conditions to identify their possible causes.
Keywords: Exposure visualisation; Ultrafine particle counts; Transport microenvironment; Real-time measurements; Video monitoring;