• Energy Research

A novel on-board system was tested to characterize size-resolved particle number emission patterns under real-world driving conditions, running in a EURO4 diesel vehicle and in a typical urban circuit in Madrid (Spain). Emission profiles were determined as a function of driving conditions. Source apportionment by Positive Matrix Factorization (PMF) was carried out to interpret the real-world driving conditions. Three emission patterns were identified: (F1) cruise conditions, with medium-high speeds, contributing in this circuit with 60% of total particle number and a particle size distribution dominated by particles >52 nm and around 60 nm; (F2) transient conditions, stop-and-go conditions at medium-high speed, contributing with 25% of the particle number and mainly emitting particles in the nucleation mode; and (F3) creep-idle conditions, representing traffic congestion and frequent idling periods, contributing with 14% to the total particle number and with particles in the nucleation mode (<29.4 nm) and around 98 nm. We suggest potential approaches to reduce particle number emissions depending on particle size and driving conditions. Differences between real-world emission patterns and regulatory cycles (NEDC) are also presented, which evidence that detecting particle number emissions <40 nm is only possible under real-world driving conditions.

Mean annual biomass burning contributions to the bulk particulate matter (PM(X)) load were quantified in a southern-European urban environment (Barcelona, Spain) with special attention to typical Mediterranean winter and summer conditions. In spite of the complexity of the local air pollution cocktail and the expected low contribution of biomass burning emissions to PM levels in Southern Europe, the impact of these emissions was detected at an urban background site by means of tracers such as levoglucosan, K(+) and organic carbon (OC). The significant correlation between levoglucosan and OC (r(2)=0.77) and K(+) (r(2)=0.65), as well as a marked day/night variability of the levoglucosan levels and levoglucosan/OC ratios was indicative of the contribution from regional scale biomass burning emissions during night-time transported by land breezes. In addition, on specific days (21-22 March), the contribution from long-range transported biomass burning aerosols was detected. Quantification of the contribution of biomass burning aerosols to PM levels on an annual basis was possible by means of the Multilinear Engine (ME). Biomass burning emissions accounted for 3% of PM(10) and PM(2.5) (annual mean), while this percentage increased up to 5% of PM(1). During the winter period, regional-scale biomass burning emissions (agricultural waste burning) were estimated to contribute with 7±4% of PM(2.5) aerosols during night-time (period when emissions were clearly detected). Long-range transported biomass burning aerosols (possibly from forest fires and/or agricultural waste burning) accounted for 5±2% of PM(2.5) during specific episodes. Annually, biomass burning emissions accounted for 19%-21% of OC levels in PM(10), PM(2.5) and PM(1). The contribution of this source to K(+) ranged between 48% for PM(10) and 97% for PM(1) (annual mean). Results for K(+) from biomass burning evidenced that this tracer is mostly emitted in the fine fraction, and thus coarse K(+) could not be taken as an appropriate tracer of biomass burning.

Biomass burning (BB) including forest, bush, prescribed fires, agricultural fires, residential wood combustion, and power generation has long been known to affect climate, air quality and human health. With this work we supply a systematic review on the health effects of BB emissions in the framework of the WHO activities on air pollution. We performed a literature search of online databases (PubMed, ISI, and Scopus) from year 1980 up to 2020. A total of 81 papers were considered as relevant for mortality and morbidity effects. High risk of bias was related with poor estimation of BB exposure and lack of adjustment for important confounders. PM10 and PM2.5 concentrations originating from BB were associated with all-cause mortality: the meta-analytical estimate was equal to 1.31% (95% CI 0.71, 1.71) and 1.92% (95% CI -1.19, 5.03) increased mortality per each 10 μg m-3 increase of PM10 and PM2.5, respectively. Regarding cardiovascular mortality 8 studies reported quantitative estimates. For smoky days and for each 10 μg m-3 increase in PM2.5 concentrations, the risk of cardiovascular mortality increased by 4.45% (95% CI 0.96, 7.95) and by 3.30% (95% CI -1.97, 8.57), respectively. Fourteen studies evaluated whether respiratory morbidity was adversely related to PM2.5 (9 studies) or PM10 (5 studies) originating from BB. All found positive associations. The pooled effect estimates were 4.10% (95% CI 2.86, 5.34) and 4.83% (95% CI 0.06, 9.60) increased risk of total respiratory admissions/emergency visits, per 10 μg m-3 increases in PM2.5 and PM10, respectively. Regarding cardiovascular morbidity, sixteen studies evaluated whether this was adversely related to PM2.5 (10 studies) or PM10 (6 studies) originating from BB. They found both positive and negative results, with summary estimates equal to 3.68% (95% CI -1.73, 9.09) and 0.93% (95% CI -0.18, 2.05) increased risk of total cardiovascular admissions/emergency visits, per 10 μg m-3 increases in PM2.5 and PM10, respectively. To conclude, a significant number of studies indicate that BB exposure is associated with all-cause and cardiovascular mortality and respiratory morbidity.

Combustion of residual forest biomass (RFB) derived from eucalypt (Eucalyptus globulus), pine (Pinus pinaster) and golden wattle (Acacia longifolia) was evaluated in a pilot-scale bubbling fluidised bed reactor (BFBR). During the combustion experiments, monitoring of temperature, pressure and exhaust gas composition has been made. Ash samples were collected at several locations along the furnace and flue gas treatment devices (cyclone and bag filter) after each combustion experiment and were analysed for their unburnt carbon content and chemical composition. Total suspended particles (TSP) in the combustion flue gas were evaluated at the inlet and outlet of cyclone and baghouse filter and further analysed for organic and elemental carbon, carbonates and 57 chemical elements. High particulate matter collection efficiencies in the range of 94-99% were observed for the baghouse, while removal rates of only 1.4-17% were registered for the cyclone. Due to the sand bed, Si was the major element in bottom ashes. Fly ashes, in particular those from eucalypt combustion, were especially rich in CaO, followed by relevant amounts of SiO2, MgO and K2O. Ash characteristics varied among experiments, showing that their inorganic composition strongly depends on both the biomass composition and combustion conditions. Inorganic constituents accounted for TSP mass fractions up to 40 wt%. Elemental carbon, organic matter and carbonates contributed to TSP mass fractions in the ranges 0.58-44%, 0.79-78% and 0.01-1.7%, respectively.

Biomass burning (BB) is a significant source of particulate matter (PM) in many parts of the world. Whereas numerous studies demonstrate the relevance of BB emissions in central and northern Europe, the quantification of this source has been assessed only in few cities in southern European countries. In this work, the application of Positive Matrix Factorisation (PMF) allowed a clear identification and quantification of an unexpected very high biomass burning contribution in Tuscany (central Italy), in the most polluted site of the PATOS project. In this urban background site, BB accounted for 37% of the mass of PM10 (particulate matter with aerodynamic diameter<10 μm) as annual average, and more than 50% during winter, being the main cause of all the PM10 limit exceedances. Due to the chemical complexity of BB emissions, an accurate assessment of this source contribution is not always easily achievable using just a single tracer. The present work takes advantage of the combination of a long-term daily data-set, characterized by an extended chemical speciation, with a short-term high time resolution (1-hour) and size-segregated data-set, obtained by PIXE analyses of streaker samples. The hourly time pattern of the BB source, characterised by a periodic behaviour with peaks starting at about 6 p.m. and lasting all the evening-night, and its strong seasonality, with higher values in the winter period, clearly confirmed the hypothesis of a domestic heating source (also excluding important contributions from wildfires and agricultural wastes burning).

Abstract Seven fuels (four types of wood pellets and three agro-fuels) were tested in an automatic pellet stove (9.5 kWth) in order to determine emission factors (EFs) of gaseous compounds, such as carbon monoxide (CO), methane (CH4), formaldehyde (HCHO), and total organic carbon (TOC). Particulate matter (PM10) EFs and the corresponding chemical compositions for each fuel were also obtained. Samples were analysed for organic carbon (OC) and elemental carbon (EC), anhydrosugars and 57 chemical elements. The fuel type clearly affected the gaseous and particulate emissions. The CO EFs ranged from 90.9 ± 19.3 (pellets type IV) to 1480 ± 125 mg MJ−1 (olive pit). Wood pellets presented the lowest TOC emission factor among all fuels. HCHO and CH4 EFs ranged from 1.01 ± 0.11 to 36.9 ± 6.3 mg MJ−1 and from 0.23 ± 0.03 to 28.7 ± 5.7 mg MJ−1, respectively. Olive pit was the fuel with highest emissions of these volatile organic compounds. The PM10 EFs ranged from 26.6 ± 3.14 to 169 ± 23.6 mg MJ−1. The lowest PM10 emission factor was found for wood pellets type I (fuel with low ash content), whist the highest was observed during the combustion of an agricultural fuel (olive pit). The OC content of PM10 ranged from 8 wt.% (pellets type III) to 29 wt.% (olive pit). Variable EC particle mass fractions, ranging from 3 wt.% (olive pit) to 47 wt.% (shell of pine nuts), were also observed. The carbonaceous content of particulate matter was lower than that reported previously during the combustion of several wood fuels in traditional woodstoves and fireplaces. Levoglucosan was the most abundant anhydrosugar, comprising 0.02–3.03 wt.% of the particle mass. Mannosan and galactosan were not detected in almost all samples. Elements represented 11–32 wt.% of the PM10 mass emitted, showing great variability depending on the type of biofuel used.

Source apportionment of atmospheric PM1 is important for air quality control, especially in urban areas where high mass concentrations are often observed. Chemical analysis of molecular inorganic and organic tracer compounds and subsequently data analysis with receptor models give insight on the origin of the PM1 sources. In the present study, four source apportionment approaches were compared with an extended database containing inorganic and organic compounds that were measured during an intensive sampling campaign at urban traffic and urban background sites in Barcelona. Source apportionment of the combined database, containing both inorganic and organic compounds, was compared with more conventional approaches using inorganic and organic databases separately. Traffic emission sources were identified in all models for the two sites. The combined inorganic and organic databases provided higher discrimination capacity of emission sources. It identified aerosols generated by regional recirculation of biomass burning, secondary biogenic organic aerosols, harbor emissions, and specific industrial emissions. In this respect, this approach identified a relevant industrial source situated at NE Barcelona in which a waste incinerator plant, a combined-cycle power plant, and an industrial glass complex are located. Models using both inorganic and organic molecular tracer compounds improve the source apportionment of urban PM.

This study evaluates geochemical and oxidative potential (OP) properties of the respirable (finer than 4 μm) fractions of 22 powdered coal samples from channel profiles (CP4) in Chinese mined coals. The CP4 fractions extracted from milled samples of 22 different coals were mineralogically and geochemically analysed and the relationships with the OP evaluated. The evaluation between CP4/CP demonstrated that CP4 increased concentrations of anatase, Cs, W, Zn and Zr, whereas sulphates, Fe, S, Mo, Mn, Hf and Ge decreased their CP4 concentrations. OP results from ascorbic acid (AA), glutathione (GSH) and dithiothreitol (DTT) tests evidenced a clear link between specific inorganic components of CP4 with OPAA and the organic fraction of OPGSH and OPDTT. Correlation analyses were performed for OP indicators and the geochemical patterns of CP4. These were compared with respirable dust samples from prior studies. They indicate that Fe (r = 0.83), pyrite (r = 0.66) and sulphate minerals (r = 0.42) (tracing acidic species from pyrite oxidation), followed by S (r = 0.50) and ash yield (r = 0.46), and, to a much lesser extent, Ti, anatase, U, Mo, V and Pb, are clearly linked with OPAA. Moreover, OPGSH correlation was identified by organic matter, as moisture (r = 0.73), Na (r = 0.56) and B (r = 0.51), and to a lesser extent by the coarse particle size, Ca and carbonate minerals. In addition, Mg (r = 0.70), B (r = 0.47), Na (r = 0.59), Mn, Ba, quartz, particle size and Sr regulate OPDTT correlations. These became more noticeable when the analysis was done for samples of the same type of coal rank, in this case, bituminous.