CONVENTION ON LONG-RANGE TRANSBOUNDARY AIR POLLUTION Expert Group on Techno-economic Issues (13th meeting) The need for better control systems and fast technological renewal in small combustion appliances burning wood Stefano Caserini s.caserini@arpalombardia.it
Outline: Domestic wood contribution to emissions and air quality Emission factors and critical points Trade off GHG /PM Regulations Conclusions
Italy Lombardy 9 Mil. inhabitants 56 Mil. inhabitants
Global mean tropospheric nitrogen dioxide (NO 2 ) vertical column density between January 2003 and June 2004, as measured by the SCIAMACHY instrument on ESA's Envisat http://www.esa.int/esacp/sem340 NKPZD_index_1.html Lombardy Region
Wood combustion has been identified in the Italian and Lombardy emission inventory as an important source of primary PM10 emissions 100% 80% 60% 40% 20% no fuel others fuel oil natural gas wood kerosene LPG gasoil refinery gas diesel coal Lombardy 2005 emissions from all SNAP groups by fuel 0% PM2.5 PM10 TSP SO2 NOx NMVOC CO CO2 eq gasoline INEMAR: AtmospheRic EMission INventory Data available on the INEMAR web page for a public review www.ambiente.regione.lombardia.it/inemar/inemarhome.htm
The role of wood combustion as a source of PM has been confirmed by studies on the presence of wood tracers in air samples. Although secondary aerosol formation is more relevant than primary emissions for both coarse and fine PM fraction, wood burning is relevant also in urban areas and strongly contributes to high peak of PM. PM from biomass burning, reconstructed by levoglucosan (a marker for biomass combustion) concentration, compared to total PM measured. PM (µg/m³) 120 100 80 60 40 20 0 PM (reconstructed) PM (measured) (winter 2005) B.Fontana Sondrio Cantù Milan Mantova Source : Fermo and Vecchi, PARFIL Project, Lombardy Region
Emissions in Lombardy by SNAP group (year 2005) 100% 80% 60% 40% 20% 0% SO2 NOx VOC CH4 CO CO2 N2O NH3 PM2.5 PM10 TSP Combustion in energy and trasformation industries Combustion in manufacturing industry Extraction and distribution of fossil fuels Road Transport Waste treatment and disposal Other sources and sinks Non-industrial combustion plants Production processes Solvent and other product use Other mobile sources and machinery Agriculture
Domestic wood combustion in small installations is the dominant contribution to SNAP groups 2 - domestic heating Lombardy 2005 emissions in the SNAP groups 2 - domestic heating, by fuel 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% PM2.5 PM10 TSP SO2 NOx NMVOC CO CO2 eq fuel oil natural gas wood kerosene LPG gasoil
Critical point in the assessment of emission from small wood combustion installations - Type of appliance - System of measure (hot / cold) - Emissions during transient conditions (start up) - Burning cycle - Type of wood
Since 2003 the emission inventory consider four or five type of wood burning appliances Lombardy 2005 emissions in the SNAP groups 2 - domestic heating, 100% 90% by SNAP level III (type of appliances) BAT system burning pellets 80% 70% 60% 50% 40% 30% 20% 10% 0% PM2.5 PM10 TSP SO2 NOx NMVOC CO CO2 eq Pellets plant or BAT system burning wood Innovative low emission system and boiler Traditional oven, closed fireplace or insert Open fireplace Residential boilers < 50 MW
Emission factors are highly dependent on the type of appliance EF from a literature survey (EEA Emission Inventory Guidebook + other sources) PM10 NO X NMVOC SO 2 CO PAH g GJ -1 g GJ -1 g GJ -1 g GJ -1 g GJ -1 mg GJ -1 Open fireplace 500 70 5,650 13 5,650 280 Traditional oven, closed fireplace or insert Innovative low emission system and boiler Pellets plant or BAT system burning wood 250 70 1,130 13 5,650 280 150 60 560 13 2,260 280 50 65 85 13 800 0.2 + Automatic pellets stove BAT Source: Literature survey (EEA Emission Inventory Guidebook + other sources)
Distribution of wood combustion systems in Italy Innovative stove 4% Automatic pellets stove 3% Closed fireplace 20% Open fireplace 45% Traditional stove 28%
% number of appliances 55% 27% 27% 35% 16% 19% 7% 3% 5% 6% 36% 36% 46% 19% 19% 27% 4% 3% 7% 1% 72% 16% 6% 3% 2% 66% 18% 11% 1% 3% 55% 19% 19% 5% 2% 36% 36% 19% 7% 1% Open fireplace Traditional stove Closed fireplace Innovative stove Automatic pellets stove Source: Caserini S. et al. (2007) "New insight into the role of wood combustion as key PM source in Italy and in Lombardy region 15th US-EPA International Emission Inventory Conference. Raleigh, NC
Wood consumptions by appliance (%) 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Piemonte Liguria Valle d'aosta Lombardy Veneto Trentino Friuli Emilia Romagna Toscana Marche Umbria Sardegna Lazio Abruzzo Molise Campania Puglia Sicilia Calabria Basilicata Open fireplace Traditional stove Closed fireplace Innovative stove Automative pellets stove Source: Caserini S. et al. (2007) "New insight into the role of wood combustion as key PM source in Italy and in Lombardy region 15th US-EPA International Emission Inventory Conference. Raleigh, NC
EF from natural gas and gas oil are much lower than the best EF from wood burning PM10 NO X NMVOC SO 2 CO PAH g GJ -1 g GJ -1 g GJ -1 g GJ -1 g GJ -1 mg GJ -1 Open fireplace 500 70 5,650 13 5,650 280 Traditional oven, closed fireplace or insert Innovative low emission system and boiler Pellets plant or BAT system burning wood 250 70 1,130 13 5,650 280 150 60 560 13 2,260 280 50 65 85 13 800 0.2 Natural gas 0.2 50 5.0 0.5 25 n.a. Gas oil 5.0 50 3.0 100 20 75 Fuel oil 40 150 10 150 16 75 Source: Literature survey (EEA Emission Inventory Guidebook + other sources)
Filterable /condensable PM measurement A lot of old PM emission factors from wood combustion rely only on the filterable fraction of PM, thus underestimating the true emission factor. It is important to consider the condensable fraction, material that is vapor phase at stack conditions, but which condenses and/or reacts upon cooling and dilution in the ambient air to form solid or liquid PM immediately after discharge from the stack. PM emission factor (g/gj) Pellets Hot smoke 27 32 C 60 Wood stove 66 93 Source: Stazione Sperimentale Combustibili, 2007 - Parfil Project, Lombardy Region
Methodologies for the measurement of condensable fraction have been defined only in the last years 2006-09 27 CEN/TC 295/WG5 N 70/2006 Residential solid fuels burning appliances. Test methods
Emissions during transient conditions (start up) 700 Open fireplace - with beech-wood 100 600 PM T flue gas 90 cumulat PM% Stoked fire 80 PM (mg/nm 3 ), T ( C) 500 400 300 200 Refilling 70 60 50 40 30 cumulat PM% 20 100 10 0 0 500 1000 1500 2000 2500 3000 3500 s 0 Source: Angelino E. et al. (2008) Experimental investigations of the influence of transitory phases on small-scale wood combustion emissions, in press
Emissions during transient conditions (start up) Closed fireplace - with beech-wood. PM (mg/nm 3 ), VOC (ppm) 1400 1200 1000 800 600 400 200 PM VOC T flue gas Refilling Stoked fire 200 180 160 140 120 100 80 60 40 20 T C 0 0 500 1000 1500 2000 2500 3000 3500 0 s Source: Angelino E. et al. (2008) Experimental investigations of the influence of transitory phases on small-scale wood combustion emissions, in press
For road transport emission factor assessment, different reference cycles has been defined Vehicle speed (km/h) 70 60 50 40 30 20 10 ARTEMIS urban Urban dense Free-flow urban Congested, stop Congested, Flowing low speed stable 0 0 100 200 300 400 500 600 700 800 900 1000 Cycle tim e (s)
Wood Burning cycle 50 Temperature near a closed fireplace 45 Temperature ( C) 40 35 30 25 20 15 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 0.00 2.00 4.00 Source: Caserini S. preliminary data 6.00 8.00 10.00
Outline: Domestic wood contribution to emissions and air quality Emission factors and critical points Trade off GHG /PM Regulations Conclusions
There are net savings of GHG emissions using biomass instead of traditional fuels, notwithstanding the supply distances. PM, PAH and VOC emissions from small domestic devices need to be diminished, otherwise the widespread use of biomass could be critical in reaching more stringent PM standards. LCA results in terms of impact categories per tonne of dry biomass Scenario/device Greenhouse effect Human toxicity Air acidification Photochemical oxidant formation kg CO 2 eq kg 1.4-DB eq kg SO 2 eq kg C 2 H 4 eq 1 Open fire -80.4 2870 3.05 44.5 2 Traditional wood stove -698 2801 2.26 11 3 Low emission wood stove -932 2782 1.17 5.17 4 Pellet stove BAT -1080 96.3 1.63 0.762 5 CHP plant - 8 MW -1020-106 -2.28-0.305 6 CHP plant - 8 MW (SRF) -850-69 -1.26-0.195 7 CHP plant - 100 MW -914-116 -2.68-0.227 Source: Caserini S. et al. LCA of domestic and centralized biomass combustion: the case of Lombardy (Italy), submitted
Controls Primary measures Secondary measures Cyclone separators Electrostatic precipitators Catalytic converters SCI manufacturer are skeptic about their real efficiency on the field
Regulations 1/2
Regulations 2/2
EF from SCI targets proposal for the Po Valley area PM10 Concentration in smokes mg/m^3 (13 %O2) PM10 Emission factors * g/gj mg/kwh Natural gas 0.2 Gasoil 5 Fuel oil 40 Literature data Open fireplace 740 500 1805 Traditional oven, closed fireplace or insert 370 250 903 Innovative low emission system and boiler 222 150 542 Pellets plant or BAT system burning wood 104 70 253 BAT system burning pellets 44 30 108 Denmark (21/12/2006) (cold) 900 600 Denmark (21/12/2006) (cold) 900 600 Austria (law limits - art. 15 B-VG) 74 50 181 Austria - Limit for incentivations 30 20 72 Germany (Munich) 75 51 183 Target proposal (13% O2) 24 16 59 Target proposal (11% O2) 30 16 59 * It was considered a PCI equal to 17 GJ/t (dry beech wood) and a specific production of dry smokes equal to 11.5 m^3/kg at 13 % O2 Resulting conversion factors: 0.676 (g/gj) / (mg/m^3); 2.44 (mg/kwh) / (mg/m^3); 0.277 (g/gj) / (mg/kwh)
Conclusions /1 Positive aspects The number of regulation that limits the emissions of particulate matter produced by small devices burning biomass is increasing Some regulation distinguishes by type of sampling (cold / hot, regimen / transitory) There are also some interesting (and probably new) procedures about test certificates, controls, verification and the manufacturer s registry.
Conclusions /2 Negative aspects There are no limits clearly distinguished by typologies (e.g. open fireplaces / inserts / stoves / pellets). The limits appear to be high with respect to the potential of existing technologies, in particular for automatic stoves Some limit for small plants seems too high Also limits for centralised plants (i.e. 150 mg/m^3, 74 g/gj) seems to be in any way too high, and it doesn t take into account the slightest possibility of abatement systems, that could be already available for plants of a certain size. This kind of prescription, with these limits, would yield very few benefits in the Po Valley context. Better control systems for PM, PAH and COV are needed
Thank You
The use of biomass in district heating systems, that use flue gas treatment system, can cause a consistent reduction of impacts for all the considered categories. Actual role of biomass combustion in Lombardy in terms of emissions Biomass consumption Greenhouse effect PM 10 emissions CO VOC t year -1 t CO 2 eq year -1 t year -1 t year -1 t year -1 Open fire 417,747-23,511 2,588 29,242 1,462 Traditional wood stove 1,582,700-773,307 4,902 110,789 5,539 Low emission wood stove/boiler 480,035-328,986 661 10,200 834 Pellet stove BAT 19,518-19,059 10 200 0 Subtotal residential devices 2,500,000-1,144,864 8,161 150,431 7,835 CHP plants?10 MW 70,000-49,980 2 7.4 2.5 CHP plant 100 MW 230,000-136,643 0.45 34.4 0.36 Subtotal CHP plants 300,000-186,623 2.4 41.7 2.8 TOTAL 2,800,000-1,331,487 8,164 150.473 7,838 Source: Caserini S. et al. (2008) LCA of domestic and centralized biomass combustion: the case of Lombardy (Italy)