Experimental study on atmospheric pollutant emissions from heating systems, in Italy Promoted by the Italian Ministry for the Environment, in coperation with : The Lombardy Region, the Piedmont Region, the Italian Oil Union, Assopetroli, ENEA, CTI, SSC, IPASS The original text of this study, in Italian, has been summarized and translated by Franco Del Manso (Italian Oil Union) and reviewed by Tiziano Pignatelli (ENEA) 1
Extended Abstract General Considerations Atmospheric pollution due to anthropogenic activities is threatening public health, especially in urban areas with high density of population; for this reason there are numerous initiatives underway in Italy, both at the regional and national levels, to limit emissions coming from different sources: heating, vehicular traffic and industrial activities. In this perspective, a study was undertaken with the aim of providing, with sound scientific support, the measures addressing the domestic sector (for heating purposes). The study was promoted by the Ministry of the Environment, with the participation of Italian oil industry associations (Unione Petrolifera, Assocostieri and Assopetroli), the Region of Lombardy, the Region of Piedmont, ENEA (Agency for the New Technology, Energy and the Environment), CTI (Italian standardization body for industry and civil sectors), Stazione Sperimentale per i Combustibili (SSC Research institute on fuels) and IPASS (Perugia s University Consortium for environment and sustainable development). For the purposes of the study, a Scientific Technical Committee was set up (TSC), by a decree, by the Ministry of the Environment, and the activities, subject of the testing activities, were entrusted to and carried out by the Stazione Sperimentale per i Combustibili (SSC) and IPASS. Before this study, the emission factors for combustion of the main fuels, used in the sector, were the only data available in the literature. However, wherever these data were available, they were neither very accurate nor up-to-date; moreover, there was a lack of data, in literature about the more recently introduced fuels. On the contrary, this study was designed to evaluate the global behavior of heating unit/fuel systems, in relation to environmental pollution, and thus, to indicate emission factors, associated with their use, for the assessment of the impact on the environment and the human health. The study looked at the fossil fuels which are the most widely used in Italy, for heating purposes, as well as some alternative and innovative types of fuels such as: natural gas, gas oil, biodiesel, emulsions of gas oil and water, low sulphur fuel oil (sulphur < 0.3%), emulsions of low sulphur fuel oil (sulphur < 0.3%) and water, and a blend of low sulphur fuel oil (sulphur < 0.3%) with biodiesel. The comparison aimed at evaluating the quantity and quality of atmospheric emissions of the various combustion apparatuses, while operating with the above mentioned fuel types. 2
The testing conducted, both in the laboratory and on site, dealt with boiler types representing the real distribution of domestic heating units in Italy. The results of these lab and on site tests provide important information concerning the combustion effects, produced at urban and suburban scale, places where the most of the residential housing and commercial activities are concentrated (and where, consequently, the use for heating purposes is higher) and certain industrial installations. There is a clear correlation between the environmental impact of domestic heating and the distribution of heating typologies (also because of the low levels of emission from chimneys), as well as with the time period under consideration (coldest week, January, winter, cold semester, and year). As a consequence, different and more diluted effects are produced when up-scaling from the municipal level, up to the provincial, regional and national levels. Data from the Literature, Qualitative Evolution and Fuel Consumptions Results of an accurate survey of the available data in literature revealed a lack of reliable and up-to-date data: the s of emission factors found, for the main fuels, were sometimes poorly significant since those data were related with far years, when the fuel chemical-physical properties were very different from today s fuel characteristics, and when technologies, nowadays considered obsolete, were implemented in the boilers. At the beginning of the seventies, the increased awareness about the potential strong impact of the air pollution on the public health, in fact, led the refinery sector to improve their products. For example, the diesel fuel, which has been used, longly, rich in sulphur, today contains less than 0.2%; similarly, fuel oil, again characterized, in the past, by high levels of sulphur, metals, asphaltenes, nowadays contains dramatically lower levels of these substances (<0.3% of sulphur), with significant environmental benefits. At the same time, new fuel products have been developed and made available on the market, with a reduced environmental consequences. Among these new fuels, the bio-diesel, a no sulphur fuel, mainly derived by the processing of vegetable oils, can be considered as a renewable energy source. And the emulsions of water in gasoil and fuel oil, allow the application of a simple technology, which has been around for decades: the addition of water to hydrocarbons to reduce certain types of pollutants. Manufacturers of heating units (boilers) also provided their contribution by improving combustion technology with innovations of great importance for the environment. As regards the consumptions, the variety of products, used in the domestic sector for heating purposes, has deeply changed in the last years toward the gaseous fuels. The general trend shows a progressive decrease in the use of liquid fuels with respect the gaseous fuels. As shown in Tab. 1, the consumption of natural gas is currently predominant (72.1%), and continuously growing, while there is a reduction in the consumption of gas oil (14.4%), along with other liquid fuels (1.5%). 3
In the annex 1, the main fuel characteristics are reported. Table 1 - Fuels tested in laboratory and on site facilities Fuel Market share Natural Gas 72 % Heating Gas oil 14 % Water gas oil emulsion 0,5 % Biodiesel 1 % Fuel Oil 2 % Water fuel oil emulsion 0,5 % Fuel oil biodiesel mix 0,5 % Experimental Results of Emission Testing Classification of the Heating Units Examined The experiment was conducted by ranking the population of examined heating units into three categories according to their nominal thermal power: - class I: 25-75 kw - class II: 75-200 kw - class III: > 200 kw In particular, the single house boilers are in class I: almost exclusively natural gas is used in these boilers. The other two classes generally comprise centralized heating boilers, in buildings with an average of 20-25 up to 50 apartments. The distribution of heating systems, installed in Italy, is shown in Table 2. Table 2 - Heating systems installed in Italy Number % Centralized Systems 4.400.000 19,7 Single House boiler 12.600.000 56,4 Other Systems 5.300.000 23,9 Total 22.300.000 100 The tests were carried out in the Laboratory (test area set up at the Stazione Sperimentale per i Combustibili, SSC) and, on site, on 23 units located in the regions Lombardy, Piedmont, Liguria, Veneto and Emilia Romagna. Measurements were taken under steady state and transient conditions. At the same time, some measurements (5% of the total) were made by SSC and IPASS in order to verify the reliability of the results, obtained with the different methodologies. 4
On site and Lab Tests under Steady State Conditions Emission factors resulting by the study revealed a good level of agreement between the s measured on site and the s monitored in the lab, as it may be derived from a comparison of average, maximum and minimum s, for each fuel. Discrepancies were found, in some cases, on CO and NO x data only. However, these pollutants are strongly dependant on the O2 concentrations, in the flues gases, as well as on the combustion temperature. As regards the correlation between emission factors and the boiler power (size), a general increase in CO emission factors was observed, in class I boilers, both on site and in the lab. However, the identification of a general trend, for classes II and III, has been difficult. Experimental emission factors, as a function of nominal power, in fact, show a decreasing trend, as the power of the unit increases. The level of SO2 measured in the emissions from gas oil was found to be in line with their sulphur content. This fact acquires significance in the case of gas oil emulsions, where diesel oil (i.e. sulphur free fuel) is used. Although s for total PM (particulate matter) and for PM10 (particles with a diameter of less than 10 µm) were higher for fuel oils, for all the examined fuels they have been found below the threshold of 50 mg/nm3. The complete set of the new emission factors measured is reported in the following tables (Tables 3 7). Table 3 - Total PM Emission Factor Fuel Laboratory (g/gj) on site (g/gj) Natural Gas < 0,1 < 0,1 0,45 Heating Gas oil 0,1 0,29 0,85 1,53 Water gas oil emulsion 0,18 0,83 0,1 1,09 Biodiesel 0,1 0,19 0,1 0,14 Fuel Oil 5,79 5,88 5,12 7,32 Water fuel oil emulsion 1,62 2,84 3,53 4,78 Fuel oil biodiesel mix 2,67 3,40 3,47 3,99 Table 4 - PM 10 Emission Factor Fuel Laboratory (g/gj) on site (g/gj) Natural Gas < 0,10 - Heating Gas oil 0,10 0,14 - Water gas oil emulsion 0,20 0,80-5
Biodiesel <0,1 - Fuel Oil 2,90 4,83 - Water fuel oil emulsion 1,68 3,25 - Fuel oil biodiesel mix 2,67 3,03 - Table 5 - NOx Emission Factor Fuel Laboratory (g/gj) on site (g/gj) Natural Gas 23-97 16-72 Heating Gas oil 37 51 54-69 Water gas oil emulsion 38 47 40 52 Biodiesel 30 43 35-56 Fuel Oil 162-167 162 234 Water fuel oil emulsion 136 144 178 214 Fuel oil biodiesel mix 144 158 225-243 Table 6 - CO Emission Factor Fuel Laboratory (g/gj) on site (g/gj) Natural Gas 1 31 4-24 Heating Gas oil 1,6 4,9 1,5 3,7 Water gas oil emulsion 1 5 0,6 17,2 Biodiesel 0,8 9,50 0,9 67,6 Fuel Oil 3,2 3,5 1,8 4,3 Water fuel oil emulsion 2,4 10,9 1,5 12,1 Fuel oil biodiesel mix 3,5 1,7 1,1 1,2 Table 7 - VOC Emission Factor Fuel Laboratory (g/gj) on site (g/gj) Natural Gas 0,1 2,17 0,22 1,75 Heating Gas oil 0,1 0,4 0,1 0,17 Water gas oil emulsion 0,1 0,29 0,1 0,63 Biodiesel 0,1 1,63 0,39 3,5 Fuel Oil 0,1 0,17 0,23 0,26 Water fuel oil emulsion 0,17 0,44 0,12 0,23 Fuel oil biodiesel mix 0,14 0,25 0,25 0,31 On site and Lab Tests under Discontinuous Conditions All the considerations mentioned above refer to steady state operating conditions. Given the fact that all the heating units operate intermittently, in real use conditions, (on/off switching, with variable periods of steady state use, for class I boilers, while the steady state periods are longer, for class II and III units, where the interruptions are activated when the system achieves the highest thermal conditions). The so-called transient conditions were also analysed, i.e., periods when the unit is being continuously 6
started up or shut down. The duration of the periodic start up and shut down cycles has been selected in order to maintain the start up phase long enough to achieve the steady state conditions, similarly to a real operating condition. Along with being an absolute novelty in these kind of studies, the results of the tests also provide concrete evidence that the type of boiler is one of key factors, when evaluating the pollutant emissions, in the final use. The analysis of the substances detected in the emissions, during these phases, show a great influence with regard to CO e COT, of low influence in the steady state phases, while no additional information come about NOx and SO2. Moreover, the relative shortness of these cycles does not allow significant sampling in the gravimetric measurement of dusts. CO Emissions were relatively higher with gasoil, water in gas oil emulsions and biodiesel, especially when used in class I boilers; intermediate s were observed for fuel oils, while for natural gas, variety of effects was found when passing from class II and III units, with relatively low average emission levels, compared to the single house boilers (class I); among the latter class, type B and type C (low NOx boilers) provide CO s comparable with the gas oils. More noticeable indications have emerged from the analysis of unburnt hydrocarbons: from the comparison, always on a relative scale, the the lowest contribution resulted attributed to the liquid oils. The intermediate s have been shown by the gas oils, when used in class II and III units, but with significant maximum s when used in class I boilers, while COT emissions produced by natural gas have been found, consistently, at average high levels. Final Considerations A comparative analysis has been performed on different fuels, used in the domestic sector for heating purposes, looking at aspects connected, not only to atmospheric pollution, but also at the risks associated with their use and relative costs, in order to verify the possibility to highlight the relative contribution, to the total impact, of each fuel. Due to the great number of variables and because of their complexity, the comparative analysis was limited to the end uses of fuel s life cycle, only. In particular, the distribution of various combustion technologies, the size of the units, the quality of fuels and their market shares have been the main aspects considered in this study. The study also has highlighted the effect of the operating transient conditions have on emissions, hence the effect of the on-off cycles frequency. These, in turn, depend on a considerable number of factors, such as the type of boiler, its size in relation to the thermal demand, the water capacity in the boiler, the type and characteristics of the boiler s thermostat, and other factors as well: all these factors are usually neglected in the designing of the unit, following accepted standard rules, which, therefore, cannot take into consideration the pecularities of different contexts in the installations. It must be remembered that the designing of boiler units is a much more complex activity than it may appear. The designing of the building has also a considerable influence on 7
environmental effect, in relation to its energy consumption. Along with the designing, the state of obsolescence and maintenance of the combustion apparatus has also a great influence on emissions As well, the renewal of the existing population of building boilers, is of greatest importance. Finally, we would like to point out the significant contribution of this experimental study in relation both to the scope of the experimental campaign, especially the on site tests, and its accurate representation of the current situation in Italy, in terms of the types fuels and heating units (boilers) being used. In this sense the emission factors resulting from the study provide an important contribution, complementing and updating what is available in the literature. Table 8 - Emission factor resulting from the study PM PM 10 NO x CO CO 2 SO 2 VOC (g/gj) (g/gj) (g/gj) (g/gj) (Kg/GJ) (g/gj) (g/gj) Natural Gas Laboratory 0,14 0,14 97,40 31,13 60,18 1,39 2,17 Field 0,45-72,32 24,43 56,87 1,39 1,75 Gheating gasoil Laboratory 0,29 0,14 51,40 4,86 73,02 45,30 0,40 Field 1,53-68,93 3,67 75,29 79,15 0,17 Water- gasoil emulsion Laboratory 0,83 0,80 46,95 5,02 78,26 15,30 0,29 Field 1,09-51,56 17,24 77,29 13,72 0,53 Biodiesel Laboratory 0,19 0,14 42,94 9,50 71,55 1,38 1,63 Field 0,14-56,41 67,59 73,56 1,58 3,50 Fuel Oil Laboratory 5,88 4,83 1.886,53 3,50 72,12 105,82 0,17 Field 7,32-232,61 4,28 5,72 115,50 0,26 Water- Fuel oil emulsion Laboratory 2,84 3,25 144,01 10,93 76,53 107,13 0,44 Field 4,78-214,00 12,01 77,92 135,79 0,23 Biodiesel fuel oil mix Laboratory 3,40 3,03 158,25 3,51 73,36 96,03 0,25 Field 3,99-242,62 1,18 74,38 101,60 0,31 The study has shown how the situation in the domestic sector, heating boilers, has gradually evolved, over time, towards a condition of lower environmental burden, although, at a local level, and during the winter season, it remains a significant source of pollution. In the light of what the European Energy Agency has proposed (EEA 2003), with regards the impact of secondary particulate on air quality parameters, the extent to which the heating systems, in the domestic sector, contributes to total levels of PM10 must be better evaluated, based upon the contribution of sulphur oxides and sulphur emissions to the formation of secondary PM. 8
Properties of Fuels tested in laboratory and on site Annex 1 Fuel oils properties Laboratory Fuel oils sample n. On site sample n. Statistical analysis 1 2 1 2 3 average min max Water % m/m 0,20 0,05 0,05 0,50 <0.05-0,05 0,50 - Total Sediment % m/m <0.01 <0.01 <0.01 0,02 <0.01-0,02 0,02 - Viscosity 50 C mm²/s 95,78 95,93 97,07 97,63 104,7 98,22 95,78 104,70 3,703 Viscosity 50 C E 12,65 12,67 12,82 12,89 13,82 12,97 12,65 13,82 0,48 Carbon Residue % m/m 3,95 4,55 3,76 2,73 5,12 4,022 2,73 5,12 0,898 Ash % m/m 0,011 0,006 0,008 0,002 0,029 0,011 0,002 0,029 0,010 Nitrogen % m/m 0,22 0,20 0,20 0,30 0,25 0,23 0,20 0,30 0,042 Sulphur % m/m 0,24 0,25 0,24 0,23 0,28 0,25 0,23 0,28 0,019 Kcal/kg 10.041 10.028 10.022 10.035 9.918 10.008 Value 9.918 10.041 51,26 Value kj/kg 42.040 41.985 41.960 42.040 41.52 41.910 41.525 42.040 218 Nickel Vanadium mg/kg <15 17,2 <15 <15 48 - <15 48 - PCB mg/kg <4 <4 <4 <4 <4 - <4 - - Water in Gas oils emulsion properties Laboratory Water gas oils On site sample n. Statistical analysis sample n. emulsion 1 2 1 2 3 4 average min max Water % m/m 11,75 12,40 11,40 11,40 11,40 10,90 11,54 10,90 12,40 0,500 K.F. Density 15 kg/m3 860,4 853,0 859,7 859,7 859,7 862,30 859,13 853,00 862,30 3,169 C Viscosity mm²/s 4,602 4,265 3,537 3,537 3,959 4,153 4,009 3,537 4,602 0,421 40 C Flash point C >100 >100 69 69 >100 >100 69 >100 Pour point C -6-24 -24-18 -24-19,20-24,00-6,00 7,823 Sulphur % m/m 0,028 0,030 0,024 0,024 0,029 0,026 0,027 0,024 0,030 0,003 kcal/kg 8.940 8.776 8.770 8.770 8.941 8.823 8.836,67 8.770 8.941 82,84 kj/kg 37.43 36.74 36.71 36.719 37.43 36.941 36.998 36.719 37.435 347 9
Water in Fuel oils emulsion properties Laboratory Water Fuel oils On site sample n. Statistical analysis sample n. emulsion 1 1 2 3 average min max Flash point C >100 >100 >100 <100 - - - - Water % m/m 13,0 9,5 13,5 16,0 13,0 9,5 16,0 2,677 Total Sediment % m/m <0.01 0,01 <0.01 0,01 - - 0,01 - Viscosity 50 C mm²/s 130,20 94,51 133,1 146,8 126,15 94,51 146,8 22,303 Viscosity 50 C E 17,16 12,48 17,57 19,38 16,65 12,48 19,38 2,941 Ash % m/m 0,005 0,025 0,004 0,060 0,024 0,004 0,060 0,026 Sulphur % m/m 0,20 0,23 0,21 0,24 0,22 0,20 0,24 0,018 Nitrogen %m/m 0,18 0,15 0,17 0,15 0,16 0,15 0,18 0,015 Lower calorific kcal/kg 8638 9.025 8.545 8.305 8.628,2 8.305 9.025 299,40 Lower calorific kj/kg 36165 37.78 35.78 34.77 36.124 34.77 37.78 1.251 Nickel +Vanadium mg/kg <15 26,3 <15 19-19 26,3 - PCB mg/kg <4 <4 <4 <4-0 0 - Biodiesel Biodiesel Properties Laboratory On site sample n. Statistical analysis sample n. 1 2 1 2 3 4 mean min max Viscosity 40 C mm²/s 4,100 4,345 4,321 4,367 4,402 4,400 4,323 4,100 4,402 0,113 Carbon Residue % m/m 0,17 n.d. n.d. 0,21 0,26 0,18-0,17 0,26 0,040 Sulfated Ash % m/m 0,001 <0.001 0,01 0,005 0,003 0,001-0,001 0,009 0,003 Water K.F. % m/m 0,043 0,032 0,09 0,096 0,079 0,048 0,065 0,032 0,096 0,027 Total contamination mg/kg 116 55 172 68 237 12 110 12 237 82,948 Acidity mg KOH/g 0,41 0,42 0,35 0,35 0,35 0,37 0,375 0,35 0,42 0,032 CFPP C -10 n.d. n.d. -7-7 -5 - -10-5 2,062 Pour point C -6 n.d. n.d. -15-10 -6 - -15-6 4,272 kcal/kg 8.937 8.940 8.95 8.950 8.950 8.970 8.949 8.937 8.970 11,554 kj/kg 37.410 37.42 37.5 37.470 37.470 37.560 37.466 37.410 37.560 51 Sulphur % m/m 0,001 <0.001 0 0,003 0,002 0,009-0,001 0,009 0,003 Ester content % m/m 92,9 94,7 92,3 95,2 95 90,7 93,467 90,7 95,2 1,801 Monoglyceride % m/m 0,74 0,76 0,51 0,63 0,63 0,35 0,603 0,35 0,76 0,153 Diglyceride % m/m 0,2 0,12 0,1 0,1 0,09 0,08 0,115 0,08 0,20 0,044 Triglyceride % m/m 0,12 0,07 0,04 0,01 0,01 0,02 0,045 0,01 0,12 0,043 Free Glycerol % m/m <0,01 0,019 0,03 0,005 0,004 <0,01-0,004 0,025 0,010 10