Development of the Consequence Model for Hazardous Substances (ESCAPE) for the Needs of Rescue Authorities Jaakko Kukkonen, Kari Riikonen, Juha Nikmo, Markku Seppänen, Antti Hellsten Finnish Meteorological Institute Fire protection fund
Introduction Starting point: Computer program ESCAPE, Expert System for Consequence Analysis using a PErsonal computer The program addresses the accidental releases of toxic or flammable gases. The model has been applied in rescue services in Finland > 20 a. The model is a useful and user-friendly tool of assessment in the analysis of accidents.
Application of models and numerical results Emergency response planning Hazard and risk analysis, including analysis of past accidents Training and improving preparedness for accidents Inspection and surveillance of installations The model results are most useful in preparing in advance for accidents not commonly used operationally.
The financiers and users of ESCAPE Ministry of the Interior, Finland The Finnish Defence Forces Fire Protection Fund, Finland Inspecta, Finland Finnish Meteorological Institute In Finland: Rescue services (22) City of Helsinki Environment Centre Ekokem Neste Radiation and Nuclear Safety Authority University of Applied Sciences (EVTEK and HAMK) Internationally: Environment Protection Agency, Lithuania Institute of Physics of Vilnius, Lithuania State Pollution Control Authority, Norway Toray Plastics. Malaysia University of Hamburg, Germany
Development of the Consequence Model for Hazardous Substances (ESCAPE) for the Needs of Rescue Authorities Aims of the project in 2013-2015 Fire protection fund General aim: More accurate and reliable evaluations of the consequences of the accidents of hazardous materials, for rescue authorities. Specific aims: 1. To update the dispersion models according to the state-of-the-art scientific knowledge, and evaluate model predictions against experimental data. 2. Improve the usability of the model in rescue services, to be more userfriendly, robust and versatile. E.g., on-line connection to weather prediction models would simplify the user s task. 3. Extend the model to work in a variety of computer environments.
Development of the Consequence Model for Hazardous Substances (ESCAPE) for the Needs of Rescue Authorities Work packages and schedule of the project 1. Dispersion models will be replaced by improved versions (2013) - update of the models (done) - evaluation of the models against experimental data (partly done) 2. Automatic use of weather prediction models (2014) - available weather prediction models: HARMONIE, AROME, ECMWF (the latter also global) - the user has also possibility to input measured weather data 3. Extension of the model to work in different operating environments (2015) - at present the model works only in Windows environment - environment-independent program version (Windows, Unix, Linux, Mac, mobile phones) - easy-to-use net application to operative rescue services => simplifies and diversifies the model usage
Management group Janne Koivukoski, chairman, SM Jaakko Kukkonen, proj.coordinator, FMI Juha Nikmo, FMI Antti Hellsten, FMI Markku Seppänen, FMI Arto Jäppinen, PvTT Markku Mäkelä, ESPL Tiina Santonen, TTL Work group Kari Riikonen, FMI Juha Nikmo, FMI Jaakko Kukkonen, FMI Mikko Aalto, FMI Kaisa Korpi, FMI Minna Rantamäki, FMI Roope Tervo, FMI Markku Seppänen, FMI Mikko Routala, LUPL SM FMI ESPL LUPL PVTT TTL Ministry of the Interior Finnish Meteorological Institute Fire and rescue services of Etelä-Savo Fire and rescue services of Länsi-Uusimaa Defence Forces Technical Research Centre Finnish Institute of Occupational Health
The processes included in the ESCAPE model Two-phase jet Release Source term Dispersion Effects Pool formation Catastrophic release X X X X X X length scale: release source term heavy gas passive dispersion 10 m 20-60 m 100-1000 m < 20 km
A block chart of the ESCAPE model Database on the properties of substances Meteorological data and description of the release Accident databases Boiling liquid expanding vapour explosion Instantaneous release Spreading and vaporisation of a liquid pool A release through a pipe or a breach Two-phase jet Heavy cloud and passive dispersion Evaluation of doses Output data, MapInfo Heavy plume and passive dispersion
Dispersion model for denser than air gas clouds Meteorology and the transport of the cloud Entrainment of air: atmospheric turbulence internal turbulence of the cloud mechanical entrainment u Gravity slumping, terrain Thermodynamical and chemical processes: pollutant and entrained air phase transitions (liquid vapour) heat flux from the ground
A model for liquid pool spreading and evaporation Atmospheric conditions Discharge Evaporation of pool: vapour pressure and latent heat flow structure Heat transfer atmosphere - pool: convection radiation Heat transfer ground - pool: conduction convection Pool spreading: gravity resistance
Current status Input data for weather conditions and terrain in ESCAPE Input data for wind, temperature and humidity Input data for determining atmospheric stability Input data for determining ground roughness
Input data of the release conditions in ESCAPE: Release type: continuous or instantaneous Chemical substance Mass of substance stored Etc. If release conditions result in liquid pool formation:
An example of the graphical output of ESCAPE: an assumed release of chlorine from a ruptured pipe in a container vessel. Input data for this case Legend for concentrations Map Genimap Oy
The model is directly applicable for these substances Acetone Chlorine Methane Acetonitrile Cumene N-butane Acrylic acid Ethane N-pentane Acrylonitrile Ethylene oxide Phosgene Ammonia Freon 12 Propane Benzene Hexane Propylene oxide Bromine Hydrogen cyanide Sulphur dioxide Butadiene Hydrogen fluoride Sulphur trioxide Butane Hydrogen sulphide Toluene Carbon disulphide Isobutane Vinyl chloride Carbon tetrachloride Methanol Additional species can be added.
Evaluation of ESCAPE against discharge trials using liquefied ammonia and sulphur dioxide (Sweden, 1982 and 1984). The Swedish National Defence Research Institute conducted ten ammonia field trials in Landskrona, Sweden in April 1982. The objective: study the flow of pressurized ammonia through pipes when released from the liquid phase of a vessel. Measurements: tank temperature and pressure, pipe exit pressure and mass flux for the ammonia trials. and five sulphur dioxide field trials in Skelleftehamn, Sweden in August 1984. The objective: measure the mass flow rate of SO 2 through an aperture and a pipe when released from the liquid space. Measurements: tank temperature and pressure, pipe exit pressure and mass flux for the sulphur dioxide trials. Refs.: Nyrén et al., 1983; Nyrén and Winter, 1983, 1986.
Mass flux error (%) Evaluation of the liquid and two-phase outflow models Of ESCAPE against the discharge trials (Sweden, 1982 and 1984) + 10 % Mass flux deviation (%) - 10 % 10 8 6 4 2 0-2 -4-6 ammonia, L = 2.636 m ammonia, L = 2.136 m -8 sulphur dioxide, L = 2 m sulphur dioxide, L = 0 m -10 350 400 450 500 550 600 650 700 Vessel pressure (kpa) Vessel pressure (kpa) -> The relative uncertainty of the discharge modelling compared with data is less than 10 % The deviation of predicted and measured mass flux in % in the ammonia and sulphur discharge trials, against vessel pressure in kpa. L = pipe length.
Evaluation of ESCAPE against the FLADIS dispersion trials Experiments on continuous releases of liquefied ammonia were carried out in the project FLADIS Field Experiments funded by EU (Nielsen et al., 1994 and 1997; Nielsen and Ott, 1996), at Landskrona in1993-1994. The objective: study dispersion in the jet stage, the heavy plume and passive dispersion. Measurements: concentration was measured on arcs of sensors distributed across the plume at distances of 20, 70 and 240 m downwind from the source; three field campaigns with a total of 27 trials were conducted, of which the data of 16 has been prepared for distribution.
Modelled concentration (mole-%) Evaluation of ESCAPE against the FLADIS field trials 5 Pointwise scatter, N = 122 4 Modelled concentration 3 2 1 Trial 16 Trial 24 0 0 1 2 3 4 5 Observed concentration (mole-%) Observed concentration Scatter plot of point-wise mean concentration for the FLADIS trials 16 and 24.
Evaluation of ESCAPE against the Thorney Island field trials The Heavy Gas Dispersion Trials at Thorney Island were organized by the Health & Safety Executive (UK) and were carried out between 1982 and 1984 (McQuaid and Roebuck, 1985; McQuaid, 1985 and 1987). The objective: study the dispersion of fixed-volume, isothermal clouds. Measurements: dose and arrival and departure times of the cloud were measured on arcs of sensors distributed across the cloud at distances of 100, 200, 220, 300, 350, 400, 450, 500, 600, 700 m downwind from the source. three continuous release trials, which divided into two phases. - phase I: 6 trials on uniform, unobstructured ground. - phase II: 10 trials in the presence of obstacles. the release gas was a mixture of dichlorodifluoromethane (CCl 2 F 2 ; freon-12) and nitrogen (N 2 ).
Downwind distance (m) An example of the evaluation of ESCAPE against the Thorney Island field trials, instantaneous releases 60 50 Thorney Island, trial 008 40 30 20 10 visual observations this study 0 0 5 10 15 20 25 30 35 40 45 50 Time (s) The cloud centroid position against time as determined visually from the overhead photographs taken from the Thorney Island trial 008 (Prince et al., 1985) and predicted by the ESCAPE model.
Cloud height (m) Comparison of the predictions of ESCAPE and the visual observations 300 250 200 150 100 50 0 600 Houston, Texas, 1976 Pasquill E NH3 u = 2 m/s m = 19 000 kg T = 10 C 500 t = 1 min C max = 7700 mg/m3 400 300 200 Downwind distance (m) t = 0 min 3 C max = 49000 mg/m 100 0 50 m A photograph taken about one minute after the Houston tanker crash (Fryer and Kaiser, 1979).
The automated combined use of ESCAPE and measurement data in a previous research project (TiTiMaKe) User interface TEKES Master server The system includes: Rescue services sensor network FMI dispersion computations Aalto population models Univ. Eastern Finland master server Visualisation tools for the end user Sensors Dispersion Population TiTiMaKe: Development of modelling methods for situational awareness ToMoVaKe: Open Platform For Environmental Monitoring ESCAPE dispersion modelling visualized. ToMoVaKe user interface Dimenteq Oy 2011.
The planned ESCAPE net application The system provides results for various users (top row) The core application includes ESCAPE combined to on-line meteorological modelling Met. duty officer A distributed system User #1 User #nn. (cf. TiTiMaKe) Harmonie HIRLAM LAPS.. Met. data extractor Core ESCAPE dispersion model
Evaluation of major fires Another model is available, called BUOYANT (Dispersion from Strongly Buoyant Sources) The dispersion of gases and particles released in major fires Includes treatments both for the plume rise, and larger scale dispersion Model output includes, e.g., concentrations of pollutants near ground level Can be used on a PC Reference: Kukkonen, J., Nikmo, J., Ramsdale, S.A., Martin, D., Webber, D.M., Schatzmann, M. and Liedtke, J., 2000. Dispersion from strongly buoyant sources. In: Gryning, S.-E. and Batchvarova, E. (eds.), Air Pollution Modeling and its Application XIII, Kluwer Academic/Plenum Publishers, pp. 539-547.
The results so far, and final expected results More accurate and reliable modelling tool for rescue services (2013) the refined model is based on the state-of-the-art science - done initial results of model evaluation look promising - done the model will be thoroughly evaluated using available measured data: A more user-friendly and versatile model (2014) less technical work required from the model user use of the best available meteorological information on-line use of the weather models, or measured values possibly, global availability of meteorological data The model will work in various information technology environments (2015) take into account all relevant environments avoid complications caused by different information tech. systems using the model over the net (possibly, via a mobile phone) possibly, use the VIRVE network (radio network of the authorities) for disseminating the dispersion maps and other information
Thank You www.fmi.fi
The propability of accident Consequence analysis Release Source term Atmospheric dispersion Consequence Risk
The state-of-the-art of source term models 1. Liquid pool models and two-phase jet models fairly good models available 2. Source terms of catastrophic releases only simple models available, challenging task 3. "Semi-instantaneous" releases no models available
Future research on heavy gas dispersion modelling thermodynamical effects the influence of complex terrain and obstacles statistical fluctuations time-dependent releases
Plume Cloud
The results of ESCAPE Dispersion diagram The height of gas cloud or plume The average concentration The temperature of gas cloud or plume The density of gas cloud or plume Tabulated results of concentration Total dose Dose in a minute Tabulated results of doses Toxicology and flammable levels
Box model cloud representation z Ur x Wind direction z y R L h x H ho Lo y Cloud shape for instantaneous release Plume shape for continuous release
Pressure (kpa) The saturation vapour pressure for some toxic and flammable substances (Kukkonen, 1990. Dissertationes No 34, The Finnish Society of Sciences and Letters). Excess of pressure (bar) 1200 10 NH 1000 3 8 800 6 600 Cl 2 4 400 C 3 H 8 SO 2 2 200 HF 0 0-40 -30-20 -10 0 10 20 30 Temperature ( C)
Height of gas cloud (m) 100 80 60 40 20 Cl 2 Continuous release G = 6.1 kg/s Pasquill B 3 m/s Pasquill D 3 m/s Pasquill D 8 m/s Pasquill E 2 m/s Pasquill D 5 m/s Pasquill F 2 m/s 0 0 200 400 600 800 1,000 Downwind distance (m) The predicted height of the gas cloud in varying meteorological conditions (Kukkonen and Savolainen, 1988. Publ. on Air Quality 4, Finnish Meteorological Institute).
Mass flux (kg/s) 10 8 Release through pipe Length of pipe = 2.0 m Diameter of aperture = 40 mm 6 4 2 Ammonia Chlorine Propane Sulphur dioxide Hydrogen fluoride 0-30 -20-10 0 10 20 30 Temperature ( C) The mass flux through a pipe from the liquid space of the container against the storage temperature (Kukkonen, 1990. Dissertationes No 34, The Finnish Society of Sciences and Letters).
Mass flux density (kg/s/cm 2 ) 3.0 2.5 2.0 1.5 1.0 0.5 Release through pipe Length of pipe = 2.0 m Diameter of aperture = 40 mm Propane Chlorine Ammonia Sulphur dioxide Hydrogen fluoride 0.0-30 -20-10 0 10 20 30 Temperature ( C) The mass flux through a pipe from the liquid space of the container against the storage temperature (Kukkonen, 1990. Dissertationes No 34, The Finnish Society of Sciences and Letters).
Mass flux (kg/s) 10 8 Cl2 Temp. = +15 C 50 mm 6 4 2 40 mm 30 mm Diameter = 20 mm 0 0 5 10 15 20 Length of pipe (m) The mass flux through pipe from the liquid space of the container against the length of pipe (Kukkonen, 1990. Dissertationes No 34, The Finnish Society of Sciences and Letters).
Effects of ammonia Concentration Time of Symptoms Injurties mg/m 3 influence 3.5-35 Noticeable odour 70 Light irritation of nose Irritation of mucous membrane 200-350 0.5-1 h Irritation of throat, Light injuries eyes and nose, cough 500 Flowing of tears Rather serious injuries 1200 0.5 h Damages in lugn Serious injuries 2500 0.5-1 h Swelling of the throat May be lethal 3500-7000 10-15 min Swelling of the throat Death
Effects of chlorine Concentration Time of Symptoms Injuries mg/m 3 influence 1.5-9 Light odour 9 8-10 h Noticeable odour Irritation of the mucous membrane 30 2 h Coughing Light injuries 150 20 min Suffocating cough Rather difficult injuries 300 10 min Suffocation cough Rather difficult injuries 900 3 min Serious injuries 3000 Suffocation Death
Distance (m) Validation of DENZ model (1/4) 60 50 40 DENZ Thorney Island Trial no 5 Pasquill B u = 4.6 m/s 30 This study 20 10 0 Experimental 0 10 20 30 40 50 Time (s) The distance of the cloud centroid from source against time (Kukkonen and Nikmo, 1992. J.Hazard.Mater. 31, 155-176).
Distance (m) Validation of DENZ model (2/4) 60 50 40 Thorney Island Trial no 7 Pasquill E u = 3.2 m/s This study 30 DENZ 20 10 Experimental 0 0 10 20 30 40 50 Time (s) The distance of the cloud centroid from source against time (Kukkonen and Nikmo, 1992. J.Hazard.Mater. 31, 155-176).
Distance (m) Validation of DENZ model (3/4) 60 50 40 30 20 10 Thorney Island Trial no 8 Pasquill D u = 2.4 m/s DENZ This study Experimental 0 0 10 20 30 40 50 Time (s) The distance of the cloud centroid from source against time (Kukkonen and Nikmo, 1992. J.Hazard.Mater. 31, 155-176).
Cloud speed (m/s) Validation of DENZ model (4/4) 6 5 Thorney Island experimental data model prediction 4 3 2 1 0 0 1 2 3 4 5 6 7 8 Wind speed (m/s) The computed and measured cloud speeds against the measured wind speed at 10 m height (Kukkonen and Nikmo, 1992. J.Hazard.Mater. 31, 155-176).
Model equations (POOL) Volume discharged: Pool volume: Pool temperature: Mass vaporised: dv dt dv dt d S S WA c V dt A Q WL Ts T c S dt dm dt WA A r 2 W ( T) u W( u z, Sc, R z, p ( T) p ) * * o o s a Q( T, t) conducted convected radiated W = evaporation rate A = area of pool c = spec. heat capacity of contaminant r = liquid density of contaminant Q = heat flux density into the pool L = latent heat of evaporation T s = temperature of source discharge
Performance evaluation of ESCAPE consequence analysis model Liquid and two-phase outflow models Measured vessel condition, pipe exit pressure and mass flux for the ammonia trials in Landskrona, Sweden (Nyrén et al., 1983). S = short pipe (2136 mm), L = long pipe (3636 mm). Trial Tank temperature ( C) Tank pressure (kpa) Exit pressure (kpa) Mass flux (kg s -1 ) S2 7.6 685 260 2.85 S3 7.4 651 248 2.72 S4 7.5 632 240 2.50 S5 7.6 615 234 2.45 S6 7.6 593 226 2.28 L7 9.3 - - 2.40 L8 9.4 652 233 2.59 L9 9.5 621 219 2.34 L10 9.1 602 210 2.21 L11 8.6 579 208 2.18
Performance evaluation of ESCAPE consequence analysis model Liquid and two-phase outflow models Measured temperatures of the tank liquid and time averages of tank pressures, pipe exit pressures and mass fluxes for the sulphur dioxide discharge trials in Skelleftehamn, Sweden (Nyrén and Winter, 1986). Trial Tank temperature ( C) Tank pressure (kpa) Exit pressure (kpa) Mass flux (kg s -1 ) P1 19.9 407 314 10.5 P3 17.2 439 304 12.0 A1 18.3 414-14.0 A2 18.2 411-14.1 A3 17.4 386-13.2
Performance evaluation of ESCAPE consequence analysis model Liquid and two-phase outflow models Trial Mass flux measured (kg s -1 ) modelled (kg s -1 ) error (%) S2 2.85 2.76-3.3 S3 2.72 2.66-2.3 S4 2.50 2.60 4.0 S5 2.45 2.55 4.1 S6 2.28 2.48 8.9 L7 2.40 2.27-5.3 L8 2.59 2.43-6.0 L9 2.34 2.35 0.5 Modelled mass fluxes q and their errors,100 q modelled q measured 1, for the ammonia (Nyrén et al., 1983) and sulphur dioxide (Nyrén and Winter, 1986) discharge trials. L10 2.21 2.30 4.0 L11 2.18 2.23 2.5 P2 10.5 9.5-9.3 P3 12.0 12.9 7.6 A1 14.0 14.6 4.0 A2 14.1 14.5 2.8 A3 13.2 13.9 5.4
Performance evaluation of ESCAPE consequence analysis model Continuous releases Performance measures for the FLADIS trials 16 and 24, where the observed (C o ) and predicted (C p ) point-wise mean concentrations (mole-%) are compared. trial max C o max C p mean C o mean C p FB NMSE MG log 10 (VG) R FAC2 16 4.56 2.55 0.22 0.15 0.39 2.88 10.2 8.65 0.95 0.45 24 2.22 1.30 0.20 0.15 0.33 2.03 20.6 11.0 0.87 0.37
Jatkuvat päästöt Laippa- ja tiivistevuodot Putkivuodot vuodot venttiilin kautta putken katkeaminen Vuodot säiliön seinämästä 8/30/2013 50
Hetkelliset päästöt Esim. säiliön repeäminen Päästön lyhyt kesto Räjähdyksenomainen Pahin mahdollinen tapaus Seurauksena voi olla BLEVE (Boiling Liquid Expanding Vapour Explosion) 8/30/2013 51
Savukaasut ilmakehän virtaukset (tuuli ja turbulenssi) Ilmaa raskaammat kaasut painovoima termodynamiikka ilmakehän virtaukset Kaksi-faasisuihkut liikemäärä termodynamiikka tuuli 8/30/2013 52
z u h 2 h u Kaasupilven etenemisnopeus = tuulen nopeus kaasupilven keskikorkeudella u c u h 2 8/30/2013 53
Pilven ja vanan yksinkertaistettu kuvaus z Ur x Tuulen suunta z y R L h x H ho Lo y Kaasupilvi Kaasuvana 8/30/2013 54
ESCAPE -ohjelman tuloksia Leviämiskuvio Kaasupilven tai -vanan korkeus Kaasupilven tai -vanan korkein maanpintapitoisuus Kaasupilven tai -vanan lämpötila Taulukoituja tuloksia pitoisuuksista BLEVE (paineaallon leviämiskuvio) Toksikologiset tiedot ja syttymisrajat 8/30/2013 55
Ilmatieteen laitoksen palvelut kemikaalionnettomuuksien varalta Leviämisselvitykset laitoskohtaiset selvitykset teollisuudelle selvitykset viranomaisille Leviämismalliohjelma ESCAPE käyttäjinä viranomaisia Palvelut akuutissa onnettomuustilanteessa Etelä-Suomen aluetoimistoissa ja keskustoimipaikassa päivystys ympäri vuorokauden: säätiedot välittömästi leviämisarvio suuronnettomuuksissa Ilmanlaatuosastolta 8/30/2013 56
Mallien ja tulosten soveltaminen Viranomaisten pelastuspalvelu Teollisuuden turvallisuusanalyysi Teollisuuslaitosten lupamenettely ja tarkastus Tapahtuneiden onnettomuuksien analysointi Mallilaskelmat ovat hyödyllisimpiä varauduttaessa ennalta onnettomuustilanteisiin. 8/30/2013 57
Leviämisarvioiden soveltamiseen akuuteissa onnettomuustilanteissa liittyy monia ongelmia Tilanteet ovat nopeita ja paikallisia Suurimmat ongelmat esiintyvät onnettomuuden lähialueella Pelastushenkilöstöllä on kova kiire Erilaisia aineita ja tilanteita on suuri määrä Riittäviä lähtötietoja arvioille ei useinkaan ole saatavilla 8/30/2013 58
Vaikutusten arviointi Lähtökohtana leviämislaskelmat Päästöjen torjunta sekä suojautumisen ja pelastustoimien vaikutus 1. Myrkylliset kaasut toksikologiset tiedot: vaikutus = f(aika,pitoisuus) 2. Syttyvät kaasut syttymisrajat palojen lämpövaikutukset räjähdysten paine- ja lämpövaikutukset 8/30/2013 59
Leviämismallien käyttö ja soveltuvuus Suunnittelu, koulutus, turvallisuusanalyysi Äkilliset onnettomuustilanteet Nopeita ja paikallisia tilanteita Suuronnettomuuksia esiintyy kerran 1-100 vuodessa Tulipalo-onnettomuuksien arvioimiseksi on omat mallinsa Maaston vaikutus 8/30/2013 60
Ilmaa raskaampien kaasupilvien leviämismallien kehittäminen Termodynamiikka ja aerosoli-ilmiöt Maastonmuotojen ja esteiden vaikutus leviämiseen Ajalliset pitoisuuden vaihtelut Leviäminen stabiilissa, heikkotuulisessa tilanteessa 8/30/2013 61
Vaarallisten aineiden kuljetukset maassamme (1982) Kuljetusluokka Maantie Rautatie Kokonaismäärä (milj. tonnia) 8.6 Mt 4.0 Mt 1. Räjähdystarvikkeet 0.1 % 0.2 % 2. Kaasut 1.4 % 9.6 % 3. Palavat nesteet 82.8 % 61.2 % 4. Helposti syttyvät aineet 2.9 % 17.8 % 5. Sytyttävästi vaikuttavat aineet 0.5 % 2.2 % 6. Myrkyt 0.4 % 3.5 % 7. Syövyttävät aineet 12.5 % 5.5 % (Lähde: Lautkaski, 1986) 100 % 100 % 8/30/2013 62
Tieliikenteessä kuljetettujen vaarallisten aineiden määrät (1992) Kuljetusluokka Osuus kokonaismäärästä (%) Muutos vuodesta 1987 (%) Kokonaismäärä 9.7 milj. tonnia 1. Räjähdystarvikkeet 0.16 +57 2. Kaasut 4.59 +62 3. Palavat nesteet 79.85 0 4. Helposti syttyvät aineet 0.34-15 5. Sytyttävästi vaikuttavat aineet 1.16 +28 6. Myrkyt 0.27-32 7. Syövyttävät aineet 13.63-12 (Lähde: Liikenneministeriö) 100 8/30/2013 63
Maailman pahimpia onnettomuuksia Paikka Aine Vuosi Kuolleet Vahingoittuneet Bhopal, Intia metyyli-isosyanaatti 1984 > 2500 tuhansia Sao Paolo, bensiini 1984 508 - Brasilia St. J. Ixhuatepec, nestekaasu 1984 452 4248 Meksiko Novosibirsk, tuntematon 1979 300 - Neuvostoliitto Los Alfaques, propyleeni 1978 216 200 Espanja Tacoa, Venezuela öljy 1981 145 1000 Xilatopec, nestekaasu 1978 100 150 Meksiko Osaka, Japani nestekaasu 1970 92 0 Gahri Dhada, maakaasu 1984 60 0 Pakistan Huimanguille, Mexico nestekaasu 1978 58 0 8/30/2013 64
Suomen pahimpia kemikaalionnettomuuksia Paikka Aine Vuosi Kuolleet Vahingoittuneet Lapua räjähdysaine 1976 43 0 Rauma kloori 1947 19 > 100 Imatra rikkidioksidi 1938 12 - Nivala rikkihappo 1988 1 3 Nilsiä kloori 1990 0 10 Uimaharju kloori 1977 0 9 Pietarsaari kloori 1978 0 4 Äetsä kloori 1993 0 2 8/30/2013 65
Myrkyllisten kaasujen ominaisuuksia Kloori Ammoniakki Rikkidioksidi Kiehumispiste -34 C -33 C -10 C Nesteen tiheys (vesi=1) Kaasun tiheys (ilma=1) Tilavuussuhde (kaasu/neste) Höyrystymislämpö (vesi=1) 1.56 0.68 1.46 2.6 0.6 2.3 400 750 500 0.13 0.61 0.17 Liukenee veteen huonosti erittäin hyvin hyvin 8/30/2013 66
Ammoniakin vaikutukset Pitoisuus Altistusaika Oireet Vahingoittuminen mg/m 3 3.5-35 Haju tuntuu 70 Lievä nenän ärsytys Limakalvojen ärtymistä 200-350 0.5-1 h Nenän, nielun ja silmien Lieviä vammoja ärsytys, yskittää 500 Kyynelvuotoa Lievää vaik. vammoja 1200 0.5 h Keuhkovammoja Vaikeita vammoja 2500 0.5-1 h Kurkunpään turvotus Hengenvaara 3500-7000 10-15 min Kurkunpään turvotus Kuolema 8/30/2013 67
Kloorin vaikutukset Pitoisuus Altistusaika Oireet Vahingoittuminen mg/m 3 1.5-9 Haju voi tuntua 9 8-10 h Haju tuntuu Limakalvojen ärtymistä 30 2 h Yskittää Lieviä vammoja 150 20 min Tukahduttava yskä Lievää vaik. vammoja 300 10 min Tukahduttava yskä Lievää vaik. vammoja 900 3 min Vaikeita vammoja 3000 Tukehtuminen Kuolema 8/30/2013 68
Paine(kPa) Ylipaine(bar) 120 10 NH 10 3 8 80 6 60 Cl 2 4 40 CH 38 SO 2 2 20 HF 0 0-40-30-20-100102030 Lämpötila( C) Kylläisen höyryn paine sekä kyllästystilan vallitessa säiliössä vallitseva ylipaine eräille myrkyllisille ja syttyville kaasuille. 8/30/2013 69
Kasuvanakorkeus(m) 10 Cl 2 Jatkuvapästö 80 G=6.1kg/s 60 Pasquil B 3m/s Pasquil D 8m/s Pasquil D 5m/s 40 20 Pasquil D 3m/s Pasquil E 2m/s Pasquil F 2m/s 0 0 20 40 60 80 1,0 Etäisys(m) Kaasuvanan korkeus etäisyyden mukana eräissä säätilanteissa (Kukkonen ja Savolainen, 1988, Ilmansuojelun julkaisuja 4, Ilmatieteen laitos). 8/30/2013 70
Propani 30 Pasquil D T=+15 C 20 u=5m/s L=1.72m 10 D=74m 0 Etäisys(m)UFL -10 0.3s -20 Sytymisrajat propanile: UFL=184g/m 3 LFL=42g/m 3 LFL 6s -30 0 20 40 60 80 10 Etäisys(m) Propaanin (nestekaasun) putkivuoto rautatiekuljetus-säiliöstä. Massavirta purkausaukossa on 21 kg/s. 8/30/2013 71