Hazardous Materials Management Background Episodic Events: 1979 - - PDF document

Hazardous Materials Management Background Episodic Events: • 1979 - Three Mile Island - Harrisburg, Pennsylvania • Partial meltdown of radioactive core Lessons • Technological systems can fail in unexpected ways • Need to plan for response to low probability emergencies • Even experts make mistakes frame December 12, 1984 Bhopal, India - Union Carbide Plant Union Carbide headquarters were in Danbury; severe impact on Con- necticut Toxic gas cloud killed or severely harmed thousands - 3,000 dead April, 1986 Chernobyl, Ukraine nuclear plant Some immigrants to Waterbury area have had health problems possibly associated with explosion. Two explosions, fire, radioactivity detected in Scandinavia frametitleNovember, 1986 Basel, Switzerland - Fire in chemical plant Released thousands of tons of chemicals into Rhine, damage reached the Netherlands. frame 1988 Pennsylvania - Million gallon diesel oil tank failed Oil entered Susquehanna River, moved to Ohio River frame 1989 Alaska - Exxon Valdex Tanker spill released 232,000 gallons of crude oil within a few miles of shore. frame Consequences • Public pressure 1

2 • Financial concerns • Regulatory developments – Superfund, Resource Conservation and Recovery Act (RCRA) frame Industry Reactions • More extensive and uniform written policies and procedures • Formal oversight processes • Quantitative methodologies for addressing probabilities and con- sequences of undersirable occurrences • Identification of “critical operating parameters” to trigger ac- tion • Extensive emergency planning frame Regulatory Reactions • Widespread availability of Material Safety Data Sheets (MSDS) listing properties, hazards and protective measures • Formation of local and statewide emergency planning commit- tees • Strict reportability requirements • Mandated environmental audits and quantitative risk analysis • Modified requirements for transport of hazardous materials frame Technical Assistance Made Available • “Orange Book” – first responder’s guide to dealing with emer- gencies • Risk analysis computer programs (1) ARCHIE - Automated Resource for Chemical Hazard In- cident Evaluation (Obsolete) (2) CAMEO - Computer-Aided Management of Emergency Operations (a) ALOHA - Areal Locations of Hazardous Atmospheres (b) MARPLOT - Mapping Application for Response, Plan- ning and Local Operational Tasks • Programs to help business evaluate opportunities for reducing their use of hazardous chemicals Exercises - Page 108 Hazardous Materials Handling Practices and Potential Accidents

3 Transportation Concerns • Tank trucks and railway cars • Trucks containing chemicals • Pipelines • Planes carrying hazardous cargoes. frame Dominant Hazards • Pool fire • Vapor cloud fire • Vapor cloud explosion • Flame jet • BLEVE – Boiling Liquid Expanding Vapor Explosion • Toxic vapor cloud • Suffocation Table 4-1 Examples of hazmat accident initiators (Page 110) frame Physical Principles and Background Basic Physics and Chemistry • Atomic number – number of protons in nucleus of an atom – determines basic chemical properties • Atomic weight – number of protons plus number of neutrons - averaged • Molecular weight – sum of atomic weights of individual atoms frame Physical Properties of Matter Three states: solid, liquid, gas Main concern here: liquids, gases, transition from liquid to gas Density: mass per unit volume Specific gravity: ratio of density to density of water Evaporation (1) Rate of evaporation is proportional to surface area. Evapora- tion occurs as molecules near the surface have sufficient kinetic energy to break through the surface. (2) Rate of evaporation increases with temperature. Boiling • vapor pressure in higher than atmospheric pressure

4 • chemical can enter vapor form from throughout the liquid Exercises Page 119 frame Characterization of Flammable Vapor Hazards Flammmable Limits - Lower Flammmable Limit - Upper Flammable Limit (Sometimes referred to as Lower Explosive Limit (LEL) and Up- per Explosive Limit (UEL)) Expressed as percentage of molecules in the vapor space If the percentage of molecules in the vapor space is either above the UFL or below the LFL, then the vapor will not burn. Flash Point - Temperature at which the percentage of vapor exceeds the LFL Volatile - Materials that readily evaporate Exercises Page 123 frame Characterization of Toxicity Hazards Acute Toxic Hazards Chronic Toxic Hazards TLV - Threshold Limit Value - Level of acceptable exposure (Based on recommendations by the American Conference of Govern- ment Industrial Hygienists, ACGIH) Subdivided into: • Under normal working conditions • Daily time-weighted averages • Short term exposures (generally 15 minutes) IDLH Values - Immediately dangerous to life or health (Published by NIOSH, National Institute for Occupational Safety and Health) Generally expressed in PPMs, parts per million; sometimes in mass per unit volume Chemical Principles (1) Under fixed conditions of pressure and temperature, a given volume of gas will contain the same number of molecules of any gaseous substance whether the molecules are small, light molecules or larger, heavy molecules. (2) Avogadro’s Number: 6 . 022 × 10 23 . This is the number of molecules contained in M grams of a substance with molecular weight M . (3) One mole of a gas under standard temperature and pressure conditions occupies a volume of 22 . 4 liters.

5 (4) The number of moles of a gas is the quotient of its mass (in grams) divided by its molecular weight. (5) The number of molecules of of a gas is the product of the number of moles × Avogadro’s Number. These principles enable one to translate between mass per unit vol- ume and parts per million. Exercises Page 128 frame Typical Quantitative Issues Three scenarios Table 4-6 – Selected information needs – Page 131 From First Scenario (Derailment, possible chemical leak) Questions: • Vapor concentrations–in various directions and at various dis- tances • Flammable and toxicity limits • Potential for flammmable or toxic vapor cloud to extend a dis- tance from the accident • Other hazards? Benefit of software on site – can give fast access to useful quantitative estimates To use software on site, need modeling specialist on site, with comput- ers and programs in the emergency vehicles From Second Scenario (Nearby manufacturing plant has chemical storage tanks) Interesting range of participants: pause fire, police, hazmat team, EPA, FEMA, environmental agencies, transportation departments, civil de- fense, public works, hospital emergency room, local companies, private contractors, university consultants, school department Modeling can play a significant role in local planning and training when analyzing hazards of a hypothetical incident From Third Scenario (Develop priorities for hazardous chemicals shipped in bulk) Different approaches are possible There are several distinct levels at which models can be useful, includ- ing site-specific response, long-range planning, training and prioritiza- tion Exercises Page 134 Question 1

6 Structure and Use of Hazmat Computer Modeling Packages Principal Input: • Chemical involved • Leak conditions – eg source, size of opening affects rate of leak- age • Duration of leak • Liquid pool area limitations – size may be constrained • Weather conditions – temperature, wind Chemical Data (MSDS) May be included in software • Dominant hazards • Hazardous concentration levels • Physical and chemical properties • Hazardous reaction or combustion products frame Geographic Data • Map of area • Sensitive facilities • Storm sewers and water bodies frame Models • Tank or pipeline discharge rate • Pool size • Evaporation rate • Vapor dispersion • Thermal radiation • Flame jet distance • Explosion overpressures • Tank pressurization (from heat) • BLEVE impacts frame Output • Hazard distances and directions • Concentrations as functions of time and location • Concentration ispleth map • Hazard zone map frame Considerations

7 • Units • Chemical data • Feasible scenarios – make sure your input is reasonable • Care with illogical input requirements • Adapting the model – it may, more likely will, not fit the situ- ation exactly frame Test Results • Against physical intuition • Against rough estimates • Against additional model calculations – varying the input slightly and checking whether the output changes as expected frame The Analysis of Typical Scenarios We will go through several of these in class and in the lab. Exercises Page 150