| OCR Text |
Show AFRC MEETING AT SALT LAKE CITY, UTAH SEPTEMBER 2015 FLAME SAFETY AND THE PARAMETERS THAT AFFECT THEM John H. Pohl, Energy International, Laguna Woods, CA, 92637 Peter E. G. Gogolek, Ottawa, Ontario, Canada James G. Seebold, Atherton, CA ABSTRACT Industry needs to keep flames lit and stable to to transfer heat and provide the heat that safely manufacturers products. The safety of gases depend on a number of factors of which most are tabulated in standard references. These factors include conditions of the flame, such as temperature, pressure, combustion properties of the gases, and oxygen concentration. A number of parameters that control flame performance are tabulated from small scale data. There are also methods to estimate most of these parameters from empirical relationships for gas mixtures. The first parameter is the heating value of the gases. The heating values of gases and liquids are tabulated. However, Hoyt Hottel (deceased) and others developed a technique to estimate the heating value of a wide range of gases using the energy of individual bonds. The heating value of gas mixtures can be estimated by combining the heating value of the gas mixtures based on composition and subtracting the energy of the bonds formed. The next parameter is the ignition temperature. This is the minimum temperature required to ignite the fuel-air mixture. It is measured at small scale by subjecting gas mixtures to increasing temperatures in a coated vessel until the mixture ignites. The minimum ignition energy is another parameter. It is the minimu m amount of energy that must be applied to a gas mixtures in a coated vessel to ignite the mixture. This includes the quench space required for ignition. The value of minimum energy to ignite gas mixtures and dead spaces are tabulated in standard reference sources. The third parameter is the flammability limits. The lower limit is a lower (lowest level of fuel in air or oxygen) and Upper Level (highest level of fuel in air) at which the flame ignites and burns in a coated vessel. The flammability limits of gas mixtures can be accessed from graphs or tabulations of the Flammability Limits. The flammability limits of mixtures can be estimated using Le Chatelier's Rule. The final parameter is the flame speed (rate of reaction of the gases that comprise the flame), it is also tabulated and can be measured at small scale by several techniques. The simplest way is to measure the velocity at which a flat flame can be blown-off. The accuracy of the flame speed measurement can be improved by reducing the pressure of the flame. The flame speed must be higher than the imposed velocity to maintain flame stability. A method has been developed by the US Bureau of Mines to estimate the flame speed of gas mixtures. INTRODUCTION Use of flames to raise steam, heat products, or heat spaces can be dangerous. Often the flame is mistakenly extinguished. Flammable gas are then fed into the enclosed container or area. The gases can reignited and may form an explosive mixture. An explosion is the rapid burning of the gas mixture with accompanying increase in pressure. As the pressure increases the reaction rate increases and the gases explodes instead of reacting more slowly to form a flame. Explosions can suddenly produce high pressures that can blow-up equipment and possibly injure people and equipment . Flame are often monitored with optically recorded emissions. Most flames will emitted light as UV or IR emissions as long as they are burning. When the flame goes out, the emissions stops. Sometimes the frequency of emissions is monitored to give a better understanding of when the fluxating flame is about to go out. The explosiveness of gases is controlled by a number of factors. The external factors are temperature, pressure, composition, and reaction rate. Small scale test have been developed to measure a number of the factors that control the flame; these values are then often put in tables. Occasionally empirical equations have also been developed to estimate these factors and the other factors for gas mixtures. The important flame factors for safety are: * High Heating Value of the gases, This value controls the heat release, temperature, and reaction rate * Temperature of the Gases, this controls reaction rate * Pressure of the gases, the pressure also influences reaction rate * Mixture strength of the fuel and Oxidizer, this measures ability of the mixture to burn * The reaction rate (fame speed) of the fuel and oxidizer, this determines the balance of reaction rate and imposed velocity HEATING VALUE The high heating value, with H2O as a liquid, of gas mixtures are measured by using a small scale calorimeter and measuring the amount of heat release through the temperature rise of the cooling water. These values are tabulated in the Handbook of Chemistry and Physics (1972) and other sources. In addition, Hoyt Hottel and others have developed an empirical technique to estimate the high heat of combustion by summing the bond energy of compounds that make up the reactants and products. A table of these values is shown in Table 4. ADIABATIC FLAME TEMPERATURE The adiabatic flame temperature can be estimated by calculating the equilibrium concentration of the burned gas and then estimating the temperature rise of the products using the heat capacity of the products. The adiabatic flame temperature is usually 200 K lower than the temperature calculated at complete combustion and significantly lower than the measured temperature. This is because the equilibrium composition includes dissociated product in the mixture and dissociation requires energy. To help compensate for this difference, partial equilibrium can be used to estimate the composition. In partial equilibrium the bimolecular reaction are activated and fast and assumed to reach equilibrium, but three body reactions are slow and remain inactivated and cannot reach equilibrium. The concentration of radical in a methane flame are calculated from measured concentrations and equilibrium equations are shown in Figure 2, Pohl(1973). EQUILBRIUM Equilibrium can be calculated by hand or using a number of available computer programs. The hand calculation usually involve solving equations representing the chemical reaction equations. A simplification of this procedure involves using measurements of stable species to calculate the radical concentrations. This procedure gives approximately the correct answer as the stable species do not differ much from their combustion values. Equilibrium values require values of species of free energy values and are tabulated in the JANAF Tables (JANAF (19611967), Handbook of Chemistry and Physics (1972-1973), and Handbook of Chemical Engineer (Perry(1973)). The Gibbs Free Energy and the Heat capacity can be empirically fit to polynomials equations. See Table 1. and Table 2. , Pohl(1973). Another way to estimate the equilibrium concentration is the calculate the values using available computer programs. These programs are usually based on minimization of free energy. There are a number of these programs available. The simplest program is one developed by a student of Professor Reynolds at Stanford. I have used this program most recently to my satisfaction. However, this program requires that all free energy data for the species of interest be in or enter into the program. The program was sent to me by an old class mate of mine at MIT, Professor Reginald Mitchell, also at Stanford. An example of calculated equilibrium for combustion of sugar cane bagasse is shown in Figure 1., (Gilfillan (2001). This figure shows the change in concentration for bu rning sugar cane bagasse at different temperatures. At a temperature of 1600 C, the concentration of CH4~0 , CO=0.3 , H2=3, H2O=32, and CO2=11 %. Figure 1. Calculated Equilibrium concentrations for sugar cane bagasse for different stoichiometric raios using the Stanford Model of Equilibrium, Gilfillan (2001). Figure 2. Radical Concentrations Estimated from Measured Concentrations in a Lean Methane Flame, Pohl (1973).. THERMODYNAMICS The Gibbs free energy is expressed as: Delta G=Deta H-TDelta S (1) where H can be expressed as Delta H=CpDeltaT and S as Delta S=Cp/T I have also in the past used computer programs developed by JANAF and NASAF. JANAF also tabulates much of the equilibrium properties of Species . I have empirically fit this data to a polynomial (see Table 1.). I was satisfied with both computer models. The actual flame temperature will be lower than the adiabatic flame temperature because the flame will lose energy and hence temperature by losing heat to the surroundings. The flame can lose heat by a number of different mechanism, such as radiation, convection and conduction. The last two can be handled conventionally as conduction and convection from a gas to a heat absorbing surface. Radiation is more difficult to handling and only lately have techniques been developed to handle the loss of heat from flames by radiation. Heat loss by radiation can be estimated by the techniques developed by Hoyt C. Hottel. The components of this procedure are shown for CO2 radiation. Other speices such as H2O, CH4, and particulars also contribute to radiation heat loss. The emission depends on the compound, the concentration of the compound, the pressure of the compound, and the path length of the compound. A small concentration of soot, ~1 %, renders the flame black. CO2 is about 10% in a normal methane flame. A technique for estimating emissivity in flames is shown in Figure 3. For a one foot diameter flame viewed perpendicular with a temperature of 2500 R, the emissivity is about 0.6. Figure 3. EMISSIVITY of CO2 in FLAMES. Hottel and Sarofim (1967) Table 1. JANAF GIBBS FREE ENERGY, Pohl (1973). Table 2. EMPIRICAL FIT OF HEAT CAPACITY AT CONSTANT PRESSURE , Pohl (1973). HEAT CAPACITY To calculate the flame temperature, one needs to write a balanced equation between the heat released by reaction and the heat absorbed by the heat capacity of the product specie. This requires the mean molar heat capacity, which is available in charts such as shown in Figure 3., Hoyt C. Hottel (ca 1975). Heat capacity can be fit with empirical parameters as shown in Table 2. Figure 3. Average Mean Molar Heat Capacity, P=0, T=60-T, Hottel (ca 1975) MINIMUM IGNITION ENERGY Another factor that affects flames is the minimum ignition energy . This factor is measured in small scale systems by applying different energies to ignitable a gas mixture in a coated vessel and measuring the minimum amount of energy required to ignite the gas mixture. FLAMMABILITY LIMITS The gas mixture must also be within the flammability limits of the reactants to burn or explode. The flammability limits of the gas mixtures have an upper and lower limit. The flammability limits can be measured in a small scale coated vessel where the mixture strength are varied, whether ignited or not, and noted. The flammability limits are tabulated in standard ta bles or are presented in graphs as shown in Figure 3. The flammability limits of mixtures can also be estimated by a technique developed by at the US Bureau of Mines facilities. FLAME SPEED The last parameter for flames is the flame speed. The flame speed is a measure of the speed of reaction of the mixture. The flame speed can be measured in a number of ways. The simplest is to measure the velocity of a flat flame and determine the velocity at which the flame blows -off. The flame can be expanded by subjecting it to low pressure which slows down the reactions an enlarges the distance. COMBUSTION TETRAHEDRON The affect of most of these value can be represented by a Tetrahedron of Combustion. Flames and explosions depend on primarily four factors: 1. fuel and heat of combustion, 2. temperature and pressure, 3. oxygen, and 4. speed of the chain reactions. These factors have been illustrated with a tetrahedron as shown in Figure 4., Baukal and Schwartz (2001). Figure 4. Combustion Tetrahedron. Baukal and Schwartz (2001). Figure 4. shows that combustion is dependent on fast reactions that release heat: these reactions raises the reaction temperature and accelerates the rates of reaction, the rate of heat release, and increase in temperature. Oxidants must also be present in sufficient quantity for the reaction to take place. The amount of oxidant that needs to be present will be discussed under flammability limits. HEAT OF REACTION The heat of reaction controls the temperature of the reaction. The heat of combustion is determined by a heat balance, adding the heat of combustion of the products and subtracting the heat of combustion of the reactants. Table 3. shows the high heating value and flame speed of some gas vapors, Baukal and Schwartz (2001). As can be seen from Table 3., the heating value of common combustion gases is relative constant at about 21,000 BTU/lb. The heat of reaction can also be estimated by adding the energy released when bond are formed and subtracting the energy required to break the bonds. A table of bond energy has been compiled for combustion gases by Hottel (ca 1975). This table is shown in Table 4. below. ADAIBATIC FLAME TEMPERATURE The flame temperatures are determined by estimating the composition of the gases and performing a heat balance. The heat balance uses the heat released from combustion product and taken up by the heat capacity of the products. The products of combustion are determined in a number of ways. The simplest and least accurate is to estimate the products of combustion as the products at compl ete combustion. A more accurate technique is to estimate the products of at combustion at equilibrium concentrations at the corresponding temperature. A final way is to estimate the composition of combustion at partial equilibrium. Partial equilibrium arises in flames when the bimolecular reactions are fast and balanced and the three body reaction are slow and are considered not to be inactive. The composition can be calculated by measuring certain stable species and calculating the radical species from equilibrium of the balanced bimolecular reaction. Table 3. High Heating Value and of Selected Gases (Vapors) and Flame Speeds, Baukal and Schwartz (2001). • FUEL • HYDROGEN 21502 ~10. • METHANE 23875 1.3 • PROPANE 21669 1.3 • ISO-BUTANE 21271 1.2 • ISO-PEN 21047 1.2 • N-HEX 20966 1.3 • BEN 18184 1.3 * ACE 21502 4.6 HHV(BTU/lb) FLAMESPEED(FT/SEC) Table 4. Bond Energies of Normal bonds found in combustion gases, Hottel (ca 1975) The adiabatic flame temperatures are Tabulated in reference works, for instance, Lewis and von Elbe (1987) and Baukal and Schwartz (2001). Table 5. shows the adiabatic flame temperatures for a number of common gases listed in Lewis and von Elbe (1987). This table is for gases in air. Limited tables also exist for gases with the adiabatic flame temperature in air compared with those in oxygen. An example of these type of tables is shown in Table 5 and 6. Table 5. Adiabatic Flame temperatures for Gases in Air. Lewis and von Elbe (1987). Table 6. Adiabatic Flame Temperature of some Gases compared with Those in Oxygen. Lewis and von Elbe (1987). FUEL AIR, T C OXYGEN, T C H2 2097 2805 CO 2108 2705 CH4 1950 2780 C3H8 1988 2822 C2H4 2088 2902 C2H2 2262 3069 PRESSURE As the temperature raises in an enclosed container increases the pressure of the gases also raises. Both the temperature and pressure increase the rate of reaction. Other important parameters are: MIXTURE STRENGTH The Mixture strength of the mixture is an important parameter. The mixture strength is discussed below under Flammability limits. The mixture strength must be between the lower and upper flammability limit for the flame to burn. AUTOIGNITION TEMPERATURE The auto-ignition temperature (AIT) is the temperature required to spontaneous ignite a flammable mixtures. The auto-ignition temperature can be determined in a small scale laboratory glassware by heating the mixtures of gases until they ignite. Often the walls are coated with a non-reacting film to minimize the effect of the vessel on ignition temperature. Some auto-ignition temperatures for common industrial gases are shown in Table 6. Baukal and Schwartz (2001), p. 340. Table 6. Auto ignition Temperatures for Gases, Baukal and Schwartz (2001). Compound BP (F) Auto ignition Temperature (F) BENZENE BUTANE CARBON MONOXIDE CYCLOHEXANE ETHANE 176 31 928 549 -313 269 -127 1128 1099 473 ETHYL ALCOHOL ETHYLENE ETHYL OXIDE 173 -155 51 685 842 804 HEPTANE HEXANE METHANE METHYL ETHYL KET 209 156 -259 175 795 437 999 759 NAPTHALENE OCTANE PROPANE 424 258 -44 979 403 871 PROPYLENE TOLUENE P-XYLENE -54 231 281 856 896 982 *MINIMUM IGNITION ENERGY These experiments are also performed on small scale and the results are listed in common reference sources such as Lewis and Von Elbe (1987). The experiments are performed on a gas mixture within the flammability limits for a gas of that composition. The gas is ignited by a spark of known energy between two terminal points of known distance and configuration. The distance between the terminals is the quench distance and the energy of the spark is the minimum ignition energy. *FLAMMABILITY LIMITS Flammability limits are determined in small, wall coated vessels where the composition of the gases can be varied and the ignition point noted. Flammability are often presented as graphs, see Figure 5. The diluents has an effect on the flammability limit due to different heat capacity and diffusive of the diluents. Figure 5. Flammability Limits and Effect of different Diluents. *FLAME SPEED Flame speed is the rate of gas reaction in a flame or explosion. It is usually measured using a flat flame and determining the imposed velocity of the gases that result in blow-off of the flame. Flame speed of mixtures has been developed by the USBM in Pittsburgh. Unfournetly, I have misplaced the paper, so I am unable to give an example of flame speed of mixture of gases or reference for it. Flame speeds are usually about 2-3 ft/sec. The two exceptions are acetylene, and hydrogen which have flame speeds of about 4 ft/sec and 10 ft/sec, respectively. REFERENCES Baukal, C.E. and R.E. Schwartz, The John Zink Combustion Handbook, CRC Press, Boca Raton, FL,(2001). Gilfillan, W. N., Design of a Laboratory Rig to Determine Slagging of Biomass Fuels, Master Degree, University of Queensland, Brisbane, QLD, Australia (2001). Hottel, H. C., "Average molar heat capacity of Fuel and Combustion gases ", Private Communication, ca 1975. Hottel, H.C., Bond Energy for Thermochemical Calculations, MIT 10.70 Notes, (ca 1975). Hottel, H.C. and A.F. Sarofim, Radiative Transfer, McGraw-Hill Book Company, NY, (1967) "JANAF Thermochemical Tables", The Dow Chemical Company, Midland, MI, (1961-1967). Kuchta, J.M., "Investigation of Fire and Explosion Accidents in Chemical, Mining, and FuelRelated Industry, USBM Bulletin 680, 1985. Lewis, B. and G. von Elbe, Combustion, Flames and Explosions of Gases, 3 rd ed., Academic Press, Inc., Orlando, FL(1987). Perry, R.H., D.W. Green, and J.O.Maloney, Perry's Chemical Engineers' Handbook, 6th ed., McGraw-Hill Book Company, New York, NY, (1973). Pohl, J.H., Kinetics of Nitric Oxide and Carbon Monoxide, S.M. Thesis, MIT, Cambridge, MA, (1973). West, R.C., ed., Handbook of Chemistry and Physics, 53 th ed., Chemical Rubber Compny (CRC) Press, Cleveland, OH, (1972-1973). |
