last minute need this tomorrow
For this assignment, you will complete the next essay of the scenario-based case study. The essay should be a one- to two-page narrative focusing on arguments that support what the authors discuss in attachment , as well as other research that you conduct related to these concepts.
The questions below may help you in focusing your essay on the one or two concepts chosen.
- The plume of a fire burning in the apartment contains three zones. What are the differences between the layers in the zones?
- When a fire plume hits a ceiling, it spreads horizontally, radiating thermal energy and accelerating fire growth. Why is this important to understand?
- A two-way flow occurs through the doorway of a fire room. The inflow through the lower portion of the opening is driven by air entrainment into the flames. The mass outflow through the upper portion of the opening is slightly higher due to the added mass of the combustion products. Why is this important for firefighters to understand?
- The ventilation pattern can change during a fire. What determines the hot upper layer temperature and optical density during a fire in a compartment with an opening?
- What is the neutral plane?
- Why is understanding the neutral plane critical to life safety?
Movement of Fire Gases
OBJECTIVES
After studying this chapter, you should be able to:
• Describe the three zones of the plume of a fire burning in the open and calculate the air entrainment into the flame and the height of the luminous flame.
• List three reasons why the nature of the ceiling jet is important.
• Calculate the mass outflow from a room in which a steady-state fire is burning.
• Estimate the minimum rate of heat release that leads a room to flashover.
• List nine reasons why calculating the smoke flow through most buildings requires a computational fire model.
Introduction
A 2007 fire in a furniture store began outside an enclosed loading dock area and spread into the retail showroom. During the early stages of the fire, its spread was slowed by the limited supply of fresh air. This underventilation led to the generation of a large mass of pyrolyzed and only partially oxidized effluent. The combustible gases flowed above the suspended ceiling of the main retail showroom and into the showroom itself, forming a hot smoke layer below the suspended ceiling. The fire at the back of the main showroom and the gas mixture below the suspended ceiling were both still rich with fuel. When the front windows were broken, the inflow of additional air rapidly increased the heat release rate of the fire and added air to the hot upper layer, enabling the ignition of the unburned fuel/air mixture. The fire swept quickly from the rear to the front of the main showroom, trapping nine fire fighters [1].
The description of each stage of this fire involves the word “flow” or a synonym of it. This is not uncommon. The flows to and from the fire determine the magnitude and direction of fire growth, and the flow from a fire transports toxic gases, aerosols, and heat to locations where they can be detected or do harm.
This chapter applies the fluid flow, heat generation, and chemical concepts developed previously to the movement of the gases by a fire. The treatment begins with the local movement within the fire plume and progresses to movement throughout a building.
Structure of a Fire Plume in the Open
The fuel flow into a fire comes from gasification of the liquid or solid that is burning. When the molecular fragments leave the fuel surface, they have almost no momentum. They rise into the air above strictly due to buoyancy (see the Flow of Fluids chapter), entraining (drawing in) cool air along the perimeter as they burn. This
air entrainment
is distinctly more than the amount of air needed for combustion. In turn, even with exothermic reactions occurring wherever the gasified fuel and air mix within the flammability limits, the overall temperature of the fire plume begins to decrease with increasing height. At some height, the plume temperature (and thus the gas density) essentially matches the temperature of the surrounding air. The buoyant force, which depends on a temperature difference, drops to zero; and the smoke spreads outward rather than being directed upward.
The overall fire plume structure can be regarded as having the three zones
Figure 12-1
[2]:
1. An always luminous flame zone. Because the flame temperature exceeds the temperature of the fuel surface, the gases accelerate upward.
2. An intermittently flaming zone. A movie of zone 2 would show fluctuating orange flames and transparent gases, with the lower part of the zone being orange more often, and the top of the zone being defined as the location where no orange is seen. The gas temperature, while fluctuating, has an average value that remains constant from the top to the bottom of this zone. Thus the upward velocity is essentially constant.
3. A buoyant plume. The plume is nonluminous. The temperature and the buoyant velocity decrease with height.
The height of the luminous flame, h (m), is the sum of the height of zone 1 and a fraction of the height of zone 2. The fraction of the zone 2 height that is included in h is the distance from the bottom of the zone to the point at which the frames would show a plume that is orange half the time and transparent half the time. The
luminous flame height
(m) has been shown experimentally to fit the following
Equation 12-1
:
Figure 12-1 Three zones of a buoyant diffusion flame.
© Robert Rathe,
www.robertrathe.com
where d is the diameter of the base of the fire (m) and Qc is the convective heat release of the flame (kW) [3]. For ordinary combustible products, the convective fraction of the total heat release is typically in the range of 0.6 to 0.7 for nonaromatic fuels and in the range of 0.3 to 0.5 for aromatic fuels [4]. The remaining fraction is radiative.
Some solutions to this equation are plotted in
Figure 12-2
. The ratio h/d can vary at least from 1 to 44. When conditions are such that h/d is less than 1, the flame breaks up into a number of small flamelets.
The turbulence intensity in zones 2 and 3 is quite high. The velocity fluctuations at the center can be on the order of 30 percent of the average velocity; the temperature fluctuations can be even greater. These fluctuations and the eventual decrease in plume temperature reflect the rate of air entrainment into the plume. Precise calculation of the rate of entrainment into a fire plume is not of practical value, because small ambient disturbances in the air near the plume can have substantial effects on the entrainment rate. A rough approximation of the air entrainment rate m′ (kg/s) for a turbulent plume of height z (m) and plume surface area A (m2) is given by
In most plumes, combustion occurs only in zones 1 and 2. At the top of zone 2, the mass of entrained air is roughly an order of magnitude greater than the mass needed for complete combustion.
In zone 3, the combustion has ceased and the height is large compared to the width of the base of the plume. The average midline temperature (relative to the ambient temperature) decreases at a rate inversely proportional to the 5/3 power of the height. The average midline velocity decreases more slowly, at a rate inversely proportional to the 1/3 power of the height. The diameter of the plume increases at a rate directly proportional to the height [2].
Fire Plume under a Ceiling
For a fire in a building, the plume will impinge on the ceiling, unless the fire is very small or the ceiling is very high. When this happens, the hot gases make a 90-degree turn and spread out radially under the ceiling, forming a
ceiling jet
Figure 12-3
. This ceiling jet is important for at least two reasons:
• Devices to detect the fire, including automatic sprinklers and residential smoke alarms, are generally mounted at heights just below the ceiling, and knowledge of the time of arrival and properties of the ceiling jet are crucial for predicting the point of actuation for a detection device.
Figure 12-2 Calculated flame height of turbulent diffusion flames versus the convective heat-release rate for two fire sizes.
Figure 12-3 A turbulent ceiling jet under an unconfined ceiling with walls remote from the fire. “X” marks a distance along a ceiling jet radius that is equal to the distance between the base of the fire and the height of the ceiling.
• The downward thermal radiation from the ceiling jet, and, a little later, from the hot ceiling itself, affects the rate of fire spread. It is also a major factor in preheating and igniting combustible items not yet involved in the fire.
If the fire is burning at steady state and if the fire centerline is far from the nearest wall (e.g., the fire is near the center of a large room), the maxima of ceiling jet velocity and temperature exist at a distance below the ceiling equal to about 1 percent of the distance from the base of the fire to the ceiling. The radial velocity of the jet progressively decreases as it moves farther away from the fire centerline.
Note
The decrease in the radial velocity of the ceiling jet occurs for three reasons:
1. The leading edge of the flow is a circle of increasing circumference, while the mass flow is unchanged.
2. The ceiling jet is generally turbulent, and mixing occurs between the jet and the air below. This entrainment slows down the jet and reduces its temperature.
3. The jet transfers heat to the ceiling, which reduces the temperature of the jet. According to the ideal gas law, the volume of the jet decreases proportionately. The mass flow is unchanged, so the velocity decreases.
Texture: Eky Studio/ShutterStock, Inc.; Steel: ©
Sharpshot/Dreamstime.com
Formulas have been developed for calculating the temperature and velocity distribution in such a ceiling jet [5]. For example, focus on the location marked by “X” in Figure 12-3. At this location, the calculated maximum velocity in the jet will have dropped to half the value near the fire centerline. The difference between the jet temperature and the ambient temperature will have dropped to approximately 40 percent of the value near the fire centerline. If the walls are much farther away than the fire-to-ceiling distance, the temperature and velocity of the ceiling jet will decay to negligibly low values before the jet encounters the nearest wall.
Filling of a Fire Compartment by Smoke
If a fire is ignited in a compartment without openings, one of two things will happen:
• The release of heat causes an increase in the pressure and temperature of the gases in the compartment, according to the ideal gas law. Ordinary construction materials can withstand a substantial increase in pressure if the pressure is applied evenly and gradually. However, windows can break in a fire because of stresses created when the viewable area of the glass is heated and expands more than the area shaded by the frame. In case of a prolonged fire, gypsum wallboard panels can crack, but typically large gaps do not open up until the cooling phase after the fire.
• If no rupture occurs, the oxygen in the compartment becomes depleted to the point that the combustion ceases.
In either case, as long as the compartment does not contain an opening, a hot layer forms near the ceiling. This upper layer becomes deeper as the burning continues. As the bottom of the smoke layer approaches the flames, the luminous flame height decreases, because the flame is attempting to extend into a region characterized by severe oxygen depletion. Meanwhile, the fire has set up a convective flow pattern, with the gas rising above the fire, traveling along the ceiling and down along the walls, and finally being re-entrained into the flames. This pattern produces two results. First, the upper and lower layers in the compartment are mixed, so the environment becomes more uniform throughout the compartment. Second, the fire entrains air that is increasingly depleted of oxygen (vitiated), and the burning rate decreases accordingly.
Smoke Flow from a Compartment with an Opening
A more common case is that of a compartment with an opening, either by design or due to a window rupture.
Figure 12-4
depicts a fire in a compartment with an open doorway. The ceiling jet has reached the walls, and a hot, smoky gas layer has formed near the ceiling. Continued burning has increased the (top-to-bottom) thickness of the layer until it extends below the top of the door opening; in addition, the hot, smoke-laden gases are flowing into the next compartment. The interface between the hot upper layer and the cool lower layer composes a somewhat wrinkled horizontal plane, called the
neutral plane
. (No such plane would form if the burning item were close to the doorway.)
Knowledge about this hot upper layer is crucial to assessment of life safety in a building fire. The layer thickness, temperature, and optical density all affect the intensity of downward thermal radiation incident on people and combustibles in the lower part of the compartment. The rate of outflow of the hot layer influences life safety in and fire spread to the adjacent space(s).
The layer temperature and optical density are determined by the heat release from the fire, the heat losses to the ceiling and walls, the fraction of the burned fuel converted to soot, and the volume of air into which the heat and soot are dispersed. Continuing with the compartment geometry depicted in Figure 12-4, assume that the burn rate has reached a steady state. Air enters the compartment through the lower part of the doorway, is entrained into the fire plume, and buoyantly flows upward into the hot gas layer. Hot gas escapes from the compartment through the upper part of the doorway.
Figure 12-4 Smoke layer from a fire in a compartment with an open doorway.
These inward and outward doorway flows are driven by pressure differences. From the Flow of Fluids chapter, recall that the difference in pressure between the top and bottom of a column of gas of height h is equal to gρh, where ρ is the gas density. The gas in the hot layer has a substantially lower density than the air in the lower part of the room or the air outside.
Figure 12-5
shows the resulting pressure variations with height. In the doorway, there is a height at which the inside and outside pressures are the same; it corresponds to the intersection of the neutral plane with the doorway opening. Above this height, the pressure is higher inside, causing an outflow. Below the neutral point, the reverse is true, and an inflow occurs.
The driver for this two-directional air flow at the doorway is the rate of air entrainment into the fire plume. This rate is proportional to the burning rate of the fire and the vertical distance between the base of the plume (the top surface of whatever is burning) and the neutral plane (the bottom of the hot layer). The greater the entrainment rate, the lower the bottom of the hot layer. The mass rate of outflow is slightly greater than the mass rate of air inflow because of the added mass of the gasified combustible(s).
The “drag” on this outflow is the heat loss from the hot layer to the ceiling and the upper portion of the walls. This heat loss depends on the factors discussed in the Heat Transfer chapter, including the thermal inertia of the walls, the heat capacity of the upper-layer gas (air plus combustion products), and the degree of turbulence in the upper layer.
Knowing all these input values, fire scientists routinely use computational fire models (discussed in the Computational Modeling of Fires chapter) to calculate the height of the neutral plane and the flows in and out of the compartment. The pertinent equations were developed from and confirmed by numerous laboratory experiments.
In such an experiment, the researcher measures the top-to-bottom temperature and pressure profiles in the doorway. The height at which the pressure differential changes from higher in the room to lower in the room locates the plane separating the inflow and outflow. The temperatures enable calculation of the gas density in each flow. An estimate of each mass flow is obtained using the square root of the average pressure differential, the gas density, the ideal gas law, and the partial area of the doorway through which that flow passes. The (unmeasured) degree of heat loss is estimated from the measured temperatures, which are lower than they would be if no heat losses occurred.
Figure 12-5 Pressure gradients at the doorway, caused by the relative densities of cold air and hot gases.
Early in the fire, the smoke mainly fills the upper layer of the room. The flow of smoke from the room might suffice to activate a smoke alarm in the next compartment, but does not put people or property in that room at risk. This situation changes when the bottom of the upper layer becomes lower than the door soffit. Smoke begins to billow out the doorway. If the fire does not run out of fuel, the fire grows quickly and the room rapidly reaches a flashover condition, perhaps within tens of seconds. At this time, the burning intensity is at a steady state.
Based on observations from numerous experiments, fire engineers have arrived at an approximate, two-stage portrayal of the smoke outflow from the room: no outflow before flashover, steady outflow after flashover. From these experiments, they also evolved a simple equation for estimating the steady inflow or outflow of hot gases through a door or window in the fire room. If the opening is of height H (m) and area A (m2), then the mass flow out of the opening, (kg/s), is given by
The value of C is generally taken as 0.5 kg/s·m5/2 [6].
Using this relationship, several engineers have developed equations for the minimum heat release rate that leads to flashover. These equations differ slightly due to differences in the assumptions regarding the thermal physics of the combustion and the heat losses. For reasonable compartment sizes and openings, they agree within approximately ±20 percent. The simplest such equation is
where is given in units of kW. For a compartment with a common doorway size (0.6 m × 2.0 m), is approximately 1.3 MW. It is a common practice to use a mnemonic of 1 MW to characterize the approximate heat-release rate that threatens a typical residential room reaching flashover.
Calculation of the flows requires one of the computational fire models discussed in the Fire and Smoke Hazards chapter in the following cases:
• The fire is not at or near a steady state.
• The combustible is located near the opening (and thus the boundary between the layers is not flat).
• The calculation focuses on a hypothetical fire rather than one that has been measured.
Smoke Movement in Buildings
If a building consists of just a series of compartments, all on the same level and all connected by open doorways, you can estimate the flows using initially zero. As the smoke moves down the corridor, it becomes diluted by the air in the corridor and is cooled both by that air and by heat loss to the ceiling and walls. At some distance from the fire room, the smoke flow essentially reaches the ambient temperature, and no demarcation between a hot layer and a cool layer is apparent. Only after a large volume of smoke has heated the corridor from the ceiling down to the height of the soffits will the uniform layer depth be realized.
1. At the beginning of the fire, some doors might be closed and some windows might be open; also, during the fire people might open or close doors or windows, or the fire might break a window. The previously described calculation methods might still be applicable if the number of changes is small and one change does not affect another. (Interacting changes might, for example, lead to cross-ventilation in a compartment.) If the effect of each change may not be independent, it is prudent to use a computation fire model.
2. If the building contains long corridors, the assumption of a corridor filling uniformly from the top down is not realistic. As the smoke leaves the fire room, the height of the smoke layer is determined by the height of the neutral plane in that doorway. In the corridor, the depth of the smoke layer is initially zero. As the smoke moves down the corridor, it becomes diluted by the air in the corridor and is cooled both by that air and by heat loss to the ceiling and walls. At some distance from the fire room, the smoke flow essentially reaches the ambient temperature, and no demarcation between a hot layer and a cool layer is apparent. Only after a large volume of smoke has heated the corridor from the ceiling down to the height of the soffits will the uniform layer depth be realized.
3. A multistory building contains stairwells, elevator and ventilation shafts, and possibly atria. These areas serve as both pathways for vertical smoke movement and repositories for large masses of smoke.
4. The building might have a sloping ceiling or a ceiling supported by beams, both of which would modify the behavior of the ceiling jet.
5. The presence of wind outside a building influences air movement within the building, if any doors or windows are open.
6. An operating heating, ventilation, or air-conditioning system has a profound effect on smoke movement in a building. Even if the system is shut down, hot gases might still move through the ducts due to buoyancy or expansion due to the fire’s heat release.
7. In a tall building, a
stack effect
might arise depending on the weather. The pressure difference between the top of the building and the bottom of the building is given by the equation ΔP = ρgh. On a cold day in winter, the temperature within the building is higher than the temperature outside, so the ideal gas law says that the indoor density, ρi, is proportionately lower than the outdoor density, ρo. The height of the building and the gravitational constant are the same outdoors and indoors. Therefore, the pressure difference indoors, ΔPi, is lower than the pressure difference outdoors, ΔPo. As a result, air would leak into the building at the lower levels and leak out at the upper levels, prior to the occurrence of a fire. The resulting upward flow inside the building would help carry smoke upward. On a hot day in summer, this effect is reversed if the building is air conditioned. In such a case, the air density is higher indoors, and the pressure difference is accordingly higher indoors. As a result, air would leak out of the building at the lower levels and leak in at the upper levels. The hot, buoyant smoke would then be flowing against the downward flow from the stack effect.
8 The force of the flow from an activated fire suppression system would change the nature of the air patterns within the fire room.
9. The hazard potential of the smoke changes through processes other than dilution with fresh air. Larger aerosol particles and droplets are removed from the upper layer, either by sticking to walls or by falling due to gravity. Hydrogen chloride and hydrogen bromide stick to some surfaces.
Lest this all seem overwhelming, the physics behind items 1 through 8 has been incorporated by experts into widely available computational fire models. For item 9, relatively few data are available regarding the kinetics of the generation and evolution of smoke aerosols and even fewer data on the loss of gases on realistic surfaces. Therefore, the yields of the smoke components serve as input data to the models, and the models simply transport and dilute the components as they flow throughout the building.
WRAP-UP
Chapter Summary
• The plume of a fire burning in the open can be depicted as containing three zones. The lowest zone is luminous, and the gases accelerate upward. The flames in the middle zone are intermittent; the average gas temperature and upward velocity are constant. The upper zone is a nonluminous buoyant plume, whose temperature and velocity decrease with height.
• When a fire plume hits a ceiling, it spreads horizontally. The ceiling jet activates automatic sprinklers and smoke alarms and radiates thermal energy onto combustibles, accelerating fire growth.
• In a fire in a compartment, the ceiling jet forms a hot, smoky upper layer, which becomes hotter, more optically thick, and deeper as the fire progresses.
• A two-way flow occurs through the doorway of a fire room. The inflow through the lower portion of the opening is driven by air entrainment into the flames. The mass outflow through the upper portion of the opening is slightly higher due to the added mass of the combustion products.
• Calculation of the smoke flow from large fires in most buildings requires the use of a computational fire model. The ventilation pattern can change during a fire, for example, and the geometry of building is rarely as simple as needed for hand calculations to be accurate.
CHAPTER
7
Fire Characteristics: Gaseous Combustibles
OBJECTIVES
After studying this chapter, you should be able to:
•
Describe the categorization of flames.
• Characterize laminar and turbulent flames.
• Define deflagration and detonation, and explain the difference between the two.
• Discuss flammability limits and burning velocity, as well as their relationship to fire hazard.
• Understand the difference between piloted ignition and autoignition.
• Explain the potential hazard from a gas leak.
• Explain the importance of chain branching in combustion chemistry.
Introduction
One evening, as one of the authors was approaching his house after a day at the fire research laboratory, he noticed an apparently vacant lot on which a house had stood just that morning. The neighbors reported that a crew doing some renovations to that house had nicked the natural gas line. Fortunately, they then went to lunch
—
which allowed them to avoid the m
ethane
explosion and fire that reduced the entire house to rubble.
Figure 7-
1
The Hindenburg disaster.
Courtesy of the US Navy.
This is a cautionary story. Gaseous fuels serve us well in a range of applications. Some of us cook indoors with natural gas (
methane
) and outdoors with
propane
. Some individuals weld with
acetylene
. However, the oxidation of these fuels generates a lot of heat, which in turn has the potential for inflicting serious harm. This relationship explains why blimps are now filled with nonflammable helium rather than potentially explosive
hydrogen
, even though hydrogen is the lighter, and thus more buoyant, of the two gases (
Figure 7-1
).
The direct fuel in nearly all flaming fires is in the gas phase. This chapter presents a general characterization of gas phase flames, regardless of the original phase of the fuel. It also includes special considerations for cases in which a gas is the original combustible. Later chapters will address how liquid and solid fuels gasify and how fire control can be achieved by attacking the condensed phase fuel.
Categorization of Flames
Flame types fall into the following categories:
• Premixed flames or diffusion flames
• Laminar flames or turbulent flames
• Stationary flames or propagating flames
• Subsonic flames (deflagrations) or supersonic flames (detonations)
Not all of the 1
6
possible combinations of these categories are common. Diffusion flames, for example, are rarely supersonic. Some of the flame types are more important to fire safety. For example, incipient residential fires are generally subsonic, laminar diffusion flames. Most fires are subsonic, turbulent diffusion flames at the time that they are particularly hazardous. Fires are stationary when the combustibles are already fully enveloped in flames. Fires are propagating when flames spread across a combustible item or from one item to another. These categories are discussed in the following sections.
Premixed versus Diffusion Flames
In a premixed flame, as the name implies, the fuel and the oxidizer (air) are uniformly mixed prior to ignition. An example would be if a gas leak occurred in a house and many minutes passed before someone came home and struck a match to light a cigarette or candle.
Note
Natural gas is predominantly methane (CH 4), plus as much as
2
0 percent of other gases, mostly small hydrocarbons. Its exact composition depends on the specific source. In the examples in this text, natural gas is assumed to be composed solely of methane. Methane is colorless and odorless. A small amount of a mercaptan (an acrid sulfur-containing compound) is added to natural gas that is delivered to homes and businesses to alert people to a gas leak and the hazard it presents.
Texture: Eky Studio/ShutterStock, Inc.; Steel: ©
Sharpshot/Dreamstime.com
Imagine that such a gas leak has occurred. The room contains
9.5
percent methane by volume and 90.5 percent air, and the gases are thoroughly mixed. Because air contains 21 percent oxygen by volume, and because 21 percent of 90.5 is 19, the compartment contains 19 percent oxygen by volume. Recall from the Physical and Chemical Change chapter that the volume percent of a gas is the same as the mole percent. Therefore, the ratio of the moles of oxygen in the compartment to the moles of methane is 19/9.5, or 2. This is a stoichiometric mixture, according to the balanced equation CH4 + 2 O2 → 2 H2 O.
If ignition of this mixture were to occur in the center of the room, perhaps due to a spark from an electric heater, then a small blue spherical flame would form around the spark and spread radially outward at a burning velocity of approximately
3
m/s (10 ft/s). If no heat losses occurred, the peak temperature in the flame would be 2230 K. Similar behavior would result if the methane percentage in air were somewhat lower or somewhat higher than 9.5 percent, except that the flame would propagate more slowly and the peak temperature would be lower. For fuel-rich mixtures (i.e., greater than 9.5 percent methane in air), there would be insufficient oxygen to completely oxidize the CH4 to CO2 and H2O, so the products would include CO, H2, and—for very rich mixtures—some solid, carbonaceous particles (soot).
A leak of propane or liquefied petroleum gas (LPG) from its storage container could give rise to the same hazard. LPG is either mostly propane, mostly butane, or a mixture of the two. Although propane and LPG are stored as liquids, propane has a boiling point of –44 °F (–42 °C), and butane has a boiling point of +30 °F (–1 °C). Thus the ullage1 in the storage container is occupied by propane or butane at a pressure greater than 1 atm, and any leakage will be gaseous.
Some other gases that might exhibit similar behavior are discussed at the end of this chapter. Moreover, as will be seen later, a stationary flame over a liquid or solid fuel has an important premixed zone. Combustion scientists have developed techniques for stabilizing a premixed flame so that it burns in a fixed (stationary) position.
Figure 7-2
shows three burner arrangements that produce stationary premixed flames. Such burner setups permit combustion scientists to measure flame properties accurately, as it is easier to measure these properties under conditions that do not vary with time. The distributions of temperature and chemical species within the flame became the basis for understanding the detailed sequences of chemical reactions and the ways they vary for different fuels. Measurement of the gas flows at which the flame stabilizes is a good way to measure the burning velocity of the flame. Because the gases are already mixed, the burning velocity is determined by the overall chemical reaction rate of the fuel and the oxidizer.
The diffusion flame stands in contrast to the premixed flame. As the name implies, the fuel and oxidizer gases in a diffusion flame initially exist as separate volumes. When these two volumes come together, each of the gases slowly diffuses into the other.
Figure 7-2 Techniques for stabilizing premixed flames. The arrows represent the flow of a premixed fuel–air mixture.
To see how a diffusion flame works, let us return to the example of a slow gas leak into a house. The gas emerging from the opening in the pipe is
100
percent methane. The gas elsewhere in the room is 100 percent air. If the ignition source is located at the gas leak, no oxygen is available to react with the methane. If the ignition source is located at the far side of the room, there is no fuel to combust. However, if the ignition source is located near the interface between the methane and the air, the fuel and the oxidizer will react, with the combustion then spreading rapidly over the interface. The flame will be sustained at the rate that the gases replenish this thin interface, and the burning rate will be determined by the diffusion rate of the gases, which is slower than the reaction chemistry. Because the early stages of the chemistry deplete the oxygen in the interface, diffusion flames are characterized by some degree of incomplete combustion and, therefore, by the formation of soot. The flame temperature is lower than that of a premixed flame, partly because the incomplete combustion releases less heat and partly because the soot is a black body that efficiently radiates heat from the flame. The temperature at the heart of the flame is on the order of
15
00 K, and the soot incandescence at this temperature appears yellow-orange to the human eye.
Because fires are predominantly diffusion flames, researchers have developed laboratory burners that can elucidate the events in this type of combustion.
Figure 7-3
shows four arrangements for creating stationary diffusion flames. Note that the drawing labeled “candle-like flame” closely resembles a Bunsen burner. In fact, if the air inlet to the Bunsen burner is closed, the Bunsen flame becomes a diffusion flame.
Laminar versus Turbulent Flames
In a laminar flame, the flow streamlines are smooth, and fluctuations in the velocity components are negligible. By contrast, a turbulent flame is characterized by significant local fluctuations in the velocity and temperature profiles
Figure 7-4
. The path of any particle in a turbulent flame is erratic, with many changes of direction, rather than the straight or gently curving line characteristic of a laminar flame. As presented in the Flow of Fluids chapter, turbulent flows are characterized by high values of the Reynolds number, whereas laminar flows are characterized by low Reynolds number values. As a general rule, a diffusion flame taller than 0.3 m (1 ft) will be turbulent, while a diffusion flame shorter than 0.1 m (4 in.) will be laminar, unless a high-velocity jet is involved.
Figure 7-3 Burners for stabilizing gaseous diffusion flames.
Figure 7-4 A laminar flame (a) and a turbulent flame (b).
© irin-k/ShutterStock, Inc. (a) and © Tigergallery/ShutterStock, Inc. (b).
The presence of turbulence in a flame enhances heat transfer and mixing and affects the rate and products of the reaction chemistry. Accordingly, rates of combustion and rates of convective heat loss to walls are considerably higher in turbulent flames than in laminar flames. These phenomena make it difficult to predict the behavior of large-scale fires, where the dimensions are on the order of meters, based on small-scale (bench-top) fire tests, where the dimensions are a few tens of centimeters or fewer.
Ignition of Gases
The first step of the ignition process is the generation of species that are so highly reactive that they can destabilize a fuel molecule. This can be accomplished by a spark or a flame (both called piloted ignition) or by raising the temperature by the infusion of a large amount of energy or enthalpy (nonpiloted ignition, thermal ignition, autoignition).
When a spark jumps a gap between a cathode and an anode, it ionizes the molecules in the gap, forming positive ions, free electrons, and fragments of the fuel molecules. This process is not characterized by a given temperature; in other words, the system is not at thermal equilibrium. As such, the minimum electrical energy for ignition is not particularly sensitive to the temperature of the surroundings. Instead, it depends on the energy of the spark, the gaseous fuel, and the equivalence ratio of the fuel–air mixture in the gap.
Experiments show that ignition of a nearstoichiometric mixture of a combustible gas and air requires much less spark energy than a lean mixture with substantial excess air or a rich mixture with substantial excess combustible. A lean mixture contains fewer fuel molecules for the reactive species to attack and more inert molecules that cool the energetic species and decrease the subsequent reaction rates. In a rich mixture, fewer oxygenated species are available to attack the fuel molecules, and the excess fuel molecules cool the energetic species and decrease the subsequent reaction rates. Thus a near-stoichiometric mixture demonstrates the greatest sensitivity to ignition; accordingly, combustion scientists use such a mixture to characterize an ease of ignition for a fuel.
Table 7-1
shows values of minimum ignition energies for a series of combustible gases and vapors mixed either with air or with pure oxygen. In both cases, the mixture compositions are near-stoichiometric.
Table 7-1 shows the following:
1. Fuel-oxygen mixtures ignite with weak sparks far more easily than fuel-air mixtures. This is due to the absence of nitrogen as a heat sink and diluent in the latter mixtures.
2. Saturated hydrocarbons (CnH2n + 2) all ignite similarly in air, due to the similarity of the C—H bond strength in these molecules.
3. Progressive unsaturation (double or triple bonds) in a hydrocarbon molecule favors much easier ignition, as in the sequence C2H6, C2H4, C2H2.
4. Certain gases, such as
carbon disulfide
, hydrogen, and acetylene, can be ignited with sparks less than one-tenth as strong as those required to ignite alkanes.
The minimum ignition energy that can ignite most fuel vapor–air mixtures is approximately inversely proportional to the square of the absolute pressure. Thus an extremely low-energy spark could ignite a combustible gas mixture at high pressure.
A second way to ignite a gas mixture is with an existing flame, such as a match. This piloted ignition introduces free radicals (O atoms, H atoms, and OH radicals) at local concentrations that are already sufficient to sustain the igniting flame.
A gas/vapor–air mixture also can be ignited, in the absence of a flame or spark, by raising its temperature sufficiently, generally by contact with a hot solid surface. The minimum surface temperature at which thermal ignition occurs can be measured for any combustible mixture, but unfortunately the result depends on the size, shape, orientation, and nature of the surface as well as the state of motion of the gas. For example, Reference [2] reports that 17 different investigations of the ignition temperature of hydrogen–air mixtures gave results varying from a low of 770 °F (410 °C) to a high of 1700 °F (930 °C); Reference [3] reports
ethyl ether
–air ignition temperatures from three sources as
36
6 °F (186 °C), 6
50
°F (343 °C), and 915 °F (491 °C). Reference [4] says that for short (less than 0.1 second) fuel–surface contact times, autoignition on metal surfaces occurred at temperatures 570 °F (
300
°C) higher than test results for longer contact times. Accordingly, any value reported for an autoignition temperature is valid for only one set of conditions. If any one experimental technique is used consistently, however, relative ignition temperatures for a series of gases can be obtained. With this caution in mind,
Table 7-2
lists some thermal ignition temperatures of gases in air. Diethyl ether is included because of its remarkably low ignition temperature. Note that carbon disulfide vapor, with an even lower ignition temperature of 194 °F (90 °C), can be ignited by a steam pipe at 212 °F (100 °C).
Table 7-1 Minimum Spark Ignition Energies of Near-stoichiometric Mixtures of Gases and Vapors in Air or Oxygen at 25 °C and 101 kPa (1 atm) [1]
|
Minimum Ignition Energy (µJ) |
|||||
|
Combustible |
Air |
Oxygen |
|||
|
methane, CH4 |
300 | 3 | |||
|
propane, C3H8 |
260 |
2 | |||
|
n-hexane , C6H14 |
290 |
6 | |||
|
ethane, C2H6 |
|||||
|
ethylene, C2H4 |
70 |
1 | |||
|
acetylene, C2H2 |
17 |
0.2 |
|||
|
carbon disulfide, CS2 |
15 |
— | |||
|
hydrogen, H2 |
1.2 |
Reproduced from: Kuchta, J.M., “Investigation of Fire and Explosion Accidents in the Chemical, Mining, and Fuel-Related Industries—A Manual,” p. 33, Bulletin 680, U.S. Bureau of Mines, Washington, D.C., 1985.
Table 7-2 Thermal Ignition Temperatures of Selected Gases and Vapors in Air [5]
|
Gas or Vapor |
Thermal Ignition Temperature (°C) |
|
540 |
|
|
450 |
|
|
225 |
|
|
n-octane, C8H18 |
220 |
|
515 |
|
|
490 |
|
|
305 |
|
|
90 |
|
|
diethyl ether, C2H5OC2H5 |
160 |
|
400 |
The same group of gases demonstrates both differences and similarities in their ease of ignition indicators for thermal ignition and spark ignition:
1. The thermal ignition data show that carbon disulfide is far easier to ignite than hydrogen, while the spark ignition data show that almost the same ignition energies are required for this pair.
2. The thermal ignition data show a progressive decrease of ignition temperature with increasing molecular weight for the saturated hydrocarbons (alkanes), while the spark ignition data show no such trend.
3. Acetylene is easier to ignite than ethane in both modes.
4. Ignition of a fuel-air mixture can be accomplished with a relatively weak spark or with a moderately hot surface.
Flammability Limits and Propagation Rates of Premixed Flames
Flammability Limits
Earlier in this chapter, we explained that the easiest to ignite and most rapidly burning fuel–air mixtures are stoichiometric in composition, and that moving to the rich or lean side of the stoichiometric point results in mixtures that are more difficult to ignite and slower to burn. Intuitively, it might be expected that a mixture that is too rich or too lean will be nonflammable, and that is so.
Note
Flammable means that something is capable of catching on fire. However, some confusion has arisen regarding the words nonflammable and inflammable. Unfortunately, the prefix in- has two opposite meanings. It can mean “not,” so inaccurate is the opposite of accurate. It can also mean “easily” or “very.” This ambiguity is unacceptable from a fire safety standpoint. Consequently, the fire protection profession strongly urges that the words flammable and nonflammable be used as appropriate, and that the words inflammable and inflammability never be used.
Texture: Eky Studio/ShutterStock, Inc.; Steel: © Sharpshot/Dreamstime.com
Suppose we have a stoichiometric mixture of methane and pure oxygen, at 77 °F (25 °C). If ignited, this mixture will burn vigorously, with a flame temperature of 5070 °F (
28
00 °C). If we had a stoichiometric mixture of methane and air (9.5 percent CH4, 19 percent O2, and 71.5 percent N2, all by volume), it would also burn, albeit not quite as vigorously, and the flame temperature would be 3560 °F (1960 °C). The flame reactions would proceed more slowly because of the lower temperature. This lower temperature arises because of the presence of inert nitrogen, which absorbs some of the heat released by the CH4 + 2 O2 reaction.
If still more nitrogen were added to the methane–air mixture, it would approach a point where the mixture would no longer be flammable, known as the lean flammability limit, or simply the lean limit. A mixture of borderline flammability—for example, 6.2 percent CH4,
12.4
percent O2, and 81.4 percent N2—would burn with a flame temperature of approximately 2200 °F (1200 °C). The hydrocarbon oxidation reactions in a flame do not occur rapidly enough at temperatures less than 2200 °F (1200 °C) to overcome either the heat losses from the flame to the surroundings or the rate of loss of free atoms and radicals in the flame by recombination. In such a situation, adding just a little more nitrogen suffices to quench the flame or keep it from igniting. Another way of looking at the lean flammability limit is that highly diluted flames propagate so slowly that free convective motion of the flame becomes larger than the propagation speed, thereby disrupting the flame structure.
A different phenomenon occurs when more methane is added to a stoichiometric mixture, creating a rich mixture. Not enough oxygen is available to react fully with all the methane, and some of the carbon in the methane is only partially oxidized, forming CO. The final reaction step that oxidizes CO to CO2 is very exothermic. Thus the partial oxidation of the rich mixture generates less heat and does not raise the temperature of the system as high as a stoichiometric mixture does. A factor contributing to the lesser temperature rise is the heat adsorption by the additional fuel.
Figure
7-5
shows the flammability limits of methane–air mixtures, as well as the effect of introducing additional nitrogen. First, look at the ordinate (“methane/volume %”). The methane–air flammability range goes from a lean limit of 5 percent to a rich limit of 15 percent. The stoichiometric composition of 9.5 percent is approximately in the middle.
Next, note the large amount of space in Figure 7-5 that is marked nonflammable. If more than about 35 percent by volume nitrogen is added to a stoichiometric methane–air mixture, the mixture crosses a boundary and enters into the region of nonflammable mixtures, regardless of the percent by volume of methane or oxygen. Furthermore, Figure 7-5 shows that somewhat smaller percentages of nitrogen could be added to lean or rich mixtures to render them nonflammable. As will be seen in the Fire Fighting Chemicals chapter, fires can be suppressed by surrounding them with inert gases. This dilution moves the fuel–air mixture down and to the right in Figure 7-5.
A plot like Figure 7-5 can be constructed for any flammable gas or vapor, and measurements of flammability limits have been determined for hundreds of substances. Reference [6] includes an extensive compilation of flammability limits, which also is reproduced in Reference [5].
Table 7-3
presents some selected values from this compilation. These flammability limits were measured at 77 °F (25 °C). At higher temperatures, the flammability limits are wider. At pressures greater than approximately 50 kPa, the flammability limits are effectively independent of pressure. In contrast, at much lower pressures, the flammability limits approach each other, meaning that it is difficult to establish a flammable mixture. At pressures less than 4 kPa, no propane–air mixture is flammable.
As is evident in Table 7-3, some substances (hydrogen, carbon disulfide, and acetylene) have extremely wide limits of flammability and, therefore, are especially hazardous. For example, acetylene—even in the absence of oxygen—can burn with a flame that produces hot carbon and hydrogen as products.
The upper flammability limit dramatically increases with oxygen enrichment of the atmosphere, but the lower flammability limit hardly budges because an excess of oxygen is present at this limit anyway. In general, the flammability limits are much wider when fuel vapors are mixed with oxygen rather than with air. For example, the lower and upper limits for methane in air are 5 percent and 15 percent, respectively, by volume, while the lower and upper limits of methane in oxygen are 5 percent and 61 percent, respectively [7].
Figure 7-5 Limits of flammability of various methane–air–nitrogen mixtures at 25 °C and 101 kPa (1 atm) [5].
Table 7-3 Limits of Flammability of Gases and Vapors in Air at 25 °C and 101 kPa [7]
|
Flammability Limits (Volume %) |
||||
|
Material |
Lower (Lean) Limit |
Upper (Rich) Limit |
||
|
acetone |
2.6 |
13 |
||
| acetylene |
2.5 |
100 | ||
|
ammonia |
15 | 28 | ||
| carbon disulfide |
1.3 |
50 | ||
| ethane |
3.0 |
12.4 | ||
|
ethanol (ethyl alcohol) |
3.3 |
* |
||
| ethanol | 36 | |||
| ethyl ether |
2.7 |
|||
| n-hexane |
7.4 |
|||
| hydrogen |
4.0 |
75 |
||
| methane |
5.0 |
|||
|
methanol (methyl alcohol) |
6.7 |
|||
| propane |
2.1 |
9.5 | ||
|
* The vapor pressure of these liquids at 25 °C is insufficient to reach the upper limit concentration. |
||||
|
Reproduced from: Zabetakis, M.G., “Flammability Characteristics of Combustible Gases and Vapors,” Bulletin 627, U.S. Bureau of Mines, Washington, D.C., 1965, p. 29. |
Burning Velocity
The rate at which a flame moves through a combustible gas mixture can have practical implications for fire safety. For example, suppose such a mixture forms and ignites within a compartment that is initially at approximately 300 K and contains only small openings (e.g., the cutout below a door or an air-conditioning vent to another compartment). The flame temperature may be six or seven times the ambient temperature, meaning that the temperature of the gas mixture will rise significantly. According to the ideal gas law, the pressure will rise proportionately.
If the mixture burns slowly, the hot gases have enough time to cool through contact with the walls. The gas will flow outward through the small openings, equilibrating the room pressure with that of the outside world. Little fresh air will enter the room, and the combustion chemistry will not proceed to the formation of CO2 and H2O. Instead, the products of incomplete combustion will include CO and other toxic compounds. The air will also be laden with unburned fuel. As noted in the previous chapter, this situation could lead to a backdraft.
Conversely, if the burning velocity is very high, the temperature and pressure will rise sharply, faster than the heat loss to the walls and flow through the small openings can abate. The force on windows will exceed the strength of the glass, so the windows will blow out. The large inflow of fresh air will enable the fire to sustain and perhaps intensify.
Given such a scenario, it is important to examine the concept of burning velocity more closely. This quantity is defined as the rate at which a planar flame moves through a stationary, quiescent flammable mixture of infinite extent. Combustion scientists now understand in considerable detail the relationship between the burning velocity of a mixture and the rates of the chemical oxidation reactions occurring in its flame. Theoretically, the burning velocity is approximately proportional to (q″·κ)1/2, where q″ is the mean rate of heat generation in the flame (per unit volume of reaction zone) and κis the thermal conductivity of the gas mixture. References [8, 9], and [10] provide additional information.
This is obviously an idealized rendition. The measurement of true burning velocities would require a very large apparatus, and the mixture within that volume would need to be motion free and stay mixed. These and other practical constraints make it difficult to make absolute measurements. Furthermore, in a real combustion situation, whether inside an internal combustion engine or a fire room, these conditions are unlikely to hold.
It is thus more instructive to grasp the magnitude of burning velocities and to recognize how they might be affected by various burning conditions.
Figure 7-6
depicts curves representing burning velocity measurements for several gases as a function of the fuel–air equivalence ratio ϕ. Recall that ϕ = 1 for a stoichiometric mixture, ϕ < 1 for a lean mixture, and ϕ > 1 for a rich mixture. At the ends of the curves, a flame could not be sustained. Reference [10] also lists burning velocities of about 100 compounds.
Figure 7-6 Laminar burning velocities of some combustibles in air at 25 °C and 101 kPa (1 atm) [9].
The first observation from Figure 7-6 is that the magnitude of the burning velocity is similar for most of the hydrocarbons. The exception is acetylene, whose higher burning velocity might be a safety concern in areas where welding or metal cutting is performed. The burning velocity of hydrogen is far higher, which is a consideration in the design of fueling stations for hydrogen-powered vehicles. The next observation is that, for the hydrocarbon fuels, the burning velocity peaks a little to the rich side of stoichiometric, and the falloff to either side is measurable, but not large. The burning velocity curve for hydrogen is quite different. The curve peaks at nearly ϕ = 2, and there is a sharp falloff toward a stoichiometric mixture.
A flame can actually move considerably faster than these measured values of the burning velocity, for multiple reasons:
• The surfaces in the small laboratory apparatus used to make these measurements slow the flame movement.
• The real-world flame raises the temperature of the gas from 300 K to perhaps 2100 K, causing a sevenfold expansion of the hot portion of the gas mixture, according to the ideal gas law. The expansion causes motion of the gas and carries the flame toward the unburned gas.
• The burning velocity is based on laminar flame propagation. A turbulent premixed flame will propagate several times faster than a laminar one.
If the combustible mixture is at an elevated temperature before ignition, the burning velocity will be greater. If the oxidant consists of pure oxygen instead of air, the burning velocity will be 5 to 10 times as great. The burning velocity also will depend somewhat on the ambient pressure.
Explosions, Deflagrations, and Detonations
To the extent that these three terms have entered common usage, they have acquired varied and subjective meanings to many people. The technical definitions of these three terms are as follows:
• An explosion is the rapid release of energy and increase in pressure when a premixed flame propagates in a confined space.
• A deflagration is subsonic combustion, so nearly all unwanted fires are deflagrations.
• A detonation is supersonic combustion.
A premixed flame may burn as either a deflagration or a detonation. Diffusion flames are typically deflagrations.
The descriptions presented so far have involved deflagrations. A deflagration propagates at subsonic velocity, less than 340 m/s. The speed with which a deflagration moves (i.e., the burning velocity) depends on the chemically controlled rate of heat release in the flame. If the deflagration occurs in a confined space, the pressure rise is relatively uniform on all the surfaces and is sustained until the environment cools or the walls are breached.
A detonation propagates at supersonic velocity, relative to the unburned gas. Its velocity is independent of the rates of the heat-generating chemical reactions. A detonation is a shock wave that, by compression, heats a reactive gas mixture to a high enough temperature to release the heat of combustion. It is possible to calculate the velocity of a detonation from knowledge of the heat of combustion and the physical properties of the mixture.
The damage caused by a detonation usually results from the very high pressure generated in such an event. For example, a detonating methane–oxygen mixture, originally at 1 atm, generates a pressure of more than 30 atm and moves at a velocity of 2500 m/s. (Detonations of liquid or solid explosives produce pressures of tens of thousands of atmospheres.) This pressure rise is sharp and short-lived, and typically moves in the direction of the combustion propagation.
In general, mixtures of combustible gases with oxygen are much more likely to detonate than mixtures with air. However, mixtures with air can detonate under suitably confined conditions or when a powerful enough initiating event occurs. For example, ignition of a gas–air mixture at the closed end of a long tube can produce an accelerating flame, becoming turbulent, with a build-up of pressure and a transition to detonation. Further information is given in Reference [11].
Chemical Mechanisms of Combustion of Gases
Elementary Chemistry
Scientists have been probing the oxidation of fuels almost since Priestley discovered oxygen, circa 1776. Today, much is known about the chemical reactions occurring in both premixed flames and diffusion flames—but much remains to be learned.
This section presents two of the basic cases of fuel oxidation chemistry. The first involves the simplest fuel, hydrogen. The second demonstrates the substantially increased makeup of the mechanism for the smallest carbon-containing fuel, methane.
Hydrogen Oxidation
The sequence of reactions in H2–O2 flames, both premixed and diffusion, is understood thoroughly. This reaction does not occur by H2 and O2 molecules colliding with each other, as that would involve the unlikely meeting of three molecules (two H2 and one O2) in just the right orientation, along with the simultaneous breaking of two hydrogen–hydrogen bonds and one oxygen–oxygen bond, all of which are very strong. Instead, hydrogen oxidation occurs by a chain reaction in which the critical steps involve two-species interactions, featuring the free atoms and radicals H, O, and OH.
The most critical reaction is
H + O2 → OH + O
In this reaction, a single H atom (produced by the ignition source) reacts with a stable molecule, O2, producing two highly reactive species, OH and O. This step is referred to as a chain branching reaction because it increases the number of reactive species.
The OH (hydroxyl radical) reacts very rapidly with H2:
OH + H2 → H2O + H
The OH is consumed in the formation of a stable water molecule, but it produces an H atom that can continue the chain reaction. Because this step keeps the overall reaction sequence going but does not increase the number of highly reactive species, it is called a chain propagation reaction.
Meanwhile, the O atom produced in the first reaction can react very rapidly with H2 to form two additional chain carriers:
O + H2 → OH + H
Thus each H atom, when introduced into an H2–O2 mixture, will be transformed by a sequence of rapid reactions (requiring a fraction of a millisecond) to form two molecules of H2O and three new H atoms and a lot of enthalpy
Figure 7-7
. It is the contribution of this branching sequence that can lead to an H2–O2 explosion. (If the sequence takes 1 ms, then in 100 ms, each H atom will generate approximately 1050 H atoms—a very large number and a very large amplification of the overall reaction rate.)
The reaction sequence continues until one or both of the reactants are consumed. Then, in chain termination reactions, the remaining H, O, and OH species recombine when they reach a surface or according to the gas-phase reactions H + O → OH and H + OH → H2O.
The entire sequence of reactions in H2–O2 flames involves only about 10 chemical species and 10 reactions. This small number is unique in the world of fire and combustion. The chemistry is the same for premixed and diffusion flames due to the simplicity of the mechanism and the very high diffusion rate of H atoms.
Figure 7-7 Chain reaction mechanism in the hydrogen–oxygen flame.
Premixed Methane–Oxygen Flame Chemistry
The burning of methane, the simplest of the hydrocarbon fuels, still involves 53 chemical species and 325 reactions [12]. As in H2–O2 flames, this burning process involves chain initiation steps, chain propagation steps, chain branching steps, and chain termination steps. Once again (and in nearly all combustion reactions in air), the chain branching reaction step, H + O2 → OH + O, is important.
Figure 7-8
shows some of the most important steps in the methane–oxygen premixed flame. (In a methane–air flame, nitrogen can be considered to be inert, unless the analysis focuses on the formation of traces of nitrogen oxides. These products are important mainly for addressing air pollution and smoke toxicity.)
As Figure 7-8 shows, the reaction proceeds through several paths, and the relatively stable intermediate CO must form before the formation of CO2. If the flame gases cool before the CO is completely converted to CO2 via oxidation by OH, then CO will appear in the products, even if an excess of oxygen is available in the environment.
Combustion of Larger Hydrocarbon Fuels
As the number of carbon atoms in the fuel increases, so do the numbers of chemical species and reactions that constitute the overall flame mechanism. For isooctane, the combustion mechanism includes nearly 900 species and 3600 reactions. Fortunately, computational techniques are available to identify which of these dominate this combustion process. This enables the use of a much simpler mechanism.
Figure 7-8 Some reaction steps in the methane–oxygen premixed flame (in addition to the H, O, OH, H2, and O2 reactions of Figure 7-7).
Specific Hazardous Gases
The following sections present information about some particularly hazardous gases. Additional information can be found in Reference [13] at the end of this chapter.
Hydrogen (H2)
From a fire viewpoint, hydrogen is among the most dangerous gases. It is odorless and burns with a flame that is almost invisible (except in a darkened room). As can be seen from the y-axis of
Figure 7-9
, hydrogen’s flammability limits are unusually wide. Mixtures of this gas with air are ignited very easily, such as with a low-energy spark. Moreover, hydrogen’s burning velocity is higher than that of any other combustible, so a hydrogen–air mixture in a suitably confined space can detonate.
Fortunately, flammable levels of hydrogen are generated in a few, well-known circumstances. Hydrogen can be released from hydrogen generation facilities, leaky compressed gas containers, and storage batteries during charging—all considerations that arise with the proliferation of hydrogen-fueled passenger vehicles. Hydrogen is also generated when acids attack metals. Both sodium and potassium, for example, react violently with water to form hydrogen.
Figure 7-9 Limits of flammability of hydrogen in air (downward propagation of flame), including the effects of added inert gases.
In addressing possible fire hazards of hydrogen, it is important to recognize that this gas is much lighter than air. Thus, hydrogen rises in the atmosphere and can concentrate in an upper area of a compartment. As a result, a locally ignitable hydrogen–air mixture may be present even if the average concentration in the compartment is below the lean flammability limit.
Also shown in Figure 7-9 is the effect of adding an inert gas to a hydrogen–air mixture. This addition has very little effect on the lean limit until the inert gas has displaced half of the original atmosphere. Therefore, the main fire safety effect of flushing or ventilating a hydrogen-containing compartment is to promote mixing and prevent localized build-up.
Acetylene (C2H2)
Acetylene is an extremely reactive, flammable gas. In its pure state, it cannot be stored at high pressure without the possibility of a highly exothermic dissociation into carbon and hydrogen. For this reason, acetylene is stored in cylinders that are first filled with a very porous mass—for example, cement, charcoal, or diatomaceous earth. The small pores limit the volume of gas in any particular space. Before putting the acetylene into the cylinder, the anhydrous filler mass, containing about 80 percent void space, is soaked with acetone or dimethylformamide. Acetylene gas readily dissolves in these liquids, so acetylene is effectively stored as a liquid. When the storage cylinder is opened, the acetylene flows out as a gas, leaving the solvent behind. Acetone dissolves 25 times its own volume of acetylene for each 14.7 psig (1 bar) pressure. An ordinary welding-type acetylene cylinder contains 5.5 gal (18.5 L) of acetone (43 lb, 19 kg) and about 20 lb (9 kg) of acetylene.
The flammability limits for acetylene gas are very wide, extending from 2.5 percent to 81 percent by volume. Under certain conditions, acetylene will dissociate at gas concentrations from 81 percent to 100 percent by volume, releasing heat in the process.
Because of the reactivity and unconventional storage method of acetylene gas, all acetylene tanks in North America must contain fusible plugs that open at about 212 °F (100 °C). Should this event occur near an ignition source, a flaming torch of burning gas will extend some distance—10 ft to 12 ft (3 m to 3.5 m)—from the opened vent. This flame need not be extinguished unless it endangers people or combustibles nearby. After a short time, the torch will die down and the cylinder will cool sufficiently that it can be moved to a safe place where it can continue to vent. Some possibility exists that the flame could propagate back into the cylinder or tank, at which time the tank will heat up and must be cooled with water sprays. Danger of explosion of the tank exists only if it becomes heated to a glowing red color.
Acetylene–air flames propagate very rapidly, and an acetylene–air mixture in a suitably confined space is capable of detonating. Acetylene in contact with copper forms copper acetylide, an extremely unstable solid that can explode; therefore, copper and copper alloys are not used in the storage or delivery of this gas.
Methane (CH4)
Methane is an odorless gas, somewhat lighter than air. It is the main constituent of natural gas, to which odorants (sulfur compounds) generally are added as an aid to leak detection. Methane also is found in coal mines, and is a major cause of coal mine explosions. Methane’s flammability characteristics are similar to those of other saturated hydrocarbon gases (ethane, propane, n-butane, isobutane), all of which are odorless (see
Figures 7-4
and 7-5).
Ethylene (C2H4)
Ethylene, also called ethene, is used widely as an industrial gas. It has a faint sweet odor. Its mixtures with air have wider flammability limits, are easier to ignite, and propagate flame more rapidly than saturated hydrocarbon gases. Ethylene flames are more luminous (sooty). Mixtures of ethylene with air can detonate, albeit not as readily as acetylene or hydrogen.
Ammonia (NH3)
Ammonia is used widely as a commercial refrigerant and a fertilizer. Some of its mixtures with air are flammable—a fact that is not commonly known. An ammonia leak is easily detectable because of its sharp pungent odor. Ammonia–air mixtures are more difficult to ignite and burn more slowly than saturated hydrocarbon–air mixtures
Figure 7-10 Anhydrous ammonia vapor escaping from a pressurized container.
WRAP-UP
Chapter Summary
• The molecules that combust in a flaming fire are nearly always in the gas phase.
• Flames can be characterized as premixed or diffusion, laminar or turbulent, and stationary or propagating. A deflagration is a subsonic flame; a detonation is a supersonic flame.
• A mixture of fuel and air is flammable if the fuel concentration is greater than the lean flammability limit and less than the rich flammability limit.
• A premixed flame has a higher burning velocity than a diffusion flame of the same fuel. The burning rate of the former is determined by the flame chemistry, while the burning rate of the latter is determined by the diffusion of fuel and air into each other.
• Piloted ignition is the ignition of a flammable mixture by a source that provides the high temperature and/or a source of the free radicals needed to initiate the flame chemistry. Autoignition, also known as thermal ignition, occurs when the temperature of the fuel–air mixture is high enough for the collision energy of the molecules to break the molecules’ bonds.
• The set of reactions for the combustion of fuel molecules includes chain initiation steps, chain propagating steps, chain branching steps, and chain termination steps. The initiation steps create the free radicals that propagate the flame; the termination steps remove the free radicals, converting them into stable species. The chain branching steps multiply the number of active species in the flame, thus determining the number of simultaneous chain propagating steps. The key chain branching reaction step is H + O2 →OH + O.
• A leak of a flammable gas can pose a serious fire hazard. Near the leak, the atmosphere is fuel rich. Far away from the leak, the atmosphere is fuel lean. If an ignition source is active in the region where the mixture is within the flammable limits, an explosion is possible.