Combustion is a chemical reaction in which an oxidant reacts rapidly with a fuel to liberate stored energy as thermal energy, generally in the form of high-temperature gases. Small amounts of electromagnetic energy (light), electric energy (free ions and electrons), and mechanical energy (noise) are also produced during combustion. Except in special applications, the oxidant for combustion is oxygen in the air. The oxidation normally occurs with the fuel in vapor form. One notable exception is oxidation of solid carbon, which occurs directly with the solid phase.

Conventional fuels contain primarily hydrogen and carbon, in elemental form or in various compounds (hydrocarbons). Their complete combustion produces mainly carbon dioxide (CO₂) and water (H₂O); however, small quantities of carbon monoxide (CO) and partially reacted flue gas constituents (gases and liquid or solid aerosols) may form. Most conventional fuels also contain small amounts of sulfur, which is oxidized to sulfur dioxide (SO₂) or sulfur trioxide (SO₃) during combustion, and noncombustible substances such as mineral matter (ash), water, and inert gases. Flue gas is the product of complete or incomplete combustion and includes excess air (if present), but not dilution air (air added to flue gas downstream of the combustion process, such as through the relief opening of a draft hood).

Fuel combustion rate depends on the (1) rate of chemical reaction of combustible fuel constituents with oxygen, (2) rate at which oxygen is supplied to the fuel (mixing of air and fuel), and (3) temperature in the combustion region. The reaction rate is fixed by fuel selection. Increasing the mixing rate or temperature increases the combustion rate.

With complete combustion of hydrocarbon fuels, all hydrogen and carbon in the fuel are oxidized to H₂O and CO₂. Generally, complete combustion requires excess oxygen or excess air beyond the amount theoretically required to oxidize the fuel. Excess air is usually expressed as a percentage of the air required to completely oxidize the fuel.

In stoichiometric combustion of a hydrocarbon fuel, fuel is reacted with the exact amount of oxygen required to oxidize all carbon, hydrogen, and sulfur in the fuel to CO₂, H₂O, and SO₂. Therefore, exhaust gas from stoichiometric combustion theoretically contains no incompletely oxidized fuel constituents and no unreacted oxygen (i.e., no carbon monoxide and no excess air or oxygen). The percentage of CO₂ contained in products of stoichiometric combustion is the maximum attainable and is referred to as the stoichiometric CO₂, ultimate CO₂, or maximum theoretical percentage of CO₂.

Stoichiometric combustion is seldom realized in practice because of imperfect mixing and finite reaction rates. For economy and safety, most combustion equipment should operate with some excess air. This ensures that fuel is not wasted and that combustion is complete despite variations in fuel properties and supply rates of fuel and air. The amount of excess air to be supplied to any combustion equipment depends on (1) expected variations in fuel properties and in fuel and air supply rates, (2) equipment application, (3) degree of operator supervision required or available, and (4) control requirements. For maximum efficiency, combustion at low excess air is desirable.

Incomplete combustion occurs when a fuel element is not completely oxidized during combustion. For example, a hydrocarbon may not completely oxidize to carbon dioxide and water, but may form partially oxidized compounds, such as carbon monoxide, aldehydes, and ketones. Conditions that promote incomplete combustion include (1) insufficient air and fuel mixing (causing local fuel-rich and fuel-lean zones), (2) insufficient air supply to the flame (providing less than the required amount of oxygen), (3) insufficient reactant residence time in the flame (preventing completion of combustion reactions), (4) flame impingement on a cold surface (quenching combustion reactions), or (5) flame temperature that is too low (slowing combustion reactions).

Incomplete combustion uses fuel inefficiently, can be hazardous because of carbon monoxide production, and contributes to air pollution.

The reaction of oxygen with combustible elements and compounds in fuels occurs according to fixed chemical principles, including

Oxygen for combustion is normally obtained from air, which is a mixture of nitrogen, oxygen, small amounts of water vapor, carbon dioxide, and inert gases. For practical combustion calculations, dry air consists of 20.95% oxygen and 79.05% inert gases (nitrogen, argon, etc.) by volume, or 23.15% oxygen and 76.85% inert gases by mass. For calculation purposes, nitrogen is assumed to pass through the combustion process unchanged (although small quantities of nitrogen oxides form). Table 1 lists oxygen and air requirements for stoichiometric combustion and the products of stoichiometric combustion of some pure combustible materials (or constituents) found in common fuels.

Fuel burns in a self-sustained reaction only when the volume percentages of fuel and air in a mixture at standard temperature and pressure are within the upper and lower flammability limits (UFL and LFL), also called explosive limits (UEL and LEL; see Table 2). Both temperature and pressure affect these limits. As mixture temperature increases, the upper limit increases and the lower limit decreases. As the pressure of the mixture decreases below atmospheric pressure, the upper limit decreases and the lower limit increases. However, as pressure increases above atmospheric, the upper limit increases and the lower limit is relatively constant.

Ignition temperature is the lowest temperature at which heat is generated by combustion faster than it is lost to the surroundings and combustion becomes self propagating (see Table 2). The fuel/air mixture will not burn freely and continuously below the ignition temperature unless heat is supplied, but chemical reaction between the fuel and air may occur. Ignition temperature is affected by a large number of factors.

The ignition temperature and flammability limits of a fuel/air mixture, together, are a measure of the potential for ignition (Gas Engineers Handbook 1965).

Combustion reactions occur in either continuous or pulse flame modes. Continuous combustion burns fuel in a sustained manner as long as fuel and air are continuously fed to the combustion zone and the fuel/air mixture is within the flammability limits. Continuous combustion is more common than pulse combustion and is used in most fuel-burning equipment.

Pulse combustion is an acoustically resonant process that burns various fuels in small, discrete fuel/air mixture volumes in a very rapid series of combustions.

The introduction of fuel and air into the pulse combustor is controlled by mechanical or aerodynamic valves. Typical combustors consist of one or more valves, a combustion chamber, an exit pipe, and a control system (ignition means, fuel-metering devices, etc.). Typically, combustors for warm-air furnaces, hot-water boilers, and commercial cooking equipment use mechanical valves. Aerodynamic valves are usually used in higher-pressure applications, such as thrust engines. Separate valves for air and fuel, a single valve for premixed air and fuel, or multiple valves of either type can be used. Premix valve systems may require a flame trap at the combustion chamber entrance to prevent flashback.

In a mechanically valved pulse combustor, air and fuel are forced into the combustion chamber through the valves under pressures less than 0.5 psi. An ignition source, such as a spark, ignites the fuel/air mixture, causing a positive pressure build-up in the combustion chamber. The positive pressure causes the valves to close, leaving only the exit pipe of the combustion chamber as a pressure relief opening. Combustion chamber and exit pipe geometry determine the resonant frequency of the combustor.

The pressure wave from initial combustion travels down the exit pipe at sonic velocity. As this wave exits the combustion chamber, most of the flue gases present in the chamber are carried with it into the exit pipe. Flue gases remaining in the combustion chamber begin to cool immediately. Contraction of cooling gases and momentum of gases in the exit pipe create a vacuum inside the chamber that opens the valves and allows more fuel and air into the chamber. While the fresh charge of fuel/air enters the chamber, the pressure wave reaches the end of the exit pipe and is partially reflected from the open end of the pipe. The fresh fuel/air charge is ignited by residual combustion and/or heat. The resulting combustion starts another cycle.

Typical pulse combustors operate at 30 to 100 cycles per second and emit resonant sound, which must be considered in their application. The pulses produce high convective heat transfer rates.

Combustion produces thermal energy (heat). The quantity of heat generated by complete combustion of a unit of specific fuel is constant and is called the heating value, heat of combustion, or caloric value of that fuel. A fuel’s heating value can be determined by measuring the heat evolved during combustion of a known quantity of the fuel in a calorimeter, or it can be estimated from quantitative chemical analysis of the fuel and the heating values of the various chemical elements in the fuel. For information on calculating heating values, see the sections on Characteristics of Fuel Oils and Characteristics of Coal.

Higher heating value (HHV), gross heating value, or total heating value includes the latent heat of vaporization and is determined when water vapor in the fuel combustion products is cooled and condensed at standard temperature and pressure. Conversely, lower heating value (LHV) or net heating value does not include latent heat of vaporization. In the United States, when the heating value of a fuel is specified without designating higher or lower, it generally means the higher heating value. (LHV is mainly used for internal combustion engine fuels.)

Heating values are usually expressed in Btu/ft³ for gaseous fuels, Btu/gal for liquid fuels, and Btu/lb for solid fuels. Heating values are always given in relation to standard temperature and pressure, usually 60, 68, or 77°F and 14.735 psia (30.00 in. Hg), depending on the particular industry practice. Heating values in the United States and Canada are based on standard conditions of 60°F (520°R) and 14.735 psia (30.00 in. Hg), dry. Heating values of several substances in common fuels are listed in Table 3.

With incomplete combustion, not all fuel is completely oxidized, and the heat produced is less than the heating value of the fuel. Therefore, the quantity of heat produced per unit of fuel consumed decreases (lower combustion efficiency).

Not all heat produced during combustion can be used effectively. The greatest heat loss is the thermal energy of the increased temperature of hot exhaust gases above the temperature of incoming air and fuel. Other heat losses include radiation and convection heat transfer from the outer walls of combustion equipment to the environment.

Air at altitudes above sea level is less dense and has less mass of oxygen per unit volume. The volume concentration of oxygen, however, remains the same as sea level. Therefore, combustion at altitudes above sea level has less available oxygen to burn with the fuel unless compensation is made for the altitude. Combustion occurs, but the amount of excess air is reduced. If excess air is reduced enough by an increase in altitude, combustion is incomplete or ceases.

When gas-fired appliances operate at altitudes substantially above sea level, three notable effects occur (see Chapter 31 of the 2020 ASHRAE Handbook—HVAC Systems and Equipment):

Altitude compensation matches fuel and air supply rates to attain complete combustion without too much excess air or too much fuel. This can be done at increased altitude by increasing the air supply amount to the combustion zone with a combustion air blower (air compensation), or by decreasing the fuel supply rate to the combustion zone by decreasing the fuel input (derating).

Power burners use combustion air blowers and can increase the air supply rate to compensate for altitude. The combustion zone can be pressurized to attain the same air density in the combustion chamber as that at sea level.

Derating can be used as an alternative to power combustion. U.S. fuel gas codes generally do not require derating of nonpower burners at altitudes up to 2000 ft. At altitudes above 2000 ft, many fuel gas codes require that burners be derated 4% for each 1000 ft above sea level (NFPA/AGA National Fuel Gas Code). Chimney or vent operation also must be considered at high altitudes (see Chapter 35 of the 2020 ASHRAE Handbook—HVAC Systems and Equipment).

In addition to reducing the gas heat content of fuel gases, reduced fuel gas density also causes increased gas velocity through flow metering orifices. The net effect is for gas input rate to decrease naturally with increases in altitude, but at less than the rate at which atmospheric oxygen decreases. This effect is one reason that derating is required when appliances are operated at altitudes significantly above sea level. Early research with draft hood-equipped appliances established that appliance input rates should be reduced at the rate of 4% per 1000 ft above sea level, for altitudes higher than 2000 ft above sea level (Figure 1).

Experience with recently developed appliances with fan-assisted combustion systems demonstrates that the 4% rule may not be required in all cases. It is therefore important to consult the manufacturer’s listed appliance installation instructions, which are based on how the combustion system operates, and other factors (e.g., impaired heat transfer).

ASHRAE research projects RP-1182 (Fleck et al. 2007) and RP-1388 (Suchovsky et al. 2011) concluded that the tested induced-draft combustion systems experienced a natural derate of 1.8% (the dotted curve in Figure 1) in gas input rate per 1000 ft increase in altitude above sea level. This gas input derate for altitude may provide safe combustion operation (less than 400 ppm of CO concentration in air-free flue gas) for induced-draft combustion systems where the combustion air drawn in by the inducer does not participate in gas entrainment or gas orifice outlet pressure that affects the gas rate. Additionally, the gas regulation reference pressure and the gas orifice outlet pressure must be at the same fluidic potential during operation. Alternatively, gas-fired systems where the gas orifice outlet pressure is affected by mechanical draft and not equal to the gas regulation pressure follow a different natural derate. Gas control systems such as the 100% premix systems using a zero-pressure (or near-zero) regulator gas delivery system and the Bernoulli principle for gas entrainment (often called constant-gas/air-ratio or tracking systems) have a natural derate of 3.1% per 1000 ft of elevation. These combustion systems can follow a 1:1 ratio with barometric pressure changes (the solid curve in Figure 1). These ASHRAE research projects showed that some gas-fired products as currently designed and constructed can be installed and operated safely and acceptably at some higher altitudes with no modifications to the sea-level gas orifices, gas manifold pressure, etc.

It is important for appliance specifiers to be aware that the heating capacity of appliances is substantially reduced at altitudes significantly above sea level. To ensure adequate delivery of heat, derating of heating capacity must also be considered and quantified.

By definition, fuel gas HHV value remains constant for all altitudes because (in North America) it is based on standard conditions of 14.735 psia (30.00 in. Hg), dry, and 60°F (520°R). Some fuel gas suppliers at high altitudes (e.g., at Denver, Colorado, at 5000 ft) may report fuel gas heat content at local barometric pressure instead of standard pressure. This can be calculated using the following equation:

where

HC	=	local gas heat content at local barometric pressure and standard temperature conditions, Btu/ft3
HHV	=	gas higher heating value at standard temperature and pressure of 520°R and 14.735 psia, respectively, Btu/ft3
B	=	local barometric pressure, psia (not corrected to sea level: do not use barometric pressure as reported by weather forecasters, because it is corrected to sea level)
Ps	=	standard pressure = 14.735 psia

For example, at 5000 ft, the barometric pressure is 12.23 psia. If the HHV of a fuel gas sample is 1000 Btu/ft³ (at standard temperature and pressure), local gas heat content is 830 Btu/ft³ at 12.23 psia barometric pressure 5000 ft above sea level.

Therefore, local gas heat content of a sample of fuel gas can be expressed as 830 Btu/ft³ at local barometric pressure of 12.23 psia and standard temperature, or as 1000 Btu/ft³ (HHV). Both gas heat contents are correct, but the application engineer must understand the difference to use each one correctly. As described earlier, local heat content HC can be used to determine appliance input rate.

When gas heat value (either HHV or HC) is used to determine gas input rate, the gas pressure and temperature in the meter must also be considered. Add the gage pressure of gas in the meter to the local barometric pressure to calculate the heat content of the gas at the pressure in the meter. Gas temperature in the meter also affects the heat content of the gas in the meter. Gas heat value is directly proportional to gas pressure and inversely proportional to its absolute temperature in accordance with the perfect gas laws, as shown in the following example calculations for gas input rate with either the HHV or local heat content method.

Natural Gas. This is a nearly odorless, colorless gas that accumulates in the upper parts of oil and gas reservoirs. Raw natural gas is a mixture of methane (55 to 98%), higher hydrocarbons (primarily ethane), and noncombustible gases. Some constituents, principally water vapor, hydrogen sulfide, helium, liquefied petroleum gases, and gasoline, are removed before distribution.

Natural gas used as fuel typically contains methane, CH₄ (70 to 96%); ethane, C₂H₆ (1 to 14%); propane, C₃H₈ (0 to 4%); butane, C₄H₁₀ (0 to 2%); pentane, C₅H₁₂ (0 to 0.5%); hexane, C₆H₁₄ (0 to 2%); carbon dioxide, CO₂ (0 to 2%); oxygen, O₂ (0 to 1.2%); and nitrogen, N₂ (0.4 to 17%).

The composition of natural gas depends on its geographical source. Because the gas is drawn from various sources, the composition of gas distributed in a given location can vary slightly, but a fairly constant heating value is usually maintained for control and safety. Local gas utilities are the best sources of current gas composition data for a particular area.

Heating values of natural gases vary from 900 to 1200 Btu/ft³; the usual range is 1000 to 1050 Btu/ft³ at sea level. The heating value for a particular gas can be calculated from the composition data and values in Table 3.

For safety purposes, odorants (e.g., mercaptans) are added to natural gas and LPG to give them noticeable odors.

Liquefied Petroleum Gases (LPG). These gases consist primarily of propane and butane, and are usually obtained as a by-product of oil refinery operations or by stripping liquefied petroleum gases from the natural gas stream. Propane and butane are gaseous under usual atmospheric conditions, but can be liquefied under moderate pressures at normal temperatures.

Commercial propane consists primarily of propane but generally contains about 5 to 10% propylene. Its heating value is about 21,560 Btu/lb, about 2500 Btu/ft³ of gas, or about 91,000 Btu/gal of liquid propane. At atmospheric pressure, commercial propane has a boiling point of about −44°F. The low boiling point of propane allows it to be used during winter in the northern United States and southern Canada. Tank heaters and vaporizers allow its use also in colder climates and where high fuel flow rates are required. American Society for Testing and Materials (ASTM) Standard D1835 and Gas Processors Association (GPA) Standard 2140, which are similar, provide formulating specifications for required properties of liquefied petroleum gases at the time of delivery. Propane is shipped in cargo tank vehicles, rail cars, and barges. It is stored at consumer sites in tanks that comply with requirements of the ASME Boiler and Pressure Vessel Code or transportable cylinders that comply with requirements of the U.S. Department of Transportation.

HD-5 propane is a special LPG product for use in internal combustion engines under moderate to high severity. Its specifications are included in ASTM Standard D1835 and GPA Standard 2140.

Propane/air mixtures are used in place of natural gas in small communities and by natural gas companies to supplement normal supplies at peak loads. Table 4 lists heating values and specific gravities for various fuel/air ratios.

Commercial butane consists primarily of butane but may contain up to 5% butylene. It has a heating value of about 21,180 Btu/lb, about 3200 Btu/ft³ of gas, or about 102,000 Btu/gal of liquid butane. At atmospheric pressure, commercial butane has a relatively high boiling point of about 32°F. Therefore, butane cannot be used in cold weather unless the gas temperature is maintained above 32°F or the partial pressure is decreased by dilution with a gas having a lower boiling point. Butane is usually available in bottles, tank trucks, or tank cars, but not in cylinders.

Butane/air mixtures are used in place of natural gas in small communities and by natural gas companies to supplement normal supplies at peak loads. Table 4 lists heating values and specific gravities for various fuel/air ratios.

Commercial propane/butane mixtures with various ratios of propane and butane are available. Their properties generally fall between those of the unmixed fuels.

Manufactured gases are combustible gases produced from coal, coke, oil, liquefied petroleum gases, or natural gas. For more detailed information, see the Gas Engineers Handbook (1965). These fuels are used primarily for industrial in-plant operations or as specialty fuels (e.g., acetylene for welding).

Renewable Gases. There are two primary processes that produce renewable gas from various feedstocks: anaerobic digestion and thermal gasification. Both processes are discussed here in general.

Anaerobic Digestion (AD). In this process, complex organic matter (source material) is broken down into simpler constituents, directly through microbial action and in the absence of oxygen. The degradation process usually occurs in some form of tank, called a digester or reactor. Organic matter, perhaps first pretreated by grinding or by mechanical or chemical hydrolysis, enters the tank and is held there for a predefined length of time. For systems based on animal manure, this time ranges from a few days to a few weeks; for systems that use energy crops, residence time can be up to several tens of days. During that period, microbial activity breaks down the organic matter, and the resultant gaseous products contain a large fraction of methane and carbon dioxide along with trace amounts of other gases. Eventually, the material is expelled from the digester and replaced by new feed matter to continue the digestion/degradation process. The new organic matter may replace the entirety of the resident matter in batch, or it may replace it semicontinuously, depending on the reactor and on the collection and processing of the source matter.

The four stages of anaerobic digestion are as follows:

Wastewater treatment plant (WWTP) gases are generated from waste liquids and solids from household or commercial water usage or from industrial processes. Depending on the architecture of the sewer system and local regulation, it may also contain stormwater from roofs, streets, or other runoff areas. The contents may include anything expelled (legally or not) from a household that enters the drains. If stormwater is included in the wastewater sewer flow, it may also contain components collected during runoff (e.g., soil, metals, organic compounds, animal waste, oils, solid debris such as leaves and branches).

Processing influent to a large wastewater treatment plant typically has three stages: mechanical, biological, and sometimes chemical processing. The goal of such treatments is to prepare solids (treated sludge) and liquids (treated effluent) output from the WWTP that is environmentally safe and able to be landfilled (treated solids) or returned to the environment (treated effluent). One step in the processing of the wastewater sludge may be anaerobic digestion, from which methane can be produced.

Landfill gas (LFG) derives from municipal solid waste (MSW) in landfills. In the United States, the primary federal law currently controlling disposal of solid and hazardous waste is the Resource Conservation and Recovery Act (RCRA), which sets criteria under which landfills can accept municipal solid waste and nonhazardous industrial solid waste. It also prohibits open dumping of waste, and ensures that hazardous waste is managed from the time of its creation to the time of its disposal.

For energy production, MSW can be used in either of two ways: it can be gasified directly through thermochemical processes [see the section on Thermal Gasification (TG)], or it can be deposited into a landfill and undergo AD. Although most surveys indicate the material composition of all landfill constituents, the important components of MSW for the production of landfill gas are the organic fractions, which form the substrates that anaerobically decompose in the landfill. Typical or approximate contents of the organic fractions of MSW appear in Table 5.

Landfill gas is the result of these anaerobic processes acting on the organic matter within a landfill. In a sense, the landfill itself substitutes for an anaerobic digester tank: a closed volume that contains putrescible matter and over time becomes devoid of oxygen.

Methane and carbon dioxide are the principal components of the gas, though the overall composition of raw LFG can vary depending on the materials in the landfill: it can contain significant amounts of hydrogen sulfide as well as trace amounts of ammonia, mercury, chlorine, fluorine, siloxanes, and volatile metallic compounds.

Variation in source MSW, its organic contents, temperature conditions, moisture conditions, compaction densities, landfill operational procedures, and other landfill attributes account for variations in LFG content. Typical compounds and their reported concentration ranges are shown in Table 6. Methane concentration is generally reported as being around 55 mol%, and carbon dioxide is often measured at 40%. Nitrogen, hydrogen, oxygen, and hydrogen sulfide are found in smaller but significant quantities.

The composition of raw biogas can vary depending on the materials being digested. Methane and carbon dioxide are the principal components of landfill gas, though the overall composition of raw LFG can contain significant amounts of H₂S as well as trace amounts of ammonia, mercury, chlorine, fluorine, siloxanes, and volatile metallic compounds (EPA 2017).

However, the composition of biogas generated from dairy manure tends to be more consistent because the dairy industry is regulated as a producer of milk for human consumption. Typical compounds and their reported concentration ranges for digester-based biogas are shown in Table 7. Methane concentration can be as high as 70% but is generally reported at around 60%. Landfill gas, unless the landfill is specifically designed for gas production, typically has a slightly lower methane fraction (e.g., around 55%). Adding food wastes into a manure-based digester (codigestion) seems to improve biogas production and may increase methane concentration, but is not addressed in this chapter. CO₂, the other major biogas component, is often measured around 40%. Nitrogen, hydrogen, oxygen, and H₂S are found in smaller quantities.

Similarly to natural gas, biogas derived from biomass feedstocks also must undergo one or more cleanup processes to remove unwanted components and to upgrade it suitably for natural gas pipelines. Quality control is required to prevent or minimize entry of raw, unconditioned biogas or less-than-pipeline-quality biomethane into the natural gas grid. Many methods and processes can be used to remove contaminants from subquality gas streams. Some are appropriate for use on farms, and others are only economical at gas flows measured in millions of standard cubic feet per day (MMSCFD) and where sulfur removal rates are measured in tons per day. The ability of a process to remove unwanted compounds depends on many factors, and assessment of the true practicality of a method for a given application requires careful evaluation.

Thermal Gasification (TG). This process encompasses a fairly broad range of processes and reactions that convert carbonaceous feedstocks (coal, heavy oils, wood, biomass, sludge, etc.) into a mixture of gases, primarily hydrogen, carbon monoxide, steam, carbon dioxide, some methane, small amounts of ethane and higher hydrocarbons, small amounts of hydrogen sulfide, and nitrogen (if gasification is conducted with air). Depending on feedstock and operating conditions, TG of biomass typically generates tars and oils that are undesirable by-products.

Thermal gasification is conducted in reducing (substoichiometric or incompletely combusting) atmospheres. Some of the process heat for the endothermic gasification reactions is typically provided by burning some of the carbon in the feedstock. Process heating can be direct or indirect. Indirect heating of the gasifier is called all-thermal gasification. A typical range of syngas compositions from oxygen- or air-blown operation is presented in Table 8.

This mixture of gases is known as synthesis gas or syngas, and can be further catalytically converted into methane to generate refinery gas (RG). The syngas can also be converted into liquid products by Fischer-Tropsch synthesis [see DOE (2017) for a description] for use as transportation fuel, or transformed into a host of chemical products such as methanol, dimethyl ether, fuel gas/town gas, ethylene/propylene, or acetic acid. It can also be combusted directly in a gas turbine to drive a generator. In some cases, a catalyst is included with the feedstock to accelerate reactions and allow a reduced operating temperature. TG can be carried out at temperatures in the range of 1200 to 2000°F and at pressures ranging from ambient to greater than 1000 psig. If the TG process is conducted at ambient or fairly low pressure, then the product RG must be compressed so it can be injected into the transmission or distribution line at the appropriate pressure.

Fuel oils for heating are broadly classified as distillate fuel oils (lighter oils) or residual fuel oils (heavier oils). ASTM Standard D396 has specifications for fuel oil properties that subdivide the oils into various grades. Grades No. 1 and 2 are distillates; grades 4, 5 (Light), 5 (Heavy), and 6 are residual. Specifications for the grades are based on required characteristics of fuel oils for use in different types of burners.

Grade No. 1 is a light distillate intended for vaporizing-type burners. High volatility is essential to continued evaporation with minimum residue. This fuel is also used in extremely cold climates for residential heating using pressure-atomizing burners.

Grade No. 2 is heavier than No. 1 and is used primarily with pressure-atomizing (gun) burners that spray oil into a combustion chamber. Vapor from the atomized oil mixes with air and burns. This grade is used in most domestic burners and many medium-capacity commercial/industrial burners. A dewaxed No. 2 oil with a pour point of −58°F is supplied only to areas where regular No. 2 oil would jell. Grade No. 2—low sulfur is a relatively new category that has a sulfur content of 0.05%. Lower fuel sulfur content reduces fouling rates of boiler heat exchangers (Butcher et al. 1997).

Grade No. 4 is an intermediate fuel that is considered either a heavy distillate or a light residual. Intended for burners that atomize oils of higher viscosity than domestic burners can handle, its permissible viscosity range allows it to be pumped and atomized at relatively low storage temperatures.

Grade No. 5 (Light) is a residual fuel of intermediate viscosity for burners that handle fuel more viscous than No. 4 without preheating. Preheating may be necessary in some equipment for burning and, in colder climates, for handling.

Grade No. 5 (Heavy) is a residual fuel more viscous than No. 5 (Light), but intended for similar purposes. Preheating is usually necessary for burning and, in colder climates, for handling.

Grade No. 6, sometimes referred to as Bunker C, is a high-viscosity oil used mostly in commercial and industrial heating. It requires preheating in the storage tank to allow pumping, and additional preheating at the burner to allow atomizing.

Low-sulfur residual oils are marketed in many areas to allow users to meet sulfur dioxide emission regulations. These fuel oils are produced (1) by refinery processes that remove sulfur from the oil (hydrodesulfurization), (2) by blending high-sulfur residual oils with low-sulfur distillate oils, or (3) by a combination of these methods. These oils have significantly different characteristics from regular residual oils. For example, the viscosity/temperature relationship can be such that low-sulfur fuel oils have viscosities of No. 6 fuel oils when cold, and of No. 4 when heated. Therefore, normal guidelines for fuel handling and burning can be altered when using these fuels.

Another liquid fuel of increasing interest is biodiesel. It is made from biological sources (e.g., vegetable oils, used cooking oils, tallow). ASTM Standard D6751 addresses biodiesel; requirements are largely similar to those for petroleum diesel (cetane number, flash point, etc.; see the section on Types and Properties of Liquid Fuels for Engines). In practice, biodiesel is almost always blended, most often with ASTM heating oils when used for stationary heating applications, because of cost and cold-flow properties of 100% biodiesel. However, the benefits of a renewable fuel that has very low net carbon dioxide emission in its life cycle, reduced particulate and sulfur emissions, and lower NO_x emissions in many heating applications balance the need for mixing.

Fuel oil grade selection for a particular application is usually based on availability and economic factors, including fuel cost, clean air requirements, preheating and handling costs, and equipment cost. Installations with low firing rates and low annual fuel consumption cannot justify the cost of preheating and other methods that use residual fuel oils. Large installations with high annual fuel consumption cannot justify the premium cost of distillate fuel oils.

Characteristics that determine grade classification and suitability for given applications are (1) viscosity, (2) flash point, (3) pour point, (4) water and sediment content, (5) carbon residue, (6) ash, (7) distillation qualities or distillation temperature ranges, (8) specific gravity, (9) sulfur content, (10) heating value, (11) carbon/hydrogen content, (12) aromatic content, and (13) asphaltene content. Not all of these are included in ASTM Standard D396.

Viscosity is an oil’s resistance to flow. It is significant because it indicates the ease with which oil flows or can be pumped and the ease of atomization. Differences in fuel oil viscosities are caused by variations in the concentrations of fuel oil constituents and different refining methods. Approximate viscosities of fuel oils are shown in Figure 2.

Flash point is the lowest temperature to which an oil must be heated for its vapors to ignite in a flame. Minimum permissible flash point is usually prescribed by state and municipal laws.

Pour point is the lowest temperature at which a fuel can be stored and handled. Fuels with higher pour points can be used when heated storage and piping facilities are provided.

Water and sediment content should be low to prevent fouling the facilities. Sediment accumulates on filter screens and burner parts. Water in distillate fuels can cause tanks to corrode and emulsions to form in residual oil.

Carbon residue is obtained by a test in which the oil sample is destructively distilled in the absence of air. When commercial fuels are used in proper burners, this residue has almost no relationship to soot deposits, except indirectly when deposits are formed by vaporizing burners.

Ash is the noncombustible material in an oil. An excessive amount indicates the presence of materials that cause high wear on burner pumps.

The distillation test shows the volatility and ease of vaporization of a fuel.

Specific gravity is the ratio of the density of a fuel oil to the density of water at a specific temperature. Specific gravities cover a range in each grade, with some overlap between distillate and residual grades. API gravity (developed by the American Petroleum Institute) is a parameter widely used in place of specific gravity. It is obtained by the following formula:

where Sp Gr at 60/60°F is the ratio of the mass of a given volume of oil at 60°F to the mass of the same volume of water at 60°F. The API gravity of water at 60°F is 10.0.

Air pollution considerations are important in determining the allowable sulfur content of fuel oils. Sulfur content is frequently limited by legislation aimed at reducing sulfur oxide emissions from combustion equipment; usual maximum allowable sulfur content levels are 1.0, 0.5, or 0.3%. Table 9 lists sulfur levels of some marketed fuel oils. Research (Lee et al. 2002a, 2002b) suggests that fuel sulfur content affects the sulfate content of particulate emissions, which are reported to be associated with adverse health effects.

Sulfur in fuel oils is also undesirable because sulfur compounds in flue gas are corrosive. Although low-temperature corrosion can be minimized by maintaining the stack at temperatures above the dew point of the flue gas, this limits the overall thermal efficiency of combustion equipment. The presence of sulfur oxides in the flue gas raises the dew point temperature (see the section on Combustion Calculations).

For certain industrial applications (e.g., direct-fired metallurgy, where work is performed in the combustion zone), fuel sulfur content must be limited because of adverse effects on product quality. Sulfur contents of typical fuel oils are listed in Table 9.

Heating value is an important property, although ASTM Standard D396 does not list it as one of the criteria for fuel oil classification. Heating value can generally be correlated with the API gravity. Table 10 shows the relationship between heating value, API gravity, and density for several oil grades. In the absence of more specific data, heating values can be calculated as shown in the North American Combustion Handbook (1978):

Distillate fuel oils (grades 1 and 2) have a carbon/hydrogen content of 84 to 86% carbon, with the remainder predominantly hydrogen. Heavier residual fuel oils (grades 4, 5, and 6) may contain up to 88% carbon and as little as 11% hydrogen. An approximate relationship for determining the hydrogen content of fuel oils is

ASTM Standard D396 is more a classification than a specification, distinguishing between six generally nonoverlapping grades, one of which characterizes any commercial fuel oil. Quality is not defined, as a refiner might control it; for example, the standard lists the distillation temperature 90% point for grade No. 2 as having a maximum of 640°F, whereas commercial practice rarely exceeds 600°F.

The primary stationary engine fuels are diesel and gas turbine oils, natural gases, and LPGs. Other fuels include sewage gas, manufactured gas, and other commercial gas mixtures. Gasoline and the JP series of gas turbine fuels are rarely used for stationary engines.

Only properties of diesel and gas turbine fuel oils are covered here; properties of natural and liquefied petroleum gases are found in the section on Gaseous Fuels. For properties of gasolines and JP turbine fuel, consult texts on internal combustion engines and gas turbines. Properties of currently marketed gasolines can be found in ASTM Standard D4814.

Properties of the three grades of diesel fuel oils (1-D, 2-D, and 4D) are listed in ASTM Standard D975.

Grade No. 1-D includes the class of volatile fuel oils from kerosene to intermediate distillates. They are used in high-speed engines with frequent and relatively wide variations in loads and speeds and where abnormally low fuel temperatures are encountered.

Grade No. 2-D includes the class of lower-volatility distillate gas oils. They are used in high-speed engines with relatively high loads and uniform speeds, or in engines not requiring fuels with the higher volatility or other properties specified for grade No. 1-D.

Grade No. 4-D covers the more viscous distillates and blends of these distillates with residual fuel oils. They are used in low- and medium-speed engines involving sustained loads at essentially constant speed.

Property specifications and test methods for grade No. 1-D, 2-D, and 4-D diesel fuel oils are essentially identical to specifications of grade No. 1, 2, and 4 fuel oils, respectively. However, diesel fuel oils have an additional specification for cetane number, which measures ignition quality and influences combustion roughness. Cetane number requirements depend on engine design, size, speed and load variations, and starting and atmospheric conditions. An increase in cetane number over values actually required does not improve engine performance. Thus, the cetane number should be as low as possible to ensure maximum fuel availability. ASTM Standard D975 provides several methods for estimating cetane number from other fuel oil properties.

ASTM Standard D2880 for gas turbine fuel oils relates gas turbine fuel oil grades to fuel and diesel fuel oil grades. Test methods for determining properties of gas turbine fuel oils are essentially identical to those for fuel oils. However, gas turbine specifications limit quantities of some trace elements that may be present, to prevent excessive corrosion in gas turbine engines. For a detailed discussion of fuels for gas turbines and combustion in gas turbines, see Chapters 5 and 9, respectively, in Hazard (1971).

Commonly accepted definitions for classifying coals are listed in Table 11. This classification is arbitrary because there are no distinct demarcation lines between coal types.

Anthracite is a clean, dense, hard coal that creates little dust in handling. It is comparatively difficult to ignite, but burns freely once started. It is noncaking and burns uniformly and smokelessly with a short flame.

Semianthracite has a higher volatile content than anthracite. It is not as hard and ignites more easily. Otherwise, its properties are similar to those of anthracite.

Bituminous coal includes many types of coal with distinctly different compositions, properties, and burning characteristics. Coals range from high-grade bituminous, such as those found in the eastern United States, to low-rank coals, such as those found in the western United States. Caking properties range from coals that melt or become fully plastic, to those from which volatiles and tars are distilled without changing form (classed as noncaking or free-burning). Most bituminous coals are strong and nonfriable enough to allow screened sizes to be delivered free of fines. Generally, they ignite easily and burn freely. Flame length is long and varies with different coals. If improperly fired, much smoke and soot are possible, especially at low burning rates.

Semibituminous coal is soft and friable, and handling creates fines and dust. It ignites slowly and burns with a medium-length flame. Its caking properties increase as volatile matter increases, but the coke formed is weak. With only half the volatile matter content of bituminous coals, burning produces less smoke; hence, it is sometimes called smokeless coal.

Subbituminous coal, such as that found in the western United States, is high in moisture when mined and tends to break up as it dries or is exposed to the weather; it is likely to ignite spontaneously when piled or stored. It ignites easily and quickly, has a medium-length flame, and is noncaking and free-burning. The lumps tend to break into small pieces if poked. Very little smoke and soot are formed.

Lignite is woody in structure, very high in moisture when mined, of low heating value, and clean to handle. It has a greater tendency than subbituminous coals to disintegrate as it dries and is also more likely to ignite spontaneously. Because of its high moisture, freshly mined lignite ignites slowly and is noncaking. The char left after moisture and volatile matter are driven off burns very easily, like charcoal. The lumps tend to break up in the fuel bed and pieces of char that fall into the ash pit continue to burn. Very little smoke or soot forms.

The characteristics of coals that determine classification and suitability for given applications are the proportions of (1) volatile matter, (2) fixed carbon, (3) moisture, (4) sulfur, and (5) ash. Each of these is reported in the proximate analysis. Coal analyses can be reported on several bases: as-received, moisture-free (or dry), and mineral-matter-free (or ash-free). As-received is applicable for combustion calculations; moisture-free and mineral-matter-free, for classification purposes.

Volatile matter is driven off as gas or vapor when the coal is heated according to a standard temperature test. It consists of a variety of organic gases, generally resulting from distillation and decomposition. Volatile products given off by heated coals differ materially in the ratios (by mass) of the gases to oils and tars. No heavy oils or tars are given off by anthracite, and very small quantities are given off by semianthracite. As volatile matter increases to as much as 40% of the coal (dry and ash-free basis), increasing amounts of oils and tars are released. However, for coals of higher volatile content, the quantity of oils and tars decreases and is relatively low in the subbituminous coals and in lignite.

Fixed carbon is the combustible residue left after the volatile matter is driven off. It is not all carbon. Its form and hardness are an indication of fuel coking properties and, therefore, guide the choice of combustion equipment. Generally, fixed carbon represents that portion of fuel that must be burned in the solid state.

Moisture is difficult to determine accurately because a sample can lose moisture on exposure to the atmosphere, particularly when reducing the sample size for analysis. To correct for this loss, total moisture content of a sample is customarily determined by adding the moisture loss obtained when air-drying the sample to the measured moisture content of the dried sample. Moisture does not represent all of the water present in coal; water of decomposition (combined water) and of hydration are not given off under standardized test conditions.

Ash is the noncombustible residue remaining after complete coal combustion. Generally, the mass of ash is slightly less than that of mineral matter before burning.

Sulfur is an undesirable constituent in coal, because sulfur oxides formed when it burns contribute to air pollution and cause combustion system corrosion. Table 12 lists the sulfur content of typical coals. Legislation has limited the sulfur content of coals burned in certain locations.

Heating value may be reported on an as-received, dry, dry and mineral-matter-free, or moist and mineral-matter-free basis. Higher heating values of coals are frequently reported with their proximate analysis. When more specific data are lacking, the higher heating value of higher-quality coals can be calculated by the Dulong formula:

where C, H, O, and S are the mass fractions of carbon, hydrogen, oxygen, and sulfur in the coal obtained from the ultimate analysis.

Other important parameters in judging coal suitability include

Stoichiometric (or theoretical) air is the exact quantity of air required to provide oxygen for complete combustion.

The three most prevalent components in hydrocarbon fuels (C, H₂, and S) are completely burned as in the following fundamental reactions:

In the reactions, C, H₂, and S can be taken to represent 1 lb mole of carbon, hydrogen, and sulfur, respectively. Using approximate atomic masses (C = 12, H = 1, S = 32, and O = 16), 12 lb of C are oxidized by 32 lb of O₂ to form 44 lb of CO₂, 2 lb of H₂ are oxidized by 16 lb of O₂ to form 18 lb of H₂O, and 32 lb of S are oxidized by 32 lb of O₂ to form 64 lb of SO₂. These relationships can be extended to include hydrocarbons.

The mass of dry air required to supply a given quantity of oxygen is 4.32 times the mass of the oxygen. The mass of air required to oxidize the fuel constituents listed in Table 1 was calculated on this basis. Deduct oxygen contained in the fuel, except the amount in ash, from the amount of oxygen required, because this oxygen is already combined with fuel components. In addition, when calculating the mass of supply air for combustion, allow for water vapor, which is always present in atmospheric air.

Combustion calculations for gaseous fuels are based on volume. Avogadro’s law states that, for any gas, one mole occupies the same volume at a given temperature and pressure. Therefore, in reactions involving gaseous compounds, the gases react in volume ratios identical to the pound mole ratios. That is, to oxidize hydrogen in the preceding reaction, one volume (or 1 lb mole) of hydrogen reacts with one-half volume (or 0.5 lb mole) of oxygen to form one volume (or 1 lb mole) of water vapor.

The volume of air required to supply a given volume of oxygen is 4.78 times the volume of oxygen. The volumes of dry air required to oxidize the fuel constituents listed in Table 1 were calculated on this basis. Volume ratios are not given for fuels that do not exist in vapor form at reasonable temperatures or pressures. Again, oxygen contained in the fuel should be deducted from the quantity of oxygen required, because this oxygen is already combined with fuel components. Allow for water vapor, which increases the volume of dry air by 1 to 3%.

From the relationships just described, the theoretical mass m_a of dry air required for stoichiometric combustion of a unit mass of any hydrocarbon fuel is

where C, H, S, and O are the mass percentages of carbon, hydrogen, sulfur, and oxygen in the fuel.

Analyses of gaseous fuels are generally based on hydrocarbon components rather than elemental content.

If fuel analysis is based on mass, the theoretical mass m_a of dry air required for stoichiometric combustion of a unit mass of gaseous fuel is

If fuel analysis is reported on a volumetric or molecular basis, it is simplest to calculate air requirements based on volume and, if necessary, convert to mass. The theoretical volume V_a of air required for stoichiometric combustion of a unit volume of gaseous fuels is

where CO, H₂, and so forth are the volumetric fractions of each constituent in the fuel gas.

Illuminants include a variety of compounds not separated by usual gas analysis. In addition to ethylene (C₂H₄) and acetylene (C₂H₂), the principal illuminants included in Equation (8), and the dry air required for combustion, per unit volume of each gas, are as follows: propylene (C₃H₆), 21.44; butylene (C₄H₈), 28.58; pentene (C₅H₁₀), 35.73; benzene (C₆H₆); 35.73, toluene (C₇H₈), 42.88; and xylene (C₈H₁₀), 50.02. Because toluene and xylene are normally scrubbed from the gas before distribution, they can be disregarded in computing air required for combustion of gaseous fuels. The percentage of illuminants present in gaseous fuels is small, so the values can be lumped together, and an approximate value of 30 unit volumes of dry air per unit volume of gas can be used. If ethylene and acetylene are included as illuminants, a value of 20 unit volumes of dry air per unit volume of gaseous illuminants can be used.

For many combustion calculations, only approximate values of air requirements are necessary. If approximate values for theoretical air are sufficient, or if complete information on the fuel is not available, the values in Tables 13 and 14 can be used. Another value used for estimating air requirements is 0.9 ft³ of air for 100 Btu of fuel.

In addition to the amount theoretically required for combustion, excess air must be supplied to most practical combustion systems to ensure complete combustion:

The excess air level at which a combustion process operates significantly affects its overall efficiency. Too much excess air dilutes flue gas excessively, lowering its heat transfer temperature and increasing sensible flue gas loss. Conversely, too little excess air can lead to incomplete combustion and loss of unburned combustible gases. Combustion efficiency is usually maximized when just enough excess air is supplied and properly mixed with combustible gases to ensure complete combustion. The general practice is to supply 5 to 50% excess air, depending on the type of fuel burned, combustion equipment, and other factors.

The amount of dry air supplied per unit mass of fuel burned can be obtained from the following equation, which is reasonably precise for most solid and liquid fuels:

where

Dry air supplied = unit mass per unit mass of fuel

C = unit mass of carbon burned per unit mass of fuel, corrected for carbon in ash

CO2, CO, N2 = percentages by volume from flue gas analysis

These values of dry air supplied and theoretical air can be used in Equation (9) to determine excess air.

Excess air can also be calculated from unit volumes of stoichiometric combustion products and air, and from volumetric analysis of the flue gas:

where

U	=	ultimate carbon dioxide of flue gases resulting from stoichiometric combustion, %
CO2	=	carbon dioxide content of flue gases, %
P	=	dry products from stoichiometric combustion, unit volume per unit volume of gas burned
A	=	air required for stoichiometric combustion, unit volume per unit volume of gas burned

Because the ratio P/A is approximately 0.9 for most natural gases, a value of 90 can be substituted for 100(P/A) in Equation (11) for rough calculation.

Because excess air calculations are almost invariably made from flue gas analysis results and theoretical air requirements are not always known, another convenient method of expressing Equation (9) is

where O₂, CO, and N₂ are percentages by volume from the flue gas analysis, dry basis.

The theoretical (or ultimate, stoichiometric, or maximum) CO₂ concentration attainable in the products from the combustion of a hydrocarbon fuel with air is obtained when the fuel is completely burned with the theoretical quantity of air and zero excess air. Theoretical CO₂ varies with the carbon/hydrogen ratio of the fuel. For combustion with excess air present, theoretical CO₂ values can be calculated from the flue gas analysis:

where CO₂ and O₂ are percentages by volume from the flue gas analysis, dry basis.

Table 15 gives approximate theoretical CO₂ values for stoichiometric combustion of several common types of fuel, as well as CO₂ values attained with different amounts of excess air. In practice, desirable CO₂ values depend on the excess air, fuel, firing method, and other considerations.

The mass of dry flue gas produced per mass of fuel burned is required in heat loss and efficiency calculations. This mass is equal to the sum of the mass of (1) fuel (minus ash retained in the furnace), (2) air theoretically required for combustion, and (3) excess air. For solid fuels, this mass, determined from the flue gas analysis, is

where

Dry flue gas = lb/lb of fuel

C = lb of carbon burned per lb of fuel, corrected for carbon in ash

CO2, O2, CO, N2 = percentages by volume from flue gas analysis

The total dry gas volume of flue gases from combustion of one unit volume of gaseous fuels for various percentages of CO₂ is

where

Dry flue gas = unit volume per unit volume of gaseous fuel

CO2 = percentage by volume from the flue gas analysis

Excess air quantity can be estimated by subtracting the quantity of dry flue gases resulting from stoichiometric combustion from the total volume of flue gas.

Water vapor in flue gas is the total of the water (1) contained in the fuel, (2) contained in the stoichiometric and excess air, and (3) produced from combustion of hydrogen or hydrocarbons in the fuel. The amount of water vapor in stoichiometric combustion products may be calculated from the fuel burned by using the water data in Table 1.

The dew point is the temperature at which condensation begins and can be determined using Figure 3. The volume fraction of water vapor P_wv in the flue gas can be determined as follows:

where

Vw	=	total water vapor volume (from fuel; stoichiometric, excess, and dilution air; and combustion)
Vc	=	unit volume of CO2 produced per unit volume of gaseous fuel
Pc	=	percent CO2 in flue gas

Using Figure 4, the dew points of solid, liquid, or gaseous fuels may be estimated. For example, to find the dew point of flue gas resulting from the combustion of a solid fuel with a weight ratio (hydrogen to carbon-plus-sulfur) of 0.088 and sufficient excess air to produce 11.4% oxygen in the flue gas, start with the weight ratio of 0.088. Proceed vertically to the intersection of the solid fuels curve and then to the theoretical dew point of 115°F on the dew-point scale (see dashed lines in Figure 4). Follow the curve fixed by this point (down and to the right) to 11.4% oxygen in the flue gas (on the abscissa). The actual dew point is 93°F and is found on the dew-point scale.

The dew point can be estimated for flue gas from natural gas having a higher heating value (HHV) of 1020 Btu/ft³ with 6.3% oxygen or 31.5% air. Start with 1020 Btu/ft³ and proceed vertically to the intersection of the gaseous fuels curve and then to the theoretical dew point of 139°F on the dew-point scale. Follow the curve fixed by this point to 6.3% oxygen or 31.5% air in the flue gas. The actual dew point is 127°F.

The presence of sulfur dioxide, and particularly sulfur trioxide, influences the vapor pressure of condensate in flue gas, and the dew point can be raised by as much as 25 to 75°F, as shown in Figure 5. For a manufactured gas with an HHV of 550 Btu/ft³ containing 15 grains of sulfur per 100 ft³ being burned with 40% excess air, the proper curve in Figure 5 is determined as follows:

This curve lies between the 0 and 3 curves and is close to the 3 curve. The dew point for any percentage of excess air from zero to 100% can be determined on this curve. For this flue gas with 40% excess air, the dew point is about 160°F, instead of 127°F for zero sulfur at 40% excess air.

The method just presented is useful for calculating the steady-state efficiency of a heating system or device. Unfortunately, the seasonal efficiency can be significantly different from the steady-state efficiency. The primary factor affecting seasonal efficiency is flue loss during the burner-off period. The warm stack that exists at the end of the firing period can cause airflow in the stack while the burner is off, which can remove heat from furnace and heat exchanger components, the structure itself, and pilot flames. Also, if combustion air is drawn from the heated space within the structure, the heated air lost must be at least partly replaced with cold infiltrated air. For further discussion of seasonal efficiency, see Chapters 10 and 33 of the 2020 ASHRAE Handbook—HVAC Systems and Equipment and Chapter 19 of this volume.

Combustion processes constitute the largest single source of anthropogenic (human-caused) air pollution. Pollutants can be grouped into five categories:

Emission levels of nitrogen oxides and products of incomplete combustion are directly related to the combustion process and can be controlled, to some extent, by process modification. Emissions from fuel contaminants are related to fuel selection and are slightly affected by the combustion process. Emissions from additives must be considered in the overall evaluation of the merits of using additives.

Carbon dioxide as a pollutant has gained attention because of its suspected effect on global warming. Carbon dioxide is produced by HVAC&R equipment (either directly or as a result of generating the electric power to operate the HVAC&R equipment), transportation, industry, and other sources. Carbon dioxide emissions can be minimized by increasing appliance operating efficiencies and using fuels with higher hydrogen content.

Nitrogen oxides are produced during combustion, either (1) by thermal fixation (reaction of nitrogen and oxygen at high combustion temperatures), or (2) from fuel nitrogen (oxidation of organic nitrogen in fuel molecules). Unfortunately, high excess air and high flame temperature techniques, which ensure complete fuel combustion, tend to promote NO_x formation. NO levels in flames where the reactants are premixed tend to peak with excess air levels around 10%. Higher excess air levels generally reduce the amount of NO_x and flame temperatures.

Table 16 lists some NO_x emission factors for fuel-burning equipment. Differences in emissions are caused by flame temperature and different levels of fuel nitrogen. Carbon monoxide emissions depend less on fuel type and typically range from 0.03 to 0.04 lb/10⁶ Btu of heat input. For gas-fired commercial and industrial boilers, particulate emissions range from 0.005 to 0.006 lb/10⁶ Btu. For distillate-oil-fired commercial and industrial boilers, particulates are typically 0.014 lb/10⁶ Btu. For residential oil-fired equipment, particulate emission factors are 0.003 lb/10⁶ Btu. For residual-oil-fired equipment, particulate emissions depend on the sulfur content and, to a lesser extent, the mineral content. For a sulfur content of 1%, the particulate emission rate is typically 0.083 lb/10⁶ Btu.

Emission levels of products of incomplete fuel combustion can be reduced by reducing burner cycling, ensuring adequate excess air, improving mixing of air and fuel (by increasing turbulence, improving distribution, and improving liquid fuel atomization), increasing residence time in the hot combustion zone (possibly by decreasing the firing rate), increasing combustion zone temperatures (to speed reactions), and avoiding quenching the flame before reactions are completed.

Relative humidity of combustion air affects the amount of NO_x produced and must be considered when specifying acceptable NO_x emission rates and measuring NO_x production during appliance tests.

The relative contribution of each of these mechanisms to the total NO_x emissions depends on the amount of organic nitrogen in the fuel. Natural gas normally contains very little nitrogen. Virtually all NO_x emissions with gas firing are due to the thermal mechanism. Nitrogen content of distillate oil varies, but an average of 20 ppm of fuel NO_x is produced (about 20 to 30% of the total NO_x). Levels in residual oil can be significantly higher, with fuel NO_x contributing heavily to the total emissions.

Thermal fixation depends strongly on flame maximum temperature. For example, increasing the flame temperature from 2600 to 2800°F increases thermal NO_x tenfold. Therefore, methods to control thermal NO_x are based on methods to reduce the maximum flame temperature. Flue gas recirculation is perhaps the most effective method for commercial and industrial boilers. In gas-fired boilers, NO_x can be reduced 70% with 15 to 20% recirculation of flue gas into the flame. The NO_x reduction decreases with increasing fuel nitrogen content. With distillate-oil firing, reductions of 60 to 70% can be achieved. In residual-oil-fired boilers, flue gas recirculation can reduce NO_x emissions by 15 to 30%. The maximum rate of flue gas recirculation is limited by combustion instability and CO production.

Two-stage firing is the only technique that reduces NO_x produced both by thermal fixation and fuel nitrogen in industrial and utility applications. The fuel-rich or air-deficient primary combustion zone retards NO_x formation early in combustion (when NO_x forms most readily from fuel nitrogen), and avoids peak temperatures, reducing thermal NO_x. Retrofit low-NO_x burners that control air distribution and fuel air mixing in the flame zone can be used to achieve staged combustion. With oil firing, NO_x reductions of 20 to 50% can be obtained with low-NO_x burners. Application of flue gas recirculation and other control methods to residential, oil-fired heating systems was reviewed by Butcher et al. (1994).

The following are some methods of reducing NO_x emissions from gas-fired appliances (Murphy and Putnam 1985):

Radiation screens or rods (flame inserts) surrounding or inserted into the flame absorb radiation to reduce flame temperature and retard NO_x formation. Proprietary appliance burners with no flame inserts have been produced to comply with the very strict NO_x emission limitations of California’s Air Quality Management Districts.

The U.S. EPA sets limits on air pollutant emissions (Source Performance Standards) from boilers larger than 10 million Btu/h of heat input. In addition, states set emission regulations that are at least as strict at the federal limits and may apply to smaller equipment.

The EPA’s automobile emission standard is 1.0 g of NO₂ per mile, which is equivalent to 750 ng/J of NO_x emission. California’s maximum is 0.4 g/mile, equivalent to 300 ng/J. California’s Air Quality Management Districts for the South Coast (Los Angeles) and the San Francisco Bay Area limit NO_x emission to 14 ng/J of useful heat for some natural gas-fired central furnaces.

For further discussion of air pollution aspects of fuel combustion, see EPA (1971a, 1971b).

These battery-powered electronic instruments, also known as flue gas analyzers (FGAs), have been widely used for over 30 years. Their sensors measure various gases found in flues of combustion appliances, whether fired by gas, oil, or solid fuels.

Users place the analyzer’s probe in the flue of an appliance to extract combustion gases for measurement. These gases pass through a water trap to create a dry gas for measurement, and then have excess particles removed by a filter before entering the PCA with help from its internal pump.

The PCA’s sensors measure or calculate real-time readings of oxygen (O₂), carbon monoxide (CO), and carbon dioxide (CO₂). Most PCAs also measure temperature, pressure, and draft; some measure the ratio of CO to CO₂, detect gas leaks, or measure other gases such as nitric oxide (NO), nitrogen dioxide (NO₂), sulfur dioxide (SO₂), and hydrocarbons (CH₄) and transfer test results to a printer, PC, smartphone, or tablet. If in doubt, ask the manufacturer which model is appropriate for a specific application.

Pay particular attention to the user manual’s care instructions. These include switching on in outdoor air to allow sensors to tare (zero out) correctly, not leaving a PCA overnight in a cold vehicle, and ensuring the PCA is annually recertified by an authorized service provider to keep its metrological performance within specification.

An annual calibration certificate is essential. As a minimum, it should include

All calibrated PCAs should have a label or sticker applied including the name of the company who performed the work and the PCA’s next service date. The label should reference the PCA using a serial number or certificate number, be tamperproof, and be placed where it can easily be seen. PCAs generally rely on service software, usually only available from the manufacturer or an approved service center. Using nonauthorized companies may invalidate the PCA’s warranty and may lead to incorrect replacement parts being used.

An American PCA performance standard, AHRI Standard 1260P, is in development, but most PCAs are already certified to European Standard CEN Standard 50379, which has three parts:

CEN Standard 50379 defines which gases are included (CO, NO, SO₂, O₂, and CO₂ if fitted to the PCA) and what accuracy, response, and performance criteria PCAs must meet. The standard also tests for sensor cross sensitivity, long life, drop, and vibration to ensure reliability in normal applications under competent use. Some PCAs are used to measure ambient levels of CO and CO₂ in residential or commercial environments. In Europe, these PCAs must meet the performance requirements of CEN Standard 50543; some PCAs are already certified to this standard as well as to CEN Standard 50379.

Fuel-burning systems that cycle on and off to meet demand cool down during the off cycle. When the appliance starts again, condensate forms briefly on surfaces until they are heated above the dew-point temperature. Low-temperature corrosion occurs in system components (heat exchangers, flues, vents, chimneys) when their surfaces remain below the dew-point temperature of flue gas constituents (water vapor, sulfides, chlorides, fluorides, etc.) long enough to cause condensation. Corrosion increases as condensate dwell time increases.

Acids in flue gas condensate are the principal substances responsible for low-temperature corrosion in fuel-fired systems. Sulfuric, hydrochloric, and other acids are formed when acidic compounds in fuel and air combustion products combine with condensed moisture in appliance heat exchangers, flues, or vents. Corrosion can be avoided by maintaining these surfaces above the flue gas dew point.

In high-efficiency, condensing-type appliances and economizers, flue gas temperatures are intentionally reduced below the flue gas dew-point temperatures to achieve efficiencies approaching 100%. In these systems, surfaces subjected to condensate must be made of corrosion-resistant materials. The most corrosive conditions exist at the leading edge of the condensing region, especially areas that experience evaporation during each cycle (Stickford et al. 1988). Draining condensate retards the concentration of acids on system surfaces; regions from which condensate partially or completely drains away before evaporation are less severely attacked than regions from which condensate does not drain before evaporation.

The metals most resistant to condensate corrosion are stainless-steel alloys with high chromium and molybdenum content, and nickel-chromium alloys with high molybdenum content (Stickford et al. 1988). Aluminum experiences general corrosion rather than pitting when exposed to flue gas condensate. If applied in sufficiently thick cross section to allow for metal loss, aluminum can be used in condensing regions. Most ceramic and high-temperature polymer materials resist the corrosive effects of flue gas condensate. These materials may have application in the condensing regions, if they can meet the structural and temperature requirements of a particular application.

In coal-fired power plants, the rate of corrosion for carbon steel condensing surfaces by mixed acids (primarily sulfuric and hydrochloric) is reported to be maximum at about 122 ± 18°F (Davis 1987). Mitigation techniques include (1) acid neutralization with a base such as NH₃ or Ca(OH)₂; (2) use of protective linings of glass-filled polyester or coal-tar epoxy; and (3) replacing steel with molybdenum-bearing stainless steels, nickel alloys, polymers, or other corrosion-resistant materials. Other elements in residual fuel oils and coals that contribute to high-temperature corrosion include sodium, potassium, and vanadium. Each fuel-burning system component should be evaluated during installation, or when modified, to determine the potential for corrosion and the means to retard corrosion (Paul et al. 1988).

If fuel-burning appliances accumulate condensate that does not evaporate, the condensate must be routed into a trapped drainage system. Because the condensate may be acidic, the drainage system must be suitable and environmentally acceptable. Condensate freezing must be considered in cold climates.

During development of a new boiler, furnace, or other gas-fired appliance, tonal noise can be unacceptable. Because the frequency of the tone is equal to a resonance frequency of the system, this problem is often called a combustion resonance, but this term is misleading: changing the appliance’s resonance frequency merely changes the frequency of the tone without much effect on the amplitude.

The proper term is combustion-driven oscillation, which is caused by feedback instability. Pressure oscillations in the combustion chamber (which manifest themselves as objectionable noise) also interact with the flame, modulating the instantaneous rate of combustion, which, in turn, causes more pressure oscillations (Putnam 1971). This feedback involves the acoustic response of the combustion chamber and of the fuel-air supply system, as well as that of the flame. For some combinations of response properties, the feedback loop is unstable. Instabilities may result from either perturbation of the mixture flow or of the air/fuel ratio. Instability because of a fluctuating air/fuel ratio is more likely to occur at low frequencies (Herrin et al. 2012).

The feedback loop suggested by Baade (1978) and modified by ASHRAE research project RP-1517 (Herrin et al. 2012) is very useful for understanding and solving oscillation problems. Baade identified three transfer functions that must be determined. Transfer functions Z and H are acoustic and can be determined experimentally or by simulation. Transfer function G_f is related to the flame and is best determined experimentally, though a few models are available. Figure 7 shows a schematic of Baade’s feedback loop stability model for predicting combustion oscillations. Perturbations to the volume velocity of the flame, which are external to the feedback loop, are indicated by q̃_ext. The driving point impedance Z of the combustion chamber is the ratio of the oscillating pressure p̃ to the volume velocity in the combustion chamber q̃_tot or Q̃. The transfer function H relates the perturbation of the mixture flow q̃_i to the acoustic pressure in the combustion chamber p̃. As shown in Figure 7, the acoustic impedance of the chamber relates the flame perturbations to the sound pressure in the chamber. Acoustic impedance is primarily a function of the geometry of the combustion chamber and length of the exhaust. The sound pressure in the chamber in turn modulates either the mixture supply or air/fuel ratio. The respective transfer functions relating the sound pressures to the mixture supply or air/fuel ratio fluctuations are mainly controlled by the geometry of the intake. Mixture supply or air/fuel ratio fluctuations then drive the flame perturbations completing the feedback loop. The flame transfer function, which relates flame perturbations to the mixture supply and air/fuel ratio fluctuations, is related to several factors, including the burner type.

Predicting instability in a design is generally not practical for domestic or small commercial appliances because there is not enough information to predict the acoustic response of some of the components, particularly the flame.

A model of the feedback loop is very useful, however, for solving existing oscillation problems, where the only concern is the particular frequency at which the oscillation occurs. Reducing the response of the flame, air/fuel mixture supply, or combustion chamber at that frequency should be the focus. This concept can be easily demonstrated with a small brazing torch in a tube of variable length (Baade 1987, 2004).

In some systems, the flame can be modified to reduce its response at the oscillation frequency. Often, this involves simply changing the fuel/air ratio further away from the stochiometric ratio (Elsari and Cummings 2003; Goldschmidt et al. 1978), thus lengthening the flame, which can also be done by increasing the size of burner ports (Matsui 1981; Schimmer 1979). Other possibilities for reducing flame response are using a suitable mix of differently sized burner ports (Kagiya 2000) and modifying the heat transfer characteristics of the burner matrix (Schreel et al. 2002).

The fuel supply system response can be reduced by avoiding resonance at or near the frequency of oscillation (Kilham et al. 1964) or by tuning the supply system to an antiresonance at that frequency (Neumann 1974). Higher-flow-resistance burners add acoustic damping to the fuel supply system. Designs for this can be evaluated by modeling the mixture supply system using transmission matrices (Munjal 1987) and computer programs for matrix multiplication, which are widely available (Baade and Tomarchio 2008).

Be careful to avoid any unnecessary resonances in the combustion chamber (Herrin et al. 2012). Structural resonances are sometimes the primary cause of a combustion oscillation, and can be identified by using an impact hammer and measuring the components’ vibrational response using an accelerometer. Increasing damping can effectively resolve instabilities, if the system can increase damping enough: any damping less than the critical amount will have very little effect. Acoustic damping or sound absorption can be added by installing perforated panels or fiber. However, damping treatments are only likely to be effective for instabilities at high frequencies.

In some systems, the oscillation frequency may be a function of the flue pipe length. In such cases, consider changing the length as well as adding damping.

For large systems, active feedback may be used to eliminate combustion oscillations (Sattinger et al. 2000). However, active feedback is not likely to be cost effective for residential and small commercial systems.

Soot deposits on flue surfaces of a boiler or heater act as an insulating layer over the surface, reducing heat transfer to the water or air. Soot can also clog flues, reduce draft and available air, and prevent proper combustion. Proper burner adjustment can minimize soot accumulation. Using off-specification fuel can contribute to soot generation.