Getting the most out of your combustion system means focusing on higher productivity, lower costs, reduced environmental impact and increased product quality. Readers have seen how thermal capture can save costs and increase productivity, but what environmental impact does this technology have? Is there any effect on product quality? The good news is that the development of energy-efficient industrial burners can now meet air-quality standards while giving operators better control of how and where the heat is delivered.

Let’s take an in-depth look at the technical basis details for enabling the use of preheated air burners that meet environmental and quality standards. We’ll look first at methods for controlling NOx emissions, with details from simple Excel burner models in downloadable workbook NOxControl.xlsx. Then we’ll look at ways of using burner controls to enhance heat transfer to a furnace’s load and maintain temperature uniformity in a furnace.

The Nature of a Combustion Flame

If a combustible mixture of fuel and air is ignited, a flame can develop. In the reaction zone, called the flame front, the temperature rises quickly to near adiabatic temperature. The flame can be attached within or close to the burner to give stable and controlled combustion. Figure 1 depicts a stable flame from a premixed burner. One of the advantages of this type of flame is the ability to direct the heat to where it’s most needed. A disadvantage is the greater production of NOx, owing to its long-lasting higher temperature.

Various techniques have been developed to modify the type of stable flame depicted in Figure 1. The goal is to minimize conditions that promote NOx formation while dissipating the heat more quickly and evenly throughout the furnace. Figure 2 depicts a flame disrupted by increasing the air and fuel velocity. Part of the fuel or oxidant is introduced, which creates turbulence in or at the boundary of the main combustion zone, or fuel and oxidant nozzle configuration is changed. When secondary air or fuel is used, partial mixing of the primary streams occurs in the burner, while the rest occurs at the nozzle. For higher preheated air temperatures, the fuel and air are mixed entirely at the nozzle. A turbulent combustion zone effectively entrains furnace products of combustion (POC).

One of the interesting consequences of nozzle-mixed combustion occurs when the burner gas velocity exceeds the rate of propagation of the flame front, and the flame “lifts off” its attachment point at the burner. Any admixture of furnace gas POC to the air lowers the oxygen content of the air/fuel mixture, further slowing the flame propagation rate and increasing the lift-off propensity.

If the furnace blanket-gas temperature is below the air/fuel self-ignition temperature (about 1500°F/820°C), flame liftoff calls for immediate cessation of fuel flow to avoid filling the furnace with a combustible mixture. Above this temperature, however, the air/fuel mixture is safely ignited by contact with the surrounding (hot) POC, giving a large, diffuse and invisible combustion zone stretching out into the furnace.^[1-3]

This so-called “flameless oxidation” is defined as:  A stable combustion without a visible flame and with defined recirculation of hot combustion products.This condition is sometimes called “distributed” combustion.^[4] The result is extensive mixing of POC and burner streams that slows the rate of combustion, thus giving time for heat to be transferred to the surroundings instead of raising the combustion zone temperature.

Stack-Gas NOx Emissions

Three oxides of nitrogen (NOx) – nitric oxide (NO), nitrogen dioxide (NO₂) and nitrous oxide (N₂O) – are considered important atmospheric pollutants. The predominant NOx emission from thermal processing is NO formed by the reaction between nitrogen and oxygen during combustion and is referred to as “thermal NOx.” NOx emitted from ordinary natural gas combustion burners is typically in the range of 100-500 vppm.

Thermal NOx itself is not regarded as a significant pollutant from either a health or economic welfare standpoint. Its significance as an air pollutant results from the chemical reactions in which it participates once it leaves the source. These reactions produce nitrogen dioxide (NO₂), which is a contributor to ground-level ozone, acid rain and small particulate formation. These concerns are the reason for the controls that have been placed on nitrogen oxides and for the emission tax, which has been levied by some countries.^[5]

NOx Generation

Combustion engineers have been successful in developing products that are effective in curbing NOx emission. Table 1 shows trends in natural gas use, NOx production and, even more important, the mass of NOx produced per mm BTU. Several important variables influence thermal NOx formation.^[6] They are flame temperature, air preheat, furnace temperature, excess air and chemical environment within the flame.

Flame temperature is the most important variable, and it is related to the other factors. Nitrogen oxides form at combustion temperatures through a complex chain of chemical reactions involving atomic and dissociated chemical species. The rate of thermal fixation of molecular nitrogen becomes significant at temperatures exceeding about 2700°F (1480°C). (Note: The amount of NOx produced also depends on the residence time. The critical temperature can be higher for very short residence times or lower for long times.)

The concentration of nitric oxide (NO), the most prevalent oxide of NOx formed during combustion, depends on the concentrations of oxygen and nitrogen, the reaction temperature and the time in the peak-temperature zone. The oxygen dependence is on the square root of the oxygen concentration and a definite peak in nitric oxide formation occurs slightly above the stoichiometric ratio of air (oxygen) to fuel.

Combustion-zone temperature is important because the NO concentration rises exponentially with temperature. When burning natural gas, equilibrium concentrations of thermal NOx exceed 3,000 vppm when using cold air or 4,000 ppm for preheated air. These levels are seldom seen in practice because reaction kinetics limit the approach to equilibrium. Burners unequipped with low-NOx features typically produce thermal NOx levels of about 100 ppm to 300 ppm.

The NOx level is highest in the hottest part of the combustion flame, where temperatures approach the AFT. Owing to a short residence time at that temperature, however, the NO formation reaction does not approach its equilibrium value. In addition, the flame temperature drops rapidly as its POC mixes with the cooler blanket gas that surrounds it, which leads to partial decomposition of the formed NOx in the blanket gas.

The NOx concentration in the flue gas depends on three factors: temperature reached during combustion, time at temperature and percent O₂ in the POC. The effects of these factors are nearly impossible to quantify, so it’s more common to measure the concentration and devise techniques to lower it. With the strong dependence of NOx formation on temperature, the use of air preheat for fuel efficiency works counter to NOx mitigation. An increase in air preheat from ambient to 500°F (260°C) increases NOx formation by a factor of approximately two.

NOx reduction-control technologies concentrate on preventing the formation of NOx during the combustion process, either by limiting the access of the reaction to oxygen or by reducing the reaction temperature. The methods and technology options involve operating changes/combustion modifications to limit NOx formation. NOx control has been practiced since the late 1970s and manufacturers have developed control equipment that, in most cases, can meet the required NOx emission levels for new installations. The advantage of combustion-modification technologies is that they can reduce the formation of NOx without incurring a continuing operating penalty.

Low NOx burners (LNB) can reduce the available oxygen, cool the combustion zone or both, thereby mitigating the emissions of NOx. Low NOx is typically defined as any value of NOx that is under 100 ppm. The most common values are between 9 and 100 ppm. LNBs typically provide NOx reduction in the range of 50-75% and some recently developed burners are achieving nearly 90% reduction as compared to uncontrolled operation. LNB are generally classified as internal exhaust-gas recirculation (EGR), staged-air or staged-fuel burners. Often more than one feature is incorporated in an ultra-low-NOx burner.

Burner Operation

In conventional combustion systems, the well-defined flame shown in Figure 1 has a sharp gradient of temperature and chemical composition across the flame front, which balances heat generation inside the flame and quenching at the boundary of the front due to convection. A combusting flame with laminar flow will entrain a small amount of the surrounding gas. At a higher velocity, the flame becomes turbulent and entrains a greater amount of the surrounding gas. NOx control involves managing the combustion zone’s character to control its oxygen content and temperature.^{[7, 8]}

Minimizing Excess Air

While operation with excess air is used to ensure completion of combustion, it can result in added NOx emissions. Figure 3 shows that the equilibrium AFT NOx-emissions peak at excess air levels of about 30% above the stoichiometric air/fuel ratio. Therefore, reducing the excess air toward the stoichiometric ratio can reduce potential NOx emissions. On the other hand, the AFT increases as the stoichiometric air/fuel ratio is approached, thus increasing the kinetics of NOx formation. Current practice is to control excess air to improve combustion efficiency while lowering the flame temperature by other means. Excess-air levels that give a POC with 2-3 volume %O₂ (dry) appear to provide the best balance of maximum thermal efficiency and low NOx and CO emissions.

Exhaust Gas Recirculation

Exhaust (or flue) gas recirculation (EGR) is a method of NOx reduction that lowers the temperature of the flame and, therefore, reduces thermal NOx.^[9,10] A portion of the exhaust gas (i.e., the blanket gas POC) is recirculated into the active combustion volume, cooling it and lowering the oxygen content. This process may be either external or induced, depending on the method used to move the exhaust gas. Although internally induced POC is “hot,” it is cooler than the AFT.

The best source of exhaust gas is the furnace itself. Some burners have inspiration ports located upstream of the combustion-air nozzle that use the Venturi effect to draw in POC from the furnace blanket gas. Other designs change the shape and formation of the flame by using plates to control airflow. A combustion zone of the type depicted in Figure 2 entrains adjacent POC by the eddy effect. (Note: An eddy is the swirling of a fluid and the reverse current created when the fluid flows past an obstacle or adjacent stagnant fluid. The moving fluid creates a space devoid of downstream-flowing fluid, creating a swirl of fluid on the edges of the stream. Eddy currents are most easily seen when a jet of water enters a pool of stagnant water. The eddy flow creates vortices which have areas of low pressure, which draws in fluid from the surroundings.)

Let’s take a look at the specifics of a cold-air EGR burner on an 1830°F (1000°C) furnace using methane (CH₄) as a fuel at 10% excess air (air/fuel ratio = 10.5). A heat balance from downloadable Excel workbook NOxControl.xlsx indicates an equilibrium AFT of 3470°F (1910°C). The POC from combustion of 2.2 lb-mole (one kg-mole) of CH₄ produces 25.35 lb-mole (11.5 kg-mole) of POC with the composition shown in Table 2. The LHV is 760,220 Btu (802,330 kJ), of which 48% escapes as flue gas heat.

For NOx control purposes, we are most concerned with the AFT, which occurs at the flame front of the area where the air-fuel mixture burns. In the absence of EGR, this will be close to 3470°F (1910°C) for cold-air use and close to 4426°F (2440°C) for air heated to 1470°F (800°C). The AFT decreases with increasing EGR because the exhaust gas absorbs heat, enlarges the combustion-zone volume and decreases the %O₂ in the oxidant. If combustion occurred adiabatically in one zone, keeping the AFT safely below the NOx formation threshold of 2700°F (1480°C) would require about 12 moles of EGR while about double that for air heated to 1470°F (800°C).

An Excel model of an EGR burner can estimate the amount of EGR needed to keep the AFT safely below the threshold temperature. Considering that the combustion process occurs over a length of several meters from the nozzle, the model divided the combustion volume into three zones, all surrounded by the furnace-blanket gas. Each zone contained an embedded heat exchanger for calculating the temperature drop caused by the transfer of heat from each zone to the load and loss.

Air, natural gas and recycled exhaust gas are mixed in the first two zones and combust as they flow through. The first-zone POC transfers out one-third of the overall net available heat and then passes into the second combustion zone. Here, any remaining methane is combusted with secondary air, more exhaust gas is added and one-third of the overall net available heat is transferred out. The POC passes into a final mixing zone where exhaust gas is added, and the last third of the net available heat is transferred. The POC then exit to the blanket gas at 1830°F (1000°C), eventually passing out the flue. All three EGR streams originate from blanket gas. Figure 4 depicts the flowsheet.

The POC exiting zone 3 (stream #10) has the blanket-gas composition (Table 2) and the blanket-gas temperature. The model assumes that the air, exhaust gas and methane streams simultaneously mix and combust and that no heat is lost to the surroundings before combustion is complete. Stream #5 enters the HX 1 at the AFT, transfers the net available heat to the load and loss, and exits at the stream #6 temperature. The AFT and stream #6 temperature is calculated by making a heat balance on the zone-1 combustion process and the HX. This process is repeated for the other zones until all of the material and heat-balance constraints are satisfied. Because of calculational convenience, kg-mole units were used for stream flows.

Table 3 shows model results for using cold air, and Table 4 shows results for air preheated to 1470°F (800°C). Downloadable Excel workbook NOxControl.xlsx shows the heat-balance details.

This rather simple model gives some insight on the amount of EGR required to appropriately lower the combustion-zone temperature. About double the EGR is required when going from unheated air to air preheated to 360°F (200°C) below the furnace temperature. Also, compare the difference in the flowrate of streams 1-4 and 11 for the two cases.

A number of variations exist for using EGR to lower NOx, such as nozzle designs to control the extent of exhaust-gas mixing with air and fuel, and using high-velocity air and fuel streams to create more turbulence at the blanket gas/combustion zone interface. The patent literature is replete with descriptions of burners employing these variations.

Staged-Combustion Burners

Early LNB burner designs involved separating the combustion process into stages of different air/fuel ratio.^[11] The staged-air burner bypasses some of the combustion air around the conventional-burner combustion zone so that only a portion of the combustion air is available in the primary flame zone. The theoretical basis for air staging is that the initial combustion of the fuel takes place in a fuel-rich zone in which the flame temperature in the primary combustion zone is somewhat reduced. More importantly, the readily available oxygen concentration is reduced to an extremely low value, thereby greatly reducing the NOx formation tendency. The remaining portion of the combustion air (50-75%) is introduced downstream in one or more additional zones where combustion is completed. NOx formation in these zones is reduced because the inerts from the primary zone reduce the flame temperature.

Staged-fuel burners also separate the combustion zone into two regions. The first is a lean primary region in which the total quantity of combustion air is supplied with a fraction of the fuel. In a typical staged fuel LNB, 40 to 70% of the fuel is bypassed around the primary combustion region. Combustion in the primary region therefore takes place in the presence of a large excess of air (and some entrained POC). This substantially lowers the combustion temperatures below the NOx threshold formation temperature. The remaining fuel is introduced downstream in one or more additional zones where combustion is completed by oxygen left over from the primary region. NOx formation in the secondary zone is reduced because the inerts from the primary zone reduce the flame temperature. As in the case for EGR burners, the patent literature contains competing claims about the relative reductions in NOx emissions for fuel-staged vs. air-staged burners. Burner technology continues to develop, so readers are advised to contact burner manufacturers for specifications, cost, and operational requirements when selecting a LNB that will attain local emission levels.

Let’s take a look at the specifics of a staged-air burner using methane with 10% excess air on a furnace operating at 1830°F (1000°C) and air heated to 1470°F (800°C). (Note: Calculations are shown for heated air operation only because it’s relatively easy to meet the low-NOx limit for unheated air operation. Workbook NOxControl.xlsx does not contain a model for a staged fuel burner.)

Workbook NOxControl.xlsx contains the model and calculational details, and Figure 5 depicts the flowsheet. The calculational basis is 2.2 lb-mole (1 kg-mole) of methane at an air/fuel ratio of 10.5. The model consists of three combustion zones and three heat exchangers but without EGR. The entire natural gas flow enters combustion zone 1 along with about 50% of the stoichiometric air. The POC thus contain substantial amounts of CO and H₂ in addition to CO₂ and H₂O. These gases distribute the oxygen amongst themselves by means of the reversible water-gas shift reaction,^[12] equation [1]. The value of Keq varies slightly with temperature and is approximately 3 near 3000°F (1650°C).
 

CO₂ + H₂ →CO + H₂O; Keq = pCO x pH₂O / pCO₂ x pH₂      [1]
 

Both the staged fuel and the staged air LNB inevitably entrain exhaust gas into the combustion volume, or EGR can deliberately be made a part of a staged-air or staged-fuel burner, thus resulting in so-called ultra-low-NOx operation. Ultra LNB technology is capable of achieving lower-NOx emissions than LNBs, on the order of 10-30 vppm NOx. This can lead to NOx reductions of about 90% compared with conventional uncontrolled raw gas burners. Owing to ongoing burner development, details of these designs are best obtained from burner manufacturers. It’s important to recognize that LNBs tend to have longer flames than conventional burners, and this needs to be considered when retrofitting low-NOx burners in existing fired heaters. NOx reduction modifications are not as simple as they first appear.

Managing Burner Heat Output

The customary way to control the amount of heat a particular burner delivers is to modulate the fuel flow while keeping the air/fuel ratio fixed to maintain thermal efficiency. Most burner manufacturers will quote a nominal or standard burner capacity, either in terms of its firing rate (BTU/hour) or scfh airflow. In addition, the maximum capacity may be listed as well as the turndown ratio. Operating at the burner’s nominal capacity is likely to result in the best overall performance.

In practice, load demand, start-up operation or temporary burner maintenance may require some or all of a furnace’s burners to be operated away from nominal capacity. When this happens, the heat distribution in the furnace may deteriorate. Product quality requires all parts of the load, no matter where they are located in the furnace, to reach and stay at the desired temperature. Therefore, uniformity and properly controlled heat-transfer rates are important furnace characteristics.

Heat-Transfer Considerations

Each item in a furnace load is surrounded by a stagnant film of POC that retards convective heat transfer. While radiation is likely the major mechanism for transferring heat to the load, convection is still important, especially when parts of the load are poorly visible to the gas blanket or furnace walls. Thus, it is beneficial to have a high POC velocity exiting from the outlet of the burner to disrupt the stagnant gas films that hinder convective heat transfer.

A second advantage of higher velocity is stirring of the furnace atmosphere to level the temperature throughout the furnace. A high-velocity burner stream is effective in a third way – it entrains surrounding POC better than the same burner-stream amount flowing at a lower velocity. Thus, modulating the burner’s firing rate by turning down the flow of the fuel-air mixture (called amplitude modulating) can create a misdistribution of furnace heat and decreased convective heat transfer.

One of the downsides of using preheated air is a decrease in POC flow when switching from unheated to heated air for the same heat to load. For example, when using unheated air, a 1600°F (870°C) furnace firing at 500,000 BTU/hour at 15% excess air produces 6,600 scf of POC. For a heat loss of 45,000 BTU/hour, the heat to load is 242,930 BTU/hour. If the air was heated to 1200°F (650°C), the same heat to load would be delivered at a firing rate of 343,960 BTU/hour for a POC production rate of 4,500 scf, which is 69% of the unheated air flow. This lower flow, plus the accompanying decrease in entrained POC, can mean a less-uniform temperature distribution in the furnace and a decrease in convective heat transfer. Figure 6 shows the trend.

Modulating Burner Heat Output by Frequency (Pulse) Control

As an alternative to amplitude modulation, the burner can be switched between two states, with the cycling of the burner controlling the heat input.^[13,14] The burners are fired at high fire (i.e., the optimum firing rate) for a period of time, then cycled to either low fire (high-low control) or turned off (on-off control). This cycle is repeated quite frequently to maintain temperature uniformity, but at the cost of reduced equipment lifetime.

Another advantage of pulse firing is that some burners do not perform well at low turndown, so that thermal efficiency can decrease during load soak periods. With pulse firing, each burner can operate for 25% of the time at optimum firing rate, instead of having all of the amplitude-modulated burners operating at 25% of their optimum firing rate.

For certain applications, significant advantages can be achieved by applying pulse-firing technology. Specialized controllers or a PLC will be necessary to get the most out of pulsed-fired technology. As always with new equipment, there will need to be a strong working relationship between the end user and the equipment supplier.

Conclusions

Burner technology has evolved to the point where thermal-capture technology is able to deliver significant fuel savings by preheating the combustion air to within 300°F (170°C) of the furnace POC temperature. However, the higher the preheated air temperature, the more difficult the NOx-reduction problem becomes. EGR models such as those developed for this article can calculate the EGR flow necessary to keep the combustion zone temperature below the NOx formation threshold. The staged combustion model can calculate the stream flows that are necessary to maintain the presence of a fuel-rich zone until the last (air-rich but cooler) combustion stage is reached. If desired, EGR streams can be added to the staged combustion model.

The increasingly common use of regenerative burners presents a particularly difficult NOx reduction problem to burner designers. Regenerative burners have excellent thermal efficiency because they achieve very high preheat temperatures. This likely requires a higher degree of EGR flow than used in the NOxControl.xlsx example or the addition of EGR to staged combustion. Even if the NOx level is higher for regenerative burners, however, their increased thermal efficiency (high ε value) means that the lb NOx/mm BTU can actually be less than for using a recuperative burner (lower ε value) for the same furnace temperature.

Operations that require high turndown periods may benefit from going to pulse-firing burner operation. While pulse-fired burners are on high flow, the flame can stir the furnace blanket gas more effectively than burners operating at turndown. Pulse-firing is applicable to both recuperative and regenerative thermal-capture operations.

References

1. Delacroix, Franck, "The Flameless Oxidation Mode: An Efficient Combustion Device Leading Also to Very Low NOx Emission Levels"

2. Wünning, Joachim, “Flameless Oxidation,” 6^th HiTACG Symposium - 2005, Essen, Germany, 17-19 October 2005

3. Milani, Ambrogio and Joachim Wünning, “What is Flameless Combustion?,” IFRF Online Combustion Handbook #171, 2002

4. Kang, Taekyu et. al, “Flameless Combustion Burner,” U.S. Patent #8915731, 12/23/2014

5. U.S. EPA, EPA-456/F-99-006R “Nitrogen Oxides (NOx), Why and How They Are Controlled,” November 1999

6. ExxonMobil Research And Engineering Company - Fairfax, VA, “Control Of Nitrogen Oxides,” December 2003

7. Bloom Engineering, “Low NOx Burners,” November 2009

8. Pisano, Stephen, "Emerging Ultra-Low-NOx Burner Technology for the Heat Treat Industry," Industrial Heating, August 2013, p. 61

9. Schalles, David, “Combustion Application Experience With Highly Vitiated Air Burners,” AFRC 2013 Industrial Combustion Symposium, September 2013

10. Baukal, Charles, “NOx 101: A Primer on Controlling This Highly Regulated Pollutant,” John Zink Co., February 2008

11. Reed, Richard, “Modification of Burner Internals,” North American Combustion Handbook, Vol. II, Third Edition, p. 167 (1997)

12. Wikipedia contributors, “Water gas shift reaction,” Wikipedia, the Free Encyclopedia, September 2013

13. Palkovic, Paul, “Pulse Firing Basics,” Hauck Manufacturing/Elster-Kromschroder, presented at 2011 Structural Clay Products Division Meeting

14. Curry, Dan, “The Basics of Pulse Firing,” Industrial Heating, October 2011, p. 73