Abstract
The article contains a comparative analysis of two types of burners used in different methods of fuel-air mixture preparation: (1) vortex mixing and (2) mixing with transverse jets. The analysis was carried out in order to determine which one of the two burning devices is more efficient and has better environmental performance. In device no. 1, conditions for the fuel-air mixture formation are created by vortex turbulence. The basic principle lying at the core of this design is a vortex flow inside, which provokes a more intense mixing of fuel and air. Moreover, preliminary physical and thermal treatment of the fuel-air mixture has a positive effect on its environmental performance. In contrast, in device no. 2 based on transverse jets’ active mixture formation is achieved through collision of air and fuel flows at an angle close to 90°. The research was based on an experiment carried out with the use of a laboratory firing stand. Flue gas samples were analyzed in order to compare the main harmful air emission indicators with TESTO 350-XL gas analyzer. A propane-butane mixture of 60% C3H8 (propane) and 40% C4H10 (butane) was used as the main fuel. Some indicators were determined after the experiment: measurement units conversion from “ppm” to “mg/m3,” excess air ratio α and equivalence ratio φ, flue gas concentrations recalculation taking oxygen into account, fuel calorific value, and heat release rate. The analysis results are as follows: (i) the swirl burner shows better performance in terms of nitrogen oxides (NOx) emissions; there is a 1.75-fold difference in total NOx emissions compared to the cross jet burner; (ii) the burner on transverse jets is 10 times more efficient than the swirl burner in terms of carbon monoxide (CO) emissions.
1. Introduction
Humanity’s fossil fuel burning technology has reached tangible heights. Most modern burners are capable of burning known design fuels with efficiency level above 92%. New issues arise as modern combustion devices become increasingly sophisticated and new operating conditions are formed. This marked the beginning of competition for fractions of a percent of efficiency. In terms of environmental performance, the situation is roughly the same. Emission levels have been reduced to an attainable minimum. All further attempts to reduce emissions have insignificant success. In order to stimulate progress in this area, countries adopt legislations to lower emission thresholds. For example, the CIS Interstate Standard [1] provides that “the nitrogen oxides content shall not exceed 150 mg/m3 (73.17 ppm) for a gas turbine plant without regeneration and 200 mg/m3 (97.56 ppm) for a gas turbine plant with heat regeneration (in exhaust gases at 0°C, 0.1013 MPa, and reference oxygen concentration of 15%)” and “carbon monoxide content in exhaust gases shall not exceed 300 mg/m3 (240 ppm).” The US EPA (Environmental Protection Agency, USA) standards are the most stringent –205 mg/m3 (100 ppm) of nitrogen oxide emissions for equipment with a capacity up to 3.5 MW. As the capacity increases to 110 MW and more, the emission threshold decreases to 30 mg/m3 (15 ppm) [2]. The maximum carbon monoxide emission is set at 63 mg/m3 (50 ppm) [3]. It also depends on the equipment installed capacity. The European Union directive [4] sets medium environmental requirements: NOx 50 mg/m3 (24 ppm) and CO 100 mg/m3 (80 ppm) for natural gas; NOx 120 mg/m3 (60 ppm) and CO 100 mg/m3 (80 ppm) for other gas types. Similar emission standard was adopted in China in 2020 [5]. According to the standard, average concentrations of NOx emissions during combustion should not exceed 150 mg/m3 (73.17 ppm).
A rich variety of burner devices have been developed by science and technology so far, featuring high efficiency and permissible level of harmful emissions into the environment. For example, the medium-capacity power boilers mostly use universal burners without premixing to flare high-calorific low- or high-pressure natural gas [6]. In external mixing burners, the active mixing zone is localized at the base of flame to ensure a stable combustion process. This is characteristic of any of the many burner designs, especially those which aim to create a vortex or jet ejection enhancing the heat and mass transfer [7]. In contrast to the described devices, burners with premixing [8] allow significant reduction of harmful emissions without loss of combustion efficiency of gaseous and, in some cases, liquid fuels, provided they are mainly used in combustion chambers of gas turbines. The most commonly used device of this kind is lean premixed prevaporized burner [9], which has a great potential for carbon monoxide (CO) and nitrogen oxide (NOx) emissions reduction [10–12]. It provides a homogeneous lean mixture of air and fuel that can burn at lower temperatures thus reducing thermal NOx emissions. However, the key feature of lean mixture combustion is its sensitivity to external disturbances resulting in flame stabilization problems [13–15] or thermo-acoustic instabilities [16, 17]. The device is greatly limited to certain types of fuel and fields of application.
An alternative device is the porous burner that has shown excellent performance in terms of low emission levels, high combustion stability, increased power density, and the ability to operate at ultralow combustion rates [18]. One of the disadvantages of this design is limited maximum heat output. It is therefore not studied further in this research. Another promising area of development is turbulent flows creation by separated (transverse) jets. Eventually, harmful emissions reduction can be observed [19, 20]. In [21], the effect of jet inclination on mixing processes compared to transverse flow was studied using the turbulence model. The effect of using axial fuel injection with variable orifices in swirl burners was investigated for flashback resistance [22]. Flashback resistance optimization is achieved through increase in the diameter of axial fuel jet contributing to vortex displacement [23]. Although axial fuel injection along the centerline has been shown to have a wider performance, the technique can result in significantly higher NOx emissions [24] due to localization of high temperature zones. The use of central air injection is a suitable alternative to central fuel injection in terms of NOx emissions reduction, where distributed reaction conditions in the combustion chamber can be achieved through air injection rate control [25].
The flameless combustion regime is also well known, which delivered on the promise of ultralow NOx and CO emissions levels. However, the high temperatures of the stoichiometric flames would produce the greatest NOx emissions [26].
Unlike other methods, MILD (moderate or intense low oxygen dilution) combustion produces a much more uniform temperature distribution with a lower peak value. MILD combustion has larger CO production rate than traditional combustion and generates a higher CO peak value in the reaction zone [27]. MILD combustion regime was achieved for a wide range of external parameters with reduced combustion peak temperatures and very low NOx and CO emissions in a wide operational window. Scaling the combustor at higher heat intensity renders the system very complex and makes its implementation highly challenging [28].
A large amount of central air injection can increase flashback resistance and significantly reduce NOx volumes [29]. Thus, the practice of initial fuel and air injection into the preignition chamber has every chance to prove effective.
The authors of the theoretical study [30] propose a method for reducing the formation of NOx by organizing fuel zoning and diffusion combustion; that is, gas fuel is mixed with part of the heated air, and then the products of incomplete combustion diffusely burn in the rest of air. This principle allowed them to reduce nitrogen oxides to 312.89 mg/m3 (152.6 ppm) according to the computer simulation data. The transfer of this principle from the heat flue in a coke oven to burners for various purposes can lead to a positive result, especially if it is experimentally verified. This idea is taken into account by the authors of the present article in the burner device no. 1 in an adapted version.
Xu et al. [31] describe an equipment which adopts a self-circulation method to reduce harmful NOx emissions. It allows them to significantly reduce nitrogen oxides by limitation of the oxygen concentration and temperature in the combustion zone. The recommendation by [32] ratio of flue gas (circulation) to air is 0.6, which made it possible to reduce NOx to 128.42 mg/m3 (62.64 ppm) and prevent significant CO emissions. Circulation is provided by using the energy within a high-pressure motive fluid (gas fuel or air) to entrain a low-pressure suction fluid (flue gas) to an intermediate discharge pressure (ejector effect). This theoretical study notes that low velocities of parallel fuel and air flows do not provide the necessary mixing of fuel and air. In addition, the high pressure of the motive fluid is the main condition for the existence of self-circulation, which is associated with the need for a compressor or other high-power supercharger. It can complicate and increase the cost of the burner. As an alternative, the vortex is able to achieve efficient mixing using lower linear velocities, since high vortex velocities are created by swirling, as is carried out in the burner device no. 1 of the present research.
The studies described above are summarized as follows:(1)Efficient mixing of fuel and air is necessary to reduce CO emissions, that is, to increase the thermal efficiency of combustion. It is carried out in(a)vortex flow due to the intensity of turbulence [8](b)cross jets due to collision of flows [19](c)parallel jets due to the speed or pressure of fluid [31](2)It is necessary for reducing NOx emissions,(a)reducing the temperature in local areas by mixing with rest of air [9](b)distributing the flame over the volume of the combustion chamber [18](c)circulation of flue gas into the chamber space [31]
Two different designs of burners were created within the scope of this research. They work on the basis of different methods of fuel-air mixture preparation: (1) vortex mixing and (2) mixing with cross jets. Parallel jets are not considered according to [31]. The two methods are compared in terms of their efficiency and environmental performance, and the newly created burner devices have the aim to provide efficient combustion with low harmful emissions of NOx and CO.
The authors of this article have made previous attempts to create a universal burner device in terms of fuel selection (prototype of device no. 1) [33].
Vortex turbulence is known to create favorable conditions for fuel-air mixture formation [34]. Reynolds number is considered to be the main indicator of turbulence. The basic principle lying at the core of this design is the vortex flow inside, which provokes a more intense mixing of fuel and air. Experiments [35] have shown that physical and thermal pretreatment of fuel-air mixture has a positive effect on its environmental performance due to the uniform distribution of the flame in the burner space with a moderate swirling. As an alternative to vortex turbulence technology, employees of the Kazan State Power Engineering University (Kazan, Tatarstan) developed the design of a burner with transverse jets (device no. 2) where active mixing is achieved through collision of air and fuel flows at an angle close to 90°. Effective environmental performance of this device is achieved by breaking up local high temperature zones with cross jets. The design is simpler in execution and operation, all flows are direct, but unlike device no. 1, device no. 2 is only suitable for combustion of gaseous fuel.
This paper provides comparison of the two combustion principles: vortex combustion and transverse jet combustion. The relevant burner devices were tested on a special stand in order to determine harmful NOx and CO emissions. The propane-butane mixture is representative of LPG (liquified petroleum gas) mixtures that widely used a different scale of operation. LPG has a wide variety of uses in many different markets.
2. Materials and Methods
2.1. Object of Study
The study focused on the combustion process in two fundamentally different burner designs. The dimensions of both units were designed to ensure the same fuel consumption.
2.1.1. Device No. 1
The swirl burner device [36, 37] works similar to the cyclone boiler by Babcock and Wilcox [38], the difference being that the fuel and air inlet into prechamber 1 (Figure 1) is not axial but tangential. This type of feed is known to create intensive swirling flow inside the chamber. Fuel and air are fed through inlets 2 and 3 and begin to actively mix and react in chamber 1. Vortex is actively generated in chamber 1 in the form of a mini tornado with deep rarefaction created on the axis. Low pressure contributes to fuel involvement (suction) in circulation and fuel-air medium release into the outlet nozzle 4. Nozzle 4 has a recess inside chamber 1, which helps to reduce the flow aerodynamic resistance at the outlet from chamber 1 due to “soft” friction of the gaseous medium against itself rather than against the chamber wall.
Device no. 1 was originally designed as a universal burner, which is capable of burning various liquid and gaseous fuels without significant modifications to the design (Figure 2). The versatility and unpretentiousness of the burner in relation to the type of fuel is one of the main advantages of the unit.
2.1.2. Device No. 2
The cross jet burner is based on the principle of fuel and air flows collision at an angle close to 90°, which facilitates active mixing. The jet trajectory depends on momentum ratio between the two streams.
The device (Figure 3) consists of two main parts: the stationary part (flange) 2, bolted to the combustion chamber 6, and the movable part 1, sliding on the stationary part in axial direction and moved by device 4.
There are holes on the inner wall of the movable part 1 for air access, 4 mm in diameter, arranged in a staggered pattern. Changing the burner position relative to the flange makes the passage area larger or smaller, and consequently the speed of transverse air jets changes at a constant flow rate, since part of the holes is blocked by the flange and the output speed through the remaining “open” holes either increases or decreases. This, in turn, changes the mixing and combustion process in the furnace.
Device no. 2 was designed as a burner, which is capable of burning gaseous fuels only (Figure 4).
2.2. Experimental
The experimental study of swirl burner and cross jet gas burner operation was carried out with the use of a laboratory firing stand in order to compare the main emission indicators through flue gas sample analysis with the TESTO 350-XL gas analyzer.
A propane-butane mixture of 60% C3H8 (propane) and 40% C4H10 (butane) was used as the main fuel which is the most readily available, predictable, and flammable.
The firing stand in Figure 5 enables analysis of the resulting combustion products under different gas burner operating conditions.
The main element of the firing stand is a gas-tight combustion chamber 1 where gaseous fuel supplied from cylinder 17 is combusted in the burner 2. Valve 20 is used for gas flow regulation. Combustion air is supplied by the air flow rate booster 13 regulated by the autotransformer 3. The flare is ignited through the igniter 5. Hatches 6 are provided for flare observation. Differential manometers 14 and 15 measure air and fuel consumption. Flue gas samples are collected through the hatch 21. Pump 24 is used to pump flue gas samples through the sampling line 22. The shut-off valve 23 controls the flue gas flow direction: to the pump and into the air or to the sampling probe of the gas analyzer. A millivoltmeter 10 is used to determine the flue gas temperature.
The flue gas temperature at the stand outlet was determined with an MR-64-02 millivoltmeter and a thermocouple. Type K thermocouple ( Nickel-Chromium vs Nickel-Aluminium) was used in the experiment. The permissible measurement deviation limit is ΔE; mV is calculated according to the following formula [39]:where t is the measuring temperature, °C.
At the temperature values obtained, the measurement limit was 0.16 mV, which corresponds to 4.03°C for a chromel-alumel thermocouple. The relative error is δt = 3.1%.
Millivoltmeter MR-64-02 is a panel profile instrument with a two-position regulating device of magneto-electric system. Millivoltmeter MR-64-02 is designed for operation at ambient temperature from +10°C to +35°C and relative humidity up to 80%. The basic permissible error of the millivoltmeter at all scale marks does not exceed δmV = 1.5% [40].
The second part of the firing stand is the universal portable TESTO 350-XL measuring system, consisting of a control module, analyzer, and sampling probe. The measuring system is shown in Figure 6.
The TESTO 350-XL is used to collect and analyze flue gas samples obtained in the combustion chamber of the firing unit. The control module 2 (Figure 6) of the gas analyzer displays measurement data and enables control of the device. The sampling probe 5 is used to take a gas sample and deliver it to the measuring device. The operating principle of the TESTO 350-XL is based on electrochemical measuring cells.
During the preliminary phase of the experiment, the TESTO 350-XL was calibrated in strict accordance with the manufacturer’s instructions [41]. Without calibration before the experiment, it can lead to gross inaccuracies in measurements.
The measurement range of TESTO 350-XL is shown in Table 1.
The device draws flue gases in (flow rate of approx. 0.8 l/min) through the gas sampling probe using an integrated pump. After gas is cleaned and dried, it flows to the electrochemical cells. The measured parameters are continuously shown on the display: O2, oxygen, %; CO, carbon monoxide, ppm; NOx, nitrogen oxides, ppm; and TG, flue gas temperature, °C.
Flue gases flow from the combustion chamber into the gas discharge pipe with internal diameter d = 130 mm.
The experiment was carried out stepwise with two burner designs in order to compare the main indicators of flue gas samples.(1)The fuel flow controller enabled three fuel flow settings:(i)0.358 kg/h(ii)0.438 kg/h(iii)0.506 kg/h(2)The air flow rate was set to achieve stable combustion. The oxidant content was gradually increased with air flow regulator, and several positions were measured for further graphic analysis. A 2-minute interval was maintained between measurements, during which a flue gas sample was pumped through the sampling line.
The estimate of the permissible error of the measuring system is calculated as the rms sum of the errors of the components of the measuring system:
For laboratory and semiindustrial emissions measurement experiments, the value of the relative error limit of the measuring system δMS = 6.07% is considered acceptable.
Each experiment was repeated 5 times. The relative statistical error of the measurement has been determined. The relative error of the measuring system does not exceed the permissible error.
2.3. Calculations
(1)Conversion from “ppm” to “mg/m3.” The concentration unit commonly used in the entire range of TESTO gas analyzers is “ppm.” The ppm is the volumetric concentration, i.e., the number of molecules of a measured substance out of a million gas molecules and is independent of the temperature or pressure of the environment. In order to compare the results of this experiment with similar research, “ppm” had to be converted to “mg/m3.” NOx and CO concentrations were converted from weight “mg/m3” to volumetric “ppm” dimension and vice versa using the following formulas: where 2.05 is the density of nitrogen dioxide under normal conditions, (kg/m3); 1.25 is the density of carbon monoxide under normal conditions, (kg/m3). “NOx” refers to a combination of NO, NO2.(2)Excess air ratio is defined as follows: where Ga is the actual airflow measured in the experiment (ma3/s); В f is the fuel consumption measured in the experiment (mf3/s); G o is the theoretical airflow rate (): where С3Н8 and С4Н10 are the volumetric concentrations of fuel components, %v. Depending on the industry sector, the term “equivalence ratio φ” is often used in science instead of “excess air ratio.” Equivalence ratio is the ratio of actual fuel/air ratio to stoichiometric fuel/air ratio. It is defined as follows: Both ratios are presented later in the figures together.(3)Flue gas concentrations recalculation with oxygen. According to the oxygen formula, excess air ratio can be calculated as follows: where О2 is the oxygen content in flue gases, %. The normal (reference) value of in swirl burners is 3%, which corresponds to an approximate value α = 1.17 (φ = 0.85). In gas turbine combustion chambers, for example, the oxygen content in the combustion chambers of gas turbines is = 15%, which corresponds to α = 3.5 (φ = 0.29) in excess air conditions. When harmful emissions were measured, NOx and CO values were defined at the actual oxygen content in combustion products, which depends on the excess air ratio and combustion efficiency and is determined in the experiment. Therefore, for emissions comparison, NOx and CO values are converted to the reference oxygen content in combustion products . where is the reduced NOx concentration at oxygen content , %; is the measured (actual) concentration of the test gas; is the measured oxygen concentration in combustion products. The carbon monoxide (CO) concentration in flue gases has been recalculated in a similar way.(4)Determination of the Reynolds number. Reynolds number is the main indicator of turbulence intensity. Also, turbulence in turn directly influences the mixing efficiency of fuel and air [34]. Reynolds number is calculated by the following formula: where υ is the gas flow velocity, m/s; l is the characteristic size. The diameter applies to the pipe, m; is the kinematic viscosity of the gas flow, m2/s. Let us calculate the Reynolds numbers for the two devices. Device 1 operates at Reynolds numbers Re = 5809.19/7678.26 (Table 2), while device 2 exhibits lower Re = 1728.51/4697.66 (Table 3). Obviously, device 1 has larger turbulence foci than device 2. As a consequence, mixing in unit 1 is more intense.(5)Fuel calorific value. The lower fuel calorific value Qd, MJ/m3 was calculated using the following formula: where C3H8 is the propane content in the fuel, %; C4H10 is the butane content in the fuel, %. The result of the experiment is as follows:(6)Fuel and air flowrate was determined at the diaphragm due to pressure drop ΔP in the narrowed passage. As we know from Bernoulli’s equation, the flow kinetic component is determined with the following formula: where Δh is the pressure drop at differential manometer 14, 15 (Figure 5), m; is the flow velocity, m/s; is the acceleration of gravity, m/s2; ξ is the drag coefficient, which for a diaphragm with thickened edges in a straight pipe (duct) (Figure 7) is calculated with the following formula: where ρ is the density of the medium flowing through the section, kg/m3; F is the cross-section area, m2; D r is the reduced diameter, m: P о is the section perimeter, m; l is the diaphragm length, m; ĺ is the reduced diaphragm length ĺ = l/Dr, m; λ is the adjustment factor λ = 0.02; τ is the adjustment factor (ĺ); φ is the degree index ; Conversion of formula (12) produces equation for flow velocity [m/s] and flow rate G [m3/s] calculation depending on the head Δh:(7)Heat output (kW) of the combustion unit was determined according to the following formula: where is the fuel consumption measured in the experiment, (m3/s); is the calorific value of the fuel, kJ/m3; η is the thermal efficiency of combustion in the furnace space.(8)Heat release rate (kW/m3) of the furnace volume was determined using the following formula:where is the furnace chamber volume, m3.
In order to determine the furnace volume VC, half of the experimental stand volume has to be deemed actively involved in combustion. The stand has a power reserve, therefore part of its volume is not involved. We take the following stand dimensions: 0.38 × 0.38 × 0.7 m.
The steam drum boilers heat release rate is usually deemed equal to 140–150 kW/m3 for boilers designed for coal combustion. For gas-and-oil-fired boilers, the range is 200–230 kW/m3, and for gas-tight boilers, somewhat higher [42]. Thus, the burners under consideration have heat release rates (100–142 kW/m3) close to the standard ones for steam boiler furnaces.
3. Results
The results of experiments with propane-butane mixture combustion in the innovative swirl burner are summarized in Table 2.
The results of experiments with propane-butane mixture combustion in the cross jet gas burner are summarized in Table 3.
Upon recalculation of measured NOx and CO emissions taking into account the O2 concentration in flue gases, the results were further converted into graphs. Figure 8 shows the dependences of total nitrogen oxides emissions on excess air ratio for all the given fuel consumption by both burners. Similarly, Figure 9 contains data on CO emissions depending on excess air ratio for all the given fuel consumption levels for both burners.
Figure 10 shows an overlay of nitrogen oxides and carbon monoxide emissions to determine the optimum excess air ratio interval for device no. 1. Figure 11 contains similar data for device no. 2.
The results of analysis of fuel consumption impact on NOx and CO emissions are graphically shown in Figures 12–15.
4. Discussion
Comparison of environmental performance of the two different burner devices is presented below. Upon initial examination of NOx and CO emissions dependence diagrams, it becomes obvious that stable operation of device no. 1 requires larger excess air amounts than that of device no. 2. The stable combustion range for device no. 1 is α = 1.4/2.0 (φ = 0.5/0.71), whereas device no. 2 shows stability at α = 0.5/1.0 (φ = 1.0/2.0). The significant difference in excess air requirements can be explained by differences in the operating principle of the burner devices. Device no. 1 needs an intensive vortex generated among other things by active air flow. Device no. 2 uses direct flows, and turbulent acceleration is created by narrowing cross-section of air flow openings rather than by air flow increase. As a result, device no. 2 operates in a lower excess air ratio interval, on average lower by 0.9.
In terms of total nitrogen oxides in Figure 8, device no. 1 is superior to device no. 2. The NOx values are on average between 25 and 35 ppm (device no. 1). However, device no. 2 values are between 40 and 65 ppm, on average 1.75 times the NOx values of device no. 1. The difference is achieved through abundant dilution of combustion products with conditionally cold air in device no. 1, which reduces the overall temperature level and the intensity of thermal nitrogen oxides formation. It should be noted that device no. 1 does not exceed worldwide requirements to nitrogen oxide emission levels. In addition, active air flow breaks up local zones of increased fuel concentration and temperature. It leads to uniform distribution of heat in the combustion chamber.
As expected, the CO levels (Figure 9) are inversely related to NOx. The average carbon monoxide emissions by both units are strongly dependent on fuel consumption. Device no. 2 generates nitrogen monoxide emissions in the range of 20/160 ppm. Device no. 1 significantly exceeds these figures, 350/1600 ppm, thus reaching 10 times higher emission levels than device no. 2. The significant CO excess in device no. 1 demonstrates the obvious fuel underburning. Combustion is incomplete due to unacceptably small dimensions of the swirl burner (device no. 1) insufficient for intense swirl generation at the given fuel flow rate. The flow of fuel-air mixture entering the chamber does not have enough time to swirl and bursts into the outlet nozzle taking unburned fuel with it. Device no. 1 has limitations in terms of minimum dimensions and consequently minimum fuel consumption. The obvious solution is that the device must have larger dimensions, similar to prototype [33].
Analysis of dependencies in Figure 10 shows that an optimum range of excess air values for Device no. 1 exists. The average value α decreases as fuel consumption increases, as shown in Figure 16. A regression relationship with polynomial degree 2 was obtained for this curve, allowing calculation of α with average relative deviation error 3.4∙10−4%:where ВF is the fuel consumption, kg/h.
Equation (18) allows us to calculate the excess air ratio for device no. 1 and any fuel consumption with high degree of accuracy.
Figure 11 also confirms that, in device no. 2, excess air ratio decreases as fuel consumption goes up. The dependence of excess air ratio α on fuel consumption for cross jet burner is graphically shown in Figure 17. Regression equation (19) allows us to calculate α with average relative deviation error of 0.057% for any fuel consumption ВF, kg/h:
The regression equations (18) and (19) were obtained by the least squares method.
Figures 12–15 provide an estimate of the fuel consumption effect on emission levels. Figure 12 shows that, in device no. 1, an increase in fuel consumption does not significantly change NOx emissions having a greater impact on the required excess air ratio. In case of device no. 2 (Figure 13), the relationship between NOx and fuel consumption is more pronounced: the amount of NOx goes down as fuel consumption grows, while the fuel consumption impact on air ratio is negligible. More fuel evenly fills the furnace volume preventing formation of local high temperature zones and reducing nitrogen oxides output.
Carbon monoxide emissions in device no. 1 are obviously reduced as fuel consumption increases (Figure 14). Larger amounts of fuel are forced to stay in the burner vortex for longer periods of time and burn out more intensively. At the same time, an increase in fuel consumption brings down the required excess air ratio.
Figure 15 demonstrates the fact that, in device no. 2, the CO emissions are almost independent of the fuel consumption level (curves are merged). The only impact factor in this case is excess air ratio. As excess air ratio increases, the CO emissions decrease.
5. Conclusion
Comparison of two fundamentally different burner designs for combustion of small volumes of combustible gas has showed that(1)in terms of NOx emissions, the swirl burner is superior to the cross jet burner with a 1.75-fold difference in total nitrogen oxides(2)in terms of carbon monoxide (CO) emissions, the cross jet burner is 10 times more efficient than the swirl burner
The cross jet burner was originally manufactured as laboratory equipment and its dimensions as well as combustion air flow rate were selected in strict compliance with the expected lowest emission levels. During the experiment, a considerable advantage of the device over the swirl burner was observed in terms of CO emissions. This is due to the fact that ultrasmall dimensions of the swirl burner fail to ensure sufficient vortex formation inside the chamber and promote direct outflow of unburned mixture, although the basic idea behind the swirl burner design is to organize the vortex inside in order to keep the mixture in the reaction zone. This problem can be successfully solved by increasing the size and consequently the output of the swirl burner. Device no. 2 must be at least twice as large in size and performance. The test results for a larger device are not discussed in this paper.
Comparison of NOx produced by the two burners leads to the conclusion that the flow swirling according to the “tornado” principle in the swirl burner significantly reduces emissions (by 80–90%) due to a longer stay of the fuel-air mixture in the combustion chamber promoting the atomic nitrogen reduction reactions. In addition, the process of combustion products dissolution in excess air has a significant impact. The stable combustion range for device no. 1 is α = 1.4/2.0 (φ = 0.5/0.71), whereas device no. 2 shows stability at α = 0.5/1.0 (φ = 1.0/2.0). The excess air is drawn in from the space around and is not heated intentionally to have an ambient temperature. As a result, the overall temperature level at the flare core is reduced having a positive effect on nitrogen oxides formation. The excess air ratio in swirl burner 1 is on average 0.9 times higher than that in cross jet burner 2. Nitrogen dioxide formation is generally more sensitive to temperature than nitrogen monoxide formation.
As a result of the experiment, it was proved that the swirl burner previously used only for combustion of liquid fuels, both traditional and alternative, can effectively compete with specialized burners operating exclusively on gas. The minimum output limitations of this burner must be taken into account.
The experiment also highlighted the need to develop a larger swirl burner prototype to achieve the most efficient swirl within the device and burn the fuel-air mixture more completely while maintaining high efficiency and flexibility to operate on different types of fuel without major design changes.
Despite the relatively small dimensions of both units, in the vortex burner device, as compared to its peer, fuel and air input can be changed to a larger extent while stable combustion without flame failure is maintained. In case of the cross jet burner, air supply could not exceed 8.46 kg/h, as at higher levels, the flame became unstable and fluctuation of most indicators increased. The steadier operation of device no. 1 with higher fuel-air mixture consumption made it possible to achieve higher output, higher heat release rate of 142 kW/m3 as compared to device no. 2 with the value of 100 kW/m3.
The main conclusion is that swirl burner 1 can be widely used in the power sector with a minimum output limit. The cross jet burner 2 is also widely applicable in the industry despite being limited to gaseous fuel alone. Therefore, the choice of device will depend on external factors: type of fuel, problems that have to be solved, etc.
Data Availability
The data used to support the finding of the study are included in the paper.
Conflicts of Interest
The authors declare that they have no conflicts of interest.