Experimental Study of Fuel Lean Reburning for NOx Reduction in Oxygen Enhanced Combustion

Transcription

1 Experimental Study of Fuel Lean Reburning for NOx Reduction in Oxygen Enhanced Combustion CHANG HWAN HWANG*, SEUNG WOOK BAEK + and HAK YOUNG KIM** Aerospace Engineering Korea Advanced Institute of Science and Technology 33 Gwahak-ro, Yuseong-gu, Daejeon 3-71 KOREA *propman@kaist.ac.kr, + swbaek@kaist.ac.kr, **dions99@kaist.ac.kr Abstract: - This experimental study is focused on a fuel-lean reburn system to control NOx emission from combustion process. In the conventional reburning process, ~2% reburn fuel of total heat input is used. As a consequence, about 4% NOx emission reduction is achieved. And it requires additional burnout air to restrict CO emission level in acceptable limit. However, in the fuel-lean reburn system, by contrast, the amount of injected reburn fuel into the reburning zone is low enough to maintain overall fuel-lean condition in the furnace, so that no additional burnout air system is required, and CO emission can be maintained at almost zero level. kw lab scale combustor is used to examine the formation characteristics of NOx with various oxygen enhanced combustion conditions. LPG (Liquefied Petroleum Gas) was used as main fuel and reburn fuel. Finally, the current fuel-lean reburn system, even with only an amount of reburn fuel of 13% of total heat input, was observed to achieve a maximum of 48% in NOX reduction. Key-Words: - fuel lean reburning system; NOx; CO; oxygen enhanced combustion; 1 Introduction Nowadays, environmental pollutants emission has been recognized as most prominent threat to the human health. Especially, nitrogen oxides formed in industrial facilities is one of hazardous pollutants. Nitrogen oxides are comprised of nitric oxide (NO), nitrogen dioxide (NO2) and nitrous oxide (N2O). Among these a mixture of NO and NO2 is collectively denoted by NOx [1]. A high concentration level of NOx in the atmosphere creates such problems as acid rain or respiratory ailments. Also it can be significantly harmful to human eyes and ozone layer. On the other hand, carbon monoxide (CO) is also known as another dangerous pollutant for people and living environment. Therefore, a discretionary control of these pollutants is one of major issues in combustion research. Over the last decade, diverse methods for the purpose of restricting air pollutants have been proposed using many experiments or numerical technique [2]. Among many other effective ways to control the NOx and CO emission, a conventional reburning technique has been successfully implemented in many industrial combustion systems [3]. The basic conceptual procedure of the conventional reburning system was well described by Wendt et al [4]. Zeldovich [] explained a mechanism of NOx formation in the primary combustion zone. Thermal NO formation dominantly depends on its local gas temperature and reactant composition. At sufficiently high temperature, oxygen reacts with atmospheric nitrogen to form thermal NO. While prompt NO is produced through CHi radical reaction with molecular nitrogen in flame front, fuel NO is generated by oxidation of nitrogen contained in the fuel. In order to reduce NOx emission, reburn fuel is injected at a downstream of primary combustion zone to create a fuel rich zone so called the reburning zone [2]. The amount of reburn fuel is usually used about 2~3% of the total heat input to the furnace. Previous researchers [6] showed that NOx may be reduced up to % by reburning process. Thereafter, usually additional air is introduced into the downstream of the reburning zone in order to complete combustion of any unburned hydrocarbons and carbon monoxide (CO) remaining in the product gases. The use of additional air ensures that unburned hydrocarbon and carbon monoxide are exhausted by the oxidation reaction. However, the adoption of additional air requires extra injection equipment installed which results in increasing operating and maintenance costs. Moreover, the conventional reburning technique involves a longer furnace length enough to comprise at least three combustion zones. In this respect, an improved reburn system is in high demand to discard the secondary air supply The fuel-lean reburn system is simple and effective technique compare to the conventional reburning process. The fuel-lean reburn system has recently been proposed by Breen and Hura and Miller et al. [7,8]. Main objective is to achieve almost the same NOx reduction as in conventional reburning process, but with less reburn ISSN: ISBN:

2 fuel input than in conventional reburning process, while maintaining CO emission level within acceptable limits. Fuel-lean reburn system was developed by Energy System Associates (ESA). ESA reported that the fuel-lean reburn technology was expected to achieve 3~4% NOx reduction with only 7% reburn fuel of the total heat input without any significant modification in the primary combustion process [9,1]. A main difference between conventional reburning and fuel-lean reburn system is that the fuel-lean reburn system always maintains overall fuel lean condition (equivalence ratio less than 1) in a whole furnace region, including reburning zone. Therefore, burnout zone is no longer necessary. NOx reduction reactions in the fuel-lean reburn system occur in local fuel rich eddies generated by the injected reburn fuel. Mixing rate between the injected reburn fuel and product gas is an important parameter in determining the effective NOx removal. A complete burnout of unburned hydrocarbon and carbon monoxide is achieved by reaction with remaining oxygen in the overall fuel-lean furnace so that additional air supply is no longer necessary. Therefore, this fuel-lean reburn system is regarded as a great potential in attaining appropriate NOx reduction with a lower operating cost than the conventional reburning system. In this study, kw lab scale burner has been used to investigate the NOx reduction characteristics of fuel-reburn system with several combustion methods. The first is conventional combustion, second is oxygen enhanced combustion. LPG (Liquefied Petroleum Gas) is used as reburn as well as main fuel. LPG consists of 9% of propane and % of other mixture such as butane, methane and etc. Experiments are composed of two experimental cases. While the first one examines the effects of fuel-lean reburn system on NOx reduction, in the second one the fuel-lean reburn system is examined in oxygen enhanced combustion. NOMENCLATURE S θ D D h ƒ re ω λ T λ 1 Swirl number Van angle Nozzle diameter Van hub diameter Reburn fuel fraction Oxygen Enrichment Ratio Total air ratio Primary air ratio 2 Experimental setup Figure 1 shows a schematic drawing of the experimental equipments. It consists of three parts. First part is a providing section for fuel and oxidizer. In this study, LPG is used as a main and reburn fuel. While normally, only dry air is used as oxidizer, oxygen is added in oxygen enhanced combustion. Fuel and oxidizer are provided and controlled by separated mass flow controller. Second part is a laboratory size experimental furnace with a burner. The furnace is vertically oriented, while the burner is installed at the bottom of furnace so that the flame is established in upward direction and product gases exit from the top of furnace. Furnace is made of stainless steel, which can tolerate a thermal load above kw/hr. The furnace is 1.2m long and.m in diameter. It had been designed to provide a space enough to re-circulate products gases. To prevent heat loss through furnace wall, it is insulated using ceramic wool (Cerakwool) with thickness of 4mm inside the furnace. The thickness of insulation is about.4m around the furnace wall. Therefore, the effective diameter of the furnace is.42m. On the outer wall of the furnace, eleven ports are made with an interval of.1m along the axial direction for measuring gas phase temperature distribution inside the furnace with R-type thermocouple up to 16 C. In order to inject reburn fuel, several ports are installed along the axial direction. Around the furnace there are also 6 ports for gas injection at each axial location. In order to examine the effect of reburn fuel injection location on NOx reduction rate and heat transfer characteristics, its reburn fuel injection height is varied along 7 locations in the axial direction. The reburn fuel nozzle is of flat type with spray angle 9 of which flow rate is controlled using a mass flow controller. To maximize mixing rate between injected reburn fuel and product gas, nozzles with a small size diameter (.66mm) are used. In order to establish LPG diffusion flame, a coaxial burner is fabricated and installed. While main fuel is supplied from the inner pipe with a diameter of 4mm, oxidizer is provided through its concentric pipe. Meanwhile, before oxidizer is supplied into the burner, it goes through a stabilizing chamber for consistent distribution. For flame stabilization, burner is equipped with a vane swirler to generate swirl induced recirculation zone. Swirler vane angle is here 4 degree and the swirl number is defined as follows [11]. 2 1 ( D S = 3 1 ( D h h 3 / D) tanθ 2 / D) (1) λ 2 Secondary air ratio ISSN: ISBN:

3 Fig.1 Schematic drawing of the experimental equipments Where D is an inner diameter of the swirler, while D h is the herb diameter of the swirler and θ is the swirler vane angle. In this study, swirl number is set to.76. The burner is also outfitted with quarl section to improve stabilizing flame. In the meanwhile, under the condition of oxygen enhanced combustion, the diameter of the oxidizer pipe is varied with oxygen enrichment ratio between O2 and total oxidizer as follows [12]. ω = volume flow rate of O2 in the oxidizer total volume flow rate of oxidizer The last part of the experimental setup is the measurement section. R-type thermocouple is made of a bare-wire Pt/13% Rh-Pt, which is sheathed in a straight length of twin-bore ceramic cladding of.3mm (o.d). Thermocouple has an unavoidable error due to the radiative heat loss from thermocouple bead. Therefore, measured temperature is here calibrated using thermocouples with three different sizes of bead (.3~.mm) as in the following. The thermocouple signals are recorded using a personal DAQ 6 (Iotech Co.) Gas temperature is measured at the same position with thermocouples having different bead sizes. Then the results are extrapolated to the value at zero bead size [13]. In order to analyze composition of product gases, the gas analyzer (Greenline MK2) is used. The product gas in furnace is sampled using a water-cooled stainless steel probe, then dehydrated through water tap in the gas analyzer and analyzed according to CO, NO, NO2 and O2 gases. Toxic gases (CO and NOx) pass through the diffusion barrier to the cell cathode, where it is electrochemically oxidized, thus generating a voltage proportional to the gas concentration. And then, a reverse of counter voltage proportional to the voltage at the air cathode is produced by the measurement circuit, and thereby, the concentration is derived and displayed. The product gases are locally sampled by using a water-cooled stainless steel probe. The gathering port is located at 1.1m away from the burner tip. The accuracy of the gas analyzer is about 9%. (2) 3 Experimental Conditions Two experimental cases are conducted to consider the effect of fuel-lean reburn system on NOx reduction and CO emission. All experimental cases are performed in thermally steady state, which is monitored by observing a variation in furnace wall temperature. The amount of main fuel in experiments is not changed to fix the thermal input condition. However, the amount of oxidizer is properly controlled to keep fuel-lean condition in overall furnace region even in reburning zone. Case 1 is tested to figure out fundamental chemical characteristics of fuel-lean reburn process. Reburn fuel injection location is fixed at.m from furnace bottom. For the case 2, the effective of fuel-lean reburn system on NOx reduction is examined under condition of oxygen enhanced combustion. The reburn fuel fraction is defined by Lee and Baek [14]. the amount of the reburn fuel f re= the amount of the main fuel + the amount of the reburn fuel In the experiment, it has the value from to 13. The effectiveness of fuel-lean reburn system for reducing NOx emission is compared with conventional reburning system. In case 1-1, the effect of reburn fuel fraction on NOx reduction phenomena is investigated. Experimental case 1-2 is considered for evaluating the effect of the amount of primary oxidizer on NOx reduction and CO emission. Variation of the amount of primary oxidizer is expressed using the outlet O2 concentration. case Primary zone(φ) ~.89 Reburn zone(φ).86~.98.84~.98 f re(%) Injection point(m) (ω) ~ ~.7.3~.7 Table 1. Experimental condition.21~.42.3 Case 2 includes two sub-cases to take account of the effect of fuel-lean reburn system when the combustion is oxygen-enhanced. In order to make oxygen enhanced combustion, the primary air supplied to the primary combustion zone is replaced by pure oxygen. An increase in the oxygen enrichment ratio previously defined represents a decrease in the amount of nitrogen in the oxidizer. In order to clarify the relationship between the reburn fuel injection location and CO emission for various oxygen enrichment ratio (ω), case 2-1 is selected for fuel-lean (3) ISSN: ISBN:

4 reburning system. And case 2-2 investigates the effects of reburn zone length on NOx reduction and CO emission. Meanwhile, the overall experimental conditions are listed in Table 1. 4 Experimental results and discussion 4.1 Case1 Most experiments are conducted in slightly fuel lean condition. Theoretically, fuel lean reburn system always has to maintain fuel lean-condition in the whole furnace region even including reburing zone. Therefore, the amount of oxidizer injected in the primary zone is carefully controlled to meet fuel-lean condition. Experimentally, a flame length is observed to be located about.3~.4m away from the burner tip, while the reburn fuel is injected at.m away from the burner tip. Based on this, the NOx reduction is represented for different reburn fuel fractions in Figure 2. In this experimental case 1-1, outlet O2 concentration for the case without reburning is maintained at 3.4% so that the maximum reburn fuel fraction is limited to.13 due to the precondition of fuel-lean reburn system. When the reburn fuel is injected, the hydrocarbon radicals known as key radicals are generated. These hydrocarbon radicals react with NO to form N2. As shown in Figure 4, the NOx reduction steadily increases until the reburn fuel fraction reaches about.13. A rule of general application for NOx reduction mechanism is as follows; [2,6]. C, CH, CH 2 +NO HCN+ (4) HCN+O NCO + H () NCO + H NH + CO (6) NH + H N +H 2 (7) N + NO N 2 +O (8) As shown in Eq. (4), the partial oxidation and pyrolysis of the reburn hydrocarbon fuel result in a formation of CHi radicals. These radicals react with NO, thereby generating HCN radicals, which are known as the intermediate species to initiate NOx removal reaction in the reburning zone. HCN radicals undergo several reactions and then eventually nitrogen oxides are changed into nitrogen molecules. The formation of HCN radicals extremely depends on the concentration of hydrocarbon radical. Therefore, the fuel rich condition in the reburning zone can lead to more reduction of nitrogen oxides. However, as mentioned above, the fuel-lean reburn system does not form fuel rich region even if reburn fuel is injected. The most important aspect for achieving high level of NOx reduction in the fuel-lean reburn system is how to mix the injected reburn fuel and product gas. Depending on the mixing process, fuel rich eddies are locally generated and then NOx reduction reactions occur in fuel rich eddies. However, additional air is not required because most of furnace region is still in fuel-lean condition, which makes it possible to accomplish the complete combustion of the unburned hydrocarbon and carbon monoxide. In order to achieve the high performance in NOx reduction, the mixing rate between reburn fuel and product gas, injection nozzle type and stoichiometry of the primary combustion zone have to be carefully controlled. The amount of oxidizer at the primary combustion zone CO Emission Level (ppm) Reburn Fuel Fraction (%) Fig.2 Effect of reburn fuel fraction on NOx reduction Injection Point :.3m, CO Emission Injection Point :.m, CO Emission Injection Point :.7m, CO Emission O 2 Concentration in product gas (%) Fig.4 CO emission and exit O 2 concentration against Effect of O 2 concentration in product gas plays an important role in fuel-lean reburning system. In this study, the equivalence ratio in the primary zone is maintained at.91 when conventional reburning system is used. So it is very useful to consider the impact of amount of oxidizer on NOx reduction efficiency and CO oxidation in fuel-lean reburning system. Experimental case 1-2 is performed to seek the influence of the varying amount of oxidizer in the primary combustion zone. Reburn fuel fraction is fixed at.1. The amount of ISSN: ISBN:

5 O 2 Concentration in product gas (%) Fig.3 NOx reduction and NOx emission level against O 2 concentration in product gas oxidizer is expressed by measuring O2 concentration in the product gas without reburning process. Figure 3 and 4 show that the effect of increase in O2 concentration on NOx reduction and CO emission. The NOx reduction reaches 3%, when the O2 concentration in product gas is 2.7 %. However, when the O2 concentration in product gas increases up to. %, NOx reduction decreases down to 12 %. The characteristic of CO emission is totally different from NOx emission. An increase in the amount of oxidizer can reduce CO emission, since O2 and OH radicals are so abundant that CHi radicals formed by injecting reburn fuel have more chance to react with O2 and OH than NO. Therefore, NOx reduction efficiency is decreased, while complete combustion of CO is more easily achieved due to OH radicals. 4.1 Case1 The experimental case 2-1 is conducted to examine CO emission level against the oxygen enrichment ratio (ω) in the primary zone. Reburn fuel fraction is fixed at.13, while reburn fuel injection point is varied from.3m to.7m away from the burner tip. Exit O2 concentration in the product gas is measured by changing the OER values with reburning. As shown in Figure, regardless of OER, CO emission is almost negligible at.3 and. m of injection point. This is because the measured gas temperature at reburn fuel injection point ranges in optimal temperature for CO oxidation. As in Figure, the exit O2 concentration is another useful factor to achieve zero CO emission. The exit O2 concentration is observed to steadily decrease as the oxygen enrichment ratio (ω) increases, even if the same amount of reburn fuel is injected. However, at.7 m, CO emission reaches ppm for.21 of OER, and then it decreases to 2 ppm for.3 of OER. Afterwards, for.42 of OER, the exit O2 concentration reaches as low as %, the CO emission increases to 6 ppm. Therefore, in order to maintain low CO emission, the optimal reburn fuel injection point and as proper O2 concentration in the product gas are needed to be taken into consideration. In the meanwhile, among others Lee and Baek [14] reported that the reaction time is another key factor to control NOx reduction and CO emission. Case 2-2 examines the effect of reburn fuel injection point on NOx reduction and CO emission. In this study, the reaction time is adjusted by reburn zone length, which is measured as a distance between reburn fuel injection point and product gas retrieval point which is fixed at 1.2 m away from the burner tip. CO Emission Level (ppm) NOx Redcution (%) 1 Injection Point :.3m Injection Point :.m Injection Point :.7m Exit O 2 Concentration (%) Oxygen Enrichment Ratio (ω) Fig. Effect of oxygen enrichment ratio on CO emission and exit O 2 concentration NO Reduction (%) CO Emission (ppm) Reburn Fuel Injection Point (m) CO Emission Level (ppm) Exit O 2 Concentration (%) Fig.6 Effect of reburn fuel injections point on NO X reduction and CO emission While the reburn fuel faction is.13, the amount of oxidizer is controlled for the exit O2 concentration to be as low as 1. %. Based on these, CO emission characteristic is examined by adjusting the reburn zone length. Overall gas temperature at the reburn fuel injection point is placed in optimal range so that the effect of temperature on CO emission is neglected. As shown in the Figure 6, while NOx reduction is not significantly ISSN: ISBN:

6 influenced by the reburn zone length and temperature distribution, CO emission rapidly increases to ppm at reburn fuel injection point of.7 m. In other words, when the reburn zone length becomes longer than.m, CO emission is observed to be less than ppm in this study. Conclusions In this paper, an experimental investigation was performed to understand the the fuel-lean reburn system characteristics in oxygen enhanced LPG flame. Practically meaningful results could be obtained from several experimental conditions. In the fuel-lean reburn system, whole furnace region was maintained in fuel-lean condition even in reburning zone. For the case of conventional reburning, NOx reduction reached 44%, when reburn fuel fraction was.22. However, the fuel-lean reburn system could attain up to 4% of NOx reduction with just.13 of reburn fuel fraction, since the fuel-lean reburn system had more reaction length than the conventional reburning process. Consequently, energy consumption using fuel-lean reburn system was about 4% compared with conventional reburning. Based on the current experimental results, CO oxidation reaction was influenced by the location of reburn fuel injection point. And if there exists sufficient O2 in exhaust stream, CO can be burnt out. Therefore, CO emission could be restricted to low level when the reburn fuel injection point was located at.3~.m away from the burner tip in this study, even if the reburn fuel fraction reached.13. NOx reduction as well as CO emission was influenced by the amount of oxidizer in the primary combustion zone. As the amount of oxidizer increased, NOx reduction was found to decrease, since CHi radical formed by injecting reburn fuel reacted with oxygen rather than NOx. The oxygen enhanced combustion could achieve a high thermal efficiency by eliminating nitrogen gas from oxidizer. When the oxygen enrichment ratio reached.42, NOx formation became 33ppm. In general, NOx reduction increased by increasing the reburn fuel fraction. However, its increasing rate was different for each OER condition. When outlet O2 concentration was maintained at as low as 1. %, CO emission could be limited to low level. Also, extending reaction length positively led to a reduction in CO emission. Acknowledgment This work is the outcome of a Manpower Development Program for Energy & Resources supported by the Ministry of Knowledge and Economy (MKE) References: [1] Fenimore, C. P. "Formation of nitric oxide in premixed hydrocarbon flames," Thirteenth Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, pp , [2] Miller, J.A.; Bowman, C.T. "Mechanism and modeling of nitrogen chemistry in combustion," Prog. Energy Combust. Sci, vol., pp , 1989 [3] Takahashi, Y.; Sakai, M.; Kunimoto, T.; Ohme, S.; Haneda, H.; Kawamura, T.; Kaneko, S. Proc, the 1982 Joint Symposium on Stationary NOX Control. EPRI Report No. CS-3182, [4] Wendt, J. O. L.; Sternling, C. V.; Matovich, M., "A. Reduction of sulfur trioxide and nitrogen oxides by secondary fuel injection," Proc. Combust. Instit. vol.14, pp , [] Zeldovich, Y. B. "The oxidation of nitrogen in combustion and explosions," Acta Physicochim. URSS, vol.21, No.4, pp , [6] Smoot, L.D.; Hill, S.C.; Xu, H. "NOx control through reburning," Prog. Energy Combust. Sci. (4), 24, [7] Breen, B.P.; Hura, H.S. "Method and apparatus for NOx reduction in flue gases," United States patent, 1. [8] Miller, C.A.; Touati, A.D.; Becker, J.; Wendt, J.O.L. "NOx abatement by fuel-lean reburning: Laboratory combustor and pilot-scale package boiler results," Symposium (International) on Combustion, vol.2, pp , [9] Frederiksen, R. "Fuel Lean Gas Reburn (FLGRTM) Technology for Achieving NOx Emissions Compliance: Application to a Tangentially-Fired Boiler," 1998 Joint- American/Japanese Flame Research Committee International Symposium, [1] Hura, H.S.; Breen, B.P. "Nitrogen oxide reduction by gaseous fuel injection in low temperature, overall fuel-lean flue gas," Unite States patent, [11] Gupta, A.K.; Lilley, D.G.; Syred, N. Swirl Flows. Abacus Press, London [12] Baukal, C.E.Jr. Oxygen-Enhanced Combustion. CRC press, New York, [13] Baek, S.W.; Kim, H.S.; Kang, S.H. "Effects of addition of solid particle on thermal characteristics in hydrogen-air flame," Combust. Sci. Tech, vol.74, pp , 2. [14] Lee C.Y.; Baek S.W. "Effect of hybrid reburnig/sncr strategy on NOx/CO reduction and thermal characteristic in Oxygen-Enriched LPG flame," Combust. Sci. and Tech., vol.179, pp , 7. ISSN: ISBN: