Bench Scale Testing of Low-NOx LBG Combustors

1982 ◽  
Vol 104 (1) ◽  
pp. 120-128 ◽  
Author(s):  
W. D. Clark ◽  
B. A. Folsom ◽  
W. R. Seeker ◽  
C. W. Courtney

The high efficiencies obtained in a combined gas-turbine/steam-turbine power cycle burning low Btu gas (LBG) make it a potentially attractive alternative to the high sulfur emitting direct coal-fired steam cycle. In the gasification process, much of the bound nitrogen in coal is converted to ammonia in the LBG. This ammonia is largely converted to nitrogen oxides (NOx) in conventional combustors. This paper examines the pressurized bench scale performance of reactors previously demonstrated to produce low NOx emissions in atmospheric laboratory scale experiments. LBG was synthesized in a catalytic reformer and fired in three reactors: a catalytic reactor, a diffusion flame, and a stirred reactor. Effects of scale, pressure, stoichiometry, residence time, and preheat were examined. Lowest NOx emissions were produced in a rich/lean series staged catalytic reactor.

Author(s):  
Haoyang Liu ◽  
Wenkai Qian ◽  
Min Zhu ◽  
Suhui Li

Abstract To avoid flashback issues of the high-H2 syngas fuel, current syngas turbines usually use non-premixed combustors, which have high NOx emissions. A promising solution to this dilemma is RQL (rich-burn, quick-mix, lean-burn) combustion, which not only reduces NOx emissions, but also mitigates flashback. This paper presents a kinetics modeling study on NOx emissions of a syngas-fueled gas turbine combustor using RQL architecture. The combustor was simulated with a chemical reactor network model in CHEMKIN-PRO software. The combustion and NOx formation reactions were modeled using a detailed kinetics mechanism that was developed for syngas. Impacts of combustor design/operating parameters on NOx emissions were systematically investigated, including combustor outlet temperature, rich/lean air flow split and residence time split. The mixing effects in both the rich-burn zone and the quick-mix zone were also investigated. Results show that for an RQL combustor, the NOx emissions initially decrease and then increase with combustor outlet temperature. The leading parameters for NOx control are temperature-dependent. At typical modern gas turbine combustor operating temperatures (e.g., < 1890 K), the air flow split is the most effective parameter for NOx control, followed by the mixing at the rich-burn zone. However, as the combustor outlet temperature increases, the impacts of air flow split and mixing in the rich-burn zone on NOx reduction become less pronounced, whereas both the residence time split and the mixing in the quick-mix zone become important.


Author(s):  
Ralph A. Dalla Betta ◽  
James C. Schlatter ◽  
Sarento G. Nickolas ◽  
Martin B. Cutrone ◽  
Kenneth W. Beebe ◽  
...  

The most effective technologies currently available for controlling NOx emissions from heavy-duty industrial gas turbines are either diluent injection in the combustor reaction zone, or lean premixed Dry Low NOx (DLN) combustion. For ultra low emissions requirements, these must be combined with selective catalytic reduction (SCR) DeNOx systems in the gas turbine exhaust. An alternative technology for achieving comparable emissions levels with the potential for lower capital investment and operating cost is catalytic combustion of lean premixed fuel and air within the gas turbine. The design of a catalytic combustion system using natural gas fuel has been prepared for the GE model MS9OOIE gas turbine. This machine has a turbine inlet temperature to the first rotating stage of over 1100°C and produces approximately 105 MW electrical output in simple cycle operation. The 508 mm diameter catalytic combustor designed for this gas turbine was operated at full-scale conditions in tests conducted in 1992 and 1994. The combustor was operated for twelve hours during the 1994 test and demonstrated very low NOx emissions from the catalytic reactor. The total exhaust NOx level was approximately 12–15 ppmv and was produced almost entirely in the preburner ahead of the reactor. A small quantity of steam injected into the preburner reduced the NOx emissions to 5–6 ppmv. Development of the combustion system has continued with the objectives of reducing CO and UHC emissions, understanding the parameters affecting reactor stability and spatial non-uniformities which were observed at low inlet temperature, and improving the structural integrity of the reactor system to a level required for commercial operation of gas turbines. Design modifications were completed and combustion hardware was fabricated for additional full-scale tests of the catalytic combustion system in March 1995 and January 1996. This paper presents a discussion of the combustor design, the catalytic reactor design and the results of full-scale testing of the improved combustor at MS9OOIE cycle conditions in the March 1995 and January 1996 tests. Major improvements in performance were achieved with CO and UHC emissions of 10 ppmv and 0 ppmv at base load conditions. This ongoing program will lead to two additional full-scale combustion system tests in 1996. The results of these tests will be available for discussion at the June 1996 Conference in Birmingham.


Author(s):  
Haoyang Liu ◽  
Wenkai Qian ◽  
Min Zhu ◽  
Suhui Li

Abstract To avoid flashback issues of the high-H2 syngas fuel, current syngas turbines usually use nonpremixed combustors, which have high NOx emissions. A promising solution to this dilemma is rich-burn, quick-mix, lean-burn (RQL) combustion, which not only reduces NOx emissions but also mitigates flashback. This paper presents a kinetics modeling study on NOx emissions of a syngas–fueled gas turbine combustor using RQL architecture. The combustor was simulated with a chemical reactor network (CRN) model in chemkin-pro software. The combustion and NOx formation reactions were modeled using a detailed kinetics mechanism that was developed for syngas. Impacts of combustor design/operating parameters on NOx emissions were systematically investigated, including combustor outlet temperature, rich/lean air flow split, and residence time split. The mixing effects in both the rich-burn zone and the quick-mix zone were also investigated. Results show that for an RQL combustor, the NOx emissions initially decrease and then increase with combustor outlet temperature. The leading parameters for NOx control are temperature-dependent. At typical modern gas turbine combustor operating temperatures (e.g., &lt;1890 K), the air flow split is the most effective parameter for NOx control, followed by the mixing at the rich-burn zone. However, as the combustor outlet temperature increases, the impacts of air flow split and mixing in the rich-burn zone on NOx reduction become less pronounced, whereas both the residence time split and the mixing in the quick-mix zone become important.


1980 ◽  
Vol 102 (2) ◽  
pp. 459-467 ◽  
Author(s):  
B. A. Folsom ◽  
C. W. Courtney ◽  
M. P. Heap

The low Btu gas (LBG) combined gas and steam turbine power cycle is a potentially attractive alternative to the direct coal fired steam cycle because of the potential for low sulfur emissions and high overall cycle efficiency. However, LBG may contain ammonia (NH3) which could be converted to nitrogen oxides (NOx) under typical combustion conditions. This paper examines the effects of LBG composition and combustor design on NOx emissions. Low Btu gases of varying compositions were synthesized from bottled gases and fired in three atmospheric pressure flame reactors: diffusion flame reactor, flat flame reactor and catalytic reactor. Nitrogen oxide emissions were found to be most sensitive to the concentrations of NH3 and hydrocarbon fuel gas in the synthetic LBG. Lowest NOx emissions were produced by the diffusion flame reactor operating at near stoichiometric conditions and the catalytic reactor operating fuel rich.


1997 ◽  
Vol 119 (4) ◽  
pp. 844-851 ◽  
Author(s):  
R. A. Dalla Betta ◽  
J. C. Schlatter ◽  
S. G. Nickolas ◽  
M. B. Cutrone ◽  
K. W. Beebe ◽  
...  

The most effective technologies currently available for controlling NOx emissions from heavy-duty industrial gas turbines are diluent injection in the combustor reaction zone, and lean premixed Dry Low NOx (DLN) combustion. For ultralow emissions requirements, these must be combined with selective catalytic reduction (SCR) DeNOx systems in the gas turbine exhaust. An alternative technology for achieving comparable emissions levels with the potential for lower capital investment and operating cost is catalytic combustion of lean premixed fuel and air within the gas turbine. The design of a catalytic combustion system using natural gas fuel has been prepared for the GE model MS9OO1E gas turbine. This machine has a turbine inlet temperature to the first rotating stage of over 1100°C and produces approximately 105 MW electrical output in simple cycle operation. The 508-mm-dia catalytic combustor designed for this gas turbine was operated at full-scale conditions in tests conducted in 1992 and 1994. The combustor was operated for twelve hours during the 1994 test and demonstrated very low NOx emissions from the catalytic reactor. The total exhaust NOx level was approximately 12–15 ppmv and was produced almost entirely in the preburner ahead of the reactor. A small quantity of steam injected into the preburner reduced the NOx emissions to 5–6 ppmv. Development of the combustion system has continued with the objectives of reducing CO and UHC emissions, understanding the parameters affecting reactor stability and spatial nonuniformities that were observed at low inlet temperature, and improving the structural integrity of the reactor system to a level required for commercial operation of gas turbines. Design modifications were completed and combustion hardware was fabricated for additional full-scale tests of the catalytic combustion system in March 1995 and January 1996. This paper presents a discussion of the combustor design, the catalytic reactor design, and the results of full-scale testing of the improved combustor at MS9OO1E cycle conditions in the March 1995 and January 1996 tests. Major improvements in performance were achieved with CO and UHC emissions of 10 ppmv and 0 ppmv at baseload conditions. This ongoing program will lead to two additional full-scale combustion system tests in 1996. The results of these tests will be available for discussion at the June 1996 Conference in Birmingham.


Author(s):  
D. Kroniger ◽  
M. Lipperheide ◽  
M. Wirsum

Addition of hydrogen (H2) to gas turbine fuel has recently become a topic of interest facing the global challenges of CO2 free combustion. As a drawback, Nitrogen oxide (NOx) emissions are likely to increase in hydrogen-rich fuel combustion which in return limits the use of the technology. In the course of this development, a model-based quantification of NOx emission increase by fuel flexibility may identify possible operation ranges of this technology. This paper evaluates the effect of an increased hydrogen fraction in the fuel on the NOx emissions of a non-premixed 10 MWth gas turbine combustor. A simple reactor network model has been set up using a perfectly stirred reactor (PSR) to simulate the flame zone and a plug flow reactor (PFR) to simulate the post flame zone. The change of residence time in the flame zone is accounted for by an empirical expression. The model is validated against data from high-pressure test rig experiments of an industrial non-premixed gas turbine combustor. The model results are in good agreement with the experimental data. Based on the model results, a fundamental correlation of the effect of hydrogen on the NOx emissions is formulated.


Author(s):  
Nathan T. Weiland ◽  
Todd G. Sidwell ◽  
Peter A. Strakey

The U.S. Department of Energy’s Turbines Program is developing advanced technology for high-hydrogen gas turbines to enable integration of carbon sequestration technology into coal-gasifying power plants. Program goals include aggressive reductions in gas turbine NOx emissions: less than 2 ppmv NOx at 15% oxygen and 1750 K firing temperature. The approach explored in this work involves nitrogen dilution of hydrogen diffusion flames, which avoids problems with premixing hydrogen at gas turbine pressures and temperatures. Thermal NOx emissions are partially reduced through peak flame temperature control provided by nitrogen dilution, while further reductions are attained by minimizing flame size and residence time. The injector design includes high-velocity coaxial air injection from lobes surrounding the central fuel tube in each of the 48 array units. This configuration strikes a balance between stability and ignition performance, combustor pressure drop, and flame residence time. Array injector test conditions in the optically accessible Low Emissions Combustor Test & Research (LECTR) facility include air preheat temperatures of 500 K, combustor pressures of 4, 8 and 16 atm, equivalence ratios of 0.3 to 0.7, and three hydrogen/nitrogen fuel blend ratios. Test results show that NOx emissions increase with pressure and decrease with increasing fuel and air jet velocities, as expected. The magnitude of these emissions changes deviate from expected NOx scaling relationships, however, due to active combustor cooling and array spacing effects. At 16 atm and 1750 K firing temperature, the lowest NOx emissions obtained is 4.4 ppmv at 15% O2 equivalent (3.0 ppmv if diluent nitrogen is not considered), with a corresponding pressure drop of 7.7%. While these results demonstrate that nitrogen dilution in combination with high strain rates provides a reliable solution to low NOx hydrogen combustion at gas turbine conditions, the injector’s performance can still be improved significantly through suggested design changes.


1980 ◽  
Author(s):  
J. R. Grant ◽  
T. E. Holladay ◽  
F. H. Boenig ◽  
R. L. Duncan

Industrial turbines fired on medium heating value (MHV) gas (nominally 300 Btu/scf) synthesized from coal offer an attractive alternative means of producing electrical power in the future. Peak flame temperatures resulting from combustion of this MHV gas in conventional diffusion flame combustors may be comparable to those of natural gas, yielding undesirably high concentrations of NOx. This paper describes an EPRI-sponsored program conducted to demonstrate a MHV gas turbine combustor capable of meeting EPA NOx requirements without water injection. Program objectives were to design, fabricate, and test three MHV combustor configurations and to demonstrate NOx emissions concentrations of 15 ppmv (dry basis) or less at a burner inlet pressure of 1.27 atm: Design of the combustors was based on a lean-premix fuel metering concept. Tests were conducted in a single-can combustor rig at simulated engine conditions ranging from 40 to 125 percent of engine baseload (74 MW).


Author(s):  
Thomas James Gill ◽  
Lukai Zheng ◽  
Emamode A. Ubogu ◽  
Ihab Ahmed ◽  
Bhupendra Khandelwal

It is a well-known fact that NO2 has far more harmful effects as compared to NO. NO2 creates ozone, which causes eye irritation and exacerbates respiratory conditions. This leads to an increased emergency departments’ visits and hospital admissions for respiratory issues, especially asthma. Under current situation, majority of regulations deal with total NOx emissions, without looking at the break-up of NO2 and NO. However, there is a feeling in emissions regulation community to implement regulations on NO2 emissions. There are standards to measure total NOx emissions. However, these standards are not equipped enough to measure NO2 emissions accurately. The effect of sample line length on NO2 emissions is not fully understood to date. Also, the standards only suggests maximum of 10 seconds residence time regardless of what the line length is. In this study, a systematic experimental test campaign has been conducted to understand the effect of sample line length on NO2, NO distribution. The residence time was maintained below 10 seconds in accordance with the SAE ARP1256D standards. A Rolls-Royce gas turbine combustor and different calibration cylinders have been used to study the effect of sample line length. A numerical study has also been done to predict the conversion of NO2 to NO. It has been found that with increasing sample line length, more NO2 gets converted to NO and overall NO2 emissions show a reduction, whereas this would not be the case at engine exhaust. This effect of sample line length can be used as a loophole in giving lower NO2 emissions readings.


Sign in / Sign up

Export Citation Format

Share Document