Search

This study reports on the performance of the Ansaldo AE-T100 commercial micro gas turbine (mGT) when fueled by ammonia-methane blends instead of its design fuel, natural gas. This micro gas turbine was used as the experimental platform to understand effects of ammonia addition, and its hardware was only marginally updated. Ammonia was added to the main fuel line in varying proportions, but the pilot flame fuel line was only fed with methane. Experiments were performed for constant electrical output power and turbine outlet temperature of 60 kWe and 645°C, respectively. Important operational parameters and exhaust emissions were continually monitored during the turbine's operation. Results show that stable operation on the mGT was possible until the volume fraction of ammonia in the fuel blend reached XNH3 = 0.63. To reach this ammonia fraction, it was necessary to increase the power of the pilot flame that provides a continuous ignition source at the base of the main flame. Concentrations of carbon monoxide and unburned hydrocarbon measured in the exhaust did not increase significantly until XNH3 = 0.22. However, above this value, concentrations increased rapidly, which is indicative of a drop in the combustion efficiency and, in turn, in the thermal efficiency. For XNH3 = 0.63, the measured thermal efficiency was ∼0.23, which is significantly lower than that found for operation with pure methane, ∼0.27. Although the CO2 concentration was found to decrease linearly when the ammonia fraction was increased, measurements reveal that NOx emissions increased rapidly, with a maximum NOx concentration of 2161 ppmvd. The concentration of N2O also increased rapidly when the ammonia fraction was increased. Due to N2O's very large global warming potential, this more than canceled the benefits associated with the reduction of CO2 emissions. Consequently, results showed that, even though stable operation of the Ansaldo AE-T100’s mGT in its original configuration is possible with ammonia-methane blends at least up to 60 kWe, hardware modifications will be required to comply with current NOx regulations and ensure sufficiently low N2O emissions.

Ammonia as a fuel offers the potential to avoid carbon emissions, but its combustion is hindered by low reactivity. Here, the extinction limits of NH3 and NH3 plus reactivity enhancers were measured in the counterflow laminar non-premixed flames. A stable NH3-N2 flame was established with an oxygen-enriched oxidizer stream, and when the fuel was blended with CH4, C2H6, C3H8, and H2. For blended mixtures, results showed that CH4 has the least potential to enhance the stability of NH3 flames compared to the other additives. The extinction limits of C2H6 and C3H8 blended NH3 flames are nearly identical. At low percentage addition, H2-blended flames extinguish earlier than those blended with C1-C3 hydrocarbons, but this trend is reversed at higher H2 blends. Experimental conditions were simulated using Okafor et al. 2018 model and an extended Zhang et al 2021 model developed here. The models captured the measured trends, including the crossover between NH3-H2 and NH3-C2/C3 hydrocarbon fuels. Quantitatively, both models under-predicted the extinction limits of NH3-N2/enriched oxidizer flame. Better quantitative agreement is observed for the blended fuels using the model developed here. Discrepancies have been observed in the reported rates for reactions involving HNO (+OH, H), and if addressed, could improve models' capability in predicting extinction behavior in non-premixed flames. Numerical analyses were carried out to understand the kinetic coupling between NH3 and H2/C2-C3 in counter-flow flames. Extinction limits of NH3-C2-C3/H2 flames are shown to be affected by H abstraction and NH3 related chain termination reactions, heat producing reactions, and chain branching reactions. It has also been observed that at high blending ratios, C2H6/C3H8 addition in NH3 flames reduced the peak H and OH concentration via recombination and termination reactions, which compete with branching pathways. H2-blended flames are mostly influenced by reactions producing active radicals.

This work presents an experimental study on the influence of the pilot flame characteristics on the flame morphology and exhaust emissions of a turbulent swirling flame. A reduced-scale burner, inspired by that fitted in the AE-T100 micro gas turbine, was employed as the experimental platform to evaluate methane (CH4) and an ammonia-methane fuel blend with an ammonia (NH3) volume fraction of 0.7. The power ratio (PR) between the pilot flame and the main flame and the fuel composition of the pilot flame was investigated. The pilot power ratio was varied from 0 to 20% for both fuel compositions tested. The NH3 volume fraction in the pilot flame ranged from pure CH4 to pure NH3 through various NH3–CH4 blends. Flame images and exhaust emissions, namely CO2, CO, NO, and N2O were recorded. It was found that increasing the pilot power ratio produces more stable flames and influences most of the exhaust emissions measured. The CO2 concentration in the exhaust gases was roughly constant for CH4-air or NH3–CH4–air flames. In addition, a CO2 concentration reduction of about 45% was achieved for XNH3 = 0.70 compared with pure CH4, while still producing stable flames as long as PR ≥ 5%. The pilot power ratio was found to have a higher relative impact on NO emissions for CH4 than for NH3–CH4, with measured exhaust NO percentage increments of about 276% and 11%, respectively. The N2O concentration was constant for all pilot power ratios for CH4 but it decreased when the pilot power ratio increased for NH3–CH4. The pilot fuel composition highly affected the NO and N2O emissions. Pure CH4 pilot flames and higher power ratios produced higher NO emissions. Conversely, the NO concentration was roughly constant for pure NH3 pilot flames, regardless of the pilot power ratio. Qualitative OH-PLIF images were recorded to further investigate these trends. Results showed that the pilot power ratio and the pilot fuel composition modified the flame morphology and the OH concentration, which both influence NO emissions.