Ammonia is expected to be a hydrogen carrier that has potential as a carbon-free fuel. Ammonia is known as a nonignitable fuel, and it is not easy to hold ammonia flames under atmospheric conditions. A demonstration test with the aim of showing the potential of ammonia-fired power plants was conducted using a micro gas turbine. A 50-kW-class turbine system firing kerosene was selected as a base model. More than 40 kW of power generation was achieved by firing ammonia gas or a mixture of ammonia and methane by modifying the combustor, the fuel control device, and the gas turbine startup sequence. The prototype bifuel combustor is a swirl combustor employing a non-premixed flame and a decreased air flow rate near a gas fuel injector for flame holding. Ammonia combustion in the prototype bifuel combustor was enhanced by supplying hot combustion air and by modifying the air inlets. However, the exhaust gases from the ammonia flames had high NOx concentrations. NOx removal equipment using selective catalytic reduction can reduce NOx emission levels to below 10 ppm from more than 1000 ppm (converted value of NOx to 15% O2) as already reported. However, downsizing of NOx removal equipment should be achieved for practical use. Therefore, a low NOx combustor was developed. As the first step of the development of the combustor, flame observation in the gas turbine combustor was tried. Although the observation area was limited, an inhomogeneous swirling orange flame of ammonia gas was observed. Then, a combustor test rig was prepared for a detailed observation of ammonia flame under various combustion conditions. The combustor test rig used a regenerative heat exchanger for heating combustion air, and it used an orifice for pressure drop instead of a turbine. Combustion air and cooling air were supplied from two air compressors. At startup of the combustor test rig, a spark plug was used to ignite non-premixed methane and air. After heating the regenerative heat exchanger, ammonia gas was supplied to the combustor instead of methane gas. The exhaust gases from the combustor were analyzed using FTIR (Fourier transform infrared spectroscopy) under various conditions, such as methane firing, methane–ammonia firing, and ammonia firing. Although there are several concepts for NOx reduction, a rich–lean combustion method was applied first for ammonia firing. The rich–lean combustor modified from the prototype bifuel combustor also could burn ammonia well in cases of both methane–ammonia firing and ammonia firing. The rich–lean combustor succeeded in reducing NOx emission from methane–ammonia combustion to half the value measured in the case of the prototype bifuel combustor.

This content is only available via PDF.
You do not currently have access to this content.