(618a) Methods for Low NOx Combustion in Ammonia/Natural Gas Dual Fuel Gas Turbine Combustor | AIChE

(618a) Methods for Low NOx Combustion in Ammonia/Natural Gas Dual Fuel Gas Turbine Combustor

Authors 

Onishi, S. - Presenter, IHI Corporation
Ito, S., IHI Corporation
Uchida, M., IHI Corporation
Kato, S., IHI Corporation
Saito, T., IHI Corporation
Fujimori, T., IHI Corporation
Kobayashi, H., Tohoku University

Ammonia
is a carbon-free fuel, so it has potential to reduce carbon dioxide emission
from power plants when used as a fuel. However, combustion characteristics of ammonia
are notably different from hydrocarbon fuels, especially regarding NOx emission[1]. The nitrogen atom of the ammonia molecule may cause
high NOx emission. Therefore, special techniques to reduce NOx emission are
essential for gas turbine combustors which burn ammonia and natural gas. The
results of our previous study [2] showed the characteristics of NOx emission in
single-stage combustion. In this study, the concept for low-emission combustion
in two-stage combustion has been examined numerically and experimentally.

Numerical
study was conducted to investigate the combustion characteristics in two-stage combustion.
A reactor network model comprising two perfectly stirred reactors (PSR) was
calculated with CHEMKIN-PRO  to evaluate NOx emission in short computation
time. The reaction mechanism constructed by Tian et al. [3] was used. In the
primary zone, methane/ammonia mixture is supplied as fuel. Then, only ammonia
is supplied to the secondary zone (Fig.1). Air, methane and ammonia flow splits
to primary and secondary zone are changed as parameters while total amounts of
methane, ammonia and air flow rate are fixed. Figure 2 shows the numerical
results in terms of NOx concentration in secondary zone against equivalence
ratio in primary zone for three ammonia flow splits. These results revealed that there are two methods to achieve
low NOx combustion, namely rich-lean combustion and combination of lean-burn
and secondary ammonia supply. In the first method, NOx is produced only in the primary
zone in fuel-rich condition and burned gas is diluted in the secondary zone
with secondary air. Therefore, the NOx formation in the primary zone is dominant
for emission. In the second method, lean combustion condition in the primary
zone leads to low temperature and low oxygen concentration in the secondary
zone. These conditions lead to low NOx concentration at the combustor outlet.

These predicted combustion characteristics have been examined
experimentally. Experimental results showed that NOx emission behavior is in
good agreement with the numerical results.

Fig. 1 Schematic of reactor network model.

Fig. 2 Numerical results of NOx concentration
against ammonia flow splits and equivalence ratio.

Reference

[1] M. T. Javed, N. Irfan, B.M. Gibbs, J.
Environmental Management 83: pp. 251-289 (2007).

[2] S. Ito,
S. Kato, T. Saito, T. Fujimori and H. Kobayashi, Proceedings of the Fifty-Third
Japanese Combustion Symposium (in Japanese), 2015

[3] Z. Tian, Y. Li and P. Glarborg, Combust. Flame
156: 1413-1426 (2009).