So here is the bad news: Combustion plant emissions of oxides of nitrogen or mono-nitrous oxides, which primarily consist of nitric oxide (NO) and Nitrogen Dioxide (NO2), also called NOx, are a source or participant in the formation of air polluting smog, ozone, and acid rain. In industrial and urban areas NOx reacts with moisture and ammonia to create nitric acid which can cause respiratory disease and aggravate heart disease in humans.

Nitric acid dissolved inmoisture has also been linked to acid rain. In combinationwith volatile organic compounds, NOx participates in the formation of ozone, which can cause damage to lung tissue and reduction in lung function. Clearly, NOx emissions are bad and should be controlled. However, as with all emissions regulation the question is: How much is sensible?

Fundamentally there are three major thermo-chemical pathways to formNOx in a combustion plant:Thermal, fuel, and prompt NOx. Thermal NOx is formed in any high temperature air-combustion process because of the disassociation of nitrogen (N2), which makes up about 78% by volume of our air, into individual atoms of nitrogen, which in the presence of oxygen can react to form oxides of nitrogen (NO or NO2).

Fuel NOx comes from the combustion of nitrogen-bearing fuels, such as oil and coal, and in these types of plants can make up the majority of NOx exhaust emissions. However, fuel NOx is usually not a major contributor of NOx emissions in gas-fired plants.

Finally, prompt NOx is associated with the reaction of atmospheric nitrogen with fuel hydrocarbon radicals. While this process is not as well understood as the other two, it only contributes small amounts of NOx to overall emissions and is only relevant for low, single digit NOx emission regulations.

Now the good news

Most modern gas turbines control NOx emissions by maintaining a low flame temperature through the dilution of the fuel gas with compressor air upstream of the primary combustion zone in a process called lean premixed combustion (aka dry low NOx). This results in NOx emissions of 5-50 ppm, depending on the size, type and method used. By comparison, older combustion systems exceeded NOx emissions of 300- 400 ppm. We have come a long way over the last 20 years.

Most large gas turbine power plantsmust meet emissions requirements below 10 ppm and, in some major urban (non-attainment) areas below 5 ppmor even lower—a recent DOE solicitation specified below 2 ppm NOx. Turbines below 50 MW are often allowed to have up to 25 ppm. It is important to recognize that designing and operating a gas turbine at these low-NOx emissions is complex and places limitations upon operating range and maintenance costs.

Furthermore, when comparing different plant types based on their contribution on a kg NOx per producedMWhr, gas turbine air permits requiremuch lower levels than other plant technologies. For example, air permits for reciprocating gas engines typically allow NOx emissions 3-10 times higher than those for gas turbines. This intuitively seems to make little sense.

The question thus becomes: Where do further reductions cross the line of diminishing returns versus investment and operating cost?Does a drop from15 ppmto 9 ppmsignificantly reduce the tons of NOx emitted by gas turbine plants? For a 200 MW power plant, lowering NOx emissions from 250 ppmto 15 ppmmeans about 5,500 tons/year ofNOx, but a further lowering to 9 ppmonly yields another 100 tons/year. What really matters is total tons per year of NOx. These single-digit reductions provide little overall improvement. However, small changes in regulations have a huge effect on the design, equipment/monitoring/operating/maintenance costs, and operating range.

Accurately measuring NOx below 10 ppm, for example, is difficult. Instrument providers list measurement uncertainty as +/- 1 ppm. However, this is the direct sensor uncertainty. When the added uncertainty of the other measurement parameters are included (flow, oxygen, moisture, temperature and barometric pressure) as well as the uncertainty introduced by sampling lines and un-even flow distributions in the stack, the measurement uncertainty can easily exceed +/-3 ppm.Thus it is difficult to provide a low, single-digit guarantee.

Add-on technology to achieve very low emissions such as a selective catalytic reduction (SCR) system can be costly. In this device, NOx is scrubbed from the flue gas with ammonia or urea and reduced to nitrogen and water using a catalyst.

Above 95% NOx reduction efficiency can be achieved, but that requires over-injection of ammonia. NOx is reduced, but ammonia or urea is emitted.One could argue that this provides limited environmental benefit. However, if the conversion efficiency is kept below 70%, ammonia slip is minimal.

One final consideration: To achieve low thermal NOx, the flame temperature has to be reduced to avoid nitrogen disassociation. Although firing and flame temperatures in the combustor are somewhat independent, this still indirectly limits the firing temperature of the gas turbine. From basic thermodynamics we know that the hotter a gas turbine operates the higher its efficiency, i.e., by reducing flame temperatures plant efficiency is automatically limited, which results in higher CO2 emissions. Ultra low NOx lean premixed combustors also tend to have slightly higher CO and unburned hydrocarbon emissions.

Low NOx emissions may be desirable, then, but may not be practical from an operations perspective, and sometimes may even have an environmentally adverse effect. Current NOx requirements are already stringent. There is little practical sense or environmental improvement by further lowering of NOx requirements.


Klaus Brun is the Machinery Program Director at Southwest Research Institute in San Antonio, Texas. He is also the past Chair of the Board of Directors of the ASME International Gas Turbine Institute and the IGTI Oil & Gas applications committee.

Rainer Kurz is the manager of systems analysis for Solar Turbines Incorporated in San Diego, CA. He is an ASME Fellow since 2003 and past chair of the IGTI Oil & Gas applications committee.