Hydrogen

When it comes to hydrogen, we’re probably overestimating NOx emissions. Here’s why

Researchers have found that a standard method to calculate NOx emissions can result in big errors when applied to hydrogen.

When it comes to hydrogen, we’re probably overestimating NOx emissions. Here’s why
(Earth as seen from the International Space Station. Credit: NASA)

Like natural gas, combusting hydrogen at high temperatures produces nitrogen oxide whose emissions need to be permitted by an environmental regulator and controlled.

But hydrogen’s molecular makeup means that it produces nitrogen oxide (NOx) differently than methane-based natural gas.

Researchers have found that a standard method used across the industry to calculate NOx emissions from gas-fueled power plants can result in potentially big errors when applied to hydrogen as an energy source.

The error may complicate the case for introducing more hydrogen into the electric power generating mix.

The error first came to light a year ago in a monograph published by the researchers from Georgia Tech and the Electric Power Research Institute (EPRI). A more detailed paper from the team came out last summer, published by the American Society of Mechanical Engineers. And the results were highlighted in a session at the recently concluded POWERGEN International conference in Orlando.

In a nutshell, the research lays out the case that many studies are calculating NOx emissions incorrectly “by as much as 40% against high-hydrogen systems.”

The error could be a relatively simple one to fix, and likely would require environmental regulators to change how they evaluate hydrogen in comparison to other fuels.

The researchers noted out that subpart KKKK of the U.S. Environmental Protection Agency’s code lays out federal emissions limits for NOx and sulfur dioxide (SO2). They said that the primary source of NOx in most gas-fired systems is the ambient air itself. Air’s dominant parts (nitrogen and oxygen) react together under high-temperature conditions. And while one benefit of hydrogen is that its combustion releases no CO2, combusting H2 does generate NOx, a result of heating air to high temperatures.

For years in the U.S., air quality permits have been used to regulate and limit permissible emissions from power plants. In practice, these permitted levels have been based on what the researchers said is the net mass production of regulated pollutants. 

An overlooked correction

And here’s where a bit of chemistry and math are needed to understand the problem.

The researchers say that in evaluating emissions, the volumetric stack concentrations of pollutants–and not their actual mass production rates–are measured. These measurements are done using continuous NOx analyzers located at individual plants. 

As a result, work needs to be done to convert volumetric measurements (referred to in shorthand as ppmv) to a mass basis (referred to either as lb/MMBtu or lb/hr).

In their July 2022 paper the researchers wrote, “Quantifying such differences is not a difficult task, but the importance of doing so does not seem to be widely appreciated by the combustion community.” Part of the challenge, they said, is that the combustion community has “traditionally focused on relatively standardized fuels that display minimal variability.”

(Image credit: 123rf).

To help explain this, the researchers noted that gas turbine emission codes often define allowable fractions of exhausted NOx based on a standardized method for preparing samples. In this process, combustion products are sampled and the water is removed. The dry sample that results then is mathematically corrected to 15% O2 (for gas turbines) before measuring the NOx levels. 

This correction is done, the researchers said, in order to evenly evaluate systems with varying levels of excess air. In the United States, typical stack exit permits range from 3-30 ppmv at 15% O2. But—and this is a crucial point—the mathematical idea known as “the constant of proportionality” that exists between pollutants’ mass production and this measured ppmv value depends on what fuel is being analyzed.

This point is well known in the environmental community and typically is corrected by what is known in industry jargon as an “F-factor.”

Flame problem

In their longer ASME paper, the authors said the challenge associated with fuels that have a high percentage of hydrogen comes about in lean, premixed combustion systems. Modern dry low NOx combustors operate in a lean premixed regime. In this regime, the flame temperature can be controlled to minimize NOx production; the flames themselves are stabilized by balancing both flame propagation and flow velocity. 

But H2 flames behave very differently from CH4 flames. For one thing, H2 flames can have up to 10x higher flame speeds at a given equivalence ratio. For another, they can exhibit extreme sensitivity to what the researchers said were flame stretch and thermal-diffusive instability. Such effects cause H2-fueled systems to be more prone to flame flashback than their natural gas counterparts. 

What’s more, the effects also shift parameter regions where combustion instability occurs. That means that operational experience developed for natural gas systems need to be adjusted for fuels containing H2.

However, as work progresses to look at various low NOx combustion technologies, including hydrogen, this correction often is not applied. The researchers said that as a result, technology evaluations are “improperly comparing measured NOx ppmv emissions between one fuel and another.”

The researchers said a simple correction can address the overestimation. Credit: Georgia Tech and EPRI

In particular, many evaluations directly compare NOx emissions from methane/hydrogen blends based on NOx ppmv concentration values. And while corrections are largely negligible for many fuels (such as between natural gas and diesel), corrections can be “very substantial” for hydrogen.

That’s because at conditions with equal power output, hydrogen/methane fuel blends result in combustion products that have higher proportions of H2O (water) and O2 (oxygen) than pure methane alone. So, even when mass production rates of NOx emissions are the same, systems that use more hydrogen will have higher composition-based value when reported in terms of ppmv at 15% O2.

That can be a problem for plants looking to use a blend of methane and hydrogen. In short, a volume-based (ppmv) measurement can erroneously indicate higher NOx emissions from H2-blended systems for the exact same mass of NOx as a methane-fired plant. The researchers pegged the correction at about 7% for a 50%/50% H2/CH4 blend; 17% for an 80%/20% blend; and 37% at 100% H2.

By comparison, similar corrections for other common fuels are “generally negligible,” the researchers said. A comparison between methane and diesel fuel yielded a correction of only around 2%.