Errors from the air: The trials and tribulations of developing ammonia catalysts
When Shelley Minteer at the University of Utah first got started studying how bacterial enzymes called nitrogenases produce ammonia, she noticed something funny. Some days, the complexes wouldn’t produce ammonia. On other days, they’d produce a lot. The culprit? The cleaning lady.
“We would see spikes in production the days she cleaned the floor,” Minteer says.
Nitrogen and ammonia are all around us. Nitrogen makes up 78% of the air we breathe, and nitrogen-containing molecules like ammonia are in numerous plastics, textiles—and cleaning supplies. These molecules can stick to tubing, gloves, and glassware. “It‘s very difficult, if not impossible, to get all of the contaminants of ammonia out of all the samples,” Minteer says. Contaminants also include other nitrogen-containing compounds, such as nitrites and nitrates, which can easily react to make ammonia.
The field of new ammonia-producing catalysts is still young, says Lauren Greenlee, a chemical engineer at the University of Arkansas. “The catalysts just are not very efficient.” Scientists make small amounts of a catalyst and then test it in small-scale setups. “The problem is that the amounts of ammonia that are actually produced by many electrocatalysts are not much larger than what you might measure in the background.”
So how do you know if the ammonia you’re measuring actually came from your catalyst instead of from contaminants in the lab? Without proper controls, you don’t, Minteer says. If part of that ammonia is coming from the background, scientists might think that their catalyst is working well when it may not be.
Currently, journals don’t require data on specific control experiments to publish data from an ammonia-producing catalyst. Whether the journals should require those controls is a matter of debate in the community. “I’ve talked to some people who have argued that we should wait and not do controls,” Greenlee says. Maybe, these members of the field argue, the catalyst community will move forward, and catalysts will get more efficient so that the difference between what the catalyst is producing and the amount of ambient ammonia will become larger.
While that may happen, that wait-and-see approach has issues, Greenlee says. If a group reports a high-performing catalyst, other researchers may start working with it, thinking that it’s an improvement. “But what if that’s not the right direction to go because the group didn’t measure their background correctly?” Greenlee asks.
Greenlee thinks that researchers should run controls and take background measurements for every catalyst on every day they run experiments. Such controls would include running experiments with isotopically labeled molecules as a final evaluation of successful catalysts so scientists know where the nitrogen in ammonia came from. Papers should also report the results from these control experiments. “Even if a lab is doing appropriate controls, it’s very hard to tell as a reviewer” because they’re not adequately reported in the paper, she says.
“There are surely errors made in the history of science,” says Karthish Manthiram, a chemical engineer at the Massachusetts Institute of Technology. “As long as everyone admits to their errors, we all move forward together.”