Hacker News Reader

Comment by tonyarkles | Hacker News Reader

From doing a bit of digging into this myself after hearing it being promoted on the radio, I both agree and disagree. Working through the $/acre for on-site produced ammonia using locally-generated electricity (I forget if I was looking at PV or Wind, doesn’t matter) the payback period was surprisingly quick, like 2 or 3 years. One of the big reasons is that you don’t actually need a very big system; you only need on-farm ammonia a few days a year but you have all year long to harvest and store enough for those few days.

View on HN

Then you have 1,000 separate ammonia generation systems that need upkeep, repairs, etc. It is doing chemical reactions under pressure to produce presumably ammonium nitrate which is explosive. The engineers you would need probably aren’t cheap, and storing large quantities of ammonium nitrate all year is sort of sketchy. Timothy McVeigh took out an entire federal building with about 4800 pounds of the stuff mixed with car fuel.

It’s usually a lot cheaper and safer to condense all that demand into a purpose-built facility where they can hire an engineer to look after all the generation, have centralized container storage, backup parts, etc.

None of that prevents putting the panels or windmills on farmer land, though. You could imagine some sort of “get a discount on ammonia if we can put a solar panel in your field” scheme. Or, even more simply, put up your own panels, sell the power back to the grid, use that money to buy fertilizer from the factory buying your power off the grid.

I don’t know the exact process but I don’t think ammonium nitrate actually gets involved. The system I was looking at (Green Lightning) actually doesn’t make ammonia but rather liquid nitrogen fertilizer

Plasma to take N2 + O2 -> NO2 and NO gas

The NO and NO2 are dissolved in water to become HNO2 and HNO3, which then become NO2- and NO3- ions in the water.

It’s a low pH solution which stabilizes it and then you dilute right before application so that you don’t burn the crops.

I am curious what kind of plasma is used, as there are many ways to create plasma under various conditions, with very different efficiencies, selectivity, yield, etc.

Among the worst is the Birkeland-Eyde process, hot plasma by electric arcing, like it happens naturally during a thunderstorm. Even microwaving the air is more energy efficient. And amongst the most promising are transient non-equilibirum / cold plasma approaches using ns pulse generators.

Edit: So according to this video "Green Lightning" is indeed using the Birkeland-Eyde process. In other words, there is no way they even come close to their energy efficiency and yield claims. https://youtu.be/8TOnyAwJyvg?si=D9hat_XAfvuAAmpB&t=165

That is the Birkeland-Eyde process, which was superceded by the Haber-Bosch process because it is much more energy effective.

As far as I understand, the reasons for using methane as the hydrogen source in industrial processes is because of costs and the difficulty in scaling electrolysis industrially.

Right, makes perfect sense.

So this is pretty far outside of my area of expertise, but https://pmc.ncbi.nlm.nih.gov/articles/PMC8133363/ suggests that there may be an evolution of Birkeland-Eyde that isn’t nearly as bad as the original process was. I haven’t gone through this in detail and even if I had I probably don’t have the background to do it justice.

That sounds like the Birkland-Eyde process. It's cool because it uses only electricity and air and seems to produce something from nothing, but it's not as efficient as Haber-Bosch.

Reply on news.ycombinator.com