Considering Ammonia

Kevin Kilty

Among the substances on which modern society depends are many with a large “carbon footprint”.  This makes them targets of the carbon is bad club. A couple of days ago a copy of the American Institute of Physics (AIP) newsletter Scilight came into my inbox, and one of the featured articles involved research into using a cold plasma over water to fix a number of nitrogen species in the water. This got me thinking about the production of ammonia, which is one of those maligned but indispensable substances. During the summer before I started college in 1971, and for the next three summers, I worked as a maintenance roustabout in a fertilizer and explosives plant. My search of the WUWT archive didn’t turn up an article about ammonia per se, and so this essay was born.

To say that ammonia makes fertilizers and explosives doesn’t do justice to its importance. Probably one-half the human race depends on anhydrous ammonia or fertilizers derived from it for their food. In the form of high density ammonium nitrate prill it provides a safe and economical explosive.[1]  Since our elites are obsessed with remaking the modern world, a good question to ask is “what alternatives do they have to produce this essential substance without using fossil fuel?”

 Estimates are that industry produces about 150 million metric tons of ammonia per year, and that this activity releases around 1.2% of the human emissions of CO₂ each year. A new role suggested for ammonia is as a carrier of hydrogen for a transportation fuel. The ammonia molecule (NH3) contains three hydrogen atoms which can be released via reforming onboard a vehicle to provide hydrogen for a fuel cell to then power electrical motors. If technology should go this direction, then the manufacture of ammonia will increase some 30 times over its present amount, taking the need for efficient and economical manufacture from merely desirable to imperative.

How we make ammonia

There are two separate issues in the quest for a carbon free manufacture of ammonia. The first is to find effective ways of producing hydrogen. The second is to find effective ways of combining this hydrogen with nitrogen to produce ammonia. In the course of this brief essay I will use the terms “manufacture of ammonia” or “nitrogen fixing” interchangeably. 

A source of hydrogen

Many people probably think of making hydrogen by passing an electrical current through water in the process of electrolysis — a demonstration they probably saw in high school chemistry. However, industrial sources of hydrogen more often come from fossil fuels, or in fact any source of carbon, by using the water gas-shift reaction or the alternative dry reforming of natural gas using reactions between methane and CO₂.

The water-gas reaction uses steam (generated with fossil fuels typically) which reacts at high temperature with carbon to produce hydrogen, contributed by water vapor, and carbon monoxide. When followed by the shift reaction, the carbon monoxide reacts with additional steam to produce carbon dioxide and yet more hydrogen. With coal as its feedstock the water-gas shift reaction produces two molecules, or a bit more, of hydrogen and one molecule of carbon dioxide per atom of carbon. Using natural gas (methane) as a feedstock will produce two extra hydrogen molecules. Since natural gas as a feedstock has better economics in most cases, it has come to dominate the feedstock for ammonia production worldwide. I will mention no more about the reforming process from here on to concentrate on ammonia synthesis. However, since chemical reactions rarely run to completion, it should be apparent that the output gas stream from reforming natural gas, or from the water-gas shift reactions, will contain water vapor, methane, carbon dioxide and carbon monoxide in addition to hydrogen. This has important considerations for ammonia synthesis.

The Haber-Bosch process

The introduction of the Haber-Bosch process in 1913 was a world changing event. Figure 2 shows a simplified diagram of the process. Components needed to make ammonia, nitrogen from air typically, and hydrogen are put into a reactor at high pressure (20-40 MPa) and high temperature (400-650C) in the presence of iron catalysts to carry out this ideal model of the reaction

N₂ + 3H₂ -> NH₃ – Energy

That energy leaves on the right side of this equation tells us the reaction is exothermic — it releases heat. Note also that the left side of the equation consists of four kMoles of gasses (one of nitrogen and three of hydrogen) while the right side involves only two. Between these two observations we can employ Le Chatelier’s principle to predict that high pressures will push equilibrium toward the products, ammonia, which is good, but that high temperature will push equilibrium toward reactants. The reason for high temperature is to nudge the reaction to occur at a useful rate. It is imperceptibly slow at room temperature.

The real mixture of gasses input to a reactor consists also of argon from the air, and CO₂ and carbon monoxide, CO, from chemical equilibrium in the reforming or water-gas reactions. Carbon monoxide is the most troublesome component because it will poison the iron catalyst.[2] To remove it the gas is put through a methanation step to turn carbon monoxide into methane. This represents an energy loss, but is unavoidable. Carbon dioxide is easily captured and can be sold as a coproduct to make fertilizer grade or cattle feed grade urea,[3] or it can be sold for carbonated beverage production.

In a single pass through a reactor only about 20% of the reactants combine to produce ammonia and so after refrigeration of the output stream to remove ammonia, the residual gasses are compressed again and go back in the reactor for another pass. Thus, the origin of the name Haber-Bosch loop.

Minimum energy input

No matter how we produce ammonia there is a minimum energy input required by the first and second laws of thermodynamics. We can estimate this minimum input by studying the process of burning ammonia in oxygen

4NH₃ + 3O₂ -> 2N₂ + 6H₂O – Gibbs Free Energy

From a chemistry tables book (CRC, Lange’s or Perry’s tables) we find that for each kilogram of nitrogen we obtain a maximum of 19.2 Mj of free energy. Thus, reversing this process, for each kilogram of nitrogen fixed as ammonia we must supply a minimum of 19.2 Mj of energy. This is true no matter what process we employ – there is no magical route requiring no energy. The current best practice of Haber-Bosch ammonia production requires 27Mj per kilogram. The difference of 7.8Mj between best practice and minimum comes from losses (irreversibilities in the parlance of the engineer) in the compression work involved, the heat losses from high temperatures in reactors, and even from the work needed to separate the product ammonia from other gasses in the output stream. There are many irreversibilities involved, but nonetheless the process energy efficiency is 70%!

Seventy percent sounds pretty good, but for global warming advocates the sin of the Haber-Bosch is that it produces any CO₂ at all. Even 1.2% of global emissions is intolerable.

Alternative production routes for ammonia

One route to greening ammonia production is using a fully electrical process.[4] First, we would produce hydrogen through electrolysis of water, then provide high pressure using electrically driven compressors rather than the more typical steam turbines or methane fueled industrial engines. All of the components of this version of the Haber-Bosch process are at a technical readiness level (TRL) of 9 (see [5] for an explanation TRL) and in that sense are ready for deployment. However, this version of the process is not market ready as the energy inputs are three times the fossil fuel version of Haber-Bosch. Electrolysis of water, for instance, is itself only about 60% efficient. No one makes hydrogen this way except those with extremely cheap hydroelectric sources. 

Most natural fixation of nitrogen occurs through symbiotic bacteria on the roots of certain plants, legumes mainly, and other bacteria. This suggests that fermentation might be another route to nitrogen fixing, or even ammonia production. By comparison world-wide production of beer is about 190 million metric tons; so there are analogous bio-industrial processes at this scale.

Additionally it is possible to use metallorganic compounds in a chemical mimicking of biological pathways. Both of these ideas have very low TRL, and resemble the fine chemicals industries which are noted for their very large waste to product ratios.[6] This is an impediment to making ammonia by these routes.

Finally, the AIP reading suggestion [7] which initiated this essay, promotes making reactive nitrogen species using a cold plasma in air above a water surface followed by dissolving the produced species in the water. This method presents some advantages when ammonia is used as a fertilizer as the nitrogen fixing can be done on site and put through irrigation water. However, the energy input per kilogram of nitrogen quoted in the article, 3.5Gj/kg, is one-hundred times that of the Haber-Bosch process. Taking a small 25kW turbine as typical of what a person might install near an irrigated field, and assume 20% capacity factor during the irrigation season (see this for typical summer c.f.), this energy figure represents only about four kilograms of nitrogen in a month. In other words, unless the energy input is improved tremendously, it is neither a technically ready nor market ready process. 

Ammonia as a transportation fuel

The maximum work output indicated in Equation 2, 19.2 Mj per kg, is what we could obtain, sans irreversibilities, in using ammonia as a transportation fuel. This represents only about one-half the energy density of comparable fossil fuels like diesel or gasoline (40+ Mj/kg). Moreover, using ammonia as a transportation fuel requires either direct use in ammonia fuel cell or its conversion back to nitrogen and hydrogen for use in a hydrogen fuel cell. Reforming of ammonia can be done by cracking at 400C, or done by electrolysis, with the addition of a minimum 5.6 Mj/kg of electrical energy. In addition the electrolysis conversion requires quite a bit of overvoltage (another irreversibility) on electrodes in order to make the process run at a useful rate. Either way, in a vehicle this energy has to come from the ammonia fuel itself. Both electrolysis and thermal reforming leave a trace of ammonia in the hydrogen stream, which is a poison to the membranes, and electrodes of fuel cells. Finally there are serious issues of needed cryogenics or high pressure involved in storing ammonia onboard a vehicle.

The alternative to using a hydrogen fuel cell is to use ammonia in a direct fuel cell.  According to [8] the technology pieces for this are at a TRL of 6 (ready for a demonstration project) but they provide this assessment without much justification and a careful reading of their analysis could convince a person that the true TRL is much lower.

Small is beautiful?

One of the great handicaps of renewable energy is its intermittent availability and overproduction at inconvenient times. One potential solution to these problems is to dedicate renewable energy sources to nitrogen fixing for fertilizer or fuel production. [4,7,9] Can a small scale version of some nitrogen fixing technology be constructed on a very local basis, even down to individual wind turbines?

Here we run into two deep problems with “Small is Beautiful” ideas. First, how does one gather small units of product to distribute it without having collection costs dominate the whole endeavor? Advocates of ammonia as a transportation fuel speak of reusing much of the current distribution system for fossil fuels to distribute ammonia. However, none of a collection system for ammonia exists at all.

Second, I spoke early in this essay about irreversibilities as the source of the 30% of losses in the current Haber-Bosch process. All sorts of new irreversibilities will appear in small scale systems resulting from starting, stopping, ramping up, ramping down, enhanced parasitic losses, and so forth. Imagine the efficiency and expense of a home furnace just getting the plenum warm enough to start its fan, when it’s called to shut down — and does this repeatedly.

At one time in China the communist authorities thought backyard blast furnaces were a wonderful small is beautiful idea. What they got was a lot of poor quality iron. Considering the low state of readiness of many alternative means of producing ammonia, or their poor economics at present, I predict ammonia made by the remarkably efficient pairing of methane reforming with a Haber-Bosch loop will dominate industry for many decades to come. 

Notes:

1-Academics, it seems, only think of explosives as war materiel. One paper credited ammonia in explosives as having determined current geopolitical borders! This is silly and misses the point. Without safe and inexpensive explosives mining and construction would return to the 1850s.

2-Carbon monoxide poisons people and catalysts, and for similar reasons.

3-Urea plants are often co-located with ammonia plants for this reason.

4- Smith, et al Energy Environ. Sci., 2020, 13, 331

5- TRL ranges from 1-“a mere idea” to 9-“ready for deployment.” For an interesting discussion of the meaning of readiness level see: https://serkanbolat.com/2014/11/03/technology-readiness-level-trl-math-for-innovative-smes/

6-N. Cherkasov et al.,Chemical Engineering and Processing 90 (2015) 24–33. http://dx.doi.org/10.1016/j.cep.2015.02.004

7-Subramanian et al, Plasma-activated water from DBD as a source of nitrogen for agriculture: Specific energy and stability studies, J. Appl. Phys. 129, 093303 (2021); https://doi.org/10.1063/5.0039253

8-https://www.intechopen.com/predownload/40233

9-Marika Wieliczko, and Ned Stetson, Hydrogen technologies for energy storage: A perspective Published online by Cambridge University Press:  09 December 2020