Applied Chemistry Laboratory, Imperial Institute of Technology. Saint Petersburg.
Every morning when Dr. Alexánder Borísovich Lebedev opened the laboratory door, he read the room by smell before reading anything else: paraffin oil burning under the crucibles, which meant the high-pressure furnace catalysts had run through the night within normal parameters. Hydrogen sulfide in perceptible quantity, indicating that the iron ore from the new batch still carried impurities that needed to be removed before it could be used as a base for the iron oxide catalyst. And below all of that, barely distinguishable but present, the metallic bite of compressed nitrogen escaping from the copper pipe fittings, which meant the gas manifold connections needed retightening before the next cycle began.
Forty years working in chemistry laboratories had taught Lebedev that problems announced in advance were the best kind, because you still had time to do something about them.
He hung his coat on the hook by the door, put on his thick cotton lab coat, and reviewed the log notebook that his assistant Mitia always left on the entry bench with the night-shift observations. Mitia Yegorovich Platonov was twenty-two years old, had small handwriting, and possessed the virtue of recording exactly what he had observed without adding interpretations that weren't his to make. It was the most valuable quality in a laboratory assistant, and Lebedev knew this because he had fired three people before finding someone who had it.
The notation read: "pressure stable in main reactor for twelve hours, catalyst temperature within range, estimated conversion yield at end of cycle sixteen percent molar."
Sixteen percent.
Lebedev closed the notebook with a controlled motion and stood looking at the reactor for a few seconds.
A sixteen percent molar conversion of nitrogen to ammonia meant that of every hundred nitrogen molecules entering the reactor under current conditions, sixteen ended up combined with hydrogen to form ammonia. The remaining eighty-four exited through the discharge pipe unreacted, process flow wasted, needing to be recompressed and recirculated.
In industrial terms, it was a percentage that made the process economically unviable due to the energy consumption that recirculation required. Fritz Haber, according to the technical reports Lebedev had obtained through academic channels, was working with conversions of fifteen to eighteen percent in his Karlsruhe reactors. Carl Bosch's target at BASF for industrial scale was to reach sustained operating conditions that made the process profitable at any conversion above twelve percent, provided the catalyst lasted long enough and the pressure could be maintained continuously.
Lebedev had sixteen. A number that was in the right range and that at the same time resolved none of the three fundamental problems sitting on his table.
Mitia appeared through the laboratory's rear door with two cups of tea, punctual as always.
"The number two reactor catalyst has been running forty-eight hours continuously." He said, setting one cup on the log bench. "Activity has dropped eight percent from the first twelve hours."
"Eight percent?"
"Calculated against the batch's initial yield."
Lebedev took the cup. Catalyst degradation was the second problem.
An iron oxide catalyst with potassium and alumina promoters had a reasonable service life under laboratory conditions. Under continuous industrial reactor conditions, with synthesis gas passing through at two hundred atmospheres of pressure and temperatures between four hundred and five hundred degrees Celsius, the catalyst's active surface progressively compacted, the microscopic pores where the reaction occurred sealed themselves off, and activity fell. The question Lebedev had spent three months unable to answer was how quickly it fell, and whether it could be regenerated or had to be replaced.
Forty-eight hours of continuous operation with an eight percent drop in activity was a degradation rate that made the process non-regenerable with available methods. It meant stopping the reactor every two days, cooling the system, removing the spent catalyst, loading fresh catalyst, and restarting. Each shutdown represented eight hours of downtime and an amount of energy to reheat the system equivalent to several days of production.
Not viable at industrial scale. Not yet.
"Sit down," said Lebedev.
Mitia took his assigned stool by the calculation bench, where there were always several half-used sheets of paper and a slide rule that Mitia handled with the fluency of someone who has practiced enough to not have to think about the procedure.
Lebedev went to the wall chalkboard, which had written on it in four colors of chalk the sequence of reactions in the synthesis process with the current operating conditions for each stage and the yield data from the last twenty cycles.
"Let's go over this week's problem three." He said, pointing to the lower right corner where there was a column of numbers headed: CONTAMINANTS - IMPACT ON CATALYST.
The third problem was the one that worried Lebedev most in practical terms, because it depended less on the reactor engineering and more on the quality of the raw materials they could obtain in Russia at this particular time.
The ammonia synthesis process required purified nitrogen and hydrogen. The nitrogen came from air separation by fractional cooling, a process the laboratory had managed to handle reasonably well using the liquefaction equipment Lebedev had obtained from the Moscow Institute of Physics through two material exchanges that had required more negotiation than some diplomatic treaties. The hydrogen was the problem.
The hydrogen for the process was produced by steam reforming of coke coal, a reaction that generated hydrogen and carbon monoxide.
Carbon monoxide was a poison for the iron catalyst: it adsorbed onto the active sites of the surface and blocked them, reducing activity irreversibly. It had to be removed through a second water-gas shift reaction followed by absorption with potassium hydroxide solution.
The coke coal arriving in Saint Petersburg from Russian mining basins had variable sulfur levels, and sulfur compounds were even more toxic to the catalyst than carbon monoxide. Each different coal batch required adjusting the hydrogen purification process, which introduced variabilities that made it impossible to standardize the reaction conditions.
Haber worked in Germany with Ruhr coke coal, uniform in composition, low in sulfur, processed by BASF with systematic quality controls. Lebedev worked with whatever arrived.
"The Donbass coal batch we received last week." Said Lebedev. "What does the elemental analysis show for sulfur?"
Mitia searched through the papers on the bench.
"Point eight percent by weight. The previous batch was point three."
"Nearly three times as much." Lebedev marked the figure on the board beside the catalyst contamination data. "If we don't correct the hydrogen purification process before using that coal in the next cycle, we'd kill the catalyst in under twenty-four hours of operation."
"Could we increase the residence time in the potassium hydroxide absorption column?"
"We could. But that reduces the hydrogen production flow rate and we have to re-optimize the feed ratio to the reactor. Two days of adjustment, minimum."
Mitia wrote in his notebook with the small, precise handwriting Lebedev appreciated.
"Is there a coal source with more consistent sulfur content in the region?"
"In the Donbass there's natural variability from different geological strata. Cardiff coal has very low sulfur and consistent composition, but supply from Great Britain has been under price restrictions for months that make it prohibitively expensive for laboratory use at the scale we need."
Lebedev didn't mention the reason for those price restrictions, though he knew it. Reports coming from the Ministry of Industry referred to pressure on European raw material suppliers to impede Russian access to certain inputs. Lebedev was not a political man, but he understood market logic well enough to know that when someone blocks access to a critical raw material, the solution isn't to look for more of that material, it's to find a way not to need it.
He went to the sample bench and took out a flat wooden box with twenty-four glass bottles stoppered with corks, each with a handwritten label. It was the collection of mineral and soil samples he had requested from the Institute's Geology Department six weeks earlier.
Inside were samples of lignite from the Moscow region, peat from the northern marshes, coal from the Ural basins, and at the far right of the box, six bottles labeled: NATURAL GAS - CAUCASUS - SOURCE: BAKU.
"Monday we start the natural gas reforming trials." Said Lebedev. "If the Caucasus methane gives us hydrogen with fewer sulfur contaminants than the coke, we switch the hydrogen source."
Mitia looked at the bottles.
"Has the high-pressure piping equipment arrived from the Putilov workshop?"
"It came in yesterday. It's in the storeroom unchecked. That's the first job of the morning."
The laboratory storeroom was a long, narrow room behind the practical classroom, with pine wood shelves reaching the ceiling and a smell of protective oil and the leather gaskets used on pressure pipe fittings. The Putilov workshop order consisted of twenty meters of seamless steel tubing with a thirty-millimeter internal diameter, twelve ninety-degree elbows of the same caliber, six quick-shutoff needle valves, and a glycerin-filled high-pressure gauge with a scale up to three hundred atmospheres.
Lebedev checked each piece against the purchase order. The pipes were fine. Four of the twelve elbows showed a minor defect: the threaded connection fittings had a rough surface finish that didn't correspond to the tolerance specified in the order, meaning that tightening the joint would risk a leak.
He marked the four elbows with red chalk.
Mitia noted the order lot number and the defect description.
"Return?"
"Return the four and request urgent replacement. Put it in the requisition letter this afternoon to the procurement department." Lebedev examined the pressure gauge, a piece of French manufacture, with a large dial and a stainless steel needle that moved without perceptible resistance when he turned it gently. "The gauge is fine."
They put the usable components in their shelf positions. The four defective elbows went into the returns drawer beside the door.
Lebedev washed his hands in the storeroom basin and returned to the main laboratory, where reactor number one was beginning the cooling cycle prior to opening for catalyst inspection.
The catalyst inspection routine was the most delicate work of the day. It required reducing the reactor temperature from the four hundred degrees Celsius of operation, roughly seven hundred and fifty Fahrenheit, to below fifty Celsius before the upper lid could be opened, which took four hours of controlled cooling.
During that time, Lebedev used the opportunity to review the previous day's calculations and prepare the day's experiments.
That morning he had spread on the central bench the results of the comparison he had been preparing for three weeks: his own process's yield data against the published figures he knew from Haber's work, fragmentary, since most of the Karlsruhe research hadn't been published yet, and against the theoretical calculations that thermodynamics established as the maximum achievable limit at different pressure and temperature conditions.
The result of that comparison was, in technical terms, a picture of where the laboratory stood relative to where it should be.
Thermodynamics set the ceiling. At two hundred atmospheres of pressure and four hundred and fifty degrees Celsius, the maximum possible equilibrium conversion was twenty-five percent molar. Any value below that was theoretically achievable with the right catalyst and correct conditions. Haber had published values of fifteen to eighteen percent in his early 1909 experiments, before his work passed to BASF and ceased to be publicly accessible. Lebedev's laboratory was at sixteen.
What the table showed was that the laboratory's results were comparable to the German state of the art from 1909. What it didn't show, but what Lebedev knew, was that Haber and Bosch had advanced considerably since then. The problem wasn't whether the process worked: it worked. The problem was the distance between a laboratory reactor running forty-eight hour cycles and a continuous industrial process producing fixed nitrogen in quantities sufficient for agricultural-scale fertilizer production.
He pulled from the bench drawer a sheet with the Agriculture Ministry figures sent to him three weeks earlier when Lebedev had requested consumption data to calculate the necessary production scale.
Russia imported annually between one hundred fifty thousand and two hundred thousand tons of Chilean sodium nitrate, the saltpeter used as nitrogen fertilizer in the southern fields. The price had risen forty percent over the past five years due to increased global demand and transportation conditions from South America. Part of that supply also passed through commercial channels where the influence of certain companies over the intermediaries created additional markups and delays that in early-planting years could mean the difference between applying fertilizer at the optimal moment and applying it late.
An industrial ammonia synthesis process on Russian soil, using Russian raw materials, would eliminate that dependence. Ammonia could be converted into ammonium nitrate or sodium nitrate through well-understood secondary processes. The nitrogen that the Volga and Ukrainian fields needed could be produced from atmospheric nitrogen and from the country's own coal or natural gas.
It was a line of reasoning Lebedev had found clear from the beginning. The difficulty wasn't in understanding why it needed to be done. It was in the three problems on the board and the chain of secondary problems each one generated.
[Author's Note: When Lebedev says Russia needs 200,000 tons annually, we're talking about 14,000,000 gold rubles per year leaving Russia toward Chile and British companies. To put that in perspective: in 1912, the Tsar could have bought nearly two Gangut-class battleships, the pride of the Imperial Navy, with 14 million rubles.]
At eleven in the morning, the visitor Lebedev had been expecting since the previous Monday arrived at the laboratory.
Konstantin Dmitrievich Smirnov, professor of applied thermodynamics at Moscow University, sixty-two years old, with a wide yellowish-white mustache and the habit of not greeting anyone but simply starting to discuss whatever was on his mind the moment he crossed the threshold, was a quality Lebedev considered a virtue, since it indicated the man hadn't come to waste time.
"I've gone over the data you sent me." Smirnov said, setting his leather briefcase on the only free bench in the laboratory. "The catalyst problem is a combination of sulfur poisoning and thermal sintering. Two independent mechanisms that reinforce each other."
"I already knew that."
"What you may not know is that your deactivation curve data is consistent with a catalyst service life of between sixty and ninety hours of continuous operation if you eliminate the sulfur from the hydrogen feed. Without sulfur, the dominant mechanism becomes sintering alone, which is slower."
Lebedev leaned against the bench.
"Sixty to ninety hours with clean hydrogen?"
"With the promoters you're using and at your reactor temperatures, yes. With hydrogen from Caucasus methane reforming, which has inherently less sulfur than Donbass coal, you could be at the upper end."
"And regeneration?"
"Sintering isn't regenerable with currently available methods. But if the catalyst lasts ninety hours instead of forty-eight, the replacement cost changes." Smirnov opened the briefcase and took out several sheets of handwritten calculations. "I worked out the economic viability figures for those conditions."
Lebedev took the sheets. They were four pages of orderly calculation, with the assumptions explicitly listed in the margins and the data sources indicated for each parameter. The work of someone who distinguished between what he knew and what he was assuming.
The numbers showed that with the catalyst lasting ninety hours, with hydrogen produced by low-sulfur methane reforming, and with the reactor pressure maintained at one hundred and eighty atmospheres, the production cost of one ton of ammonia was forty-eight rubles. The equivalent import price of Chilean sodium nitrate in terms of fixed nitrogen was, in the 1912 market, between sixty and seventy rubles per ton delivered to Saint Petersburg.
There was a margin. Not a wide one, but sufficient for the process to be economically viable if built at large enough scale to spread the fixed equipment costs.
"What minimum scale would you need for the fixed costs not to destroy the margin?" asked Lebedev.
"With the conservative assumptions I used, you'd need a plant with a production capacity of at least ten metric tons of ammonia per day, about twenty-two thousand pounds, for the process to be profitable without state subsidy." Smirnov sat on Mitia's stool, who had gone to the glassblowing workshop to fetch some flasks. "With initial subsidies for equipment investment, the minimum viable scale would drop to five tons per day."
"How much does it cost to build a ten-ton-per-day plant?"
"I haven't calculated that. That's process engineering, not thermodynamics. But to give you an order of magnitude, the Germans are investing in their first industrial plant at Oppau a figure close to one million marks in high-pressure equipment. At the current exchange rate, that's on the order of four hundred thousand rubles just in equipment, not counting the civil construction or utility infrastructure."
Four hundred thousand rubles was a figure Lebedev had no authorization to spend. The laboratory's annual budget, including staff, materials, and maintenance, was twelve thousand rubles.
"I need a meeting with the Ministry of Industry," said Lebedev.
"That's more complicated than the catalysis," Smirnov replied in a neutral tone.
"Probably. But we now have the viability numbers. If the process works in the laboratory, if the costs are comparable to imports, and if the necessary scale is financially within the State's reach, the justification exists. What's missing is for someone with authority to make the investment decision to see those three elements in the same document."
Lebedev went to the chalkboard and began erasing the catalyst contamination data section he had started the morning with. Mitia came back in with the flasks and saw the board being erased with the methodical motion of a professor closing one topic before opening the next.
Smirnov watched Lebedev work on the new scheme appearing on the chalkboard: three columns headed PROBLEM, CURRENT STATUS, and NEXT STEP.
"How long have you been on this project?" Smirnov asked.
"Seven months since I received the brief from the Institute."
"And who originated the brief?"
"The Ministry of Industry through the Department of Industrial Chemistry."
The brief said verbatim: evaluate the technical viability of the ammonia synthesis process from atmospheric nitrogen with raw materials available in Russian territory and produce a report with recommendations for industrial scale.
"What did they tell you about the timeline?"
"That the report was needed urgently. That was seven months ago."
Smirnov paused with the weight of someone who knows well how ministerial bureaucracy works.
"Do you know why there's urgency?"
Lebedev filled the first row of the table on the board before answering.
"The tractors from the new agricultural program will triple the cultivable surface area. More cultivable land means more land to fertilize. And more land to fertilize at current Chilean nitrate prices means that the cost of agricultural inputs will grow proportionally with the expansion. If there's no domestic source of fixed nitrogen by the time those lands are in full production, the tractor program will have solved the tillage problem and created a new one in fertilization."
Smirnov studied him with the attention of someone checking whether a chain of reasoning has weak links.
"You've thought about this beyond what the ministerial brief asks."
"The ministerial brief describes the technical problem. The underlying problem is that an agricultural reform that increases potential production but creates dependence on imported inputs doesn't control the supply that growth will need." Lebedev wrote in the second column of the board: CONVERSION 16%, CATALYST LIFE 48H, H2 SOURCE VARIABLE. "Fertilizers are the third element of the system, after the tractor and the land. If the third element isn't available under stable conditions, the first two don't function in a sustained way."
"Have you told the Ministry that in those terms?"
"I'll tell them in the report. With Smirnov's numbers." Lebedev pointed to the sheets the professor had brought. "The report will have the quantitative foundation it didn't have before. We start drafting it this afternoon."
Mitia looked up from the flasks he was organizing.
"This afternoon?"
"This afternoon. The reactor cooling cycle ends at two. Until then, prepare the elemental analysis of the Caucasus methane batch, I want the sulfur data before noon."
The cooling cycle ended at two-fifteen. Lebedev and Mitia opened the reactor's upper lid with the four-arm torque wrench that hung on the hook beside the equipment, unscrewing the eight locking bolts in the reverse order to which they had been tightened, and carefully removed the lid to set it on the wooden support marked in red paint to indicate which side should face up, to avoid contaminating the sealing gasket.
Inside the reactor, resting on the support grid, the catalyst bed looked as Lebedev had seen it dozens of times before, a cylinder of dark gray-black granulate with bluish tones on the upper edges where the temperature had been highest during the cycle. The smell on opening was sharp, nitrogenous, with a background of sulfur that was the signature of the poisoning mechanism Smirnov had described.
Lebedev took a sample from the bed with the steel spatula, deposited it in the glass bottle marked for that cycle, sealed the bottle, and set it in the analysis tray.
"Tomorrow we'll have the specific surface area data for the spent catalyst and we can compare it with the fresh catalyst. If sintering is the dominant mechanism, the active surface will have fallen more than thirty percent. If it's sulfur poisoning, the fall will be smaller but there will be detectable adsorbed sulfur compounds in the analysis."
Mitia took notes.
"Do we load fresh catalyst for the next cycle?"
"Yes. But with the activation procedure we modified last month, reduction with hydrogen diluted in argon for two hours first, before moving to reaction conditions. We activated the previous batch too quickly and that may have contributed to the initial sintering."
They began loading the fresh catalyst with the spatula, distributing the granulate in layers over the grid with the uniformity Lebedev insisted on, because uneven bed distribution created hot spots that degraded the catalyst locally.
While they worked, Smirnov had spread his calculations across the empty bench at the back of the room and was reviewing some of the figures with a pencil, adding margin notes.
"There's one figure I didn't have when I calculated the costs," Smirnov said without looking up from the paper. "How much fixed nitrogen does Russia need to eliminate dependence on Chilean saltpeter?"
"The Ministry figures I mentioned. Between one hundred fifty thousand and two hundred thousand tons annually of sodium nitrate equivalent."
"And converted to ammonia as an intermediate product, accounting for the conversion losses in the downstream processes..."
"Between eighty and one hundred thousand tons of ammonia per year." Smirnov worked the calculation in the paper's margin. "With plants of ten metric tons per day capacity, you'd need between twenty-two and twenty-eight plants operating in parallel to cover that demand."
"Over a ten-year horizon, yes. Not all at once."
"The first step is a pilot plant at industrial scale, say fifty metric tons per day, that allows the process to be validated with real equipment before committing to the investment of twenty-odd plants." Smirnov looked up. "Is that in the report you're writing this afternoon?"
"It will be."
"Then you also need a site location. An ammonia synthesis plant needs natural gas or low-sulfur coal, water in quantity for the steam, and rail access to distribute the product. Where does Russia have all three?"
Lebedev thought for a few seconds while placing the last layer of fresh catalyst.
"The Volga region has all three. There's natural gas in the southern subsoil, the rivers provide the water for steam, and the Volga basin rail network connects to the main agricultural centers. Tsaritsyn or Samara would be candidates."
"Put that in the report with the technical justification. Don't make whoever reads it do that calculation on their own."
Lebedev closed the reactor lid and began tightening the bolts in the correct order.
"Ministerial bureaucrats don't read things that require calculation," said Lebedev.
"Exactly. That's why you give them the number and the location already worked out, not the argument for why they should calculate them." Smirnov packed his papers into the briefcase. "If the report says: the first plant should be built in Tsaritsyn, capacity fifty metric tons per day, estimated construction and equipment cost of four hundred thousand rubles, construction timeline eighteen months, return on investment through import substitution in three years, that's a report someone can sign a decision on. If the report says: the process is viable under certain conditions that require further evaluation, nobody signs anything."
That was the difference between chemistry and administration. Lebedev knew it, but sometimes needed someone to remind him from outside the laboratory.
Smirnov left at four. Lebedev and Mitia spent the rest of the afternoon drafting the report at the back table, with the board data, the professor's calculations, and the laboratory log records spread out across the wood.
Reactor number one was already in the activation cycle for the new catalyst, with the diluted hydrogen mixture circulating at low pressure and moderate temperature. The system's sound was different from what it made at full operation, quieter, without the pressurization cycles that at full working conditions made the nearest bench vibrate slightly.
Lebedev wrote and Mitia copied what his supervisor drafted in clean handwriting, asking focused questions when any figure needed verification before going into the final document.
At six in the evening, the report had twelve pages. It included the state of the art for the process with Haber's published data compared against the laboratory's results; the analysis of the three technical problems with the resolution status of each; the economic viability calculations with the assumptions explicitly stated; the proposal for the pilot plant with the Tsaritsyn location and the investment and return figures; and an appendix with the raw data from the last twenty reactor cycles.
On the last page, under the heading RECOMMENDATION, Lebedev wrote:
"Ammonia synthesis from atmospheric nitrogen under high-pressure and high-temperature conditions is technically viable using raw materials available in Russian territory, provided the hydrogen source is natural gas from the Caucasus basin rather than mineral coal with highly variable sulfur content. The current molar conversion yield of sixteen percent is comparable to the published German state of the art from 1909. The principal limitation is catalyst service life, which with clean hydrogen would reach sixty to ninety hours of continuous operation according to Prof. Smirnov's calculations, sufficient for an industrial pilot plant. The estimated investment for a pilot plant of fifty metric tons per day in Tsaritsyn is four hundred thousand rubles. At that production level, the cost per ton of fixed nitrogen is competitive with imports even before considering the supply security premium. It is recommended that planning for the pilot plant be initiated immediately, subject to confirmation of the catalyst life data with methane reforming hydrogen, to be obtained in the test cycle initiated today."
He read the paragraph once. He left it.
Mitia copied the final page in clean handwriting with his small, orderly script.
Lebedev signed the original and the copy. The copy was for the laboratory archive. The original would go to the Ministry of Industry in the official envelope that the Institute's director personally delivered each week.
"Do you think they'll read it?" Mitia asked while sealing the envelope.
"Someone will read it. If the numbers are clear enough, it'll reach the desk of someone with authority to say yes." Lebedev put on his coat. "What we can't control is how long that process takes. What we can control is that when someone at that level asks for more information, the laboratory is ready to provide it."
Mitia nodded.
"Tomorrow we have the surface analysis of the spent catalyst and the sulfur data from the Caucasus methane."
"Tomorrow at nine we review both results. And if the methane gives the profile I'm expecting, we start the first complete cycle with that hydrogen source."
Lebedev turned off the lamp at his bench and left the laboratory. Reactor number one was still in activation, its constant murmur filling the empty room.
Outside, the afternoon had already closed and the streets of Saint Petersburg were lit with their gas lamps, casting yellow circles on the packed snow of the sidewalks.
The ammonia the reactor would produce over the next few cycles was an amount that would fit in four or five laboratory glass bottles. The distance between that and the eighty thousand tons per year Russia needed was a gap that couldn't be closed in a single winter. But it couldn't be closed, either, without first passing through the sixteen percent conversion rate, the catalyst life data, the report to the Ministry, the Tsaritsyn pilot plant that didn't yet exist.
It closed by going. Nothing more.
[Nemryz: If you've enjoyed this story and want to read ahead, I have more chapters available on my patreon.com/Nemryz. Your support helps me continue writing this novel and AU. Thank you for reading!]