West Point Chemists Re-Create Medieval Gunpowder Recipes

Making gunpowder is a bit like cooking, except more explosive. Gunpowder makers in the 14th and 15th centuries used black powder brought to Europe from China, then mixed its three ingredients together one by one: saltpeter (also known as potassium nitrate), charcoal, and sulfur. But they also made some chef-like improvisations, including a splash of brandy, vinegar, or varnish.

Now a group of experts at the US Army Military Academy at West Point has re-created these medieval recipes and tested the craft gunpowder in a replica cannon. They found that early gunpowder took a lot of experimentation to get right—and that gives them insights into how modern-day bombmakers might be using similar trial-and-error methods to assemble explosive devices.

The project started when West Point history professor Cliff Rogers was looking through the Feuerwerkbuch (German for “firework book”), a collected set of anonymous manuscripts. Rogers says the Feuerwerkbuch is a practical handbook for master gunners, discussing how to process the ingredients for gunpowder, how to make it, and how to load and fire a cannon. The manuscripts were assembled over a number of decades when the technology of gunpowder and artillery was changing rapidly; the book included recipes from the year 1336 to its publication in 1420 and used descriptive terms like “common,” “better,” and “still better” to describe the combustive properties of each mixture.

Rogers asked his colleague Dawn Riegner, a professor of chemistry, to fact-check one recipe that included an unusual ratio of sulfur, saltpeter, and charcoal. “The main goal was to check on the interpretation of one particular recipe that just seemed wrong,” says Riegner, who was lead author on the team’s paper, published this month in the journal ACS Omega. The issue turned out to be a translation error, not a scientific one, but that had piqued their interest. “Then it became: Well, what about all these other ingredients that the medieval gunners were putting in, and what was the thought process?” Riegner says. “Did these folks who have no chemistry degrees know what they were doing? Did they have a hypothesis about what these new ingredients would do for them, or how mixing them would help them?”

Riegner and Rogers decided to re-create these early recipes and find out whether they would still work. Riegner worked in her chemistry lab with her daughter, an engineering undergraduate at Stevens Institute of Technology, who was home during the Covid-19 pandemic last year. “We started mixing the ingredients in the lab, starting together the dry mixes,” she recalls. “And then, when needed, when expressed in the recipe, we would add different wet solutions as well, whether it was water or varnish or vinegar.”

Once they came up with a final product, the mother-daughter team put the material in a chamber containing pure oxygen to test the “bomb calorimetry” of the gunpowder, which is a measure of the amount of heat energy produced by its ignition. Tessy Ritchie, chief chemist at West Point, analyzed chemical residues from the lab and field tests.

Riegner says this part of the project ran into some obstacles. The ingredients used in the lab were of scientific quality, meaning they were extremely pure. But the sulfur and potassium nitrate used in the 14th and 15th centuries would have been more contaminated. That might have been one reason that gunpowder cooks added extra ingredients—the team found that, over time, the recipes started using larger amounts of sulfur in order to replace the more expensive saltpeter, which was difficult to obtain. The sulfur needed to be purified, hence the use of the other additives, Riegner says.

They might also have been used to turn the dry ingredients into a wet paste that was later dried and purified into gunpowder. And there’s a third theory: The researchers believe the alcohol in the brandy also might have supplemented the organic compounds in the early gunners’ charcoal and improved its burning. But the modern-day experiment wasn’t able to accurately determine the effects of these additives, because the researchers were starting with higher-quality ingredients. “None of them really improved the energetics,” Riegner says.

Lt. Robert Seals and Major Edwin Den Harder set up a replica of an early-15th-century cannon at a West Point gunnery range.

Courtesy of Dawn Riegner

Next, they wanted to check how well the recipes worked in the field. Cadet Robert Seals, an undergraduate student at West Point, worked with Rogers to secure a small grant to build a replica of a German short-barreled Steinbüchse stone-throwing gun, originally built around the year 1400, which they had made at a North Carolina foundry. The cannon has a 2-foot-long cylinder with a wider opening at the front chamber where the ball is placed. The powder is placed in the rear chamber and ignited with a fuse, and the entire cannon is supported by a wooden frame. For the field trials, a mason provided small stone balls as projectiles, like those that were fired into castles and walled cities during medieval sieges.

The early medieval gunners who developed this kind of weapon learned over time that the cannonball is thrown by gas pressure, not flame, and that charcoal from a willow tree prepared in a closed container is far superior to oak charcoal burned and fabricated in a traditional pit. The medieval gunpowder recipes were generally lower in saltpeter and higher in sulfur than modern ones.

Rogers says these cannons were a big advancement over existing weaponry during the late Middle Ages. “If you are up on a tower and up against a bunch of knights with really strong armor that might be able to withstand a crossbow bolt or an arrow,” Rogers says, “they won’t withstand a four-inch stone ball with 1,500 joules of energy.”

The team hauled their replica cannon to a military firing range at West Point and tested several mixtures to see which one worked best. Because the field was laced with buried bullets and unexploded ordnance, the researchers weren’t allowed to walk out and measure how far each ball traveled. The team made five shots, and were able to make rough estimates of the distances using video imagery, but ultimately didn’t have enough information to know if one mixture worked better than any other. Riegner says she hopes to do further research into this question.

Kathleen Riegner and Lt. Robert Seals collect gunpowder residue from the cannon for chemical analysis.

Courtesy of Dawn Riegner

Still, they did learn something important: Over time, their research shows, the newer recipes evolved to blast the stone ball at a cooler ignition temperature. Throughout the centuries, gunners had been grappling with how to get rid of the heat produced by igniting gunpowder, because they faced the risk of an overheated cannon exploding, or a flame of hot gases escaping from the back of the weapon. Riegner measured the temperature of the cannon and found the two earliest recipes produced the most heat, although the differences were only a few degrees Celsius.

“The most interesting aspect was that these recipes work,” says Dan Spencer, a medieval military historian and author based in the United Kingdom, who was not involved in the new study. And that, he feels, could only be resolved by field testing. “It's difficult to know from reading a text whether that would actually work or not,” he says.

West Point researchers fired five stone cannonballs during the test of various medieval gunpowder recipes.

Courtesy of Dawn Riegner

Spencer says that the 14th and 15th centuries were hothouses for gunpowder experimentation. That was also due in part to the 100 Years’ War, which pitted France against England between 1337 and 1453 and spurred the development of armaments. During the earlier part of the 14th century, gunpowder was the domain of alchemists who sourced their saltpeter and sulfur from traders arriving from China and India. But later, European gunpowder became widely accessible, and the recipes changed as weapons evolved. Cannon makers lengthened the barrels to give them a longer and more accurate range, while handheld guns were developed at the same time. These weapons needed different mixtures of gunpowder that didn’t create as much heat, but would still launch a ball through enemy lines.

Rogers says the experiment was eye-opening for someone who has spent much of his career studying medieval warfare. Over time, the gunpowder formulas became more efficient at firing projectiles and less dangerous to the gunners, Rogers wrote in an email to WIRED. “From the field tests, we learned that medieval loading methods and powder formulations could indeed work well in practice,” he wrote. “We were able to throw a four-inch stone ball at a higher velocity than a crossbow bolt, meaning the projectile, with its much greater mass, would have been very lethal even against a man armored with high-quality plate armor that would typically not be penetrated by any muscle-powered bolt or arrow.”

Riegner’s expertise is in forensic chemistry and developing techniques to remotely detect improvised explosive devices (IEDs), like the kind that were a threat to US troops during conflicts in Iraq and Afghanistan. The exercise in re-creating medieval recipes helped Riegner become a better problem solver when it comes to understanding how current explosives are made, she says.

“Today, people are making bombs in their kitchens,” Riegner says. “They are using totally different materials, but I think we can apply some of the techniques that we did in our lab for these gunpowders to these new materials, and to understand the thought process: Why did they add this? And why are they tweaking that?”

Update 10-6-2021 12:09 pm: This story was updated to correct the spelling of Tessy Ritchie's name.