An Alchemist Tries to Save the Soul of Rum From His Boston Kitchen

Stephen Shellenberger lives in a narrow apartment on a quiet one-way street in Cambridge, Massachusetts. One gets the feeling that the 35-year-old does not cook much. His kitchen table and counter are overrun by dozens of liquor samples in small, brown lab bottles, along with an array of stainless steel and glass equipment, some of which looks as if it escaped a high school laboratory circa 1958, and all of which appears inscrutable.

Welcome to Shellenberger’s time machine, the place where he’s exploring the depths of distilling past. There is a device that measures surface tension, a collection of round bottomed flasks, and tall glass apparatus clamped to stainless rods. Here, between the sink and the stove, is his pride and joy—a birectifier, which consists of glass tubes and coils and heating elements. It’s a device he rescued from obscurity and is trying to usher back into circulation.

His goal: to improve the distilling world—especially to give craft distillers an affordable path toward excellence by drawing on the forgotten work of others. And he makes a pretty good argument that the future lies in the past.

Shellenberger is an independent researcher, and supports himself as a restaurant manager and running a side business recasting antique doorknobs and the like. When earning a livelihood isn’t distracting him, he’s rooting through yellowing documents in search of lost secrets. Once found, he transcribes, scans, translates (when necessary), and summarizes his finding on his blog, Boston Apothecary.  

Shellenberger is essentially a latter-day alchemist.

Alchemists of yore believed that answers to life’s greatest mysteries could be found in the past. “The alchemist believed that the ‘ancients’ knew the secrets,” wrote F. Sherwood Taylor in his 1974 history of alchemy, “and his principal endeavor was to understand the meaning of their books. Modern science, on the other hand, looks forward to the time when her efforts will make known the things that have never been known.”

Shellenberger’s research didn’t take him quite to ancient Egypt, but to a lost golden age of spirits research, dating roughly from the 1930s to the 1970s. Many liquor industry scientists at the time devoted themselves to basic research, understanding yeast and distillation technology and things like the role of osmotic pressure during fermentation.

Shellenberger is especially fascinated by the rum research undertaken in Puerto Rico.

In particular, he’s curious about the work of Rafael Arroyo, a Louisiana State University graduate and long-time head of the chemistry department of the Agricultural Experiment Station at Rio Piedras in Puerto Rico. Arroyo spent much of his career studying sugar cane and the rum that came from it—it’s likely he knew more about the chemistry of fermentation and distillation than just about anyone before or since. He held patents for a fermentation process for producing butyric acid—a compound that’s essential for giving bold rums their overripe pineapple aroma—and was the author of the book Studies in Rum, published in 1945 by the ‪University of Puerto Rico.

Arroyo died in 1949 at the age of 57, and his studies began a long slide into obscurity. His research papers moldered in filing drawers and libraries; even his seminal book was available chiefly in rare book reading rooms, where it couldn’t be checked out.  

Shellenberger assembled a bibliography of Arroyo’s writing, and then began gathering his papers wherever he could find them—searching online, or requesting copies through a helpful interlibrary loan program at his local library. For others, he searched for long out-of-print journals on used book and auction websites. Then he scanned and uploaded PDFs of the articles to his blog, aiming to start a conversation with curious others. His site has links to several dozen of these documents (“The Arroyo fermentation process for alcohol and light rum from molasses,” from a 1949 Sugar Journal; Arroyo’s 1942 patent for “Ethanol Fermentation of Black Strap Molasses”). He also translated from the Spanish Arroyo’s 1938 circular entitled simply “Rum Manufacture” and uploaded that to his site. (“It is possibly the single greatest short read on rum production any new distiller can do,” Shellenberger writes.) He also convinced someone at Harvard to scan Arroyo’s book, and he posted a PDF of that for all to see as well.

The golden age of rum research entered its twilight in the 1970s. The production environment was changing. The best minds in the industry, which once would have been drawn to basic research, were deployed instead to deal with treating production waste as society turned to environmental concerns. The scientists also were assigned to address issues raised by health inspectors, who wanted replace old wooden tanks with stainless steel ones, and sweep out all the spiders. Alchemy was elbowed aside by society and science.

In reading through Arroyo’s work, Shellenberger came across a passing reference to a thing called a birectifier. “He described it in the book, but you’d skip over it in a heartbeat,” he says. “I didn’t know how significant it was.”

But he started digging, and eventually found a German journal that featured an engineer’s rendering; it was used for the analysis of dessert wine.

The birectifier was first developed by Dr. Curt Luckow of the Berlin Institute of Fermentology. A 1939 report of the Agricultural Research Station called it “an invaluable apparatus for the evaluation of commercial and aging rums.” As Arroyo noted in a report, “chemical analysis alone, as usually practiced, is almost entirely ineffective in the appraisal of the rum aroma, nor does it give an indication as to which particular constituents of the non-alcohol number of rums are mainly responsible for the nature of their flavors.” The birectifier filled that void.

At heart, the birecrifier allows a distiller to dissect a product, dividing it into eight fractions for analysis, largely by aroma. Each fraction offers clues about the process, and how best to tweak it—switching up the yeast strain, for instance. Arroyo revered what he called “rum oil,” a mixture of essential oils that gave can a rich rum considerable depth, and which he found often appeared in the fifth fraction. If a rum yielded a “naked” fifth fraction, with virtually no aroma, then the producer should head back to the drawing board.

The more Shellenberger read about the birectifier, the more intrigued he became. He eventually shifted from archival researcher to kitchen laboratory scientist. With descriptions and sketches of the birectifier in hand, he set out to rebuild this vanished bit of laboratory equipment. “I called 30 glass manufacturers,” he says. “Most didn’t reply.” Eventually he found an amenable glass maker, and started assembling other parts of the device, such as the heating elements.

He then began running all manner of spirits through it, and posting regular reports on what he was uncovering: “Birectifier Analysis of a Demerara Rum” was one. “Birectifier Analysis of a Role Model Vermouth” was another.

Shellenberger also got into the business of offering the birectifier to craft distillers. “I get a lot of emails from new distillers who are frustrated that they have no laboratory and no money to pay for consultants,” he says. “This is the most affordable analysis thing you can do.” And unlike sending samples out to a pricey lab for gas chromatography-mass spectrometry analysis—another way to dissect a product—he insists his lower-cost analysis actually reveals more about quality, since the high-tech approach overlooks the “rum oils.”

The cost of a birectifier is $2,100. So far, he’s sold four. Shellenberger suspects much of the traffic of people accessing his posts and papers on his website is not from distillers but from consumers seeking to educate themselves and broaden their understanding of what they’re drinking.

“I probably have ten times the readership with niche consumers as I do with distillers,” he admits. He suspects that educated drinkers revel in complexity, whereas distillers are more interested in “trying to weirdly dumb things down” into bullet points. “They want to reduce it all to a series of grunts,” he says.

And Shellenberger insists that what he turns up is accessible and not terribly technical. (Although sentences such as this, from a blog post earlier this year, may suggest otherwise: “I am also recreating Seagram’s botanical assay lab for essential oil yield calculation to scale botanical charges”).

“If people talk about the difference between an ale and lager, why can’t they talk about two different yeasts?” he says. “This is rum’s slow food moment. This is a case of the vinyl DJ versus the MacBook.”