Birectifier Analysis of Haitian Orange Peel

Follow along: IG @birectifier

This was an extremely cool first draft of a project because I’ve always had a fascination with this type of peel, however I had never encountered them before because they are typically unavailable to small scale purchasers. These are decrepit looking dried orange peels that almost look like a dehydrated lime, but they smell so wonderful right in the bag. There is a lot of promise in these peels, because besides contributing to the aroma of a gin, they could also be used for orange liqueur production which is a big staple of any cocktail bar. They are a little tricky to work with because of all the barely soluble terpenes, but better first hand knowledge of the processes at different levels of sophistication could help make beautiful products as well as right some wrongs.

These peels get sent all over the world as a commodity where wealthier nations take very little risk when processing them, use century old methods, completely in the public domain, and then claim incredible markups in markets that are insufficiently competitive. The vast majority of the profits from these peels should be made by the farmer who endures significant risk and not the rectifier who does very little that is novel. No small nation that produces high quality peels should ever import an orange liqueur. It should all be produced domestically. There is very little that is special about the leading brands.

The goal becomes to gain insights into working with these agricultural products so we can bust up the colonial cash cows and create a more competitive market for the peels that strengthens the position of the farmer. We can gain insight in this post, but sadly we cannot resolve all the best practices for these peels in one experiment.

I have previously covered birectifier analysis of a triple-sec as well as highlighted the production advise of flavor chemist Joseph Merory which concerns terpene removal. I have also explored the perceptual puzzle of terpene removal which hints at a contrast enhancement phenomenon.

My protocol here was to redistill 40 grams of peels in 250 ml of neutral 40% ABV spirit (100 ml of absolute alcohol). Ten more mL of water was used for rinsing the volumetric flask. I only broke up the peels enough to get them into the boiling flask.

One place I think I went wrong was only using 100 ml of absolute alcohol because I suspect the sponginess of the dehydrated peels somehow held onto ethanol altering the typical birectifier ethanol curve. As ethanol was held onto slightly by the peels. The ethanol concentration in the vapor was prematurely depressed pulling steam volatile congeners forward into fraction 4. Typically the ethanol concentration stays high then dives off a cliff as fraction 5 begins.

This could be solved by either inputting more absolute alcohol like 110 ml or possibly just changing to fraction 5 by temperature instead of volume to isolate aroma from the essential oil and separate it from generic terpene aroma that is apparently still abundant in fraction 4.

My suspicion is that these peels have an excessive abundance of undesirable insoluble generic terpenes because there is so much pith. If they were processed with a peeler as opposed to harvesting the entire skin, pith and all, there may be less terpenes and they would be easier to use.

My next suspicion is that this may not be a very good sample in terms of quality and condition. My understanding is the sample was over a year old. I expected more essential oil aroma but I also do not have much frame of reference or another sample to compare to. If they degrade in quality, that may be all the more reason to process them as close to home as possible.

Previous times where I have observed an abundance of insoluble terpenes in the first and second fraction, they have eventually settled upon dilution and after 24 hours could be decanted. This did not happen in this case and they appeared to settle somewhat but never to the point they could be decanted.


Next up was supposed to be a trial of Clevenger analysis, but the Clevenger actually broke. It featured a glass 3-way stopcock that cracked in half. I’m having it repaired, but the lesson is to never buy a device that doesn’t have a modern teflon stopcock.

I tried to machine a replacement, but I think they warped the bore by trying to convert a 2-way housing to a 3-way. Import glassware will always waste your time and break your heart.

We will have to return to this and I will have to acquire more samples.


Fraction 1: Extremely turbid with no sign of terpenes settling after 24 hours. Somehow it smells kind of muted and not as zesti as fraction 2.

Fraction 2: Very turbid with no sign of terpenes settling after 24 hours. Generically zesti aroma, but not as acute as others I’ve experienced before.

Fraction 3: Visually flocks of insoluble terpenes that start to settle out.

Fraction 4: Beautiful orange-y aroma that likely belonged in fraction 5 but ended up at the end of 4 due to how the remainder of the ethanol distributed itself. Visually there are flocks of insoluble terpenes that have settled to the bottom. I have never seen these flocks persist all the way to a fraction 4. No droplets of essential oil, possibly because of the ethanol level. Only slightly acrid on the palate which implies not much concentration.

Fraction 5: Not so much orangey, but a type of tightly wound aroma that reminds me of certain estery fraction 5’s from rum. I would not call this concentrated at all. Perfectly clear upon dilution. However, the undiluted fraction appears to be faintly turbid relative to the rest.

Fraction 6: Not as much aroma as I imagined. Fairly neutral. Perfectly clear.


Products like lemoncello I consider living liqueurs because their fragile unseparated terpenes define their character becoming something very comparable to the fruit directly off the tree. These special aromas are ephemeral and actually break down over the short term giving the product a shelf life which is partly why lemoncello is often stored in the freezer.

Triple-secs and other products like Rose’s lime cordial are the converse and are defined by removal of the most volatile terpenes. They are often distilled, diluted, decanted or centrifuged, and then finally re-distilled. A sort of neon glowing fruit character exists underneath the terpenes. The most volatile terpenes essentially cast olfactory shadows which is a feature for lemoncello with beautiful proportions or a hindrance with other types of peels where a different effect is desired.

A spectrum of abstraction exists where the expression is most ordinary or extraordinary. Lemon without it’s terpenes can feel artificial and non-culinary, like Pledge brand furniture cleaner, while limes may also feel plebian, cheap, and artificial like Rose’s cordial. Oranges may have more range and we could see a living orange-liqueur like Creole Shrubb or a modernist masterpiece like Combier triple-sec. Can you imagine the novelty of a triple-sec before the advent of electric light or ubiquity of artificial flavors?

I think I would pay top dollar for a Haitian made triple-sec with a Clairin base. I wonder if they would go full terpene removal or let it live a little?

Long ago, probably on eGullet, I proposed a “toy” class of spirits & liqueurs. These would be products produced at the highest level on an extremely small scale. At the time, the idea was mainly aimed at home distillers and the new category of bar program distilling that was now legal outside of the U.S. The point was to explore the unique potential (as well as technical limitations) of extremely small scale as well as the unique opportunities for storytelling that would go with it. Back then, I was trying to get oranges from the Lyman estates greenhouses which had among the oldest citrus in the country. I would likely be making an orange liqueur from less than a bushel of fruit, but the storytelling would be incredible.

Flash forward ten plus years and the toy class can mean something a little different. It could become a product produced at less than 1000 liters by a major distillery with international distribution, but never intended to leave the estate which now hosts significant agro-tourism. Toy products may never scale up, but they may support brand storytelling on social media and be exclusively featured at a new generation of festivals to generate enthusiasm. Development and production sharpens the skills of the distillery team and supports important farmers in the community that are vital reservoirs of local know how.

One of my favorite toy liqueurs is the Berry Hill pimento dram by Wray & Nephews. It barely existed outside of the airport duty free shop in Jamaica, but has always been thought by cocktail enthusiasts to be the best version of a pimento dram produced. Trade lore used to say that it was produced for years by a lone woman and it was her family recipe. How do you get more honest and authentic than that? Next step would be telling more of the story and weaving it into social media and agro-tourism. Berry Hill provides us a template that could be used for fine orange liqueur.

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.

search previous next tag category expand menu location phone mail time cart zoom edit close