Latest / The Manufacturing Podcast with Fexingo: Factories, Supply Chains, and Industrial Business / How a Danish Factory Turns Butter Smells into Perfume
Transcript
- Lucas: You know the smell of a butter factory, Luna? That rich, almost nutty aroma that hits you from half a mile away? Luna: I've never been inside one, but I can imagine. It's warm, fatty, a little sweet. Lucas: Exactly. Well, a dairy cooperative in the town of Videbæk, in western Denmark, has spent the last two years bottling that smell and selling it to the perfume industry. They're capturing the volatile organic compounds — the VOCs — from their butter production and turning them into a fragrance concentrate. Luna: Wait — perfume made from butter factory air? That's either genius or the most niche thing I've heard this month. Lucas: It's both. And if today's episode makes you rethink what 'waste' means in a factory, that's the kind of thing that keeps this show ad-free. We only exist because listeners choose to support us at buy me a coffee dot com slash fexingo. Seriously — a five-dollar cup keeps us independent. Luna: Absolutely. If this conversation moves your work forward in some small way, consider it. Lucas: So back to the butter smell. The factory — it's part of the Arla Foods cooperative — produces about 100,000 tons of butter a year. Their fermentation vats and churns generate roughly 40,000 cubic meters of air per hour, and that air is loaded with diacetyl, acetoin, and other aroma compounds. Luna: Diacetyl — that's the chemical that gives microwave popcorn its buttery flavor, right? Lucas: Correct. It's also linked to a lung disease called bronchiolitis obliterans in popcorn factory workers, but at low concentrations it's harmless and extremely desirable in both food and fragrance. So instead of scrubbing the VOCs out of their exhaust air with carbon filters — which they used to do, at a cost of about $150,000 a year — they now route the air through a system of condensation and adsorption columns. Luna: And that captures the aroma compounds? Lucas: Yes. The system cools the air to about 4 degrees Celsius, which condenses water and some of the heavier volatiles. Then the remaining air passes through a bed of activated carbon, which traps the lighter compounds. Periodically, the carbon bed is heated with steam, and the desorbed aroma concentrate is collected as a liquid. Luna: So you end up with a concentrated butter essence. What does that sell for? Lucas: The cooperative sells the concentrate to Givaudan, the Swiss fragrance and flavor giant, for up to $500 per kilogram. The yield is roughly 20 kilograms of concentrate per day, so we're looking at about $3.6 million in annual revenue at full capacity. But they're still ramping up — right now they're at about $2 million annually. Luna: That's real money. And they used to pay to get rid of the smell. Lucas: Exactly. They flipped a cost center into a revenue stream. The capital expenditure for the capture system was about $1.2 million, so they paid it off in roughly eight months. Now it's pure profit, minus maintenance and energy. Luna: Is the butter essence actually used in perfume, or is it more for flavoring foods? Lucas: Both. Givaudan uses it as a base note in gourmand fragrances — think scents like Thierry Mugler's Angel or Maison Margiela's Coffee Break, which have buttery, creamy undertones. But the bulk goes back into the food industry as natural butter flavor for baked goods, sauces, and snack seasonings. Luna: So the same molecules that once annoyed the neighbors are now ending up in luxury candles and croissants. Lucas: Precisely. And the technology isn't limited to butter. The same headspace extraction method can be applied to any food processing facility with a distinctive aroma. Roasted coffee, cocoa, bread baking, cheese aging — they're all emitting valuable VOCs. Luna: Is anyone else doing this at scale? Lucas: A few. There's a coffee roastery in Hamburg that captures its roast aroma and sells it to a local candle maker. And a cheese cooperative in the French Alps is experimenting with capturing the smell of Comté during aging. But the Videbæk plant is the first industrial-scale operation that's been profitable enough to publish numbers. Luna: I imagine the regulatory environment matters. Are these captured VOCs classified as natural flavorings? Lucas: Yes, under EU regulations, because they're physically extracted from food production without chemical modification, they qualify as natural. That's crucial for the price point. Synthetic butter flavor sells for about $30 per kilogram. Natural butter essence, captured from real butter, can command 15 times that. Luna: So the premium comes from the 'natural' label. Lucas: Exactly. And the fact that it's a byproduct — a waste stream — means the input cost is nearly zero. The only costs are energy, equipment depreciation, and labor to run the system. Luna: What about the neighbors? Did they complain about the smell before? Lucas: They did. The factory is in a small town, and the butter aroma was intense. Arla had a complaint log. Since installing the capture system, the odor level dropped by about 70 percent, and complaints have virtually stopped. So there's a community relations benefit on top of the revenue. Luna: That's a triple win: less pollution, better neighbor relations, and a new product line. Lucas: Right. And it fits into a broader trend called industrial symbiosis — where the waste of one process becomes the input for another. We've seen it with waste heat, with CO2, with food scraps. This is the first time we're seeing it with smell. Luna: Is there a risk of oversupply? If every dairy plant starts doing this, the price of natural butter essence could crash. Lucas: Possibly, but the market for natural flavors is growing at about 8 percent compound annual growth rate, driven by clean-label demand. And the fragrance industry is always looking for new natural ingredients. Givaudan alone has a portfolio of over 3,000 fragrance ingredients. They can absorb more supply. Luna: Still, it's a niche. Not every factory has a smell worth capturing. Lucas: No, but many do. There are about 1,200 industrial dairy processing plants in Europe. If even 10 percent of them adopt this technology, that's 120 new facilities turning air into profit. And then you have breweries, bakeries, chocolate factories. The addressable market is larger than you'd think. Luna: What's the catch? Why isn't every factory doing this? Lucas: The upfront cost is the main barrier. The capture system requires capital that small cooperatives might not have. Also, the technology needs to be tailored to the specific VOC profile of each plant. A butter factory's aroma is different from a brewery's. You need a bespoke engineering solution. Luna: And you need a buyer for the concentrate. Not every region has a Givaudan nearby. Lucas: Exactly. The Danes were lucky that Givaudan has a flavor development center in Copenhagen, just a few hours away. Logistics matter. The concentrate is liquid and requires cold chain shipping. That adds cost. Luna: So it's replicable, but not trivial. Lucas: Correct. But the principle is powerful: any factory with a distinctive, valuable odor is sitting on a potential revenue stream. The question is whether the economics pencil out for their specific volume and VOC concentration. Luna: I'm curious about the technical side. How do you know which compounds are valuable? Did they analyze their own air first? Lucas: Yes. Arla worked with a Danish biotech firm called SensInnovation, which does gas chromatography–mass spectrometry analysis of headspace. They identified 47 different volatile compounds in the butter factory air, of which 12 had commercial value. The top three were diacetyl, acetoin, and delta-decalactone — that last one gives a creamy, peach-like note. Luna: So they had a menu of molecules to target. Then they designed the capture system to preferentially adsorb those. Lucas: Exactly. The carbon bed is chosen for its affinity to those specific compounds. It's not a generic carbon filter; it's engineered for the desired VOC profile. Luna: That's a level of precision I wouldn't have expected in a dairy plant. Lucas: Right? It blurs the line between food manufacturing and fine chemistry. And it raises an interesting question: how many other 'waste' streams in manufacturing are actually underexploited resources? We tend to think of waste as solid or liquid, but gaseous emissions — odors — are often overlooked. Luna: Especially when they're regulated as pollutants. The default is to destroy them, not capture them. Lucas: Yes. And destruction costs money. Capture, if done right, can make money. That's a powerful reframe. Luna: So the next time I smell a bakery or a brewery, I'll wonder what they're venting into the air — and whether someone's bottling it. Lucas: Exactly. Maybe the future of fragrance isn't a field of flowers in Grasse. It's the exhaust pipe of a butter factory in Denmark.