Does Fragrance Oil Burn Off While a Candle Burns?

“It burns off” is three different claims wearing one sentence

Makers say “the fragrance burns off” to mean at least three unrelated things, and each has a different fix. The first is the pot: a worry that heating wax to 85°C destroys the oil before it ever reaches a jar. The second is the flame: a belief that the burning wick incinerates fragrance that should have gone into the room. The third is the fade: the real, common observation that a candle reads weaker in its last third than in its first.

Only the third is a genuine phenomenon worth managing, and it is mostly not about the flame at all. The first is largely a misreading of what temperature does to a blended fragrance oil. The second is true but harmless, because the flame's share is set by how fast the candle burns, not by some separate appetite for fragrance.

Getting this wrong is expensive in a specific way. If you believe the flame is eating your oil, the intuitive response is to put more in. Moving a 200 g candle from 8% to 10% with Roasted Coffee adds ₹14.95 per candle — ₹747.60 across a batch of fifty — and changes nothing about the proportion the flame takes, because the flame takes a proportion. The wrong model does not just fail to fix the problem; it charges you a monthly fee for not fixing it.

The right model is a mass balance. Fragrance oil is a liquid dissolved in wax. It cannot disappear. It can only be somewhere — in the jar, in the room, or in the products of combustion — and once you can name the four routes and put a rate on them, most of the mythology falls away.

The four routes fragrance takes out of a jar

Two things follow immediately. First, route 3 is proportional. The wick cannot distinguish fragrance from wax; it draws the mixture. If the jar is 7.41% oil by mass, so is what goes up the wick, near enough. Doubling the load doubles what reaches the room and doubles what the flame takes, in the same ratio. There is no load at which you outrun the flame.

Second, route 2 competes with route 3, and the competition is decided by the pool, not by the oil. A wide, warm, shallow pool gives evaporation a lot of surface and a lot of time before the wax reaches the wick. A narrow, deep pool with an oversized wick pushes more of the mixture up the wick per hour. That is the real mechanism behind the observation that an over-wicked candle can smell burnt and thin rather than strong: more is being combusted and less evaporated, and the jar is emptying faster while it happens. The live guide on wick size and hot throw covers the sizing side.

What nobody publishes — not CSI, not any supplier honest about it — is the split between routes 2 and 3 as a percentage. It depends on wick, jar diameter, pool depth, room temperature and how long each session runs. Any blog that gives you a clean number for it has invented it. What you can do instead is measure the total, which is the next section.

What one hour of burning actually takes out of the candle

A single-wick soy container candle in a jar of around 6.5 to 7 cm typically consumes somewhere near 4 to 7 grams of wax an hour once the pool is established — that is general candle-making practice, not a CSI specification, and your own jar may sit outside it. Take 5 g an hour as an illustrative working figure for the arithmetic below, then replace it with your own measured number using the protocol further down.

The poured mix is 200 g of wax plus 16 g of oil, so oil is 16 ÷ 216 = 7.41% of everything in the jar. If mass leaves at 5 g an hour and fragrance leaves with it, oil leaves at 0.37 g an hour and 1.48 g over a four-hour session. The whole 16 g is spent after roughly 43 hours of burning.

That table is the honest commercial reply to “does it burn off”. Yes, at somewhere between ₹1.14 and ₹2.83 an hour, depending on which oil you chose — and that is not waste, it is the product being consumed as intended. A customer burning a Roasted Coffee jar for four hours on a Sunday evening has used ₹5.54 of fragrance oil. The candle cost ₹167.19 in materials to make.

Notice which line moves the total. The wax is ₹101.48 in every row. The oil ranges from ₹49.09 to ₹122.34 for the same 16 g. Across a fifty-jar batch the Roasted Coffee version is ₹8,359.40 of raw material. If you are worried about fragrance being consumed, the lever with real money on it is which page you buy from, not how long your customer burns the candle.

What changes between burn one and burn twenty

That ladder assumes fragrance leaves in exact proportion to wax, which is the simplest honest model and close enough for planning. In practice the more volatile molecules leave a little ahead of the rest, so the wax remaining late in the candle is slightly leaner in top notes and relatively richer in heavy base notes. That is a real effect, it is why a vanilla or oud candle often reads deeper and flatter near the bottom, and it is also why the live explainer on why fragrance smells different after burning is about character rather than strength.

The five things that actually change as a candle burns down

The pool gets hotter. Less wax below the flame means less thermal mass. In the last third, the jar and the pool run warmer for the same wick, which tends to increase evaporation, not decrease it. Many candles read at their strongest around the middle.

The flame sits lower in the jar. Glass walls shield the pool from room air, convection over the surface slows, and scented air has to climb out of a deeper well. That reduces what reaches you even while evaporation at the surface holds up.

Hang-up narrows the pool. Wax left clinging to the walls from earlier sessions shrinks the usable diameter. A candle that made a 6.5 cm pool at session one may only make 5.5 cm at session eight — 33.18 cm² down to 23.76 cm², a 28% cut in the surface that does all the work. This is the single biggest cause of genuine late-life weakness, and it is a wick and first-burn problem, covered in the live guide on tunnelling with the correct wick.

The wick changes shape. Carbon builds on the tip, the flame shrinks or flickers, and an untrimmed wick can both soot and shorten the pool. A trim to 5–6 mm before each session is general practice and costs nothing.

Your nose adapts. By session eight the household has smelled this scent for thirty hours. Adaptation is real and it is not a formulation problem. Ask someone who has just come in from outside to score it before you change anything.

Weigh, burn, weigh: put your own number on it in one evening

The whole question — how fast does fragrance leave my candle — reduces to one measurement anybody can take with a kitchen scale. You are measuring total mass loss, then converting it to oil using the proportion you poured. Do it three times across the candle's life and you will know more about your own product than any blog can tell you.

Two readings that mean something once you have the numbers. If your candle loses far more than about 7 g an hour, the wick is probably too large for the jar: you are buying a shorter candle and sending more of the mixture up the wick. If it loses under about 3 g an hour with a pool that never reaches the wall, the wick is too small, and the fix is the wick ladder, not the fragrance.

The two places you genuinely lose fragrance — and neither is the flame

Loss one: the uncapped jar

This is route 1, and it is the only route where a gram lost is a gram wasted. A cured candle has a solid surface, but volatile top notes continue to leave it, and an open jar on a shelf in a Chennai showroom in April loses them much faster than a capped one in a cupboard. Nobody publishes a figure for how much, so treat the arithmetic as illustrative: if a jar shed even 15% of its top-note load before sale, that is ₹8.97 of Roasted Coffee per 200 g candle thrown away for want of a lid.

The fixes are all cheap and all boring. Cap jars as soon as the wax sets. Keep cured stock in a closed carton, not on an open rack. Keep it out of direct sun and off a west-facing wall in summer. Rotate stock so that the oldest jars sell first. And if you sell lidless vessels, shrink-wrap or box them. The live guide to fading fragrance and curing mistakes goes further into the storage side.

Loss two: the long, hot, open melt

Adding fragrance at the right temperature does not destroy it. The Luxury Soy Wax Chunks page asks for fragrance at 80–90°C with a slow two-minute stir, and the Eco Soy CSI 400 page asks for 85°C within an 80–90°C window. Those temperatures are well below the flashpoints CSI's oil pages publish — Sandalwood 203°C, Vetiver 206°C, Eucalyptus Spearmint 199°C, British Rose 192°C. Flashpoint is a transport, storage and handling property in any case, not a ranking of throw and not the point at which scent disappears.

What does cost you is time at temperature with the lid off. A fragranced melt held at 85°C in a wide open pan for twenty minutes while you fill forty jars is evaporating from a large hot surface the entire time, and the last jar is not the same as the first. Two of CSI's pages publish notably low flashpoints — Palo Santo at 93°C and Gardenia Tuberose at 98°C, whose page says to add below that figure — and for those, add at the cool end of the window and pour quickly. For everything else: add, stir two minutes, pour within a few minutes, and cover the pot between pours. The live guide on what temperature to add fragrance covers the rest.

What this means for how much oil you buy

If fragrance left candles at some mysterious accelerated rate, the answer would be to over-dose every jar as insurance. It does not, so the answer is the opposite: buy the load your two ceilings allow, buy it in the pack size your volume justifies, and spend the saved money on the things that actually govern route 2 — a correctly sized wick, a jar wide enough for the room, and the patience to cure.

Order the levers by how much they change what the customer smells per rupee. Pool area first: a wick one size up, or a wider jar, changes the evaporating surface directly and costs ₹3.54 more per candle between C1 and C2. Cure second: free, and the Luxury Soy Wax Chunks page itself asks for at least 48 hours and ideally one to two weeks, with 14 to 21 days the standard soy practice by family. Load third, up to the lower of the wax ceiling and the oil page's stated range. A different oil last, and only after a controlled comparison.

On pack size, the arithmetic in this page gives you a clean planning rule. One kilogram of fragrance oil at 8% dresses 62 candles of 200 g, or 83 candles of 150 g. If you sell forty jars a year, a 1 kg pack is more than two years of stock, and route 1 — evaporation from a half-used bottle opened and closed thirty times — becomes your real enemy. Keep bottles tightly closed, in a cool dark cupboard, and buy the 100 gm or 500 gm rung until the volume justifies the kilo.

And for a festive run, plan backwards from the burn, not forwards from the pour. Diwali 2026 falls on 8 November 2026. A 14 to 21 day cure plus courier time across India means pouring by about mid-October, capping immediately, and storing in closed cartons — which is the same advice as route 1, arriving from a different direction.

Frequently asked questions