If Fragrance Oil Migrates During Heat Exposure and Is Reabsorbed After Cooling, Can the Candle’s Burn Performance Still Change?

Where migrated oil goes when the candle cools

When a candle warms, the wax softens from the outside in, and oil that is loosely held moves along the easiest path: upward to the surface and outward to the glass. When the candle cools, the wax firms again and the visible oil disappears from view. Some of it soaks back into the top layer of wax, some stays as a film too thin to see, some collects in the narrow gap where wax meets glass, and a little evaporates.

What is unlikely is a perfect return to the even spread the candle had when it first set. After a heat cycle, the top few millimetres and the outer edge may hold more oil than before, and the core may hold slightly less. Nothing on the surface tells you how much moved. That is why a candle that looks recovered can still burn differently.

This is a general mechanism, not a measurement of your candle. Some formulas move very little oil in a cycle and burn exactly as before. Others shift enough to notice. The point of this page is to give you a way to find out which kind you have.

What can change in the burn

The first burn melts the top layer first, which is where migrated oil tends to gather. If that layer holds more oil than the rest of the candle, the first melt pool may be oilier than the candle was designed for. Makers who see a difference typically describe some of the following; your candle may show none of them.

None of these is certain, and none should be assumed from looks alone. They are the things to watch for in the comparison below.

The twin-candle burn test

The fair test compares candles that are identical except for their heat history. Pour them in one batch, split them, heat only half and burn them in pairs.

Burn the pairs in the same room, at the same time of day, with wicks trimmed to the same length before each lighting. A pair burned on different evenings, or one near a fan and one in a corner, compares rooms rather than heat histories. Keep a fixed distance between the jars so neither flame disturbs the other.

Use plain descriptions and record both candles in the same words. If all four cycled candles match their controls, the heat history is not changing your burn. If even two of four burn worse in the same way, treat it as a real effect.

Why the first burn matters most

The first time a buyer lights a candle sets their view of it. If a heat-cycled candle smokes, flares or throws strongly for an hour and then fades, the buyer does not know that its twin on your shelf would have behaved better. They judge the product in front of them, and that product has a heat history you cannot see.

This is also when safety margins are thinnest. A larger flame early on is still a flame in a jar on someone's table. Your wick was chosen for the candle you poured, not for a candle whose oil has been moved around by a hot van. If the twin test shows a larger first-burn flame, take that seriously even if it settles later.

So when you run the twin test, give the first burn the closest attention. Watch the first fifteen minutes of each pair without distraction, note the flame size against a fixed mark behind the jars, and photograph both flames together. Later burns matter too, but the first one is where a heat history is most likely to show.

If you cannot run the full test

If eight candles is too many for a small range, run two pairs instead of four. Two pairs give you less confidence but still show a large effect. What you must not do is burn one cycled candle on its own and decide from memory how the uncycled ones behaved; without a twin beside it, small differences in room, draught and wick trim will mislead you.

What the test costs and what you decide

A worked example uses Roasted Coffee at 8% in CSI 464, in the matte 200 ml jar with a Thin C1 Eco wick. One candle is 184 × ₹0.5074 (₹93.36) wax, 16 × ₹3.738 (₹59.81) oil from the 1 kg pack, ₹70.80 jar and ₹5.90 wick: ₹93.36 + ₹59.81 + ₹70.80 + ₹5.90 = ₹229.87 per candle in materials. Eight candles cost 8 × ₹229.87 = ₹1,838.96, excluding packaging, labour and GST. The coffee oil's own page lists 6–10% for soy and asks for 48–72 hours of cure, so 8% sits inside what it recommends.

Set that against what the answer protects. If heat-cycled candles burn with more soot or a weaker throw, every parcel that meets that heat is delivering a worse candle than the one you tested at your bench.

If the cycled candles burned worse, do not sell the remaining heat-exposed stock through parcels while you reformulate. Stock that never left your cool room is still the candle you tested at the bench; stock that sat in a hot van or a warm stockroom may now be the cycled version.

Whatever you find, keep the scorecards. A twin test on today's formula is the reference you will compare every future change against.

Where this page fits

CSI already has guides that cover parts of this question. This page answers one sweating decision; these cover the neighbouring questions:

Three things to buy for this fix

Frequently asked questions