Why Do Essential Oils Fade in Candles?

Three complaints hide inside the words "it faded"

When a maker says an essential oil faded in a candle, they mean one of three quite different things, and the three are almost never the same problem. The first is bottle fade: the oil smelt sharp and alive when it arrived and reads flat two months later, before it ever met wax. The second is cure and shelf fade: the candle smelt lovely unlit on day two and reads faint unlit on day sixty. The third is burn fade: the candle smells right for the first twenty minutes and then seems to disappear while it is still burning, or throws nothing at all from a pool that looks perfect.

These are three distinct losses, and only the third is really about the candle. Bottle fade is a storage question. Cure and shelf fade is largely about how much of the lighter material was still in the wax by the time the jar reached the customer — the same process that makes cold throw settle over a fortnight. Burn fade is about what is left in the molten pool after the light material has gone, which is a question about the composition of the scent itself.

Essential oils are unusually exposed at all three points, and the reason is not mystical. An essential oil is a distilled or expressed plant material. Whatever is in it is what the plant put there. A candle fragrance oil is a formulated blend, built deliberately from materials of different weights so that something is still evaporating at hour three. Nobody formulates a lemon; you get what the peel contained, and a great deal of what a lemon peel contains is very light.

Why a distilled plant oil gives up its scent faster than a blended fragrance oil

A scent reaches a nose only when molecules leave the liquid and travel through the air. Light, small molecules leave easily; heavy, large ones leave grudgingly. That is the whole mechanism, and it explains both why a candle smells at all and why it stops smelling.

A perfumer building a candle fragrance oil chooses materials across that whole weight range on purpose. The light ones give you the first impression — the moment the customer opens the lid, and the first ten minutes of a burn. The middle ones carry the body of the scent through the first hour. The heavy ones — woods, resins, musks, vanillin-bearing materials — barely evaporate at all on their own, but they keep releasing slowly for hours and they hold lighter materials back by simply being in the way. A blend is engineered to have all three layers still doing something at hour three.

A single-plant essential oil has whatever weight distribution that plant happens to have. Citrus peel oils are dominated by very light terpenes; that is exactly why a fresh lemon smells so immediate and so brief. Mint and eucalyptus are similarly front-loaded. Lavender sits in the middle. Woods, roots and resins — vetiver, patchouli, cedar, frankincense — are heavy and genuinely persistent, which is why they are the essential oils that behave best in wax and also, as it happens, among the most expensive per gram.

This is also why the honest comparison is not "natural versus synthetic". It is "unweighted versus weighted". You can see the same logic on CSI's own pages: the Vetiver Fragrance Oil page says vetiver builds slowly and asks you to allow a full cure before assessing throw. Heavy material is slow at both ends — slow to leave the wax during cure, slow to leave the pool during a burn. That slowness is the thing you are buying.

The second mechanism is oxidation. Light terpenes react with oxygen, warmth and light, and the reaction products smell different — flatter, sometimes faintly turpentine-like. This is the mechanism behind bottle fade, and it is the one most affected by where you live. A sealed bottle in an air-conditioned room in Bengaluru and the same bottle on a shelf in a Nagpur workshop in May are not having the same year.

Which aroma families hold on, and which ones were always going to be brief

This ladder is general candle-making practice, not a CSI measurement, and no supplier — including this one — publishes a laboratory persistence figure per family. Treat it as the order in which to place your expectations, then confirm with your own jars.

Two honest qualifications. First, "holds longer" and "smells stronger" are not the same judgement — a heavy woody scent can hold for four hours at a level a customer still describes as quiet, while a bright citrus can fill a room for forty minutes and be called strong. Second, several of the scents Indian buyers ask for most — strawberry, bubblegum, most bakery notes, aquatics — have no essential-oil equivalent at all. There is no strawberry essential oil. If your range depends on those, the essential-oil question does not arise; see making scented candles without essential oils.

Heat: the pot, the pool, and a store room in Nagpur in May

The pot is the smallest of the three heat exposures and the one makers worry about most. Soy work sits in a narrow band: the Luxury Soy Wax Chunks page asks you to add fragrance at 80–90°C and stir slowly for two full minutes; the Eco Soy CSI 400 page gives 85°C within an 80–90°C window with a gentle two-minute stir; the Luxury Soy Wax CSI 464 page gives 80–90°C as the working range for melting and pouring. Two minutes of stirring in that band is not where a candle loses its character, and the temperature guide explains why the window sits where it does.

What does cost you in the pot is dwell time. Wax held at 85°C for twenty minutes while you hunt for a wick sticker, or a melter left on while you take a phone call, gives the lightest molecules twenty minutes of open surface to leave from. The fix is not a lower temperature — that brings its own problems with binding and pour finish — it is being ready before you open the bottle: jars wicked, scale zeroed, oil weighed out, lid or foil to hand.

The melt pool is the second exposure, and it is unavoidable: a candle works by evaporating fragrance off a warm liquid surface. A three-hour burn is three hours of evaporation, and the light material goes first. This is why a candle that smelt vivid on lighting can seem muted by hour two even though the pool is larger and hotter than it was. Nothing has gone wrong. The scent has simply moved down its own weight ladder.

The third exposure is the one that actually causes most Indian fade complaints, and it happens after the candle leaves you. Delhi, Ahmedabad and Nagpur summers run past 40°C, and a metal-roofed store room or a cupboard against a west wall runs hotter. A closed delivery van parked in the sun for two hours in May is an oven. Coastal humidity in Mumbai, Kochi and Kolkata does less to the scent directly but encourages makers to store jars open "to let them breathe", which is exactly wrong. A candle with its lid on is a sealed system; a candle with its lid off is slowly perfuming your shelf instead of the customer's room.

Flash point does not explain fading — and what it is actually for

The story goes that essential oils have low flash points, that adding them to hot wax "burns them off", and that this is why the candle is weak. It is a tidy story and it is not what flash point means. Flash point is the lowest temperature at which the vapour above a liquid can ignite in the presence of a spark or flame. It is a transport, storage and handling property — it decides how a drum is shipped and labelled. It is not a boiling point, not an evaporation rate, and not a ranking of hot throw.

It is easy to check that it cannot be a throw ranking. On CSI's own pages, Vetiver gives 206°C and Sandalwood 203°C, while British Rose gives 192°C, Ylang Ylang 193°C and Jasmine 185°C. Those are four very different scent behaviours sitting inside twenty degrees of each other. Meanwhile Citronella gives 132°C and is a famously persistent, pungent material. Flash point is telling you about vapour and ignition, not about how long a scent lasts in a room.

Two things the number is genuinely good for. It confirms that the 80–90°C add window is comfortably below the flash point of most candle fragrance oils — so you are not doing anything dramatic by adding at 85°C. And it flags the small number of materials whose flash points sit close to, or inside, that window, which are worth handling differently.

Sixteen of the ninety-two oil pages print a numeric flash point. The rest do not, and where a page does not publish a figure we will not estimate one — ask on WhatsApp and the documentation can be sent. The same applies to essential oils bought anywhere: if you want the number, ask the supplier for the safety data sheet rather than taking it from a blog. And note what the presence of the number does not tell you: none of these figures says anything about whether the scent will still be there at hour three.

What it costs to chase a fading scent — and why the ceiling stops you before your budget does

Take the series unit: 200 g of Luxury Soy Wax CSI 464 at ₹507.40 per kilogram, which is ₹101.48 of wax, one Eco Candle Wick Thin (C1) at ₹5.90, and fragrance at 8% of wax weight, which is 16 g. With Eucalyptus Spearmint at 1 kg pricing (₹3.07 per gram) that is ₹49.09 of oil and ₹156.47 of materials in the jar, before the container. Fragrance is about 31% of that.

Now push the load. Going from 8% to 10% takes you from 16 g to 20 g and adds ₹12.27 per candle — about 7.8% on the materials bill, or ₹1,227.20 across a hundred candles. That is affordable. The problem is that you have now arrived at the ceiling. The CSI 464 page states up to 10% of total wax weight, and the Eucalyptus Spearmint page states up to 10% in soy. Both say 10%, so 10% is the limit, and there is nowhere else to go.

This is the two-ceilings rule that runs through the whole fragrance series: the wax grade has a stated maximum and the oil page has a stated range, and the lower of the two governs. Moving to a wax with more headroom does not always help, because the oil page may be the binding constraint. Luxury Soy Wax Chunks and Premium Coconut Soy CSI 468 both state up to 13% — but with an oil whose own page stops at 10%, you still stop at 10%. Eco Soy CSI 400 does not publish a fragrance ceiling at all; ask on WhatsApp rather than assuming one.

Two more things the arithmetic says plainly. Pushing a light-family oil from 8% to 12% would cost ₹24.54 a candle, or ₹2,454.40 across a hundred — and it is outside what both pages state, so it is money spent on a formulation neither page supports. And the gap between families is smaller than makers expect: 16 g of Citrus Lemon is ₹45.31 against ₹80.43 for Sandalwood, a difference of ₹35.12 per candle. Choosing the family that holds is roughly the price of a cup of tea per candle. Choosing to pour more of a family that does not hold is both dearer and capped.

If the honest answer is that you want the character of a distilled plant oil and you accept the arc that comes with it, that is a legitimate choice and CSI lists a Butters and Essential Oils collection alongside the fragrance oil range. What this page will not do is tell you that a heavier load rescues a light material. It does not; it just costs more and, past the stated figures, risks seepage and a wick that cannot cope.

The 90-day fade test: four readings, one variable, a written verdict

Fade is a claim about change over time, so it can only be settled by measuring the same jars twice. Everything else is memory, and memory of smell is unreliable. This protocol costs ₹1,032.83 in materials for six jars across two oils, excluding containers, and gives you a defensible answer by the end of the third month.

Add a third scent and the test costs ₹1,490.91 for nine jars — worth it if you are choosing a range rather than checking one oil. Keep the sheet. In six months, when a customer says a candle "smelt stronger last time", a scoring sheet with dates on it is the difference between a guess and an answer.

What to reach for instead, including the oils whose pages state essential-oil content

There are three honest routes out of a fade problem, and only one of them involves buying a different kind of material. Route one is to fix storage and cure, which costs nothing. Route two is to move to a heavier scent family and accept a different character. Route three is to use a formulated oil that sits in the same aromatic territory as the essential oil you wanted — which is what most Indian makers end up doing once they have costed the alternative. A few CSI pages also describe their oils as containing essential oils; those are the closest thing to a middle path, and the fragrance oil versus essential oil guide sets out the wider comparison.

If you do want to work with distilled plant oils directly, CSI lists a Butters and Essential Oils collection and a separate carrier oil range; prices and pack sizes are on those collection pages rather than here, because this page is about candles and the candle usage rates for those materials are not published on them. That is a real limitation and we would rather say it than invent a percentage. For beeswax work specifically, the beeswax and essential oil guide is the live starting point.

One last practical note for anyone building a range for long-lasting performance. Do not fill a catalogue entirely with heavy woods because they score well at hour three. Customers buy on the first sniff in a shop or on a reel, which is a cold-throw, top-note moment. A sensible Indian range carries two or three bright scents that sell fast and are honestly described as fresh and lively, and three or four heavy ones that carry an evening — and tells the customer which is which on the label.

Frequently asked questions