My White Candle Turns Beige After Adding Fragrance Oil — Why?

Day 0 or day 10? The first question

Before you test anything, find the pour date and think back. Was the candle already off-white when you took it out of the cooling area the next morning? Or did it look right for a week and drift? And is the colour even all round, or darker on the side that faced a window? Those three answers split the causes cleanly, and they cost nothing. Makers who skip this step often run the wrong test: they compare two oils for throw when the beige was a jug problem, or change wax when the colour was building slowly from light.

This guide deals with the first row: beige that is there from the start. It is the one you can settle in a week with four jars. The slower changes in rows two and three are a time-course problem, and the day 0 to day 30 log for them is in Can Fragrance Oil Change Candle Colour During Curing?. If you would rather know in advance which oils on CSI's list are more likely to drift, the note-by-note risk map is in Which Fragrance Oils Cause Candle Discoloration?.

Five things that turn white wax beige

Here they are in the order we would suspect them for day-0 beige. The order is our judgement, based on how commonly each is cited in candle-making practice for day-0 tint and how cheap each is to rule out; CSI has no discolouration measurements for any oil or wax, and nobody should imply otherwise.

1 — The oil's own colour

Many fragrance oils are not water-clear. They range from nearly colourless to pale yellow or light amber in the bottle, depending on what is in them. At 8% a 150 g candle holds 12 g of oil, and a coloured oil at that dose can tint pale wax on its own, with no reaction at all. CSI's product pages, in the text we hold, do not state an oil's colour, so the only way to know is to look: pour a few drops onto white paper, or 5 ml into a small clear glass held against a white sheet, beside a glass of water. If it is visibly yellow in the glass, some of that yellow will be in the wax. The tell-tale sign in a candle is proportion: halve the dose and the tint should roughly halve.

2 — Vanillin and similar components

Vanillin, and some other aromatic components, commonly cause yellowing or browning over time, faster in light. That is usually a slow change, so it more often explains a candle that turned beige over a week than one that was beige on day 0; an oil heavy in vanillin can, though, also be darker in the bottle. Some CSI oil pages state their vanillin content, and it is worth reading before you blame anything else. Vanilla Bean's page states "Vanillin Content: 2.5% (Potential discoloration; test with lighter waxes or consider adding color stabilizers)"; CSI does not currently list a colour stabiliser or UV inhibitor, so that advice cannot be followed with a CSI product. At the other end, twelve CSI pages state 0% or no vanillin, and the Gardenia page states "Vanillin content 0.2% - minimal discoloration risk". Those are page claims. A page that says 0% vanillin is telling you one common cause is absent, not that the candle will stay white; test anyway. The mechanism in vanilla oils is explained in Why Does Vanilla Fragrance Oil Turn Candle Wax Brown?.

3 — A trace of dye

It takes very little dye to shift white wax. CSI's dye reviewers say as much: one says a purple dye is so "concentrated" that "a little amount gives the candle good pastel shade", and a brown-dye reviewer notes shades "light to dark depending upon the amount of dye you add". A jug used for a brown or yellow batch and wiped rather than washed, a stirring stick reused, or trimmings from a coloured candle melted back into the white stock can leave a warm cast that looks exactly like fragrance tint. The test is the unscented jar poured from the same jug: dye contamination shows there too.

4 — Heat during the melt and pour

Too-hot addition, or long holding at high temperature, tends to drive off lighter top notes and may darken some oils. Wax that sits on the heat far longer than it needs to is a risk we would rather avoid too, though we have no measurements to put a number on it. We will not invent a temperature: follow your wax page. CSI 464's page states a pour temperature of 80°C–90°C (it also lists "Melting Point 80°C – 90°C" in its table); CSI 468's states 58°C–60°C; Eco Soy CSI 400's says to add fragrance at 85°C and stir for 2 minutes. If you have no reading for your beige batch, you cannot rule heat out. What happens when oil goes in too hot is covered in What Happens If Fragrance Oil Is Added to Candle Wax Too Hot?.

5 — The wax was never white

Soy wax is cream to ivory, not paper-white, and adding anything to it can shift how the cream reads. CSI's own soy wax pages describe it that way: CSI 464 has "a creamy appearance" and a "smooth, opaque finish"; Eco Soy CSI 400 lists its appearance as "Creamy White"; CSI 468 describes a "rich, opaque appearance". If your reference for "white" is a shop-bought paraffin pillar, a soy container candle will never match it, whatever oil you use. We come back to this below, because it is the one suspect that no purchase from us, oil or otherwise, fully removes.

Is it actually the fragrance oil? The unscented jar

Melt your usual wax in your usual jug to your usual temperature, weigh 150 g into your usual jar, add nothing, and pour. Cool it where your beige candles cooled. Next morning, stand it beside a beige candle from the same wax on a sheet of white paper and photograph both.

If the unscented jar is the same beige, the oil is cleared. The cause is in the wax, the jug, the temperature or the cooling. Buying a different fragrance will give you a different-smelling beige candle. If the unscented jar is clearly paler, the oil is implicated, but not yet convicted: you still need to know whether it is the oil's own colour (a dose effect), a reaction with heat (a process effect), or something about oils in general in your wax (a system effect). That is what the other three jars in the panel are for.

Two cautions. First, compare jars poured the same day from the same wax lot; a new bag of wax can differ slightly from the last. Second, compare set wax to set wax. Molten and warm wax looks darker and more translucent than it will once it has set, so judge no earlier than the next morning, and do not panic at the colour in the jug.

The four-jar beige panel, costed

The panel settles branch A of the tree. Everything is poured from one melt of your usual wax, in one clean jug, at your wax page's temperature, into identical jars, and cooled together. Candles are 150 g with load as a percentage of total weight.

Jar C is a colour diagnostic, not a recommended production load; 4.0% is simply half your usual dose, chosen so the difference is easy to see. If your usual load is 10%, make B 15 g and C 7.5 g, and keep D at your usual load too. Never go above the lower of your wax page's and oil page's stated maximum: CSI 464 states up to 10%.

Add the Pen Thermometer if you do not own one (₹566.40 in all). Figures exclude labour, freight, packaging and GST. Wicks are optional for a colour test, but a wicked jar is a usable candle afterwards, and you may want to burn jar B later to see whether the melt pool darkens. The jars are the dearest line: wash and reuse them for the next test.

Reading the panel: what each result means

"Pale" and "beige" here are relative to each other in the same photo, not absolute colours. If A is ivory and B is a shade warmer, that shade is your oil's contribution. Whether that shade is a problem is a business decision, not a chemistry one: for many ranges an even, stable ivory is perfectly saleable, and customers rarely see a jar beside an unscented control.

When the oil is convicted

You have three honest options. Lower the load, if throw allows: a candle at 6% carries a quarter less oil than one at 8%, and roughly a quarter less of its colour, but judge throw at the lower load before you commit (the load ladder is its own topic in this series). Change the oil, starting with a 50 g test of a candidate poured beside your current oil, never a straight switch to 500 g. Change the vessel, so the wax colour stops mattering: a matte white or coloured jar. Which one is right depends on how much the scent is worth to your range. Whether it is worth dropping a vanilla scent from a white range altogether is a separate policy question, answered in Should I Avoid Vanilla Fragrances When Making White Candles?.

How white can soy wax be?

When the unscented jar is the same colour as your beige candle, the argument moves to the wax. Here is what CSI's container wax pages say about appearance, verbatim where they say anything, and what they do not.

Three practical points follow. First, set your standard against your own unscented jar, not against a photograph of someone else's candle; your customer is comparing your candle with your last candle. Second, frosting changes how colour reads: the white crystalline bloom that soy can develop on the surface makes a jar look paler and patchier, so a batch with less frosting can look slightly "warmer" by comparison. That is a finish effect, not discolouration. Third, switching wax to chase whiteness changes everything else: throw, wick, load ceiling, temperatures and adhesion. Fully refined paraffin is commonly whiter than soy, in our judgement, but it is a different product with different handling, and the page does not state a colour or a fragrance ceiling.

What to buy, in the order you need it

In the order the panel uses them: a thermometer to settle the heat suspect, a comparison oil for jar D, a replacement candidate to test only if your oil is convicted, and a vessel for the maker who would rather keep the scent. Oils are chosen on page vanillin wording, a stated soy load inside CSI 464's 10% ceiling, a stated cure, notes, price per gram and stock at every size on 25 September 2026. CSI has no colour measurements for any of them; the page wording is the page's claim. Candle counts assume 150 g at 8%.

White candles for Diwali orders

Diwali falls on 8 November 2026, 44 days after these prices were checked, and white and ivory candles are a common choice for gifting sets. The panel gives you a day-1 verdict on day-0 causes, and a day-7 look at whether anything is drifting. That leaves time for one tested production batch, if you start this week. Dispatch time depends on your location; confirm it on WhatsApp before you promise a customer a date.

Worked example: 41 candles at 8% from 500 g of Eucalyptus Spearmint (₹1,652.00), with CSI 464, 150 g clear jars and C1 wicks, come to ₹8,635.17, or ₹210.61 per candle, before labour, freight, packaging and GST. In White Matte Glass Jars instead (at ₹70.80 each, and remembering their different fill), the same bill of materials comes to ₹7,667.57. Oil per candle at the 500 g price: ₹39.65 for Eucalyptus Spearmint, ₹79.86 for Neroli.

Frequently asked questions