A few hours after you pour, a loaf can turn dark and translucent from the middle outward, like a slow ripple, and stay that way for a while before going opaque again as it cools. That's gel phase. It changes how the bars look and how soon you can cut them, and it changes nothing about whether they are good soap.
The advice everywhere is to insulate if you want it and not if you don't. That's true and it is roughly a third of the story, because whether a batch can reach gel at all is mostly decided by the mold you poured it into — and none of the pages ranking for this term mentions mold size at any point.
What gel phase actually is
Saponification is exothermic: the batch heats itself. Left to accumulate, that heat takes the batter to somewhere around 75–85 °C (165–185 °F) — well past the point where the young soap softens — and the loaf becomes a hot, translucent gel. It cools over the following hours and sets opaque again.
It is easy to read "gel phase" as a step in the chemistry, like trace. It isn't. Gel is the same saponification you already have, running hot enough to melt the soap it is making. A batch that never gels saponifies just as completely, on almost the same schedule — it simply does it below the melting point, so nothing ever goes translucent. That is why the finished bars are equally good soap, and why "did it gel?" is a question about appearance and timing rather than about whether the batch worked.
What gelling actually changes
| Gelled | Ungelled | |
|---|---|---|
| Color | Deeper and more saturated. The main reason people force it — natural colorants especially. | Paler, matte, creamy. Sometimes exactly what you want. |
| Surface | Slightly shiny. | Matte, and more prone to soda ash. |
| Unmolding | Firm sooner — often the same day. | Slower; a soft recipe can need three or four days. |
| Texture | Slightly harder early on; the difference fades over the cure. | Softer at first, and easier to cut cleanly. |
| Cure | No real difference. Both need the same weeks for water to leave. | Same. |
| Quality | No difference. Neither is milder, harder-wearing or better lathering than the other once cured. | |
The one genuine hazard is on the gelled side: milk, honey and other sugars scorch at gel temperatures, which is why those recipes are usually kept cool deliberately.
Whether it gels is mostly your mold
Here is the part nobody publishes. Heat is produced throughout the batter, so in proportion to its volume. Heat is lost through the outside, so in proportion to its surface. Whether the temperature climbs to gel or leaks away is therefore decided by surface ÷ volume — a number that is fixed the moment you choose a mold, and that varies by a factor of four across molds people use every week:
| Mold | Batter | Oils | Surface per 100 mL | What happens |
|---|---|---|---|---|
| 5 lb slab12 × 9 × 2.5 in | 4,425 mL | 2,951 g | 46.8 cm² | Gels whether you want it to or not. Plan around it. |
| 2.5 lb loaf10 × 3.5 × 2.5 in | 1,434 mL | 956 g | 61.9 cm² | The common case. Gels reliably if covered, usually not if left bare in a cool room. |
| Tube mold3 in across, 8 in tall | 927 mL | 618 g | 62.3 cm² | Tall and narrow is efficient — behaves like a loaf twice its size. |
| 1 lb loaf8 × 2.5 × 2.5 in | 819 mL | 547 g | 72.8 cm² | Marginal, and so the classic partial gel. Insulate it or chill it — don’t leave it in between. |
| 12-cavity mold82 mL per cavity | 984 mL | 656 g | 138.1 cm² | Will not gel on its own. Needs an oven or a heating pad if you want it to. |
| 24 guest cavities35 mL per cavity | 840 mL | 560 g | 183.4 cm² | Effectively cannot be gelled by insulation alone. |
Exposed surface of the batter — mold walls, base and the open top — per 100 mL of batter, computed by this site's engine from the filled dimensions; oil weights at a 33% lye solution. Cavities are treated as cubes of their volume, which is the smallest surface a real cavity could have, so the last two rows are conservative. Put your own mold through the mold volume calculator for its dimensions.
A 12-cavity mold sheds heat over 2.2 times as much surface per unit of batter as the standard loaf, and 3.9 times as much as a slab. That is the whole explanation for something every soapmaker notices and no guide accounts for: the same recipe, on the same day, gels in a loaf and refuses to in individual cavities. It also explains why a tall narrow tube behaves like a much bigger mold, and why the 1 lb loaf — sitting awkwardly at 72.8 — is the one that gives you a half-gelled bar.
Scale a loaf up and gel gets easier, because volume grows faster than surface. Scale a cavity mold up and nothing changes at all: six cavities and twelve cavities have exactly the same surface per 100 mL — 138.1 cm² either way — because you have added separate small blocks of batter rather than a bigger one. Each cavity is thermally on its own. So the usual advice to "make a bigger batch if you want it to gel" is sound for a loaf and completely empty for cavities, where the only routes are an external heat source or accepting an ungelled bar.
The other levers
- Soaping temperature. The head start. Batter that goes into the mold at 50 °C needs far less self-heating to reach gel than batter poured at 30 °C.
- Insulation. Slows the loss side. A cover, a towel, a cardboard box — worth more than it sounds, because it also stops the draft that carries heat off the open top.
- Water. More water is more thermal mass to carry heat, and it lowers the temperature at which the soap softens. The classic bar at 25% carries 431 g of water per kilo of oils against 216 g at 40% — which is why steeply discounted recipes resist gelling and full-water ones fall into it. The water discounting guide covers what else that choice costs.
- Sugars. Honey, milk, beer and added sugar all raise the peak temperature. Useful if you want gel, and the reason those recipes are the ones most likely to overheat.
- Mold material. Wood insulates, silicone is middling, metal is a heat sink. A silicone liner inside a wooden box is the warmest common setup.
How to force gel
- Soap warm — both parts around 43–54 °C (110–130 °F).
- Cover and insulate as soon as it's poured: a board or piece of cardboard over the mold, then a towel or blanket.
- Add heat if the geometry is against you. A heating pad or seedling mat under the mold for the first hour, or the oven method: warm the oven to about 75 °C (170 °F), turn it off, put the covered mold in and leave it overnight. For a cavity mold this is the only thing that reliably works.
- Check after an hour or two. Gel spreads from the center; if the middle is dark and the edges are not, add insulation now rather than later.
How to prevent gel
- Soap cool — room-temperature lye solution and oils barely melted, as low as the recipe's hard fats allow before you risk false trace instead (the trace guide covers that floor).
- Don't cover it. Leave the mold open in the coolest room you have.
- Use a water discount, which removes thermal mass.
- Refrigerate or freeze. Into the fridge for 24 hours, or the freezer for a few hours then the fridge, is the reliable method — and the standard one for milk and honey soaps. Let it come back to room temperature before cutting.
- Expect more soda ash, and lay plastic wrap on the surface, which prevents ash without adding meaningful warmth.
Partial gel: the one outcome nobody wants
A dark translucent circle in the middle of a pale bar. It is purely cosmetic — the soap is fine throughout — but it looks like a mistake because it is one: the batch got hot enough to gel in the center and lost the heat before the gel reached the edges.
Which is exactly what the table above predicts. The center of a loaf is the furthest point from every surface, so it is always the last place to lose heat and the first to gel. Molds in the middle of that range — the 1 lb loaf, an uninsulated 2.5 lb loaf in a cool room — are the ones that get partway and stall. The fix is to commit in one direction:
- Toward gel: soap warmer, insulate properly, add a heat source. Half-hearted insulation is the reliable route to a partial gel.
- Away from gel: the fridge. Cold enough, fast enough, and no part of the loaf gets there.
- Wide shallow molds need more help than their size suggests, because a slab's center is close to a large flat surface on two sides. A tall narrow mold holds its heat better than a broad one of the same volume.
Can it be fixed? Within about a day, yes: put the whole loaf in a 75 °C (170 °F) oven for half an hour and let it gel through. After that, it's a rustic bar, or a rebatch, or you cut it and use it yourself — it washes exactly as well as a bar that looks right.
So which should you do?
- Force gel for bright or natural colors, for a faster unmold, and any time you're making a loaf big enough that fighting it is work.
- Prevent gel for milk, honey, beer and fruit purées, which scorch; for pastel and creamy looks; and for piped or textured tops you don't want to slump.
- Either for a plain first batch. Pick one on purpose, note what you did, and compare — that's a more useful experiment than reading about it.
The short version
- Gel is heat, not a chemical stage. The same reaction, running hot enough to melt its own soap.
- Your mold decides more than your blanket. 46.8 cm² per 100 mL for a slab against 183.4 for guest cavities.
- Bigger batches gel more easily — in a loaf. In a cavity mold, not at all.
- Neither bar is better soap. Color, shine and unmold time; that's the list.
- Commit. Partial gel is what happens to people who do half of each.