Scope. This page provides a controlled reader-run protocol and does not claim site-run measurements. For cooking, cooling, storage and reheating, follow the current official guidance linked on the page.
This page is a measurement guide rather than a temperature chart. It defines the fields to record for one specific sauce formula and explains why a number from a different formulation is not automatically transferable.
The worksheet
Use one of these provenance labels beside every value you record:
measured— taken here with a calibrated probe, with a log reference in the cell.cited— reproduced from a named source with a retrieval date in the cell.reader measurement— taken on your equipment using the protocol below; record the formula, instrument, date and replicate count.
Worksheet fields: Emulsion; Continuous phase; What holds it together; Working range; Break temperature; Rescue viability; Provenance. Copy these labels into a notebook or spreadsheet and fill one row per run.
Record Celsius as the canonical reading and add Fahrenheit alongside it if useful to your audience.
Two structural notes about the table are worth reading before you use it. First, the working range and the break temperature are two different measurements, and the gap between them is the useful number. A sauce with a wide gap forgives inattention. A sauce with a narrow gap does not, and that is a property of the sauce, not a comment on the cook. Second, calculate the gap only after measuring both ends under the same conditions: a break point on its own tells you what to fear and a working range on its own tells you what to aim at.
What every emulsion on this list has in common
An emulsion is two liquids that do not mix, forced into a state where one is dispersed through the other as droplets small enough to stay suspended. Left alone, that state is not where the system wants to be. Every emulsion in the kitchen is temporarily prevented from doing what it would rather do, which is separate.
Three things prevent it:
Droplet size. Smaller droplets rise, fall and merge more slowly. This is why the same ingredients whisked by hand and blitzed with a stick blender produce sauces with visibly different stability. It is a mechanical variable, not a chemical one, and it is the one home cooks control best.
An emulsifier. A molecule with an affinity for both phases parks at the boundary and makes merging harder. Egg yolk is the classic one. Mustard, garlic, dairy proteins and cocoa solids all do a version of the same job with different strengths.
Viscosity of the continuous phase. Droplets move slowly through thick liquid. This is why a reduction that has concentrated the gelatine in a stock produces a pan sauce that holds together far better than the same operation with water, and why starch stabilises sauces that would otherwise split.
Break any of the three and the sauce breaks. Heat attacks two of them at once, which is why temperature dominates this page: it thins the continuous phase and, above a certain point, it denatures the protein doing the emulsifying.
Temperature-stabilised versus mechanically-stabilised
This is the real dividing line on the list, and it explains why a filled-in break column looks so uneven.
Mechanically-stabilised emulsions are held together by droplet size and an emulsifier that is already in its final state. Mayonnaise is the archetype. The yolk is raw, the oil is dispersed very finely, and the sauce is stable at room temperature indefinitely by kitchen standards. What breaks it is ratio (too much oil for the emulsifier present) or shock (adding oil faster than the shear can disperse it). Temperature matters at the extremes and is not the everyday enemy.
Temperature-stabilised emulsions are held together partly by being at the right temperature. Beurre blanc and beurre monté are the clearest cases: butterfat needs to be in a particular physical state for the sauce to exist at all. Too cold and it solidifies out. Too hot and it separates cleanly into fat and water in front of you. These sauces have a working range in the literal sense, a band you must stay inside, and they are the ones that break during service while you are looking at them.
Hollandaise sits in both camps, which is exactly why it is the sauce everybody fails. It has a mechanically-stabilised component (dispersed butterfat, yolk emulsifiers) and a temperature-stabilised one (yolk that must be cooked enough to thicken and not so much that it scrambles). Two failure modes, two different rescues, one sauce. It gets its own page: hollandaise in detail.
The practical consequence: before you try to fix a broken sauce, work out which camp it is in. A mechanically-stabilised sauce that has broken has a ratio or shear problem and needs rebuilding. A temperature-stabilised sauce that has broken has been somewhere it should not have been, and the fix starts with getting it back into the band.
Why the break temperature differs so much across the list
Do not expect independently measured break points to cluster. The spread is the interesting part, and it comes from four things.
What the emulsifier is made of. Protein emulsifiers denature with heat and stop working. Emulsifiers that are not protein-dependent do not care in the same way. A sauce leaning on yolk protein has a ceiling that a sauce leaning on mustard mucilage does not.
How much water is in the system. Water is what keeps the continuous phase continuous. Sauces that lose water as they sit on the heat drift towards a state where there is not enough of it to carry the fat, and they break for a reason that has nothing to do with a threshold temperature being crossed. A sauce held for twenty minutes and a sauce made in three minutes are not the same sauce.
Fat percentage and acid level. Both change stability, but this page supplies no universal effect size. Record both for every run, and do not assume a range measured with one combination holds when you change either. Substituting a lower-fat dairy product into a sauce whose range was measured with a higher-fat one is a different experiment.
Whether the break is reversible. Some sauces separate and can be brought back with agitation because nothing has been permanently altered. Others break because a protein has cooked, and cooked protein does not uncook. That distinction is the entire content of the rescue column.
How to read the rescue column
Use the rescue column for a viability observation, not a promise. State what happened in your attempts and how many attempts you made. If you have not run the rescue, leave it blank rather than inventing a success rate.
Record the judgement from a fixed set:
| Verdict | Meaning |
|---|---|
| Routine | Came back in most attempts using the named method |
| Possible | Came back in some attempts; depends on how far gone it was |
| Unlikely | Rarely came back; usually better to rebuild from scratch |
| Gone | The mechanism is irreversible. Do not spend twenty minutes on it |
The one thing worth saying without any measurement is that "gone" is a real category. Emulsions that broke because a protein cooked do not come back, and advice that implies every split sauce is recoverable costs people more time than the sauce was worth. The ordered routes, the diagnosis that tells you which one to use, and the explicit stop-here point are in how to rescue a broken emulsion.
How to measure it
Use this protocol so the numbers remain interpretable:
- One sauce, made to a fixed formula with fat percentage and acid content recorded.
- Probe in the sauce, not in the pan, not in the water bath.
- Temperature raised in small steps with a hold at each step, and the state of the sauce recorded photographically at each hold.
- The working range recorded as the band across which the sauce is stable and behaves as intended.
- The break temperature recorded as the point at which separation begins and does not reverse with agitation.
- Replicated, with the replicate count recorded, because a single break point measured once is an anecdote.
- Failures recorded alongside successes.
Keep the first run deliberately narrow. The following require separate protocols rather than extra rows in the same comparison:
- Emulsions built on ingredients you cannot source consistently, where batch variation would make a single measurement misleading.
- Any range across varying fat percentages, which multiplies the work by the number of percentages tested.
- Hot emulsions held over extended periods, where water loss confounds the temperature reading and the test needs a different design.
- Anything requiring a claim about food safety, which is a question for a named food-safety authority and not a bench question.
For cooking and food-safety decisions, use a validated recipe and current official guidance rather than extrapolating from an incomplete measurement.
In Plain English
What it means. Every classic emulsion is the same physical problem with different ingredients. Some are held together by mechanics and some by temperature, and knowing which is which tells you what will break it and whether it can be saved.
Who it helps. Cooks who learned each sauce as a separate recipe and never saw the shared rules, and anyone who wants one place to look up a working range instead of five.
When to use it. Before you cook, when you are deciding how much attention a sauce will need. Mid-crisis, go to the rescue page instead; this one is a reference, not a triage tool.
What to do next. Work out whether your sauce is mechanically or temperature-stabilised. That single classification predicts most of its behaviour, and it is available to you right now without any numbers at all.
Where To Go Next
The rest of this section, in the order most people need it:
- How To Rescue A Broken Emulsion — three rescues in the order to try them, and an explicit stop-here point.
- Pan Sauces At The Right Temperature — the off-heat window that saves the last thirty seconds of dinner.
- Thickeners Compared: Behaviour By Agent And Degree — activation range and failure mode for each thickening agent.
- Reductions Without Bitterness — why pan geometry matters more than heat level.
- Your Stock Is Cloudy Because You Boiled It — the mechanism behind cloudy stock and the threshold that causes it.
- Stock Below The Simmer: Temperature, Time, Yield — the target band, the hourly return, and a cited cooling protocol.
Next step
The break points on this page are one case of a wider habit: record a failure temperature alongside every target, because the temperature at which something stops working is more useful than the temperature at which it works.
Start with hollandaise in detail, the sauce that sits in both camps, or browse everything in this section if you would rather read by topic.