Ash Content and Burnout Behaviour of Casting Resins
Resin datasheets lead with wax content. That number is close to worthless for predicting whether a pattern will cast cleanly. What decides the outcome is how much solid residue is left when the thermal cycle ends, at what temperature the material stops producing gas, and whether the shell can vent that gas fast enough. A resin advertised at 60 % wax can outperform one advertised at 70 %, and both can fail on the same casting if the burnout cycle is wrong.
This note separates the numbers that matter from the number that sells, gives a measurement protocol a shop can run without a laboratory, and maps the common burnout defects back to specific causes.
Three different residues, routinely confused
The confusion starts with vocabulary. Three distinct quantities get called "residue":
- Wax content is the mass fraction of a meltable or low-softening-point component, usually measured by extraction or differential scanning calorimetry. It describes how the material behaves in the first 200 °C. It says nothing about what remains at 600 °C.
- Ash is the inorganic residue left after complete thermal decomposition, typically reported as a percentage of the original mass after heating to 750 °C or above in air. It is silica, ceramic, talc, or oxide from pigments and fillers. Good casting resins sit well below 0.1 %.
- Char is organic carbon that has not oxidised. It is not ash, and it does not appear in an ash figure measured in an inert atmosphere. A resin can legitimately claim 0.02 % ash and still leave visible carbon on the cavity wall.
The commercially important quantity is the sum of char and ash that survives the actual cycle, in the actual atmosphere, at the actual loading density. That number is measured, not quoted.
What the burnout curve actually looks like
A cured filled photopolymer does not melt out like injection wax. Thermogravimetric analysis under a slow ramp typically shows three regions, and the boundaries shift with filler load and photoinitiator package but the shape is consistent:
| Range | What happens | Consequence for the shell |
|---|---|---|
| 150–300 °C | Softening, thermal expansion, onset of scission in the lowest-molecular-weight fraction | Pattern expands against the cavity; early gas release is small in volume but high in condensable compounds |
| 300–500 °C | Bulk polymer chain scission; the mass-loss rate peaks somewhere in this band | Peak gas evolution. This is where pressure builds and where ramp rate matters most |
| 500–750 °C | Oxidation of char, slow and oxygen-dependent | Residual carbon is removed here — or is not removed, if air exchange is poor |
The mass-loss peak in the middle band is the single most useful number on the whole curve. Put a hold there. Most shops do the opposite: they creep through 200–300 °C because that is what they learned with injection wax, then run quickly through 400 °C to save time. That inverts the correct priority. The 150–300 °C band needs to be slow only for thick sections, where thermal lag across the pattern is significant; the 300–500 °C band needs dwell regardless of section thickness, because gas evolution is a bulk reaction and does not care about geometry.
How much gas, and why geometry decides the outcome
A first-order estimate is enough to see the scale. A typical acrylate-based casting resin is roughly 60 % carbon and 7 % hydrogen by mass. Complete conversion of one gram to carbon dioxide and water yields about 0.05 mol of CO2 from the carbon and 0.035 mol of water from the hydrogen, which together occupy on the order of one litre at standard conditions. A 4 gram ring therefore releases several litres of gas inside a cavity whose only exit is the sprue.
Spread over two hours that is a trivial flow rate. Concentrated into a twenty-minute window, on a pattern with a 1.5 mm sprue and a filler network that has already slumped and partly sealed the channel, it is not. The failure is not average flow; it is peak flow against a restricted path.
Two geometry factors dominate. First, surface area to mass ratio: a flat 0.8 mm plate vents easily; a solid 5 mm-thick signet head does not. Second, the drain path: a pattern printed with a single thin sprue on a heavy section is the worst case, because the first material to soften plugs its own exit. This is the practical argument for heavier sprues and for orienting the pattern so the largest cross-section is closest to the sprue, not furthest from it.
A burnout residue test any shop can run
Thermogravimetric analysis is the right tool, but a shop without access to it can get most of the answer with a balance and a furnace.
- Prepare five gram samples of the same resin, cured under the production protocol, in shallow open dishes. Cure matters: an undercured sample will show a false low residue because unreacted monomer boils off before it can char.
- Weigh to 0.001 g before and after, on the same balance, at the same temperature. Do not substitute a different balance mid-test; the difference between a good and a marginal resin is often below the resolution of a cheap scale.
- Run the production cycle, not a laboratory cycle. Same ramp, same holds, same atmosphere, same furnace loading. A residue figure measured with a fast ramp in a clean empty furnace is not the figure the shop will experience.
- Weigh the dish after cooldown in a desiccator; a warm dish picks up moisture and reads heavy. Residue percentage is final mass divided by initial mass.
- Repeat three times. Single measurements on filled resins scatter, and a 0.03 % difference between two resins is meaningless if the repeatability of your method is 0.05 %.
- Record the colour and texture of the residue, not only the mass. Grey powder is inorganic ash. Black film or flakes are char, which means the oxidation step failed and the cycle needs more air or a higher hold — the resin may be fine.
Run the same test twice: once in air, once with the flask door cracked or the vent open exactly as in production. If the two results differ materially, the shop has an atmosphere problem, not a material problem.
Matching defects to mechanism
| Defect | Mechanism | Discriminating check |
|---|---|---|
| Shell cracks radiating from the pattern surface | Internal gas pressure exceeds shell green or fired strength | Crack faces are clean and the pattern is partly intact — pressure, not thermal shock |
| Crack pattern following the flask wall instead | Thermal shock from ramp rate, not gas | Craze lines on the outer wall, pattern fully consumed |
| Rough or pitted casting surface | Char or ash left on the cavity wall | Cavity wall is grey or black after burnout; residue test shows black film |
| Black inclusions in the metal | Char flushed into the metal during pour | Inclusions are soft and smeary under a file, unlike oxide |
| Gas porosity concentrated near the pattern | Decomposition products still evolving when metal enters | Porosity follows the pattern surface, not the sprue base |
| Clean surface but cast is oversized | Thermal expansion of the pattern stressing the cavity | Residue test normal; the error scales with section thickness |
| Localised erosion of the cavity wall | Filler fluxing with the investment at high hold temperature | Glazed, glassy patch on the wall rather than a rough one |
The first three rows are where nearly all production losses sit, and they have different fixes. Pressure cracks need a slower decomposition ramp or a better drain path. Thermal shock cracks need a slower initial ramp. Surface roughness from residue needs more oxygen in the 500–750 °C band, which usually means fewer flasks per load or a longer hold, not a different resin.
Air exchange is a loading problem
Char oxidation is a chemical reaction with oxygen, and a full furnace is an oxygen-starved furnace. Two shops running identical cycles with identical resin can get different residue simply because one stacks flasks mouth to mouth and the other leaves 20 mm between them. Before blaming the material for a black cavity, halve the load and repeat the cycle on the same resin. If the cavity comes out white, the resin was never the problem.
The same logic applies to venting the chamber. A burnout furnace with a closed damper during the oxidation hold will not remove carbon no matter how long the hold runs, because the reaction consumes the available oxygen within minutes and then stops.
What to ask before changing resin
Skip the wax percentage. Ask for the mass-loss curve, or run the five-gram test yourself, and get four numbers: residue at 750 °C in air, temperature at which 95 % of the mass is gone, temperature of maximum mass-loss rate, and whether the residue is ash or char. Those four numbers tell you how to write the cycle. A resin with a mass-loss peak at 430 °C and 0.02 % oxidisable residue will cast cleanly on a well-vented cycle; a resin with the same ash figure but a peak at 500 °C needs a hold twenty minutes later and a slower ramp to get there.
Change the resin only after the cycle has been matched to it. Most "bad batch" complaints resolve into a cycle that was tuned for a different decomposition profile, and the cheapest fix is a rewritten ramp, not a new drum of material.
Ready to discuss your project?
Tell us what you need — our engineering team replies within 24 hours.