Callister — Materials Science and Engineering
Ceramics — why geopolymer breaks the way it does
The Uktena and Ogua chapters both landed on "grout the channels solid" as a rule — this is where that rule comes from, and why a moon-eye brick behaves nothing like a steel beam.
See the build: uktena →Concepts pulled in
- Ceramic failure mode
- Compression vs. tension strength
The Uktena and Ogua chapters (Hibbeler doc) both landed on the same conclusion: unreinforced geopolymer masonry can't be trusted in tension, so the built-in rebar channels have to be grouted and doweled solid. That conclusion was stated as a rule there. This chapter is where it comes from — the material-science reason a moon-eye brick behaves nothing like a steel beam.
Ceramics (which geopolymer chemically is, whatever it looks like) fail in a fundamentally different way than metal. A steel post yields — it bends, stretches, gives visible warning before it breaks. A ceramic doesn't yield; it's held together by ionic/covalent bonds that don't let planes of atoms slide past each other the way metallic bonds do. That means a ceramic is strong in compression (squeezing those bonds together, which they resist well) and weak in tension (pulling them apart, which a tiny surface flaw can turn into a crack that runs instantly — no warning, no bending first). This is not a manufacturing defect to engineer away; it's what the material is.
Why this matters for every geopolymer build in the guide
It's the reason the Snawfus slab problem was about bearing (compression) rather than tension, the reason the Uktena/Ogua walls needed steel specifically to carry the tension the brick itself can't, and the reason a brittle material's failure is sudden rather than gradual — which is also why the "verified" standard leans harder on testing for ceramics/masonry than it does for wood: a wood beam gives you sag as a warning sign; a geopolymer wall does not.
Rule of thumb carried forward
Treat any geopolymer element as roughly 8–10x stronger in compression than in tension (a typical ratio for cementitious/ceramic materials — concrete is commonly cited around 10x), and never rely on the material's own tensile strength for anything load-bearing, even when the hand-calc says it would technically hold. That's the standing design rule this chapter hands to every future geopolymer chapter, the same way the NDS slenderness cap became a standing rule from the fire tower chapter.
Open items for a later pass
Your own cubes, once tested, can also be broken in a quick tension/splitting test (a cube or cylinder loaded on its side, not squeezed flat) to check this 8–10x ratio against your actual mix rather than the general rule.
Modern Monticello · Technical Library · Worked calculation
Ceramics — why geopolymer breaks the way it does
Callister — Materials Science and Engineering · Chapter 3
Project or part: (not named)
Check the compression-to-tension ratio
Enter your compression cubes and a splitting test to see whether your mix follows this chapter's 8 to 10 times rule. The table starts with example numbers (the published low end, 3,900 psi, for all three cubes). Those are not test results. Clear them and enter your own.
| Age | Cubes | Average (psi) | Lowest (psi) | Spread (± psi) |
|---|---|---|---|---|
| 28 day | 3 | 3,900 | 3,900 | 0 |
Each cube: load ÷ face area (15.5 in² for this size). Spread is the sample standard deviation.
Splitting test (check the 8 to 10 times rule from Chapter 3)
A cylinder lying on its side, squeezed along its length until it splits. T = 2P ÷ (π × length × diameter).
Believable-range check from published data (Survey of Mechanical Properties of Geopolymer Concrete, a review of hundreds of results, PMC8398852; R² about 0.7, so wide scatter): at 3,900 psi compressive, expect roughly 372 psi splitting tension and 527 psi bending strength. A result far outside that is a reason to check the test, not a design value. These are guides, never a substitute for your own cubes.
Worked example from the Modern Monticello Technical Library, using published design values and simplified checks. Not a stamped engineering design: your site, materials and loads will differ, and anything that could hurt someone if it fails deserves a licensed engineer's review. Provisional or example values are placeholders, not measurements of your material. Sources for every number: modernmonticellobuild.com/technical-library/callister-geopolymer-ceramics
Print this calculation
Prints just the calculator above, with your numbers and the working, on a clean white page headed with this chapter and today's date. Add a name if you want it on the sheet, for a build file or a plan set.