A2 vs D2: trading wear resistance for toughness, and what it does to the quote
A2 and D2 sit next to each other on every tool steel chart, both air hardening, both cold work, both landing in the high fifties HRC. Designers often treat them as interchangeable and pick whichever the supplier has in stock.
They are not interchangeable, and the difference shows up in two places: how the tool eventually fails, and what the machining costs.
The composition difference
| A2 (1.2363, SKD12) | D2 (1.2379, SKD11) | |
|---|---|---|
| Carbon | ~1.0% | ~1.5% |
| Chromium | ~5% | ~12% |
| Molybdenum | ~1% | ~0.8% |
| Vanadium | ~0.25% | ~0.9% |
| Typical working hardness | 57-62 HRC | 58-62 HRC |
That 12% chromium in D2 is the whole story. It forms a heavy population of large chromium carbides — hard particles distributed through the matrix. They resist abrasion superbly. They also act as internal stress concentrators, which is why D2 chips where A2 would deform.
A2, with a third of the chromium, has far fewer and finer carbides. Less abrasion resistance, considerably more toughness.
How each one fails
This is the practical difference.
D2 fails by chipping. Edges break away in small pieces under impact or side load. It holds its edge almost indefinitely against sliding abrasion, then loses a chunk the first time the press hiccups or a slug jams.
A2 fails by wearing and deforming. Edges round over gradually. It gives you warning, and it tolerates misalignment and shock that would break a D2 punch.
So the choice is really about the failure mode you can live with:
- Long production runs, clean material, well-aligned press, abrasive workpiece → D2. Wear is the enemy and D2 wins that fight.
- Impact, interrupted cuts, thin sections, less-than-perfect setup, short runs → A2. Toughness is the enemy’s enemy.
A classic case is a blanking die for thin stainless: D2 for the die block, A2 for a slender punch that has to survive being slightly off-centre.
What it does to machining
Soft-annealed, both machine, but not equally:
A2 at around 200-220 HB cuts reasonably. Machinability rating in the region of 65-70% of 1% carbon steel. Ordinary carbide, ordinary feeds.
D2 at 220-255 HB is noticeably worse — around 50-55%. Those chromium carbides are harder than the carbide in your cutting tool, so you are effectively grinding as much as cutting. Tool life drops, and the shop prices inserts accordingly.
Hard, the gap widens further. At 60 HRC both want CBN, but D2 eats edges faster. On a large die block, insert consumption on D2 is a visible line item, not a rounding error. This is the main reason a D2 quote comes back higher than an A2 quote for the same geometry. See D2 vs H13 for the other end of the comparison.
Heat treatment and distortion
Both are air hardening, which is the point of the A and D families — they harden in still or forced air rather than an oil or water quench, so distortion is far lower than O1 or W1. See oil vs air hardening for what that buys you.
Between the two, A2 is slightly more forgiving:
- A2: austenitise around 950-980 °C, air cool, temper 175-540 °C depending on target hardness. Very stable dimensionally.
- D2: austenitise around 1000-1030 °C, air cool, temper 200-540 °C. Also stable, but the higher austenitising temperature and heavier carbide load make it slightly more prone to retained austenite. A double or triple temper is normal practice, and skipping it is how D2 parts grow in service.
If dimensional stability over years matters — gauge blocks, precision fixtures — specify the multiple temper explicitly. It is cheap at the heat treater and expensive to discover later.
What to put on the drawing
Name the grade and the target hardness, and name the equivalent you will accept — D2, SKD11 and 1.2379 are close enough that insisting on one specific designation can add weeks to a Vietnamese or Chinese supply chain for no functional gain. The grade equivalents table covers the cross-references.
And say what the tool does. A shop that knows the punch sees impact will suggest A2 if you specified D2, and that conversation is worth having before the block is cut, not after the first one chips. Send the drawing with the application described and you will get that feedback in the quote.