AISI D2 / 1.2379 / SKD11 Spheroidising: The 60°C Austenitising Drop That Reshapes the Carbides

Cold-Work Tool Steel · Field Note from FUSHUN METAL

AISI D2 / 1.2379 / SKD11 Spheroidising: The 60 °C Austenitising Drop That Reshapes the Carbides

In a recently published trial on Cr12Mo1V1 — the standard AISI D2 / DIN 1.2379 / JIS SKD11 cold-work die steel — two spheroidising anneals were run on the same heat, with the same chemistry and the same three-stage sequence. The first held the austenitising soak at 920 °C for 8 h; the second at 860 °C for 8 h. The first produced a network of grain-boundary carbides and a globule rating of ‘poor’. The second produced fine, evenly distributed spheroidised carbides rated level 4 on the GB/T 1299 reference chart. The single change — a 60 °C drop in the austenitising temperature — moved the same alloy from failed to fully spheroidised. That is the number this note is built around, and it has direct consequences for anyone buying or heat-treating Cr12Mo1V1 / D2 / 1.2379 / SKD11 die blocks, rolls, or shear blades.

1. Grade cross-reference: Cr12Mo1V1 and its international equivalents

The trial under discussion uses the Chinese GB/T 1299 designation Cr12Mo1V1. The grade is the direct Chinese equivalent of the US AISI D2 (UNS T30402) and the European 1.2379 / X155CrVMo12-1, and is functionally interchangeable with the Japanese SKD11 (JIS G4404). All four are 12 %-Cr, Mo-V, air-hardening, high-carbon, high-chromium cold-work tool steels with the same operating envelope and the same metallurgical traps. The table below is what an international procurement or QA team should match when they are buying any one of them.

The trial bar (Table 1 in the source) and the published spec windows for D2 and 1.2379 sit on top of each other on every controlled element:

The trial bar sits closer to the high-Mo / high-V / low-W end of the family (Chinese Cr12Mo1V1 and US D2) than to the low-Mo / low-V / low-W European 1.2379 end. That difference matters for the result, but it is small enough that the published spheroidising insight carries across all four designations with the same logic.

2. The 0.5 °C/s rule: where the CCT curve puts the line

The published CCT curve was measured on a Gleebel-3500, with samples austenitised at 950 °C / 10 min and then cooled at twelve discrete rates between 0.02 °C/s and 50 °C/s. The critical transformation temperatures on the trial bar came in at A_c1 = 827 °C and A_c3 = 870 °C. Three transformation fields are visible on the curve:

• Pearlite (P) in the 600–827 °C high-temperature nose — diffusion-controlled transformation of undercooled austenite, Fe and C both mobile.
• Bainite (B) in the 300–500 °C intermediate-temperature field — C diffuses, Fe does not. The carbon-enriched austenite that survives the pearlite nose reappears here as bainite plates.
• Martensite (M) below the M_s line (around 190 °C) — diffusionless transformation of any austenite that survives the pearlite and bainite noses.

The single most important number on the curve is the critical cooling rate of 0.5 °C/s. Below that rate, undercooled austenite has time to clear the pearlite nose and the bulk transformation product is soft pearlite. Above that rate, the austenite skips past both the pearlite and bainite noses and lands as martensite. The published micrographs (Fig. 2, a–i) show the transition happening between 0.30 °C/s (mixed P + B + M, faint pearlite still visible) and 0.50 °C/s (full martensite).

3. The pearlite–martensite boundary, in numbers

Table 3 in the source reports Vickers hardness and dominant microstructure for all twelve cooling rates. The shape of the data is the point. The published numbers, rounded, are:

Three observations matter for the specifier. First, 0.05 → 0.1 °C/s is the bainite onset: hardness jumps from 244 to 349 HV, a 43 % rise in a single step. Second, 0.3 → 0.5 °C/s is the martensite onset: from 587 to 667 HV, with the first all-martensite structure at 0.5 °C/s. Third, once you are in the martensite field, faster cooling does not make the steel harder: 10 °C/s gives 673 HV, 50 °C/s gives 642 HV, essentially the same plateau. The lesson is that for a spheroidising anneal (which is what we actually want, not martensite) the only correct target is a cooling rate well under 0.05 °C/s — the published subcritical anneal is run in the 0.02–0.04 °C/s range, in the middle of the safe pearlite window.

4. The spheroidising surprise: 1# vs 2# anneal

The published trial compares two three-stage spheroidising anneals on the same bar. The only deliberate changes between them are the austenitising temperature and the longest spheroidising hold (Table 2 in the source):

The two microstructures tell the same story as the numbers. The 1# anneal (920 °C soak, 12 h spheroidise) shows white blocky carbides strung along the prior-austenite grain boundaries in the optical micrograph and a network of dark cavities in the SEM — the classic signature of a high-temperature austenitising soak that has dissolved too many primary carbides, followed by a hold that is too short for them to re-precipitate as globules. The 2# anneal (860 °C soak, 16 h spheroidise) shows fine, evenly distributed spheroidised carbides with no grain-boundary network and only a small amount of untransformed residual austenite.

5. Why it works: the ‘undissolved carbide seeds’ mechanism

The published explanation is the right one, and it is worth being explicit about because it generalises to every D2 / 1.2379 / SKD11 spheroidising anneal the shop will ever run. Spheroidisation in a high-C, high-Cr cold-work die steel is a two-step process: dissolution of the as-cast carbide network during austenitising, followed by re-precipitation of carbon on existing particle cores during the sub-critical hold. Both steps are necessary, and they have to be balanced.

1. If the austenitising soak is too hot or too long (1# anneal: 920 °C / 8 h), the primary M₂₃C₆ carbides dissolve into the austenite, the carbon concentration in the matrix becomes uniform, and the undercooled austenite has no pre-existing particle cores on which to nucleate globular cementite. The result is either retained grain-boundary carbide network (slow cool) or martensite with brittle grain-boundary decoration (fast cool).
2. If the austenitising soak is at or just below A_c3 and short enough to retain undissolved primary carbides (2# anneal: 850–860 °C / 8 h, well below the 870 °C A_c3), the matrix is carbon-enriched but carbon-heterogeneous, with a population of undissolved M₂₃C₆ cores left in place. During the subsequent 700–720 °C pearlitic transformation, carbon re-precipitates on those cores. During the 800–820 °C spheroidising hold, surface tension drives the rod-like or blocky cementite to break up into spheres (the ‘Rayleigh instability’ of a cylinder, applied to carbide morphology). The 16 h hold is what gives the break-up time to complete.
3. The CCT curve explains why the upper temperature matters so much. A soak above A_c3 (870 °C on this bar) risks dissolving the carbide population entirely; a soak well below A_c1 (827 °C) would not austenitise at all. The published ‘best’ window is 850–860 °C, sitting 10–20 °C below A_c3 and 25–35 °C above A_c1 — close enough to A_c3 to give uniform austenite, far enough below it to leave the undissolved carbide seeds in place.

6. What this means for buyers and heat treaters

If you are buying Cr12Mo1V1 / D2 / 1.2379 / SKD11 die blocks, rolls, shear blades, or extrusion tooling, the published data argue for four changes to the usual conversation:

1. Ask for the actual austenitising temperature, not just the word ‘annealed’. An ‘annealed’ bar that was soaked at 920 °C is not the same product as an ‘annealed’ bar soaked at 860 °C. The first will have a partial grain-boundary network that survives into the quench and shows up as brittle streaks in service; the second will be uniformly spheroidised. A mill certificate that says ‘annealed to ≤ 255 HB’ tells you hardness but not microstructure.
2. Pin the austenitising window relative to A_c1 / A_c3, not in absolute °C. A 1.2379 European bar with a slightly different Cr / Mo ratio will have a different A_c1 / A_c3 than a high-Mo D2. The published ‘850–860 °C’ window is specific to a Cr12Mo1V1-type high-Mo chemistry. The right rule for any of the four equivalents is ‘austenitise at A_c3 − (10 to 20) °C for 4–8 h, then run the standard sub-critical spheroidising sequence’.
3. Treat the spheroidising hold as the rate-limiting step. 4 h and 8 h are insufficient. 12 h gives a partial globule rating. 16 h is what produces the level-4 rating the GB/T 1299 chart treats as fully spheroidised. If the shop’s standard cycle is ‘anneal to ≤ 255 HB in 12 h total’, the bar is leaving the furnace with incomplete spheroidisation, and the customer is paying for a quench-and-temper performance that the microstructure cannot deliver.
4. Cool through the pearlite nose, not past it. The published critical cooling rate is 0.5 °C/s. A sub-critical anneal that drops the bar into still air from 800 °C can easily hit 1–3 °C/s through the pearlite nose, producing a mixed P + B + M structure with hardness above 500 HV. Furnace cool at ≤ 30 °C/h, or hold at 700–720 °C for the isothermal pearlite step, is what keeps the structure on the correct side of the 0.5 °C/s line.

A spheroidised Cr12Mo1V1 / D2 / 1.2379 / SKD11 bar is not the same as an ‘annealed’ bar. The difference is the austenitising temperature (60 °C), the spheroidising hold length (4 h), and the cooling rate through the pearlite nose. The published data show the failure mode (grain-boundary network, hardness 666 HV after a mistaken fast cool) and the success mode (level-4 globules, hardness 220–240 HV after a 16 h isothermal anneal) on the same bar, in the same paper. The spec just has to ask for the success mode.

7. The point of view

In an AISI D2 / 1.2379 / SKD11 bar, the alloy is the easy part. The hard part is the anneal. A spheroidising soak that runs 60 °C too hot, or 4 h too short, or 10× too fast through the pearlite nose will silently move the same steel from a level-4 globule rating to a brittle grain-boundary network — on a mill certificate that still says ‘annealed to 240 HB’. The 0.5 °C/s critical cooling rate and the A_c3 − (10 to 20) °C austenitising window are the only two numbers a procurement spec actually needs to control. They are now in the public domain; the spec just has to use them.

Source: Qiu Guijuan (Heye Technology). CCT curves and spheroidising-anneal process of Cr12Mo1V1 cold-work die steel. Hebei Metallurgy, 2023, (12). (Original Chinese; trial on a Φ300 mm VIM + VAR bar; 400 MN-class rolling-mill process-route design.)