Question 8 of 8: Heat Treatment of a 0.35%C Steel; Austempering; 2024 Aluminum Tempers

Question 8: Heat Treatment of a 0.35%C Steel; Austempering; 2024 Aluminum Tempers (20 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

(a)(i) Two possible as-cast microstructures at 400 HB. For a 0.35%C steel, 400 HB is far above the hardness of any pearlitic/ferritic structure (which tops out well below 300 HB even for fine pearlite), so the as-cast structure must have cooled fast enough, in at least some sections of the casting, to bypass the pearlite nose. Two credible microstructures are: (1) untempered martensite (or largely martensitic with some retained austenite), from sections that cooled faster than the critical cooling rate; and (2) fine (lower) bainite, from sections that cooled fast enough to miss the pearlite nose but not fast enough to reach $M_s$ before starting to transform isothermally in the bainite range — both microstructures are hard, brittle, and difficult to machine, consistent with the foundry's observation.

(a)(ii) Isothermal (subcritical) anneal for 250 HB max. To guarantee a soft, 250-HB-max, fully pearlitic structure regardless of casting section thickness, the castings are first fully austenitized (heated above the upper critical temperature, $A_3$, and held long enough for a homogeneous austenite structure), then rapidly cooled (quenched) down to a subcritical isothermal holding temperature high in the pearlite range (well above the bainite region, e.g. in the low 600s °C) — fast enough to avoid forming any bainite or martensite on the way down — and held isothermally at that temperature until the TTT diagram's "finish" curve is crossed, i.e. until transformation to coarse pearlite is complete. Because coarse pearlite forms at a higher isothermal temperature (slower nucleation, coarser lamellae) than fine pearlite, it is measurably softer, safely under the 250 HB ceiling. The castings are then cooled to room temperature in any convenient way (air cool), since the transformation is already complete and no further hardening reaction can occur.

(b) Austempering of Belleville washers. The parts are austenitized, then quenched rapidly — fast enough to avoid the pearlite nose — into a molten-salt bath held at a temperature just above $M_s$, within the bainite range (typically $250\!-\!400^\circ$C for a plain-carbon steel), and held isothermally at that temperature until transformation to bainite is complete (crossing the TTT finish curve), then air cooled to room temperature. This is used specifically because (i) it produces tough, reasonably hard bainite directly, without the separate temper step a conventional martensitic quench would need, and (ii) because the part is held at a uniform elevated temperature throughout the transformation rather than quenched all the way to room temperature, thermal gradients and the associated volume-change-driven residual stresses/distortion are much smaller than in a conventional quench-and-temper — important for a thin, precision, springy part like a Belleville washer, where warping or quench cracking would ruin the part's function. Limitations: the part's thickest section must still be able to cool from the austenitizing temperature down to the salt-bath temperature faster than the pearlite nose (limiting austempering to relatively thin sections or steels with enough hardenability); it requires a molten-salt-bath facility and tight time/temperature control, and it works only over the alloy's own bainite "bay", so it is not applicable to steels whose TTT nose is too far to the left (too little hardenability) for practical section thicknesses.

(c) 2024-T4 vs. 2024-T6 vs. 2024-T8. All three tempers start from the same solution treatment: heating the alloy into the single-phase solid-solution field (dissolving the CuAl$_2$/other Cu-rich strengthening phases into the aluminum matrix) and quenching fast enough to retain a supersaturated solid solution (SSSS) at room temperature. From that common starting point:

T4 (naturally aged) means the SSSS is simply held at room temperature for an extended period (days to weeks); Cu atoms slowly cluster into coherent GP (Guinier–Preston) zones, giving a moderate strength increase that continues to develop slowly over time (2024 is well known for aging substantially even at room temperature). T6 (artificially aged) reheats the SSSS to a moderate elevated temperature (roughly $150\!-\!190^\circ$C) for a controlled time, accelerating precipitation to a finer, more uniform dispersion of coherent/semi-coherent precipitates at (or near) the peak-strength condition much faster than natural aging — for many Al–Cu alloys this gives a higher, more reproducible strength than T4, though for 2024 specifically the room-temperature-stable T4 condition is already close to peak strength, so the T4/T6 difference for this particular alloy is smaller than for age-hardening alloys that respond poorly to natural aging. T8 (cold worked, then artificially aged) adds a cold-working step (e.g. stretching or rolling) to the SSSS before artificial aging; the cold work introduces a high density of dislocations, which (i) contributes its own strain-hardening strength increment and (ii) provides many more heterogeneous nucleation sites for the aging precipitates, giving a finer, denser precipitate distribution and the highest strength of the three tempers — at the cost of the lowest ductility/toughness, since both strengthening mechanisms (dislocation forest hardening and precipitation hardening) act together to restrict further plastic flow. Overall expected ranking: strength $T8 > T6 \gtrsim T4$; ductility $T4 \gtrsim T6 > T8$.