Question 2 of 7: Constructing TTT Curves, the Origin of the "C" Shape, and TTT vs. CCT Procedure

Question II: Constructing TTT Curves, the Origin of the "C" Shape, and TTT vs. CCT Procedure (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.

II.1 — (i) Experimental construction of a TTT curve

A TTT (time–temperature–transformation) curve is built from a large batch of small, identical specimens of the same steel, transformed one at a time under strictly isothermal conditions:

1. Austenitize every specimen identically. Heat each specimen above its upper critical temperature and hold long enough to obtain a fully homogeneous, single-phase austenite structure of uniform grain size.
2. Quench rapidly to a chosen sub-critical hold temperature. Transfer a specimen quickly (to avoid any transformation en route) into a constant-temperature bath — typically a molten salt or lead bath — held at a fixed temperature $T_1$ somewhere between $A_1$ (727 °C) and the martensite-start temperature $M_s$.
3. Hold isothermally for a chosen time, then quench to room temperature. After a set holding time $t$ at $T_1$, quench the specimen into water/brine to instantly freeze whatever fraction had transformed at $T_1$ and to convert any still-untransformed austenite to (fresh) martensite, which is readily distinguished from the isothermal product under the microscope.
4. Determine the transformed fraction. Examine the quenched specimen metallographically (or by hardness, dilatometry, or magnetic saturation) to measure the fraction transformed at $T_1$ after time $t$. Repeat with fresh specimens at a series of times to bracket the "start" ($\sim$1% transformed) and "finish" ($\sim$99% transformed) times at $T_1$.
5. Repeat the whole procedure at a series of hold temperatures. Cover a range of $T_1$ values spanning just below $A_1$ down to just above $M_s$, obtaining a start-time and finish-time at each temperature.
6. Plot the diagram. Plot temperature (linear, vertical axis) against time (logarithmic, horizontal axis); join all the start points into a single "start" curve and all the finish points into a "finish" curve, and add horizontal lines at $M_s$ and $M_f$.

II.2 — (ii) Why the TTT curve is "C"-shaped

The time needed to start the diffusional (pearlite/bainite) transformation is controlled by two competing, temperature-dependent factors — the thermodynamic driving force for transformation and the atomic diffusivity that carbon (and, for bainite, substitutional atoms) must have to redistribute and let the new phases nucleate and grow:

Near $A_1$ (small undercooling, high temperature). Diffusivity is high, but the driving force ($\Delta G_v$, proportional to the undercooling below $A_1$) is very small, so the nucleation rate is very low. Transformation is slow because there is little thermodynamic incentive to nucleate the new phase, even though atoms can move quickly. Transformation start time is therefore LONG.

Near $M_s$ (large undercooling, low temperature). The driving force is now large, but atomic diffusivity has fallen off exponentially (Arrhenius behaviour) — carbon and substitutional atoms can barely move, so nucleation and growth are both sluggish even though there is ample thermodynamic incentive. Transformation start time is again LONG.

At an intermediate temperature (the "nose," typically $\sim$550 °C for a eutectoid steel). Here there is a favourable balance — enough driving force AND enough diffusivity for both nucleation and growth to proceed rapidly. This is where the transformation start time is SHORTEST.

Because transformation time is long at both the high-temperature and low-temperature ends and reaches a minimum at an intermediate temperature, plotting time (log scale) against temperature necessarily traces out a curve that bulges toward short times in the middle and swings back toward long times at both ends — the characteristic "C," or nose, shape.

II.3 — (iii) TTT vs. CCT: procedural differences

A TTT curve is built entirely from ISOTHERMAL holds, as described in Part (i): each data point comes from a separate specimen quenched rapidly to and then held at ONE constant temperature, with transformation progress measured purely as a function of time at that fixed temperature. Many specimens and many discrete hold temperatures are needed, and the resulting curve strictly describes transformation kinetics only under isothermal conditions.

A CCT curve is instead built from specimens that are austenitized and then cooled CONTINUOUSLY, each at its own constant cooling RATE (e.g., a set of different linear or Newtonian cooling rates imposed on a dilatometer), rather than being held at a fixed temperature. Transformation start and finish are recorded as the TEMPERATURES (not fixed hold times) at which the transformation begins and ends along each continuous-cooling path, typically detected in real time from the length (dilatometric) change as the specimen cools, or by interrupting several duplicate continuous-cooling runs at different points and examining microstructure/hardness. Because the temperature keeps falling throughout the whole transformation (rather than being held fixed at the most favourable nose temperature), diffusional transformation under continuous cooling is slightly suppressed relative to an equivalent isothermal hold, so CCT curves are shifted to longer times (down and to the right) relative to the TTT curve for the same steel, and CCT curves have no separate "finish" branch below the nose for high enough cooling rates (the austenite bypasses the diffusional transformation entirely and goes to martensite). Since almost all real heat treatments (quenching, normalizing, air cooling) are continuous-cooling processes, CCT curves are the practically relevant diagram for predicting the as-quenched microstructure, while TTT curves remain the more fundamental representation of isothermal transformation kinetics.