The Time-Temperature-Transformation (TTT) diagram — also called the isothermal transformation diagram or S-curve — is the most important single tool in steel heat treatment metallurgy. It maps, for a given steel composition and prior austenite grain size, the kinetics of austenite decomposition as a function of temperature and time under isothermal (constant-temperature) conditions. Every C-curve, every nose, every Ms line encodes the thermodynamic driving forces and diffusion kinetics that govern what microstructure forms when austenite is quenched and held at any temperature. Reading and interpreting the TTT diagram is a core competency for any metallurgist, heat treatment engineer, or materials scientist working with steels.

How to Read the TTT Diagram

The TTT diagram has temperature on the vertical axis (linear scale, °C) and time on the horizontal axis (logarithmic scale, seconds). The diagram is constructed for a specific austenite composition and prior austenite grain size, and is valid only for isothermal conditions — meaning the steel is assumed to be quenched instantaneously from the austenitising temperature to the holding temperature, arriving there before any transformation has occurred.

The C-Curves: Pearlite and Bainite Transformation Kinetics

Two sets of C-shaped curves appear on a typical TTT diagram for a plain or low-alloy steel. Each set has a start curve (conventionally defined at 1% transformation) and a finish curve (99% transformation). The region to the left of any start curve represents untransformed supercooled austenite; between start and finish, transformation is in progress; to the right of the finish curve, the transformation is essentially complete.

• Upper C-curves (pearlite): Occur at temperatures just below A1 (727°C) down to approximately 550°C for eutectoid steel. Pearlite is a lamellar eutectoid product of ferrite and cementite, formed by a diffusion-controlled cooperative growth mechanism. The interlamellar spacing decreases as the transformation temperature decreases, giving progressively finer and stronger pearlite. See the Pearlite Colony Growth article for the full treatment of lamellar spacing and colony nucleation kinetics.
• Lower C-curves (bainite): Occur between approximately 550°C and Ms (around 230°C for eutectoid steel). Bainite forms by a mixed shear-diffusion mechanism: the ferrite sublattice forms by shear (like martensite) but carbon is then rejected by diffusion and precipitates as carbides. Upper bainite (400–550°C) has inter-lath cementite films; lower bainite (<350°C) has intra-lath carbides on specific habit planes. The Bainite Microstructure article covers this in full crystallographic detail.

Why the C-Shape Arises: Competing Arrhenius Kinetics

The characteristic C-shape is a direct consequence of two competing temperature-dependent factors that together control the overall transformation rate:

Overall transformation rate: R(T) ∝ N_dot × v  (nucleation rate × growth velocity)

Nucleation rate:  N_dot ∝ exp(−ΔG* / kT) × exp(−Q_diff / RT)
  ΔG* ∝ 1 / ΔT²   (activation barrier decreases as undercooling ΔT increases)

Growth velocity:  v ∝ D × ΔG / RT
  D = D₀ × exp(−Q_D / RT)   (diffusivity: Arrhenius, increases with T)

Near A1 (small ΔT):   ΔG* very large → near-zero nucleation → slow transformation
Near Ms (large ΔT):   D → 0 → near-zero growth → slow transformation
At nose (optimal ΔT): N_dot and v both significant → maximum transformation rate

The product of these two opposing exponentials creates
a maximum at intermediate temperature → the C-curve nose.

Transformation Products and Their Properties

Martensite: Athermal Transformation and the Koistinen-Marburger Equation

Martensite is unique among steel transformation products: it forms by a diffusionless, athermal shear transformation that requires no time at temperature. Instead of a C-curve, the TTT diagram shows two horizontal lines — Ms and Mf — whose temperatures are fixed solely by the chemical composition of the austenite. The crystallographic mechanism involves simultaneous cooperative atomic displacements that preserve the iron lattice topology while trapping all dissolved carbon in interstitial positions of the body-centred tetragonal (BCT) cell. The tetragonality ratio c/a increases linearly with carbon content: c/a = 1 + 0.046 × (wt% C).

The Koistinen-Marburger Equation

The fraction of martensite formed at any temperature below Ms was empirically quantified by Koistinen and Marburger (1959) by systematic dilatometry measurements on a wide range of steels:

Koistinen-Marburger (K-M) equation:
  f_M = 1 − exp[−α × (Ms − T)]

Where:
  f_M  = volume fraction of martensite (0 to 1.0)
  Ms   = martensite start temperature (°C)
  T    = quench/hold temperature (°C, must be < Ms)
  α    = 0.011 K⁻¹   (empirical constant; valid for carbon steels)
         [Some authors use 0.0110–0.0115 depending on steel]

Application — eutectoid steel (Ms ≈ 230°C):

  At room temperature (T = 25°C):
    f_M = 1 − exp[−0.011 × (230 − 25)]
        = 1 − exp[−0.011 × 205]
        = 1 − exp(−2.255)
        = 1 − 0.1046 = 0.895   →  89.5% martensite; 10.5% retained austenite

  At T = −80°C (dry ice treatment):
    f_M = 1 − exp[−0.011 × 310]
        = 1 − exp(−3.41)
        = 1 − 0.033 = 0.967   →  96.7% martensite; 3.3% retained austenite

  To achieve < 2% retained austenite:
    0.02 = exp[−0.011 × (230 − T)]  →  T ≈ −105°C  (liquid nitrogen territory)

Factors Controlling Ms Temperature

Ms is the single most important datum on the TTT diagram for quench design. It is controlled by austenite composition through the Andrews (1965) empirical equation, which the calculator above implements:

Andrews (1965) Ms equation:
  Ms (°C) = 539 − 423·C − 30.4·Mn − 17.7·Ni − 12.1·Cr − 7.5·Mo

Approximate elemental effects on Ms:
  Carbon (C):        −423°C / wt%   (strongest depressant — interstitial)
  Manganese (Mn):    −30.4°C / wt%
  Nickel (Ni):       −17.7°C / wt%
  Chromium (Cr):     −12.1°C / wt%
  Molybdenum (Mo):   −7.5°C / wt%
  Silicon (Si):      +8°C / wt%     (mild raiser)
  Cobalt (Co):       +10°C / wt%    (raiser)

Examples:
  AISI 1080 (0.80C, 0.75Mn):            Ms ≈ 539 − 338 − 23 = 178°C
  AISI 4140 (0.40C, 0.90Mn, 1.0Cr, 0.20Mo): Ms ≈ 539 − 169 − 27 − 12 − 1.5 = 329°C
  D2 tool steel (~1.5C, 12Cr, 0.8Mo):    Ms ≈ 539 − 635 − 145 − 6 = −247°C
  (D2 Ms is below −100°C → deep cryogenic treatment essential)

Effect of Alloying Elements on the TTT Diagram

The most commercially significant effect of alloying is the shift of C-curves to longer times, increasing the time available to cool through the transformation temperature range without forming pearlite or bainite. This is quantified as hardenability — the depth to which martensite can be formed in a quenched bar. Different elements affect pearlite and bainite kinetics differently:

The Mo ‘Bay’: Separation of Pearlite and Bainite C-curves

In steels containing molybdenum (and to a lesser extent, other strong carbide formers), the pearlite and bainite C-curves can become completely separated on the TTT diagram, creating a temperature–time “bay” between them where neither transformation is fast. Molybdenum specifically retards grain-boundary nucleation of pearlite far more strongly than it retards the bainite shear mechanism. This bay is exploited in two important commercial heat treatments: martempering (quench into the bay, hold until temperature equalises through the section, air cool for uniform martensite) and stepped bainite treatments for thick sections in alloy steels like 300M, 4340, and 4330V.

Critical Cooling Rate and Hardenability

The critical cooling rate (CCR) is the minimum cooling rate that avoids the pearlite C-curve nose entirely, producing 100% martensite on continuous cooling. On the TTT diagram it is found by drawing the continuous cooling curve that is tangent to the pearlite start (Ps) nose — the line from the austenitising temperature that just touches the fastest point of Ps without crossing it.

Critical cooling rates for representative steels:

Steel           Composition (simplified)     CCR (°C/s)   Quench for 25mm bar
AISI 1080       0.80C, 0.75Mn                ~140          Water only; limited section
AISI 1045       0.45C, 0.75Mn                ~400          Water, small sections only
AISI 4140       0.40C, 0.90Mn, 1.0Cr, 0.20Mo ~30          Oil quench (any section)
AISI 4340       0.40C, 1.8Ni, 0.80Cr, 0.25Mo  ~5           Oil or even polymer quench
AISI H13 (tool) 0.40C, 5.0Cr, 1.5Mo, 1.0V    ~1           Air quench, large sections
M2 (HSS)        0.85C, 6.0W, 5.0Mo, 4.0Cr     ~0.5         Air quench

The Jominy end-quench test (ASTM A255) measures hardenability empirically:
  A standard bar (25.4mm × 101.6mm) is austenitised and end-quenched.
  Hardness is measured every 1.6mm (1/16") from the quenched end.
  Cooling rate at each position is known from standard charts.
  The resulting hardenability curve (H-band) is compared to the TTT diagram.

Grossmann H-factor (hardenability):
  H = α / (2λ)   (simplified)
  where α = heat transfer coefficient, λ = thermal conductivity
  H = 0.2: still water; H = 0.5: oil; H = 1.0: agitated oil; H = 5.0: agitated water

Industrial Applications: Austempering and Martempering

Austempering

Austempering is a heat treatment designed directly from the TTT diagram to produce a fully bainitic microstructure with superior toughness at equivalent hardness compared to tempered martensite, and with greatly reduced distortion compared to quench-and-temper.

Austempering procedure (design from TTT diagram):

Step 1 — Austenitise: Heat to A3 + 30°C (hypoeutectoid) or A1 + 30°C (hypereutectoid)
         Soak time: typically 30 min + 15 min per 25mm section thickness

Step 2 — Quench into bainite bath: Molten salt (KNO₃/NaNO₂) or hot oil
         Target temperature: within bainite field (250–450°C typically)
         Quench must be fast enough to miss the pearlite nose entirely
         → This limits austempering to section sizes below ~12–15mm for
           plain carbon steels; up to 50–75mm for alloy steels with
           C-curves pushed right by Mn, Mo, Ni additions

Step 3 — Hold isothermally until Bf (bainite finish):
         Read Bf time from TTT diagram + 25% safety margin
         Examples for 0.8%C: 250°C → ~60 min; 350°C → ~20 min; 450°C → ~5 min

Step 4 — Air cool to room temperature
         No martensite transformation occurs → no distortion, no quench cracking

Result: Lower bainite at 45–58 HRC → K_IC typically 40–70 MPa√m
        vs. tempered martensite at equivalent HRC → K_IC typically 20–50 MPa√m
        Superior fatigue resistance; minimal distortion; suitable for thin gears,
        springs, and precision components

Martempering (Marquenching)

Martempering eliminates the distortion caused by conventional direct quenching by equalising the temperature throughout the section before the martensite transformation begins. The steel is quenched into a bath held just above Ms, held until the core and surface reach the same temperature (within the “bay” where neither pearlite nor bainite is forming rapidly), then air-cooled. Martensite forms uniformly through the entire section during the slow air cool rather than progressively from the surface inward, dramatically reducing the differential volume changes that cause warping and cracking.

Martempering procedure:

Step 1 — Austenitise: Same as conventional quench-and-temper

Step 2 — Quench into marquench bath:
         Temperature: Ms + 20°C to Ms + 50°C  (above Ms; in the bay)
         Medium: Molten salt (KNO₃/NaNO₂), hot oil, or hot polymer
         Hold time: sufficient to equalise temperature through section
                    (~1 min per 25mm section for salt; read from TTT Bs curve
                    to ensure bay is maintained — must not form bainite)

Step 3 — Air cool to room temperature:
         Martensite forms uniformly throughout section during this step
         Minimal thermal gradient → minimal distortion vs. direct water/oil quench

Step 4 — Temper immediately (standard Q&T temper cycle)

Limitation: Only works if the steel composition provides a bay (gap between
            pearlite and bainite C-curves). Mo is the most effective element
            for creating this bay. 4140, 4340, 52100 are all routinely martempered.

TTT vs CCT Diagrams: The Practical Distinction

The TTT diagram is constructed under perfectly isothermal conditions — a theoretical ideal that is never exactly achieved in industrial practice, where parts cool continuously from austenitising temperature to ambient. The CCT (Continuous Cooling Transformation) diagram is the practical counterpart: it maps the transformation products formed at different constant cooling rates through the transformation temperature range.

For designing quench processes, selecting quenchant media, or predicting microstructure in different positions of a large forging, the CCT diagram is the more directly applicable tool. For the full treatment of CCT diagram construction and cooling curve overlay, see the CCT Diagram guide. For the underlying austenite microstructure from which all transformations begin, see the Austenite in Steel article.

Prior Austenite Grain Size: Effect on TTT Kinetics

Prior austenite grain size (PAGS) has a significant and often overlooked effect on TTT diagram position. Because pearlite and bainite both nucleate preferentially at austenite grain boundaries, the number of nucleation sites per unit volume is inversely proportional to grain size. A finer grain (higher ASTM number) means:

• More nucleation sites → higher nucleation rate → faster overall transformation → C-curves shift slightly to the left (shorter times).
• Worse hardenability: harder to avoid pearlite nose in a given section size.
• Better toughness: finer grain gives better impact resistance via the Hall-Petch relationship.

TTT diagrams published by atlas sources (ASM Boyer, Verein Deutscher Eisenhüttenleute) specify the grain size at which they were determined. Using a diagram determined at ASTM grain size 7 for a steel that was actually austenitised to ASTM 3 (coarser) will overestimate the speed of transformation and may lead to incorrect quench selection. For safety-critical parts, always verify the actual PAGS matches the conditions used to generate the TTT data. For more on grain size measurement and its mechanical significance, see Grain Boundaries.