Delta Ferrite in Stainless Steel Welds — WRC-1992 Diagram and Ferrite Number
Delta ferrite content in austenitic stainless steel weld metal sits at the intersection of solidification metallurgy, hot cracking prevention, and long-term elevated-temperature performance. Too little delta ferrite leaves the weld metal susceptible to solidification cracking; too much accelerates sigma phase embrittlement in service. The WRC-1992 constitution diagram and the Ferrite Number (FN) system are the practitioner's primary tools for navigating this balance — from filler metal selection through welding procedure qualification and in-service fitness-for-purpose assessment. This article develops the thermodynamic basis, the WRC-1992 equivalents and their limitations, solidification mode classification, and the practical measurement and control methods required by ASME, AWS, and European pressure equipment standards.
Key Takeaways
- Delta ferrite (δ-ferrite) is a BCC iron phase retained in austenitic stainless steel weld metal when the Creq/Nieq ratio exceeds approximately 1.48, preventing solidification hot cracking by suppressing low-melting grain boundary films of sulphide and phosphide impurities.
- The WRC-1992 diagram uses the equivalents Creq = Cr + Mo + 0.7Nb and Nieq = Ni + 35C + 20N + 0.25Cu, giving more accurate FN predictions than the earlier Schaeffler and DeLong diagrams, especially for nitrogen-bearing grades.
- A Ferrite Number (FN) of 3–8 is the widely accepted target for austenitic weld metal: sufficient to prevent hot cracking while limiting sigma phase formation in elevated-temperature service (600–900 °C exposure range).
- Four solidification modes are defined by Creq/Nieq: A (fully austenitic, hot-crack susceptible), AF, FA (recommended), and F (fully ferritic). FA mode is preferred for general stainless welding.
- FN is measured in production by calibrated magnetic instruments (Fischer Feritscope or Magne-Gage) traceable to AWS A4.2 / ISO 8249 reference blocks, not converted directly to volume percent above 10 FN.
- Dilution of austenitic filler by wrought base metal reduces FN; the actual weld deposit composition must be calculated from filler + base metal + dilution percentage and plotted on the WRC-1992 diagram to verify compliance.
WRC-1992 Ferrite Number Calculator
Enter weld deposit composition (wt%) — use nominal filler metal certificate values, adjusted for dilution if known.
Phase Diagram Basis: The Fe-Cr-Ni Ternary System
Delta ferrite in austenitic stainless steels is best understood by reference to the Fe-Cr-Ni ternary phase diagram at compositions near 18%Cr–8%Ni. In this region of the ternary, the liquidus surface contains two primary solidification fields: a ferrite (δ) field at higher Cr:Ni ratios and an austenite (γ) field at lower Cr:Ni ratios. The boundary between them passes near the 18-8 composition, and the exact position of a given weld metal relative to this boundary determines whether the alloy solidifies primarily as ferrite or primarily as austenite.
The key thermodynamic feature is that delta ferrite, stabilised at high temperature by Cr (a ferrite-stabilising BCC-former), is the equilibrium high-temperature phase above approximately 1350–1400 °C for compositions in the FA solidification field. As the weld metal cools, delta ferrite partially transforms to austenite via the peritectic and solid-state reactions. The fraction that does not transform is retained as residual delta ferrite at room temperature. Understanding the Iron-Carbon Phase Diagram and its extension into the ternary system provides the thermodynamic foundation for interpreting constitution diagrams.
Ferrite-Stabilising vs Austenite-Stabilising Elements
| Element | Type | Crystal structure effect | WRC-1992 coefficient | Relative potency |
|---|---|---|---|---|
| Chromium (Cr) | Ferrite-stabilising | Expands δ-ferrite loop, stabilises BCC | 1.00 (in Creq) | Reference |
| Molybdenum (Mo) | Ferrite-stabilising | Strong BCC stabiliser, isoelectronic with Cr | 1.00 (in Creq) | Equivalent to Cr |
| Niobium (Nb) | Ferrite-stabilising | Carbide former; stabilises ferrite lattice | 0.70 (in Creq) | 0.7× Cr |
| Silicon (Si) | Ferrite-stabilising | BCC stabiliser; not in WRC-1992 equivalents | — | Moderate |
| Titanium (Ti) | Ferrite-stabilising | Strong BCC stabiliser; not in WRC-1992 | — | Strong |
| Nickel (Ni) | Austenite-stabilising | Stabilises FCC γ-austenite | 1.00 (in Nieq) | Reference |
| Carbon (C) | Austenite-stabilising | Expands FCC; powerful γ-stabiliser | 35 (in Nieq) | 35× Ni |
| Nitrogen (N) | Austenite-stabilising | Interstitial; very strong FCC stabiliser | 20 (in Nieq) | 20× Ni |
| Manganese (Mn) | Austenite-stabilising | Stabilises γ, but weak (not in WRC-1992 Nieq) | — | Weak (<1×) |
| Copper (Cu) | Austenite-stabilising | Stabilises FCC; relatively weak | 0.25 (in Nieq) | 0.25× Ni |
Note that the DeLong diagram (1974) used a Nieq = Ni + 30C + 0.5Mn, which significantly underestimated nitrogen's contribution. For modern nitrogen-bearing austenitic grades (0.1–0.25% N) such as 316LN, 304LN, and duplex grades, the WRC-1992 coefficients give substantially more accurate FN predictions.
Solidification Modes: A, AF, FA, and F
The constitution diagram divides stainless steel weld metal compositions into four distinct solidification modes based on the Creq/Nieq ratio. Each mode has characteristic hot cracking susceptibility, residual delta ferrite level, and microstructural development sequence.
Mode A — Fully Austenitic
When Creq/Nieq is below approximately 1.25, the weld metal solidifies entirely as austenite with no primary ferrite. The solidification grain boundaries that form during cooling retain all the sulphur, phosphorus, and silicon that segregates during solidification, forming low-melting eutectic films (Ni–S eutectic at 645 °C; Ni–P eutectic at 880 °C). Solidification contraction during cooling from the liquidus exerts tensile stress on these boundaries; if the liquid films persist to low enough temperature, solidification (hot) cracking occurs. Fully austenitic welds (≤0 FN) are inherently susceptible and are used only where service conditions demand it (e.g., dissimilar metal welds into nickel-base alloys, cryogenic applications requiring maximum toughness).
Mode AF — Primary Austenite with Peritectic Ferrite
At Creq/Nieq of approximately 1.25–1.48, primary austenite solidifies and undergoes a peritectic reaction with residual liquid to produce a small amount of ferrite at grain boundary triple points. This partially disrupts the grain boundary liquid film network but provides limited protection against cracking. Weld metal in AF mode typically shows 1–3 FN.
Mode FA — Primary Ferrite with Austenite Transformation (Recommended)
At Creq/Nieq of approximately 1.48–1.95, the weld metal solidifies as primary delta ferrite. On cooling through the solid state, most of this ferrite transforms to austenite via the eutectoid-like reaction δ → γ, leaving a residual network of delta ferrite stringers (typically 3–12 FN) in an austenite matrix. Hot cracking susceptibility is low because: (a) impurity segregation occurs in the ferrite field where S and P solubility is higher; (b) the δ/γ transformation disrupts the grain boundary network; (c) shrinkage cracking is resisted by the two-phase microstructure. FA mode is the standard recommendation for most austenitic stainless steel welding.
For most structural and pressure vessel austenitic stainless steel applications, fillers are selected to produce 3–8 FN in the deposited (undiluted) condition. ASME BPVC Section II SFA-5.4 (SMAW) and SFA-5.9 (bare wire) certification records include certified FN values for this reason. ASME Section IX considers FN an essential variable for some P-number 8 (austenitic stainless) welding procedures.
Mode F — Fully Ferritic
Above Creq/Nieq of approximately 1.95, the weld metal remains largely ferritic with minimal austenite transformation. This regime applies to high-chromium ferritic stainless steels and the ferrite-rich side of duplex grades. Delta ferrite levels are very high (>50 FN or not measurable in the normal FN scale), and grain boundary embrittlement mechanisms shift from hot cracking to 475 °C embrittlement and sigma phase precipitation.
WRC-1992 Equivalents and Prediction Equations
The WRC-1992 diagram, developed by Kotecki and Siewert and published in Welding Journal in 1992, uses the following Cr and Ni equivalents:
WRC-1992 Chromium Equivalent:
Creq = Cr + Mo + 0.7×Nb (all in wt%)
WRC-1992 Nickel Equivalent:
Nieq = Ni + 35×C + 20×N + 0.25×Cu (all in wt%)
Ferrite Number (WRC-1992 empirical fit):
FN = 14.3×(Creq/Nieq) − 13.7 [valid approximately 0–20+ FN range]
Solidification mode boundaries (approximate):
Creq/Nieq < 1.25 : A mode (fully austenitic, FN = 0)
1.25 ≤ Creq/Nieq < 1.48 : AF mode (1–3 FN)
1.48 ≤ Creq/Nieq < 1.95 : FA mode (3–12 FN, recommended)
Creq/Nieq ≥ 1.95 : F mode (high FN, duplex/ferritic)
The WRC-1992 diagram was calibrated on a dataset of GTAW and SMAW welds. It does not include silicon in the Creq (silicon is a ferrite stabiliser; its omission slightly underestimates FN in high-Si deposits). Manganese is excluded from the Nieq (it is a weak austenite stabiliser; its omission slightly overestimates FN in high-Mn deposits). For highly alloyed grades, super-duplex, or compositions far outside the original calibration range, thermodynamic calculation software (Thermo-Calc CALPHAD approach) should supplement constitution diagram predictions.
Comparison: Schaeffler, DeLong, and WRC-1992
| Diagram | Year | Creq formula | Nieq formula | Key limitation |
|---|---|---|---|---|
| Schaeffler | 1949 | Cr + Mo + 1.5Si + 0.5Nb | Ni + 30C + 0.5Mn | No nitrogen term; predicts %FV not FN; poor accuracy for N-bearing grades |
| DeLong | 1974 | Cr + Mo + 1.5Si + 0.5Nb | Ni + 30C + 0.5Mn + 30N | N coefficient (30) too low for high-N grades; still predicts vol% not FN |
| WRC-1992 | 1992 | Cr + Mo + 0.7Nb | Ni + 35C + 20N + 0.25Cu | Omits Si (slight FN underestimate) and Mn (slight FN overestimate); most accurate for mainstream grades |
Hot Cracking Prevention: Mechanism in Detail
Solidification cracking in austenitic stainless steel welds is the primary practical reason for controlling delta ferrite. The mechanism is well characterised and explains why FA solidification mode is effective:
The Impurity Film Mechanism (Fully Austenitic Welds)
During solidification of an austenitic stainless steel, elements with strong partition coefficients concentrate in the last liquid to solidify. Sulphur, phosphorus, and silicon all partition strongly to the liquid, building up at the solidification cell and dendrite boundaries. When sufficient concentration is reached, these form low-melting compounds:
Critical low-melting films in austenitic stainless welds:
Ni-S eutectic: m.p. ≈ 645 °C (catastrophically low in Ni-rich austenite)
Ni-P eutectic: m.p. ≈ 880 °C
Ni-Si eutectic: m.p. ≈ 1143 °C (less critical but contributes)
Solidification temperature range (typical 304/308):
Liquidus ≈ 1455 °C
Solidus ≈ 1400 °C (A mode) / 1420 °C (FA mode)
BTR (Brittle Temperature Range) where cracking occurs:
A mode: film still liquid down to ~1100–1200 °C
FA mode: impurities dispersed; cracking window greatly reduced
Why FA Mode Prevents Cracking
In FA mode, primary solidification as delta ferrite provides two protective effects. First, sulphur solubility in BCC ferrite (approximately 0.018 wt% S at 1000 °C) is substantially higher than in FCC austenite (approximately 0.003 wt% S), so sulphur remains dissolved in the solid ferrite rather than enriching at boundaries. Second, as the solid-state δ → γ transformation proceeds on cooling, the grain boundaries of the as-solidified ferrite structure are disrupted and replaced by the austenite grain boundaries of the new phase — which contain no accumulated S or P films. The result is a microstructure without the critical low-melting boundary films that enable crack propagation.
Some service conditions mandate fully austenitic weld metal (0 FN): dissimilar metal welds between austenitic stainless and nickel-base alloys (where delta ferrite can provide a crack propagation path under thermal cycling), cryogenic applications below −196 °C (delta ferrite undergoes BCC-to-martensite transformation on cooling, reducing toughness), and certain nuclear service specifications. In these cases, hot cracking risk must be managed by strict control of S + P + Si + Sn impurity levels in filler metal and base metal, and by welding technique (low heat input, stringer beads, no weaving).
Sigma Phase Formation from Delta Ferrite
The upper bound on acceptable delta ferrite content is primarily set by sigma phase embrittlement risk in elevated-temperature service. Sigma phase (σ) is a topologically close-packed (TCP) intermetallic compound with approximate composition (Fe54Cr36Mo10) that precipitates from delta ferrite in the temperature range 600–900 °C. Because delta ferrite is the Cr-enriched constituent in austenitic stainless weld metal, it is the thermodynamically preferred nucleation site for sigma phase.
Transformation Reaction and Kinetics
Sigma formation from delta ferrite (solid-state):
δ(BCC, Cr-rich) → σ(tetragonal, (Fe,Cr)) + γ'(FCC, Ni-enriched austenite)
Peak transformation rate: 700–800 °C
Incubation time at 700 °C: as short as 100–500 h for 10+ FN weld metal
>2000 h for 3–5 FN weld metal
Sigma phase hardness: 900–1200 HV (extremely hard and brittle)
Charpy energy after sigma formation: can drop from >100 J to <10 J
The practical implication is that weld metal with high delta ferrite (above 10–15 FN) is unsuitable for prolonged service in the sigma-forming temperature range, regardless of its room-temperature properties. ASME Code Case N-382 and related documents address sigma phase assessment for pressure vessel internals and piping. In heat exchanger and reactor vessel applications exposed to process temperatures of 600–800 °C, FN must be limited to 3–5 FN to provide acceptable sigma phase resistance.
By contrast, at ambient and low service temperatures, sigma phase risk is negligible and the upper FN limit of 8–10 FN may be relaxed for corrosion or cracking resistance reasons. This is the basis for the standard 3–8 FN target window: the lower bound is set by hot cracking prevention; the upper bound by sigma phase risk in moderate-temperature service.
Ferrite Number Measurement: Instruments and Standards
The distinction between Ferrite Number (FN) and ferrite volume percent (%FV) is important in practice. Below approximately 10 FN, FN and %FV are approximately numerically equal. Above 10 FN, the relationship becomes non-linear because FN is defined by magnetic response — a property that does not scale linearly with volume percent above moderate ferrite fractions. The FN scale is therefore an instrument calibration scale, not a direct physical quantity.
Fischer Feritscope
The Fischer Feritscope (and equivalent eddy-current/magnetic induction instruments from other manufacturers) is the standard instrument for production FN measurement. It operates by measuring the impedance change in a high-frequency coil placed on the sample surface; the magnetic ferrite content changes the coil response, which is converted to FN via a calibration curve traceable to AWS A4.2 (USA) or ISO 8249 (international) certified reference blocks. Measurement procedure:
- Calibrate with certified reference blocks spanning the expected FN range before each measurement session.
- Ensure the measurement surface is smooth, clean, and free from oxide scale; light grinding with 120-grit paper is acceptable.
- Take a minimum of 5 readings per weld sample, repositioning the probe between readings.
- Report the arithmetic mean ± standard deviation; discard readings that differ from the mean by more than 2 FN (single outlier).
- The probe must be held perpendicular to the surface; curvature and proximity to edges or weld toes introduce error.
Magne-Gage (Balance Instrument)
The Magne-Gage uses a calibrated spring balance to measure the attractive force between a small permanent magnet and the ferromagnetic delta ferrite phase. It is highly accurate at low FN (<5 FN) where the Feritscope response can be less sensitive, making it the preferred instrument for qualifying very low-ferrite weld procedures (cryogenic and nuclear applications with FN ≤3 requirements). Calibration per ANSI/AWS A4.2 is required.
Metallographic Point Counting
ASTM E562 systematic manual point counting or image analysis of colour-etched cross-sections (Beraha's reagent or electrolytic oxalic acid etch differentiates austenite from ferrite by colour) gives volume percent ferrite directly. This method is used to confirm instrument measurements in procedure qualification records, or where instrument access is limited. It requires proper sectioning through the weld metal cross-section to avoid sampling bias.
Controlling Delta Ferrite in Production Welding
Filler Metal Selection and Certification
For common austenitic grades, filler metals per AWS A5.4 (SMAW), A5.9 (solid wire/strip), A5.22 (flux-cored), and A5.30 (consumable inserts) are required to state the diffusible hydrogen level and, for austenitic grades, the certified FN on the test certificate. The certified FN is determined by testing undiluted pad deposits using the standard test method in AWS A4.2, not calculated from nominal composition. Purchasing specifications for pressure equipment should request certified FN in the range 3–8 FN (or narrower as required) from the electrode manufacturer.
| AWS Filler Grade | Base metal application | Typical undiluted FN | Solidification mode | Notes |
|---|---|---|---|---|
| 308L | 304/304L | 5–9 FN | FA | Standard general-purpose; low-carbon to prevent sensitisation |
| 316L | 316/316L | 5–10 FN | FA | Mo addition for pitting resistance; check FN with high Mo |
| 309L | Dissimilar (C-steel to SS) | 8–15 FN | FA / F | Higher Cr, higher FN; dilution critical in transition layer |
| 347 | 321/347 (stabilised) | 4–9 FN | FA | Nb stabilises against sensitisation; check Nb in Creq |
| 310 | 310S (high-temp) | 0 FN | A | Fully austenitic; strict S+P control required; hot crack risk |
| 312 | Hard-facing, dissimilar | 20–40 FN | F | Very high Cr/Fe ratio; used as buffer layer in dissimilar welds |
Dilution Calculation
The actual weld deposit composition must account for dilution from the base metal. For a two-component system (filler + base metal), each alloying element in the weld metal is calculated as:
Weld deposit composition (for element X):
X_weld = X_filler × (1 − D) + X_base × D
where:
D = dilution fraction (decimal; 0.25 = 25% dilution)
X_filler = element content in undiluted filler (wt%)
X_base = element content in base metal (wt%)
Typical process dilution (single-pass butt welds):
GTAW (autogenous / fused): 30–50%
GTAW (with filler): 15–25%
SMAW: 20–30%
GMAW (spray transfer): 25–35%
SAW: 30–50%
After computing X_weld for all elements, calculate Creq and Nieq
and plot on WRC-1992 diagram to verify FN compliance.
This calculation is required when qualifying welding procedures under ASME Section IX for P-No. 8 austenitic stainless materials with FN requirements, and when evaluating overlay weld metal (cladding) where base metal dilution in the first layer is particularly high. See also the companion guide on HAZ Microstructure for the base metal side of the stainless steel weld cross-section.
Special Cases: Dissimilar Metal Welds and Cladding
In dissimilar metal welds between carbon or low-alloy steel (typically ASME P-No. 1 or 4) and austenitic stainless steel, the carbon and low-alloy base metal provides a very different dilutant composition. Carbon from the base metal migrates into the weld metal near the fusion line (carbon migration under the chemical potential gradient), and iron dilution reduces both Creq and Nieq. This can drive the near-fusion-line weld metal into the A or AF solidification mode even when the filler metal is specified to produce 8 FN undiluted. Buffer layers (309L or 312) are used to manage this transition. Refer to the Hydrogen-Induced Cracking guide for the concurrent risk of cold cracking in the heat-affected zone of the carbon steel side.
Service Performance: Corrosion, Toughness, and Elevated Temperature
Corrosion Resistance
Delta ferrite in austenitic weld metal is typically enriched in chromium relative to the surrounding austenite. In the as-welded condition, this Cr-enriched ferrite can actually improve localised corrosion resistance in some environments. However, heat treatment or in-service exposure in the sensitisation range (400–800 °C) promotes chromium carbide precipitation at the ferrite-austenite interface, depleting the adjacent region of chromium below the 12% threshold for passivity. This sensitisation renders the weld metal susceptible to intergranular stress corrosion cracking in chloride or other oxidising environments. Low-carbon (L-grade) fillers and solution annealing after welding address this risk. For highly corrosive service, also consider the Corrosion Mechanisms and Pitting Corrosion guides.
Cryogenic Toughness
Delta ferrite has a BCC crystal structure and is therefore subject to the ductile-to-brittle transition characteristic of all BCC metals. At cryogenic temperatures (below approximately −100 °C), delta ferrite in austenitic weld metal loses toughness and can provide crack initiation sites under impact loading. For LNG (liquefied natural gas, −163 °C), liquid nitrogen (−196 °C), or liquid hydrogen (−253 °C) service, specifications frequently restrict FN to ≤3 FN or require fully austenitic (0 FN) weld metal, accepting the hot crack risk and managing it through impurity control.
Elevated Temperature and Creep
In the creep range (above approximately 550 °C for austenitic stainless), delta ferrite transforms progressively to sigma phase, carbides, and secondary austenite. As discussed in Section 6, sigma phase formation causes severe embrittlement. Long-term elevated-temperature service is therefore the primary driver for the upper FN limit in pressure vessel and petrochemical applications.
Frequently Asked Questions
What is delta ferrite and why does it form in stainless steel weld metal?
Delta ferrite (δ-ferrite) is a body-centred cubic (BCC) iron phase that is stable at high temperatures in austenitic and duplex stainless steels. In austenitic stainless steel weld metal, it forms as a primary solidification phase when the Creq/Nieq ratio exceeds approximately 1.48 (FA or F solidification mode). Rather than fully transforming to austenite on cooling, a residual fraction of delta ferrite is retained at room temperature as elongated stringers in the austenite matrix. Its formation is controlled by the balance of ferrite-stabilising elements (Cr, Mo, Si, Nb, Ti) against austenite-stabilising elements (Ni, Mn, N, C) in the weld metal composition, as predicted by the WRC-1992 constitution diagram.
What is the WRC-1992 diagram and how does it differ from the Schaeffler diagram?
The WRC-1992 (Welding Research Council 1992) diagram uses revised Cr and Ni equivalents (Creq = Cr + Mo + 0.7Nb; Nieq = Ni + 35C + 20N + 0.25Cu) that more accurately predict delta ferrite in fully austenitic and low-ferrite compositions, particularly for nitrogen-bearing grades where nitrogen's coefficient (20 in WRC-1992 vs 30 in DeLong vs absent in Schaeffler) is critical. The Schaeffler diagram (1949) lacks a nitrogen term entirely, significantly overestimating FN in modern nitrogen-bearing grades. WRC-1992 expresses results in Ferrite Numbers (FN) calibrated to magnetic instruments per AWS A4.2, replacing the volume percent ferrite scale of the Schaeffler diagram. For standard austenitic grades (304, 316, 308L, 316L) without high nitrogen, all three diagrams give broadly similar predictions; for 304LN, 316LN, and duplex grades, WRC-1992 is required.
What Ferrite Number range is recommended for austenitic stainless steel weld metal?
For most austenitic stainless steel weld metal, 3–8 FN is the widely accepted target range, specifying sufficient delta ferrite to prevent solidification hot cracking (FA solidification mode) while limiting sigma phase embrittlement risk in service at 600–900 °C. Specific applications adjust this window: cryogenic service below −100 °C typically requires ≤3–5 FN to maintain impact toughness of the BCC delta ferrite; petrochemical high-temperature service above 600 °C may specify ≤3 FN to minimise sigma phase; corrosion-resistant applications may specify 3–6 FN. ASME BPVC and AWS standards require the certified FN to be reported by electrode manufacturers on test certificates for austenitic grades.
What are the four solidification modes in stainless steel weld metal?
The four modes are classified by Creq/Nieq ratio: Mode A (fully austenitic, Creq/Nieq <1.25, 0 FN, highest hot cracking susceptibility), Mode AF (primary austenite with peritectic ferrite, Creq/Nieq 1.25–1.48, 1–3 FN, moderate risk), Mode FA (primary ferrite transforming partially to austenite, Creq/Nieq 1.48–1.95, 3–12 FN, low hot cracking risk, recommended for general use), and Mode F (fully ferritic, Creq/Nieq >1.95, very high FN, applicable to ferritic and duplex grades). Most austenitic stainless filler metals are formulated to produce FA mode solidification in undiluted deposits.
Why does delta ferrite prevent hot cracking in austenitic stainless steel welds?
Delta ferrite prevents solidification hot cracking through two mechanisms. First, sulphur and phosphorus have approximately six times higher solubility in BCC delta ferrite than in FCC austenite; when the weld solidifies in FA mode (primary ferrite), these impurities remain distributed in the solid rather than segregating to solidification boundaries to form low-melting eutectic films (Ni–S eutectic at 645 °C, Ni–P eutectic at 880 °C). Second, the solid-state δ → γ transformation disrupts the continuous solidification grain boundary network and the boundary planes rotate, eliminating the long-range connectivity of any boundary films that might have formed. Together, these effects prevent the sustained liquid film at boundaries that is the physical prerequisite for solidification cracking under contraction stress.
What is sigma phase and when does it form from delta ferrite?
Sigma phase is a hard (900–1200 HV), brittle topologically close-packed intermetallic compound with approximate composition (Fe,Cr)x(Ni,Mo)y that forms from delta ferrite in the temperature range 600–900 °C (peak rate near 700–800 °C). Delta ferrite, being enriched in chromium and molybdenum, is the thermodynamically preferred nucleation site. The transformation reaction δ → σ + γ' produces sigma phase at the ferrite-austenite interface. Sigma phase formation causes severe impact embrittlement (Charpy energy can fall below 10 J from >100 J), sensitises the microstructure to intergranular corrosion, and is irreversible without full solution annealing above 1000 °C. It is the primary reason delta ferrite content is limited to ≤8–10 FN in austenitic stainless weld metal intended for elevated-temperature service.
How is Ferrite Number measured in production welds?
Ferrite Number is measured using calibrated magnetic instruments, primarily the Fischer Feritscope (eddy-current/magnetic induction) and the Magne-Gage (spring balance). The Feritscope is the most common production instrument; it must be calibrated with AWS A4.2 or ISO 8249 certified reference blocks at the start of each measurement session. At least 5 readings per location should be taken and averaged; outliers more than 2 FN from the mean are discarded. The measurement surface must be clean and smooth. FN is not directly equal to volume percent ferrite above approximately 10 FN, and the two scales must not be interchanged without correction. Metallographic point counting per ASTM E562 gives volume percent ferrite directly and is used for procedure qualification confirmation.
What is the effect of dilution on delta ferrite content in stainless steel welds?
Dilution of filler metal with wrought austenitic base metal shifts the deposit composition toward the base metal composition, which is always fully austenitic (near 0 FN on the WRC-1992 diagram). Since fillers are formulated to produce 3–8 FN undiluted, significant dilution reduces the Creq/Nieq ratio and can move the composition toward the A or AF mode. For high-dilution processes (SAW or GMAW, 30–50% dilution), the weld metal composition must be calculated from filler + base metal + dilution and plotted on WRC-1992 to verify FN compliance. If the calculated FN falls below 3, a higher-FN filler or lower-dilution technique is required. 309L (high Cr) is often used for the first layer of stainless cladding precisely because its high Creq ensures adequate FN even at 40–50% dilution into carbon steel base.
How does cooling rate affect delta ferrite content in stainless steel weld metal?
Unlike in carbon steel, delta ferrite content in austenitic stainless weld metal is primarily composition-dependent rather than cooling-rate-dependent. However, faster cooling reduces the time available for the solid-state δ → γ transformation, retaining more delta ferrite than equilibrium predictions. High heat input (slow cooling) allows more complete transformation, slightly reducing FN. For conventional arc welding processes, this effect is typically 1–3 FN. For rapid processes such as laser or electron beam welding (cooling rates >100 °C/s), retained FN can be 2–5 FN higher than arc welding of the same composition. This effect must be considered when qualifying autogenous laser weld procedures on compositions that are close to the lower FN boundary.
Recommended Reference Books
Welding Metallurgy of Stainless Steels — Folkhard
Comprehensive reference on the physical metallurgy of all stainless steel families, solidification modes, delta ferrite control, and sigma phase embrittlement in welded structures. Essential for welding engineers.
View on AmazonASM Handbook Vol. 6: Welding, Brazing and Soldering
The industry-standard welding reference including detailed coverage of stainless steel welding metallurgy, constitution diagrams, ferrite control, and procedure qualification requirements.
View on AmazonStainless Steels — Lula (ASM International)
Thorough treatment of austenitic, ferritic, martensitic, duplex, and precipitation-hardening grades covering metallurgy, corrosion, and fabrication including welding procedure guidance and delta ferrite implications.
View on AmazonCorrosion of Stainless Steels — Sedriks (2nd Ed.)
Definitive reference on corrosion behaviour of all stainless steel families: sensitisation, stress corrosion cracking, pitting, crevice corrosion, and the role of weld microstructure including delta ferrite in corrosion performance.
View on AmazonDisclosure: MetallurgyZone participates in the Amazon Associates programme. If you purchase through these links, we may earn a small commission at no extra cost to you. This helps support free technical content on this site.
Further Reading & Related Topics
References
- Kotecki, D.J. and Siewert, T.A., “WRC-1992 Constitution Diagram for Stainless Steel Weld Metals: A Modification of the WRC-1988 Diagram,” Welding Journal, 71(5), 171-s–178-s, 1992.
- Folkhard, E., Welding Metallurgy of Stainless Steels. Springer-Verlag, 1988.
- AWS A4.2M/A4.2: Standard Procedures for Calibrating Magnetic Instruments to Measure the Delta Ferrite Content of Austenitic and Duplex Ferritic-Austenitic Stainless Steel Weld Metal. American Welding Society.
- ISO 8249: Welding — Determination of Ferrite Number (FN) in Austenitic and Duplex Ferritic-Austenitic Cr-Ni Stainless Steel Weld Metals. ISO Geneva.
- Lippold, J.C. and Kotecki, D.J., Welding Metallurgy and Weldability of Stainless Steels. Wiley-Interscience, 2005.
- Sedriks, A.J., Corrosion of Stainless Steels. 2nd ed. Wiley-Interscience, 1996.
- ASM Handbook Vol. 6: Welding, Brazing, and Soldering. ASM International, 1993.
- ASME BPVC Section IX, QW-404: Welding Variables for Austenitic Stainless Steel, current edition.
