​‌​​‌​

Hastelloy C-22 / (2.4602) / UNS N06022 / AMS 5766

​‌​​‌​

Hastelloy C-22 is known as the second most widely used and most readily available of the Hastelloy materials. It is a nickel alloy formed from a combination of high proportions of nickel, molybdenum, chromium and tungsten. The material is supplied as bar, sheet, tube, welding wire and electrode. It belongs to the austenitic group of metals and has excellent corrosion resistance both at room temperature and at high temperatures.

Alloy C22 generally contains around 22% chromium. It also contains around 14% molybdenum and around 3% tungsten. The chromium in the material keeps it resistant to acids such as nitric acid, while the molybdenum and tungsten allow it to last for long periods in environments such as sulphuric acid and hydrochloric acid. Formed largely from nickel, this nickel alloy can also operate in many different environments even at high temperatures. Thanks to these superior properties it is used to protect the steel tubes and other components of coal-fired boilers and of waste-to-energy boilers.​‌​​‌​

It is widely used in many different components in pressure vessels where temperature and pressure are present, in plants carrying out chemical production processes, in filters and pollution control equipment, in marine components, in paper production and in water treatment plants. Resistant to acids such as sulphuric acid and hydrochloric acid, it is a very widely used Hastelloy grade.

What Hastelloy C-22 Is — and Where It Really Stands Against C-276​‌​​‌​

​‌​​‌​

Hastelloy C-22 (UNS N06022 / W.Nr. 2.4602 / EN-DIN NiCr21Mo14W / ISO NiCr21Mo14W3) is a single-phase FCC austenitic, solid-solution strengthened Ni-Cr-Mo-W alloy; it is not precipitation hardenable. Nominal Ni 56 · Cr 22 · Mo 13 · W 3 · Fe 3. The trade names describe the same material: HASTELLOY® C-22®, VDM® Alloy 22 / Nicrofer 5621 hMoW, INCONEL® alloy 22, ATI 22™. The cast counterpart is ASTM A494 grade CX2MW, and that is a SEPARATE UNS number (N26022) — not the same material.

One-line identity: C-22 is a deliberate re-balance of C-276 — Cr raised from 16 to 22 %, Mo lowered from 16 to 13 %, W from 4 to 3 %. Chromium buys the passive film that survives oxidising chloride and wet chlorine; molybdenum and tungsten buy reducing acid resistance. C-22 is not an upgrade, it is a TRADE. Carbon and silicon are also held deliberately low (C ≤0.015 %, Si ≤0.08 %) to suppress grain-boundary carbide and silicide precipitation in the weld HAZ — this is what makes the alloy usable as welded. Every figure in the table below is from the originator’s own brochures and the SAME test suite; cross-alloy numbers from mixed publishers are not comparable.​‌​​‌​

Saturated wet chlorine — C-22’s clearest win​‌​​‌​

The originator’s test: 3000 ppm Cl⁻, pH 1.5, Cl₂ gas bubbled through, 30 days, final pH 0.9. The results run one way: at the liquid/vapour interface at 65 °C, C-276 suffered severe weld-metal attack while C-22 showed no attack at 40× magnification; in liquid at 80 °C, C-276 showed severe weld corrosion plus intergranular HAZ attack, C-22 gave 1.5 mpy and zero localised attack; in vapour at 65 °C C-276 pitted, C-22 gave 2.5 mpy. Because C-276 had already failed lower, the 95 °C test was never run on it; C-22 gave 0.9 mpy there. Chlorine, hypochlorite, chlorine dioxide and bleach stages are C-22’s real market. The same logic applies to oxidising acid-chloride mixtures: in boiling Green Death (11.5 % H₂SO₄ + 1.2 % HCl + 1 % FeCl₃ + 1 % CuCl₂), the legacy datasheet gives C-22 3 mpy · C-276 42 mpy · C-4 837 mpy · alloy 625 1815 mpy — a 14× margin over C-276 in the same test; and in 5 % HNO₃ + 25 % H₂SO₄ + 4 % NaCl, boiling: 12 mpy (0.30 mm/y), where neither a stainless nor C-276 survives. If an oxidising species is present together with chloride, C-22 is the right member of the family.

C-22’s honest ceiling — where alloy 59 and alloy 686 are ahead​‌​​‌​

C-276 beats C-22 in hot reducing acid (the rows above); it is also the older grade, widely stocked, and the one listed in ASTM F467 — N06022 is NOT on that list. But the real ceiling lies elsewhere: alloy 59 (N06059) and alloy 686 (N06686) beat C-22 decisively on thermal stability and on the crevice resistance of a weldment.

Standards by Product Form

​‌​​‌​

​‌​​‌​

Product Forms With NO Standard — the Commercially Valuable Section

This is the section your sales engineers should memorise: when a customer asks for “C-22 wire to ASTM” or “code-certified C-22 bolting”, the honest answer always comes from the same place.​‌​​‌​

Chemical Composition​‌​​‌​

ASTM B575 / B574 limits (wt %), the same table carried into B619 / B622 / B626 / B366 / B462 / B564: Ni balance (~56 nominal) · Cr 20.0–22.5 · Mo 12.5–14.5 · W 2.5–3.5 · Fe 2.0–6.0 · Co ≤2.50 · Mn ≤0.50 · V ≤0.35 · Si ≤0.08 · C ≤0.015 · S ≤0.02 · P ≤0.02. Sources disagree on phosphorus (≤0.02 % in three publishers, ≤0.025 % in two); copper ≤0.5 % appears only in the originator’s nominal listing and is ABSENT from the reproduced ASTM tables — do not guarantee a Cu limit. Nominal is a TARGET chemistry; the specification is the bands above — which is why a heat at 5.2 % Fe is fully conforming and “iron is limited to 3 %” is false.

Mechanical Properties

​‌​​‌​

Physical Properties​‌​​‌​

The source divergences in this section are real and have not been hidden: publishers do not agree on density, melting range, thermal conductivity, resistivity or specific heat. For weight calculations and thermal design, use the figure from the mill that supplied the material.

Heat Treatment and Thermal Stability

​‌​​‌​

​‌​​‌​

C-22 has exactly ONE heat treatment: a full solution anneal plus rapid quench. This alloy has NO stress-relief treatment. C-22 is a metastable solid solution; everything above 12 % Mo wants to precipitate out on cooling. The purpose of the anneal is to FREEZE the supersaturated solid solution; slow cooling recreates the very problem the anneal was performed to remove.

Welding

C-22 is designed to be used as welded — not as a slogan but a direct consequence of the chemistry: C ≤0.015 % and Si ≤0.08 % are held low so the HAZ does not sensitise on a normal weld thermal cycle. But the claim is narrow, and below we state exactly where it ends.​‌​​‌​

What actually goes wrong​‌​​‌​

1. Hot cracking — the dominant weld defect mode. The originator: “the most common type of weld cracking encountered is HOT CRACKING, which is associated with the presence of liquid in the microstructure” and “weldments with increased joint thickness are more susceptible“; also “large CONCAVE weld beads that place the weld surface in tension tend to promote solidification cracking” — which is why the convex-bead rule exists.
2. Sluggish weld pool. “Ni- and Co-base molten weld metal is comparatively ‘sluggish’ … not as fluid.” Consequence: joint preparations must be OPENED UP relative to stainless practice — wider included angle, larger root gap; otherwise lack of fusion is inevitable.
3. Dilution — the classic C-22 overlay failure. Overlaying onto carbon steel or 6Mo stainless dilutes Cr and Mo in the first layer below the alloy’s own limits. Mill overlay data on 6Mo, Green Death CPT: alloy 59 >85 °C · alloy 22 >85 °C · C-276 only 55 °C — the C-22 overlay holds up, but the test was run at a qualified dilution level. Specify TWO LAYERS or require a measured Fe content on the wetted surface.
4. Molybdenum micro-segregation into interdendritic regions is the seed for TCP precipitation.
5. DO NOT claim a hot-cracking RANKING. Alloy 59 has been published as showing lower sensitivity than C-276 and C-4 in Modified Varestraint testing; C-22 does not appear in the published ranking — it could not be independently verified.

The “as-welded” claim — what it covers and what it does not​‌​​‌​

The published claim is narrow and should be quoted narrowly: “HASTELLOY® C-22® alloy exhibits excellent resistance to the key inorganic acids, even in welded form.” The measured penalties, however, are real (weld metal / wrought base metal at 66 °C, mm/y): 30 % H₂SO₄ 0.02 / 0.01 · 50 % H₂SO₄ 0.24 / 0.02 — 12× WORSE · 70 % 0.26 / 0.28 — equal · 90 % 0.47 / 0.34; at 38 °C, 15 % HCl 0.28 / 0.24 · 20 % HCl 0.26 / 0.20. In ASTM G28: G28A 64.3 / 30.0 — 2.1× and G28B 14.2 / 4.5 mpy — 3.2×.
What the claim does NOT cover, all four measured: (1) crevices at a weldment in heavy chloride — at 70 000 ppm Cl⁻ / pH 1 / 105 °C the alloy 22 weldment suffered crevice attack at 0.44 mm/y while alloy 59 did not; (2) uniform rates in strong reducing acid — the 12× penalty above; (3) any material that has seen 538–816 °C AFTER welding (PWHT, adjacent passes, fire, hot service); (4) high heat-input welds — the claim is PREDICATED on heat input ≤8 kJ/cm and interpass ≤93–120 °C. Exceed those limits and the claim lapses with them.

Machining​‌​​‌​

A caveat that must be stated openly: the speed/feed table the originator publishes is for the corrosion-resistant alloy family, not specifically for C-22, and it dates from the uncoated-carbide and HSS era. Both it and modern coated-carbide figures are below — print both and NEVER average them.

Corrosion — Where It Wins and WHERE IT FAILS

PREN, and why it is not a design criterion here. There is no nitrogen in N06022. Using PREN = %Cr + 3.3 × (%Mo + 0.5 × %W): 65.4 at the ASTM minimum, 76.1 at the maximum, ≈70 at nominal. One publisher prints 65.38–76.13, exactly reproducing this formula; but the same publisher gives a second column headed “PREN” with alloy 22 = 46, C-276 = 45, alloy 59 = 47, alloy 686 = 51 — evidently a different formula. DO NOT print a bare PREN number without the formula. More importantly: PREN is a ranking index for stainless steels; for Ni-Cr-Mo alloys it correlates poorly with measured CPT/CCT and is not a design criterion.​‌​​‌​

Frequently Asked Questions​‌​​‌​

Our EPC specified C-276. Can we substitute C-22 and save money?

Only after you identify which medium governs. The two alloys are not ranked but traded — and both directions have been measured on the same test suite.
C-22 is decisively better in oxidising chloride: crevice temperature in acidified 6 % FeCl₃ 80 vs 55 °C; Yellow Death CCT 75 vs 60 °C; boiling Green Death 3 vs 42 mpy — a 14× margin; and in saturated wet chlorine at pH 0.9 C-276 weldments were severely attacked at 65 °C while C-22 showed no localised attack at 95 °C.
C-276 is decisively better in hot reducing acid: boiling 50 % H₂SO₄ 3.64 vs 9.98, boiling 5 % HCl 3.63 vs 8.99 mm/y — roughly 2.5–2.7× better, because the higher Mo and W are what matter there and the extra chromium does nothing. Phosphoric reverses again (0.66 vs 1.68 in boiling 85 %).
Decision rule: if your service is a chlorinated oxidiser, a bleach or chlorine dioxide stage, an FGD scrubber or a mixed nitric-plus-chloride stream, C-22 is the correct substitution. If your governing case is hot HCl or H₂SO₄ — including a reducing scenario that only occurs during upsets — do not substitute. Check the code route: both are 677 °C in ASME Sec. VIII Div. 1, but if the specification invokes ASTM F467 fasteners, N10276 is listed and N06022 is NOT. Get the substitution approved against the governing medium, in writing.​‌​​‌​

The fabricator wants to stress-relieve our C-22 vessel after welding. Is that acceptable?

No — and it is one of the most damaging things that can be done to the alloy. C-22 is solid-solution strengthened; there is no beneficial precipitation to develop and no tempering to perform. The originator states explicitly that for these alloys “PWHT in the 538 to 816 °C (1000 to 1500 °F) temperature range should be avoided”, because that range precipitates secondary phases with a detrimental effect on corrosion resistance.
The mechanism is μ phase. Ageing studies across 593–760 °C for up to 16 000 hours found μ phase forming as a continuous film on the grain boundaries at 593 °C and covering all grain boundaries plus the bulk matrix from 649 to 760 °C, with grain-boundary carbides and the Ni₂(Cr,Mo) ordered phase. The corrosion consequence is quantified: ASTM G28 Method A rises from roughly 30–35 mpy mill-annealed to 872 mpy after sensitisation at 871 °C. That is a 25-fold degradation and it is INVISIBLE — the vessel looks identical and passes hydrotest.
Any “stress relief at 590–650 °C for 1–2 h” on a distributor datasheet sits inside the μ-phase window and is supported by NO mill publisher. The alloy has exactly one heat treatment: a full solution anneal at 1105–1135 °C followed by rapid quenching. If residual stress genuinely must be removed, the whole fabrication has to be re-annealed and re-quenched, usually impractical — which is precisely why C-22 is designed to be used as welded: preheat at ambient, interpass ≤93–120 °C, heat input ≤8 kJ/cm.​‌​​‌​

The mill test certificate shows carbon at 0.013 %. Our German client rejected it. Who is right?

Both — and this is a standards divergence rather than a quality problem. ASTM B575 and B574 permit carbon up to 0.015 %, as reproduced by three independent publishers. The German/EU route — VdTÜV Werkstoffblatt 479 for 2.4602, the basis of PED compliance — limits carbon to 0.010 %. So a heat at 0.013 % is a fully conforming ASTM heat and a non-conforming VdTÜV heat. The divergence is not accidental: lower carbon further suppresses grain-boundary carbide precipitation in the weld HAZ — the basis of the as-welded corrosion claim.
Three more divergences catch buyers the same way: phosphorus (ASTM 0.02 % max versus 0.025 % — read the invoked edition); tensile, where ASTM gives a minimum only (≥690 MPa) but VdTÜV gives a RANGE (690–950 MPa), so an unusually strong heat can pass ASTM and fail the European route; and HB ≤240 plus Rp1.0 ≥335 MPa, which ASTM does not have.
The remedy is commercial, not technical: for any order that may end up in a PED-scope vessel, specify dual certification to ASTM B575/B574 AND VdTÜV WB 479 (edition 2021-11-30) at enquiry stage, with EN 10204 3.1 or 3.2 as required, and make the mill confirm the tighter limits BEFORE the heat is melted.​‌​​‌​

Common datasheet errors — check these before you order

1. “The CCT of C-22 is 102 °C.” Only in one legacy datasheet, and in an unspecified “NaCl-HCl solution”. The current brochure publishes 80 °C in acidified 6 % FeCl₃; another mill 75 °C in ASTM G48 C and D; a distributor 58 °C in 10 % FeCl₃ and 105 °C in Green Death. Six numbers, five solutions — never publish a crevice temperature without its test solution. The same confusion exists between tests: “Green Death” and “ASTM G28 Method B” are NOT the same thing (Green Death is 11.5 % H₂SO₄ + 1.2 % HCl + 1 % FeCl₃ + 1 % CuCl₂; another mill’s crevice test uses 11.9 / 1.3 / 1 / 1 and labels G28 B “Modified Green Death”; G28 B is a more concentrated solution again).
2. “G28A = 872 mpy, so C-22 has poor intergranular resistance.” That figure is for material deliberately sensitised at 871 °C; mill-annealed C-22 is 30–35 mpy. The converse is equally true: “G28A = 30 mpy” published without the condition is just as misleading for anyone assessing a fabrication that has seen heat. Always print the condition. Note also that in one distributor table the G28 rows are shifted by one alloy; the mill paper assigns alloy 22 = 872 / 17 and C-276 = >500 / 339.
3. Do not use aggregator databases or AI-generated distributor pages. One aggregator publishes density 8.9 g/cm³ (against every mill’s 8.61–8.70) and modulus 220 GPa (against 206–209) — modelled, not measured; one publisher’s 11.1 × 10⁻⁶ /K expansion is likewise a lone outlier (the other five give 12.4). A widely indexed AI-written page publishes “50 % H₂SO₄ boiling: C-22 = 0.8 mm/y” against the mill’s 9.98 — order-of-magnitude errors, all in the optimistic direction; another puts tungsten’s range under manganese and swaps sulphur with silicon. Verify every corrosion rate against a mill brochure.
4. “Iron is strictly limited to less than 3 %” and “molybdenum 14 %” — BOTH FALSE. ASTM specifies Fe 2.0–6.0 %; 3 % is the nominal, not a limit, and a heat at 5.2 % Fe is fully conforming. For molybdenum the nominal is 13 % and the range 12.5–14.5 %; the 14 % some pages print sits near C-276’s Mo band and blurs the one difference that matters.
5. “Stress relieve at 590–650 °C.” Found on distributor pages and supported by NO mill publisher; that window sits inside the μ-phase field. C-22 has NO stress-relief treatment — the only heat treatment is a 1105–1135 °C solution anneal plus rapid quench.
6. “Max code temperature 1250 °F” and “the alloy is good to 677 °C.” 1250 °F is true for ASME Sec. I and Sec. VIII Div. 1 and 2; it is 800 °F (427 °C) for B31.3, B31.1 and Sec. III Class 1 — publishing only 1250 °F misleads every piping enquiry. And that figure is a code allowable-stress ceiling, not a metallurgical stability statement: μ phase forms from 593 °C and the originator advises avoiding 538–816 °C altogether. Treat ~540 °C as the practical long-term ceiling.
7. Scope errors: B574, F467 and A494. B574 does not cover bar or wire — its scope is rod, 5/16 to 3½ in dia.; bar and billet go to B472, and there is no ASTM wire specification for N06022. There is no bolting specification either: F467 lists N10276, N04400, N04405, N05500, N06059, N06625 and N06686, not N06022; C-22 fasteners are machined from B574 rod and carry no Sec. II Part D bolting allowable. A494 CX2MW is UNS N26022 — a CAST alloy that cannot be certified as wrought N06022.
8. Five divergences between mills that must NEVER be averaged. Carbon: ASTM ≤0.015 % vs VdTÜV ≤0.010 % (VdTÜV also imposes 690–950 MPa, HB ≤240 and Rp1.0 ≥335 MPa). Air-cool threshold: 10 mm vs ~1.5 mm — 6.7×; demand a water quench above ~1.5 mm and have the quench medium stated on the MTC. Hot working: 1232 → 954 °C vs 1100 → 900 °C. Re-anneal after cold work: ~7 % outer-fibre elongation vs 15 % — and annealing BELOW ~7–10 % causes ABNORMAL GRAIN GROWTH. Interpass: 93 °C vs 120 °C, use 93 °C when both apply; and the originator gives NO numeric heat-input limit.
9. “As-welded corrosion resistance equals base metal”, and scale errors. The published claim is narrower, and the measured penalties are real: weld metal is 12× worse in 50 % H₂SO₄ at 66 °C, 2.1× worse in G28A, 3.2× worse in G28B, and an alloy 22 weldment suffered crevice attack at 70 000 ppm Cl⁻ / pH 1 / 105 °C. Also, four scale errors: the originator’s 27 m/min turning is not a modern tooling speed (coated carbide runs 70–90 m/min) — print both, never average, and 7 rpm tapping is NOT a typo; PREN is quoted with two incompatible formulas (65–76 versus 46); a single “yield strength” figure is meaningless (minimum 310 · typical plate 345–372 · typical sheet 407–434 · weld metal 524 MPa); and ASTM A262 rates (68 and 137 mpy) look alarming and are not — A262 is a stainless test.
10. Do not publish what has not been published. The following could not be independently verified for this material; write “could not be independently verified” instead of printing a number: the NACE MR0175 hardness cap and environmental envelope for N06022 · the ASME Section IX F-Number for ERNiCrMo-10 / ENiCrMo-10 · ASME Sec. II Part D allowable stress values · the covered electrode’s UNS number (W86022 or N06022) · B574 cold-finished rod properties · P-phase and σ-phase windows · a TTT curve and time-to-sensitisation · a hot-cracking ranking for C-22 · whether ASTM F468 includes N06022 · magnetic permeability after cold work · a hypochlorite corrosion-rate table · quantitative FGD rates.​‌​​‌​

​‌​​‌​