​‌​​‌​

Hastelloy B-3 / (2.4600) / UNS N10675

​‌​​‌​

Hastelloy B-3 (UNS N10675), also widely known as Alloy B-3, is a very special material within the Hastelloy group of nickel alloys. Not in very widespread use among the Hastelloy materials, this grade is chosen particularly where pure hydrochloric acid, hydrobromic acid and sulphuric acid are present, because it withstands acids of this kind over long periods.

The material contains approximately 65% nickel, approximately 29% molybdenum and around 3% of elements such as tungsten and cobalt that raise its resistance to very demanding conditions. All of these elements make the grade difficult to produce, and because they are all expensive elements these materials are correspondingly costly.​‌​​‌​

Machinability: Hastelloy B-3 is an alloy of high strength thanks to its high nickel content and molybdenum additions. This can create some difficulties in terms of machinability, but efficient results can be obtained with careful machining methods.

Turning and milling: Cutting tools — hardened steel and carbide inserts are generally used, as this type of tooling provides resistance to high hardness and temperature. Cutting speed — low and medium cutting speeds generally give better results, while excessively high speeds can accelerate wear and reduce the quality of the machined surface. Cooling — using cutting fluids helps prevent overheating and improves machining quality; it is important that the cutting fluid is of a type that will not cause chemical reactions.​‌​​‌​

Weldability: The alloy can be welded by the TIG and MIG processes. The most important factor to observe during welding is the risk of high thermal stress and oxidation, so cleanliness before and after welding is important. High thermal stress during welding can reduce the corrosion resistance of the alloy, so the use of shielding gases and control of process temperatures are recommended.

Heat treatment: The alloy is generally not heat treated, since it already has high temperature capability. Instead, attention should be paid to the temperature tolerances of the material during welding and forming operations.​‌​​‌​

​‌​​‌​

What Hastelloy B-3 Is — and the Real Difference Between B, B-2 and B-3

Hastelloy B-3 (UNS N10675 / W.Nr. 2.4600 / DIN NiMo29Cr) is a wrought, single-phase face-centred-cubic (γ) nickel–molybdenum solid-solution alloy: nominally 65 Ni – 28.5 Mo, with deliberately small Cr and Fe and essentially no carbon (C ≤0.010 %). It contains no meaningful chromium — 1.0–3.0 % — and that single fact explains both everything it is outstanding at and everything it fails at. It is a reducing-acid alloy, not a general-purpose corrosion alloy.​‌​​‌​

Buyers and even some datasheets conflate two entirely different failure mechanisms. They are not the same problem and they were solved thirty years apart.

The measured result — the numbers that belong on the page​‌​​‌​

B-2: Ni₄Mo forms rapidly at ~750 °C. B-3: “it takes several hours (at around 650 °C) to induce deleterious second phases“, and the phase that eventually forms is the slower-forming Ni₃Mo rather than Ni₄Mo. The patent’s hardness data: experimental alloys at 5–5.5 at.% SAE “did not significantly harden even after 24 hours at 700 °C“, against B-2’s 10+ HRA in 0.5–1.0 h. The commercially decisive sentence is in the patent: heating and cooling times “may safely be about ten times slower than the times recommended for B-2 alloy.” That is what lets you weld heavy sections, slow-cool and air-cool thin sections.

The corrosion consequence (the mill’s own comparison test): after exposure at 700 °C, tested in boiling 60 % H₂SO₄ — B-3: no cracking at 24 h. B-2: intergranular cracking at 3 h. Long-term stability: after exposure at 540 °C, elongation 45.6 % at 4,000 h, 47.1 % at 8,000 h, 43.7 % at 16,000 h — essentially no ductility loss in two years.​‌​​‌​

The caution that must be printed: B-3 is slower, not immune. Several hours at 650 °C still produces Ni₃Mo. B-3 is not a high-temperature alloy and is not intended for sustained service in the 500–900 °C band. Keep two claims separate as well: freedom from knife-line attack comes from the low carbon inherited from B-2; freedom from HAZ ordering embrittlement is B-3’s new contribution.

Standards by Product Form​‌​​‌​

Product Forms With NO Standard — the Commercially Valuable Section​‌​​‌​

Chemical Composition​‌​​‌​

ASTM B333 / B335 (the same table carried into B619/B622/B626/B462/B564), wt%: Ni ≥65.0 · Mo 27.0–32.0 · Fe 1.0–3.0 · Cr 1.0–3.0 · Ni + Mo 94.0–98.0 (a compound limit distributors routinely omit) · C ≤0.010 · Si ≤0.10 · Mn ≤3.0 · Co ≤3.0 · W ≤3.0 · Al ≤0.50 · Ti ≤0.20 · V ≤0.20 · Nb ≤0.20 · Ta ≤0.20 · Cu ≤0.20 · Zr ≤0.10 · P ≤0.030 · S ≤0.010.

Mechanical Properties​‌​​‌​

Physical Properties​‌​​‌​

Heat Treatment and Thermal Stability​‌​​‌​

Welding

What actually goes wrong

1. Welding over cold work. The single most-cited failure mode. If cold work exceeds ~7 % and has not been annealed out, B-3 is “very susceptible to cracking in the welded region during subsequent fabrication/welding“. Solution-anneal cold-formed heads, bent pipe, rolled shells and expanded tube ends BEFORE you weld them.
2. Heat-input creep in multipass welds. The nose is at 650 °C with a several-hour incubation, so one weld does not hurt it — but a heavy multipass joint held warm, plus a slow stress-relief, plus a hot-forming reheat, accumulates. Track cumulative time in 500–900 °C, not just peak temperature.
3. The weld-metal penalty in sulphuric acid. The mill’s own weld-versus-base data at 93 °C: in 50 % H₂SO₄ the weld corrodes at 0.13 mm/y against 0.04 mm/y for base metal — a 3.3× penalty; in 70 % H₂SO₄, 0.03 against 0.01. But in HCl the weld and base metal are identical (5 % HCl: 0.30 mm/y both; 10 % HCl: 0.29 both). So in HCl the weld is not a weak point; in sulphuric acid it is — a genuinely useful distinction almost nobody publishes.
4. Cleanliness. Components must be “stress-free, clean, and free of contaminants” before welding. Sulphur, lead, zinc and copper contamination cause hot cracking in all high-nickel alloys.​‌​​‌​

Machining

A caveat to state plainly: the mill publishes machining parameters for its corrosion-resistant alloy family rather than a B-3-specific table. Publish the figures below as “starting parameters for solution-annealed corrosion-resistant alloys, including B-3“, not as B-3 data.​‌​​‌​

Corrosion — Why the Behaviour Is Bimodal​‌​​‌​

B-3 has about 1.5 % chromium. It does not form a chromium-oxide passive film. Its resistance in acids comes from molybdenum raising the hydrogen overpotential and suppressing the reducing cathodic reaction. The consequence: superb where the only cathodic reaction available is hydrogen evolution; very poor the moment a stronger oxidant is present. Every entry below follows from that one sentence.

Frequently Asked Questions

We already run Hastelloy B-2. Is it worth paying for B-3, or is this just a newer part number?​‌​​‌​

It is worth it, and the reason is narrow and specific: thermal stability during fabrication and upset conditions, not corrosion resistance in service. In clean, uncontaminated hydrochloric acid the two alloys perform very similarly; if that were the whole story you should keep buying B-2.
What differs is what happens to the metal between roughly 500 and 900 °C. B-2, having had its carbon removed, orders very quickly into the brittle Ni₄Mo β phase — the mill measured over ten Rockwell-A points of hardening in half an hour to an hour at 700 °C. That happens in a weld heat-affected zone, in the centre of a slow-cooled heavy section, during hot forming, and during any process upset. The result is grain-boundary embrittlement and an intergranular corrosion path. The mill’s own side-by-side test, after exposure at 700 °C and then testing in boiling 60 % sulphuric acid, found B-2 cracking intergranularly at three hours while B-3 was still uncracked at twenty-four.
B-3 achieves this by deliberately adding back what B-2 removed — minimum 1.0 % each of chromium and iron, up to 3 % manganese, and aluminium — which diverts the reaction to the far slower Ni₃Mo. The patent states the practical consequence bluntly: heating and cooling times can be about ten times slower than for B-2.
So: if you are buying plate to weld into a vessel, to hot-form, to weld heavy sections, or you have process excursions into that temperature band, B-3 removes a real and documented failure mode. If you are buying thin, already-annealed sheet for a non-welded application in cold acid, the advantage is smaller. Price the risk, not the alloy.

Your page says B-3 resists hydrochloric acid at all concentrations and temperatures. Our HCl line is carbon steel upstream. Are we fine?​‌​​‌​

No, and this is the failure we see most often. That “all concentrations and temperatures” claim is true only for pure, deaerated, uncontaminated hydrochloric acid, and your carbon-steel upstream line guarantees you do not have that.
B-3 contains about 1.5 % chromium. It has no passive oxide film. Its resistance comes from molybdenum suppressing the hydrogen-evolution reaction — which works beautifully as long as hydrogen evolution is the only cathodic reaction available. Introduce a stronger oxidant and the protection simply is not there.
Ferric ion is exactly that oxidant, and carbon steel in HCl manufactures it continuously. The published number: 20 % HCl at 93 °C corrodes B-3 at 0.30 mm/y. The same acid with 50 ppm of ferric ion corrodes it at 2.0 mm/y — about 6.6 times faster. Fifty ppm is a trace: one rusty flange, one steel valve trim, one carbon-steel pump casing produces it. Cupric ion, from any copper-bearing component, does the same. Both the originating mill and every serious European mill sheet carry the sentence “not recommended in the presence of ferric or cupric salts“.
Practically, you have three options. Replace the upstream carbon steel so the acid stays clean. Fit and actually monitor a ferric/cupric analysis on the stream — set an alarm, not a quarterly sample. Or select a nickel-chromium-molybdenum alloy such as C-276 or C-22, which tolerates the oxidant at the cost of some performance in pure HCl. We will quote B-3 for this line, but we will put the ferric-ion limitation on the order acknowledgement, because the alloy is not the variable — the contamination is.

The customer’s specification says B-3, ASME Section VIII, service to 450 °C. Can we supply it?​‌​​‌​

Not as a pressure-retaining material at that design temperature — and the reason is a code limit, not a metallurgical one. Those are different things, and it matters which one you are hitting.
Metallurgically, B-3 is perfectly serviceable at 450 °C. The mill publishes typical tensile data to 649 °C, and exposure data showing elongation still at 43.7 % after 16,000 hours at 540 °C. The alloy is not in trouble at 450 °C.
The problem is code coverage. ASME Section VIII Division 1 accepts B-3 only to 427 °C (800 °F). There are no allowable stresses above that. ASME B31.3 stops at the same 427 °C. If your customer is on the European route, VdTÜV Werkstoffblatt 517 stops at 400 °C, lower still. And Section XII, for transport tanks, stops at 343 °C. So at 450 °C there is no allowable stress to design with, whichever code you are on.
Three further coverage facts that catch people out: B-3 is not accepted in Section VIII Division 2, not in Section I, and not in B31.1. If the vessel is a Division 2 design, B-3 cannot be the pressure boundary without a code case.
Realistic answers: get the design temperature reduced to 427 °C if the process genuinely allows it; move to a Ni-Cr-Mo alloy such as C-276 or alloy 625 that has code coverage further up; or use B-3 as a non-pressure-retaining liner or internal on a code-approved backing material. What you must not do is quote “B-3 is good to 650 °C” from a mechanical-property table and let it become a design temperature. That is a capability figure, and the code will not accept it.

Common datasheet errors — check these before you order​‌​​‌​

1. “AMS 5891” quoted as a B-3 plate specification — WRONG. AMS 5891 covers bars, forgings and rings of a 60Ni-22Cr-2Mo-14W alloy, a completely different material. There is no AMS specification for B-3.
2. “AMS 5396” quoted as a B-3 equivalent — WRONG. That is the specification for 65Ni-28Mo-5.5Fe-0.4V (N-12MV) investment castings — the cast alloy-B composition, with the 5.5 % iron and 0.4 % vanadium that B-3 specifically excludes.
3. ERNiMo-7 / ENiMo-7 quoted as the filler for B-3 — WRONG; that is the B-2 filler. Correct: ERNiMo-10 / ENiMo-10, both UNS N10675.
4. ASME P-Number given as 112 — WRONG. B-3 is P-No. 44 / F-No. 44.
5. W.Nr. confusion: 2.4600 = B-3, 2.4617 = B-2, 2.4800 = alloy B. And 2.4695 / 2.4696 are FILLER-metal numbers, not base-metal grades.
6. EN name: the mill and two German producers all publish NiMo29Cr — that is the majority and the mill position. NiMo30Cr is genuinely correct — for the welding wire (2.4695), which is very likely the origin of the error.
7. “Maximum working temperature 400 °C” published as a material property. 400 °C is the VdTÜV 517 code ceiling; ASME VIII Div. 1 allows 427 °C; the mill publishes mechanical data to 649 °C and long-term stability data at 540 °C. Three different numbers, three different meanings — label every one.
8. Column-misaligned distributor tables. A widely mirrored B-3 PDF shows yield 1370 MPa (that is the solidus in °C), tensile 216 MPa (that is the modulus in GPa), expansion 50 (that is the elongation %) and hardness 11.2 HRC (that is the thermal conductivity). Sanity-check any table where yield exceeds tensile.
9. Hardness scale confusion. B-3 is 92–95 HRBW, roughly 15–17 HRC. Anyone reading “95” as HRC will reach a nonsensical conclusion. Always print the scale.
10. “B-3 is immune to embrittlement” — FALSE. It is slower: several hours at ~650 °C against minutes for B-2, forming Ni₃Mo rather than Ni₄Mo. It is still a Ni-Mo alloy with an ordering reaction.

​‌​​‌​

​‌​​‌​