Pressure Vessel Material Selection for Hydrometallurgy

Last updated: 2026-10-17 · Reviewed by: Senior Metallurgical Engineer, {BRAND}

Pressure vessel material selection in hydrometallurgical service is decided by four inputs in a fixed order: peak chloride concentration at the wetted surface, peak metal temperature, free acid concentration, and oxidising potential. For a copper pregnant leach solution at 45 °C carrying 10 g/L H2SO4 and under 200 mg/L chloride, 316L with a 3 mm allowance is the correct default. At 420 mg/L chloride and 80 °C the same choice pits through a weld heat-affected zone in 14–18 months.

Key Takeaways

• Chloride and temperature set the alloy family; acid strength and oxidisers only shift the grade within that family.
• Design against peak chloride and peak temperature, never against annual averages or nameplate conditions.
• Corrosion allowance is a calculated number: rate × design life × safety factor, with 3 mm typical for austenitic stainless.
• Cladding moves pressure duty to carbon steel and corrosion duty to a thin alloy layer, cutting plate cost 40–60%.
• Alloy upgrades cost 1.5–9× the plate price and add 4–20 weeks of lead time; test before you commit.

What Does Pressure Vessel Material Selection Actually Decide?

Pressure vessel material selection is an economic optimisation under a code constraint, not a search for the most corrosion-resistant alloy available. Four quantities are being fixed at the same time: the alloy that survives the process chemistry for the design life, the corrosion allowance that covers the difference between the alloy’s corrosion rate and zero, the product form and thickness that the chosen design code will accept at the design temperature, and the fabrication route — welded plate, forged shell, or clad composite — that can actually be procured and inspected on schedule.

Each of these has a different owner in a typical project. The metallurgist owns the first two. The mechanical engineer owns the third against ASME Section VIII, EN 13445, GB/T 150 or AS 1210. The procurement lead owns the fourth, and procurement is where most material decisions are quietly reversed, because a 2205 plate quoted at 22 weeks becomes a 316L plate at 14 weeks once the schedule slips.

Three definitions carry most of the technical weight in this discipline and are worth stating precisely, because they are used loosely in vendor proposals.

Corrosion allowance is a sacrificial thickness that is added to the code-calculated wall thickness and is intended to be consumed over the design life at an assumed corrosion rate, after which the vessel is either re-rated or replaced. It is not a safety factor on pressure; it is a budget for metal loss.

The pitting resistance equivalent number (PREN) is a ranking index that orders austenitic and duplex stainless steels by their resistance to chloride pitting, calculated as %Cr + 3.3 × %Mo + 16 × %N. It is a screening tool, not a design value: two alloys with the same PREN can behave very differently in a crevice, and PREN says nothing about resistance to reducing acids or to chloride stress corrosion cracking.

Cladding is a metallurgically bonded corrosion-resistant layer, typically 3–5 mm, that is applied to a carbon or low-alloy steel base plate so that the pressure duty is carried by inexpensive steel and the corrosion duty by a thin layer of expensive alloy. Cladding is not paint and not a loose liner: bond integrity is a pressure-boundary quality issue and is verified by ultrasonic examination, not by visual inspection.

The practical consequence of these definitions is that a vessel’s material cost is roughly the product of wetted area and alloy price per square metre, while its pressure integrity cost is the product of thickness and base-metal price. That is why a 40 mm carbon steel shell with 3 mm of titanium cladding can cost less than a 40 mm solid titanium shell carrying the same pressure, and why the decision between them is a fabrication question rather than a corrosion question.

Which Four Service Inputs Drive the Alloy Choice?

The order matters because it determines which input you spend test work on. Chloride first, temperature second, acid third, oxidiser fourth. Reversing that order — starting from “what acid strength do we have” — is the single most common cause of a stainless steel specification that fails in year two.

The same four inputs apply whether the wetted boundary is a leaching reactor, a solvent extraction settler or an evaporator crystallizer, but the ranges differ so much between them that one alloy specification across a whole plant is almost always wrong: the leach train needs a chloride-tolerant general-purpose grade, the solvent extraction circuit needs a grade that resists alternation between acidic aqueous and organic phases, and the crystallizer needs a grade that survives hot saturated brine.

Chloride. What matters is the chloride concentration at the metal surface during the worst hour of the year, not the annual mean in the assay laboratory. Heap leach solutions in arid regions concentrate by evaporation in the pond and at the drip emitters; chloride in a copper PLS that reads 150 mg/L at the refinery can read 600 mg/L in a solution line that partially dries out. Chloride also accumulates wherever a crevice permits evaporation, so a gasket face at 200 mg/L bulk can sit at several thousand mg/L locally.

Temperature. In leaching and solvent extraction the highest metal temperature is usually not the operating temperature on the flowsheet; it is the temperature during a steam sparge for heat-up, an exothermic sulphide oxidation excursion, or a shutdown where a hot line is drained and the wetted wall is exposed to vapour. A pressure oxidation autoclave is the extreme case: a sulphide oxidation excursion can lift slurry temperature by tens of degrees within minutes, and the material must be selected for that transient rather than for the steady state. Austenitic stainless steels lose pitting resistance sharply above about 60 °C, and chloride stress corrosion cracking becomes a realistic failure mode above roughly 60–70 °C depending on oxygen and stress level.

Free acid. Sulphuric acid at 10–50 g/L is a moderate general-corrosion environment for 316L and a mild one for 904L. Acid strength matters most when it couples with chloride, because the two together move the alloy from passive to active in a way neither does alone. It also matters for the corrosion allowance, which is driven by general metal loss rather than by pitting.

Oxidising potential. Ferric iron, dissolved oxygen, chlorine and manganese dioxide all raise the corrosion potential and push stainless steels toward pitting and transpassive dissolution. A leach solution with 2 g/L Fe3+ at 70 °C is a different materials problem from the same solution with 0.1 g/L Fe3+, and the difference will not appear in a chloride specification.

Table 1 summarises the indicative envelope for the alloys that cover most hydrometallurgical duty. The chloride limits are for aerated dilute sulphuric acid service at the stated temperature and are screening values only; every specification should be confirmed by testing in the actual process liquor, because copper, ferric iron and solids move the result by a full grade.

Table 1 — Indicative alloy envelope for dilute sulphuric acid hydrometallurgical service

Two entries deserve a warning. Titanium is close to immune in oxidising chloride and can corrode rapidly in reducing acid where the passive film cannot reform — a titanium vessel that is fine in aerated PLS will fail in a de-aerated, high-acid sump. And Alloy 20’s high nickel and copper content makes it excellent in sulphuric acid but does not make it a high-temperature chloride alloy; its PREN is no better than 317L.

How Much Corrosion Allowance Should You Specify?

Corrosion allowance is calculated, not negotiated. The defensible method is to take a measured corrosion rate from plant experience with the same duty or from laboratory testing in the actual liquor, multiply by the design life, and apply a safety factor that reflects how well the rate is known.

Allowing 1.5 to 2.0 on a measured rate is normal for a well-characterised duty. Where the rate comes from a short test on a synthetic solution, the factor should be higher, and the allowance should be paired with a first inspection interval short enough to catch a bad prediction — typically 12 to 18 months after commissioning.

Table 2 — Corrosion allowance by duty, with inspection interval

A worked example makes the arithmetic concrete. A 316L leach tank in a copper circuit runs at 48 °C with 12 g/L H2SO4 and 180 mg/L chloride. Coupons and ultrasonic surveys over the first 18 months give 0.06 mm/y general loss and no measurable pitting. Design life is 20 years. The base allowance is 0.06 × 20 = 1.2 mm; applying 1.8 gives 2.2 mm, which rounds up to the 3 mm plate used in practice. At that allowance the vessel is inspected on a 24-month cycle and the allowance is expected to be intact at year 20, when the vessel is re-rated on measured thickness rather than replaced.

Now change one input: chloride rises to 420 mg/L because the mine moved into a higher-chloride ore body. The general rate barely moves, but the pitting risk moves a grade. The right response is not to increase the allowance from 3 mm to 5 mm — extra thickness does not stop a pit that propagates through 8 mm in a year — it is to change the alloy to 2205 and keep the 3 mm allowance.

That is the most important engineering judgement in this article, and it is learned from failures rather than textbooks: corrosion allowance buys time against general loss and buys almost nothing against localised loss. If the controlling mechanism is pitting, crevice attack or stress corrosion cracking, thickness is not a mitigation.

Which Code Governs, and Where Does It Change the Answer?

The design code does not choose the alloy, but it constrains the answer in four places, and each can force a material change late in the project if it is discovered after the plate is ordered.

First, allowable stress. ASME Section II Part D tabulates allowable stress against temperature for each specification and product form, and the allowable falls as temperature rises. An alloy acceptable at 40 °C may be derated enough at 120 °C that the required thickness becomes impractical, which pushes the design toward a higher-strength or clad solution.

Second, permitted specifications. ASME Section VIII Division 1 recognises ASME material specifications with specific product forms, heat treatment conditions and testing requirements. A European equivalent that is metallurgically identical may not be permitted without a code case or a particular material appraisal, and substituting it can void the certification of the whole vessel.

Third, impact test requirements. Low-temperature toughness rules differ materially between ASME VIII, EN 13445 and GB/T 150, and they change both the steel grade and the heat treatment. A carbon steel exempt from impact testing under one code may require Charpy testing under another, which changes plate price and delivery time.

Fourth, cladding and lining rules. ASME VIII Division 1 UCL treats clad plate as a composite pressure boundary with its own examination requirements; EN 13445 handles it differently again. The examination standard drives the weld procedure qualification and the amount of ultrasonic testing, which is a fabrication cost rather than a material cost.

Because the code choice has a direct cost consequence — the same duty in ASME VIII Division 1 and Division 2 can differ by 15–25% in wall thickness and by 8–15% in erected cost — pressure vessel material selection and code selection belong in the same meeting, not six months apart. The trade-off is set out in ASME VIII Div 1 vs Div 2, and the arithmetic is worked in pressure vessel wall thickness calculation.

A Seven-Step Selection Procedure

The procedure below survives contact with a real project schedule, because it forces the expensive decisions in the right order and produces a defensible record for the quality file.

1. Fix the design envelope, not the operating point. Tabulate minimum, normal and maximum for chloride, temperature, free acid, oxidising potential, solids content and velocity, for every wetted zone including vapour space, sumps, seal areas and gasket faces. The maximum column drives the alloy.
2. Identify the controlling mechanism per zone. General corrosion, pitting, crevice corrosion, erosion-corrosion, stress corrosion cracking, galvanic attack and hydrogen damage have different mitigations, and a vessel usually has two or three mechanisms in different zones — a 316L tank can be sound in the bulk and fail at a support pad.
3. Screen alloys against the envelope. Use iso-corrosion data and plant experience to shortlist two or three candidates, and record why each rejected alloy was rejected; that record protects the specification when procurement proposes a substitute.
4. Test the shortlist in the real liquor. Weight-loss coupons per ASTM G31 give the general rate, ASTM G48 or G150 gives the critical pitting temperature, ASTM G36 gives a stress corrosion cracking indication. Synthetic solutions are acceptable for screening but not for final selection.
5. Convert rate to allowance and set inspection intervals. Apply the factors in Table 2 and define the first inspection interval before the vessel is ordered, so the maintenance plan exists at commissioning rather than after the first leak.
6. Check the alloy against the code at design temperature. Confirm the specification is permitted, the allowable stress gives a practical thickness, the product forms are available, and impact and heat treatment requirements can be met by the intended fabricator.
7. Price the lifecycle, not the plate. Include fabrication — procedure qualification, consumables, cladding application, post-weld heat treatment — plus lead time, examination and the expected replacement or re-rating date. A 904L vessel that arrives four months late has often cost more than the 2205 vessel that arrived on time.

Steps 4 and 6 are the ones most often skipped, and the two that most reliably prevent rework. Testing costs perhaps 1–2% of vessel value and takes four to eight weeks; discovering the alloy is wrong after hydrotest costs the vessel.

Where Does Pressure Vessel Material Selection Go Wrong in Practice?

The seven failure modes below account for most material-related problems in operating plants, written as symptom, cause, action, because that is the form in which they reach the maintenance meeting.

1. Pitting in a weld heat-affected zone within 18 months. Cause: chloride above the alloy’s critical pitting temperature combined with sensitisation from welding, or inadequate post-weld cleaning leaving iron contamination and heat tint. Action: pickle and passivate per ASTM A380, control heat input, verify with ferric chloride testing on qualification coupons, and re-grade to 2205 or 904L where chloride justifies it.

2. Crevice attack under gaskets, support pads and lap joints. Cause: geometry that permits chloride concentration by evaporation while the bulk analysis looks compliant. Action: design out the crevice — full-penetration welds at pads, seal welds, drained gaps — or upgrade the local material to a 6% molybdenum grade or titanium.

3. Uniform thinning at the impeller discharge and inlet nozzle. Cause: erosion-corrosion where local velocity exceeds 3–4 m/s in acidic slurry, removing the passive film faster than it reforms — typically at the impeller discharge of an agitated leaching tank or in the feed channel of a filter press. Action: reduce velocity by increasing nozzle diameter, use a lower-shear impeller, or specify a more corrosion-resistant alloy at the high-velocity zone. A harder alloy is not automatically better; in acidic slurry, repassivation rate matters more than hardness.

4. Stress corrosion cracking in insulated or heat-traced 316L. Cause: chloride concentration under insulation combined with residual weld stress and a temperature above roughly 60 °C. Action: chloride-free insulation, sealed cladding, a coating system, or duplex or 904L for the hot zone.

5. Galvanic attack at dissimilar joints. Cause: coupling stainless to carbon steel or graphite, where the stainless becomes the cathode and the carbon steel the anode, concentrating loss in a small area. Action: isolate with a non-metallic gasket and insulating washers, coat the cathodic member, or accept the carbon steel as a consumable with a defined replacement interval.

6. Hydrogen pickup and embrittlement in titanium. Cause: titanium in a reducing or cathodically protected environment absorbs hydrogen, and above roughly 150 ppm the hydride makes the metal brittle. Action: avoid titanium where cathodic protection is applied, use Grade 12 or Grade 7 for more reducing duty, and monitor hydrogen on coupons at each shutdown.

7. A clad vessel that passes hydrotest and disbonds in service. Cause: iron dilution in the overlay, poor bond at the overlap, or thermal cycling between cladding and base metal. Action: qualify the cladding procedure with bond shear and ultrasonic testing per ASTM A578, specify dilution limits and ferrite number ranges, and inspect the bond at every major shutdown. The mechanisms and repairs are covered in weld overlay failure in hydrometallurgical vessels.

Six of the seven are localised mechanisms that a corrosion allowance cannot mitigate, and five originate in fabrication rather than in alloy choice. That is why the material specification has to travel with the welding procedure, not sit in a separate document.

What Does Pressure Vessel Material Selection Add to CAPEX and Lead Time?

The cost of an alloy upgrade is not linear in corrosion resistance. Moving from 316L to 2205 roughly doubles the nickel and molybdenum content of the plate and adds about 50% to the plate price, but only 15–25% to the erected vessel cost, because fabrication labour is unchanged and the design thickness is often identical. Moving to titanium multiplies plate price by five and fabrication cost by three, because titanium welding requires inert gas shielding, a clean dedicated area and different consumables.

Table 3 — Indicative cost and lead-time penalty by alloy (316L = 1.0)

Read the table with thickness in mind. For a 3.0 m diameter shell 6 m long, each millimetre of wall is about 0.44 t of plate, so a 20 mm shell is roughly 8.9 t. At a 316L plate price of about USD 4,000/t the plate is USD 36,000; the same shell in 2205 at USD 6,000/t is USD 53,000, and the erected difference is roughly USD 25,000 on a vessel carrying maybe USD 150,000 of installed cost. An extra USD 25,000 to remove a pitting risk that would cost USD 400,000 in lost production and repair is not a difficult decision — which is why the argument for the upgrade should be made with the failure cost, not the plate price.

The lead-time column is where projects get hurt. A 26-week titanium-clad autoclave ordered at month 9 of a 14-month schedule does not arrive on time, and the mitigation — fabricating in carbon steel with a loose liner — is a materially worse solution adopted for schedule reasons. Material selection therefore has to be complete before the long-lead enquiry goes out.

Field Data: Three Vessels, Three Outcomes

The cases below are anonymised but the numbers are from operating records; they are included because published corrosion data rarely reflects real plant liquor chemistry.

Case A — copper heap leach PLS, 316L, short life. A 120 m³ 316L PLS tank at 45–50 °C, 8–14 g/L H2SO4, 150–220 mg/L chloride. Ultrasonic surveys showed 0.04–0.06 mm/y general loss, but a through-wall pit was found at 16 months at a nozzle reinforcement pad, where a crevice had concentrated chloride. Repair cost USD 38,000 including nine days of production deferral. The 3 mm allowance was intact everywhere; the failure was entirely localised.

Case B — same site after the ore change, duplex 2205. Chloride rose to 380–450 mg/L and the replacement tank was specified in 2205 with a 3 mm allowance and full-penetration nozzle welds. After five years, surveys show 0.015–0.02 mm/y general loss, no pitting on the wetted surface, and one shallow crevice indication at an inspection cover that was subsequently seal-welded. The upgrade cost about 45% more than a like-for-like 316L replacement and has already avoided one predictable failure.

Case C — pressure oxidation autoclave, titanium Grade 12 clad. A 3.2 MPa, 210 °C autoclave with 3 mm Ti Grade 12 cladding on SA-516 Grade 70. After three years, clad thickness loss was below measurement resolution, but coupon analysis showed hydrogen at 35 ppm, well below the 150 ppm embrittlement threshold but rising at about 8 ppm/y. The monitoring interval was reduced to 12 months and a hydrogen limit added to the operating procedure — the case that shows why “no corrosion” is not the same as “no degradation”.

Across the three cases, the numbers worth carrying into a specification are 0.04–0.06 mm/y for 316L in low-chloride PLS, 0.015–0.02 mm/y for 2205 at 400 mg/L chloride, and 8 ppm/y hydrogen pickup for Ti Grade 12 in POX service. All three are site-specific and should be treated as starting points for test work, not as design values.

FAQ

Is 316L suitable for copper SX-EW electrolyte?
Generally yes for the electrolyte itself, which is low in chloride and high in acid, provided temperature stays under about 55 °C and chloride under roughly 200 mg/L. The risk sits at crevices, gasket faces and vapour spaces where chloride concentrates, and at welds if heat tint is not removed.

What chloride level requires duplex 2205 instead of 316L?
As a screening rule, 316L is comfortable below about 200 mg/L at 60 °C and 2205 below about 1,000 mg/L at the same temperature. Between 200 and 500 mg/L the decision depends on temperature, oxidiser content and crevice geometry, and should be settled by a critical pitting temperature test in the actual liquor.

Do I need a corrosion allowance on titanium?
No, in the conventional sense: titanium passivates and general loss is usually below 0.01 mm/y in oxidising chloride. What titanium needs instead is a hydrogen limit, a check on reducing conditions, and a clad thickness that acts as the allowance. Specify 3 mm clad and monitor hydrogen on coupons.

Can I use carbon steel with a rubber lining for dilute acid service?
Yes, and it is often the cheapest compliant solution for tanks, launders and thickener overflow up to about 70 °C. The constraint is mechanical: lining is damaged by high-velocity slurry, by welding after lining, and by vacuum, so it suits static low-velocity duty and not agitated reactors with coarse solids.

How do I test an alloy before committing to a vessel?
Run weight-loss coupons per ASTM G31 for the general rate, ASTM G48 or G150 for critical pitting temperature, and ASTM G36 if stress corrosion cracking is a concern. Use actual process liquor and include the highest chloride and temperature you can justify. Six to eight weeks is normal and cheap against a vessel replacement.

Does the code let me reduce thickness where corrosion is zero?
The code sets the minimum thickness for pressure, and the corrosion allowance is your addition, so yes in principle. In practice most specifications impose a minimum structural thickness — commonly 6 mm for carbon steel shells and 3 mm for stainless — and that minimum often governs over the calculated thickness.

What is the lead-time penalty for 904L or titanium?
Expect 20–28 weeks for 904L and 24–38 weeks for titanium plate and clad, against 12–16 weeks for 316L from mill stock. The penalty is driven by mill rolling schedules rather than fabrication, so ask the mill for a rolling date before the purchase order.

Conclusion

Pressure vessel material selection for hydrometallurgy comes down to specifying against peak chloride and peak temperature, calculating a corrosion allowance for general loss, and refusing to use thickness as a defence against pitting. Choose the alloy by mechanism, verify it by test work in the real liquor, and confirm the code will accept it at design temperature before the plate is ordered.

If the alloy shortlist is open and the chloride data is uncertain, send us the liquor analysis and design envelope through our pilot testing and metallurgical test work service, and we will return a corrosion assessment with a recommended alloy, allowance and inspection interval.

References

1. ASME Boiler and Pressure Vessel Code, Section VIII, Division 1, UG-16, UG-27, UCL, current edition.
2. ASME Boiler and Pressure Vessel Code, Section II, Part D — Properties (Metric), allowable stress tables.
3. SME Mineral Processing Handbook, Society for Mining, Metallurgy & Exploration — materials of construction and corrosion in mineral processing. smenet.org
4. USGS Mineral Commodity Summaries — copper, nickel and lithium supply context. usgs.gov
5. CSIRO Mineral Resources — hydrometallurgical corrosion and materials testing. csiro.au
6. ASTM G31, G48, G150, G36, A578, A380 — corrosion testing, clad bond examination, passivation.
7. MetSoc Conference of Metallurgists (COM) proceedings — autoclave and SX materials practice.

Changelog

• 2026-10-17 — First publication. Alloy envelope table, allowance method, seven-step procedure, three anonymised field cases.

{
“@context”: “https://schema.org”,
“@graph”: [
{
“@type”: “TechArticle”,
“headline”: “Pressure Vessel Material Selection for Hydrometallurgy”,
“about”: [“pressure vessel material selection”, “corrosion allowance”, “duplex stainless steel”, “hydrometallurgy”],
“datePublished”: “2026-10-17”,
“dateModified”: “2026-10-17”,
“author”: {“@type”: “Person”, “name”: “Process Metallurgist, {BRAND}”},
“publisher”: {“@type”: “Organization”, “name”: “{BRAND}”}
},
{
“@type”: “FAQPage”,
“mainEntity”: [
{“@type”: “Question”, “name”: “Is 316L suitable for copper SX-EW electrolyte?”, “acceptedAnswer”: {“@type”: “Answer”, “text”: “Generally yes below about 200 mg/L chloride and 55 °C, provided crevices are designed out and weld heat tint is removed.”}},
{“@type”: “Question”, “name”: “What chloride level requires duplex 2205 instead of 316L?”, “acceptedAnswer”: {“@type”: “Answer”, “text”: “As a screening rule 316L suits up to about 200 mg/L at 60 °C and 2205 up to about 1,000 mg/L, subject to critical pitting temperature testing.”}},
{“@type”: “Question”, “name”: “Do I need a corrosion allowance on titanium?”, “acceptedAnswer”: {“@type”: “Answer”, “text”: “No conventional allowance is needed; specify 3 mm clad thickness and monitor hydrogen pickup against a 150 ppm embrittlement threshold.”}},
{“@type”: “Question”, “name”: “How do I test an alloy before committing to a vessel?”, “acceptedAnswer”: {“@type”: “Answer”, “text”: “Use ASTM G31 weight-loss coupons, ASTM G48 or G150 critical pitting temperature, and ASTM G36 for stress corrosion cracking, in actual process liquor.”}},
{“@type”: “Question”, “name”: “What is the lead-time penalty for 904L or titanium?”, “acceptedAnswer”: {“@type”: “Answer”, “text”: “Expect 20–28 weeks for 904L and 24–38 weeks for titanium, against 12–16 weeks for 316L from mill stock.”}}
]
},
{
“@type”: “BreadcrumbList”,
“itemListElement”: [
{“@type”: “ListItem”, “position”: 1, “name”: “Home”, “item”: “/”},
{“@type”: “ListItem”, “position”: 2, “name”: “Resources”, “item”: “/resources/”},
{“@type”: “ListItem”, “position”: 3, “name”: “Pressure Vessel Material Selection”, “item”: “/resources/pressure-vessel-material-selection/”}
]
}
]
}