316SS, CD4MCu, Alloy 20 or Hastelloy: Choosing Pump Metallurgy

Most pump material decisions are inherited rather than made, and the inherited answer is very often 316 — including in the services where 316 is exactly wrong.

Start with what actually attacks the metal

Four questions decide the alloy, and they need answering in this order:

1. What is the fluid, and at what concentration? Sulphuric acid at 10% and at 95% attack completely different materials.
2. At what temperature? Corrosion rate roughly doubles for every 10–20°C. A material that is fine cold can be unusable hot.
3. Are chlorides present, and how much? This is the question that most often gets skipped, and it is the one that kills austenitic stainless.
4. Are there solids, and are they abrasive? Erosion–corrosion is a different mechanism, and hardness starts to matter as much as passivity.

The materials, in ascending order of when you need them

Carbon steel

Correct for hydrocarbons, non-corrosive process fluids, and any service where corrosion is genuinely not the mechanism. Cheap, strong, easy to weld and repair. It rusts in water service, which is why it is not the answer for water treatment even though it is the answer for a great deal of oil and gas duty.

316 stainless steel

The default, and correct far more often than not: organic acids, most caustics, general chemical service, food and pharmaceutical duty. The molybdenum over 304 buys real resistance to reducing conditions and moderate chlorides.

Where it quietly fails: chlorides. 316 pits and crevice-corrodes in chloride-bearing service, and above roughly 60°C in chloride solutions it becomes vulnerable to chloride stress-corrosion cracking — which is the dangerous one, because it produces sudden through-wall cracking with essentially no wall loss to warn you. Seawater, brine, produced water, and chloride-dosed cooling water all belong in this category. So does anything that concentrates chlorides by evaporation.

CD4MCu — the duplex

A cast duplex stainless: roughly half ferrite, half austenite, with copper. Two properties matter.

First, chloride resistance far beyond 316 — markedly better pitting and crevice resistance and greatly improved resistance to chloride stress-corrosion cracking. Second, strength and hardness at roughly twice the yield of 316, which makes it genuinely erosion-resistant.

That combination is why CD4MCu is the standard answer for abrasive slurry in a corrosive carrier — mine dewatering, phosphate and fertiliser processing, seawater, brine, and flue-gas desulphurisation. It is also why it appears as a casing and impeller option across the SPP and SPPX self-priming range, where solids and chlorides usually arrive together.

Limit: duplex grades embrittle above roughly 300°C (570°F) through sigma-phase formation. Duplex is not a high-temperature material.

Alloy 20 (Carpenter 20 / CN7M)

Developed for one problem: sulphuric acid. High nickel with copper and molybdenum, and it resists sulphuric across a very wide concentration range where both 316 and duplex fail, particularly in the hot mid-concentration band that is most aggressive.

Also strong against phosphoric acid, many sulphates, and mixed-acid service. Specify it when sulphuric is present at any meaningful concentration — the cost premium over 316 is trivial against an unplanned pump failure on an acid line.

Hastelloy B and C

The two are not interchangeable and the distinction is critical.

Hastelloy B (nickel–molybdenum) is for reducing conditions: hydrochloric acid at all concentrations and temperatures, sulphuric at moderate concentration, phosphoric, and hydrogen chloride. It has essentially no chromium, so it is poor in oxidising service — nitric acid or ferric ions will attack it rapidly.

Hastelloy C (nickel–molybdenum–chromium, as in C-276) handles both oxidising and reducing conditions, which is what makes it the general-purpose answer for mixed acids, wet chlorine, hypochlorite, ferric chloride, and any service where the oxidation state changes or is not reliably known. It is the practical ceiling for pump metallurgy.

Both are expensive. Specify from the actual chemistry, not as insurance — and note that specifying Hastelloy B where the service is oxidising is worse than specifying 316.

The mistakes worth naming

• Ignoring chlorides. The most common expensive error. Ask for a chloride number, in ppm, before selecting austenitic stainless. Ask again about upset and concentration conditions.
• Specifying on the normal case. Corrosion is driven by the worst hour, not the average one. Cleaning cycles, CIP chemicals, upsets and start-up conditions are frequently more aggressive than normal operation and are frequently left off the data sheet.
• Assuming higher alloy is always safer. Hastelloy B in oxidising service is a clear counter-example. So is duplex above 300°C.
• Matching the casing and forgetting everything else. Wear rings, shaft sleeve, fasteners and gaskets sit in the same fluid. A 316 pump with the wrong gasket still leaks.
• Ignoring galvanic pairing. Dissimilar metals in a conductive fluid will find each other, and the smaller anode will go first.
• Forgetting the seal faces. Silicon carbide against silicon carbide is the durable answer for abrasive service and is what the SPP and SPPX cartridge seals use. Carbon faces will not survive it regardless of casing alloy.

A short selection path

1. Non-corrosive or hydrocarbon → carbon steel.
2. General chemical, low chloride, moderate temperature → 316SS.
3. Chlorides present, or abrasive solids in a corrosive carrier, below 300°C → CD4MCu.
4. Sulphuric or phosphoric acid → Alloy 20.
5. Hydrochloric or strongly reducing, reliably non-oxidising → Hastelloy B.
6. Mixed acids, wet chlorine, oxidising or variable → Hastelloy C.
7. Not sure, or the fluid changes → send the analysis, including chlorides, temperature range and upset conditions. Guessing here is expensive in both directions.

All six are available across the Pinnacle 8896 ANSI range, and Pinnacle-Flo carries stock in carbon steel, stainless, CD4, HAS-C276 and Alloy 20 specifically so that the right material does not become a lead-time argument.

General engineering reference. Conditions vary by application; confirm any selection against your actual duty point, fluid analysis and operating range before purchase. Pinnacle-Flo application engineers will review your data on request.