Tell us what is in the tank. We will tell you what survives it.

Water with alkalinity under about 30 mg/L, pH under 7 and an LSI below -1 is the most under-specified service in the storage-tank business. It leaches cement and every calcium-bearing lining, it drives corrosion at each holiday in an organic coating, and it attacks galvanized bolts and hardware before it attacks the shell.

Specify an inert, holiday-free system. A factory-coated fusion-bonded epoxy bolted shell, holiday-tested panel by panel before it ships, is the system we supply for this water; a high-build 100% solids epoxy applied in the field with 100% holiday testing is the equivalent on an existing tank; 316L stainless is third. Glass-fused-to-steel also suits the chemistry and is worth pricing as a comparison, but it is a separate system, not a variant of the epoxy one, and its coating cannot be renewed once it is spent. Rule out cement-mortar, zinc-rich primer in immersion, and any thin-film epoxy under 16 mils. Then take care of the details that actually fail: stainless or fully encapsulated bolts and anchors, and cathodic protection on any epoxy-lined tank.

In balanced potable water — LSI between -0.5 and +0.5, alkalinity 50 to 200 mg/L, pH 7 to 8.5 — both systems are correct and the decision belongs to budget, schedule and access. An NSF/ANSI 61-listed immersion epoxy per AWWA D102 Inside Coating Systems is the baseline and it can be renewed in place for the life of the shell.

Chemistry is what separates them. Glass is inert, so it pulls ahead in aggressive low-alkalinity water, in high chlorine or chloramine residual, in iron, manganese and H2S groundwater, and anywhere the tank is drained and refilled often. Epoxy pulls ahead where the water is hard and scaling, where the budget is tight, and where the ability to recoat matters more than the coating being fused. Glass has a hard pH window of roughly 3 to 11 and it cannot be renewed in the field; epoxy is limited to about 120 to 140 °F wet.

The liquid is rarely the problem. H2S in the headspace — typically 500 to 5,000 ppm — dissolves in condensate on damp steel above the waterline and becomes sulfuric acid, so the gas-liquid line and the dome are where linings fail first.

For mesophilic digestion at 95 to 100 °F: glass-fused-to-steel with a vapor-zone-rated enamel, novolac epoxy second, 316L stainless in the dome and gas zone third. A factory-coated fusion-bonded epoxy shell will carry the liquid zone at these temperatures, but the headspace has to be specified as its own lining rather than assumed to come with the tank. At thermophilic temperatures of 125 to 140 °F, standard epoxy is already past its wet-heat rating and polyurea has no place in the gas zone at all — confirm the enamel's temperature rating in writing, insulate and heat-trace, and specify inspection ports. A gas holder seeing 100% biogas should be treated as chemical service, not water service: 316L or a phenolic novolac.

Saturated sodium chloride at near-neutral pH is a chloride problem, not a pH problem, and it is the service where stainless steel most often gets specified by mistake. Chloride pitting and crevice corrosion do not care that the pH is 7.

Vinyl ester or FRP first — which is why so many brine makers are FRP by default — then glass-fused-to-steel, then polyurea. Rule out any stainless, galvanized and cement. The same logic governs RO reject above 1,000 mg/L chloride, seawater intake at 19,000 mg/L, and FGD scrubber water above 10,000 mg/L; 316L is only defensible below roughly 1,000 mg/L chloride and at low temperature, and 304 below about 200 mg/L.

A fire protection tank per NFPA 22 is full, stagnant and rarely exchanged, so it fails from microbial corrosion and oxygen-cell attack at the waterline rather than from chemistry. On a new bolted tank, specify a factory-coated fusion-bonded epoxy interior, NSF 61 listed and compliant with NFPA 22. On an existing welded tank, an NSF 61-listed field epoxy per AWWA D102 does the same job. Glass-fused-to-steel is the other factory option and a fair comparison, but it is a different coating system with different renewal economics. Galvanized bolted tanks are reasonable on small volumes. Avoid cement-mortar in soft water.

The lining is only half the answer. NFPA 25 requires a three-year interior inspection, and the water itself needs either treatment or periodic exchange — a holiday-free lining in stagnant, nutrient-rich water still grows MIC nodules. If foam concentrate or AFFF is stored with the water, move up to fused glass or a novolac epoxy: surfactants drive creep at any holiday, and the PFAS transition means the foam chemistry on your site is probably changing.

Stainless resists corrosion through a passive chromium oxide film, and chloride ions break that film down locally, which starts a pit that then feeds itself. The alloy's general corrosion resistance is irrelevant once pitting initiates — and temperature accelerates it.

Practical screening limits: 304 below about 200 mg/L chloride, 316L below about 1,000 mg/L and at low temperature. Above that, chloride-immune systems take over — glass-fused-to-steel, vinyl ester or FRP — or, in seawater, a duplex alloy rather than 304 or 316. This is the rule the selector on this page applies when you enter a chloride number: over 1,000 mg/L and it removes 304 and 316L stainless from the ranked list and tells you why.

Use it to build the short list and to catch the systems that should never have been on it — that is what a screening matrix is good for, and the eliminations are usually worth more than the recommendations.

Do not use it as the basis of a submittal. Every figure here is a typical published range, and the actual limits belong to the specific product data sheet for the system you assign. Final lining selection requires a current water analysis and written confirmation from the coating or tank manufacturer that the system is rated for your pH, temperature, chloride level, vapor-zone exposure and disinfectant. Send us the analysis and we will get that confirmation in writing.