Molybdenum halves corrosion of stainless spring wire in hot acidic cleaners

Cold-drawn stainless steel spring wire can lose most of its metal over 60 days only after the first month has passed, a 432-specimen immersion study reports, and the molybdenum-bearing grade SUS316 cuts the corrosion rate of molybdenum-free SUS304 by roughly two- to five-fold under the harshest tested conditions. The work, published in Results in Engineering, is the first full-factorial immersion test on SUS304 and SUS316 spring wire across the pH, temperature and exposure window that acidic clean-in-place service in food, beverage and pharmaceutical plants actually involves.

The research team, led by Lun-Chuan Liao and Chen-Yang Cheng, wound helical compression springs from commercial 0.8 mm wire of both grades and immersed them in three acids — citric, glycolic and hydrochloric — at initial pH values of 3.2, 4.0 and 5.0, and at 25, 40, 55 and 60 °C, all against a fixed 10 wt% sodium chloride background. With six replicate springs per combination, the design covered 72 cells and 432 specimens, each weighed at 15, 30 and 60 days following ASTM G31. The springs hung uncompressed in individual tubes, and every specimen was inspected at 10× magnification for stress-corrosion cracks at each weighing.

The choice of acids reflects industrial practice. Cleaning-in-place cycles expose equipment to organic acids such as citric and glycolic, residual hydrochloric acid from descaling, and chloride traces from feedwater, typically at 40–70 °C. Unlike mineral acids, which attack stainless steel mainly through proton-driven dissolution of the passive film, organic acids carry carboxylate and hydroxyl groups that chelate iron and chromium ions, weakening the metal–oxygen bonds of the protective chromium-rich layer. Citric acid, with three carboxylic groups, can bind a metal ion at several sites at once; glycolic acid binds at a single site. The study tested whether this difference in coordination geometry shapes the corrosion response.

The headline numbers are striking. The highest rate anywhere in the design was SUS304 in citric acid at pH 3.2 and 60 °C, at 1.8 × 10⁻² millimetres per year, about 18 micrometres per year. SUS316 under the same condition read 7.6 micrometres per year, a 2.4-fold reduction. In glycolic acid at pH 4.0, the molybdenum advantage was larger still: SUS316 cut the SUS304 rate by 3.8 to 5.3 times at 55 and 60 °C. At 25 °C, however, the two alloys could not be told apart at all, because nearly all readings sat at the balance’s 1-milligram detection floor, equivalent to about 0.95 micrometres per year.

Statistical analysis confirmed that interactions, not main effects, dominate the behaviour. In a four-way analysis of variance on the day-60 rates, the acid-by-pH interaction was the largest single term, explaining 14.1% of the variance, ahead of acid-by-temperature and acid type. The alloy main effect was the smallest of all ten terms and not statistically significant. That does not mean molybdenum does nothing; rather, its benefit concentrates in the hot, aggressive cells and registers through interactions. Across eleven cells at 40 °C and above where both alloys corroded measurably, a regression of log SUS316 rate on log SUS304 rate gave a slope of 0.47, below unity, meaning SUS316 gains less corrosion rate than SUS304 as conditions harshen.

The two organic acids responded to pH in opposite directions, the study’s most chemically intriguing finding. On SUS304 at 60 °C, citric acid became steadily more aggressive as pH fell, rising from 0.48 micrometres per year at pH 5.0 to 18.0 at pH 3.2. Glycolic acid peaked in the middle, at pH 4.0, reaching 12.7 micrometres per year — well above its rates of 3.7 at pH 3.2 and 5.2 at pH 5.0. The glycolic peak sits near the acid’s pKa of 3.83, where its protonated and deprotonated forms coexist, consistent with a monodentate binding mechanism that requires both forms. Citric acid, which retains at least two coordination sites across the whole range, instead tracked the overall pool of protonated acid. The authors stress this coordination explanation is proposed, not demonstrated by the weight data.

Perhaps the most practically consequential result concerns timing. In six SUS304 cells, between 89 and 100% of the 60-day mass loss arrived after day 30, and at most 11% in the first 15 days. A two-part incubation-propagation model, in which nothing happens until an incubation time and loss then climbs at a steady rate, was strongly preferred over a straight line by both Akaike and Bayesian information criteria in all six cells, and a paired specimen-level test confirmed every individual spring lost mass faster in the second half of the exposure. A corrosion test stopped at day 30 would therefore have recorded almost none of the eventual damage. The authors recommend that any new acid chemistry be qualified over at least 60 days.

Specimen-to-specimen scatter added a further layer. In SUS304 citric acid at pH 3.2, the range-to-mean ratio of mass loss across six replicate springs was about 1.3, with the worst specimen at 60 °C losing 33 milligrams against 8 for the best — the signature of a few large localised pits dominating each specimen. In glycolic acid at pH 4.0 the same ratio was just 0.38, with all specimens within 22% of the mean, suggesting distributed, more uniform dissolution. A Brown–Forsythe test confirmed the dispersion difference was statistically significant. The authors link this contrast to ligand geometry: multidentate citrate could concentrate attack at clustered sites, while monodentate glycolate spreads it out.

Translating the rates into engineering guidance, the authors took 5 micrometres per year as the allowance on the corrosion rate between days 30 and 60. Citric cleaners at pH 3.2 are limited to 40 °C on both alloys, since both exceed the allowance at 55 °C and above. Raising the cleaner to pH 4.0 extends the SUS304 limit to 55 °C. Glycolic cleaners are where alloy choice matters most: SUS316 holds the allowance up to 60 °C at every glycolic pH tested, whereas SUS304 exceeds it from 40 °C at pH 4.0. Hydrochloric acid, supplying protons but no metal-binding chemistry, stayed within a factor of three of the detection floor at every pH.

The authors are careful about limits. No chloride stress-corrosion cracking was found on any specimen, but the 60-day exposure, small surface area and unloaded geometry all work against crack initiation, so the data do not clear cold-drawn spring wire for 10 wt% NaCl service at 60 °C or above. The reported pH values describe initial solutions only, since free acid was consumed well before day 60 in the aggressive cells. Activation energies could be reliably fitted for only four of eighteen alloy-acid-pH combinations; the best-determined value, 45 ± 19 kilojoules per mole for SUS304 in glycolic acid at pH 4.0, indicates the corrosion is limited at the metal–oxide–solution interface rather than by transport through the liquid. Flow, applied stress and wet–dry cycling of real cleaning service lie outside the static immersion test, and the practical relevance to plant operation remains qualitative.

Subject of Research: Corrosion behaviour of cold-drawn SUS304 and SUS316 stainless spring wire in acidic chloride solutions

Article Title: Corrosion resistance of SUS304 and SUS316 spring wire in multi-acid chloride environments

Article References: Liao, L.-C., Cheng, C.-Y., Yang, Y.-L., Chen, T.-C., & Rajendran, S. (2026). Corrosion resistance of SUS304 and SUS316 spring wire in multi-acid chloride environments. Results in Engineering, 32, Article 113752. https://doi.org/10.1016/j.rineng.2026.113752

Image Credits: AI Generated

DOI: 10.1016/j.rineng.2026.113752

Keywords: SUS304, SUS316, stainless steel, corrosion, spring wire, citric acid, glycolic acid, pitting, clean-in-place, passive film, molybdenum, ASTM G31