Hard Water Leaves No Scum on Rooibos, but It Does Age the Aspalathin
The standing advice is to brew with filtered water because minerals in hard water dull tea. A 2023 study that tested rooibos directly against black and green tea found the usual reason for that advice does not apply to the red bush at all, though a different, real effect on rooibos's own chemistry does.
The short answer: no, not the way it does for real tea, and a study that put rooibos through the exact test settled it directly. Water hardness is not, however, entirely beside the point. It changes something in rooibos, just not the thing the standard advice warns about.
The advice you have probably heard
Search for rooibos brewing tips and one line turns up often: use filtered or soft water, because the minerals in hard water bind to tea's polyphenols and flatten the flavour. The SA Rooibos Council's own brewing guide says exactly this, citing the general chemistry of tea. It is honest about where the claim comes from: general tea chemistry, the kind Bryn Mawr College chemistry professor Michelle Francl lays out in her 2024 book on the subject, not a rooibos-specific test. The council's own page allows that the effect "may" carry over to rooibos "if the water is particularly hard," the same way it does for Camellia sinensis. That is a reasonable guess, carried across from one plant to a different one. It had not, when that page was written, actually been checked.
What "hard water dulls tea" is really describing
The mechanism behind the advice has a name: tea scum, the thin, often iridescent film that can form on the surface of a strong cup of black or green tea, particularly one made with mineral-rich water. It is not dirt and it is not spoilage. It forms when calcium and magnesium ions in hard water complex with the tea's own polyphenols, the catechins and theaflavins that give real tea its astringency and colour, into an insoluble film that sits on the surface and, some tea drinkers believe, dulls the cup underneath it. The chemistry is real. The question this Ministry wanted answered is whether rooibos, a different plant with a different set of polyphenols entirely, does the same thing.
The direct test, on rooibos itself
In 2023, a team at ETH Zurich, publishing in the journal Soft Matter, set out to measure exactly this across several tea types, rooibos among them. They mixed three artificial tap waters to standard hardness levels used in food-science testing: a soft water at 50 milligrams of calcium carbonate per litre, a moderate water at 100, and a hard water at 200, the hard version carrying roughly 2 millimoles of calcium and two-thirds of a millimole of magnesium per litre alongside the bicarbonate that hard water always brings. They brewed black tea, green tea, and rooibos in each, then measured the surface directly with interfacial rheology, a technique built to detect and quantify exactly the kind of thin film tea scum is.
Black and green tea both formed measurable films, strengthening as the water got harder. In soft and moderate water, black tea's film actually ran stronger than green tea's, and only in the hardest water tested did green tea's overtake it. Rooibos formed none, in any of the three. Across all three water hardness levels, the researchers recorded no visible layer and no measurable surface strengthening on the rooibos infusion at all. The mineral-complex film that hard water is known to build on real tea simply did not appear on the red bush, soft water or hard.
Why the red bush skips it
The explanation sits in what rooibos is made of. Tea scum forms because calcium and magnesium ions complex specifically with catechins and theaflavins, the polyphenol family Camellia sinensis carries. Rooibos's polyphenols are a different family entirely, dihydrochalcones, chief among them its own signature compound, aspalathin. The Soft Matter researchers note plainly that the precise polyphenol profile of a tea is what decides whether a film forms, and rooibos's profile does not supply the catechins and theaflavins the reaction needs. No supply, no film, at any hardness the study tested. This is a difference in the molecule, not the water.
Where the water still does something real
None of that means water is irrelevant to rooibos. It means the specific worry, the one the general tea advice was built on, does not transfer. A separate line of research on rooibos's own chemistry finds a real effect, running through a different door entirely.
Aspalathin, rooibos's defining molecule, is chemically fragile in water. A 2015 study that purified it directly from the plant found it held up far better in an acidic solution than a neutral one, 91 percent of it still present after 29 hours in the acidic version, only 45 percent surviving that long in the neutral one. And a 2023 study out of South Africa's Agricultural Research Council, working through the reaction kinetics in detail, found that raising the temperature or the pH of the water both sped up aspalathin's own breakdown, and that the metal ions naturally present in an aqueous solution catalyse that breakdown further. The proof came from the opposite direction: adding a metal-binding chelator, citric acid, citrate, or EDTA, sharply slowed the degradation back down, evidence that it is the metal ions doing the damage in the first place.
This is not the tea-scum mechanism. It is a slower, chemical one, about how long aspalathin itself survives in the cup and in any bottled or stored rooibos, rather than about a film on the surface. A freshly poured, freshly drunk cup is not sitting in solution long enough for this to matter much. A rooibos concentrate or ready-to-drink iced tea sitting on a shelf for weeks is a different story, and it is exactly why manufacturers of bottled rooibos iced tea lean on acids like citric or ascorbic acid, which do double duty as flavour and as aspalathin's protection.
The direct test, now run on rooibos itself
Rooibos's finished cup has since had its own full sensory panel across different waters, the same test already run on its Cape fynbos cousin, honeybush, brewed from the unrelated genus Cyclopia. A 2025 study from the Agricultural Research Council group brewed honeybush with six real water types, deionised water, two bottled spring waters, bottled reverse-osmosis water, brackish water, and ordinary tap water, then ran the infusions through a trained sensory panel and chemical analysis. Most of the waters, tap included, produced infusions indistinguishable from the deionised-water baseline. Only the two waters at the extreme end, one with a high pH of 8.3 and one high in dissolved minerals classed as brackish, produced a measurably different, off-character cup, described as smelling of stewed dates rather than honeybush's usual floral sweetness, with lower phenolic content to match.
The same research group ran the equivalent test on rooibos directly, published in February 2026. Infusions brewed with water in a pH range of 5.3 to 7.8 and with electrical conductivity below 10 millisiemens per metre came out close to the deionised-water baseline, sensorially and chemically. Only water sitting at both extremes at once, a high pH and a heavy mineral load together, left a cup you would actually notice: the floral sweetness faded, a strange smell crept in, a faint salt on the tongue, a heavier body, and a colour gone darker than usual. The pattern holds across both plants: unremarkable water behaves like pure water, and only the genuine extremes do measurable harm.
The cup you drink today, and the jug you keep
Cape Town's own water utility classifies its municipal supply as slightly hard, 17 to 60 milligrams of calcium carbonate per litre depending on the area, comfortably inside the range the Soft Matter study tested and found no scum on. Most ordinary tap water in most places sits well short of the genuine extremes, high alkalinity or true brackish salinity, that the honeybush study flagged as the ones that matter. For a cup you intend to drink today, plain tap water is very unlikely to be the problem the filtered-water advice implies it is.
Where water chemistry genuinely earns a second thought is anything meant to keep: a jug of cold-brewed rooibos left in the refrigerator for days, or a homemade iced tea concentrate. There, a squeeze of lemon or a splash of any acidic mixer is not only flavour. It is doing real, measured chemistry, protecting the same fragile molecule that gives rooibos its antioxidant reputation for as long as the jug lasts. The standard method remains what it always was: a full boil, a generous steep, and a kettle filled from whatever tap is nearest. The Ministry has checked, and the water was never the worry it was made out to be.
Sources
- Tea film formation in artificial tap water, Giacomin, Chen, Hack and Fischer, Soft Matter, 2023, on the interfacial rheology test across three water hardness levels finding no film on rooibos.
- Reaction kinetics of aspalathin degradation and flavanone isomer formation in aqueous model solutions, de Beer, Human, Van der Rijst and Joubert, Food Research International, 2023, on temperature, pH, and metal-ion catalysis of aspalathin's own breakdown in water.
- Isolation of aspalathin and nothofagin from rooibos using high-performance countercurrent chromatography, de Beer et al., Journal of Chromatography A, 2015, on aspalathin's measured stability in acidic versus neutral solution.
- The Sensory and Physicochemical Properties of Honeybush Tea Depend on the Brewing Water, Van Schoor, Moelich, von Pressentin du Preez, Human, de Beer and Joubert, Journal of Food Science, 2025, on the six-water sensory and chemical comparison.
- Sensory and physicochemical quality of rooibos (Aspalathus linearis) infusions depend on brewing water characteristics, de Beer, Moelich, von Pressentin du Preez, Human and Joubert, South African Journal of Botany, vol. 191, 2026, pp. 193 to 203, on the direct rooibos sensory panel across water of varying pH and mineral content.
- An expert chemist's guide to brewing the perfect cup of tea, SA Rooibos Council, on the standing filtered-water advice and its general-tea-chemistry basis.
- Q&A with the author of Steeped, The Chemistry of Tea, World Tea News, on Michelle Francl's general tea-chemistry account of hard water and tea scum. Her book, published by the Royal Society of Chemistry in 2024, is the source SA Rooibos Council's brewing advice draws on.
- City of Cape Town, official reply on municipal water hardness, on the city's own water utility classifying its supply as slightly hard.