Does baking soda ruin your enamel

The short version: the evidence on whether baking soda harms enamel is thin and indirect, but what exists points to abrasion, not chemical erosion, as the real risk.

The evidence you need does not exist neatly. No human trial has compared baking soda toothpaste to a non-abrasive control and measured enamel loss over years. What we have are ex vivo studies and ingredient analyses. The key question is whether baking soda damages enamel chemically (dissolves it, like acid) or physically (scratches it away, like sandpaper). Dental enamel is hydroxyapatite, a calcium phosphate mineral that dissolves below pH 5.5. Baking soda solution is mildly alkaline—about pH 8—so it cannot cause chemical erosion.

The real question is abrasion. The whitening toothpaste review [18] states plainly that the main whitening effect of baking soda pastes is "direct physical removal of extrinsic stain by abrasives." Abrasion depends on particle size, shape, force, and frequency. One clinical trial found that a 65% sodium bicarbonate paste was equivalent to a regular Crest paste at preventing stain over 8 weeks [18], suggesting comparable abrasivity in that formulation. Another trial found a high-concentration baking soda paste did not reduce plaque more than a control paste [16]. So the abrasion is enough to remove stain, but controlled studies show it does not outperform standard toothpaste for plaque.

The 2023 systematic review on mucositis rinses [8] shows that sodium bicarbonate mouthwash is used without reported enamel damage in cancer patients who rinse multiple times daily for weeks. But those patients have soft tissue, not enamel, as the primary concern.

The biggest gap is that nobody has published a head-to-head study measuring enamel thickness loss from baking soda brushing versus standard toothpaste in real human mouths over years. The closest is the whitening paste data [18]: one 12-week study showed a baking soda paste outperformed a regular paste on whitening without reporting enamel damage as an adverse event. But "no reported damage" is not "no damage"—adverse event reporting in short toothpaste trials is not designed to detect slow enamel wear.

My call: baking soda toothpaste is unlikely to chemically erode your enamel, but long-term abrasion risk is unstudied and should not be assumed zero. If you brush with a paste containing large or sharp baking soda particles aggressively, you are risking physical wear. Confidence: moderate—the chemistry is clear, but the long-term abrasion data simply does not exist.