The widespread loss of stony reef-building coral populations has been compounded by pervasive recruitment failure, i.e., the low or absent settlement and survival of coral juveniles. To combat global coral reef stressors and rebuild coral communities, restoration practitioners have developed workflows to rear and settle vulnerable coral larvae in the laboratory and subsequently outplant settled juveniles back to natural and artificial reefs. These workflows often make use of the natural biochemical settlement cues present in crustose coralline algae (CCA), which can be presented to swimming larvae as extracts, fragments, or live algal sheets to induce settlement. In this work, we investigated the potential for inorganic chemical cues to complement these known biochemical effects. We designed settlement substrates made from lime mortar (CaCO3) and varied their composition with the use of synthetic and mineral additives, including sands, glasses, and alkaline earth carbonates. In experiments with larvae of two Caribbean coral species, Acropora palmata (elkhorn coral) and Diploria labyrinthiformis (grooved brain coral), we saw additive-specific settlement preferences (>10-fold settlement increase) in the absence of any external biochemical cues. Interestingly, these settlement trends were independent of bulk surface properties such as surface roughness and wettability. Instead, our results suggest that not only can settling coral larvae sense and positively respond to soluble inorganic materials, but that they can also detect localized topographical features more than an order of magnitude smaller than their body width. Our findings open a new area of research in coral reef restoration, in which engineered substrates can be designed with a combination of organic and inorganic additives to increase larval settlement, and perhaps also improve post-settlement growth, mineralization, and defense.