Iron ore sintering is an important source of "dioxins", polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs). This paper reports on attempts to identify materials, conditions, and mechanisms responsible for PCDD/F formation (i) by investigating salient properties of ores (viz., with respect to oxidation, condensation, and chlorination of model organics) and (ii) by mimicking the industrial process on a microscale with real-life materials. Principles of Design of Experiments (DOE) are employed. The reactivities of iron ores differ greatly. Limonite/goethite "soft" ore is a very active oxidation catalyst (e.g., for benzene and phenol), a property that may be useful in cleaning up crude sintering process offgases, whereas hematite/magnetite "hard" ore is not. The latter, however strongly promotes condensation of phenol to dibenzofuran. A newly built lab-microscale sintering facility could satisfactorily imitate the large-scale process, in part or as a whole. Results obtained with realistic feed mixtures point at dioxin formation in the sinter bed at levels significant enough to explain a major part of the outputs observed in the real-life process. With approximately 8 ppm (wt) of chloride added as NaCl, the PCDD/F output doubled, but with the same proportion of chlorine administered as C2Cl4, the dioxin output was over 2 orders of magnitude larger. The use of process reverts, etc. containing chlorinated organics should therefore be avoided. PCDD/F congener patterns are also reported and compared with those observed in practice.