The behavior of vaporous hydrogen peroxide (VHP) was examined in clean, room-scale galvanized steel (GS) and polyvinylchloride-coated steel air ducts, to understand how it might be used to decontaminate larger ventilation systems. VHP injected into the GS duct decreased in concentration along the length of the duct, whereas VHP concentrations in the polyvinylchloride coated duct remained essentially constant, suggesting that VHP decomposed at the GS surface. However, decomposition was reduced at lower temperatures (approximately 22 degrees C) and higher flow rates (approximately 80 actual cubic meter per hour). A computational fluid dynamics model incorporating reactive transport was used to estimate surface VHP concentrations where bioaerosol contamination is likelyto reside, and also showed that VHP decomposition was enhanced at bends within the duct, compared to straight sections. Use of G. stearothermophilus indicators, in conjunction with model estimates, indicated that a concentration-contact time of approximately 100 mg/L H2O2(g) x min was required to achieve a 6 log reduction of indicator spores in clean GS duct, at 30 degrees C. When VHP is selected for building decontamination, this work suggests the most efficacious strategy may be to decontaminate GS ducting separately from the rest of the building, as opposed to a single decontamination event in which the ventilation system is used to distribute VHP throughout the entire building.