Growing biomass on non-agricultural land could potentially deliver renewable energy services without displacing land from food production, avoiding the social and environmental conflicts associated with bioenergy. A variety of derelict underutilized and neglected land types are possible candidates, sharing a number of challenges for agronomy, including contaminants in soils, potential uptake and dispersion through energy use. Most previous field trials have grown woody biomass species during phytoremediation. Five one-hectare brownfield sites in NE England, were each amended with c.500 t ha(-1) of green-waste compost, planted with short-rotation coppice willow, Miscanthus, reed canarygrass and switchgrass,(1) and then harvested for 3-5 years. Critical issues for the economic and environmental viability of energy production on brownfield land were investigated: The yields achieved on non-agricultural land; the potential for fuel contamination; the suitability for use and potential markets for any biomass produced. RCG appears best suited to the challenging soil conditions found on non-agricultural land, outperforming other species in ease of establishment, cost, time to maturity, yield and contamination levels. Invasive spreading and low melting ash compositions were not observed. Annual yields of 4-7 odt ha(-1) from the second growth season were found consistently across a range of previously-developed, capped or former landfill sites, with a gross annual energy yield of 97 GJ ha(-1) at contamination levels acceptable for domestic pellets. The analogy with marginal agricultural land suggests that this species and approach could help boost biomass production while avoiding the natural capital "nexus" related to global food-fuel-land-water limits. (C) 2015 The Author. Published by Elsevier Ltd. This is an open access article under the CC BY license.