The potential role of nuclear fission to meet increased future energy demand while reducing greenhouse gas emissions and controlling nuclear proliferation is assessed. The World Energy Council projection for an environmentally driven future is used, which projects deployment of nearly 3 TW(e) of nuclear generation by 2100, with concurrent reduction of global CO2 emissions to one-third of present levels. We simulate three scenarios based on this demand curve that rely on evolutionary and advanced systems of reactors. The scenarios differ only in fuel cycle choice between once-through, transmutation, and breeding. We show that the cost of nuclear power will likely remain a minimum using the once through fuel cycle, which, we argue, also minimizes proliferation risks. The other two fuel cycle choices have the benefits of decreased waste production and increased uranium resource utilization, but these come at a price that is probably not acceptable unless the cost of repository space increases dramatically, or the cost of building advanced transmuting or breeding reactors can be reduced to a level lower than that of constructing new plants with contemporary technology. The importance of choice of discount rate in allocating resources to advanced nuclear technologies is discussed. The linkage of fuel cycle choice with the international non-proliferation regime is emphasized.
