The first goal of the proposed research is to produce, characterize and simulate nonlinear plasma accelerator structures that are larger (hundreds of microns) and more amenable to injection of low-energy spread (ΔE/E << 1 %) electron bunches from an external linac --- and thus more likely to yield the high-quality GeV bunches required for x-ray free-electron lasers (XFELs) and colliders --- than previously studied laser wakefield accelerators (LWFAs). The research will utilize unique terawatt CO2 laser and 70 MeV linac instrumentation at the Accelerator Test Facility (ATF) at Brookhaven National Laboratory (BNL). The research would use our dedicated experimental station (built using our current DoE HEP funding) at ATF beamline to develop a stable plasma-source with advanced diagnostics and an external electron injector with parameters suitable for plasma accelerators as well as for plasma-bubble diagnostics. The ATF plans to advance CO2 laser pulse duration to 200-500 fsec and peak power to 10-20 TW open a clear path toward ultimate goal of this project - CO2-laser-driven wakefield accelerators with high quality electron beam. A significant part of this research will involve our undertaking advanced simulations of CO2-laser-driven plasma wakefield and electron acceleration. The most important feature of ATF’s plasma accelerators will be the relatively large size of the bubble (~ 300 μm), the properties of which can be studied precisely via external e-beam injection. The combination offers a sound foundation for generating very stable high-quality beams in plasma-driven accelerators, with parameters comparable to those realized in modern conventional accelerators. The proposed work will yield the first comprehensive study of bubble-regime LWFAs driven by CO2 lasers.