if i recall correctly, the model, with various different parameters, predicts from about a third to half of input energy being emitted in x-ray.
the 80% figure for the onion capturing the x-rays might be unrealistic. if we capture its energy with photo-voltaic cells, we can get 10 to 20%. if we capture its energy with a heat engine, we can get about 55%. (but i have not fully explored the possibilities).
the present set of parameters for attempting break-even is going to try mostly to capture energy from the exit beam. since it consists of a pulse of charged particles, it can be transformed at high efficiency (today’s high voltage transformers routinely achieve 98% or better).
