Description
Achieving GeV-class proton beams with high laser-to-proton conversion efficiency remains challenging because the accelerating field rapidly dephases from the ion population and the bunch thermally debunches during sheath-dominated expansion. We propose and numerically demonstrate a dual-pulse micronozzle acceleration (DP–MNA) scheme that alleviates this efficiency–energy trade-off by enforcing source–field synchronization in a confined, source–field-separated target.
In DP–MNA, a tightly focused prepulse extracts and pre-bunches a compact proton front from a hydrogen rod placed at the entrance of an aluminum micronozzle. After a controlled delay $\Delta t$, a larger-spot main pulse deposits most of the laser energy into the nozzle cavity and drives a quasi-static axial electric field. Within a delay-defined synchronization window, the proton front is injected into and remains embedded within a narrow accelerating channel that is advected downstream with the bunch. This phase locking suppresses longitudinal dephasing and thermal debunching, preserving bunch integrity and extending the effective acceleration stage.
Fully relativistic particle-in-cell simulations with the EPOCH code show that, at moderate main-pulse intensities of the order of $10^{21}$ W/cm$^{2}$, DP–MNA reaches GeV-class proton cutoffs ($\varepsilon_{p,\mathrm{max}} \approx 0.8$–$1.0$ GeV) with a total conversion efficiency $\eta \approx 20\%$. Crucially, the efficiency into the application-relevant high-energy component remains exceptionally high: $\eta_{>100} \approx 12$–$13\%$ for protons above 100 MeV, indicating preferential energy loading into a compact, directed proton population rather than quasi-thermal sheath expansion. Compared with an unconfined dual-pulse hydrogen-rod target under identical laser conditions, DP–MNA develops a markedly harder spectral tail; the cutoff extends to $\geq 760$ MeV versus $\sim 460$ MeV without nozzle confinement. A systematic scan over $\Delta t$ and relative pulse energy sharing identifies a robust synchronization band at $\Delta t \approx 0$–$40$ fs, within which both $\eta$ and $\eta_{>100}$ are simultaneously maximized.
These results establish femtosecond timing control combined with geometric confinement—rather than brute-force intensity scaling—as a practical design principle for compact, high-yield GeV-class proton drivers, with direct relevance to pion/muon production, laser-based neutron sources and accelerator-driven subcritical systems.