Plasma double layer development during high power EUV exposure

Manis Chaudhuri 1 , Pavel Krainov 2 , Dmitry Astakhov 2 , Andrei M. Yakunin 1 , and Mark van de Kerkhof 1,3 · 2026

Read Plasma double layer development during high power EUV exposure by Manis Chaudhuri 1 , Pavel Krainov 2 , Dmitry Astakhov 2 , Andrei M. Yakunin 1 , and Mark van de Kerkhof 1,3 (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-31797.txt.

Introduction

Chaudhuri and four colleagues confirm, in simulation, that plasma double layers form at the edge of regions exposed to intense extreme ultraviolet light. A double layer is a thin structure of separated positive and negative charge that holds a sharp voltage drop. Whether one appears during fast, high-power EUV exposure matters for lithography plasmas.

Three-dimensional particle-in-cell simulations show the layer exists only while the EUV is on, for roughly 70 nanoseconds in their runs, and vanishes soon after switch-off. A threshold beam energy near 0.1 millijoules must be crossed first. Past that point the layer strengthens from weak to strong as power rises. The fingerprint could affect how EUV plasmas are modeled and controlled.

Catalog record

CREATORManis Chaudhuri 1 , Pavel Krainov 2 , Dmitry Astakhov 2 , Andrei M. Yakunin 1 , and Mark van de Kerkhof 1,3
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE19 KB
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2606.31797

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Plasma double layer development during high power EUV exposure Manis Chaudhuri 1 , Pavel Krainov 2 , Dmitry Astakhov 2 , Andrei M. Yakunin 1 , and Mark van de Kerkhof 1,3 Abstract The development of electrostatic plasma double layer (DL) at the boundary of Extreme Ultra-Violet (EUV) exposed and un-exposed region in the bulk volume has been confirmed by 3DPIC (Particle-In-Cell) simulations in the context of fast transient high power EUV exposures. It is found that the DL exists only for short time scale during EUV-ON time period ( ∼ 70 ​ ns \sim 70{\rm ns} ) and disappears soon after EUV is OFF. Such DL fingerprint appears above a certain critical value of EUV beam energy ( ∼ 0.1 ​ mJ \sim 0.1{\rm mJ} ) and it transforms from weak-to-strong DL with further increase of EUV power. Plasma double layers (DL) are self-organized, non-neutral and nonlinear structures consisting of adjacent regions of net positive and negative space charge that sustain a localized electrostatic potential drop within an otherwise quasi-neutral plasma Langmuir1929 ; Alfven_book1981 ; Hershkowitz1985 ; Alfven1986 . Early laboratory and theoretical studies established DL as self-consistent solutions of the plasma kinetic equations, capable of maintaining electric fields over only a few Debye lengths while mediating significant energy transfer between fields and particles Sagdeev1966 ; Block1978 ; Schamel1986 ; Raadu1989 . They have been observed in different laboratory plasmas such as mercury discharge Torven1971 ; Strangeby1973 , Q-machines Sato1976 ; Sato1982 , double plasma (DP) devices Quon1976 ; Sekar1985 , triple plasma (TP) devices Coakley1978 , helicon devices Hairapetian1990 ; Charles2003 to laser pellet ablation plasmas Hora1984 . Such laboratory experiments have demonstrated that DL can form spontaneously as current-driven DL (CDDL) Sato1980 ; Sato1981 ; Cartier.Marlino1984 ; Hershkowitz2005 or current-free DL (CFDL) Perkins1981 ; Hatakeyama1983 ; Hairapetian1988 ; Boswell2006 ; Liberman2006 ; Chen2006 ; Charles2007 ; Takahashi2007 ; Ahedo2009 ; Thakur2009 ; Takahashi2011 . Observations in space plasmas significantly broadened the relevance of DL physics such as discrete auroral arcs Mozer1997 ; Ergun1998 ; Andersson2002 , the solar wind Carlqvist1982 and in planetary magnetospheres Carlqvist1982 ; Temerin1982 . These findings firmly established DL as a natural plasma phenomenon operating across many orders of magnitude in scale Hultqvist1971 ; Torven1976 ; Stenzel1982 ; Yamamoto1985 ; Williams1986 ; Lennartsson1987 ; Lindberg1988 ; Raadu1988 . Figure 1: Spatial distributions of (a) electron density ( N_{e} ), (b) ion density ( N_{i} ) during EUV exposure ( E_{b} = 1mJ) at 40 ns with two boundaries between exposed and unexposed area (c) Spatial distribution of electrostatic potential and electric field at one of the two boundaries mentioned before (d) The spatial distribution of delta ( N_{i}-N_{e} ) density which clearly shows DL formation at the boundary. The DL characteristics along the Line of Interest (LOI) are shown as (e) Delta density ( N_{i}-N_{e} ) variation and (f) the variation of electrostatic potential and electric field Theoretically, DL arises as nonlinear solutions to the Vlasov–Poisson system or reduced fluid models, maintained by a balance between particle trapping, current continuity, dissipation and non-Maxwellian velocity distributions which are often linked to electron holes as well as other phase-space structures Schamel1972 ; Newman2001PRL ; Singh2011 ; Torven1999 ; Forslund1967 ; Okuda1973 ; Joyce1978 ; Newman2001 . The stability of quasi-static DL structures depends on two specific mechanisms: charge separation and pressure balance. Depending on the resulting potential drop, a key distinction is made between “weak” ( v_{d}<v_{th} ) and “strong” ( v_{d}>v_{th} ) DLs where v_{th} and v_{d} are particle thermal speed and drift speeds respectively. In this context, the Particle‑in‑cell (PIC) simulations have been central in establishing DL as robust, self‑consistent solution of the kinetic plasma equations, free from assumptions about distribution functions or closure relations Forslund1967 ; Okuda1973 ; Joyce1978 ; Newman2001 ; Goldman2003 ; Decyk2003 ; Lapenta2010 ; Divin2012 . These simulations reveal that DL acts as localized electrostatic accelerators, efficiently converting field energy into directed particle kinetic energy while regulating current flow. PIC results further indicate that double‑layer stability and strength depend sensitively on boundary conditions, mass ratios, and dimensionality, raising open questions regarding their formation thresholds and interaction with turbulence. The goal of this work is to explore a new parameter regime associated with fast transient DL structural development at the interface of EUV exposed and un-exposed regions at high EUV power using 3DPIC. Figure 2: The variation of delta ( N_{i}-N_{e} ) density within DL for different EUV beam energy is shown in the left panel. The corresponding variations of electrostatic potential and electric field within DL for different EUV beam energy are shown in the right panel. The inset shows plots with zoom-in scales. The kinetic approach has been adopted to address non-Maxwellian electron energy distribution function (EEDF) by solving Boltzmann-Vlasov equations for relevant plasma species: \frac{\partial f_{a}}{\partial t}+{\bf v}\nabla_{\bf r}f_{a}+\frac{q_{a}{\bf E}_{a}}{m_{a}}\nabla_{\bf v}f_{a}=St(f) (1) coupled with Poisson equation: {\bf\nabla}.{\bf E}=-4\pi\rho (2) Here f_{a} is the distribution function for speciaes a (= electron/ion), St(f) is the collision integral with Monte Carlo collision scheme (MCC). {\bf E} is the electric field which is coupled with electrostatic potential ( \phi ) as {\bf E}=-\nabla\phi . The model follows explicit in time energy conserving formulation Langdon.1973.energy-conserving . One of the favorable property of this formulation is that only plasma frequency need to be resolved accurately and the cell size can be much larger than the Debye length Powis.2024.accuracy . That allows to model cooling of the electrons due to collisions with H 2 and avoid the need to have too fine mesh in 3D. A brief overview of the model is presented here and further details are reported in prior works Ast15 ; Ast16_2 . The code rely on the unstructured tetrahedral mesh to represent geometry. The tetrahedral shape is chosen as base element due to its robustness within available meshes for the cases of complex geometries. The Poisson equation is solved by the finite element method with linear base elements. This choice in combination with energy conserving formulation leads to the constant E-field in the cells. That follows from the fact that spatial derivatives of the linear base functions for the potential are constants. To counteract large numerical noise induced by cell-constant E-field a sub stepping scheme in time domain is used. The code checks, if particle crosses the cell boundaries on each time step. If no cell-crossing is detected, then the equation of motion of such particle is computed with standard Verlet integrator. If the particle crosses the cell boundary, than the motion of the particle is computed according to the time spend in each cells. That fractional times are saved for each particle that crosses cell boundaries and used in the velocity update part of the Verlet integrator. That way there is a smooth transition between fast particles, that may cross many cells in one step and slow particles, that stay in the same cell for many time steps. The collisions are sampled for each particle on each time step with help of the Nanbu scheme Nanbu.1994.simple . The cross-section set used to compute collisions probabilities includes collisions of electrons (e) and ions with H 2 . Due to low degree of dissociation (expected to be less than 10^{-4}

Read the complete text