A detailed ray-tracing simulation of the high resolution microbeam at the AIFIRA facility
Introduction
The need for more accurate and higher resolution charged particle probes in certain applications, e.g. Ion Beam Analysis at the sub-cellular level, ion beam lithography, ion beam induced charge, etc., presses the technology of today to its limits. The development of a high resolution beamline on a sub-micrometer scale requires the use of a reliable simulation toolkit able to track charged particles in electromagnetic fields. There exist a number of different softwares which are able to simulate beam transportation; among them the Geant4 toolkit [1], [2] provides the flexibility to include all the parameters in full physical detail. This paper is a continuation in the series of papers describing the development and testing of Geant4 for the simulation of ion transport in fully described beam lines [3], [4], [5], [6], [7].
Section snippets
Beam lines
This paper mainly concerns the nanobeam line under development; however, as simulations are based on measurements on the CENBG microbeam line, a short description of it will be given accordingly. The lens configuration of the microbeam line is of the Dymnikov type, i.e. four quadropoles mounted in a series, allowing equal transverse demagnifications Dx = Dy ≈ 10 [4]. The microbeam line has been in use since 1989 and was reassembled during 2005 at the new accelerator facility on the 10° exit pole of
Magnetic field models
Modelling of the magnetic field in the quadrupoles is of main importance in a precise simulation as its imperfections and intrinsic aberrations will influence directly on beam resolution. The quadrupole field, used to simulate beam transport in the nanobeam line, has previously been based on a 3D-mesh calculated by the OPERA3D® software [7]. The drawback of such a model is the fact that the resolution has to be limited to a certain point, in our case, a granularity of 1 mm. In the current paper,
Measurement
In order to create a realistic model of the initial beam delivered by the accelerator, the emittance in horizontal (x, θ) and vertical (y, ϕ) planes were measured in the microbeam line. This was done by cutting out angular parts of the beam with slits H1 and V1 simultaneously with the microbeam object and scanning the cut-outs with a beam switch over the diaphragm C2 (see Fig. 1). For a more thorough description of this procedure, consult [7]. The slits after the 90o magnet waist were fully
Beam transmitted current
Estimations of transmitted current have been computed earlier under the assumption of a Gaussian angular distribution with a maximum incoming divergence of 0.09 mrad [7]. To compare the impact of the phase space model presented in the previous chapter, new simulations have been performed and the results are compared to the above in Table 3. Energy distribution is computed from a Gaussian distribution of mean 3 MeV and FWHM of 4 × 10−5 in ΔE/E[7].
The estimations using the hollow core phase space
Grid shadow images
A widely used and relative straightforward method to analyse the aberrations of a certain quadrupole system is the grid shadow technique [10], [11], [12], [13]. Image coordinates are obtained from a grid placed at the image plane and the divergence coordinates are acquired from the corresponding grid shadow pattern, cast on an ion luminescent screen placed a couple of hundred millimetres downstream.
For alignment purposes it would be an advantage to have information on grid shadow patterns
Electrostatic deflection
The electrostatic scanning plates for the nanobeam line are currently being designed with the help of Geant4 modelling. Particle deflection was studied while changing the length ratio between x and y scanning plates in order to find the optimum value in terms of scan size for a given voltage under the condition that the scanning plate system total length equals 52 mm and the gap between plates should be greater or equal to 4 mm. The resulting model of the scanning plates was used to simulate the
Discussion
The modelled emittance produces a higher brightness than the one measured on the HVEE 3.5 MeV Singletron™ [9]. However, as the measurement of the emittance in the vertical (y, ϕ) plane gave an unreliable result, we proceeded by assuming a static phase space in all the planes around the optical axis. This might of course not be the case and the brightness will change accordingly by the change of the emittance. The object box under construction (at C0 in Fig. 1) will provide the possibilities of
Acknowledgements
We wish to thank G.W. Grime and M.J. Merchant for the calculation of the OM-50® quadrupole parameters used in Enge’s model.
This work is supported by the CELLION Marie Curie Research Training Network, MRTN-CT-2003-503923.
References (13)
- et al.
Simulation of ion propagation in the microbeam line of CENBG using GEANT4
Nucl. Instr. and Meth. B
(2003) A comparison of ray-tracing software for the design of quadrupole microbeam systems
Nucl. Instr. and Meth. B
(2005)- et al.
Geant4 simulation of the new CENBG micro and nanoprobes facility
Nucl. Instr. and Meth. B
(2006) Monte Carlo simulation of the CENBG microbeam and nanobeam lines with the Geant4 toolkit
Nucl. Instr. and Meth. B
(2007)- et al.
Grid shadow pattern analysis of the Shanghai nuclear microprobe
Nucl. Instr. and Meth. B
(1995) Summary of the workshop on the grid shadow method
Nucl. Instr. and Meth. B
(1995)
Cited by (14)
Magnetic quadrupole simulations for focusing the electron beams emitted by a plasma focus device
2020, Radiation Physics and ChemistryCitation Excerpt :In Table 4 the main keywords used for the charged particle's transport simulation in the quadrupole magnetic field (Bull et al., 2004; Bull, 2011) are reported. MCNP6 offers the possibility of first to apply a magnetic field in the considered geometry and then transports the particle in it (see references Incerti et al., 2006; Incerti et al., 2007; Incerti et al., 2005; Andersson et al., 2008) for more details on Monte Carlo transport simulations in magnetic fields), with a direct magnetic field tracking utilizing numerical integration methods (Goorley, 2012; Goorley, 2014; Booth et al., 2003; Bull et al., 2004; Bull, 2011). For models including the quadrupole fields, the user can add the effect of the magnet fringe fields (Muratori et al., 2015; Zhou, Tang, Chen, Wang; Koscielniak and Johnstone, 1386), approximately inserting the hard-edge kicks to the particle entering and leaving the magnetic field cells.
Single-stage quintuplet for upgrading triplet based lens system: Simulation for Atomki microprobe
2017, Nuclear Instruments and Methods in Physics Research, Section B: Beam Interactions with Materials and AtomsCitation Excerpt :As a result, different shapes of the beam current distribution on the target that represents microprobe resolution were obtained. In contrast with the simulation techniques applied in the work [23], this one has additional advantage because it permits a ratio of the beam spot size at the base of the current distribution to the FWHM size to be varied and the shapes more relevant for two modes of the microprobe to be formed. The same approach was also used for the triplet-based Atomki microprobe with its initial geometry in order to validate our method by comparing the resolution obtained by simulation with the experimental one.
Effect of magnetic quadrupole lens alignment on a nuclear microprobe resolution
2016, Nuclear Instruments and Methods in Physics Research, Section B: Beam Interactions with Materials and AtomsCitation Excerpt :These new systems are characterized by a shorter working distance and/or greater number of lenses in the system. Overall, the separated single-stage [2–10] and two-stage PFSs [11–16] with quadrupoles arranged along the PFS axis at some distances are generally used. Higher demagnification factors are obtained due to the unique PFS geometry.
Five magnetic quadrupole lenses with four power supplies as a single-stage lens system of a nuclear microprobe
2015, Nuclear Instruments and Methods in Physics Research, Section B: Beam Interactions with Materials and AtomsCitation Excerpt :The second way of improving the OHET microprobe implies a two-stage system with more than three lenses. Facilities with 5 magnetic lenses were proposed at Surrey [4] and Bordeaux [5,6]. Experimental results of the CENBG nanobeam line are presented in [7].
Combination of electromagnetic physics processes for microdosimetry in liquid water with the Geant4 Monte Carlo simulation toolkit
2012, Nuclear Instruments and Methods in Physics Research, Section B: Beam Interactions with Materials and AtomsCitation Excerpt :The Geant4 Monte Carlo simulation toolkit [1–3], an object oriented and open-source set of libraries, is able to simulate particle physical interactions with matter using a rich variety of electromagnetic, hadronic and optical physics models. In the past years, Geant4 has been utilized for the simulation of experimental setups dedicated to ion beam analysis, including ray-tracing (for the design of high resolution irradiation facilities at the sub-micrometre level [4]) and dosimetry (at the single- and multi-cell scales using realistic geometrical models of biological cells [5,6]). Geant4 is currently being extended for microdosimetry simulation capabilities in biological media within the framework of the Geant4-DNA project [7], which was originally initiated by the European Space Agency for the modelling of biological effects of radiation during manned long duration space exploration missions.
First results obtained using the CENBG nanobeam line: Performances and applications
2011, Nuclear Instruments and Methods in Physics Research, Section B: Beam Interactions with Materials and AtomsCitation Excerpt :Such a high resolution nanobeam line, based on a two stage focusing system, has been recently developed on the AIFIRA facility (“Applications Interdisciplinaires des Faisceaux d’Ions en Région Aquitaine”) at CENBG (Centre d’Etudes Nucléaires de Bordeaux-Gradignan). A long working distance doublet–triplet configuration was chosen as it presents a good compromise between high demagnification and low spherical aberrations [8,9]. First order calculations allow predicting demagnifications of 65 and 100 in the vertical and horizontal directions, respectively.