High-resolution Rutherford backscattering spectrometry with an optimised solid-state detector
Introduction
Rutherford Backscattering Spectrometry (RBS) is well established as an ion beam analysis method, and has found application in a variety of fields such as geoscience, forensics, and materials science [1]. Its principle attraction lies in the ability to analyse a sample in a non-destructive and rapid manner, yielding key quantities such as the standard-free identification and quantification of elements, the evaluation of contaminant profiles, and the measurement of thin film thicknesses. It is particularly well-suited to the measurement of heavy elements in a light matrix, with sensitivities around 10 ppm routinely achieved [2]. However, the rapid advancement of science and fabrication technologies in the 21st century has started to limit its utility. The problem lies in the reduction of sample feature sizes that produce characteristics below the energy resolution of standard RBS systems — typically within the range when using a passivated planar detector (PPD) connected to standard instrumentation [3].
A number of high energy-resolution RBS systems have been demonstrated by several groups in order to study samples which demand enhanced measurement precision. Electrostatic spectrometers have enabled the surface reconstruction of NiSi(111) systems [4] and observation of island formation following InAs epitaxy on GaAs [5]. Likewise, magnetic spectrometers have demonstrated monolayer depth resolution for single crystal PbTe(001) systems [6], and allowed accurate thickness measurements of ultra-thin high- [7] and SiO2 [8] gate dielectrics. In addition, heavy ion time-of-flight spectrometers have resolved indium profiles in InGaAs heterostructures [9], with isotopic resolution for heavier elements such as Ga [10]. While yielding greatly improved relative energy resolutions compared to a PPD, their more widespread adoption may be hindered by their large physical footprint, system complexity, and cost of installation.
We present here a new high energy-resolution detection system designed for easy and cost-effective integration into existing nuclear probe infrastructure. It is based on a modified solid-state detector integrated within a compact, custom-designed charge-sensitive preamplifier that is specifically engineered for ultra-low noise performance. Its functionality is demonstrated with silicon-based systems due to their unmatched dominance in modern semiconductor technology, producing devices with complex features, such as thin-gate stacks [11] and ultra-scaled diffusion barriers [12], that would be difficult to analyse with conventional RBS systems. Furthermore, silicon forms the basis of a number of promising solid-state quantum computer architectures [13], [14], [15] which require sufficient depletion of the nuclear spin 1/2 29Si isotope. By reducing the concentration of 29Si below 800 ppm to create a semiconductor “spin vacuum”, exceptionally long spin coherence times can be achieved [16], [17]. Analysis by high-resolution RBS may represent a promising method to measure this depletion factor.
Section snippets
Detector energy resolution
The energy resolution of an RBS detection system depends on the scattering geometry, properties of the incident ion beam, features of the sample, and finally factors associated with ion–matter interactions between the backscattered ion and detector. Each of these is now described in turn.
Geometrical straggling arises due to the angular dependence of the kinematic factor and the finite solid angle of the detector [18]. This introduces a source of energy broadening of width
Low-noise preamplifier
The custom-designed in-vacuum preamplifier is shown in Fig. 1. Computer-aided design was employed to create a compact, fully enclosed system with dimensions 5 × 3 × 3 cm. The in-vacuum footprint is similar to a conventional PPD and considerably smaller than existing compact high-resolution RBS systems [31], [32]. The design is based on [18], [33] and consists of three key components: the detector, the junction field effect transistor (JFET), and the operational amplifier (op-amp). All
Preamplifier baseline noise
Before application to detect ions, non-destructive evaluation of the system noise baseline took place via the detection of x-rays emitted by a 57Co source. The 14.4 keV emission line was used for the measurement. After correcting for , an FWHM of was determined. This does not represent the ultimate obtainable resolution because a further decrease in can be achieved with modest cooling. This was verified in a separate measurement, but was ultimately not
Conclusion
We have presented a comprehensive treatment of the sources of energy broadening in a typical RBS experiment employing MeV light ions and a solid-state detector. We have shown that optimisation of these sources through the use of a custom-designed detection system incorporating a modified silicon p–i–n photodiode and a low-noise charge-sensitive preamplifier to be a viable method for achieving high-resolution RBS. An energy resolution of was demonstrated using 2.2 MeV He ions,
CRediT authorship contribution statement
S.G. Robson: Conceptualization, Methodology, Software, Investigation, Formal analysis, Writing - original draft. A.M. Jakob: Conceptualization, Methodology, Software, Resources. D. Holmes: Resources. S.Q. Lim: Resources. B.C. Johnson: Methodology, Resources. D.N. Jamieson: Conceptualization, Supervision, Funding acquisition.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
We gratefully acknowledge the technical assistance of S. Gregory and R. Szymanski. We thank T. Schenkel of the Lawrence Berkeley National Laboratory for providing the enriched silicon. This research was funded by the Australian Research Council Centre of Excellence for Quantum Computation and Communication Technology (CE170100012). This work was performed in part at the Melbourne Centre for Nanofabrication (MCN) in the Victorian Node of the Australian National Fabrication Facility (ANFF). We
References (61)
- et al.
The CIBA high resolution RBS facility
Nucl. Instrum. Methods Phys. Res. B
(2006) - et al.
Heavy ion backscattering analysis with a time-of-flight detector
Nucl. Instrum. Methods Phys. Res. B
(1990) - et al.
Time-of-flight telescope for heavy-ion RBS
Nucl. Instrum. Methods Phys. Res. B
(2007) - et al.
Optimizing the rutherford backscattering spectrometry setup in a nuclear microprobe
Nucl. Instrum. Methods Phys. Res. B
(2013) - et al.
High resolution imaging with high energy ion beams
Nucl. Instrum. Methods Phys. Res. B
(1993) - et al.
The novel ultrastable HVEE 3.5 MV singletron accelerator for nanoprobe applications
Nucl. Instrum. Methods Phys. Res. B
(1997) - et al.
Energy resolution of silicon detectors: approaching the physical limit
Nucl. Instrum. Methods Phys. Res. B
(1994) - et al.
A review of ion beam induced charge microscopy
Nucl. Instrum. Methods Phys. Res. B
(2007) - et al.
Observed energy dependence of Fano factor in silicon at hard X-ray energies
Nucl. Instrum. Methods Phys. Res. A
(1999) - et al.
Monte Carlo method for simulating -ray interaction with materials: A case study on Si
Nucl. Instrum. Methods Phys. Res. A
(2007)