Developing the IBA equipment to increase the versatility of the CNA

Accepted Manuscript Developing the IBA equipment to increase the versatility of the CNA Y. Morilla, M.C. Jiménez-Ramos, J. García López, J.A. Labrador, F.R. Palomo, I. Ortega-Feliu PII: S0168-583X(11)00723-3 DOI: 10.1016/j.nimb.2011.07.079 Reference: NIMB 58283 To appear in: Nucl. Instr. and Meth. in Phys. Res. B Please cite this article as: Y. Morilla, M.C. Jiménez-Ramos, J. García López, J.A. Labrador, F.R. Palomo, I. Ortega- Feliu, Developing the IBA equipment to increase the versatility of the CNA, Nucl. Instr. and Meth. in Phys. Res. B (2011), doi: 10.1016/j.nimb.2011.07.079 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. PROGRAM ID: PA47 1 Developing the IBA equipment to increase the versatility of the CNA Y. Morilla (a), M.C. Jiménez-Ramos (a), J. García López (a), J.A. Labrador (a), F.R. Palomo (b) and I. Ortega-Feliu (a) (a)Centro Nacional de Aceleradores, Universidad de Sevilla. Thomas Alva Edison 7, E-41092 Sevilla, Spain. [email protected]; mcy[email protected]; [email protected]; [email protected]; [email protected] (b)Departamento de Ingeniería Electrónica, Escuela Superior de Ingenieros, Universidad de Sevilla. Descubrimientos s/n, E-41092 Sevilla, Spain. [email protected].es Abstract The Centro Nacional de Aceleradores (CNA) in Sevilla-Spain currently contains three accelerators: two Tandems and a compact 18/9 Cyclotron. Next year, a 60Co irradiation system will also be installed. This equipment will expand the range of irradiation experiments in the Centre using both charged particles as well as gamma-ray photons. Originally, based on the 3 MV tandem accelerator, the CNA was considered as a multidisciplinary research Centre devoted basically to the materials characterization by IBA techniques. Nowadays, the validity of our low-energy accelerators has been verified not only to perform analysis but also for irradiation testing, a very promising field of work since radiation effects are aggravated with the technology decreasing scales. In this work, the last innovations and modifications of our laboratory will be briefly described, emphasizing with respect to high energy PIXE experiments and its versatility to carry out irradiation tests. Keywords: irradiation facility, high energy PIXE, low LET irradiation test 1. Introduction The 3 MV tandem accelerator possibilities have been assessed in order to increase its PROGRAM ID: PA47 2 application fields. For instance, the particle radiation effects are a fundamental problem in the life of electronic devices for space applications and other harsh environments, which is aggravated with the technology shrinking towards smaller and smaller scales. Nowadays, the suitability of low-energy accelerators for irradiation testing is a fact. Moreover, the use of a nuclear microprobe is especially helpful to check specific elements in electronic devices [1]. Simple and inexpensive changes have been undertaken in our laboratory since its installation in 1998 [2]. The extension of our laboratory, including a Cyclone 18/9 in 2004, was at the origin to design a portable beam line for modification and irradiation of materials [3]. In general, the research that requires the use of protons and deuterons with energies above 6 MeV, must be carried out in the cyclotron beam transport line. This beamline was installed in one of the target ports with two major purposes: to perform irradiation tests and to complement the analysis of materials by Particle Induced X-ray Emission (PIXE) using high energy protons (HEPIXE) [4, 5]. Recently, the cyclotron versatility has increased with the installation of a new external beam line to carry out the experiments in air, together with the set-up of a degraders system to modify the beam energy at choice. In the first section, we will present some improvements accomplished in our 3 MV tandem accelerator and a realistic availability of species to determine the vulnerability of a certain digital circuit. In the second section, the new external beam line installed in the cyclotron will be briefly described. In addition, preliminary 18 MeV PIXE experiments, both in air and in vacuum, will be discussed. 2. The 3 MV tandem versatility 2.1. Equipment improvements Two different endstations, the microprobe and the portable line for irradiation and ion implantation, have been used to carry out irradiation tests, both in dynamic and static regime. Recently, in the portable line there is the possibility to heat up to 500ºC and to cool down to PROGRAM ID: PA47 3 liquid nitrogen temperature using two specially designed sample holders (manufactured by Criolab-Equipamento Criogénico e de Laboratório Lda.), which can be useful to simulate working extreme conditions. The microprobe focusing system installed at the CNA is based on the Oxford Microbeam endstation OM2000 [6]. The possible ions that can be focused magnetically depend on the focal length of the microbeam system, the ion energy (E), mass (m), charge state (q) and the magnetic rigidity of the lens system available. In our microprobe, the accessible combinations are given by the mass-energy product mE/q2=12 of the quadrupole lenses. Two new stabilized power supplies (maximum 150 A), manufactured by GRELCO, have been installed in order to focus heavy ions, with high charge state, capable to induce electronic single event effects (SEE). Up to now, we have obtained a minimum size of 10x10 µm2 for O5+ beam at 18 MeV. An opto-electrical feedthrough with 51 pins conectors, coupled to one flange, has been placed to bias and monitor “in situ” the response of the devices. The control of very low particle flux is crucial to perform irradiation tests, and this flow can be dramatically decreased using a set of micrometrical slits. Ion fluxes as low as a few particles/cm2s can be readily obtained and measured using a surface barrier silicon detector located at θ=0º. The individual pulses are processed with a simple amplifier plus a SCA and counted using a rate-meter. Once the appropriated flux is attained, the target chip is placed in position to intercept the beam and the flux stability during the irradiation process is monitored by observing the Si K-line in the PIXE spectrum recorded with a Si(Li) detector at 135º. We have already employed this methodology, with different oxygen beams, to carry out SEU (Single Event Upset) experiments on ASIC circuit designed in 0.6 µm integration scale [7]. 2.2. Species availability study After checking the usefulness of our tandem to produce SEU, we plan to start a complete study in order to obtain the SEU Cross Section for circuits using technology of 130 nm PROGRAM ID: PA47 4 integration scale. Our target will be a particular chip, based on VLSI HCMOS9_GP technology (STMicroelectronics), which has been designed and manufactured by the Electronic Engineering Department (ESI-Sevilla University). First of all, we have studied the possible set-up combinations, based on ion range simulations by SRIM2010 [8], in order to cover the necessary LET range [9]. The layers compositions (Al, Polysilicon and Si3N4) and thicknesses have been introduced into the simulation code to obtain ion range data for different species, energies and incident angles, using 1000 MonteCarlo runs per specie. The ions and energies have been chosen in function of a realistic estimation of beam currents from our ion sources and the data bases provided by NEC. Only the angles 0º, 45º and 60º have been taken into account according to the standard JEDEC JESD57. The worst case has been considered, since the ion has to go through the entire structure (15500 Å) up to the active region, which has an equivalent thickness of 2700 Å (Si). For the species and energies available at the CNA we observe a clear dependence of LET with depth, LET(z), so it is calculated the mean value in the active volume corresponding to the silicon well. In Table 1 are shown the definitive combinations which cover the full range of LET from 1 to 50 MeV.cm2/mg, taking <LET(z)> with z between 15500 Å and 18200 Å. As seen, in order to achieve the 130 nm SEU Cross Section experiments in our facility using different species it is absolutely necessary to tilt the sample holder. For these experiments, a new sample holder for our microprobe allowing a monitored full rotation is being manufactured. 3. The cyclotron versatility 3.1. The external beam line Up to now, the experiments in the cyclotron [5], which supplies 18 MeV protons and 9 MeV deuterons, were done in vacuum by using the portable irradiation chamber. Recently, an energy degrader system, consisting in a carousel with capacity for introducing up to four films PROGRAM ID: PA47 5 in different positions, has been assembled to the line to extend the energy range of the accelerator. The films are easily interchangeable and the whole system can be monitored by remote control. On the other hand, an elementary external beam line has been developed to have the possibility to perform experiments in air. The new line is customized with the necessary elements for each experiment without much complexity. One of its key items is the exit flange, which is internally covered with a 5 mm carbon film to avoid its activation. Several nozzles are available, where different materials windows can be adapted, according to the study concerned, and the nozzles can accommodate graphite collimators with various hole diameters. They are essential to perform the PIXE experiments with a reduced beam size (~ 1mm) and beam current (~ pA). Although there is not possibility for scanning, diverse irradiation areas can be achieved by playing with the material and thickness of the exit window and the target distance. The study of the beam homogeneity in the irradiated area was carried out by direct measurements of the beam current on metallic pins symmetrically distributed on a Teflon base, using a Brookhaven 1000c Current Integrator. For instance, for a 50 µm tungsten window, 2 cm diameter nozzle and a target distance of 100 cm, a maximum beam with 12 cm diameter size has been obtained. In that case, the final proton energy due to different losses at the window and air was 13.5 MeV. However, three discrete areas were identified with a current density variation around 30% between them: a hot core with 2 cm of diameter, a less intense ring from 2 to 8 cm, and a softer one from 8 to 12 cm. The current uniformity was found to be better than 10% in each sector, and the estimation for the dose error was lower than 15%. To measure very low fluxes, well below the sensitivity our current integrator (max. 1-10 pA), particle or scintillation detectors can be used. Moreover, an ionization chamber is being designed to be placed before the exit window, in order to cover the direct monitoring of a PROGRAM ID: PA47 6 wide range of beam currents (≤ 104 p/s). 3.2. High Energy PIXE High Energy PIXE presents some advantages compared to the commonly used 1-4 MeV protons, especially when it comes to the detection of high-Z elements and thick targets, as the significant increase of the K- and L-lines X-ray production cross-sections and the decrease of the stopping power at higher proton energies. A strong reduction of the target irradiation damage is expected because lower ion beam currents can be used, and higher depths exploration is possible [10, 11]. Nevertheless, the use of high energy protons produces an enlarged background due to Bremsstrahlung and γ-rays from nuclear reactions [12], ultimately leading to an important decrease in the sensitivity of HEPIXE compared to the standard technique. From our experience with 18 MeV protons, the (p, xn) nuclear reactions at this energy considerably alter the spectra not only by the emission of low energy γ-rays, but also by the emission of the characteristic X-rays of the (Z+1) elements. These effects must be taking into account in the analysis of the spectra. It is especially important in the case of high- Z elements, since the (p, xn) nuclear reactions cross-sections are of the same order of magnitude as the K X-ray lines PIXE cross-sections (Figure 1), and the de-excitations through internal conversion of the produced radioisotopes are favored. From the analytical standpoint, performing the experiments in air adds some advantages over the use of a vacuum chamber: no restricted sample size, lower temperature reached in the target, and reduction of the background in the X-ray spectra. As an evidence of the background decrease, Figure 2 shows the PIXE spectra from a gold standard using 18 MeV protons acquired in air and in vacuum, respectively. The X-ray spectra were obtained with a LEGe detector with 50 mm2 active area, 25 µm thick Be window and placed at θ= 135º. The spectra analysis was carried out using the GUPIXWIN code [13]. As a recent application of our new set-up to perform HEPIXE in air, we can mention the PROGRAM ID: PA47 7 study of “hot particles” (high activity concentration particles) bigger than 1 mm. For this kind of samples, composed mainly of Pu and U, it is essential the analysis of their K-lines due to the low transmission of L-lines through the sample. The use of 18 MeV proton beam is a good option as the ion range is about 2 mm and therefore information from the whole particle can be obtained. 5. Conclusions As a result of different simple improvements, our facilities at the Centro Nacional de Aceleradores have considerably increased their versatility. We have verified that currently our accelerators are valuable tools not only to perform ion beam analysis, but also for irradiation testing in the low LET region. The LET range available, particularly for protons and deuterons, has been also increased with the installation of a variable energy degraders system in the cyclotron. It is particularly interesting the possibility to use an adapted nuclear microprobe, which allows us to evaluate the behaviour under ion irradiation of specific elements inside electronic circuits. Finally, the installation of a new external beam line to perform analysis in air has remarkably reduced the background of the X-ray spectra for HEPIXE experiments. 6. Acknowledgements This work is partially supported by the Projects TEC2010-22095-C03-02 and TEC2010- 22095-C03-01 (Spanish MICIIN). 7. References [1] G. Vizkelethy, B. Doyle, D. Brice, P. Dodd, M. Shaneyfelt, J. Schwank, Instr. and Meth. B 231 (2005) 467-475. [2] J. García López, F.J. Ager, M. Barbadillo Rank, F.J. Madrigal, M.A. Ontalba, M.A. Respaldiza, M.D. Ynsa, Nucl. Instr. and Meth. B 161-163 (2000) 1137-1142. PROGRAM ID: PA47 8 [3] M.A. Respaldiza, F.J. Ager, A. Carmona, J. Ferrer, M. García-León, J. García López, I. García-Orellana, B. Gómez-Tubío, Y. Morilla, M.A. Ontalba, I. Ortega-Feliu, Nucl. Instr. and Meth. B 266 (2008) 2105-2109. [4] J. Tombs, M. A. Aguirre, R. Palomo, J. M. Mogollón, H. Guzmán, J. Nápoles, A. Rodríguez-Pérez, J. A. Rodríguez, A. P. Vega-Leal, Y. Morilla, J. García López, Proceedings of the 9th European Conference Radiation and Its Effects on Components and Systems (RADECS 2007). [5] J. García López, I. Ortega-Feliu, Y. Morilla, A. Ferrero, Nucl. Instr. and Meth. B 266 (2008) 1583-1586. [6] G.W. Grime, M. Dawson, M. Marsh, I.C. McArthur, F. Watt, Nucl. Instr. and Meth. B 54, (1991) 52-63. [7] F.R. Palomo, Y. Morilla, J.M. Mogollón, J. García López, J.A. Labrador, M.A. Aguirre, Nucl. Instr. and Meth. B (2011), in press corrected proof. [8] J. Ziegler, J. Biersack, M. Ziegler, SRIM The Stopping and Range of Ions in Matter, SRIM Co., 2010. Available from <http://www.srim.org>. [9] V. Correas, F. Saigné, B. Sagnes, J. Boch, G. Gasiot, D. Giot, P. Roche, IEEE Transactions on nuclear science, Vol. 54 Nº 6 (2007) 2413-2419. [10] McKee JSC. Intl. J. of PIXE 1993; 3(3): 187-214. [11] A. Denker, J. Opitz-Coutureau, J.L. Campbell, J.A. Maxwell, T. Hopman, Nucl. Instr. and Meth. B 219-220 (2004) 130-135. [12] Johansson SAE, Johansson TB. Nucl. Instr. Meth. 1976; 137: 473-516. [13] GUPIXWIN. http://pixe.physics.uoguelph.ca/gupix [14] Keller KA, Lange J, Münzel H. in Schopper H. (ed.), Q-values and Excitation Functions of Nuclear Reactions, part c: Estimation of Unknown Excitation Functions and Thick Target Yields, Springer Verlag: Berlin, Heidelberg, New York, 1974.