Scholarly article on topic 'Material Transformation: Interaction between Nuclear and Electronic Energy Losses'

Material Transformation: Interaction between Nuclear and Electronic Energy Losses Academic research paper on "Materials engineering"

CC BY-NC-ND
0
0
Share paper
Academic journal
Procedia Materials Science
OECD Field of science
Keywords
{"swift heavy ions" / "nuclear energy loss" / "electronic energy loss" / "interaction between nuclear and eelctronic" / defects / sputtering / "SiO2 " / Fe / Ti.}

Abstract of research paper on Materials engineering, author of scientific article — M. Toulemonde, W. Assmann, Y. Zhang, M. Backman, W.J. Weber, et al.

Abstract The interaction between nuclear and electronic energy losses induced by an individual ion are described and illustrated by four different experiments. Each experiment shows the different behaviors of the combined interactions. Defects created by nuclear energy loss are annealed by electronic energy loss in Fe by ions in the GeV energy regime, showing a competitive interaction (1+1<1). The sputtering of Ti is enhanced by electronic excitation induced by ∼100 MeV ions supporting a synergetic interaction (1+1>2). Damage cross section in crystalline and amorphous SiO2 is enhanced by ions in the MeV regime, evidencing a cooperative interaction (1+1>1 and <2). Moreover Molecular Dynamic calculations show that defects created by nuclear and electronic collisions appear to be additive (1+1 =2) in this same range of beam energy.

Academic research paper on topic "Material Transformation: Interaction between Nuclear and Electronic Energy Losses"

CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Procedía Materials Science 7 (2014) 272 - 277

2nd International Summer School on Nuclear Glass Wasteform: Structure, Properties and Long-

Term Behavior, SumGLASS 2013

Material transformation: Interaction between nuclear and electronic

energy losses

M. Toulemondea*, W. Assmannb, Y. Zhangc,d, M. Backmand, W.J. Weberd,c, C. Dufoura

and Z.G. Wange

a CIMAP-GANIL (CEA-CNRS-ENSICAEN-Univ. Caen), Bd H. Becquerel, 14070 Caen (F). b Ludwig-Maximilians-Universität München, Am Coulombwall 1, 85748 Garching (D) c Mat. Sci. & Tech. Div., Oak Ridge National Laboratory, Oak Ridge, TN 37831 (USA) d Dep. of Materials Sci. & Engineering, Univ. of Tennessee, Knoxville, TN 37996 (USA) e IMP, CAS, 509 Nanchang Road, Lanzhou 730000 (R.P. China)

Abstract

The interaction between nuclear and electronic energy losses induced by an individual ion are described and illustrated by four different experiments. Each experiment shows the different behaviors of the combined interactions. Defects created by nuclear energy loss are annealed by electronic energy loss in Fe by ions in the GeV energy regime, showing a competitive interaction (1+1<1). The sputtering of Ti is enhanced by electronic excitation induced by ~100 MeV ions supporting a synergetic interaction (1+1>2). Damage cross section in crystalline and amorphous SiO2 is enhanced by ions in the MeV regime, evidencing a cooperative interaction (1+1>1 and <2). Moreover Molecular Dynamic calculations show that defects created by nuclear and electronic collisions appear to be additive (1+1 =2) in this same range of beam energy.

© 2014Publishedby ElsevierLtd.This isanopen access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/3.0/).

Selection and peer-review under responsibility of the scientific committee of SumGLASS 2013

Keywords: swift heavy ions; nuclear energy loss; electronic energy loss; interaction between nuclear and eelctronic; defects; sputtering; SiO2; Fe; Ti.

* Corresponding author.

E-mail address: toulemonde@ganil.fr

2211-8128 © 2014 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Selection and peer-review under responsibility of the scientific committee of SumGLASS 2013 doi: 10.1016/j.mspro.2014.10.035

1. Introduction

Experimental investigations with heavy projectiles where nuclear (Sn) and electronic (Se) energy losses are interacting together are reviewed. The interaction of the two energy loss phenomena may be either competitive, i.e. the addition of the two is smaller than the effect of each one (1+1<1), or cooperative, i.e. the addition is larger than the effect of each one but lower than the sum of the two (1+1>1 and <2), additive, i.e. equal to the sum of the two, 1+1=2 or synergetic, i.e. the addition is larger than the sum of the two (1+1>2). In order to exemplify the different cases, the results of four separate experimental investigations are resume in the present paper, including model descriptions: 1) defects created by nuclear collisions in Fe and annealed by electronic excitations with GeV ions demonstrates the competitive interaction [1]; 2) the atomic sputtering of Ti induced by ions of ~100 MeV showing a synergistic effect [2]; and 3) damage formation in crystalline [3] and amorphous [4] SiO2 demonstrates a clearly cooperative interaction and possibly additive. The details of the four experiments [1-4] and the applied model descriptions [1,2,4,5] are presented and discussed in the original papers .

2. Defect annealing of Fe

The number of defects Nexp created in Fe by swift heavy ions in the GeV energy regime has been measured by Dunlop and al. [6] using in-situ electrical resistance measurements, and the results are shown as a function of Se in Fig. 1a. They compared Nexp to Nd, the total number of defects created calculated using the binary collision approximation (BCA) theory [6,7] assuming a displacement threshold of 25 eV (Fig. 1a). Within the experimental conditions, it appears that Nexp decreases for Se > 15 keV/nm, while Nd increases. Since such an effect appears very clearly for high values of Se, it has been proposed that there is an annealing effect induced by Se. Within the framework of the inelastic thermal spike model (i-TS) [8,9], it is supposed that the energy deposited by Se on the electrons is thermalized by electron-electron interactions before its transfer to the lattice, leading to a short and significant increase of the lattice temperature. The initial radial distribution of the defects (Ndi(R)) versus radial distance from the ion path was calculated by Wang et al. [7] and a thermal process induced by Se is superimposed on this radial distribution [1] (Fig. 1b). Assuming that motion of the defects can be activated by a thermal process [10], the number of stable defects (Nda(R)) versus radial distance decreases (Fig. 1b). The total number of stable defects after annealing (Nda) is calculated [1,7] for all the irradiations by integrating over R and compared to the experimental results [6] (Fig. 1a). The calculations lead to a decrease of the number of defects (Nda) versus Se like in the experiment. Multiplying Nda by 0.43 [1] to normalized Nda to the experimental points, it is possible to describe the experiment. This factor can be qualitatively explained as the annealing of defects created within the recombination volume [11]. This illustrates the competitive effect between Se and Sn, since defects induced by nuclear collisions are annealed by the electronic excitation. Such competitive phenomenon was also observed in SiC [12,13] by sequential irradiations.

5 10-2- (b)

Nb of defects afte\ annealing N (R)

0 10 20 30

electronic energy loss (keV/nm)

radial distance R (nm)

Fig. 1: (a) number of defects versus Se: The squares (Nexp) are the experimental measurements [1], Nd (black line) the calculated number of defects from binary collision approximation BCA [7], Nda (black dotted line) is the number of defects after annealing process by the electronic energy loss [7,8], Ndac (blue dotted line) is equal 0.43*Nda [8]. (b) Density of defects [7,8] versus radial distance R from the ion path before

(Nd(R), (full black line) and after annealing (Nda(R) (red dotted line).

3. Sputtering of Ti

The sputtering experiment of Ti was performed by Mieskes et al. [2] in ultra high vacuum (UHV, ~10-8 Pa) using Au ions with energies between 100 MeV and 275 MeV and at an incidence angle of 18° relatively to the sample surface. The composition of the surface was continuously analyzed by elastic recoils detection analysis (ERDA) in order to ensure a surface free of oxygen. In this range of energy, Sn decreases from 0.08 keV/nm to 0.04 keV/nm, while Se increases from 22 keV/nm to 30 keV/nm (Fig.2a). The measured sputtering yield, corresponding to the experimental conditions, are presented in fig. 2a showing that it is nearly constant value (dotted red line) within the experimental errors, suggesting that it results from an interaction of Se and Sn. Two models exist to describe the sputtering by Sn. the first is TRIM cascade [14], which calculates the number (An) of sputtered atoms by nuclear collision cascade, and the second is the elastic collisions spike model (ECS) [15], which assumes a transient thermal process induced by the incident ion. The values of An from TRIM are plotted in fig. 2b and are too low relative to the experimental sputtering rate. It is even worse for the ECS model, which yields a value of An less than 0.01 atoms per incident ion (not reported on the Fig. 2b). The i-TS model, previously described for defect annealing in Fe by Se, is extended to calculate the number of sputtered atoms (Ae) by electronic excitations. It also gives a too low sputtering rate [2] (blue dotted line in Fig.2b). Even the sum of An and Ae cannot reproduce the experiment. The only way to interpret the measured sputtering rate is to assume that the thermal process from the nuclear collision acts in synergy with the thermal process from electronic excitation in order to produced the observed number of sputtering atoms (Ae+n). Such a model, including both electronic and nuclear energy deposition, is now called the unified thermal spike model (u-TS) [4]. The sputtering of Ti at high energy is clearly a synergistic effect since Ae+n>

Au energy (MeV) 109

combination of the thermal processes induced by S and Sn

20 25 30

energy loss (keV/nm)

100 150 200 250 300

Au energy (MeV/u)

Fig. 2: (a) Sputtering yield of Ti by Au ions, electronic energy loss (Se) and nuclear energy loss (Sn) versus beam energy. The value of Sn has been

multiplied by 300 in order to fit in the figure. The red dotted line is the mean value of the sputtering rate which is constant within the experimental errors. (b) The sputtering yield deduced from different models is compared to the experimental data, An from Sn (black dotted line), Ae from Se (blue dotted line), Ae + An the sum of the two processes(green dotted line) and Ae+n from a combination of Se and Sn (red line) within

the unified thermal spike model framework.

4. Tracks in amorphous and crystalline SiO2

4.1. Crystalline SiO2

Crystalline SiO2 has been irradiated with Au ions between 0.5 MeV and 10 MeV [3], where the nuclear energy loss is nearly equal to the electronic energy loss. However, while the nuclear energy loss is well known in this energy regime, our present knowledge of the electronic energy loss values versus energy is not well known [16,17], and the differences in predicted values of electronic energy loss can vary by a factor of 2 depending of the beam energy. New measurements of electronic energy loss values [18] support the model developed by Sigmund [17], leading to a smaller value of Se as compared to the one of SRIM [16]. The damage cross section (a) was extracted using channeling Rutherford backscattering analysis of the damage evolution versus fluence and calculated using the Avrami formula [19] (Fig.3a). Due to a specific behaviour of the damage evolution versus fluence ()), it was

possible to deconvolute the nuclear damage cross section which follows a law equal to (1-exp(-an^)A) where an is the nuclear cross section and A the Avrami coefficient equal to 3.5 for crystalline SiO2,from the electronic damage cross section which follows a law equal to (1-exp(-an^)). These cross sections are plotted in fig.3b versus beam energy and compared to the track cross section measured at higher energy by Afra et al. [20]. It is clear that between 3 MeV and 15 MeV cooperative interactions between electronic and nuclear damage processes are observed when comparing with the fitted black line at around 10 MeV. Such cooperative interaction was also observed in spinel [21].

E 1 10-13

E 1 10-13

1 10 1 10 Au energy (MeV) Au energy (MeV)

Fig. 3: (a) damage cross section deduced from C-RBS analysis (open black points) [3] and by small angle x-rays scattering (SAXS) measurements (filled square red points) [20]. (b) Deconvolution of nuclear (blue open points) and electronic cross sections (red open square points) in the energy regime between 0.5 to 10 MeV of Au beam and from SAXS [20](filled square red points) at higher energy. The black lines (a and b) are there to guide the eyes. The dotted lines in (b) are also to guide the eyes and are deduced from an analysis of the experiment to determine the nuclear cross section (circles) and the electronic cross sections (square).

4.2. Amourphous SiO2

The compaction of the amorphous SiO2 by irradiation has been studied by infrared absorption [4,22] with Au ions over the same energy range (from 0.3 to 15 MeV) as for c-SiO2. By a statistical analysis, using a Poisson law to fit the infra-red peak intensity versus fluence, the effective cross section (a) for this compaction can be determined [4]. Such cross sections are measured for different beam energies and plotted versus Au energy in Fig. 4a. Combined with recent measurements of track radius by small angle X-rays scattering [23] and assuming a =^xR2, a U-shaped dependence of the track radius on energy is observed over the energy regime from 0.3 to 168 MeV Au. Such an evolution of the cross sections versus beam energy suggests a combined interaction of Se and Sn. At high energy the i-TS model [23,24] describes quite well the cross section for Au beam energy larger than 22 MeV (Fig. 3a). In the nuclear collision regime damage, the cross sections deduced by SRIM are too low to explain the experimental cross sections [4], and the ECS model [15] only describes the cross section at very low Au energy (Fig. 4a). For the intermediate energy regime of the Au ions, between 3 and 15 MeV, a cooperative action of Se and Sn, calculated with the unified thermal spike (u-TS) model, as done for Ti sputtering, predicts cross sections in good agreement with the experiments (Fig. 4a). This interaction is surely cooperative and possibly additive at ~10 MeV Au ion. Atomistic simulation allow have been used to describe defect formation from ballistic recoils and the i-TS model [23,24]. The binary collision approximation is used to create effective recoil spectra (energy and directions), and the inelastic thermal spike model is used to calculate the thermal energy distribution. The results of these two computational approaches are imported into a molecular dynamics environment, either separately or combined, and the damage is allowed to evolve for 100 ps and subsequently analyzed. Using this approach, it is possible to examine the individual contribution of nuclear and electronic energy deposition mechanisms or simultaneous contribution of both nuclear and electronic energy deposition mechanisms on the evolution of structural damage in the irradiated structure. The resulting defect density in a MD simulation cell of 26x26x26 nm3 size is shown in fig. 4b as a function of ion energy. These results clearly demonstrate mainly there is an additive effect on the total number of defects at intermediate ion energies if we compared the two red curves in Fig. 4b.

100 101 100 1 01

Au energy (MeV) Au energy (MeV)

Fig. 4: (a) Track cross section versus beam energy [4,23] and the inelastic thermal spike (i-TS) model calculations for pure Se [23] (dotted black line), the elastic collisions spike (ECS) model for pure Sn (dotted blue line) [15] and the u-TS model [4] for combining the i-TS and ECS models (red curve)s. (b) Defect density in SiO2 calculated by MD and created 1) by Sn ((blue dotted line) 2) by Se (black dotted line), 3) by the sum of the numbers of defect induced by Se and Sn independently (red dotted line) and 4) by a combination of Se+Sn [25] (red continuous line).

5. Conclusion

The mechanism driving the material transformation by irradiation is always a complex problem. Interaction between damage processes from nuclear and electronic energy losses can occur for individual ions [1-4,21]. These interactions can be classified in four groups: competitive [1,12,13], cooperative [3,4,21], additive in the defect creation [5] or synergistic [2] ways. In this paper, only synchronous interactions are presented, but such interaction can appears by sequential [12,26,27] or simultaneous [28] irradiations. Synchronous interaction is not expected in these last cases.

Acknowledgements

This synthesis would not have been possible without the collaboration of the researchers participating in the different papers 1 to 5 during the last 15 years: E. Paumier and H. Kucal from the CIMAP laboratory, Caen (France), H.D. Mieskes and F. Grüner from LMU, Garching (Germany), S.M.M. Ramos and B. Canut from University of Lyon I, Lyon (France), H. Bernas, C. Clerc and J. Chaumont from CSNSM, Orsay (France), C. Trautmann from GSI, Darmstadt (Germany), G.S. Li, V. Shutthanandan from PNNL Richland (USA), T.F Yan, Y.G. Wang from Peking University, Beijing (R.P. China), P. Kluth from ANU, Canberra (Australia), F. Djurabekova, K. Nordlund and O.H. Pakarinen from University of Helsinki, Helsinki (Finland). I thank all of them very deeply.

References

[1] C. Dufour, Z.G. Wang, E. Paumier and M. Toulemonde Bull. Mat. Sei. 22 (1999) 671

[2] H. D. Mieskes, W. Assmann, F. Grüner, H. Kueal, Z. G. Wang and M. Toulemonde Phys. Rev. B 67 (2003) 155414

[3] M. Toulemonde, S.M.M. Ramos, H. Bernas, C. Clere, B. Canut, J. Chaumont and C. Trautmann Nuel. Instr. Meth. B 178(2001)331

[4] M. Toulemonde,W.J. Weber, G.S. Li, V. Shutthanandan, P. Kluth, T.F. Yang, Y.G. Wang and Y. Zhang, Phys. Rev. B 83 (2011) 054106

[5] M. Backman, F. Djurabekova, O.H. Pakarinen, K. Nordlund, Y. Zhang, M. Toulemonde and W.J. Weber Nuel. Instr. Meth. B 303 (2013) 129

[6] A. Dunlop, D. Lesueur, P. Legrand and H. Dammak Nuel. Instr. Meth. B 90 (1994) 330

[7] Z.G. Wang, C. Dufour, E. Paumier and M. Toulemonde Nuel. Instr. Meth. B 115(1996) 577

[8] C. Dufour, A. Audouard, F Beuneu, J Dural, J P Girard, A Hairie, M Levalois, E Paumier and M Toulemonde J. of Phys.: Cond. Matt. 5 (1993) 4573

[9] Z.G. Wang, C. Dufour, E. Paumier and M. Toulemonde J. of Phys.: Cond. Matt. 6 (1994) 6733 and 7 (1995) 2525.

[10] G.H. Vineyard, Rad. Eff. 29 (1976) 245.

[11] M. Nakagawa, W. Mansel, K. Böning, P. Rosner, and G. Voglt Phys. Rev. B 19 (1979) 742

[12] A. Benyagoub, A. Audren, L. Thome, and F. Garrido, Appl. Phys. Lett. 89 (2006) 241914

[13] M. Baekman, M. Toulemonde, O.H. Pakarinen, N. Juslin, F. Djurabekova, K. Nordlund, A. Debelle and W.J. Weber, Comp. Mat. Sei. 67 (2013) 261

[14] J.P. Biersack, Nucl. Instr. Meth. B 27 (1987) 21, J. P. Biersack, computer code TRIM-CASCADE, HMI, Berlin, Germany, 1992.

[15] P. Sigmund and C. Claussen, J. Appl. Phys. 52 (1981) 990

[16] J.F. Ziegler, J.P. Biersack, and M.D. Ziegler, The Stopping and Range of Ions in Solids (SRIM Co., 2008).

[17] P. Sigmund, Eur. Phys. J. D 47 (2008) 45.

[18] K. Jin, Y. Zhang, H. Xue, Z. Zhu, W.J. Weber Nucl. Instr. Meth. B 307 (2013) 65

[19] M. Avrami, J. Chem. Phys. 9 (1941) 177

[20] B. Afra, M.D. Rodriguez, C. Trautmann, O.H. Pakarinen, F. Djurabekova, K. Nordlund, T. Bierschenk, R. Giulian, M.C. Ridgway, G. Rizza, N. Kirby, M. Toulemonde and P. Kluth J. of Phys.: Condens. Matt. 25 (2013) 045006

[21] R. Devanathan, N. Yu, K.E. Sickafus, M. Nastasi Nucl. Instr. Meth. B 127 (1997) 608

[22] M. C. Busch, A. Slaoui, and P. Siffert, E. Dootyhee and M. Toulemonde J. Appl. Phys. 71(1992)2596

[23] P. Kluth, C. S. Schnohr, O. H. Pakarinen, F. Djurabekova, D. J. Sprouster, R. Giulian, M. C. Ridgway, A. P. Byrne, C. Trautmann, D. J. Cookson, K. Nordlund, and M. Toulemonde Phys. Rev. Lett. 101 (2008) 175503

[24] M. Toulemonde, W. Assmann, C. Dufour, A. Meftah, C. Trautmann Nucl. Instr. Meth. B 277 (2012) 28

[25] M. Backman, F. Djurabekova, O.H. Pakarinen, K. Nordlund, Y. Zhang, M. Toulemonde and W.J. Weber J. Appl. Phys. 45 (2012) 505305

[26] A. Debelle, M. Backman, L. Thome, W. J. Weber, M. Toulemonde, S. Mylonas, A. Boulle, O. H. Pakarinen, N. Juslin, F. Djurabekova, K. Nordlund, F. Garrido, and D. Chaussende Phys. Rev. B 86 (2012) 100102

[27] P. Trocellier, S. Miro, Y. Serruys, E. Borda, H. Martin, N. Chaäbane, S. Pellegrino, S. Vaubaillon and J.P. Gallien Mater. Res. Soc. Symp. Proc. 1298 (2011) 153