1: \documentstyle[aps,floats,graphicx]{revtex}
2: %\usepackage[dvips]{graphics,monochrome}
3:
4: \begin{document}
5:
6: \newcommand{\bec}{\begin{center}}
7: \newcommand{\ec}{\end{center}}
8: \newcommand{\be}{\begin{equation}}
9: \newcommand{\ee}{\end{equation}}
10: \newcommand{\beqn}{\begin{eqnarray}}
11: \newcommand{\eeqn}{\end{eqnarray}}
12: \newcommand{\bet}{\begin{table}}
13: \newcommand{\ent}{\end{table}}
14: \newcommand{\bib}{\bibitem}
15:
16:
17: %\baselineskip 4.2mm
18:
19:
20: \wideabs{
21:
22: \title{
23: Asymmetric intermixing in Pt/Ti
24: %Interface-anisotropy induced asymmetry of intermixing in bilayers
25: }
26:
27:
28: \author{P. S\"ule, M. Menyh\'ard, L. K\'otis, J. L\'ab\'ar, W. F. Egelhoff Jr.$^{\star}$}
29: \address{Research Institute for Technical Physics and Material Science,\\
30: Konkoly Thege u. 29-33, Budapest, Hungary,sule@mfa.kfki.hu,www.mfa.kfki.hu/$\sim$sule,\\
31: $^{\star}$ National Institute of Standards \& Technology, Gaithersburg, Maryland 20899\\
32: }
33: %\email{sule@mfa.kfki.hu}
34:
35: \date{\today}
36:
37: \maketitle
38:
39: \begin{abstract}
40: The ion-sputtering induced intermixing is studied by Monte-Carlo TRIM, molecular dynamics (MD) simulations,
41: and Auger electron spectroscopy depth profiling (AES-DP) analysis in Pt/Ti/Si substrate (Pt/Ti) and Ta/Ti/Pt/Si substrate (Ti/Pt) multilayers.
42: Experimental evidence is found for
43: the asymmetry of intermixing in Pt/Ti, and in Ti/Pt.
44: In Ti/Pt we get a much weaker interdiffusion than in Pt/Ti.
45: The unexpected enhancement of the interdiffusion of the Pt atoms into the Ti substrate
46: has also been demonstrated by simulations.
47: We are able to capture the essential features of intermixing using TRIM and MD simulations for ion-beam sputtering
48: and get reasonable values for interface broadening which can be compared with the
49: experimental measurements.
50: However, the origin of the asymmetry remains poorly understood
51: yet.
52:
53:
54: {\em PACS numbers:} 79.20.Rf, 66.30.-h, 68.49.Sf, 61.80.Jh, 79.20.Ap \\
55: %{\scriptsize {\em Keywords:} intermixing, interdiffusion, interface, film/substrate, Auger depth profiling, sputtering, computer simulations, thin film, multilayer, ion-solid interaction, molecular dynamics, Ti/Pt
56: %}
57: \end{abstract}
58: }
59:
60: %61.80.Jh Ion radiation effects (for ion implantation, see 61.72.Tt, Vv, Ww
61: %68.49.Sf Ion scattering from surfaces (charge transfer, sputtering, SIMS)
62: %79.20.Ap Theory of impact phenomena; numerical simulation
63: %79.20.Rf Atomic, molecular, and ion beam impact and interactions with surfaces
64: %... ... ... Electron and ion channeling, see 61.85.+p
65: %\scriptsize{
66: %68.43.Jk Diffusion of adsorbates, kinetics of coarsening and aggregation
67: %05.45.-a Nonlinear dynamics and chaos (see also section 45 Classical mechanic
68: %s of discrete systems; for chaos in fluid dynamics, see 47.52.-j)
69: %05.40.Fb Random walks and Levy flights
70: %61.80.Az Theory and models of radiation effects,
71: %61.80.Jh Ion radiation effects,
72: %61.82.-d Radiation effects on specific materials,
73: %61.82.Bg Metals and alloys,
74: %63.20.Ry Anharmonic lattice modes
75: %66.30.-h Diffusion in solids (for surface and interface diffusion, see 68.35.
76: %66.30.Jt Diffusion of impurities
77: %68. Surfaces and interfaces; thin films and low-dimensional systems (structure and nonelectronic properties) (for surface and interface chemistry, see 82.65.+r, for surface magnetism, see 75.70.Rf)
78: %68.35.Bs Structure of clean surfaces (reconstruction)\\
79: %68.35.-p Solid surfaces and solid-solid interfaces: Structure and energetics
80: %68.55.-a Thin film structure and morphology (for methods of thin film deposition, film growth and epitaxy, see 81.15.-z)
81: %68.35.Fx Diffusion; interface formation (see also 66.30.-h
82: %79.20.Rf Atomic, molecular, and ion beam impact and interactions with surfaces\\
83: %81.40.Wx Radiation treatment (particle and electromagnetic) (see also 61.80.-x Physical radiation effects, radiation damage)\\
84: %61.46.+w Nanoscale materials: clusters, nanoparticles, nanotubes, and nanocrystals (see also 36.40.-c Atomic and molecular clusters; for fabrication and characterization of nanoscale materials, see 81.07.-b in materials science)
85: %\\
86: %79.20.Ap Theory of impact phenomena; numerical simulation\\
87: %34.50.Dy Interactions of atoms and molecules with surfaces; photon and electron emission; neutralization of ions (for surface characterization by particle-surface scattering, see 68.49.-h in surfaces, interfaces, thin films, and low-dimensional structures)\\
88: %34.10.+x General theories and models of atomic and molecular collisions and interactions (including statistical theories, transition state, stochastic and trajectory models, etc.)\\
89:
90:
91:
92:
93: \section{Introduction}
94:
95:
96:
97: Bombardment of surfaces by energetic particles often leads to atomic inter-layer mixing
98: \cite{Gnaser,Averback,Ghose,Som,Enrique,Zhou}.
99: The ion-induced inter-layer atomic transport through interfaces (ion-intermixing) has been the subject of
100: numerous studies \cite{Averback,Egelhoff,Cai}.
101: However, recent results indicate that our knowledge on the mechanism of ion-mixing
102: might be incomplete \cite{Sule_PRB05}.
103: %Also, it is necessary to understand the driving force of intermixing
104: %in order to avoid or suppress interdiffusion during thin film growth.
105:
106:
107: Ion-sputtering of surfaces by energetic particles has widely been used for depth profiling
108: of interfacial structures \cite{sputtering2}.
109: Recent advances in the use of ion-sputtering has been attracted considerable
110: theoretical and experimental attention \cite{sputtering1}.
111: In particular, ion-sputtering has been simulated by molecular dynamics
112: using sequential bombardment of metallic surfaces \cite{Karolewski,Thijsse,Zhong,Hanson}.
113: In order to get more insight into processes which govern interdiffusion
114: during repeated ion-bombardments
115: Auger electron spectroscopy (AES) depth profiling analysis (see e.g. the review refs. of \cite{Hofmann}) has been used for the
116: study of
117: broadening
118: of interfaces during ion-sputtering in combination with transmission electron
119: microscopy (TEM) \cite{Gnaser,Menyhard}.
120:
121: There is now also a widespread interest in understanding the asymmetry and the
122: anomalously wide range of intermixing in various diffusion couples \cite{Buchanan,Luo}.
123: It has also been reported that
124: interdiffusion is not driven by bulk diffusion parameters nor by
125: thermodynamic forces (such as heats of alloying) \cite{Sule_NIMB04,Buchanan}.
126: Computer simulations have also revealed that
127: mass-anisotropy of the bilayer governs ion-bombardment induced interdiffusion at the interface \cite{Sule_PRB05} and
128: greatly influences surface morphology development during ion-sputtering
129: \cite{Sule_SUCI,Sule_NIMB04_2}.
130: We also pointed out that interdiffusion takes place via ballistic jumps (ballistic mixing)
131: in various diffusion
132: couples under the effect of ion-bombardment \cite{Sule_NIMB04}.
133:
134:
135:
136:
137: We demonstrate in the present communication that using atomistic simulations
138: we are able to reproduce the experimental interfacial broadening and the asymmetry of intermixing obtained by ion-sputtering.
139: Furthermore, we would like to model and explain the enhancement of intermixing as a function of the anisotropy of the
140: interfaces.
141: We reproduce and partly explain the experimentally found asymmetry of intermixing by TRIM and MD simulations.
142:
143:
144:
145:
146:
147: \section{The setup of the measurements and the simulations}
148:
149: \subsection{The experimental setup}
150:
151: The experimental setup is the following:
152: According to the crossectional TEM (XTEM) results the thickness of the layers in
153: samples are:
154: Pt $13$ nm/Ti $11$
155: nm/Si substrate
156: (denoted throughout the paper as Pt/Ti), and Ta 21 nm (cap layer to prevent oxidation
157: of Ti)/Ti 11 nm/Pt 12 nm/Si substrate (denoted as Ti/Pt).
158: The XTEM images are shown in Fig 1.
159: %are sharp and straight and no asymmetry is observable.
160: For the sake of simplicity we consider our multilayer samples as bilayers and we study the
161: atomic transport processes at the Ti/Pt and Pt/Ti interfaces.
162: Both samples have been AES depth profiled by applying various sputtering conditions.
163: The sample has been rotated during sputtering. In the following we will outline results of $500$ eV
164: Ar$^+$ ion bombardment at an
165: angle of incidence of $10^{\circ}$ (with respect to the surface).
166: The atomic concentrations of Pt, Ti, Ta and Si were calculated by the relative sensitivity method
167: taking the pure material's values from the spectra. The oxygen atomic concentration has been
168: calculated by normalizing the measured oxygen Auger peak-to-peak amplitude to TiO$_2$
169: \cite{Vergara}. The depth scale was determined by assuming that the sputtering yield ($Y_i$) in the
170: mixed layer is the weighted sum of the elemental sputtering yields ($\sum_i X_i Y_i$).
171: %Two typical AES depth profiles recorded on samples of Pt/Ti and Ti/Pt at $500$ eV ion energy,
172: % are shown, respectively in Fig. 2. It is clear without any detailed evaluation that the Pt/Ti
173: %and Ti/Pt transitions are very different.
174: %While the Ti/Pt transition
175: %is a "normal" one, in case of Pt/Ti an unusual deep penetration of Pt to the Ti phase is observed.
176:
177: % We have carried out several experiments, using different parts of the sample
178: %and applying various sputtering conditions.
179: %In all cases we recognized oxygen. The oxygen concentration varied considerably.
180: %The oxygen at the metal/substrate interface is due to the native
181: %oxide on the Si substrate.
182: %Since the bulk level of oxygen slightly correlated with the ion current intensity
183: %(the larger
184: %the ion current intensity the lower the oxygen AES signal) part of the oxygen is
185: %contamination occurring during the AES depth profiling process.
186: %On the other hand the interface broadening does not
187: %correlate with the
188: %concentration of
189: %the oxygen
190: %in the Ti/Pt interface.
191: %Thus we conclude
192: %that the atomic transport is not affected by the presence of the slight oxygen contaminant.
193:
194: % XTEM give reasonable good estimate of the roughness of the as received sample, which was found to be about $\sim 1,5$ nm (intermixing would result in a gradual change of the contrast; since it was missing we estimate that the initial intermixing is less than $\sim 1,5$ nm). Ion-sputtering might cause surface roughening as well as ion mixing. In the case of our experimental conditions (rotated sample, grazing angle of incidence, low relative sputtering yield) the ion bombardment induced roughening is expected to be weak, but cannot be ruled out; thus part of the measured total broadening might be due to roughening.
195: The broadening of the interface is frequently characterizied by the depth resolution. The depth resolution is defined as the distance of points on the depth profile exhibiting 84 \% and 16 \% concentrations. This definiton has been introduced for the "normal" cases, when either the ion mixing or the roughening can be described by a Gaussian convolution resulting
196: in an erf fuction transition in the depth profile. The same definition used for other cases as well, however.
197:
198: If the transition does differ from the $erf$ function (e.g. when the mobility of one of the components of a diffusion couple is much higher than that of the other's) one might give also the distances between points of $84$ \% and $50$ \%, and $50$ \% and $16$ \%.
199: The ratio of these distances gives us the asymmerty of intermixing (shown in Table 1).
200:
201:
202:
203:
204: \subsection{The computational methods}
205:
206:
207: Dynamic TRIM (Transport of Ion in Matter) TRIDYN simulation has been applied to model the ion sputter-removal process (see e.g. ref. \cite{Nastasi}). The input parameters of the code used (TRIDYN \cite{TRIM}), characterizing the pure material, are: atomic number and density.
208: ($\rho_{Ti} = 56.8$ at/nm$^3$, $\rho_{Pt}=66.2$
209: at/nm$^3$), bulk binding energies ($0$ for both materials), and the surface binding energies (SBE) of Ti
210: ($4.88$ eV) and Pt ($5.86$ eV).
211: The simulation provides the atomic concentration along the depth after a given dose of bombarding ions. Having this distribution the Auger intensities of elements present can be calculated using standard equations and parameters like inelastic mean free path (IMFP), backscattering factors, primary current etc
212: (the software provided by S.Tougaard is used to calculate the IMFP based on TPP-2M).
213: Repeating this procedure (simulation of concentration distribution
214: after additional ion bombardment and calculation of the corresponding Auger intensities)
215: until the layer sputtered away results in a simulated depth profile, which
216: might be compared with the measured one. (The complete description of this calculation
217: is given in ref. \cite{MM_TRIM}).
218: As will be seen later the code describes the essence of the experiments, but fails to provide a quantitative agreement. It is not surprizing since we know that this code accounts neither for roughening nor for the recovery processes taking place after the collisional cascade.
219:
220: In order to get more insight to the mechanism of interdiffusion
221: classical molecular dynamics simulations have also been used to simulate the ion-solid interaction
222: (using the PARCAS code \cite{Nordlund_ref}).
223: Here we only shortly summarize the most important aspects.
224: A variable timestep
225: and the Berendsen temperature control is used to maintain the thermal equilibrium of the entire
226: system. \cite{Allen}. The bottom layers
227: are held fixed in order to avoid the rotation of the cell.
228: Periodic boundary conditions are imposed laterarily and a free surface is left for the ion-impacts.
229: The temperature of the atoms in the
230: outermost layers was softly scaled towards the desired temperature to provide temperature control and ensure
231: that the pressure waves emanating from cascades were damped at the borders.
232: The lateral sides of the cell are used as heat sink (heat bath) to maintain the thermal equilibrium of the entire
233: system \cite{Allen}.
234: The detailed description of other technical aspects of the MD simulations are given in \cite{Nordlund_ref,Allen} and details specific to the current system in recent
235: communications \cite{Sule_PRB05,Sule_NIMB04,Sule_SUCI,Sule_NIMB04_2}.
236: Atomic collisions have been simulated in a standard way given in refs. \cite{Nordlund_ref,ZBL}.
237: Recoils have been initialized by giving the kinetic energy of the incoming ion to a lattice atom which is nearby the
238: impact position.
239:
240:
241:
242:
243: \begin{table}
244: \caption[]
245: {
246: The experimental and TRIDYN results for depth resolution and for the asymmetry of intermixing
247: for Ti/Pt and Pt/Ti samples.
248: }
249: %{\scriptsize
250: \begin{tabular}{ccccc}
251: & exp & TRIDYN & asym. (exp) & asym. (TRYDIN) \\
252: \hline
253: Ti/Pt & 2.0 & 3.3 & 0.9 & 0.9 \\
254: Pt/Ti & 7.0 & 4.3 & 2.3 & 2.0 \\
255: %---------------------------------------------------------------
256: \end{tabular}
257: {\small
258: The exp and TRIDYN denote our measured and calculated
259: depth resolutions (nm).
260: In the 3rd and 4th columns the experimental and the calculated
261: asymmetry of mixing are given.
262: }
263: %\label{table}
264: \end{table}
265:
266:
267:
268: %------------------------------------------------------
269: \begin{figure}[hbtp]
270: \begin{center}
271: %\includegraphics*[height=6cm,width=7cm,angle=0.]{fig1a.ps}
272: %\includegraphics*[height=6cm,width=7cm,angle=0.]{fig1b.ps}
273: \caption[]{
274: The crossectional TEM (XTEM) images of the as received samples.
275: {\em Upper panel:} Ta/Ti/Pt/Si, the thickness of the Ta, Ti and
276: Pt layers are $21$, $11$ and $12$ nm, respectively.
277: {\em Lower panel:} Pt/Ti/Si, the thickness of the Pt and Ti
278: layers are $13$ and $10$ nm, respectively.
279: }
280: \label{fig1}
281: \end{center}
282: \end{figure}
283: %------------------------------------------------------
284:
285:
286: We irradiate the bilayers Pt/Ti and Ti/Pt
287: with 0.5 keV Ar$^+$ ions repeatedly with a time interval of 10-20 ps between each of
288: the ion-impacts at 300 K
289: which we find
290: sufficiently long time for the termination of interdiffusion, such
291: as sputtering induced intermixing (ion-beam mixing) \cite{Sule_NIMB04}.
292: The initial velocity direction of the
293: impacting ion was $10$ degrees with respect to the surface of the crystal (grazing angle of incidence)
294: to avoid channeling directions and to simulate the conditions applied during ion-sputtering.
295: We randomly varied the impact position and the azimuth angle $\phi$ (the direction of the ion-beam).
296: In order to approach the real sputtering limit a large number of ion irradiation are
297: employed using automatized simulations conducted subsequently together with analyzing
298: the history files (movie files) in each irradiation steps.
299: In this article we present results up to 200 ion irradiation which we find suitable for
300: comparing with low to medium fluence experiments. 200 ions are randomly distributed
301: over a $2.0 \times 2.0$ nm$^2$ area which corresponds to $\sim 5 \times 10^{15}$
302: ion/cm$^2$ ion fluence
303: and to the removal of few MLs.
304:
305: The size of the simulation cell is $11.0 \times 11.0 \times 9.0$ nm$^3$ including
306: 57000 atoms (with 9 monolayers (ML) film/substrate).
307: At the interface (111) of the fcc crystal is parallel to (0001) of the hcp
308: crystal
309: and the close packed directions are parallel.
310: The interfacial system is a heterophase bicrystal and a composite object of
311: two different crystals with different
312: symmetry is created as follows:
313: the hcp Ti is put by hand on the (111) Pt bulk (and vice versa) and various structures are probed
314: and are put together randomly. Finally the one which has the smallest
315: misfit strain prior to the relaxation run is selected.
316: The difference between the width of the overlayer and the bulk does not exceed $0.2
317: -0.3$ nm.
318: The remaining misfit is properly minimized below $\sim 6 \%$ during the relaxation
319: process so that the Ti and Pt layers keep their original crystal structure and we
320: get an
321: atomically sharp interface.
322: During the relaxation (equilibration) process the temperature is softly scaled down
323: to zero.
324: According to our practice we find that during the temperature scaling down the
325: structure
326: becomes sufficiently relaxed therefore no further check of the structure has been
327: done.
328: Then the careful heating up
329: of the system to $300$ K has been carried out. The systems were free from any serious built-in strain
330: and the lattice mismatch is minimized to the lowest possible level.
331: The film and the substrate are $\sim 2.0$ and $\sim 6.8$ nm thick, respectively.
332:
333:
334:
335:
336: In order to reach the most efficient ion energy deposition at the interface,
337: we initialize recoils placing the ion above the interface by $1$ nm (and
338: below the free surface in the 9 ML thick film) at grazing angle of incidence
339: ($10^{\circ}$ to the surface)
340: with $500$ eV ion energy.
341: In this way
342: we can concentrate directly on the intermixing phenomenon avoiding
343: many other processes occur at the surface (surface roughening, sputter erosion, ion-induced surface diffusion, cluster ejection, etc.) which weaken energy deposition at the interface.
344: Further simplification is that channeling recoils are left to leave the cell
345: and in the next step these energetic and sputtered particles are deleted.
346:
347: We used a tight-binding many body potential, developed by Cleri and
348: Rosato (CR) on the basis of the second moment approximation to the density of states \cite{CR}, to describe interatomic interactions.
349: %This tight-binding approach is formally analogous to the embedded-atom method (EAM) \cite{EAM}.
350: This type of a potential gives a good description of lattice vacancies, including atomic migration
351: properties and a reasonable description of solid surfaces and melting \cite{CR}.
352: Since the present work is mostly associated with the elastic properties,
353: melting behaviors, interface and migration energies, we believe the model used should be suitable for this study.
354: The interatomic interactions are calculated up to the 2nd nearest neighbors
355: and a cutoff is imposed out of this
356: region.
357: This amounts to the maximum interatomic distance of $\sim 0.6$ nm.
358: For the crosspotential of Ti and Pt we employ an interpolation scheme \cite{Sule_PRB05,Sule_SUCI,ZBL}
359: between the respective elements.
360: The CR elemental potentials and the interpolation scheme for heteronuclear interactions
361: have widely been used for MD simulations \cite{Sule_PRB05,Stepanyuk2,Goyhenex,Levanov}.
362: The Ti-Pt interatomic crosspotential of the Cleri-Rosato \cite{CR} type is fitted to the experimental
363: heat of
364: mixing of the corresponding alloy system \cite{Sule_NIMB04,Sule_NIMB04_2}.
365: The scaling factor $r_0$ (the heteronuclear first neighbor distance) is calculated as the average of the elemental first neighbor distances.
366:
367:
368:
369: The computer animations can be seen in our web page \cite{web}.
370: Further details are given in \cite{Nordlund_ref} and details specific to the current system in recent
371: communications \cite{Sule_PRB05,Sule_NIMB04}.
372:
373:
374:
375: \section{Results}
376:
377: \subsection{Experimental results}
378:
379: Two typical AES depth profiles recorded on samples of Pt/Ti and Ti/Pt at eV ion energy,
380: are shown, respectively in Fig. 2. It is clear without any detailed evaluation that the Pt/Ti
381: and Ti/Pt transitions are very different.
382: While the Ti/Pt transition
383: is a "normal" one, in the case of Pt/Ti an unusual deep penetration of Pt to the Ti phase is observed.
384:
385: According to the experiment a relatively weak intermixing is found in Ti/Pt ($\sigma \approx 2.0$ nm) while an unusually high
386: interdiffusion occurs in the Pt/Ti bilayer ($\sigma \approx 7.0$ nm).
387: Moreover we observe a long-range tail in upper Fig 2 for Pt in Pt/Ti while no such tail is found for Ti/Pt
388: for Pt (neither for Ti, see lower Fig 2).
389: The depth profiles also show oxygen presence as well. Does the presence of O influence the result? The answer is no which is detailed bellow.
390:
391: We have carried out several experiments, using different parts of the sample
392: and applying various sputtering conditions.
393: In all cases we recognized oxygen. The oxygen concentration varied considerably.
394: The oxygen at the metal/substrate interface is due to the native
395: oxide on the Si substrate.
396: Since the bulk level of oxygen slightly correlated with the ion current intensity
397: (the larger
398: the ion current intensity the lower the oxygen AES signal) part of the oxygen is
399: contamination occurring during the AES depth profiling process.
400: On the other hand the interface broadening does not
401: correlate with the
402: concentration of
403: the oxygen
404: in the Ti/Pt interface.
405: Thus we conclude
406: that the atomic transport is not affected by the presence of the slight oxygen contaminant.
407:
408:
409:
410: XTEM gives reasonable good estimate of the broadening of the as received sample, which was found to be about $\sim 1,5$ nm (intermixing would result in a gradual change of the contrast; since it was missing we estimate that the initial intermixing is less than $\sim 1,5$ nm). Ion-sputtering might cause surface roughening as well as ion mixing. In case of our experimental conditions (rotated sample, grazing angle of incidence, low relative
411: %------------------------------------------------------
412: \begin{figure}[hbtp]
413: \begin{center}
414: \includegraphics*[height=5.5cm,width=6.6cm,angle=0.]{fig2a.eps}
415: \includegraphics*[height=5.5cm,width=6.5cm,angle=0.]{fig2b.eps}
416: \caption[]{
417: The concentration depth profile as a function of the removed layer thickness (nm)
418: obtained by AES depth profiling analysis
419: using ion-sputtering at 500 eV ion energy ($10^{\circ}$ with respect to the surface) in P
420: t/Ti (upper Fig: 2a) and Ti/Pt (lower Fig: 2b).
421: }
422: \label{fig2}
423: \end{center}
424: \end{figure}
425: %------------------------------------------------------
426: sputtering yield) the ion bombardment induced roughening is expected to be weak, but cannot be ruled out; thus part of the measured total broadening might be due to roughening.
427:
428: Unfortunately, using MD simulations it also hard to account for interface roughening, since the wavelength of interface roughening often exceeds the lateral size of the simulation cell. Therefore, we account only for intermixing, and then we estimate the magnitude of roughening as a difference of the measured broadening $\sigma$ and the simulated one.
429:
430:
431: % The transition between the two layers can be characterized by the depth resolution defined by the distance of points exhibiting $84$ \% and $16$ \% concentrations. This definition of depth resolution was introduced to characterize the transitions, the shape of which are close to $erf$ function. The measured depth resolution can be considered as the broadening of the interface $\sigma$ in those cases when the inelastic mean free path of the Auger signal electrons is much smaller than the depth resolution. This condition holds for the case of Ti and Pt.
432:
433: The measured depth resolution can be considered as the broadening of the interface $\sigma$ in those cases when the inelastic mean free path of the Auger signal electrons is much smaller than the depth resolution. This condition holds for the case of Ti and Pt.
434: Moreover, $\sigma$ is defined as $\sigma=\sqrt{\sigma_{ro}^2+\sigma_{mix}^2}$,
435: where $\sigma_{ro}$ and $\sigma_{mix}$ are the roughening and
436: intermixing components \cite{Hofmann}.
437: Unfortunately we can not extract the components even in the as received samples,
438: hence the $\sigma \approx 1,5$ nm measured by XTEM in the parent samples might include
439: components both from interface roughening and intermixing.
440: Also, we can say not too much about the ratio of $\sigma_{ro}$ to $\sigma_{mix}$
441: after ion-sputtering. We measure an overall value for broadening.
442: We employ TRIM and molecular dynamics simulations to account for the contribution
443: of intermixing $\sigma_{mix}$ to broadening $\sigma$.
444:
445: % Two typical AES depth profiles recorded on samples of Pt/Ti and Ti/Pt at $500$ eV ion energy,
446: % are shown, respectively in Fig. 2. It is clear without any detailed evaluation that the Pt/Ti
447: %and Ti/Pt transitions are very different.
448: %While the Ti/Pt transition
449: %is a "normal" one, in the case of Pt/Ti an unusual deep penetration of Pt to the Ti phase is observed.
450:
451: % According to the experiment a relatively weak intermixing is found in Ti/Pt ($\sigma \approx 2.0$ nm) while an unusually high
452: %interdiffusion occurs in the Pt/Ti bilayer ($\sigma \approx 7.0$ nm).
453: %Moreover we observe a long-range tail in upper Fig 2 for Pt in Pt/Ti while no such tail is found for Ti/Pt
454: %for Pt (neither for Ti, see lower Fig 2).
455: %Later on we will point out that this tail in the AES spectrum is the singature of anomalously long range intermixing
456: %lenghts of Pt in the Ti phase driven by a specific mechanism characteristic of the Pt/Ti bilayer.
457: %We do not show the AES spectra in every detail, e.g. the transitions at the Pt/Si interface.
458:
459:
460:
461: \subsection{Results obtained by MD and TRIDYN}
462:
463: In Table 1 the experimental and TRIDYN results are summarized for the depth resolution
464: and for the asymmetry of mixing.
465: It is clear that the TRIDYN describes the essential features of the experiments. It predicts
466: that broadening is wider when the Pt layer is on the top of the Ti, and that the penetration of
467: the Pt into the Ti is strong, while the intermixing of Ti to the Pt phase is weaker.
468: Considering the absolute values the predicted asymmetry is in excellent agreement with that
469: of
470: given by the experiment.
471: On the other hand the experimental
472: depth resolution is much
473: larger for
474: Pt/Ti than that predicted by TRIDYN.
475:
476: It is well known that this code cannot account for the relaxation occurring after the
477: collisional cascade \cite{ZBL}.
478: Hence we also carried out MD simulations.
479:
480: Further advantage of the MD simulations is the applied
481: many body tight-binding potentials \cite{CR} which are
482: known to be more accurate and reliable than the two-body potentials used in the TRYDIN code \cite{Nastasi,Sule_NIMB04,ZBL}.
483:
484:
485: MD simulations provide $\sigma \approx 8$ ML ($\sim 2.0$ $nm$ and $\sigma \approx 16$ ML ($\sim 4.0$ $nm$ thick interface after $200$ ion impacts, respectively.
486: The computer animations of the simulations together with the plotted broadening values
487: at the interface in
488: Fig 3 also reveal the stronger
489: interdiffusion in Pt/Ti \cite{web}.
490:
491:
492: Moreover, the applied setup of the simulation cell, in particular the $2.0$ nm film thickness is assumed to be
493: appropriate for simulating broadening.
494: Our experience shows that the variation of the film thickness does not affect the
495: final result significantly, except if ultrathin film is used (e.g. if less than
496: $\sim 1$ nm thick film). At around $5$ or less ML thick film surface roughening could affect
497: mixing, and vice versa \cite{Sule_NIMB04_2}.
498: Also, we do not carry out complete layer-by-layer removal as in the experiment.
499: It turned out during the simulations that the ions mix the interface the most efficiently
500: when they are initialized $\sim 1 \pm 0.3$ nm above the interface.
501: This value is naturally in the range of the projected range of the ions.
502: Hence, the most of the broadening is coming from this regime of ion-interface distance
503: .
504: %------------------------------------------------------
505: \begin{figure}[hbtp]
506: \begin{center}
507: \includegraphics*[height=6cm,width=7cm,angle=0.]{fig3.eps}
508: \caption[]{
509: The simulated broadening at the interface in nm as a function of
510: the number of ions at 500 eV ion energy.
511: The ions are initiated from $1.0$ $nm$ above the interface.
512: The error bars denote a statistical uncertainty in the measure of broadening.
513: }
514: \label{fig3}
515: \end{center}
516: \end{figure}
517: %------------------------------------------------------
518: Initializing ions from the surface it takes longer and more ions are needed to
519: obtain the same level of damage and broadening at the interface as it has been found
520:
521: by ion-bombarding
522: from inside the film.
523:
524:
525:
526: Indeed, we find that the measured ion-sputtering induced broadening of $\sigma
527: \approx 2$ nm for Ti/Pt is in nice agreement with
528: the simulated value.
529: Hence we expect that the most of the measured $\sigma$ is coming from intermixing
530: and interface roughening contributes to $\sigma$ only slightly.
531: The nice agreement could be due to that the saturation of intermixing (broadening)
532: during ion-sputtering is insensitive to
533: the rate of mixing in the as received samples
534: ($\sigma_0 \approx 1.5$ nm interface width including the interface roughening
535: in both samples).
536: This is because
537: a sharp interface and a weakly mixed one (before bombardment in the as-received samples) lead to the same magnitude of broadening
538: upon ion-sputtering, because the binary systems reach the same steady state
539: of saturation under the same conditions (ion energy, impact angle, etc.).
540: This is rationalized by our finding that
541: during simulations we start from a sharp interface ($\sigma_0 \approx 0$) and
542: we get a very similar magnitude of $\sigma$ than by AES.
543:
544: \section{Discussion}
545:
546: Asymmetric AES depth profiles (when the broadening of A/B interface is different from that of B/A) have already been observed \cite{Hofmann,Barna}.
547: The asymmetric behavior has been explained by the large relative sputtering yield of the elements (preferential sputtering)
548: \cite{Hofmann,Barna,Berg}.
549: In the present experiment the relative sputtering yield of
550: $Y_{Pt}/Y_{Ti} \approx 0.7$ at $500$ eV (at $1500$ eV we find $Y_{Pt}/Y_{Ti} \approx 0.9$ and a similar rate of broadening)
551: therefore the mechanism is different.
552: The asymmetry of intermixing is also known during thin film growth
553: \cite{Buchanan,MM,Jedryka} as well as during concentration gradient driven intermixing processes \cite{Erdelyi}.
554: Moreover, we found high energy ion-beam mixing results which report
555: strongly asymmetric interdiffusion e.g. in Au/Cu/Au and in Cu/Au/Cu (with thin tracer impurity layers sandwitched between matrix layers), although these results have not been
556: discussed in detail \cite{Averback2}.
557: %However, this kind of a low-energy ion-bombardment induced asymmetric mixing in metallic diffusion couples has not reported yet
558: %at best of our knowledge.
559: However, no reports has been found which present asymmetric IM for ion-sputtering with
560: vanishing difference of the sputtering yield of the constituents.
561: This is interesting result because a new mechanism should be worked out
562: to explain the preferential interdiffusion of Pt in Pt/Ti.
563:
564:
565:
566: The comparison of the measured $\sigma$ with the simulated broadening
567: using the $84-16$ \% rule in both cases can only be carried out with great care.
568: In principle, these values are not comparable directly.
569: However, we make some simple assumptions at this point.
570: We expect that interface roughening has smaller contribution to broadening
571: than intermixing.
572: MD simulations supports this assumption since we find that the
573: simulated $\sigma$ is in the range of the measured values.
574: %This assumption holds because if the wavelength of interface waving would be higher than the nanoscale
575: %than the measured $\sigma$ should be larger than few times of $nm$.
576: %If the wavelenght and the amplitude of interface waving is in the nanoscale than
577: %$\sigma$ could come mostly from roughening.
578: Unfortunately we have no results for the roughening of the samples after ion-bombardment.
579: However, we expect that if the simulated $\sigma_{mix}$ is comparable with the measured
580: $\sigma$, than we can expect that the ion-sputtered $\sigma_{ro} \ll \sigma_{mix}$.
581:
582: According to the simulations, the mixing of the Pt increases with the fluence,
583: at $200$ ions irradiation (which is the highest we reached because of the
584: limited CPU time) $\sigma \approx 40$ $\hbox{\AA}$ was found. This is
585: not a saturation value (in contrast to the Ti/Pt case) hence
586: we can expect further increase in $\sigma$.
587: The simulated values of broadening are purely coming from intermixing.
588: We expect that the rest of measured $\sigma$ is coming from roughening.
589: Nevertheless, MD simulations reproduce the mixing asymmetry and
590: we are able to capture the essential features of the phenomenon.
591:
592:
593:
594: Finally we should also mention that the reason of asymmetry is not fully
595: understood yet.
596: We can rule out the asymmetry of mass effect with the following computer simulation:
597: if we interchange the atomic masses and leaving all other parameters are unchanged we
598: also get the asymmetry of mixing. Hence not mass-anisotropy is
599: responsible for the asymmetry of intermixing.
600: It could also be that some other parameters, such as e.g. atomic size difference
601: could explain asymmetric interdiffusion (there is a large difference in atomic
602: volumes between Ti and Pt, and also the interaction potential of Ti is strongly anharmonic
603: which could cause the observed anomaly of mixing).
604: E.g., it does matter if we ion-bombard the Pt or the Ti film
605: and the ion-induced injection of Pt to Ti is much easier than that of the Ti atoms to the Pt phase.
606: This could be due to e.g. the atomic size difference.
607: However, the verification of this hypothesis goes beyond the scope of the
608: present paper.
609:
610:
611:
612: \section{Conclusions}
613:
614: Performing AES depth profiling on Pt/Ti and Ti/Pt bilayers we have
615: found strong asymmetry of intermixing depending on the succession of the layers.
616: We could reproduce the mixing asymmetry by means of TRIM and MD simulations.
617: We get a nice agreement for interface broadening in Ti/Pt with experiment while for
618: Pt/Ti the discrepancy is relatively large.
619: We conclude from this that interface roughening might has a significant
620: contribution to broadening in Pt/Ti.
621: In Ti/Pt ion-sputtering increases broadening only slightly: in the as received sample
622: we find $\sigma_0 \approx 1.5$
623: nm,
624: and $\sigma \approx 2.0$ nm is measured after AES depth profiling.
625: Although, atomistic simulations reproduce the main features of
626: interdiffusion,
627: the mechanism of the asymmetry of intermixing remains, however, unexplained.
628:
629:
630:
631:
632:
633: {\scriptsize
634: This work is supported by the OTKA grant F037710
635: from the Hungarian Academy of Sciences.
636: We wish to thank to K. Nordlund
637: for helpful discussions and constant help.
638: The work has been performed partly under the project
639: HPC-EUROPA (RII3-CT-2003-506079) with the support of
640: the European Community using the supercomputing
641: facility at CINECA in Bologna.
642: The help of the NKFP project of
643: 3A/071/2004 is also acknowledged.
644: }
645:
646: \vspace{-0.7cm}
647:
648: \begin{thebibliography}{99}
649:
650: \bib{Gnaser}
651: H. Gnaser, {\em Low-energy Ion Irradiation of Solid Surfaces},
652: in Solid-State Physics, Springer (1999)
653:
654:
655: \bib{Averback}
656: R. S. Averback, T. Diaz de la Rubia, Solid State Physics {\bf 51},
657: 281. (1998).
658:
659: \bib{Ghose}
660: S. K. Ghose, D. K. Goswami, B. Rout, B. N. Dev, G. Kuri, and G. Materlik,
661: Appl. Phys. Lett. {\bf 79}, 467 (2001).
662:
663: \bib{Som}
664: T. Som, B. Satpati, P. V. Satyam, and D. Kabiraj,
665: J. Appl. Phys. {\bf 97}, 014305 (2005).
666:
667: \bib{Enrique}
668: R. A. Enrique, K. Nordlund, R. S. Averback, and P. Bellon, J. Appl. Phys. {\bf 93}, 2917 (2003)
669:
670: \bib{Zhou}
671: X. W. Zhou and H. N. G. Wadley,
672: J. Appl. Phys. {\bf 90}, 3359 (2001).
673:
674:
675: \bib{Egelhoff}
676: W. F. Egelhoff, P. J. Chen, C. J. Powell, R. D. McMichael and M. D. Stiles,
677: Prog. Surf. Sci., {\bf 67}, 355 (2001).
678:
679: %\bib{Schukin}
680: %V.A. Schukin, N. N. Ledentsov, D. Bimberg, {\em Epitaxy of Nanostructures},
681: %Springer (2004).
682:
683: \bib{Cai}
684: M. Cai, T. Veres, S. Roorda, F. Schiettekatte, and R. W. Cochrane,
685: J. Appl. Phys., {\bf 95}, 1996 (2004).
686:
687: \bib{Sule_PRB05}
688: P. S\"ule, M. Menyh\'ard, Phys. Rev., {\bf B71}, 113413 (2005).
689:
690: \bib{sputtering2}
691: C. Szakal, J. Kozole, M. F. Russo, Jr., B. J. Garrison, and N. Winograd, Phys. Rev. Lett., {\bf 96}, 216104 (2006),
692: M. G. Dowsett, Appl. Surf. Sci., {\bf 203}, 5 (2003),
693: A. Zalar, S. Hofmann, A. Zabkar, Thin Solid Films, {\bf 131}, 149 (1985).
694:
695:
696: \bib{sputtering1}
697: L. E. Rehn, R. C. Birtcher, S. E. Donnelly, P. M. Baldo, L. Funk,
698: Phys. Rev. Lett., {\bf 87}, 207601-1, (2001),
699: H. Hansen, C. Polop, T. Michely, A. Friedrich, H. M. Urbassek,
700: Phys. Rev. Lett., {\bf 92}, 246106-1 (2004).
701:
702: \bib{Karolewski}
703: M. A. Karolewski, Nucl. Instrum. Meth. in Phys. Res. {\bf 243}, 6 (2006).
704:
705: \bib{Thijsse}
706: B. J. Thijsse, T. P. C. Klaver and E. F. C. Haddeman,
707: Appl. Surf. Sci., {\bf 231-232}, 29 (2004).
708:
709: \bib{Zhong}
710: Y. Zhong, Y. Ashkenazy, K. Albe, R. S. Averback, J. Appl. Phys., {\bf 94}, 4432.
711: (2003).
712:
713: \bib{Hanson}
714: D. E. Hanson, B. C. Stephens, C. Saravanan, J. D. Kress, J. Vac. Sci. Technol.
715: {\bf A19}, 820. (2001).
716:
717: \bib{Hofmann}
718: S. Hofmann,
719: Prog. Surf. Sci., {\bf 36}, 35 (1991),
720: Rep. Prog. Phys., {\bf 61}, 827 (1998).
721:
722: \bib{Menyhard}
723: A. Barna, {\em et al.}, Appl. Surf. Sci. {\bf 242} 375 (2005).
724:
725:
726: \bib{Buchanan}
727: J. D. R. Buchanan, T. P. A. Hase, B. K. Tanner, P. J. Chen, L. Gan, C. J. Powell, and W. F. Egelhoff, Jr.,
728: Phys. Rev. {\bf B66}, 104427 (2002).
729:
730:
731: %\bib{Gomez}
732: %L. G\'omez, C. Slutzky, J. Ferr\'on, J. de la Figuera, J. Camarero, A. L. V\'azquez de Parga, J. J. de Miguel, and R. Miranda, Phys. Rev. Lett., {\bf 84}, 4397 (2000).
733:
734:
735: %Odunuga,Buchanan,Gomez,Stepanyuk2,Tersoff
736: %\bib{Tersoff}
737: %J. Tersoff, Phys. Rev. Lett. {\bf 74}, 434-437 (1995).
738:
739:
740: \bib{Luo}
741: G. M. Luo, Z. H. Mai, T. P. A. Hase, B. D. Fulthorpe, B. K. Tanner, C. H. Marrows, and B. J. Hickey , Phys. Rev. {\bf B64}, 104427-5 (2001).
742:
743:
744: \bib{Stepanyuk2}
745: H. L. Meyerheim, V. Stepanyuk, A. L. Klavsyuk, E. Soyka, J. Kirschner,
746: Phys. Rev. {\bf B72}, 113403 (2005).
747:
748:
749: \bib{Sule_NIMB04}
750: P. S\"ule, M. Menyh\'ard, K. Nordlund, Nucl Instr. and Meth. in Phys. Res.,
751: {\bf B226}, 517 (2004), {\bf B211}, 524 (2003).
752:
753:
754: %\bib{IBAD}
755: %R. A. Gonz\'ales, F. Yubero, J. M. Sanz,
756: %{\em Low Energy Ion Assisted Film Growth}, Imperial College Press, London, (2003).
757:
758:
759: \bib{Ramana}
760: C. V. Ramana, P. Masse, R. J. Smith, and B.-S. Choi, Phys. Rev. Lett.,
761: {\bf 90}, 066101, (2003).
762:
763: \bib{Gomez05}
764: L. G\'omez, C. Slutzky, J. Ferr\'on, Phys. Rev. {\bf B71}, 233402 (2005).
765:
766:
767: %\bib{Odunuga}
768: %S. Odunuga, Y. Li, P. Krasnochtchekov, P. Bellon, and R. S. Averback, Phys. Rev. Lett. {\bf 95}, 045901 (2005).
769: %
770:
771:
772: \bib{Sule_SUCI}
773: P. S\"ule, Surf. Sci., {\bf 585}, 170 (2005).
774:
775: \bib{Sule_NIMB04_2}
776: P. S\"ule, M. Menyh\'ard, K. Nordlund, Nucl Instr. and Meth. in Phys. Res.,
777: {\bf B222}, 525 (2004).
778:
779:
780: %\bib{Martin}
781: %G. Martin, Phys. Rev. {\bf B30}, 1424 (1984),
782: %G. Martin, P. Bellon, Solid State Physics, {\bf 50}, 189 (1997),
783: %R. A. Enrique, P. Bellon, Phys. Rev. {\bf B70}, 224106 (2004).
784:
785: %\bib{Lysenko}
786: %O. V. Lysenko, {\em et al.}, Phys. Rev. Lett., {\bf 89}, 126102-1 (2002).
787:
788: %\bib{Delaire}
789: %O. Delaire, T. Swan-Wood, B. Fultz,
790: %Phys. Rev. Lett., {\bf 93}, 185704-1 (2004).
791:
792:
793: %\bib{Stepanyuk1}
794: %V. S. Stepanyuk, {\em et al.},
795: %V. S. Stepanyuk, D. V. Tsivline, D. I. Bazhanov, W. Hergert, and A. A. Katsnelson
796: %Phys. Rev. {\bf B63}, 235406 (2001).
797:
798:
799: \bib{Nastasi}
800: M. Nastasi, J. W. Mayer, J. K. Hirvonen, {\em Ion-Solid Interactions: Fundamentals
801: and Applications}, Cambridge, (1996).
802:
803:
804:
805: \bib{TRIM}
806: TRIDYN, FZR-317, W. M\"oller and M. Posselt, Forschungzentrum Rossendorf, 01314 Dresden, Germany.
807:
808: \bib{MM_TRIM}
809: M. Menyh\'ard, Surf. Interface Anal. {\bf 26} 1001 (1998).
810:
811: \bib{Nordlund_ref}
812: %K. Nordlund, Comput. Mater. Sci, {\bf 3}, 448. (1995).
813: K. Nordlund, J. Tarus, J. Keinonen, S.E. Donnelly, R.C. Birtcher, Nucl. Instr. and Meth. {\bf B206} 189 (2003),
814: K. Nordlund, {\em et al.}, Phys. Rev., {\bf B57} 7556. (1998),
815: C. G. Zimmermann, M. Yeadon, K. Nordlund, J. M. Gibson, and
816: R. S. Averback, Phys. Rev. Lett. {\bf 83}, 1163 (1999).
817:
818: \bib{Allen}
819: M. P. Allen, D. J. Tildesley, {\em Comupter Simulation of Liquids}, (Oxford Science Publications, Oxford 1989)
820:
821: \bib{Goyhenex}
822: C. Goyhenex, H. Bulou, J.-P. Deville, and G. Tr\'eglia , Phys. Rev. {\bf B60}, 2781 (1999).
823:
824: \bib{Levanov}
825: N. A. Levanov, V. S. Stepanyuk, W. Hergert, D. I. Bazhanov, P. H. Dederichs, A. Katsnelson, and C. Massobrio, Phys. Rev. {\bf B61}, 2230 (2000).
826:
827: \bib{ZBL}
828: W. Eckstein, {\em Computer Simulation of Ion-Solid Interactions}, (Springer, Berlin 1991).
829:
830: \bib{CR}
831: F. Cleri, G. Mazzone and V. Rosato, Phys. Rev. {\bf B47}, 14541 (1993).
832:
833: \bib{web}
834: http://www.mfa.kfki.hu/$\sim$sule/animations/ptti.htm.
835:
836: %\bib{CuCo}
837: %P. S\"ule, {\em et al.}, submitted, cond-mat/0605130
838:
839:
840: \bib{Vergara}
841: L.L. Vergara, I.Vaquila, J. Ferron, Appl. Surf. Sci. 151. 129 (1999).
842:
843: \bib{Barna}
844: A. Barna, M. Menyhard, G. Zsolt, A. Koos, A. Zalar, P. Panjan, J. Vac. Sci. Tech. {\bf A21} 553 (2003).
845:
846: \bib{Erdelyi}
847: %G. L. Katona, Z. Erdélyi, D. L. Beke, Ch. Dietrich, F. Weigl, H.-G. Boyen, B. Koslowski, and P. Ziemann
848: G. L. Katona, {\em et al.},
849: Phys. Rev. {\bf B71}, 115432 (2005),
850: D. L. Beke and Z. Erd\'elyi,
851: Phys. Rev. {\bf B73}, 035426 (2006).
852:
853: \bib{Averback2}
854: S.-J. Kim, M-A. Nicolet, R. S. Averback, D. Peak, Phys. Rev. {\bf B37},
855: 38 (1988).
856:
857: \bib{MM}
858: M. Menyh\'ard, {\em et al.},
859: Appl. Surf. Sci. {\bf 180} 315 (2001).
860:
861: \bib{Jedryka}
862: E. Jedryka, M. W\'ojcik, S. Nadolski, D. Kubinski, M. Parsons, and H. Holloway
863: J. Appl. Phys. {\bf 91}, 7191 (2002).
864:
865:
866: \bib{Berg}
867: S. Berg, I. Katardjiev, Surf. Coat. Tech., {\bf 84}, 353 (1996).
868:
869:
870:
871:
872: \end{thebibliography}
873:
874:
875: \end{document}
876: