Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.issp.ac.ru/libm/papers/245.pdf Äàòà èçìåíåíèÿ: Tue Mar 5 21:37:39 2013 Äàòà èíäåêñèðîâàíèÿ: Fri Feb 28 00:29:15 2014 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: http www.badastronomy.com phpbb index.php

J Mater Sci (2011) 46:4349­4353 DOI 10.1007/s10853-011-5322-1

IIB 2010

Inversed solid-phase grain boundary wetting in the Al­Zn system
S. G. Protasova · O. A. Kogtenkova · B. B. Straumal · P. Zie a · B. Baretzky b

Received: 12 August 2010 / Accepted: 24 January 2011 / Published online: 8 February 2011 ñ Springer Science+Business Media, LLC 2011

Abstract The microstructure of binary Al­10 at% Zn and Al­15 at% Zn alloys after long anneals (800­4000 h) was studied between 190 and 258 °C. The contact angles between (Zn) particles and (Al)/(Al) grain boundaries (GBs) were measured. They decrease with decreasing temperature. First (Al)/(Al) GBs completely wetted by the second solid phase (Zn) appear below TwsAl0% = 205 ± 5 °C. Above TwsAl0% = 205 ± 5 °C all (Al)/(Al) GBs are incompletely wetted by (Zn) solid phase. The extrapolation of the maximal contact angle h to zero permits to obtain the TwsAl100% = 125 ± 10 °C. Below this line all (Al)/(Al) GBs has to be completely wetted by (Zn) solid phase.

Introduction Thin equilibrium GB or surface films in the one-phase area of a bulk phase diagram were first considered by Cahn [1] and Ebner and Saam [2]. They proposed the idea that the transition from incomplete to complete surface wetting is a
S. G. Protasova à O. A. Kogtenkova à B. B. Straumal (&) Institute of Solid State Physics Russian Academy of Sciences, Chernogolovka, Russia 142432 e-mail: straumal@mf.mpg.de; straumal@issp.ac.ru S. G. Protasova à B. B. Straumal ¨ Max-Planck Institut fur Metallforschung, Heisenbergstraúe 3, 70569 Stuttgart, Germany P. Zie a b Institute of Metallurgy and Materials Science, Polish Academy of Sciences, Reymonta St. 25, 30-059 Cracow, Poland B. B. Straumal à B. Baretzky ¨ ¨ Karlsruher Institut fur Technologie (KIT), Institut fur Nanotechnologie, Hermann-von-Helmholtz-Platz 1, 76344 Eggenstein-Leopoldshafen, Germany

phase transformation. Later this idea was successfully applied for GBs, and also old data on GB wetting were reconsidered from this point of view [3­6]. From the bulk Al­Zn phase diagram (Fig. 1a) it is obvious that the Al­Zn system belongs to the ``classical'' Cahn's systems with a critical point for a binary solution. GB wetting phase transformation proceeds at the temperature TwGB where GB energy rGB becomes equal to the energy 2rSL of two solid/ liquid interfaces (Fig. 2a). Above TwGB GB is substituted by a layer of the melt. As a result, the new tie-line appeared in the two-phase area of a bulk phase diagram (Fig. 1a), namely that of the GB wetting phase transition. GBs can also be ``wetted'' by a second solid phase, the reversible transition from incomplete (Fig. 2c) to complete (Fig. 2d) solid phase wetting was observed for the first time in the Zn­Al system at a certain temperature Tws [7]. The transition from incomplete to complete GB wetting by a liquid phase always proceeds with increasing temperature [3­6]. It is easy to understand because the entropy of a liquid phase is higher than that of a solid one. Therefore, 2rSL has a good reason to decrease with increasing temperature steeper than rGB (see scheme in Fig. 2a). In the first studies on the GB wetting by a second solid phase the same tendency was observed [7, 8]. Namely, the (Zn)/(Zn) GBs became completely wetted by the (Al) solid phase with increasing temperature [7]. The (Al)/(Al) GBs also became completely wetted by the Al3Mg2 phase with increasing temperature [8]. However, there is no unambiguous reason for the GB wetting by a second solid phase, why the transition from incomplete (Fig. 2c) to complete (Fig. 2d) GB wetting cannot proceed by decreasing temperature. Recently the authors observed the GB wetting followed by the dewetting in the Co­Cu system [9]. This phenomenon occurs close to the Co Curie point, and it is known that if a paramagnet matrix becomes ferromagnetic, the additional

123


4350 Fig. 1 a Al­Zn phase diagram constructed using the data [7, 14­17, 20]. Thick lines denote bulk phase transformations [19]. Thin lines denote GB phase transformations [7, 14­17, 20]. Open triangles denote TEM and DSC data for the GB prewetting line [15, 17]. Crosses mark the experimental points for the measurements of contact angle h. The inset shows the Zn-rich corner of the diagram. b and c Micrographs illustrating the morphology of GB (Zn) precipitates at b 260 °C and c 190 °C. (Al) matrix appears black

J Mater Sci (2011) 46:4349­4353

attraction between grains may ``squeeze'' the diamagnetic wetting phase from a GB [10­13]. The goal of this study is to check, whether the transition from incomplete to complete GB wetting by a second solid phase can proceed by decreasing temperature in ``pure'' case, without any additional phase transformations in a matrix like in [9]. In this case the temperature dependences of the GB energy rGB and energy of two solid­solid interfaces 2rSS would also intersect at Tws (Fig. 2b). However, in this case GB becomes completely wetted by a second solid phase below Tws and not above Tws like in refs. [7­9]. The authors started to search for this phenomenon in the Al­Zn alloys. First, it was obtained some preliminary results in the previous investigations of GB and triple junctions wetting phenomena in these alloys [7, 14­17]. Second, the nanograined Al­Zn alloys reveal the unusual ductility increase by the decreasing temperature [18].

Experimental Al-based alloys with 10 and 15 at% Zn were investigated. They were prepared of high purity components (5N5 Al

and 5 N Zn) by vacuum induction melting. The 2 mm thick slices were also cut from the [ 10 mm cylindrical Al­Zn ingots and sealed into evacuated silica ampoules with a residual pressure of approximately 4 9 10-4 Pa at room temperature. Samples were annealed at temperatures 190 °C (4000 h), 205 °C (1200 h), 242 °C (1000 h), 253 °C (800 h), and 258 °C (800 h), and then quenched in water. The accuracy of the annealing temperature was ± 2 °C. After quenching, samples were embedded in resin and then mechanically ground and polished, using 1 lm diamond paste in the last polishing step, for the metallographic study. After etching, samples were investigated by means of the light microscopy, and scanning electron microscopy (SEM). SEM investigations have been carried out in a Philips XL30 scanning microscope equipped by the LINK ISIS energy-dispersive spectrometer produced by Oxford Instruments. Light microscopy has been performed using Neophot-32 light microscope equipped with 10 Mpix Canon Digital Rebel XT camera. A quantitative analysis of the wetting transition was performed adopting the following criterion: every (Al)/(Al) GB was considered to be completely wetted only when a layer of (Zn) solid solution had covered the whole GB (Figs. 1b, 2c); if such a layer

123


J Mater Sci (2011) 46:4349­4353 Fig. 2 a, b Scheme of the temperature dependence of GB energy rGB and energy of two wetting interfaces 2rSL and/or 2rSS. a Transition from incomplete to complete wetting by a liquid phase at TwGB with increasing temperature. b Transition from incomplete to complete wetting by a second solid phase at Tws with decreasing temperature. c Scheme of the polycrystal with GBs incompletely wetted by a second solid phase (black). rGB \ 2rSS, second solid phase forms individual particles along GBs and in the triple junctions of the matrix solid phase, the contact angle h [ 0. d Scheme of the polycrystal with GBs completely wetted by a second solid phase. rGB [ 2rSS, second solid phase forms continuous layers along GBs, the contact angle h = 0

4351

appeared to be interrupted, the GB was regarded as incompletely (Figs. 1c, 2d) wetted. In case of incompletely wetted (Al)/(Al) GBs the contact angle h between (Al)/(Al) GB and (Zn) solid solution was measured (Figs. 1b, 2c). At least 100 GBs were analysed at each temperature. Typical micrographs obtained by SEM are shown in Figs. 1b, c.

Results and discussion In this study the authors measured the mean contact angle between (Al)/(Al) GBs and GB particles of (Zn) phase in the Al­10 at% Zn and Al­15 at% Zn polycrystals after long anneals below the monotectoid temperature Tmon = 277 °C (crosses in Fig. 1a). In Fig. 3 the temperature dependences of the maximal hmax (circles), mean hmean (diamonds), and minimal hmin (hexagons) values of the contact angle h are shown. h decreases with decreasing temperature. The value hmin reaches zero at TwsAl 0% = 205 ± 5 °C. The extrapolation of hmean to zero gives TwsAl 50% = 167 ± 10 °C. Figure 1b, c shows the shape of GB Zn particles at 260 and 190 °C. The extrapolation of the hmax to the low temperatures permits to estimate the temperature of the GB wetting transition as TwsAl100% = 125 ± 10 °C. It means that at the room temperature the condition of complete wetting is fulfilled for all (Al)/(Al) GBs. In other words, all (Al)/(Al) GBs should be completely substituted by the Zn layer. As a result,

two new lines of GB phase transformations appear in the Al­ Zn phase diagram (Fig. 1a). The first tie-line is the TwsAl0% = 205 ± 5 °C. Above this line all (Al)/(Al) GBs are incompletely wetted by (Zn) solid phase. Below this line the first (Al)/(Al) GBs completely wetted by (Zn) solid phase appear (Fig. 1c). The second tie-line is the TwsAl100% = 125 ± 10 °C. This is the result of extrapolation of the upper plot hmax(T) in Fig. 3. Below this tie-line TwsAl100% = 125 ± 10 °C all (Al)/(Al) GBs has to be completely wetted by (Zn) solid phase. Above this tie-line first (Al)/(Al) GBs appear which are incompletely wetted by (Zn) solid phase. Other lines of GB phase transformations obtained recently are also shown in the Al­Zn phase diagram (Fig. 1a) For example, in the (Al) ? L two-phase region of the Al­Zn system the GB transformation for the (Al) GBs wetting by Zn-containing melt occurs [20]. The completely wetted GBs in the Al­Zn polycrystals do not exist below TwGB0% = 440 °C. TwGB0% is the wetting temperature for a GB with maximal energy rGBmax. Above TwGB100% = 565 °C all high-angle GBs in (Al) are completely wetted by the melt (Fig. 2a) [16, 20]. TwGB100% is the wetting temperature for a GB with minimal energy rGBmin. Between TwGB0% and TwGB100% the wetting tie-lines for GBs with intermediate rGBmax [ rGB [ rGBmin are positioned in the (Al) ? L area (Fig. 1a). GB triple junctions (TJs) become completely wetted at the temperature TwTJ100% = 555 °C below TwGB100% [16]. The GB wetting phase transition in the

123


4352

J Mater Sci (2011) 46:4349­4353

Fig. 3 Temperature dependences of maximal hmax (circles), mean hmean (diamonds), and minimal hmin (hexagons) values of the GB contact angle h between (Al)/(Al) GBs and (Zn) particles (see scheme in Figs. 2c, d). The value hmin reaches zero at TwsAl 0% = 205 °C. The extrapolation of hmean to h = 0 gives TwsAl 50% = 167 °C. At this temperature 50% of (Al)/(Al) GBs are completely wetted by (Zn) and another 50% of (Al)/(Al) GBs are incompletely wetted by (Zn). The extrapolation of the hmax to the low temperatures until h = 0 permits to estimate the temperature of the GB wetting transition as TwsAl100% = 125 °C. Below TwsAl100% all (Al)/(Al) GBs should be completely wetted by (Zn)

(Al) ? L two-phase region of the Al­Zn system is of first order (discontinuous). The GB wetting phase transition in the (Zn) ? L two-phase region is of second order (continuous) [14]. The tie-lines for the tilt ½11" GBs with mis20 orientation angles / = 46° and / = 84° are shown in the (Zn) ? L two-phase region at Tw84° = 418 °C and Tw46° = 415 °C (Fig. 1a, see also insert in the upper right corner). Following the Cahn's generic phase diagram, the more sophisticated theories of GB phases, segregation and wetting layers were developed [21­23]. Thin films of interfacial phases were observed in GBs in metals (studies of J. Luo and co-workers [24­26]), in oxides ([27], see also concept of complexions by Harmer and co-workers [28­32]), in interphase boundaries (Kaplan and co-workers [33­35]). According to those developments the GB wetting tie-lines continue as prewetting (or GB solidus or solvus) lines in the one-phase (Al) area. Just one prewetting line for TwGB0% is shown for simplicity in Fig. 1a. The experimental evidence for the existence of a GB liquid-like phase between GB prewetting line and bulk solidus line was obtained by transmission electron microcopy [13] and differential scanning calorimetry (DSC) [36] (open triangles in Fig. 1a). In the area between GB solidus and bulk solidus, GB contains the thin layer of a GB phase. The energy gain (rGB­2rSL) above TwGB permits to stabilize

such thin layer of a GB phase between the abutting crystals, which is metastable in the bulk and become stable in the GB. The formation of metastable phase layer of thickness l leads to the energy loss lDg (Dg is the additional free enthalpy needed to create the layer of a metastable liquid phase). Finite thickness l of the GB phase is defined be the equality of the energy gain (rGB­2rSL) and energy loss lDg. In this simplest model, the prewetting GB layer of finite thickness l suddenly appears by crossing the prewetting (GB soludus) line cbt(T). Thickness l logarithmically diverges close to the bulk solidus. It is due to the fact that the thickness of a wetting phase is thermodynamically infinite in the two-phase area. Physically, in the two-phase area, its thickness is defined only by the amount of the wetting phase. Several monolayer (ML) thick liquid-like GB layers possessing high diffusivity were observed in the Cu­Bi [37­41], Al­Zn [13, 36], Fe­Si­Zn [11­13] and W­Ni alloys [25, 26]. GB liquid-like phase drastically influences also the GB segregation [38], GB mobility [42], GB energy and electrical resistivity [43, 44]. The direct HREM evidence for thin GB films and triple junction ``pockets'' of the liquid-like phase has been recently obtained in metallic W­Ni [25, 26] and Al­Zn [13] alloys. The authors can imagine that additional lines, similar to the GB solidus, can also continue the tie-lines TwsAl0% = 205 ± 5 °C and TwsAl00% = 125 °C into the one-phase (Al) area. Such continuations would form a kind of GB solvus lines for GBs with various energies. The search of such GB solvus lines will be the subject of the future study. In summary, the authors observed the GB ``wetting'' by a second solid phase. The reversible GB transition from incomplete to complete wetting by a liquid phase always proceeds with increasing temperature since the temperature dependence 2rSL(T) is always more steep than the dependence rGB(T), due to the fact that a liquid phase possesses higher entropy in comparison with a solid one. In case of solid state wetting, there are no obvious thermodynamic reasons for the drastic difference between temperature derivatives for GB energy rGB and for the energy of two solid/solid interphase boundaries 2rSS. As a result, rGB may decrease with increasing temperature quicker than the energy of two solid/solid interphase boundaries 2rSS. In this case, the transition from incomplete GB wetting to the complete one would proceed with decreasing temperature, like in the case in the Al­Zn system. The new GB tie-lines at TwsAl0% = 205 ± 5 °C and TwsAl100% = 125 ± 10 °C appeared in the Al­Zn phase diagram (Fig. 1a). Between TwsAl0% and TwsAl100% some (Al)/Al) GBs are completely wetted by the layers of solid (Al) phase. Below TwsAl100% all (Al)/(Al) GBs has to be completely wetted by (Zn) solid phase.

123


J Mater Sci (2011) 46:4349­4353 Acknowledgements Authors thank the Russian Foundation for Basic Research (contract 09-03-92481) and Israel Ministry of Science (project 3-5790) and the Program of bilateral cooperation between Russian and Polish Academies of Sciences for the financial support. Authors cordially thank Prof. E. Rabkin, Prof. R. Valiev, Prof. T. Langdon and Dr. A. Mazilkin for stimulating discussions, Mr. A. Nekrasov for the help with SEM and EPMA measurements.

4353 ´ 20. Straumal B, Lopez G, Gust W, Mittemeijer E (2004) In: Zehetbauer MJ, Valiev RZ (eds) Nanomaterials by severe plastic deformation. Fundamentals­processing­applications. Wiley, VCH Weinheim, p 642 21. Wynblatt P, Saul A, Chatain D (1998) Acta Mater 46:2337 22. Wynblatt P, Chatain D (2008) Mater Sci Eng A 495:119 23. Bishop CM, Tang M, Cannon RM, Carter WC (2006) Mater Sci Eng A 422:102 24. Luo J (2008) Curr Opin Sol State Mater Sci 12:81 25. Gupta VK, Yoon DH, Meyer HM et al (2007) Acta Mater 55:3131 26. Luo J, Gupta VK, Yoon DH et al (2005) Appl Phys Lett 87:231902 27. Luo J, Chiang Y-M (2008) Ann Rev Mater Res 38:227 28. Luo J, Dillon SJ, Harmer MP (2009) Microsc Today 17:22 29. Cho J, Wang CM, Chan HM, Rickman JM, Harmer MP (2002) J Mater Sci 37:59. doi:10.1023/A:1013185506017 30. Dillon SJ, Tang M, Carter WC, Harmer MP (2007) Acta Mater 55:6208 31. Dillon SJ, Harmer MP (2007) Acta Mater 55:5247 32. Dillon SJ, Harmer MP (2008) J Eur Cer Soc 28:1485 33. Baram M, Kaplan WD (2006) J Mater Sci 41:7775. doi: 10.1007/s10853-006-0897-7 34. Sadan H, Kaplan WD (2006) J Mater Sci 41:5099. doi: 10.1007/s10853-006-0437-5 35. Levi G, Kaplan WD (2006) J Mater Sci 41:817. doi: 10.1007/s10853-006-6565-0 36. Straumal BB, Mazilkin AA, Kogtenkova OA et al (2007) Phil Mag Lett 87:423 37. Straumal B, Valiev R, Kogtenkova O et al (2008) Acta Mater 56:6123 38. Divinski SV, Lohmann M, Chr Herzig et al (2005) Phys Rev B 71:104104 39. Chang L-S, Rabkin E, Straumal BB et al (1999) Acta Mater 47:4041 40. Straumal BB, Polyakov SA, Chang L-S et al (2007) Int J Mater Res 98:451 41. Straumal B, Prokofjev SI, Chang L-S et al (2001) Def Diff Forum 194:1343 42. Molodov DA, Czubayko U, Gottstein G et al (1995) Phil Mag Lett 72:361 ¨ 43. Scholhammer J, Baretzky B, Gust W et al (2001) Interf Sci 9:43 44. Straumal B, Sluchanko NE, Gust W (2001) Def Diff Forum 188:185

References
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. Cahn JW (1977) J Chem Phys 66:3667 Ebner C, Saam WF (1977) Phys Rev Lett 38:1486 Eustathopoulos N (1983) Int Met Rev 28:189 Straumal B, Muschik T, Gust W, Predel B (1992) Acta Metal Mater 40:939 Straumal B, Molodov D, Gust W (1995) Interface Sci 3:127 Straumal BB (2003) Grain boundary phase transitions. Nauka, Moscow (in Russian) ´ Lopez GA, Mittemeijer EJ, Straumal BB (2004) Acta Mater 52:4537 Straumal BB, Baretzky B, Kogtenkova OA et al (2010) J Mater Sci 45:2057. doi:10.1007/s10853-009-4014-6 Straumal BB, Kogtenkova OA, Straumal AB et al (2010) J Mater Sci 45:4271. doi:10.1007/s10853-010-4377-8 Rabkin EI, Semenov VN, Shvindlerman LS et al (1991) Acta Metall Mater 39:627 Noskovich OI, Rabkin EI, Semenov VN et al (1991) Acta Metall Mater 39:3091 Straumal BB, Noskovich OI, Semenov VN et al (1992) Acta Metall Mater 40:795 Straumal B, Rabkin E, Lojkowski W et al (1997) Acta Mater 45:1931 Straumal BB, Gornakova AS, Kogtenkova OA et al (2008) Phys Rev B 78:054202 Straumal B, Valiev R, Kogtenkova O et al (2008) Acta Mater 56:6123 Straumal BB, Kogtenkova O, Zieba P (2008) Acta Mater 56:925 Straumal B, Kogtenkova O, Protasova S et al (2008) Mater Sci Eng A 495:126 Valiev RZ, Murashkin MY, Kilmametov A et al (2010) J Mater Sci 45:4718. doi:10.1007/s10853-010-4588-z Massalski TB (ed) (1990) Binary alloy phase diagrams, 2nd edn. ASM International, Materials Park, p 238

123