Quasicrystals - American Chemical Society

Quasicrystals: A Short Review from a Surface Science. Perspective ... Quasicrystals are materials of both intellectual and practical importance. Altho...
8 downloads 0 Views 107KB Size
1392

Langmuir 1998, 14, 1392-1397

Quasicrystals: A Short Review from a Surface Science Perspective C. J. Jenks* and P. A. Thiel Department of Chemistry and Ames Laboratory, Iowa State University, Ames, Iowa 50011 Received July 3, 1997. In Final Form: November 4, 1997 Quasicrystals are materials of both intellectual and practical importance. Although some level of understanding can now be applied to their bulk electronic and atomic structure, the same cannot be said of their surface properties. In this article, we point out some of the key issues in the surface science of quasicrystals at present.

To a surface scientist, quasicrystals are quite unusual because of several factors, the first of which is their structure. Quasicrystals are not three-dimensionally periodic, but they can be very well ordered.1-4 The mathematical rules that are thought to dictate the atomic positions are simply rules other than the three-dimensional periodicity we know so well. Diffraction is a predictable consequence of these rules, and the high degree of translational order attainable in the quasicrystals has been demonstrated experimentally by extremely narrow diffraction linewidths, as well as the Boorman effect.5 However, the exact atomic structure is an issue of active discussion within the quasicrystalline community, even for those concerned with bulk structure. The discussion extends, naturally, to those concerned with surface structure. Even the semantics of quasicrystals are unsettled. In principle, they could be referred to simply as “crystals” because the International Union of Crystallography has recently broadened the definition of that word to “any solid having an essentially discrete diffraction diagram.” 6 Furthermore, quasicrystals are periodic if one goes to six (or higher) dimensional space, so it can be argued that they even satisfy older definitions of “crystal” that invoke periodicity. However, for now, the term “quasicrystal” seems to serve a useful function in distinguishing these materials from their more well-understood counterparts. The practical distinction is that quasicrystals exhibit symmetry elements, most commonly a 5-fold rotational axis that cannot be accommodated in the 230 space groups. To maintain this distinction, we replace “single-crystal” with “single-grain” in describing the large and perfect samples that are desirable for most surface studies. Then “single grain” is defined, more specifically, as a sample in which the rotational axes are common (i.e., a structure that shares a coherent orientation). A second unusual feature of quasicrystals has a more practical nature; that is, large (>2 mm), high-quality, single-grain samples of quasicrystals are difficult to obtain and to maintain for several reasons. Quasicrystals are (1) Stephens, P. W.; Goldman, A. I. Sci. Am. April 1991, 24-31. (2) Goldman, A. I.; Widom, M. Ann. Rev. Phys. Chem. 1991, 42, 685729. (3) Goldman, A. I.; Kelton, K. F. Rev. Mod. Phys. 1993, 65, 213-230. (4) Janot, C. Quasicrystals: A Primer; Humphreys, C. J., Hirsch, P. B., Mott, N. F., Brook, R. J., Eds., Monographs on the Physics and Chemistry of Materials; Clarendon: Oxford, 1992; Vol. 48. (5) Kycia, S. W.; Goldman, A. I.; Lograsso, T. A.; Delaney, D. W.; Black, D.; Sutton, M.; Dufresne, E.; Bru¨ning, R.; Rodricks, B. Phys. Rev. B 1993, 48, 3544-3574. (6) IUCR Acta Crystallogr. 1992, A48, 922-946.

alloys and usually are aluminum (Al) rich. The thermodynamically stable quasicrystals (which are far outnumbered by the purely metastable ones), are known not to melt congruently, and the compositional differences between the solid and liquid phases can be large. These factors can make it difficult or even impossible to grow these materials in the large, single-grain form that is most desirable for surface studies. (Phase purity is also a significant issue.) When the obstacle of sample growth can be overcome, maintenance becomes an issue. The multielement nature of quasicrystals introduces potential complications, such as surface segregation. Furthermore, standard surface cleaning procedures (sputtering, annealing, oxidation) can move the surface and near-surface composition out of the field of quasicrystalline stability (which spans only a few atomic percent), leading potentially to surface (and even bulk) transformations.7-10 In short, the characterization and treatment of these samples with the techniques typical of surface science are nontrivial tasks. Our group has devoted considerable time and effort to learning these tasks,11-13 but we continue to discover new phenomena that underscore the complexity of the topic. As an example of the controversy that can arise regarding surface preparation, Figure 1 shows scanning tunneling microscopy (STM) images of clean, five-fold surfaces of i-Al-Pd-Mn obtained by two different groups, both working in ultrahigh vacuum (UHV).14-17 In Figure (7) Kang, S.-S.; Dubois, J.-M. J. Mater. Res. 1995, 10, 1071-1074. (8) Sordelet, D. J.; Gunderman, A.; Besser, M. F.; Akinc, A. B. In Proceedings of the Conference New Horizons in Quasicrystals: Research and Applications; Goldman, A. I., Sordelet, D. J., Thiel, P. A., Dubois, J. M., Eds.; World Scientific: Singapore, 1997; pp 296-303. (9) Wehner, B. I.; Ko¨ster, U. In Proceedings of the Conference New Horizons in Quasicrystals: Research and Applications; Goldman, A. I., Sordelet, D. J., Thiel, P. A., Dubois, J. M., Eds.; World Scientific: Singapore, 1997; pp 152-156. (10) Shen, Z.; Jenks, C. J.; Anderegg, J.; Delaney, D. W.; Lograsso, T. A.; Thiel, P. A.; Goldman, A. I. Phys. Rev. Lett. 1997, 78, 1050-1053. (11) Jenks, C. J.; Delaney, D.; Bloomer, T.; Chang, S.-L.; Lograsso, T.; Thiel, P. A. Appl. Surf. Sci. 1996, 103, 485-493. (12) Shen, Z.; Pinhero, P. J.; Lograsso, T. A.; Delaney, D. W.; Jenks, C. J.; Thiel, P. A. Surf. Sci. Lett. 1997, 385, 923-929. (13) Jenks, C. J.; Pinhero, P. J.; Shen, Z.; Lograsso, T. A.; Delaney, D. W.; Bloomer, T. E.; Chang, S.-L.; Zhang, C.-M.; Anderegg, J. W.; Islam, A. H. M. Z.; Goldman, A. I.; Thiel, P. A. In Proceedings of the Sixth International Conference on Quasicrystals (ICQ6); Takeuchi, S., Fujiwara, T., Eds.; World Scientific: Singapore, in press. (14) Schaub, T. M.; Bu¨rgler, D. E.; Gu¨ntherodt, H.-J.; Suck, J.-B. Z. Phys. B 1994, 96, 93-96. (15) Schaub, T. M.; Bu¨rgler, D. E.; Gu¨ntherodt, H.-J.; Suck, J. B. Phys. Rev. Lett. 1994, 73, 1255-1258. (16) Schaub, T. M.; Bu¨rgler, D. E.; Gu¨ntherodt, H.-J.; Suck, J. B.; Audier, M. Appl. Phys. A 1995, 61, 491-501.

S0743-7463(97)00727-0 CCC: $15.00 © 1998 American Chemical Society Published on Web 01/22/1998

Quasicrystals from a Surface Science Perspective

Langmuir, Vol. 14, No. 6, 1998 1393

Figure 1. (a) Scanning tunneling micrograph of a 5-fold i-Al-Pd-Mn surface. The surface was prepared by sputtering and then heating close to the melting point, in UHV.16 The figure at the left shows the terrace-step-kink morphology. The panel at the right shows the fine structure present on the terraces, with 5-fold features highlighted by circles. A 4 Å step cuts across the upper middle of the right-hand image. The gray scale is adjusted so the two terraces are of about equal darknes, to allow the fine structure on both to be evident. Reprinted with permission from Springer-Verlag, 1995. (b) Scanning tunneling micrographs (at two different magnifications) of an i-Al-Pd-Mn surface, prepared by cleavage along a five-fold plane at room temperature.17 Reprinted with permission from the American Institute of Physics, 1996.

1A, the surface was prepared by sputter-annealing, and its structure can be interpreted in terms of the standard terrace-step-kink model of crystalline surfaces. In the other case, the rather rough topography (with a maximum corrugation of 10 Å) was obtained by cleavage in UHV. Both images have been interpreted in terms of fundamental concepts of bulk quasicrystalline structure. The differences between these images have engendered a (17) Ebert, P.; Feuerbacher, M.; Tamura, N.; Wollgarten, M.; Urban, K. Phys. Rev. Lett. 1996, 77, 3827-3830.

critical discussion, which is not yet resolved, over what is the most appropriate way to prepare a quasicrystalline surface. The final element of strangeness about quasicrystals is that many of their mechanical-physical properties are quite unusual by the standards of common metals. Some of these properties are surface related and some are of interest to the commercial sector. This intriguing element provides the main motivation for surface scientists to try to study these materials at present, in spite of the difficulties just outlined. There is great potential for

1394 Langmuir, Vol. 14, No. 6, 1998

Jenks and Thiel

Table 1. Selected Values of Physical and Mechanical Properties of Icosahedral Alloys, Compared with Relevant Benchmark Materialsa property hardness (Hv)

coefficient of friction23 (unlubricated, with a diamond pin) fracture toughness (MPa m1/2) Young’s modulus (106 psi)

thermal conductivity (W m-1 K-1)

surface energy (mJ/m2)

a

value

material

6000-10 000 750-1200 800-1000 700-800 70-200 40-105 25-45 0.42 0.37 0.32 0.05-0.2 4 1.5 1 0.3 31 29 19 10 9 390 170 50 2 2 2480 1830 50 24-25 17-18

diamond18 silica18,19 i-Al-Cu-Fe20,21 i-Al-Pd-Mn22 low-carbon steel18,19 copper18,19 aluminum18,19 copper24 aluminum alloy24 low-carbon steel24 i-Al-Cu-Fe24,25 alumina26 silica27 i-Al-Cu-Fe20,21 i-Al-Pd-Mn22 stainless steel28 i-Al-Pd-Mn22 copper28 aluminum28 i-Al-Cu-Fe20 copper28 aluminum20 low-carbon steel28 yttria-doped zirconia20 i-Al-Cu-Fe20 iron (clean)29 copper (clean)29 alumina25 i-Al-Pd-Mn (air-oxidized)25 PTFE (Teflon)25

All values pertain to room temperature conditions.

surface science to make a significant contribution in this area. The remainder of this article summarizes some of these properties, and the efforts of our group and others to understand them. Some of the physical-mechanical properties of quasicrystals, relative to selected benchmark materials, are shown in Table 1. Two ternary quasicrystalline materials were chosen for Table 1; they are icosahedral (i-)-Al-PdMn and i-Al-Cu-Fe. The latter alloy is of some commercial interest because it is an equilibrium phase with inexpensive components. The values of Vickers’ hardness, Hv,18-22 in Table 1 demonstrate that this quasicrystal is harder than any of its individual components and comparable in hardness to silica. Because sand is a common environmental abrasive, it is an important standard of comparison. The quasicrystal also exhibits a low coefficient of friction and is reputed to be relatively unreactive (“non-stick”).23-25 These properties, combined with the hardness, lead naturally to speculation that quasicrystals (18) Hutchings, I. M. Tribology: Friction and wear of engineering materials; CRC: Boca Raton, FL, 1992. (19) Chemical Engineers' Handbook; Perry, R. H., Chilton, C. H., Ed.; McGraw-Hill Book: New York, 1973. (20) Dubois, J. M.; Weinland, P. CNRS, Nancy, France, Coating Materials for Metal Alloys and Metals and Method, Eur. Patent EP 0356287 A1 and U.S. Patent 5,204,191, April 20, 1993. (21) Ko¨ster, U.; Liu, W.; Liebertz, H.; Michel, M. J. Non-Cryst. Solids 1993, 153-154, 446-452. (22) Yokoyama, Y.; Inoue, A.; Masumoto, T. Mater. Trans., JIM 1993, 34, 135-145. (23) Note, the coefficient of friction is not a property of a single material, rather a property of the entire friction measurement and all its experimental parameters (hardness of the pin, roughness of both sliding surfaces, number of passes, etc.). See, for instance, ref 18. (24) Kang, S. S.; Dubois, J. M.; von Stebut, J. J. Mater. Res. 1993, 8, 2471. (25) Dubois, J. M. In Proceedings of the Conference New Horizons in Quasicrystals: Research and Applications; Goldman, A. I., Sordelet, D. J., Thiel, P. A., Dubois, J. M., Eds.; World Scientific: Singapore, 1997; pp 208-215.

may be useful as low-friction or release coatings in industrial settings where abrasion resistance is also important. However, to date, such an application has not reached the marketplace. The results in Table 1 also illustrate other properties, including brittleness, elasticity, conductivity, and surface energy. The quasicrystal is quite brittle, illustrated by the low value of fracture toughness, which is a drawback for most applications.20-22,26,27 Its elasticity, judged by Young’s modulus, is comparable to that of typical metals.20,22,28 Although metallic, quasicrystals are poor thermal and electrical conductors, which may be put to good use in thermal barrier applications. Quasicrystals also exhibit a rather low surface energy, which is much lower than that of a typical clean metal and significantly lower than that of an oxidized metal (alumina).25,29 Finally, although not shown in Table 1, qualitative observations suggest that the quasicrystals are resistant to oxidation. These and other properties related to potential applications have been reviewed thoroughly by Dubois and co-workers.20,25,30-33 Note that other potential applications also exist (e.g., as sensors, as components of maraging steels, or as hydrogen storage materials), for which alloys other than i-Al-Cu-Fe may be best suited. The issue that is perhaps most germane to our community is the low surface energy of the quasicrystalline materials, as illustrated at the bottom of Table 1. It has been speculated that this property is related to the low coefficients of friction and oxidation resistance of quasicrystals.34,35 Rivier34 has proposed that this low surface energy is (at least in part) a consequence of the pecular electronic structure of the bulk, which is characterized by a “pseudogap” (a reduction) in the density of states at the Fermi level, EF. The presence of a pseudogap has been demonstrated, for example, by the work of Belin-Ferre´,36 which is illustrated in Figure 2. In chemists’ terms, this pseudogap is due partly to Hume-Rothery stabilization, which is analogous to a Jahn-Teller distortion at EF. In spite of the pseudogap, the quasicrystals are metals, not insulators (although there is some controversy over this statement).37 The pseudogap is now known to be a universal characteristic of quasicrystals, although it is by no means unique to the quasicrystals,38,39 and it is related (26) Kovar, D.; Ready, M. J. Am. Ceram. Soc. 1994, 77, 1928-1938. (27) Lyons, J.; Starr, T. J. Am. Ceram. Soc. 1994, 77, 1673-1675. (28) Metals Handbook: Desk Edition; Boyer, H. E., Gall, T. L., Eds.; Am. Soc. for Metals: Metals Park, OH, 1984. (29) de Boer, F. R.; Boom, R.; Mattens, W. C. M.; Meidema, A. R.; Niessen, A. K. Cohesion in Metals; North-Holland: Amsterdam, 1988. (30) Dubois, J. M.; Proner, A.; Bucaille, B.; Cathonnet, P.; Dong, C.; Richard, V.; Pianelli, A.; Massiani, Y.; Ait-Yaazza, S.; Belin-Ferre´, E. Ann. Chim. Materiaux 1994, 19, 3-25. (31) Dubois, J.-M.; Kang, S. S.; Perrot, A. Mater. Sci. Eng. 1994, A179/A180, 122-126. (32) Dubois, J. M. Phys. Scripta 1993, T49A, 17. (33) Dubois, J.-M.; Kang, S. S.; Massiani, Y. J. Non-Cryst. Solids 1993, 153-154, 443-445. (34) Rivier, N. In Proceedings of the Conference, New Horizons in Quasicrystals: Research and Applications; Goldman, A. I., Sordelet, D. J., Thiel, P. A., Dubois, J. M., Eds.; World Scientific: Singapore, 1997; pp 188-199. (35) Dubois, J. M.; Plaindoux, P.; Belin-Ferre´, E.; Tamura, N.; Sordelet, D. J. In Proceedings of the 6th International Conference on Quasicrystals (ICQ6); Takeuchi, S., Fujiwara, T., Eds.; World Scientific: Singapore, in press. (36) Traverse, A.; Dumoulin, L.; Belin, E.; Senemaud, C. In Quasicrystalline Materials; Janot, C., Dubois, J. M., Eds.; World Scientific: Singapore, 1988; pp 399-408. (37) Stadnik, Z. M.; Purdie, D.; Garnier, M.; Baer, Y.; Tsai, A.-P.; Inoue, A.; Edagawa, K.; Takeuchi, S. Phys. Rev. Lett. 1996, 77, 1777. (38) Belin, E.; Mayou, D. Phys. Scripta 1993, T49, 356-359. (39) Belin-Ferre´, E.; Fourne´e, V.; Dubois, J.-M. In New Horizons in Quasicrystals: Research and Applications; Goldman, A. I., Sordelet, D. J., Thiel, P. A., Dubois, J. M., Eds. World Scientific: Singapore, 1997; pp 9-16.

Quasicrystals from a Surface Science Perspective

Figure 2. Illustration of the reduction in the density of occupied Al3p states near the Fermi edge for a quasicrystal, compared with the same states in pure Al.66 The curve for the occupied band of the quasicrystal is farthest left in the figure. The data for the quasicrystal have the lowest intensity of the two occupied bands at zero binding energy (EF). The unoccupied p states for pure fcc Al are illustrated schematically by the rightmost curve in the panel. Reproduced with permission from E. Belin-Ferre´.

to their poor electrical conductivity (cf. Table 1). HumeRothery stabilization cannot explain the pseudogap entirely, based on detailed characterizations of electronic transport in the quasicrystals,38,40-42 but it is still thought to play a major role. At present, it is not known whether and under what conditions the pseudogap may persist in the surface and near-surface regions. Although it has been argued on symmetry grounds that the pseudogap must be present at the surface of any quasicrystal (assuming that the surface structure is bulk terminated),34 theoretical43 and experimental44,45 evidence seems inconclusive. The exact implications of the pseudogap for surface propertiessand whether it can explain all the anomalous surface phenomena, relating them in a coherent fashionsare issues of current exploration. In general, however, one might expect that those properties of a metal that depend on its surface polarizability (i.e., its surface dielectric constants) will be suppressed in the quasicrystals because of the pseudogap. In fact, Dubois and co-workers35 have demonstrated, via careful comparative measurements of contact angles, that the quasicrystalline surfaces behave much more like covalently bonded solids than like metals (i.e., the polarizability, which is characteristic of metal surfaces, is largely suppressed in the quasicrystals, as probed by these particular measurements). What implication does this behavior have for chemical reactivity at the quasicrystalline surface? For instance, can the pseudogap be related to the low reactivity toward oxygen, and the reputed low reactivity toward other materials? Addressing the oxidation resistance first, the data currently available suggest that, to a first approximation, (40) Mayou, D.; Berger, C.; Cyrot-Lackmann, F.; Klein, T.; Lanco, P.; Phys. Rev. Lett. 1993, 70, 3915-3918. (41) Janot, C. Phys. Rev. B 1996, 53, 181-191. (42) Janot, C.; de Boissieu, M. Phys. Rev. Lett. 1994, 72, 1674-1677. (43) Fasolino, A.; Janssen, T. In Proceedings of the 6th International Conference on Quasicrystals (ICQ6); Takeuchi, S., Fujiwara, T., Eds.; World Scientific: Singapore, in press. (44) Klein, T.; Symko, O. G.; Davydov, D. N.; Jansen, A. G. M. Phys. Rev. Lett. 1995, 74, 3656. (45) Davydov, D. N.; Mayou, D.; Berger, C.; Gignoux, C.; Neumann, A.; Jansen, A. G. M.; Wyder, P. Phys. Rev. Lett. 1996, 77, 3173-3176.

Langmuir, Vol. 14, No. 6, 1998 1395

the answer is “no.” We believe that the oxidation resistance has less to do with the surface energy than with the chemical nature (i.e., the Al richness) of the samples with which we have worked. For the first studies of surface oxidation,46-48 i-Al-Pd-Mn was chosen because, starting in the early 1990s, it could be grown in large, single-grain form.49-51 Later studies have indicated, however, that oxidation phenomena on i-Al-Pd-Mn and i-Al-Cu-Fe are quite similar.52,53 Hence, the i-Al-Pd-Mn, although not commercially attractive because of its precious metal content, may serve as a useful model for understanding oxidation of the other Al-rich quasicrystalline alloys. The basic finding has been that oxidation of the quasicrystalline surface, in pure oxygen or in dry air at room temperature, produces a thin, passivating layer of pure aluminum oxide, and that this accounts for the oxidation resistance of the material.46-48,54 This effect is illustrated by the X-ray photoelectron spectroscopy (XPS) data in Figure 3. Water or humidity deepens the oxide layer, and attacks metals other than Al.52,53,55 Oxidation at elevated temperatures appears to have a similar effect,46,48,56 provided the oxygen pressure is high enough that the oxygen does not simply migrate into the bulk before it can nucleate at the surface.57 These results were obtained primarily by XPS and Auger electron spectroscopy, although some data from low-energy electron diffraction are also available. Surface passivation by a pure, protective layer of aluminum oxide apparently occurs for other Al-rich alloys as well, such as the nickel aluminides.58-62 Although the oxidation resistance of the quasicrystal is understood to be zero order, there may be a first-order (46) Chang, S.-L.; Chin, W. B.; Zhang, C.-M.; Jenks, C. J.; Thiel, P. A. Surf. Sci. 1995, 337, 135-146. (47) Chang, S.-L.; Zhang, C.-M.; Jenks, C. J.; Anderegg, J. W.; Thiel, P. A. In Proceedings of the 5th International Conference on Quasicrystals (ICQ5); Janot, C., Mosseri, R., Eds. World Scientific: Singapore, 1995; pp 786-789. (48) Chang, S.-L.; Anderegg, J. W.; Thiel, P. A. J. Noncryst. Solids 1996, 195, 95-101. (49) Yokoyama, Y.; Tsai, A.-P.; Inoue, A.; Masumoto, T. Mater. Trans., JIM 1991, 32, 1089-1097. (50) Yokoyama, Y.; Miura, T.; Tsai, A.-P.; Inoue, A.; Masumoto, T. Mater. Trans. 1992, 33, 97-101. (51) de Boissieu, M.; Durand-Charre´, M.; Bastie, P.; Carabelli, A.; Boudard, M.; Bessie`re, M.; Lefe`bvre, S.; Janot, C.; Audier, M. Phil. Mag. Lett. 1992, 65, 147. (52) Jenks, C. J.; Pinhero, P. J.; Chang, S.-L.; Anderegg, J. W.; Besser, M. F.; Sordelet, D. J.; Thiel, P. A. In Proceedings of the Conference New Horizons in Quasicrystals: Research and Applications; Goldman, A. I., Sordelet, D. J., Thiel, P. A., Dubois, J. M., Eds. World Scientific: Singapore, 1997; pp 157-164. (53) Jenks, C. J.; Pinhero, P. J.; Bloomer, T. E.; Anderegg, J. W.; Thiel, P. A. In Proceedings of the Sixth International Conference on Quasicrystals (ICQ6); Takeuchi, S., Fujiwara, T., Eds. World Scientific: Singapore, in press. (54) Suzuki, S.; Waseda, Y.; Tamura, N.; Urban, K. Scripta Mater. 1996, 35, 891-895. (55) Pinhero, P. J.; Chang, S.-L.; Anderegg, J. W.; Thiel, P. A. Phil. Mag. B 1997, 75, 271-281. (56) Rouxel, D.; Gavatz, M.; Pigeat, P.; Weber, B.; Plaindoux, P. in Proceedings of the Conference New Horizons in Quasicrystals: Research and Applications; Goldman, A. I., Sordelet, D., Thiel, P. A., Dubois, J. M., Eds.; World Scientific: Singapore, 1997; pp 173-180. (57) Gavatz, M.; Rouxel, D.; Claudel, D.; Pigeat, P.; Weber, B.; Dubois, J. M. In Proceedings of the Sixth International Conference on Quasicrystals (ICQ6); Takeuchi, S., Fujiwara, T., Eds.; World Scientific: Singapore, in press. (58) Bardi, U.; Atrei, A.; Rovida, G.,Surf. Sci. 1992, 268, 39. (59) Zehner, D. M.; Gruzalski, G. R. Mater. Res. Soc. Symp. Proc. 1987, 83, 199. (60) Jaeger, R. M.; Kuhlenbeck, H.; Freund, H.-J.; Wuttig, M.; Hoffmann, W.; Franchy, R.; Ibach, H. Surf. Sci. 1991, 259, 235. (61) Gassmann, P.; Franchy, R.; Ibach, H. Surf. Sci. 1994, 319, 95109. (62) Libuda, J.; Winkelmann, F.; Baumer, M.; Freund, H.-J.; Bertrams, T.; Neddermeyer, H.; Muller, K. Surf. Sci. 1994, 318, 61.

1396 Langmuir, Vol. 14, No. 6, 1998

Jenks and Thiel

Figure 3. Bottom four rows: X-ray photoelectron spectra of the five-fold surface of i-Al-Pd-Mn, clean and oxidized under various conditions. Top four rows: spectra of the constituent elements, clean and oxidized under the same conditions as for the quasicrystal. 48 The peaks that develop at higher binding energy for Al and Mn in the top four rows (marked with arrows) signify oxidation of these elements. Comparison with the bottom four rows shows that Al in the quasicrystal oxidizes much as it would in the pure metal under these conditions, but the Mn is protected. Pd does not oxidize either in the pure metal, or in the quasicrystal under the same conditions. Reprinted with kind permission from Elsevier Science-NL, Sara Burgerhartstraat 25, 1055 KV Amsterdam, The Netherlands.

effect that is, after all, unique to the quasicrystals. Comparative studies, both for Al-Pd-Mn and Al-Cu-Fe, suggest that the quasicrystals oxidize less readily than their crystalline analogs.8,52,53 The quasicrystals appear to form layers of aluminum oxide that are thinner and purer than those formed by their crystalline counterparts.52,53 This difference requires further investigation, but it is interesting to speculate that it may have some relationship to the classic Mott-Cabrera mechanism of metal oxidation, in which the formation of an electric field across the metal-oxide interface controls the kinetics of oxidation.63 Formation of the field could be affected by the pseudogap in the quasicrystal. Another topic requiring further investigation is the spatial structure of the oxide layer on a microscopic scale. Techniques such as XPS provide no information in this (63) Cabrera, N.; Mott, N. F. Rep. Prog. Phys. 1948-1949, 12, 163.

regard. Scanning tunneling micrographs of air-oxidized i-Al-Cu-Fe indicate that the oxide formed at room temperature consists of small nodules, on the order of 10 Å wide and 5 Å high, arranged in a symmetry that may be related to the icosahedral nature of the substrate.64 Again, however, this is a topic that begs for more work. Addressing the reactivity toward other materials next, it seems that there are basically two paradigms. One is to think in terms of surface energies and contact angles, assuming that “sticking is a wetting problem; what wets sticks.”34 In this rather global approach, a low surface reactivity in general can be understood in terms of low surface energy. As already noted, the low surface energy can, in turn, be explained in terms of the pseudogap. However, surface roughness, such as that illustrated in Figure 1B or that described previously for an oxidized (64) Dubois, J. M. private communication, 1997.

Quasicrystals from a Surface Science Perspective

surface, may contribute to high contact angles as well.34 And, in general, focussing on the surface energy may overlook the fact that different adsorbates can react differently with a surface. A second approach, more natural to a chemist, might be to examine the products of reaction of a quasicrystalline surface with individual molecules on a case-by-case basis, using the arsenal of modern surface spectroscopies. We have just begun this endeavor in our laboratory. Preliminary results suggest that the surface chemistry of small, covalently bonding molecules, such as CO or methanol, on i-Al-Pd-Mn is very similar to the chemistry that would be expected with a pure Al substrate. This is consistent with other studies, which show that the surface termination of a sputter-annealed surface is Al-rich.65 Our studies of surface chemistry thus far, however, have been carried out on the clean surface only, prepared in UHV. One must always be cognizant of the presence and possible (65) Gierer, M.; Van Hove, M. A.; Goldman, A. I.; Shen, Z.; Chang, S.-L.; Jenks, C. J.; Zhang, C.-M.; Thiel, P. A. Phys. Rev. Lett. 1997, 78, 467-470. (66) Belin-Ferre´, E. private communication.

Langmuir, Vol. 14, No. 6, 1998 1397

influence of the oxide layer in “real” applications (and in contact angle measurements).35 In summary, this article has posed many questions, but only a few answers, and in so doing it has aptly reflected the current status of our understanding of quasicrystalline surfaces. This topic is important, exciting, and rapidly-moving, and more answers are sure to emerge soon. Acknowledgment. Many of the ideas in this article were engendered during discussions with E. Belin-Ferre´, J. Chevrier, J. M. Dubois, A. Goldman, N. Rivier, D. Rouxel, T. Schaub, D. Sordelet, and K. Urban. We thank them sincerely. We are also grateful to E. Belin-Ferre´, T. Schaub, and K. Urban for permission and assistance in reproducing Figures 1 and 2. This work was supported by the Director, Office of Energy Research, Office of Basic Energy Sciences, Materials Sciences Division, of the U. S. Department of Energy under Contract no.W-405-Eng-82. LA970727+