HIGHLIGHTS cence detection techniques." 21 With that in mind, this combination triggers intriguing possibilities for chemists and biochemists: Single molecules can conceivably be separated, collected. and manipulated on a chip, and possibly even made to undergo chemical reactions. Chemistry with single molecules could soon be reality. German version: Angeu Chern. 1996, 108, 931 --933 Keywords: analytical methods miniaturization - capillary electrophoresis . ~ [ l ] A selection o f r n i e w articles and monographs: J. W Jorgenson. K. D. Lukacs, Scien<r 1983, 322. 266-272; M. J. Gordon. X . Huang, S. L. Pentoney, Jr.. R. N Zare. f h f d 1988. 242, 224-228; A. G Ewing, R. A. Wdllingford. T. M. Olefirowicz. A n d Cizrm. A 1989. 61. 292-303: H. Engelhardt, W.Beck, J Kohr. T Schmiit. Angcii.. C'keni. 1993, 105. 659-680: Angeir. Cheni. Inr. Ed. En~zI 1993. 32. 619 -649: S. F. Y. Li. C'upilhrj. Eltcrrophorrsis (J C/iromu/ogr. Lihr 53) Elsckier, Amsterdam 1992, Cupillurr Ele(.rrophore.si.s Tedinolog~ [ U i r ~ n i u r o , y r .S u Ser. 64) (Ed.: N. A. Guzman) Dekker. New York, 1993; (Ed. : J. P Landers) CRC Press, Boca Honrll~ooko/ C'upillur>. Ele~~rropplio~-e.sh Raton. FL. USA. 1993  B. L. Karger. A. S. Cohen. A. Guttman, J Chroniurogr. 1989. 492. 5x5 -613: G . J. M. Bruin, A. Paulus, Anul. Methods Insrrron. 1995, 2. 3 26.  S. Terabe. K Otsuka. K Ichikawa. A. Tsuchiya, T Ando. A n d C'heni. 1984. 56, 113-116. 141 J. H. Knox. I. H. Grant. Chrotnotogruphru 1991, 32, 317 328. N. W. Smith, M. B. Evans. ihid 1994.38.649 -657: K.Schmeer. B. Behnke. E Bayer, A n d Chem 1995.67.3656-3658: B. Behnke, E. Grom. E. Bayer. L C/iromufogr.A 1995, 716, 207-213.  K. Albert. A n g e u . Chem. 1995, 1117, 699-701: Angeir. Cliem. Inr Ed. Engl. 1995, 34, 641 - 642.  D.L. Olson. T. L. Peck, A. G . Webb. R L. Mangin, J. V Sweedler. Scicwce 1995, 270. 1967-1969.  Y. M. Liu. J. V. Sweedler, J Am. Chem Soc. 1995. 117. 8x71 -8872.  D. J. Harrison, K. Fluri, K. Seiler, Z. Fan, C. S. Effenhauser. A. Manz, Science 1993,261, 895-897, C. S . Effenhauser. A. Manz, H. M. Widmer, Anul. Chrni. 1993.65. 2637-2642  C. S. Effenhauser. A. Paulus, A. Manz, H. M. Widmer. A n d C'hern. 1994, 66, 2949-2953. [lo] S. C. Jacobson, R. Hergenroeder. A. W Moore. Jr., J. M. Ramsey, A n d . C/iem. 1994. 66, 41 27 -41 32  C.S Effenhauser. A. Manz, H. M. Widmer. A n d Clitwr 1995, 67, 22842287. (121 M. Eigen, R. Riegler, Proc Norl. Acad. Sct U S A 1994, 91. 5740-5746. S . Nie. D T.Chiu. R.N. Zare. Science 1994.1766, 1018-1020. Superlattices, Thin Films, and Materials Design-A Mechanistic Approach Robert Schollhorn* The area of thin films is rapidly developing into a key technology with applications ranging from electronic materials (semiconductors, superconductors, optoelectronics etc.) to chemical sensor systems. A variety of deposition techniques for the preparation of single layers and superlattices is now available, extending from molecular beam epitaxy to chemical vapor deposition. While physicists are focussing their attention on the detection of new quantum phenomena associated with the constrained morphology in nanorange dimensions, it appears that a new avenue is opening here for solid-state chemistry through the controlled synthesis of complex metastable structures. A particularly interesting example in this line is represented by the approach documented in recent work by D. C. Johnson and coworkers which concerns the preparation of new compositionally modulated layered transition metal dichalcogenide structures characterized by the use of a rational synthesis strategy that involves a topological model for kinetic trapping and directed interfacial growth." --'I Superlattices are defined here as multilayer thin films with long-range structural coherence perpendicular to the basal plane which are prepared by time sequential deposition of components. In the area of molecular solids the formation of ordered multilayer systems with complex arrays of subunits has become an established art (for example, Langmuir-Blodgett ["I Prof. I h . R. Schbllhorn lnstitut fur Anorganische und Analytische Chemie, Sekr. C 2 Technische Universitit Berlin Strssae des 17 J u n i 135. D-10623 Berlin (Germany) Fax: Int. code +(30) 314-21106 e-mail: schoe(owap0209.chem.t~-herlin.de An,fw C'liiwi. lrir E d D i g / . 1996. 35. No 8 $:J and polymer films).14* The deposition of multilayers of metals/ alloys, covalent semiconductor materials,['] and oxometalate superconductorsr8] are well known for nonmolecular solids; however, there are only a few reports on other materials. The usual method of preparation is to deposit these phases under conditions that lead directly to the final product. A different philosophy underlies the approach used by Johnson et al. for the controlled chemical synthesis of complex metastable chalcogenide structures by superlattice precursors. The concept used is based on a two-step process. In the first step a multilayer structure is prepared by modulated sequential deposition of the chemical elements required. This structure is subsequently transformed by low-temperature thermal conversion under defined conditions into a crystalline superlattice compound with compositional modulation. The formation of the heterostructure is not based on conventional epitaxial growth but on kinetic trapping of the desired structure by the specific solid-state reaction mechanism. Layered transition metal dichalcogenides M X 2 of the CdI, or MoS, structure type were selected for investigations on this reaction model. These chalcogenides are interesting for several reasons. Their electronic properties vary from insulator to metal and superconductor behavior; charge density waves have been studied extensively with these layered phases.['* '*I Their topotactic reactivity is exceptional;[", " 1 in terms of application they are of interest, for example as electrode materials for secondary batteries and photoelectrochemical cells. as solid lubricants, and in heterogeneous catalysis. Their physical and chemical properties can be modified by isomorphous replacement of lattice cations and anions. VCH Verlrigsgr.srllschu/rftmhH. 0.69451 Weinheini. 1996 0570-0833:Y6/35U8~085YB 15.00 +25 0 859 HIGHLIGHTS Experimental investigations were initially carried out on the hinavy Nb/Se system by X-ray diffraction and differential scanning calorimetry; these demonstrated the oriented formation of the binary chalcogenide NbSe, as a result of the nucleation and growth of the product along the planar interfaces of the modulated precursor phase. The major parameters for reaction control were the diffusion length (layer thickness in the precursor state), the annealing temperature, and the annealing time. With appropriate reaction parameters a variation of the Nb/Se ratio (i.e. metal o r chalcogen excess relative to the 1 :2 stoichiometry) had no influence on the formation of stoichiometric NbSe,, in spite of the potential competition by neighboring binary phases thermodynamically stable under these conditions. This is in perfect agreement with the assumption of a low-temperature solidstate reaction with kinetic control. If the mechanism of the interfacial nucleation with oriented crystal growth of kinetic products (Fig. 1) is correct, it should be Fig. 1. The reaction pathway for the formation of chalcogenide superlattices: lowtemperature thermal transformation of amorphous chalcogenide precursor multilayers, comprising layers of niobium (black spheres) and titanium atoms (gray spheres), each separated by intervening layers of selenium atoms (white spheres) (left). by interfacial nucleation and oriented growth into the desired crystalline product superlattice after annealing (right). possible to prepare new metastable heterostructures containing ternary compounds with modulated composition perpendicular to the direction of the thin film basal plane. X-ray and DSC studies on the Ti/Nb/Se system in the stoichiometric ratio used for the formation of MX, layers demonstrated a persistence of the precursor thin film composition modulation in the course of the low-temperature annealing process and a continuous oriented growth of the product phase. An experiment with a precursor superlattice sequence corresponding to 6(Ti/2 Se)- 6(Nb/2 Se) yielded a remarkable result: the X-ray pattern of the product phase displayed all of the calculated 62 Bragg reflections of the desired superlattice (Fig. 2). The high quality of the reflections allowed a Rietveld analysis that confirmed the correlation between the precursor heterostructure sequence and the superlattice ordering of the final product. As expected high-temperature annealing of the product phase leads to quantitative conversion into the thermodynamically stable, solid solution phase with short periodicity and identical layer units of the composition Tio,5Nb, $e,. The novel concept for the controlled low-temperature solidstate synthesis of metastable phases by thin-layer precursors modulated in the nanometer range4haracterized by the competition in time windows between short-scale diffusion and heterogeneous interface nucleation with oriented growth of a metastable product phase-has thus been successfully applied by the authors for the controlled preparation of novel transition I log I 0 10 20 40 30 201" 50 60 70 80 Fig. 2. X-ray diffractogram of a dichalcogenide superlattice with a modulation sequence corresponding to six NbSe, and six TiSe, layer units All 62 Bragg rcflections of the desired superlattice are shown. metal dichalcogenide structures. A new dimension is given here for modifying the interesting physical and topochemical properties of these materials. Transition metal dichalcogenides belong to the category of van der Waals type layered solids; their structural morphology and bonding anisotropy naturally should favor the directed interfacial growth mechanism. A large variety of related layered compounds with similar o r more complex layered structures or stoichiometry are known to which this concept could be extended (e.g. transition metal dihalides, phosphorous chalcogenides, sulfide carbides etc.). It seems most likely that this synthesis strategy can similarly be applied to the preparation of metastable framework solids characterized by strongly anisometric structures, such as layer-type perovskites. The limit of application of this concept may be reached with structurally isometric solids, in which coherent directed nucleation and growth at interfaces appears to be less probable. Further aspects of general interest related to this concept are the investigation of dynamic interface processes (in situ diffraction studies), nucleation phenomena, and the transformation mechanisms of amorphous phases in low-temperature solid-state reactions whose investigation with bulk material is frequently problematic. It is important to note in this context a further line of progress achieved recently concerning modulated materials. Present techniques for the preparation of pristine thin films of nonmolecular solids are based essentially on physical or chemical vacuum techniques in which atoms, molecules or cluster aggregates are deposited from the vapor phase onto solid substrates (i.e. gas/ solid processes). A basically different approach is the use of electrochemical techniques for thin film preparation (i.e. fluid-solid conversion). There are several advantages associated with this technique: a) the strong tendency of electrochemically deposited materials-including isometric systems-for epitaxial growth, b) the possibility of working with systems not suited for vacuum deposition, c) suppression of interdiffusion between layers (room temperature), and d) precise control of deposition by current density (kinetics) and electrode potential (variable energy scale). Whereas the electrochemical deposition of multilayer systems of metals/alloys and s e m i c ~ n d u c t o r s ~has ' ~ ~been HIGHLIGHTS well investigated in the last decade, recent work by Switzer and coworkers appears to be the first successful attempt to obtain superlattices from a complex ternary inorganic material by the electrochemical deposition technique.[141 The major subject of these investigations was the ternary TI/ PbjO system with a fluorite type structure and a large variation in stoichiometry. By pulsed-current o r pulsed-potential deposition from electrolytes with appropriate ionic composition, superlattices with a modulation layer unit thickness of 3 nm could be prepared. X-ray and scanning tunneling microscopy studies confirmed the high quality of the superlattice architecture. The modulation wave lengths were in the order of mean free path dimensions of electrons. The material is a degenerate n-type semiconductor whose high electronic conductivity, which lies in the order of that of metals, is combined with the optical properties of a semiconductor; it is of interest with respect to quantum optical. electronic, and optoelectronic functions. A further remarkable recent result from the work by Switzer et al. is the pulsed anodic electrodeposition of nanometer-scale modulated “defect chemistry superlattices” of thallium(1rr) oxide (Fig. 3).“ ’] The modulation parameters here are intrinsic anion vacancies and cation Frenkel type defects. It could be i TI ,TI?Oq Fig. 3. Scheme ofdefect modulation in a thallium(~rc)oxidesuperlattice obtained by anodic pulsed deposition. Substrate = stainless steel. A = Periodicity of layer sequence. shown that the defect chemistry of TI,O, is a function of the applied electrode potential : high overpotentials were found to induce oxygen vacancy defects (T1,03 - J , whereas low overpotentials favor excess TI ions on interstitial sites of the lattice (i.e. T1.~T1203). The examples of recent work presented here underline the increasing involvement of chemistry in the area of nonmolecular materials artificially modulated on the nanometer scale, and the potential of low-temperature solid-state processes and controlled materials design. In comparison with the impressive rapid progress in “supramolecular chemistry” it now seems that some keys to doors are becoming available, providing access to a comparable tailoring approach in the domain of nonmolecular solids. German version’ Angew. Cheni 1996, I()#, 933-935 Keywords: electrochemistry materials design thin films . superlattices . 111 M. Noh, J. Thiel, D. C. Johnson, Science 1995, 270, 1181  T. Novet, D. C. Johnson, L. Fister, Adv. in Chem. Ser. 1995. 245, 425.  L. Fister. T. Novet, C. A. Grant. D. C. Johnson in Adrunces in the Sjnrhesis und Reuctii+ti. of Solids, Yo/. 2 (Ed.: T. E. Mallouk). JAI Press. London. 1994. p. 155.  R. H. Tredgold, J. Muter. Chrm. 1995, 5 , 1095.  M. Seufert. M. Schaub, G. Wenz, G. Wegner, Angew. C‘hpm. 1995, /07, 363; Angeir. Chrni. I n f . Eli. Engi. 1995, 34. 340.  I. K. Schuller, Sofid Sur e Comrnun. 1994, 92, 141 171 A. Ourmdzd, M. Schemer. M. Heinemann. J. L. Rouviere. M R S Bull. 1992, 17. 24.  D. Neerinck. K . Temst, M. Baert, E. Osquigil, C. van Hasendonck. Y Bruynserade. A. Gilabert. 1 K. Schuller, Phys. Re>*.Lerr. 1991. 67, 2577.  J. A Wilson. A. D. Yoffe, Adv. Phjs. 1969. 18, 193. [lo] F. Levy. Intercalated Layered Materiuk. D. Reidel, Dordrecht. 1979. Ill] A. J. Jacobson in SolidSfarr Chemrsfry-Cotnpounds(Eds: A. K. Cheetham, P. Day), Clarendon. Oxford, 1992. p. 183.  R. Schollhorn in Progresr i n Interculurion Reseurch (Eds.: W. Muller-Warmuth, R. Schollhorn). Kluwer, Dordrecht/Boston, 1994, p. 1. 1131 K. Rajeshwar. Adv. Muter. 1992, 4. 23. [I41 J. A. Switzer, R. J. Phillips. T D. Golden. Appl. P/ijs. Lrrr. 1995,66,819: J. A. Switzer. T. D. Golden, Adv. Muter. 1993, 5, 474; J. A. Switzer, R. J. Phillips, R. P. Raffaele in Supramoleculur. Architecture (Ed.: T. Bein). ACS Symposium Series 499. Amer. Chem. SOC.,Washington. 1992, p. 244.  R. A. van Leeuwen, C. J. Hung. D. R. Kammler, J. A. Switzer, J. P h w Chem. 1995,99,15247; J. A. Switzer, C. J. Hung. B. E. Breyfogle. M . G. Shumsky, R. van Leeuwen. T. D. Golden. Science 1994, 264. 1573.