Transparent Conducting Oxides—An Up-To-Date Overview

Transparent conducting oxides (TCOs) are electrical conductive materials with comparably low absorption of electromagnetic waves within the visible region of the spectrum. They are usually prepared with thin film technologies and used in opto-electrical apparatus such as solar cells, displays, opto-electrical interfaces and circuitries. Here, based on a modern database-system, aspects of up-to-date material selections and applications for transparent conducting oxides are sketched, and references for detailed information are given. As n-type TCOs are of special importance for thin film solar cell production, indium-tin oxide (ITO) and the reasonably priced aluminum-doped zinc oxide (ZnO:Al), are discussed with view on preparation, characterization and special occurrences. For completion, the recently frequently mentioned typical p-type delafossite TCOs are described as well, providing a variety of references, as a detailed discussion is not reasonable within an overview publication.


Introduction
Transparent conducting oxides (TCOs) are electrical conductive materials with a comparably low absorption of light. They are usually prepared with thin film technologies and used in opto-electrical devices such as solar cells, displays, opto-electrical interfaces and circuitries. Glass fibers are nearly lossless conductors of light, but electrical insulators; silicon and compound semiconductors are wavelength dependent optical resistors (generating mobile electrons), but dopant dependent electrical conductors. Transparent conducting oxides are highly flexible intermediate states with both these characteristics. Their conductivity can be tuned from insulating via semiconducting to conducting as OPEN ACCESS well as their transparency adjusted. As they can be produced as n-type and p-type conductives, they open a wide range of power saving opto-electrical circuitries and technological applications.
A still valuable overview of transparent conductive oxides is given in [1], basics to material physics of TCOs are discussed in [2], some structural investigation of TCOs was made e.g., in [3], preparation of TCOs was discussed in [4] and substitutes for the most popular transparent conducting oxide, namely ITO (indium-tin oxide), are listed in [5]. Here, based on a modern database-system, aspects of up-to-date material selections and applications for transparent conducting oxides are sketched, and references for detailed information are given. As n-type TCOs are of special importance for thin film solar cell production, ITO and the reasonably priced aluminum-doped zinc oxide (ZnO:Al) are discussed with view on preparation, characterization and special occurrences. For completion, the recently frequently mentioned typical p-type delafossite TCOs are described as well, providing a variety of references, as a detailed discussion is not reasonable within an overview publication.
As transparent conducting oxides are usually compound semiconductors-where the nonmetal part is oxygen-they are discussed along their metal elements. Metals were used as compound materials or dopants (with just a few percent content).

TCOs in General
In transparent conducting oxides (TCOs), the nonmetal part, B, consists of oxygen. In combination with different metals or metal-combinations, A, they lead to compound semiconductors, A y B z , with different opto-electrical characteristics. These opto-electrical characteristics can be changed by doping, A y B z :D (D = dopant), with metals, metalloids or nonmetals. Hence, metals can be part of the compound semiconductor itself, A, or can be a dopant, D. Scanning the periodic table of elements, with a view on the utilization of metals for TCOs, results in Table 1 (regarding just the 2nd and 3rd period, exclusively aluminum). Table 1. Published results regarding transparent conducting oxide (TCO)-layers, containing metallic elements e.g., from the 2nd and 3rd period of the periodic table of the elements (PE, excluding aluminum), including examples for the later discussed ZnO's and delafaossites (mayenites)-research with the web of knowledge using "TCO < name of element > oxide".  Outstanding good optical characteristics have been provided by tin-, indium-and zinc oxides (A = tin, indium, zinc). Well known is, for example, indium tin oxide (ITO), and the doping of zinc oxide with less than 5% aluminum (ZnO:Al). Moreover, doped delafossite and mayenite compounds are of upcoming interest (see Table 1). A variety of preparation and characterization methods was applied to investigate their different chemical structures and physical characteristics. These shall be briefly discussed.
The opto-electronic properties [129] of these TCO thin films can be changed by post process thermal annealing in an inert gas or reactive gas atmosphere [38, [130][131][132]. Especially surface and interface states can be influenced [133,134]. The deterioration of ZnO:Al thin films is discussed in [135].

Delafossite and Mayenite Type Transparent Conducting Oxides
Commonly, ITO-and ZnO-based TCO thin films are n-doped, as p-doping has been shown to be technologically more difficult. Fortunately, for delafossite compound semiconductors this is vice versa. They typically show TCO properties with semiconducting p-type characteristics. Delafossites, Cu x A y O z , are commonly ternary material combinations of copper, Cu, one (or more) further metal(s), A, (aboriginal iron, A = Fe) and oxygen, O.
Ag-Cu and Rh-Mg replacements were for example studied in the quinternary structure Cu 1−x Ag x Rh 1−y Mg y O 2 [179].
The CuA III O 2 group shows increasing band gap from A III = Al, Ga, to In. The largest gap CuInO 2 can be doped both n-and p-type but not the smaller gaps CuAlO 2 and CuGaO 2 [189]. Therefore, doping CuInO 2 with Ca results in p-type, doping with Sn in n-type semiconducting TCO thin films [188,190]. Bidirectional doping is possible for CuFeO 2 , too (p-type: Mg, n-type: Sn [191]). In addition, the electronic structure of CuAO 2 (A = Al, Ga, Y) was discussed in [192][193][194][195][196] and its luminescent properties in [197]. Defect analyses have been made with the screened-hybrid density functional theory [160].
Because of the structural anisotropy of the CuAlO 2 -crystal, anisotropic electrical conductivity was detected in [200]. Ohmic contacts between CuInO 2 and Cu are reported in [170].
The crystal structures and chemistries are by far the best investigated topics in delafossite (semi)conductor research and systematically discussed in [201,193]; the according temperature dependency is shown in [202].

Further Aspects to Technological Advances of Transparent Conducting Oxides
Reasons for technical advances in transparent conducting oxides are manifold-influencing aspects are: The investigation of adequate novel materials and material-combinations, as for example the first delafossites by Charles Friedel in 1873 (named after the French mineralogist and crystallographer Gabriel Delafosse); an increasing financial support for research according to political decisions, as for example the increased financial support of solar cell investigations and therefore of TCOs by the present nuclear power phase-out in Germany; the publication of new results, as research groups in industrial companies often reserve important information; and the efficiency of modern literature data-bases, as only included literature can be found and selected.
Therefore, technical advances in transparent conducting oxides may be illustrated researching the web of knowledge (Thomson Reuters). Applying e.g., the search item "TCO < name of element > oxide" leads to the carefully selected citation statistics, shown in Table 2. Again, the already discussed elements aluminum (Al), zinc (Zn), indium (In) and tin (Sn) show the by far highest nominal citation impacts. In order to demonstrate the technical advances in transparent conducting oxides, the gradient of citations over the years 2007 until 2011 shall be printed for these four elements in Figure 1. This indicates, that the focus of investigation was preferably set on ITO and that A TCO       Despite these four elements, let us regard the next five metals, which exhibit the most average citations per year in TCO-related publications, see Table 2, Figure 2. Hence, Cadmium (Cd) is discussed as CdO:D (D = Ga, Sn, Sm, Eu, Gd, or Dy), CdIn 2 O 4 or Cd 2 SnO 4 , where H 2 -annealing is frequently applied to widen the energy gap [203][204][205].
Copper (Cu) represents the group of doped and undoped CuO 2 and delafossites, see above. Gallium (Ga) on the one hand is used as dopant, D (about 2% at ), for ZnO and CdO. On the other hand Ga is the metallic part, A, of Ga 2 O 3 . Based on this, gallium zinc oxide (GZO: ZnGa 2 O 4 ) is produced with 90% wt of Ga 2 O 3 and 10% wt of ZnO. Moreover, aluminum gallium zinc oxide (AGZO) is a combination of aluminum zinc oxide (AZO) and GZO, respectively indium gallium zinc oxide (IGZO) a combination of IZO and GZO [206,207].
Niobium (Nb) is exclusively used as dopant, with an atomic concentration of about 3% at -6% at , primarily for TiO 2 :Nb but also for SnO 2 :Nb [208,209].
The upcoming importance of transparent conductive materials for thin film solar cells, opto-electrical interfaces, displays and opto-electrical circuitry widens the area of investigation. So, exotic dopants, such as sodium (Na) [213] and manganese (Mn) [214] for zinc oxides (ZnO), zirconium (Zr) [215], platinum (Pt) and tungsten (W) [216] for indium oxide (In 2 O 3 ), ITO and IGZO or lanthanum (La) [217] for strontium stannate La x Sr 1−x SnO 3 have been discussed in the last few years.
Finally, ultra-thin metals without any oxygen content (except natural oxidation in air at room temperature)-as for example nickel (Ni)-have been applied as optical transparent conducting materials [218].

Conclusions
Based on a modern database-system, aspects of up-to-date material selections and applications for transparent conducting oxides have been sketched; references for detailed information have been given for the interested reader. As n-type TCOs are of special importance for thin film solar cell production, indium-tin oxide (ITO) and the reasonably priced aluminum-doped zinc oxide (ZnO:Al) have been discussed with view on preparation, characterization and special occurrences. For completion, typical p-type delafossite TCOs have been described the same way, providing a variety of references, as a detailed discussion is not reasonable within an overview-publication. Moreover, absolutely unusual, novel TCO materials have been discussed and their presence and development in the world of science pointed out. Trends have been shown.
As transparent conducting oxides are usually compound semiconductors-where the nonmetal part is oxygen-they have been discussed along their metal elements. Metals were used as compound materials or dopants (with just a few percent content).