Recent Studies On Structural, Optical, And Electrical Properties Of Semiconducting Chalcogenide Thin Filmsdocx

=== Recent studies on structural, optical, and electrical properties of semiconducting chalcogenide thin films ===

Recent Studies on Structural, Optical, and Electrical Properties of

Semiconducting Chalcogenide Thin Films.

State Of Art

By

Dr/ Samia Ahmed Abd El Kawy Gad

Solid State Physics

Physics Division

Centre

2016

Abstract:

The study of glasses has been important because of their great technological applications. One class of these materials, amorphous semiconductors, has moved a great deal of interest during the last few years. This interest advances in part from the fact that solid state physics, after obtaining a signally high level of scientific understanding of crystals, can now hope for identical achievements in connection with disordered materials. Of equal importance is the fact that the metastability of amorphous semi-conductors provides them with certain unique properties that may be of considerable technological significance. This state of art is purposed to provide an overview of the field, its present standing, and its promise. The fundamental structural and electronic properties and the present level of understanding of these Properties are of primary concern. However, much of the progress in solid state physics has traditionally been motivated by technological considerations. Therefore, the principal aspects of the physics underlying the more important amorphous semiconductor devices are discussed, as well as the technological setting in which this new field finds itself.

1.Introduction:

The chalcogenide name’s arises from the Greek word "chalcos" meaning ore and "gen" meaning formation, thus the term chalcogenide is generally considered to mean ore former.[1] A chalcogenide is a chemical compound consisting of at least one chalcogen anion and at least one more electropositive element. Many metal ores exist as chalcogenides. Photoconductive chalcogenide glasses are used in xerography. Some dyes and catalysts are also based on chalcogenides. The metal dichalcogenide MoS2 is a common solid lubrican. There are three elements in Group 16 in the periodic table: sulfur, selenium and tellurium. Such glasses are considered covalently bonded materials and may be classified as network solids; in effect, the whole glass matrix acts as an boundlessly bonded molecule. There are other elements such as polonium is chemically a chalcogenide as well, it is not used in chalcogenide glasses because of its strong radioactivity and high price and oxygen is also a group 16 element, but it is not considered a chalcogenide.

Although oxide materials are the oldest known glass forming systems it has become more conventional to handle them separately from more recently discovered chalcogenide compounds. Oxide materials proceed rather differently from other chalcogenides, especially their widely different band gaps contribute to very different optical and electrical properties. Chalcogenides can place naturally as minerals; there are two compounds of the most famous being FeS2 (pyrite) and AuTe2 (calaverite).

The conventional chalcogenide glasses (contained sulfur- such as As-S or Ge-S) are strong glass and can form glasses within large concentration. The ability of Glass-formation decreases with increasing molar weight of component elements; i.e., S > Se > Te. The properties of semiconducting chalcogenide glasses were discovered in 1955 by B.T. Kolomiets and N.A. Gorunova from Ioffe Institute, USSR.[2,3] This invention started many researches and applications of this new class of semiconducting materials. There are another chalcogenide compounds such as AgInSbTe and GeSbTe can be used in rewritable optical disks and phase-change memory devices. They are brittle, glass-formers; by subjecting it to heat, they can be transformed between an amorphous and a crystalline state, so their optical and electrical properties changed and allowing the storing information.

Another famous type of chalcogenides has glass-forming regions where three chalcogens are bonded to two group 15 elements. Most stable binary chalcogenide glasses consist of a chalcogen and a group 14 or 15 element. This allows a wide range of atomic ratios. Ternary glasses announces a larger variety of atoms to be combined into the glass structure, thus giving greater engineering capacity [4]. chalcogenides can occur over a wide range of compositions, but all of which can’t exist in glassy form, it is often possible to find materials with which these non-glass-forming compositions can be alloyed in order to form a glass. An example of this is gallium sulphide-based glasses. Gallium sulphide on its own is not a known glass former; however, it readily bonds with sodium or lanthanum sulphides to form a glass, gallium lanthanum sulphide (GLS). Amorphous chalcogenide materials can be classed by the type of atoms to which they bond to form amorphous systems. One of the more well-known chalcogenide glasses is based on arsenic trisulphide, an example of a stable binary glass which preferentially exists in a glassy phase. Amorphous materials are not new now.

Iron reach siliceous glassy materials found from the Moon! (Apollo mission) Billion years old! People has been preparing glassy materials (i.e. SiO2) for thousands of years. Scientific investigations started about 70 years earlier. Zachariasen (1932) proposed that SiO2 structure can be described by a continuous random network (CRN).

A perfect crystal is that in which the atoms are arranged in a pattern that repeats periodically in three dimensions to an infinite extent.

An imperfect crystal is that in which the atoms are arranged in a pattern that repeats periodically in three dimensions to a finite extent.

Real crystal: imperfect crystal having defects like vacancy, interstitial (foreign) atoms, dislocations, impurities, etc.

1.1.Amorphous-Crystalline Transformations

The physical, chemical, and mechanical properties of amorphous materials can all be strongly affected by the transformation to the crystalline state. The changes in electrical and optical properties have already been noted. Some examples, representative of the range of these changes, may be useful. The room temperature resistivity of amorphous Ge and Si films can be as much as five orders of magnitude larger than that of the corresponding polycrystalline films. The extent of the change depends sensitively on the details of the film preparation. By dissimilarity in AsS and AsSe , the glasses are less resistive than the corresponding crystals. The electrical band gap, as determined from the temperature dependence of the conductivity, is respectively 0.2 and 0.55 eV in c-InSb and a-InSb. In c- and a-Te, the corresponding quantities are 0.33 and 0.87 eV. The index of refraction in Te and Se decreases respectively by 40 percent and 12 percent in going from the crystalline to the amorphous state. On the other hand, in Ge the index changes but slightly in the opposite direction. The previous results are representative of the simplest types of measurements. As will be seen later, even the Hall effect and thermoelectric power, whose measurements represent no difficulty in many crystalline semiconductors, are still poorly established quantitatively in amorphous materials. Further citation of experimental results in the present context might therefore be perverting. There have been qualitative observations of changes in chemical properties such as wettability, reactivity, adhesion, and solubility resulting from amorphous-crystalline transformations. Mechanical properties, such as hardness, thermal expansion, and sound velocity are similarly affected.

This unique potential for change in amorphous materials is of both fundamental and technological importance. Its measurement and clarification is a challenge and important problem for the solid state physicist and chemist; its using is a challenge for the inventive inventor. Proposals have been made to utilize these and other effects for optical mass memories, memory and threshold switches, electroluminescent displays, non-impact lithographic plates, and imaging applications including photography and copiers [5]. As an example of how these unique properties can be used technologically, the memory device takes candid advantage of the fact that glasses are energetically metastable. In the Te-based glasses, the Te rich phases tend to segregate from the rest at sufficiently high temperatures. Such temperatures can be achieved by joule heating. Phase separation can also be achieved by photocrystallization. Predicating on the maximum temperature and the rate of cooling, the glass then stabilize either into a state containing crystalline filaments or returns to its initial amorphous state. The two differ by orders of magnitude in conductivity.

1.2. Types of chalcogenides:

Alkali metal and alkaline earth chalcogenides

Alkali metal and alkaline earth monochalcogenides are as salt, being colourless and often water-soluble. The sulfides tend to undergo hydrolysis to form derivatives containing bisulfide (SH−) anions. The alkali metal chalcogenides often crystallize with the antifluorite structure and the alkaline earth salts in the sodium chloride cop.

Transition metal chalcogenides

Transition metal chalcogenides exist with many stoichiometries and many structures.[5] Most common and most important technologically, however, are the chalcogenides of simple stoichiometries, such as 1:1 and 1:2. There are cases include metal-rich phases (like Ta2S), which subject to wide metal-metal bonding,[6] and chalcogenide-rich materials such as Re2S7, which features extended chalcogen-chalcogen bonding. For classifying these materials, the chalcogenide is often viewed as a dianion, i.e., S2−, Se2−, and Te2−. In fact, transition metal chalcogenides are covalent, not ionic, as indicated by their semiconducting properties [7].

Monochalcogenides

Metal monochalcogenides have the formula M B, where M = a transition metal and B = S, Se, Te. They crystallize in one of two cops, named after the corresponding forms of zinc sulfide. In the zinc blende structure, the sulfide atoms collect in a cubic symmetry and the Zn2+ ions occupy half of the tetrahedral holes. The main alternative structure for the monochalcogenides is the wurtzite structure wherein the atom connectivities are similar (tetrahedral), but the crystal symmetry is hexagonal. A third class for metal monochalcogenide is the nickel arsenide lattice, where the metal and chalcogenide each have octahedral and trigonal prismatic coordination, respectively. This class is commonly subject to nonstoichiometry [8]. The important monochalcogenides are some dyes, especially cadmium sulfide. Many minerals and ores are monosulfides [9].

Dichalcogenides

Metal dichalcogenides have the formula MB2, where M = a transition metal and B = S, Se, Te [10]. The most important members are the sulfides. They are always dark diamagnetic solids, insoluble in all solvents, and showing semiconducting properties. In terms of their electronic structures, these compounds are usually viewed as derivatives of M4+, where M4+ = Ti4+ (d0 configuration), V4+ (d1 configuration), Mo4+ (d2 configuration). Titanium disulfide was exacted in prototype cathodes for secondary batteries, employing its ability to reversibly expose to insertion by lithium. Molybdenum disulfide, the subject of many thousand publications, is the main ore of molybdenum where it is called molybdenite. It is used as a solid lubricant and catalyst for hydrodesulfurization. The corresponding diselenides and even ditellurides are known, e.g., TiSe2, MoSe2, and WSe2.

Transition metal dichalcogenides typically take over either cadmium diiodide or molybdenum disulfide structures. In the CdI2 class, the metals shows octahedral structures. In the MoS2 class, the metals show trigonal prismatic structures [11]. The strong bonding between the metal and chalcogenide ligands, discrepancy with the weak chalcogenide – chalcogenide bonding between the layers. Owing to these contrasting bond strengths, these materials participate in intercalation by alkali metals. The intercalation process is accompanied by charge transfer, reducing the M(IV) centers to M(III).

Pyrite and related disulfides

Iron pyrite is a famous mineral, consisting of Fe2+ and the persulfido anion S22− is in discrepancy to classical metal dichalcogenides . The sulfur atoms within the disulfido dianion are bound together via a short S-S bond [12]. Transition metal disulfides (Mn, Fe, Co, Ni) almost always arrange the pyrite or the related marcasite class, in contrast to early metals (V, Ti, Mo, W) which adopt 4+ oxidation state with two chalcogenide dianions.

Tri- and tetrachalcogenides

Several metals, (Ti, V, Cr, Mn groups) also produce trichalcogenides. These materials are usually described as M4+(E22−)(E2−) (where E = S, Se, Te). A well known example is niobium triselenide. Amorphous MoS3 is generated by treatment of tetrathiomolybdate with acid:

MoS42− + 2 H+ → MoS3 + H2S

The mineral patrónite, which has the formula VS4, is an example of a metal tetrachalcogenide. Crystallographic analysis shows that the material can be considered a bis(persulfide), i.e.

V4+,(S22−)2[13].

Main group chalcogenides

Chalcogen derivatives are known for all of the main group elements except the noble gases. Usually, their stoichiometries follow the classical valence trends, e.g. SiS2, B2S3, Sb2S3. The structures of many of main group materials are dictated, not by close packing, but by directional covalent bonding [14]. The chalcogen is assigned positive oxidation states for the halides, nitrides, and oxides.

1.3. Semiconductor, Amorphous

  A substance in the amorphous solid state that has the properties of a semiconductor. Also, solid state material that can be switched from one state to another. For example, the recording layer in phase change rewritable CDs and DVDs switches between an unstructured amorphous state that absorbs light and a structured crystalline state that allows light to pass. In phase change memory, the storage cell switches between a state of low resistance to one of very high resistance. Contrast with crystalline semiconductor. Amorphous semiconductors are divided into three groups: covalent amorphous semiconductors, such as amorphous Ge and Si, InSb, and GaAs; chalcogenide glasses, such as As31Ge30Se21-Te18;; and oxide glasses, such as V2O5-P2O5, and dielectric films, such as SiOx, Al2O3, and Si3N4 [15].

The energy spectrum of amorphous semiconductors differs from that of crystal semiconductors in the presence of density “tails” of electronic states that penetrate the energy gap. According to one theory, an amorphous semiconductor should be considered as a heavily doped, heavily compensated semiconductor such that the bottom of its conduction band and the top of its valence band fluctuate. These fluctuations are large-scale and of the order of the width of the energy gap. Electrons in the conduction band and holes in the valence band are divided into a system of “droplets” located in potential wells and separated by high barriers. Electrical conduction in amorphous semiconductors is accomplished at very low temperatures by means of electron tunneling through the barriers between wells in a manner analogous to hopping conduction. At higher temperatures, electrical conduction is due to the thermal excitation of carriers into higher energy levels.

Amorphous semiconductors have various practical applications. Chalcogenide glasses are used in television camera tubes and for hologram recording because of their transparency to infrared radiation, high resistance, and high photosensitivity. Dielectric films are also used in metal-dielectric-semiconductor structures . In systems consisting of an amorphous semiconductor film between two metals, the rapid (10-l0 sec) transition (switching) of the amorphous semiconductor from a highly resistive state to a conductive state is possible when the applied voltage exceeds a threshold voltage. In particular, there exists memory switching wherein the highly conductive state is keeped even after the voltage is removed; the memory is usually erased by a short intense current pulse. The conductive state in memory systems is due to the partial crystallization of amorphous semiconductors.

1.4.Classification of amorphous semiconductors:

1. Tetrahedrally bonded amorphous semiconductors: a-Si, a-Ge, a-C(?) and their alloys like a-SiC, etc. (tathogen)

2. Chalcogenide glasses: a. a-S, a-Se, a-Te, a-SxSe1-x (pure chalcogenide)

b. a-As2Se3, a-As2S3, a-P2Se3 , etc. (pnictogen-chalcogen (V-VI))

c. a-GeSe2, a-SiS2, a-SiSe2, etc. (tetragen-chalcogen (IV-VI))

1.5. Amorphous solid

A rigid material whose structure lacks crystalline periodicity; that is, the pattern of its constituent atoms or molecules does not repeat periodically in three dimensions. In the present terminology amorphous and noncrystalline are synonymous. A solid is differentiated from its other amorphous counterparts (liquids and gases) by its viscosity: a material is considered solid (rigid) if its shear viscosity exceeds 1014.6 poise (1013.6 Pa · s).

Oxide glasses, generally the silicates, are the most familiar amorphous solids. However, as a state of matter, amorphous solids are much more spread than just the oxide glasses. There are both organic (for example, polyethylene and some hard candies) and inorganic (for example, the silicates) amorphous solids. Glasses can be prepared which extend a broad range of physical properties. Dielectrics (for example, SiO2) have very low electrical conductivity and are optically transparent, hard, and brittle. Semiconductors (for example, As2SeTe2) have intermediate electrical conductivities and are optically opaque and brittle. Metallic glasses have high electrical and thermal conductivities, have metallic luster, and are ductile and strong [16].

The obvious uses for amorphous solids are as window glass, container glass, and the glassy polymers (plastics). Less widely known but nevertheless established technological uses include the dielectrics and protective coatings used in integrated circuits, and the active element in photocopying by xerography, which depends for its action upon photoconduction in an amorphous semiconductor. In optical communications a highly transparent dielectric glass in the form of a fiber is used as the transmission medium.

It is the changes in short-range order (on the scale of a localized electron), rather than the loss of long-range order alone, that have a profound effect on the properties of amorphous semiconductors. For example, the difference in resistivity between the crystalline and amorphous states for dielectrics and metals is always less than an order of magnitude and is generally less than a factor of 3. For semiconductors, however, resistivity changes of 10 orders of magnitude between the crystalline and amorphous states are not uncommon, and accompanying changes in optical properties can also be large [17].

One class of amorphous semiconductors is the glassy chalcogenides, which contain one (or more) of the chalcogens sulfur, selenium, or tellurium as major constituents. These materials have application in switching and memory devices. Another group is the tetrahedrally bonded amorphous solids, such as amorphous silicon and germanium. These materials cannot be formed by quenching from the melt (that is, as glasses) but must be prepared by one of the deposition techniques mentioned above.

When amorphous silicon (or germanium) is prepared by evaporation, not all bonding requirements are satisfied, so a large number of dangling bonds are introduced into the material. These dangling bonds create states deep in the gap which limit the transport properties. The number of dangling bonds can be reduced by a thermal anneal below the crystallization temperature, but the number cannot be reduced sufficiently to permit doping.

1.6. Glasses and Their Uses

It is usual to restrict the specification "glass" to those amorphous solids that have been formed by cooling a liquid. However, it is dubitable that "glasses'* so defined differ sharply in microscopic character from amorphous solids with the same composition formed in other ways. Thus the terms "amorphous solids" and "glasses" will be taken to be equivalent in this state of art. Glasses can be metallic, semiconducting, or insulating. The forces bonding the atoms are similar to those found in crystals. The chemical bonding can be covalent, ionic, metallic, van der Waals, or hydrogen bonding, or combinations of these.

Most glasses, however, fall into the mainly covalent category. Because of their metastability, glasses exhibit properties that are quite unique and remarkable. They do not undergo a first-order-phase transition at the melting temperature. Instead, they soften gradually at sufficiently high temperatures and pass more or less continuously into the liquid state. The molten glass may either return to its original state if it is cooled sufficiently rapidly, or crystallize if it is cooled slowly. Glasses containing several constituents may exhibit a separation into phases having different compositions on a very minute spatial scale. These structural transformations have a qualitative effect on the electrical and optical properties in various types of glasses. These are of interest not only as phenomena in themselves but also because of their technological importance. Finally, the fact that glasses are structurally disordered suggests that their properties can be relatively insensitive to high-energy radiation and bombing. The oxide glasses are, perhaps, the most familiar.

The soda-lime-silicate glasses (mixtures of Na2O, CaO, and SiO ) are good dielectrics, thermal insulators, and optical transmitters. Because they soften gradually with increasing temperature, it is possible to pour, mold, roll, press, and

float ordinary window glass, processes that are essential in its manufacture. Many, though not all, oxide glasses are insulators with conductivities less than 10-8 Ω-1 cm-1 . This fact, as well as the natural tendency of metals to oxidize, makes these materials very useful in solid state device technology. This fact, as well as the natural tendency of metals to oxidize, makes these materials very useful in solid state device technology [18].

Semiconducting glasses were not investigated to any large degree before 1955. In contrast to the insulating glasses, the conductivity in these substances is electronic rather than ionic. As a result the conductivity is larger, ranging from 10-13 -10-3 Ω-1 cm-1. While some of the semiconducting glasses are oxides, the most widely studied examples do not contain oxygen. Instead they contain another constituent, such as S, Se, Te of group six of the periodic table. Such elements are called "chalcogens" and the glasses involving them are known as chalcogenides. The chemical bonding in such glasses is mainly covalent with a smaller ionic contribution, although cases involving mixed covalent and van der Waals binding are also frequently encountered in materials such as Se [19]. The chalcogenide glasses have received a great deal of attention because of their established or possible importance in connection with electrophotography, infrared transmitting windows, electronic switching, and electronic and optical memory applications. Work at Energy Conversion Devices, Inc., (BCD) has particularly stimulated the development of applications for chalcogenide glasses. Elemental amorphous Se has been investigated extensively in part because it forms the essential component of the photosensor involved in xerography. In practice the commercial compositions may contain some As and traces of other elements. As-Se glasses have also been studied at RCA in connection with vidicon applications. Indeed, the fundamental properties of these glasses have received considerable attention both in this country and the Soviet Union. The interest in technological applications of chalcogenide glasses has stimulated interest in other chalcogenide glass compositions, such as those belonging to the Ge-Te family [20].

Crystalline Si and Ge are among the best understood solids. Their amorphous forms are of interest particularly in connection with fundamental research directed toward discovering physical differences between the crystalline and amorphous states. The planning of these differences would be expected to be simpler in elemental glasses that contain only structural and not compositional disorder. The metallic glasses usually occur as compounds of the form A3B to A5B where A is a noble or transition metal and B is a metalloid like Si, Ge, or P. While they show a variety of interesting properties including radiation hardness, they have not as yet found significant use in electronic technology [21]. Since amorphous semiconductors have been observed to crystallize in the neighborhood of conventional metallic contacts, speculation has focused on the possibility of using amorphous metal contacts on semiconducting glasses in order to prevent this from happening. It should be confirmed that the study of glasses is important, not only for technological reasons but also, more fundamentally, because they are systems having structural and possibly compositional disorder [22]. Until very recently solid state physics has been concerned almost exclusively with crystalline materials. Considerations of disorder confirmed effects arising, for example, from lattice vibrations, point defects, and impurities and dislocations in small concentrations that only influence the crystalline properties weakly. However, during recent years, emphasis has been increasingly given to the investigation of the properties of strongly disordered materials, such as liquids, binary substitutional alloys, and amorphous materials. Clearly, an increased understanding of liquids and alloys will be of benefit to those investigating amorphous semiconductors, just as further theoretical insight concerning the materials considered in this state of art will aid those investigating liquids and alloys[23,24].

2. Characterization

2.1. Methods of preparation

Quenching

The most common preparatory method is quenching from the melt or vapor. Melt quenching rates extend from 10-2 0C/second in an annealing furnace, to 103 – 104

0C/second in strip furnaces, to 105 – 107 0C/second by the more complex splat cooling techniques. Vapor quenching rates overlap the high end of this range and rates as high as l015 0C/second has been reported. The choice of method is usually dictated by the material of interest since the faster quench rates are achieved at greater experimental complexity and cost. The number of nucleii and crystal growth rate determines the minimum required quench rate[25].

Vapor Deposition

Vapor deposition is the most commonly used technique for materials considered in this art. A number of special techniques have been developed and will be mentioned here in the context of advantages or disadvantages in the preparation of amorphous semiconductors. Numerous reviews on thin-film preparation are available for further information on specific systems, techniques, and materials.

All vacuum deposition systems consist of several basic elements, a vacuum chamber, a source of the material to be deposited, a substrate and associated fixturing. A most significant factor is the amount and nature of contaminants,

including the ever present atmospheric gases, available for incorporation into the material of interest. All pumping systems except cryosorption contribute some

foreign material, e.g., hydrocarbons, Ti, and Hg. Fortunately, mercury pumps are rarely if ever used in this application since some materials, such as Se, are excellent getters for Hg vapor. In addition, systems vary in their pumping speed for various atmospheric gases, in effect concentrating certain species. The surface of the film being deposited is exposed to sufficient background gases to condense ~ 4 monolayers per second at 10-5 Torr and at 10-9 Torr ~ 4 x 10-4 monolayers/sec. What, if any, material is incorporated, how it is incorporated and its effect must be individually considered. It is frequently noted that it becomes more difficult to quench an amorphous film at higher vacuums indicating a stabilizing effect of impurity incorporation. The substrate temperature during sample preparation is a particularly important parameter. Too low a temperature results in low-density films with poor adhesion. At higher temperatures there may be sufficient mobility only to allow complete replication of the substrate, while slightly above this there is sufficient flow to provide very smooth surfaces and higher densities. At still higher temperatures, crystallization begins and the ability to quench an amorphous phase is lost. Since physical and molecular structure can be affected by quench rate and annealing, other measured properties may vary with substrate temperature at preparation. The classical method of providing a vapor of the desired material at the substrate is evaporation. The sophistication comes in the choice of heating methods. Simple resistance or RF induction heating of a source is appealing because of its simplicity, but introduces problems of contamination and fractional distillation in multicomponent materials. The first problem can be minimized by proper choice of crucible material or eliminated by using the material as its own crucible. The most common method for this is electron bombardment where a focused electron beam causes evaporation from a small heated region on the surface of a larger piece of the material of interest [26].

In flash evaporation, material is continuously fed into a source at a rate slow enough to prevent the buildup of a pool of molten material so that the instantaneous average composition of the vapor is that of the feed material. This method is slow, inefficient, and usually results in films with numerous defects due to spatter of solid and liquid material from the source. A variation uses continuous feed to a pool of molten material with the rate controlled so that the vapor is of constant composition, although different from that of the feed material[27].

Sputtering is another method that can avoid fractionation effects and is finding ever wider applications. Solids can be transformed to a disordered state in a solid-state reaction with the energy provided by radiation (neutron, α particles, etc.),shear, or chemical reaction in processes often referred to as amorphization. The chemical reaction need not be completely solid state; in fact, reaction with, or evolution of, a vapor in an oxidation, reduction, or disproportionation reaction is often involved [28].

2.2. FUNDAMENTAL PROPERTIES OF AMORPHOUS SEMICONDUCTORS

* In any real crystal there will be defects (e.g., impurities, vacancies, interstitials,

dislocations),which scatter the Bloch waves, so that the wave vector, k, is only an approximate quantum number characterizing the states. Scattering is also produced by phonons, and at high energies or finite temperatures, by electron-electron interactions. In addition, defects can give rise to states within the energy gap of the perfect crystal [29].

* At low temperatures, the Fermi level of a perfect semiconductor lies halfway between the bottom of the conduction band and the top of the valence band. The number of charged carriers (electrons in the conduction band and holes in the valence band), and hence the electrical conductivity at low temperatures, is proportional to e-(Ea/kT) where the activation energy Ea is one half the energy gap, Eg. This form for the conductivity is called the "intrinsic" conductivity of the material. In any real material, however, at sufficiently low temperatures, the very small intrinsic conductivity will be dominated by the contribution of carriers associated with impurities or other defects, which give rise to states in the gap, and generally cause the Fermi level to be closer to one or the other side of the energy gap. The conductivity then depends strongly on the number and kind of defects present, and is called "extrinsic." In general, the more imperfect the material, the larger the value of the extrinsic conductivity, and the larger the temperature range over which the extrinsic conductivity is seen.

The principal optical absorption band in a crystalline semiconductor is associated

with transitions of an electron from a state in the valence band to a state in the conduction band having the same wave vector as the initial state. The fundamental optical absorption edge occurs at the gap energy, Eg. For crystals where the lowest state in the conduction band occurs at a different point in the Brillouin zone than the top of the valence band, optical absorption near the fundamental edge requires the aid of a phonon or crystal defect to conserve wave vector. In a real crystal, absorption well below the threshold, Eg ,can occur because of transitions from a

Filled defect level inside the energy gap to the bottom of the conduction band,

from the top of the valence band to an empty defect level, or from one defect

level to another[30].

In a crystal with a small density of defects, one would see a number of sharp thresholds in the absorption, corresponding to transitions between the isolated defect levels and the edge of the conduction or valence band. In a crystal with a large density of defects, or many kinds of defects, one would find only a relatively featureless tail in the optical absorption below the fundamental edge. An absorption tail below the fundamental edge can also result from interaction of the electronic states with lattice distortions (phonons). The position of the fundamental absorption edge is also lowered somewhat by the electron-electron interaction,

which leads to a series of electron-hole bound states (excitons) just below the

continuum of states beginning at Eg. Many features of the crystalline semiconductor persist in the amorphous. There will again be a filled valence band, roughly derived from bonding orbitals, and an empty conduction band derived from antibonding orbitals. In the amorphous material, k is not a good quantum number for the electronic states. Some remnants of k-conservation may persist, however, in that states in a given energy range may contain predominantly wave vectors associated with particular portions of the Brillouin zone of the ordered structure [31].

2.3. STRUCTURE AND BONDING IN AMORPHOUS SOLIDS

When glass is formed by cooling a liquid, it is often observed that the heat capacity and thermal expansivity drop sharply in the vicinity of Tg, as defined above. However, the temperature at which these abrv.pt changes occur is lower for

lower cooling rates, and it simply marks the point below which the amorphous system is no longer in internal configurational equilibrium. That this equilibrium

is not achieved in the glasses of ordinary experience is to be expected in view of the very large time constants noted earlier for configurational changes at T< Tg. Also, it is not likely that amorphous solids formed by the various deposition techniques are in internal equilibrium. It follows that, at the same temperature and pressure, two amorphous solid specimens with the same compositions may still differ somewhat in internal structure. This behavior is quite analogous to that of an compositionally disordered crystalline alloy at temperatures where the time constant for interpositional exchanges is very long. An Amorphous solid, if constrained from crystallizing, would presumably relax after an infinite time to an "ideal" amorphous state of minimum enthalpy and entropy. If we classify condensed materials according to the type of bonding responsible for their coherence, i.e., covalent, metallic, ionic, van der Waals,or hydrogen, every class contains some members that can be put into the amorphous solid form [21]. In general, the tendency to amorphous solid formation is greatest in some covalently bonded materials, and least in most ionic and metallically bonded materials.

The problem of whether the structure of amorphous solids is, in general, distinct and unique, or only trivially different from that of a crystalline solid, has persisted for a long while without being resolved definitively. The continuous random models for amorphous structure, of the type developed by Zachariasen [32], Bemal, and others [33], seem to be uniquely different from crystal structures. At the other extreme, there are the models based on the idea that the amorphous solid is an assembly of randomly oriented microcrystallites. For the microcrystallite models to be meaningful, it appears that the crystallite dimension should equal or exceed two unit cell dimensions. At this dimension, most of the material in the system would lie on crystallite boundaries, and the atomic configurations across these boundaries would be more important than those within the crystallites for the over-all description of the amorphous structure [34]. Model studies indicate that the atomic configurations connecting highly misoriented crystallites are quite similar to some of the configurations in the continuous random models; for example, pentagonal arrangements are often seen. This suggests the interesting possibility that the structure of a microcrystallite assembly might degenerate to a continuous random structure when the crystallite size falls below a certain limit.

Structure

Structure, as defined here, ranges in scale from macroscopic. defects (e.g .,cracks and bubbles) which are important if experimental data is to be meaningful to atomic bonding and molecular structure which are more important for theoretical understanding and modeling. Phenomena such as crystallization and phase separation can span the range from visible to the unaided eye to the angstroms limit of the most powerful microscopes available. Obviously no single tool can be recommended for all characterizations.

Macroscopic (≥ 10 Å)

Optical microscopy [35] remains a primary tool. Gross defects such as bubbles, cracks, foreign inclusions, surface contamination and reaction products, as well as crystallization and phase separation down to about 1μ, can often be easily identified. For other than very thin films or surface effects an infrared microscope is required for many materials [36]. The scanning electron microscope (Fish 70)overlaps optical microscopy and extends its range down to less than 100Ä. Using the standard secondary electron mode, surface defects can 1be studied throughout this range. Other modes of operation, such as conduction, back scattering and cathode luminescence can give additional information on surface and/or bulk properties [37] .

Microscopic (≤10Å)

The greatest handicap to structural characterization of amorphous materials is that the absence of long range order precludes complete structure determination by standard x-ray, electron and neutron diffraction methods. The scattering which is observed can be used to determine a radial distribution function (RDF)from which the number and distance of at least the first and second nearest neighbors of an average atom can be determined. Recently Mozzi and Warren succeeded in greatly improving the resolution of the x-ray method by using procedures which minimized the Compton modified scattering [38].

2.4. Dense Random Packing (DRP) of Hard Spheres

The coherence of those amorphous solids, with which this study is primarily concerned, is due mostly to covalent bonding as in amorphous germanium; a mixture of covalent and van der Waals bonding, as in amorphous selenium; or a mixture of covalent and ionic bonding, as in the soda-lime-silicate glasses. However, it may be instructive to consider first the nature of the bernal dense random packed structure (DRP structure) of uniform hard spheres. This structure has a density about 86 percent that of crystalline close packing. It has been

characterized by the distribution of its Wigner-Seitz cells (voronoi polyhedra) amongst a small group of ideal forms from which the actual forms of the cells can be derived by small distortions. From this stand point the structure can be viewed as an admixture of crystallographic cells and non-crystallographic cells (such as pentagonal dodecahedra).The unique feature of the structure is these non-crystallographic elements. When short-range interatomic interactions dominate, as in the condensation of attracting uniform hard spheres, packing to form tetrahedral holes (e.g., Rather than octahedral)will be preferred. This should almost always lead to a randomly packed structure, an expectation that was confirmed by Bennett [39] in a study of computer-generated hard-sphere structures.

2.5. Structure of covalently bound amorphous systems

In covalently bound systems, the analog of the DRP(Dense Random Packing) structure is the random network type of structure that was first proposed by Zachariasen [32]. Recent model studies have shown that these structures can account remarkably well for the pair distribution functions, densities, and configurational entropies of tetrahedrally coordinated amorphous systems. The models are constructed according to the following general procedure [40]:(1) the number of nearest neighbors, their average spacing, and the dispersion of these spacing’s around the average is made the same as in the corresponding crystal; (2) a certain distribution of distortions of the bond angles from, their ideal crystal values is allowed;(3) the surface density of dangling bonds is kept constant during the building of the model. In this way an "ideal" amorphous structure is formed, which can be enlarged indefinitely without the development of prohibitive strains.

The random network structures that have been discussed, containing no internal dangling bonds, represent ideal structures, which may be more or less closely approached by actual amorphous structures, and which might be the end states of the thermal relaxation processes discussed above. Depending upon their conditions of formation, the actual structures may contain considerable numbers of internal dangling bonds and voids. Even so, the structure of the greater part of the amorphous body might approximate the ideal according to a "swiss cheese" model

[41]. The group V elements, e.g. As and Sb can crystallize into a 3-coordinated network in which each atom is at the apex of a pyramid formed by the bonds to the three atoms with which it coordinates. These groups are bound together in puckered 2-dimensional layers that are stacked, partly by van der Waals binding, to form the crystal. A random network, which can be a fully connected 3-dimensional one, can be formed from such a 3-coordinated system by distortions of the ideal A-A-A bond angle(about 98°for the arsenic structure). The As2S3 and As2Se3 amorphous structures might be regarded as having been generated by the insertion of S or Se atoms between each As-As closest pair of the amorphous As structure.

2.6. Optical Properties

Optical properties the scientific and technological interest in amorphous semiconductors stems mainly from their electrical and optical properties. Usually, the optical properties are easier to interpret. Optical absorption reflects essentially the density of states, more or less modified by the transition probabilities between the states. In crystals, it has been possible to obtain much information about the electronic and phonon state structures from the optical measurements, and similar methods have been applied to amorphous solids [42,43]. The measurements of optical absorption and particularly the absorption edge are important especially in connection with the theory of the electronic structure of amorphous materials. In the high absorption region, Tauc et al. [44] and Davis and Mott [45] independently derived an expression relating the absorption coefficient α(υ), to photon energy, h υ.

.

2.7. Electrical Properties

Investigations of the electrical properties of amorphous semiconductors like heavily on the procedures and analyses used in studies of crystalline semiconductors like Ge, Si, and the III-V compounds. However, exploration of amorphous semiconductors encounters many unusual aspects and difficulties (some of which are already apparent in work on low-mobility materials like oxides):

The preparation of well-defined samples for electrical measurements is a cumbersome and often impossible task. Besides foreign impurities and other traps, specimens contain uncontrolled compositional and positional inhomogeneities, which may give rise to charge accumulations and potential fluctuations.

The resistivity of an amorphous semiconductor or semi-insulator is often several orders of magnitude larger than that of its crystalline counterpart [47] .

Charge carrier concentrations are hard to obtain because of the difficulties in

Measuring and interpreting the Hall coefficient.

For the chalcogenide glasses, plots of log (conductivity) vs. reciprocal temperature often show a high-temperature segment which can be retraced in successive runs, and low temperature branches which depend on previous history. In crystalline semiconductors, one would call these two parts "intrinsic" and "extrinsic." However, in the case of amorphous semiconductors these terms are not well defined. Perhaps it would be better to refer to the high and low temperature portions of the conductivity curves as structure-insensitive and structure sensitive branches, respectively [48-50].

Two kinds of mechanisms have been proposed for the d.c. electrical conductivity of an amorphous semiconductor: a "free-carrier" contribution resulting from electrons or holes thermally excited into the high mobility extended states, and a "hopping" contribution resulting from phonon-assisted tunneling between localized

States of random energy, not too far from the Fermi level. The free carrier contribution has an activation energy which is the separation between the Fermi level and the nearest mobility edge. As noted earlier, this activation energy need not be equal to one-half the optical gap, as in the crystalline case, since the width of the mobility gap need not coincide precisely with the width

of the optical gap, and since the Fermi level need not lie precisely in the middle of the quasi gap if there is a finite density of states there. For the hopping

conductivity, Mott has proposed a form [51]:

lin σ α (α3/ ρ°k T)1/4

where ρ° is the density of localized states at the Fermi level, α-1 is the length for exponential decay of the states, and the variation of these quantities with energy is neglected. The unusual temperature dependence arises from the existence of a continuum of possible activation energies for hopping between two localized states of random distance from the Fermi energy.

3.DEVICE PHYSICS Scope of Applications

Amorphous semiconductors lend themselves to a wide array of possible uses, a few of these, which have particularly stimulated fundamental work in recent

years, will be singled out for detailed discussions. These will include: (a) The use of inorganic amorphous semiconductors in electrophotography, i.e., document

copying through the use of an electrostatic image of the material to be copied. This image is produced, after corona charging, through a selective discharge by photoconductivity, and the electrostatic image is used in turn to control the deposition of charged pigment particles. (b) The use of selective crystallization (or phase separation) in amorphous films, through incident light, in image handling methods, such as photography and electrophotography.

(c) The use of the same sort of selective crystallization (or phase separation) induced by a laser beam to write digital information. The laser beam can also produce local melting and quenching and thus can restore the amorphous material to its original form. The information thus written and/or erased, can be read out through the significant change in optical properties between the amorphous and crystalline state [52].

Applications.

A CD-RW (CD). Amorphous chalcogenide materials form the basis of re-writable CD and DVD solid-state memory technology [53].

The modern technological applications of chalcogenide glasses are wide spread. Examples include infrared detectors, mouldable infrared optics such as lenses, and infrared optical fibers, with the main advantage being that these materials transmit across a wide range of the infrared electromagnetic spectrum. The physical properties of chalcogenide glasses (high refractive index, low phonon energy, high nonlinearity) also make them ideal for incorporation into lasers, planar optics, photonic integrated circuits, and other active devices especially if doped with rare earth ions. Many chalcogenide glasses exhibit several non-linear optical effects such as photon-induced refraction [54,55], and electron-induced permittivity modification some chalcogenide materials experience thermally driven amorphous crystalline phase changes. This makes them useful for encoding binary information on thin films of chalcogenides and forms the basis of rewritable optical discs [5] and non-volatile memory devices such as PRAM. Examples of such phase change materials are GeSbTe and AgInSbTe. In optical discs, the phase change layer is usually sandwiched between dielectric layers of ZnS-SiO2, sometimes with a layer of a crystallization promoting film other less common such materials are InSe, SbSe, SbTe, InSbSe, InSbTe, GeSbSe, GeSbTeSe and AgInSbSeTe [56,57].

Electrical switching in chalcogenide semiconductors emerged in the 1960s, when the amorphous chalcogenide Te48As30Si12Ge10 was found to exhibit sharp, reversible transitions in electrical resistance above a threshold voltage. The switching mechanism would appear initiated by fast purely electronic processes. If current is allowed to persist in the non-crystalline material, it heats up and changes to crystalline form. This is equivalent to information being written on it. A crystalline region may be melted by exposure to a brief, intense pulse of heat. Subsequent rapid cooling then sends the melted region back through the glass transition. Conversely, a lower-intensity heat pulse of longer duration will crystallize an amorphous region .

Attempts to induce the glassy–crystal transformation of chalcogenides by electrical means form the basis of phase-change random-access memory (PC-RAM). This emerging technology is on the brink of commercial application by ECD Ovonics. For write operations, an electric current supplies the heat pulse. The read process is performed at sub-threshold voltages by utilizing the relatively large difference in electrical resistance between the glassy and crystalline states. Examples of such phase change materials are GeSbTe and AgInSbTe.

Although the electronic structural transitions relevant to both optical discs and PC-RAM were featured strongly, contributions from ions were not considered even though amorphous chalcogenides can have significant ionic conductivities. At Euromat 2005, however, it was shown that ionic transport can also be useful for data storage in a solid chalcogenide electrolyte. At the nanoscale, this electrolyte consists of crystalline metallic islands of silver selenide (Ag2Se) dispersed in an amorphous semiconducting matrix of germanium selenide (Ge2Se3) [58].

All of these technologies present exciting opportunities that are not restricted to memory, but include cognitive computing and reconfigurable logic circuits. It is too early to tell which technology will be selected for which application. But scientific interest alone should drive the continuing research. For example, the migration of dissolved ions is required in the electrolytic case, but could limit the performance of a phase-change device. Diffusion of both electrons and ions participate in electromigration—widely studied as a degradation mechanism of the electrical conductors used in modern integrated circuits. Thus, a unified approach to the study of chalcogenides, assessing the collective roles of atoms, ions and electrons, may prove essential for both device performance and reliability [59].

In Summary, Amorphous Semiconductors are used in many applications, including Solar Cells, Thin Film Displays, Electrophotography, Switching devices ,Thin-Film Transistor (TFT) LCD Displays ,

Photovoltaics where Photoelectric effect discovered by Edmund Bequerel in 1830. Albert Einstein received the Nobel Prize for describing the nature of light and the photoelectric effect in 1905 .

Bell Laboratories made the first photovoltaic module in 1954. The space industry in the 1960s and the energy crisis in the 1970s spurred further photovoltaic development.

Amorphous Silicon is used in solar cell’s fabrication and has the following properties:

Amorphous materials have no long-range crystalline order

In 1974, researchers found that photovoltaic devices could be made using amorphous silicon by properly controlling deposition and composition

Amorphous silicon absorbs solar radiation 40 times more efficiently than single-crystal silicon – a film 1-micron thick can absorb 90% of the usable solar energy.

Amorphous silicon can be processed at relatively low temperatures on low-cost substrates making it very economical.