16 Semiconducting and Oxide Nanowires

Dr. Anchal Srivastava

Contents of this Unit

1. Introduction

2. Semiconducting Nanowires

2.1. Silica Nanowires

2.2. Germanium Nanowires

2.3. III-V Nanowires

2.3.1. GaAs Nanowires

2.3.2. InAs Nanowires

2.3.3. InP Nanowires

2.3.4. GaP Nanowires

3. Oxide Nanowires

3.1. Synthesis Methodologies

4. Summary

Learning Outcomes

After studying this module, you shall be able to
Know about semiconducting nanowires. Know about oxide nanowires.
Learn about different types of nanowires their applications and growth.

1. Introduction

A nanowire is a nanostructure, with the diameter of the order of a nanometer ( 10-9 meters). It can be also defined as the ratio of the length to width being greater than 1000. Nanowire can also be defined as the structures that have thickness or diameter constrained to tens of nanometers or less and an unconstrained length. This kind of nanostructure is also known as quantum wire because at this scale quantum mechanical effects become important.

Nanowires show many interesting properties which are missing in 3-D (three-dimensional) bulk materials. The interesting phenomenon arises due to quantum confinement of electrons which leads to discrete energy levels that are different from the traditional continuum of energy levels or bands found in bulk materials. The quantum confinement exhibited by certain nanowires manifest themselves in discrete values of the electrical conductance. Such discrete values arise from a quantum mechanical restraint on the number of electrons that can travel through the wire at the nanometer scale. These discrete values are generally referred as the quantum conductance and are integer multiples of

They are inverse of the well-known resistance unit h/e2, which is roughly equal to 25812.8 ohms, and referred to as the von Klitzing constant RK.

Fig. 1. Figure showing various nanostructures: 3D, 2D, 1D, and 0D.

When the size or the dimension of a material is continuously reduced from a large or macroscopic size, such as a meter or a centimeter, to a very small size, the properties remain the same at first, then small changes begin to occur, until finally when the size drops below 100nm, dramatic changes in properties can occur. If one dimension is reduced to the nanorange while the other two dimensions remain large, then we obtain a structure known as a two dimensional (2D) material or quantum well. If two dimensions are reduced in nanometer range and one remains large, the resulting structure is referred as one dimensional (1D) material or quantum wire. If all the three dimensions are reduced in nanometer range is called zero dimensional (0D) material or quantum dot. The word quantum is associated with these three types of nanostructures because the changes in properties arise from the quantum-mechanical nature of physics in the domain of the ultrasmall. The process of diminishing the size has been illustrated in Fig. 1. Quantum size effect in nanowires will have an impact on electronics and photonics applications. Quantum confinement occurs when the nanomaterial dimensions approach the size of an exciton in bulk crystal, called the exciton Bohr radius. This leads to an increase in band gap with a decrease in size of the nanomaterial. Keeping in the view of importance of 1D nanomaterials in this module we are interested to study one dimensional nanostructured materials (1D NSMs) or quantum wire

2. Semiconducting Nanowires:

Semiconducting nanowires are of great interest due to their potential in electronics, optoelectronics, and sensor applications. Silicon nanowires (SiNWs) have received the most attention due to the historical role of silicon in the integrated circuits (IC) industry. Continued high performance from silicon may require alternatives to thin film based planar CMOS transistors in the form of novel devices, three-dimensional structures, innovative architectures, and integration with other functional components. SiNWs may provide new avenues in these directions. Silicon becomes a direct band gap semiconductor at nanoscale dimensions and thus may have interesting optoelectronics applications. The interest in optical interconnects beyond the copper era in IC manufacturing and the desire to have all silicon-based process (instead of using III–V materials for the optical interconnect part) may benefit from mature nanowire technology if and when it becomes available. In many of the applications, Ge provides a competitive avenue and thus received serious consideration in the nanowire community. Nanowires of III–V semiconductors such as GaAS, AlxGa1−xAs, InP, InAs, and wide band gap III–V nitrides have also been reported. a wide variety of techniques including the VLS approach, laser ablation, template-guided synthesis, oxide-assisted growth, and others have been used to grow semiconducting nanowires.

2.1. Silicon Nanowires:

Silicon nanowires, also referred as SiNWs, are a type of nanowire most often formed from a silicon precursor by etching of a solid or through catalyzed growth from a vapour or liquid phase. Initial synthesis is accompanied by thermal oxidation steps to yield structures of accurately tailored size and morphology. SiNWs shows many unique properties that are missing in bulk (3D) silicon materials. These properties arise from an unusual quasi 1-D electronic structure and are the subject of research across numerous disciplines and applications. The reason that SiNWs are considered one of the most important 1D materials is that they could have a function as building blocks for nanoscale electronics assembled without the need for complex and costly fabrication facilities. SiNWs are frequently studied towards applications including photovoltaics, nanowire batteries, thermoelectrics and non-volatile memory. A large number of techniques exist to fabricate silicon nanowires. These can be classified into bottom-up and top-down fabrication techniques. In top-down fabrication, lithography is used to define the fabricated structure that is then transferred from the photo-resist to the substrate by etching or a similar way of structuring the already available material. In the bottom-up approach, the material is added to the substrate in a self-organized way.

2.2. Germanium Nanowires:

Germanium is an important semiconductor with a direct band gap of 0.8 eV and an indirect band gap of 0.66 eV. It has been receiving much attention recently either as a pure element or as an alloy with silicon (SixGe1−x) for logic and other devices. Germanium offers higher intrinsic carrier mobilities than silicon:

at room temperature, enabling faster switching and higher frequency devices. Other advantages include higher intrinsic carrier concentration, 2.4 × 1013cm−3 versus 1.45 × 1010 cm−3; larger bulk excitonic Bohr radius, 24.3 nm versus 4.7 nm for more prominent quantum confnement effects even at larger material dimensions; and compatibility with high-k dielectrics enabling integration to current semiconductor processing technology. Germanium is also compatible with III–V materials, for example, good lattice matching with GaAs. Germanium oxide exhibits interesting optical properties suitable for interesting integrated optoelectronic circuits. All these features have prompted extensive investigations of germanium nanowire (GeNW) growth, characterization, and application development.

2.3. III-V Nanowires:

III–V compound semiconductor materials are of great interest in electronics and optoelectronics for applications in high-power or high-frequency transistors, light emitting diodes, lasers, resonant tunneling diodes, solar cells, photodetectors, and others. Their usefulness in the above applications has been well established in the last two decades through thin flms and quantum wells of GaAs, AlxGa1-xAs, InxGa1-xAs, InP, GaP etc grown using techniques such as metal organic chemical vapor deposition (MOCVD) and molecular beam epitaxy (MBE). Recently, these materials have also been grown as onedimensional nanowires since quantum confined low-dimensional structures have the potential to provide superior performance in all of the above applications.

2.3.1. GaAs Nanowires:

Gallium arsenide is a direct band gap semiconductor with a band gap of 1.424 eV, electron saturation velocity of 107cm/s, and carrier mobility of 8500 and 450 cm2/V s for electrons and holes, respectively. It has been grown in the nanowire form by laser-assisted catalytic growth, MBE, MOCVD and template-guided approach.

2.3.2 InAs Nanowires:

InAs exhibits a small energy band gap (0.35 eV), high electron mobility (33000 cm2/V s), and a large quantum confnement energy. There have been several reports on the preparation of InAs nanowires. InAs growth by MOCVD with a thin layer of SiO serving as catalyst on InP(111), InAs(001), and Si(001) substrates. Fig. 2 shows SEM images of InAs nanowires on InP substrates at temperatures 540°C–660°C.

Fig. 2. InAs nanowires grown on InP(111)B by MOCVD at various temperatures. Growth conditions: pressure = 10 kPa, TMIn mole fraction = 2 × 10−6, AsH3 mole fraction = 2 × 10−4, H2 carrier gas glow = 6 L/min..

2.3.3. InP Nanowires:

Indium phosphide is a direct band gap semiconductor with a gap of 1.34 eV that has been widely investigated and used for electronics and optoelectronics applications. It has been grown in the form of nanowires using laser-assisted growth, sublimation of InP powder, MBE, and MOCVD. Fig. 3 shows results of InP nanowires synthesized using MOCVD technique, which provides vertical, uniform diameter nanowires.

Fig. 3. InP nanowires grown by MOCVD using 20 nm gold colloids. TBP/TMIn ratio = 120 and 430°C growth temperature..

2.3.4. GaP Nanowires:

Gallium phosphide is a wide band gap semiconductor with an indirect gap of 2.26 eV. The electron mobility of bulk GaP is 160 cm2/V.s. GaP is preferred for high-temperature electronics applications and light emission in the visible range. In laser catalytic growth of GaP, a GaP solid target is laser ablated to provide the source vapor and the use of gold colloids appears to provide a tight diameter distribution of grown nanowires. In general, the mean nanowire diameter is only 1–2 nm larger than that of the colloids. Sublimation of GaP powder has also been used to generate the vapor source.

Fig. 4. Schematic of laser ablation grown well-defined GaP nanowires with monodisperse gold colloids as catalysts.

Laser ablation is a laser-assisted method and the key feature of this technique is that the catalyst used to define nanowire VLS growth can be selected from phase diagram data and/or knowledge of chemical reactivity. Using GaP as the target, GaP nanowires on well-defined gold colloids coated substrates by the laser ablation growth was grown. Fig. 4 is schematic of laser ablation growth of well-defined GaP nanowires using monodisperse gold colloids as catalysts.

3. Oxide Nanowires:

Metal oxides are an important class of materials that find applications in a wide variety of fields. For example, ZnO is a direct, wide band gap semiconductor with applications in ultraviolet (UV) and near UV photonics. Many metal oxides find applications either as catalyst supports or active catalyst materials and conducting channels in gas sensors. Several metal oxides also find direct applications in electrochemical energy devices such as lithium ion batteries, electrode materials, and solar cells.

Almost all of the above applications can benefit from the availability of metal oxides in the form of nanowires due to various reasons: well-defined crystallinity and surface, reactivity, electronic properties, and high surface area. In some cases, metal oxide nanowires with diameters less than few nanometers are sought for understanding the quantum confinement effects. Even though, nanowires with radius larger than the corresponding Bohr radius may not show distinct properties compared to bulk, they are still promising due to the following reasons: fast charge transport, low thresholds for percolation, high surface area, and well-defined crystal surfaces. The above-mentioned applications are diverse and require metal oxide nanowires in different formats ranging from epitaxial nanowire arrays, vertical nanowire arrays on a variety of conducting substrates, networked thin films, highly dense thick flms, and powders. For epitaxial nanowire arrays, there should be a minimal lattice mismatch between the nucleating metal oxide crystal and the underlying single-crystal substrates. For vertical nanowire arrays, high nucleation densities are typically necessary. A more detailed discussion will be provided under different categories of the metal oxide nanowire growth schemes.

3.1. Synthesis Methodologies:

In the case of oxide nanowires, there are four important vapor-phase methodologies used for their synthesis:

(a) Foreign metal catalyst assisted

(b) Direct oxidation of large, molten metal droplets

(d) Thermal/ plasma oxidation of metal foils.

(a) Catalyst Assisted:

The CVD of metal oxides when performed on substrates coated with a thin film or layer of catalyst particles can lead to metal oxide nanowire growth with catalyst particles at their tips. However, the metal containing gas or liquid precursors are not readily available for deposition of many metal oxides. So, it is easier and inexpensive to start with either solid metal or metal oxide powder sources. The catalyst is typically deposited onto substrates as thin films (<100 nm thickness) either using sputtering or evaporation.

Fig. 5. A schematic showing a typical horizontal quartz tube reactor used for the catalyst-assisted synthesis of nanowires.

In some cases, dispersions of catalyst particles can be synthesized or purchased separately and spin-coated onto the substrates. Many of the catalyst-assisted synthesis techniques for metal oxide nanowires use thermal CVD in a quartz tube reactor as shown in the schematic in Fig. 5 This setup includes a horizontal tube heated with a two-zone furnace with one zone for evaporation of solid metal sources and another zone for growth on substrates coated with the catalyst layer.

(b) Direct Oxidation Schemes Using Low-Melting Metals:

The synthesis procedures involving direct oxidation of metals are gaining importance compared to catalyst-assisted schemes mainly because of two reasons: (1) possible contamination with foreign catalyst metals in the catalyst-assisted schemes and (2) the costs associated with the respective catalyst metals and corresponding processes.

The solubilities of oxygen and the respective oxides in the low-melting point metal melts are expected to be negligible. So, any amount of dissolution of oxygen species during direct oxidation of large, micron-scale molten metal droplets leads to spontaneous nucleation of nanometer scale, metal oxide nuclei from molten metal. Further growth of metal oxide nuclei via basal attachment leads them to one-dimensional structures. High density of nucleation followed by growth via basal attachment (from dissolved oxygen) from large, molten metal clusters results in flowery morphology in which a high density of nanowires emanates from one point. After nucleation stage, further growth of metal oxide nuclei can also occur laterally in addition to basal growth. Such growth behavior leads to polycrystalline crust formation on large molten metal clusters. It has been found that the presence of reducing species such as hydrogen or chlorine reduces the lateral propagation of metal oxide nuclei. This is schematically illustrated in Fig. 6.

Fig. 6. A schematic illustrating the effect of hydrogen or other reducing gas-phase precursors during direct oxidation schemes for bulk nucleation and growth of oxide nanowires from their respective molten metal droplets using Ga as an example.

(c) Chemical Vapor Transport or Deposition of High-Melting Metal Oxides:

In the case of high-melting point metals (for example, tungsten, tantalum, and molybdenum), their melting points are always much higher than the synthesis temperatures used in any procedure. For such metals, the respective oxide vapors can easily be produced through oxidation of the corresponding metal at temperatures much lower than their melting points of metals. The transport of the respective metal oxide vapors onto the substrates can be used to create metal oxide films similar to any CVD scheme. However, it has been shown that under particular circumstances with low oxygen partial pressure, such chemical vapor transport procedures, resulted in the synthesis of respective oxide nanowires without the aid of any external catalyst. Such a scheme can be appropriately described as “vapor–solid” since the respective metal is not molten at the synthesis temperature.

Fig. 7. A schematic of the hot-wire CVD reactor setup used for chemical vapor transport experiments using a variety of high-melting point metals as filaments.

The hot wire (or filament) CVD experiments were conducted using oxygen flow over tungsten filaments for the synthesis of tungsten oxide nanowires. As shown in Fig. 7, the reactor consists of a 2 in. diameter quartz tube with the metal f laments wound on top of two ceramic tubes, which are connected to an electrical feed through. The filament is resistively heated by flowing electric current and the temperature is controlled by controlling the applied power to the filament using a variac power supply. The quartz tube is placed inside a furnace oven. In addition to the oven, the radiation from the filaments can heat the substrates to high temperatures depending upon their distance from the source. The substrates can be easily heated up to temperatures around 550°C in 2 in. quartz tube without the use of external heating via furnace. Experiments using tungsten filaments and oxygen flow at low partial pressures in the range of 0.05 to 0.3 Torr, filament temperature around 1600°C and substrate temperatures around 400°C to 800°C resulted in the synthesis of tungsten oxide nanowires.

(d) Plasma and Thermal Oxidation of Foils:

Another concept that is interesting for synthesizing metal oxide nanowire arrays involves direct oxidation of metal foils. Here, the metal foils are oxidized either thermally or using highly dissociated oxygen plasmas. The underlying nucleation and growth modes in these cases are different from that of “tip-led” growth in catalyst-assisted and self-catalyzed schemes but are somewhat similar to that of bulk nucleation and growth from low-melting metal melts. The main difference is however that the nanowire growth occurs here with direct oxidation of solid metal foils and at temperatures lower than the melting temperatures of the respective metals. Plasma oxidation of iron foils using highly disassociated oxygen plasma was created using an inductively coupled radiofrequency (RF) plasma. These oxidation experiments were conducted in a vacuum chamber equipped with an inductively coupled RF generator with a maximum power of 5 KW, reactor pressure of 2 Pa, and experimental duration of about 2 min. The resulting nanowire arrays are shown in Fig. 8a and b.

Fig. 8. SEM images of iron oxide nanowires synthesized through plasma oxidation of an iron foil: (a) side view of an iron oxide nanowire array on top of the iron foil substrate; and (b) inset shows SEM image of a single tapered iron oxide nanowire.

4. Summary:

In this module, you study

Various kinds of one dimensional, semiconducting and oxide nanowires and their synthesis from different techniques.

Suggested Reading

(i) Text book of “Advances in III-V Semiconductor Nanowires and Naodevices” by Jianye Li, Deli Wang and

Ray R. LaPierre.

(ii) Text book of “Introduction to Nanotechnology” by Charles P. Poole and Frank J. Owens.

(iii) Yiying Wu et. al. “Germanium Nanowire Growth via Simple Vapor Transport, Chem. Mater. 2000, 12, 605-

607.

(iv) Text book of “Inorganic nanowires, applications, properties and characterization” by M. Meyyappan and

Mahendra K. Sunkara. CRC Press.

you can view video on Semiconducting and Oxide Nanowires