Scanning Probe Microscopy of InAs/InP Nanowires


Bachelor Thesis, 2011

58 Pages, Grade: 1,0


Excerpt

Contents

Abstract

1 Introduction

2 Theory
2.1 Scanning Tunneling Microscopy
2.2 Scanning Tunneling Spectroscopy
2.3 Nanowires
2.4 Quantum dots

3 Experimental background
3.1 Setup
3.2 Sample preparation
3.3 Tip production
3.4 Vacuum conditions

4 Results and Discussion
4.1 Cleaning
4.1.1 Surface
4.1.2 Nanowires
4.2 Imaging
4.2.1 InAs(111)B surface
4.2.2 InAs nanowires grown without gold particle
4.2.3 InP with InAs quantum well nanowires
4.2.4 InP with InAs quantum dots nanowires
4.3 Spectroscopy
4.3.1 InAs(111)B surface
4.3.2 Defect induced influence on spectra

5 Conclusion and Outlook

Bibliography

Acknowledgements

Abstract

In this thesis the InAs(111)B surface and III-V semiconductor nanowires are inves- tigated using scanning tunneling microscopy and spectroscopy. The morphology of InAs nanowires grown without gold particle is studied. Radial nanowire heterostruc- ture such as InP core with InAs shell are analyzed and the wurtzite [Abbildung in dieser Leseprobe nicht enthalten] top facet is identified. Furthermore nanowire heterostructures with an InP core and InAs shell with induced stacking faults possibly giving rise to quantum dots, which could be used as quantum dot lasers or for quantum information processing, are investigated. A mod- el is obtained based on morphology analysis and as top facet the wurtzite [Abbildung in dieser Leseprobe nicht enthalten] are found. Furthermore stacking faults on top of a nanowire are seen.

The analysis of the InAs(111)B surface shows the hexagonal pattern. Defects are determined to occur due to missing In atoms in the first layer. Spectroscopy next to those defects indicated no influence on the local electronic structure.

Sammanfattning

Ett av de huvudsakliga forskningsområdena vid Lunds Universitet är att undersöka och utveckla halvledande nanotrådar. Nanotrådar är endimensionella trådliknande strukturer vars tjocklek motsvarar endast ett fåtal nanometer och vars längder motsvarar endast ett fåtal mikrometer. Dessa nanotrådar, som tros vara framtidens byg- gstenar inom elektronik, fotonik, biokemiska och kemiska sensorer, anses vara det sista steget i den tekniska kapp-löpning där varje framsteg gjort elektroniska komponenter mindre och mindre.

Ett område där man hoppas på speciellt stora framsteg med hjälp av nanotrådar är framställandet av snabbare och mer avancerade datorer samt effektivare solceller. Detta betyder att forskning kring nanotrådar kan spela en avgörande roll för vår framtida förnyelsebara energiproduktion.

Speciella egenskaper kan uppnås genom att kombinera olika material vid tillverknin- gen av nanotrådarna. Sådana nanotrådar, gjorda av flera material, kallas heterostruk- turer. Vissa typer av sådana strukturer kan ge upphov till så kallade kvantprickar, of- ta kallade artificiella atomer då de har diskreta energinivåer. Detta innebär att man i teorin skräddarsyr avståndet i energi mellan dessa nivåer, vilket kan vara till avsevärd nytta för att kunna framställa fotoelektroniska apparater som lasrar och lysdioder med specifika våglängder, d.v.s. färger. En annan tillämpning kan vara att hantera krypterad information.

En hel del forskning har ägnats åt nanotrådar, men heterostrukturer med kvant- prickar på är ett nytt och relativt outforskat område. För att studera dessa måste man först ha en grundläggande förståelse för hur trådarnas ytstruktur och morfologi ser ut, d.v.s. hur atomerna är orienterade relativt varandra. Då detta kräver upplösning tillräcklig för att studera enskilda atomer är det inte helt trivialt att göra sådana undersökningar och väldigt få instrument är kapabla att göra dem även under goda förhållanden. Ett instrument som dock är väl anpassat för ett sådant ändamål är sveptunnelmikroskopet (STM). Det uppfanns 1982 av Binning och Röhrer, som fick nobelpriset gemensamt i fysik för denna bedrift fyra år senare, 1986.

Principen för mikroskopet bygger på att man mäter ändringar i den ström som, på grund av den kvantmekaniska tunneleffekten, går mellan mikroskop och den yta som undersöks. Ändringar i strömstyrkan kan direkt kopplas till hur den topografiska strukturen på ytan ser ut, och med hjälp av en dator kan en bild av ytan återskapas så att enskilda atomer framträder. Detta i sin tur kan ge detaljerad information om hur tillväxtprocessen av nanotrådarna sker. Ytterligare information kan fås med spek- troskopiska metoder genom att i väldefinierade intervall ändra spänningen mellan prov och mikroskop, vilket kan avslöja ytans speciella elektronsiska egenskaper. Såväl STM som relaterad spektroskopi beskrivs i detalj i denna kandidatuppsats. Även en experimentell modell för nanotrådar med kvantprickar läggs fram, som förklarar dess egenskaper samtöppnar för vidare studier. Slutligen undersöks även defekter i den atomära strukturen hos olika ytor med förslag till förklaringar.

Zusammenfassung

In dieser Arbeit werden die InAs(111)B Oberfläche und Nanodrähte aus III-V Halb- leitern mittels Rastertunnelmikroskopie und -spektroskopie untersucht. Die Morphologie von InAs Nanodrähten, die ohne Goldpartikel wuchsen wird be- schrieben. Nanodrähte mit radialen Heterostrukturen, wie InP Kern mit InAs Schale, werden analysiert und die Wurtzit { 10 1 0 } als obere Seitenfläche identifiziert. Weitere Heterostrukturen wie InP Kern und InAs Schale mit induzierten Stapelfehlern wer- den untersucht. Diese Stapelfehler könnten eine Ursache f ür die Bildung von Quanten- punkten sein, die in Quantenpunktlaser oder Quanteninformationsverarbeitung Ver- wendung finden können. Die Analyse der Morphologie f ührt zu einem Model dieser Nanodrähte, f ür deren obere Seitenfläche sowohl die Wurtzit { 10 1 0 } als auch { 11 2 0 } gefunden wurden. Dar über hinaus wurden Stapelfehler in der oberen Seitenfläche ge- funden.

Die hexagonale Struktur der InAs(111)B wurde beobachtet. Auftretende Gitterfehler konnten auf drei fehlende In Atome in der obersten Schicht zur ück gef ührt werden. Spektroskopie in der Umgebung dieser Gitterfehler zeigte keinen Einfluss auf die lokale Zustandsdichte.

1 Introduction

One of the main areas of research at Lund University is growth and characterization of III-V semiconductor nanowires. These are one dimensional semiconductor struc- tures made of group III, e. g. Gallium (Ga) or Indium (In), and group V materials, e. g. Arsenic (As) or Phosphorus (P). Nanowires have typically a diameter of a few nanometer and a length up to some micrometer. They are expected to be new building blocks in electronics, photonics, biochemical and chemical sensors by several studies. [12, 16] Scanning tunneling microscopy and spectroscopy contribute to the research by obtaining information about the surface and electrical properties.

Special properties can be obtained by heterostructures which are obtained by chang- ing materials during growth. [18, 21] Combining materials with different band gaps the degree of freedom for electrons can be confined possibly resulting in the creation of a quantum dot.

Nanowires and quantum dots can be seen as the perfection of miniaturization. The physics of the quantum dots made them to be key elements of future optoelectronic devices like quantum dot lasers, light emitting diodes and quantum information processing. [27, 29, 26]. Surface and electrical properties of those special nano structures have to be studied in order to realize such devices.

This experimental thesis investigates the surface structure of the InAs substrate and several nanowires including quantum well and quantum dot nanowires via scanning tunneling microscopy. Additionally, spectroscopy studies are performed on the impor- tant InAs(111)B surface acting as a substrate for all nanowires analyzed during this work. The main part of this thesis deals with radial heterostructures. First InP core nanowires with an InAs shell forming a quantum well are investigated. Then a further study is performed on InP core nanowires with induced stacking faults and an InAs shell exhibiting quantum dots.

2 Theory

2.1 Scanning Tunneling Microscopy

Scanning tunneling microscopy (STM) allows to get information about topographic and electronic properties of the conductive sample surface with very high - up to the atomic scale - spatial resolution. [27] In figure 2.1 the principle of a scanning tunneling micro- scope is drawn. Basic elements are the extremely sharp tip, in ideal case consisting of only one atom at its apex, and the piezoelectric elements which gently move the tip. [25]

Abbildung in dieser Leseprobe nicht enthalten

Figure 2.1: Schematic drawing of the Scanning tunneling microscope. In constant current mode, the distance d between tip and sample is kept constant via feedback loop.

The tip approaches the sample surface without touching it. On the sample a bias volt- age is applied resulting in a tunneling current I, which is proportional to the tunneling probability τ between the sample and the tip. Depending on the sign of the applied bias the electrons tunnel from the sample into the tip or the other way around (figure 2.2).

The tunneling current depends exponentially on the tip-surface distance d and the effective barrier height φ, the tunneling voltage V and the local density of states near

Abbildung in dieser Leseprobe nicht enthalten

Figure 2.2: Dependence of the tunneling current on the applied sample bias. (a) With no ap- plied sample bias, the Fermi level of the tip and the sample are aligned and thus no tunneling current results. (b) Positive sample bias (V > 0) results in tunneling of the electrons from the tip into the sample. (c) Negative sample bias (V < 0) resulting in electrons tunnel from the sample into the tip. In all images, the shaded area represents occupied states. The black arrows symbolize tunneling of electrons, E F,T and E F,S are the Fermi-energy of the tip and sample respectively and e the elementary charge.

Abbildung in dieser Leseprobe nicht enthalten

The magnitude of the tunneling current is of the order of some picoamperes and thus has to be amplified. In order to avoid too much noise, the amplification is done as close to the junction as possible. [3, p.23]

A change of the tip position of 0,1 nm yields a change in the tunneling probability of an order of magnitude resulting in a high vertical resolution. [18, p.7] Using the feedback mode the z -piezo gets a signal of the tunneling current at every scan point and is adjusted so that the current is kept constant. The movement of the z - piezo as a function of x and y positions is registered and leads to a topographic image. The advantage of the constant current mode is that the distance between tip and surface is constant and thus collisions with huge steps or fragments are avoided. This is the most common mode to work with the STM. [27, 7]

In order to avoid disturbances due to outer influence, the system is isolated against vibrations. Furthermore the system is operated in an ultra high vacuum (UHV) chamber with a base pressure below 1 · 10 [10] mbar.

2.2 Scanning Tunneling Spectroscopy

The development of the STM in 1982, for which Binning and Röhrer received the Nobel Prize for Physics in 1986, led to enormous progress in the surface science. In the 1980’s STM imaging with atomic resolution seemed to solve the problem of surface structure determination. It turned out that STM images not always lead to the surface structure. [7, 1] The reasons are on the one hand the limited resolution and on the other hand the voltage-dependence of the images. It is known, that the actual shape of the tip greatly influences the appearance of the objects within the image (see section 3.3). As it already can be seen in figure 2.2 and equation 2.1 the tunneling current depends on the applied voltage and the local density of states. This is to say that the image also contains information about the electronic properties of the investigated sample which can be obtained by Scanning Tunneling Spectroscopy (STS).

Only electrons from occupied states can tunnel into unoccupied states. This means, according to figure 2.2, with the applied voltage V, those states being between E F and E F + eV can contribute to the tunneling current.

An easy way to look for bias dependent changes of the surface would be to record some images with different applied biases and compare them with each other. To align those taken images, surface defects can help. Another possibility is to change the bias during the scan process after every scanned line. This dual mode leads to two different images (respectively applied voltages) in the forward and backward scan.

Another interesting procedure would be to continuously change the applied volt- age and measure the resulting tunneling current leading to an I - V -curve. The results are information about the electrical properties of the surface around the Fermi level. This spectroscopy can be done in various ways. One possibility is to keep the dis- tance tip-sample fixed and modulate the applied voltage. Therefore the feedback loop is switched off and the topographic imaging process is interrupted for the duration of the spectroscopy. [7] The voltage sweeps through the set range and the corresponding current is recorded. The spectrum can either be recorded at special areas of interest or at every pixel of the recorded image called current imaging tunneling spectroscopy (CITS), whereas CITS needs very clean samples, a stable tip and takes long time to record.

A second possibility is to also vary the sample-tip distance during acquisition of the spectra. This has the advantage that the conductance at low voltages is amplified with- out increasing the noise level. Therefore the tip is moved in a given way while modu- lating the voltage. Using equation 2.1 those data can be transformed to fixed tip-sample distance. [8]

A typical I - V -spectrum for semiconductors can be seen in figure 2.3 where also the corresponding points for the band edges and the Fermi level are given. The distance between Fermi level and the conduction band edge (E C − E F) depends on the type and amount of doping and thus the curve would be shifted with respect to the Fermi level (zero applied voltage). It can be seen that the absolute value of tunneling current increases with higher absolute value of tunneling voltage. This is due to the fact that for higher applied voltages more states contribute to the tunneling.

Additionally one can measure the differential conductance d I /d V to directly get the local density of states for a certain energy [Abbildung in dieser Leseprobe nicht enthalten]. This can be done using the

Abbildung in dieser Leseprobe nicht enthalten

Figure 2.3: Typical spectrum for semiconductors. (a) A typical I - V -curve with marked Fermi level and band gap. (b) Band situation in the sample and tip. As before, the shaded area represents occupied states. The black arrows symbolize tunneling of electrons, E F,T and E F are the Fermi-energy of the tip and sample respectively, e the elementary charge and V the applied sample bias. E C is the conduction band edge, E V the valence band edge and E g the band gap.

lock-in technique where the actual tunneling voltage is superimposed by a small highfrequency modulation voltage. The frequency and amplitude of the modulation voltage have to be chosen not to effect the feedback loop. The variation of the tunneling current can then be recorded.

The analysis is done according to Feenstra [8]. Therefore the recorded differential conductance d I /d V is normalized to the total conductance I/V (for metals or low band gap materials) or to the broadened conductance I/V (for high band gap mate- rials). The latter one has to be done because the ratio (d I /d V)/(I/V) diverges at the band edges. The broadening is done by convoluting the total conductance with an ex- ponential function. [22] This treatment of the obtained spectra as well as the averaging of several spectra is done by a MATLAB program and the different steps can be seen in figure 2.4.

The big advantage of STS compared to other spectroscopy techniques is that the tun- neling current in STM flows through a region of around 5 Å in diameter and can thus provide spectroscopic information on an atom-by-atom basis. [3, p.60] Furthermore the influence of defects and adsorbates on the surface state bands can be identified. [3, p.67]

Surface effects

All atoms within a crystal are bond to neighboring atoms in a ordered fashion. An exception are the surface atoms experiencing dangling bonds. If the surface provides electronic states positioned within the band gap the Fermi level can be pinned and

Abbildung in dieser Leseprobe nicht enthalten

Figure 2.4: The analysis of the obtained spectra. (a) Recorded I - V -spectrum. (b) Recorded [Abbildung in dieser Leseprobe nicht enthalten] -spectrum. (c) Normalized (d [Abbildung in dieser Leseprobe nicht enthalten] -spectrum the observation of the bulk electronic properties becomes more difficult. The origin of Fermi level pinning could be surface defects like contamination or adatoms. Hence, in order to get good results the surface should be clean and flat. [18, 4]

The main interest on spectroscopic features is the relative position of the valence band and conduction band edges. The occurrence of tip-induced band bending (TIBB) in the semiconductor (illustrated in figure 2.5) gives rise to problems in the quantitative description of tunneling spectra. [10] Neglecting the band bending, the energy E of a

Abbildung in dieser Leseprobe nicht enthalten

Figure 2.5: Schematic diagram of the band situation. E C and E V are the conduction band edge and the valence band edge respectively. The sample Fermi level is denoted by E F and the tip Fermi level by [Abbildung in dieser Leseprobe nicht enthalten] where V is the applied voltage. (a) Tip-induced band bending at the surface is denoted by φ. Quantum effects are shown, like (b) wavefunction tailing through a depletion region and (c) formation of accumulation states. Modified from [9] state can be related to the applied voltage at which it appears by [Abbildung in dieser Leseprobe nicht enthalten], with E F the Fermi energy. Taking band bending into account this expression has to be modified for the surface states to [Abbildung in dieser Leseprobe nicht enthalten] is the band bending at the surface. As seen in figure 2.5, the situation becomes more complicated for bulk states since the wavefunction can tail through the depletion region, figure 2.5 (b). Furthermore the formation of localized states, figure 2.5 (c), has to be taken into account. [9]

2.3 Nanowires

Nanowires are one dimensional rod shaped nanostructures. They can be grown using different techniques for example chemical beam epitaxy (CBE), molecular beam epitaxy (MBE) or metalorganic vapor phase epitaxy (MOVPE) [18].

Semiconductor nanowires are commonly grown on a (111) semiconducter substrate since they preferably grow in this direction and thus standing perpendicular to the substrate. [13] The growth mechanism for nanowires is attributed to the vapor-liquid- solid (VLS) mechanism since they have alloy droplets on their tip. [28] Therefore, a substrate, a catalyst (often a gold particle) and a vapor phase growth material (in this case group III and V vapors) are used. The growth process is illustrated in figure 2.6 and can be divided into three stages, namely metal alloying, crystal nucleation and axial growth.

Abbildung in dieser Leseprobe nicht enthalten

Figure 2.6: The vapor-liquid-solid growth mechanism with the three stages metal alloying, crystal nucleation and axial growth. [28]

During the first stage, metal alloying, the gold catalyst forms an alloy with the vapor, liquefies due to the temperature (400-500 C), and is liquid during the entire growth. In the second stage, the nanowire nucleation starts followed by the growth stage, when further condensation of the vapor at the liquid/solid interface takes place. After cooling the alloy particle solidifies on top of the nanowire. The diameter of the nanowire is due to the alloying process bigger than the initial gold particle by a few nanometers. [28, 18] The growth mechanism is not yet fully understood. Some people say that InAs and InP gold assisted wires can only grow at temperatures where the particle is solid and thus would grow in the so-called vapor-solid-solid (VSS) mechanism. This still demands some work to characterize. [17]

A possibility to grow nanowires without a gold particle is to use a silicon oxide mask on the substrate. The nanowires then grow from an e.g. Indium particle for InAs or InP nanowires. [23]

The nanowires have usually a cubic zinc blende (Zb), a hexagonal wurtzite (Wz) or a mixture of both structures. In a III-V material a single bilayer is defined by a pair of one group III and one group V atom. A normal Zb sequence is ABCABC and a normal Wz sequence would be ABAB, where each letter stands for one bilayer. A misplacement of a bilayer in a Wz sequence (ABAB | CBCB), defined as a stacking fault, gives a single sequence of Zb (ABC). A single misplacement of a bilayer in a Zb sequence creates a twin plane (ABCACBA) and is not enough to produce a stacking fault. Therefore, at least two sequential twin planes are required [Abbildung in dieser Leseprobe nicht enthalten] in order to produce a Wz sequence. [5]

The most common side facets are [Abbildung in dieser Leseprobe nicht enthalten] and [Abbildung in dieser Leseprobe nicht enthalten] for Wz structures and [Abbildung in dieser Leseprobe nicht enthalten] for Zb structures [17, 21]. In figure 2.7 the common Wz side facets [Abbildung in dieser Leseprobe nicht enthalten] can be seen.

Abbildung in dieser Leseprobe nicht enthalten

Figure 2.7: Model of the most common Wz side facets. [Abbildung in dieser Leseprobe nicht enthalten] facet with zigzag rows of the top In or As elements. (b) [Abbildung in dieser Leseprobe nicht enthalten]facet, where the rows of top layer In respectively As atoms are perpendicular to the growth direction. (c) Cross section of a hexagonal crystal perpendicular to the [0001]-direction showing the [Abbildung in dieser Leseprobe nicht enthalten] type facets. From [6]

A distinction between those two side facets is possible due to the appearance of zigzag rows of the III and V elements along the growth direction at the [Abbildung in dieser Leseprobe nicht enthalten] facet and the rows of the III and V elements perpendicular to the growth direction at the [Abbildung in dieser Leseprobe nicht enthalten] facet. [6]

2.4 Quantum dots

An electron system, which is confined in all three dimensions is called quantum dot (QD). [19] As a result of this confinement, the energy levels of the charge carriers (for semiconductors both, the electrons and holes) become quantized. The densities of states are delta function-like and thus the QD form nearly ideal zero-dimensional systems with high quantum efficiency. [26] Semiconductor QD are therefore often described as artificial atoms, what is also reflected in the fact that both free atoms and QD exhibit optical line spectra with narrow linewidth. [29]

In order to get the needed confinement for semiconductor materials both, the valence band and the conduction band should be modulated. [24] The modulation is done by construction of heterostructures, two semiconductor materials of different sizes in band gap but not to big lattice mismatches. In figure 2.8 a, the band structure of an InAs QD on InP bulk is depicted schematically. InAs/InP build a ”type-I” band alignment, that is that both electrons and holes are confined in the nanostructure. [27, 11] Investiga- tions have been made on stacked InAs QDs in InP on a planar surface [20] but not on nanowires with QDs. The nanowires used in the experiments for this thesis consist of an InP core with induced stacking faults and an InAs shell as shown in the transmission electron microscopy image in figure 2.8 b.

Abbildung in dieser Leseprobe nicht enthalten

Figure 2.8: (a) Schematic band diagram of an InAs/InP quantum dot. The quantized energies are not drawn to scale for both, the electron and hole states. (b) A transmission electron microscopy image of the nanowires with InAs/InP quantum dots. The perfect wurtzite and the stacking fault area are indicated

Formation of quantum dots could be due to induced strain in the substrate leading to energy differences on the surface and thus reconstructions on the surface or a shortening in the migration length of diffusing surface atoms. [21, 29] The electrical properties of those QD are an interesting field for investigation and therefore STS is a good technique to obtain information about this small area.

3 Experimental background

3.1 Setup

The system used during these experiments is an Omicron XA VT STM, figure 3.1. It consists basically of two parts, the analysis and the preparation chamber.

Abbildung in dieser Leseprobe nicht enthalten

Figure 3.1: Scanning tunneling microscope used during experiments

Samples and tip are introduced via the load lock. After introducing a new sample and pumping down the load lock chamber, the sample is transferred to the preparation chamber using the transfer arm. Within the preparation chamber the sample is cleaned (see section 3.2) using the homebuilt hydrogen source II since the commercial hydrogen source I is temporarily out of order. For tip preparation (section 3.3) the argon source is used. Using the mass spectrometer it can be seen which gases (up to mass number 100) are present in the preparation chamber. The experiments itself take place in the analysis chamber. In order not to affect the atmosphere in the analysis chamber during preparation processes, both chambers are separated by a valve. Once a sample or a tip is prepared to be used for the experiment it is transferred to the analysis chamber, and placed to the destination using the wobble stick. Within the analysis chamber there is space to store twelve tips or samples in the carousel. The tip and the sample used for investigations are placed on a stage which is magnetic damped. Tip and sample can be positioned and approached with the help of the camera, the light source and a screen.

3.2 Sample preparation

Since the as-grown nanowires are perpendicular to the substrate and were exposed to air they cannot be used for STM imaging without preparation. First the as-grown nanowires have to be deposited on a new substrate plate which is done by a direct con- tact method. The new substrate plate is gently placed onto the facing up growth plate. Thereafter a slight pressure against the plates yields the nanowires to be deposited on the new substrate plate. Results of this depositing method can be seen in scanning elec- tron microscopy (SEM) images, e. g. figure 3.2 b. It has been shown that this method yields a high density of aligned nanowires without bundling and with a low amount of fragmentation [18]. Those new sample plates are then glued on a sampleholder using Indium. The sample undergone those steps (figure 3.2 a) can be introduced into the STM.

Abbildung in dieser Leseprobe nicht enthalten

Figure 3.2: (a) The nanowires are deposited onto a new substrate plate which is then glued onto a sample holder (not in this image) and can be inserted into the STM. Until now the preparation was done in the normal atmosphere and thus the sample is oxidized. (b) Results of the deposition method in this case for the quantum well sample described in section 4.2

After introducing the sample one has to get rid of the native oxides on the nanowires since they are insulating. Furthermore, if tunneling is possible on the oxidized surface the image would contain only information about the oxides structures. First the sample is outgassed at a temperature of around 300 C. With the mass spectrometer one can see what elements and molecules desorb from the surface.

After outgassing the sample has to be cleaned further to fully get rid of the oxides. A common way to clean the III-V surfaces is to use atomic hydrogen and annealing temperatures between 300 C and 500 C. Controlling the annealing temperatures, times and hydrogen pressure yields good results without big changes in electronic properties. [6, 2] The molecular hydrogen is introduced via the hydrogen source II and is thermally cracked into atomic hydrogen. The pressure during the cleaning processes for experiments described in this thesis was around [Abbildung in dieser Leseprobe nicht enthalten] mbar.

3.3 Tip production

The quality of recorded images depends crucially on the quality of the tip, especially for non-flat surfaces. For the investigation of nanowires and quantum dots on nanowires, the tip should therefore be atomically sharp in order not to get multiple features as depicted in figure 3.3. A good example for tip effects can also be seen in figure 4.7. A tip consisting of several minitips leads to a repetition of the recorded features or objects may occur broader than they are since the tip from which electrons tunnel might change during the scan.

To produce a sharp tip, a tungsten wire of 0,38 mm diameter is chemically etched, using a 8% sodium hydroxide (NaOH) solution as electrolyte while a voltage between the wire and a steel cathode is applied. After etching the tip is cleaned by rinsing with distilled water and isopropanol. The tungsten tips get contaminated during the etching process and a common way to remove those contaminations and to further sharpen the tip is sputtering with Argon ions [15]. Therefore Argon gas is introduced into the preparation chamber via the argon source at energies around 3 kV and gets ionized using 19 mA filament current. The duration of the sputtering process is two times 15 minutes at a pressure of about [Abbildung in dieser Leseprobe nicht enthalten] mbar.

[...]

Excerpt out of 58 pages

Details

Title
Scanning Probe Microscopy of InAs/InP Nanowires
College
LMU Munich
Grade
1,0
Author
Year
2011
Pages
58
Catalog Number
V178801
ISBN (eBook)
9783656009795
ISBN (Book)
9783656010494
File size
4208 KB
Language
English
Tags
scanning, probe, microscopy, inas/inp, nanowires
Quote paper
Stephan Pröller (Author), 2011, Scanning Probe Microscopy of InAs/InP Nanowires, Munich, GRIN Verlag, https://www.grin.com/document/178801

Comments

  • No comments yet.
Read the ebook
Title: Scanning Probe Microscopy of InAs/InP Nanowires



Upload papers

Your term paper / thesis:

- Publication as eBook and book
- High royalties for the sales
- Completely free - with ISBN
- It only takes five minutes
- Every paper finds readers

Publish now - it's free