We're sorry but this page doesn't work properly without JavaScript enabled. Please enable it to continue.
Feedback

Doping dependence of the surface phase stability of polar O-terminated (000) ZnO

00:00

Formale Metadaten

Titel
Doping dependence of the surface phase stability of polar O-terminated (000) ZnO
Serientitel
Anzahl der Teile
40
Autor
Lizenz
CC-Namensnennung 3.0 Unported:
Sie dürfen das Werk bzw. den Inhalt zu jedem legalen Zweck nutzen, verändern und in unveränderter oder veränderter Form vervielfältigen, verbreiten und öffentlich zugänglich machen, sofern Sie den Namen des Autors/Rechteinhabers in der von ihm festgelegten Weise nennen.
Identifikatoren
Herausgeber
Erscheinungsjahr
Sprache

Inhaltliche Metadaten

Fachgebiet
Genre
Abstract
The dependence of hydrogen coverage on the bulk doping concentration is investigated for the polar O-terminated (000) ZnO surface. We use hybrid density-functional theory in combination with ab initio thermodynamics to determine a doping-dependent phase diagram of this surface. For hydrogen coverages lower than 50% dangling oxygen bonds remain at the surface, where they subsequently become charged by bulk electrons. For such charged surfaces, a computational first-principles approach is presented, with which long-range band bending can now be included in first-principles supercell calculations. In this work, we use a modified and extended version of the recently introduced charge-reservoir electrostatic sheet technique (Sinai et al 2015 Phys. Rev. B 91 075311) to incorporate band bending effects directly into our first-principles calculations. This allows us to investigate the effect of space charge layers and the resulting band bending on the hydrogen coverage of the ZnO (000) surface. After introducing a new implementation of CREST, we show that the structure and stability of polar ZnO surfaces are indeed sensitive to the amount of free charge carriers in the bulk. At low doping concentrations our results corroborate the previously reported (2 × 1) hydrogen phase, at higher doping concentrations the hydrogen coverage diminishes notably.
WeltzeitSchwingungsphaseNyquist-KriteriumNiederspannungsnetzGleitlagerVideotechnikAM-Herculis-SternBlatt <Papier>SchreibzeugWarmumformenSubstrat <Mikroelektronik>DotierungHalbleiterKopfstützeGasdichteChannelingSpiegelobjektivUrkilogrammErsatzteilSauerstoff-16CrestfaktorComputeranimationBesprechung/Interview
NivelliergerätFermionEnergielückeElektrischer LeiterValenzbandSupraleitungStrukturelle FehlordnungWasserstoffatomDotierungThermalisierungElektronEnergielückeSauerstoff-16KompressibilitätAngeregter ZustandDrahtbondenMaterialColourRuderbootSubstrat <Mikroelektronik>KühlkörperKalenderjahrCrestfaktorComputeranimation
FermionNivelliergerätElektrischer LeiterEnergielückeKlangeffektSupraleitungLiegeradSubstrat <Mikroelektronik>RaumladungsgebietDiagramm
EnergielückeFermionElektrischer LeiterRaumladungRaumladungsgebietErsatzteilHandyKopfstützeBlechElektrostatische AufladungCrestfaktorKlangeffektMaßstab <Messtechnik>GleichstromSchreibzeugRegentropfenSteinmetzDiagramm
FermionElektrischer LeiterEnergielückeFeinstblechLadungstrennungColourKlangeffektEnergielückePaarerzeugungBlechElektrostatische AufladungKonzentrator <Nachrichtentechnik>Diagramm
Elektrischer LeiterEnergielückeLadungstrennungFeinstblechCrestfaktorKlangeffektKonzentrator <Nachrichtentechnik>Initiator <Steuerungstechnik>KombinationskraftwerkSchwingungsphaseWasserstoffatomHauptsatz der Thermodynamik 2KugelstrahlenDiagramm
KugelstrahlenKonzentrator <Nachrichtentechnik>Sauerstoff-16WasserstoffatomAtomistikSpannungsabhängigkeitSpezifisches GewichtKristallgitterStoff <Textilien>Hauptsatz der Thermodynamik 2Computeranimation
OberflächenspannungWasserstoffatomStoff <Textilien>SpannungsabhängigkeitWasserstoffatomKugelstrahlenSpezifisches GewichtLeistungssteuerungKristallgitterLuftstromStoff <Textilien>SpannungsabhängigkeitKonzentrator <Nachrichtentechnik>ElektronDotierungComputeranimationDiagramm
WasserstoffatomSpannungsabhängigkeitStoff <Textilien>Konzentrator <Nachrichtentechnik>KugelstrahlenBuntheitWasserstoffatomComputeranimationDiagramm
CrestfaktorSchreibzeugHalbleiterFernordnungWerkzeugWarmumformenBlatt <Papier>EnergielückeFACTS-AnlageSatz <Drucktechnik>Isostatisches HeißpressenBesprechung/Interview
Transkript: Englisch(automatisch erzeugt)
Hi, I'm Simon Erker from Graz University of Technology. I'm here to introduce our recent paper published in New Journal of Physics. The main question we address in our work is how long-range bandbanding can be included in a density functional theory calculation of a doped semiconductor substrate.
In our paper we present a new implementation of the recently published Crest method. We describe and explain our approach in detail on the example of the polar oxygen-terminated zinc oxide surface.
To demonstrate the capability of Crest, we calculate that to our knowledge first doping-dependent surface phase diagram of the oxygen-terminated zinc oxide surface. The oxygen-terminated zinc oxide surface has been extensively studied theoretically over the years. In thermodynamic equilibrium and for typical experimental conditions, most studies predict a surface where every second
row of surface oxygen atoms is decorated with hydrogen atoms, resulting in a 50% hydrogen overlayer. This overlayer completely saturates the dangling bonds and therefore stabilizes this polar surface. But what happens now when we remove one of the hydrogen atoms?
This leaves behind a dangling bond at the surface that causes a surface state within the bandgap of the zinc oxide. I want to note that in terms of explaining Crest, this surface state could also be caused by a different surface defect or a molecular absorber as well.
If we have three charge carriers available in the substrate, which might be from intentional or unintentional doping of the zinc oxide, the dangling bond may be saturated by electrons from the doped bulk material. This leads to a charging of the surface and an upward bandbanding of the conduction and the valence band.
This well-known bandbanding effect is a macroscopic effect and the corresponding space charge region might reach far into the substrate. Because of these large length scales involved, a direct description of space charge layers using first principle DFD calculations are in general computationally not feasible.
Our new Crest method makes it possible to treat systems like this within a regular DFD calculation. The idea of Crest is to divide such a system in a surface part that is described by exact quantum mechanics and the rest of the space charge region that can be described by classical electrostatics.
This classical part needs to be designed in a way that it reproduces the electric field arising from the ionized donors in the space charge region and at the same time we need to introduce the correct amount of mobile charge carriers into our system. In our Crest implementation, this is achieved by introducing a charged sheet that mimics the electrostatic effect of bandbanding.
The properties of this sheet can be calculated by classical electrostatic equations. With this we can determine the energy contribution of bandbanding effects for any doping concentration and in combination with up-initio thermodynamics are able to compute doping-dependent surface phase diagrams.
We do this for the oxygen-terminate zinc oxide surface by continuously reducing the hydrogen coverage and are able to quantify the hydrogen desorption for a specific doping concentration and environmental conditions. We use up-initio thermodynamics to determine the lowest energy structure for a specific hydrogen chemical potential.
As one can see from this graph, which is calculated for a doping concentration of 10 over the power of 20 electrons by cubic centimeters, that by reducing the hydrogen chemical potential, surfaces with lower hydrogen coverages are stabilized.
Graphs like this can be calculated for various doping concentrations and from that a two-dimensional phase diagram with the doping concentration as a variable can be created. On the y-axis on this graph we now plotted the doping concentration as the variable. The different colors in this plot represent the various considered hydrogen coverages.
What we find is that at low doping concentrations we find no significant deviation from the 50% hydrogen overlay. But once we go to higher doping concentrations, the hydrogen coverage at the surface is reduced. With this I would like to finish my short overview where I introduce the
CREST method as a new important tool to treat band bending at semiconductor surfaces. In order to make our work more accessible, we added a simple step-by-step manual for our CREST implementation for the FHI Ames code. Thank you for listening and I hope you enjoy our paper.