The epitaxial growth of the polar GaP(100) on the nonpolar Si(100) substrate suffers from inevitable defects at the antiphase domain boundaries (APDs), resulting from mono-atomic steps on the Si(100) surface. Stabilization of Si(100) substrate surfaces with As is a promising technological step enabling the preparation of Si substrates with double atomic steps and reduced density of the APDs. In this paper, 4-50-nm-thick GaP epitaxial films were grown on As-terminated Si(100) substrates with different types of doping, miscuts, and As-surface termination by metalorganic vapor phase epitaxy (MOVPE). The GaP(As)/Si(100) heterostructures were investigated by X-ray photoelectron spectroscopy (XPS) combined with gas cluster ion beam (GCIB) sputtering and by hard X-ray photoelectron spectroscopy (HAXPES). We found residuals of As atoms in the GaP lattice (similar to 0.2-0.3 at.%) and a localization of As atoms at the GaP(As)/Si(100) interface (similar to 1 at.%). Deconvolution of core level peaks revealed interface core level shifts. In As core levels, chemical shifts between 0.5 and 0.8 eV were measured and identified by angle-resolved XPS measurements. Similar valence band offset (VBO) values of 0.6 eV were obtained, regardless of the doping type of Si substrate, Si substrate miscut or type of As-terminated Si substrate surface. The band alignment diagram of the GaP(As)/Si(1 0 0) heterostructure was deduced.

•Epitaxial GaP(0 0 1) layers were grown on single-domain and two-domain Si(0 0 1) surfaces by MOVPE.•Heterostructures were investigated by hard X-ray photoelectron spectroscopy (HAXPES).•Core level peak broadening is correlated with band bending in the heterostructure.•Inter-diffuse layer (IDL) structure model of single-domain GaP/Si(0 0 1) interface was suggested.•Non-monotonic band bending profiles were derived by a newly suggested parametrized polynomial function (PPF) approach.

GaP is a preferred candidate for the transition between Si and heterogeneous III-V epilayers as it is nearly latticematched to Si. Here, we scrutinize the atomic structure and electronic properties of GaP/Si(0 0 1) heterointerfaces utilizing hard X-ray photoelectron spectroscopy (HAXPES). GaP(0 0 1) epitaxial films with thicknesses between 4 and 50 nm are prepared by metalorganic vapor phase epitaxy on either predominantly single-domain (SD) or two-domain (TD) Si(0 0 1) surfaces. The antiphase domain content in the GaP films is in situ controlled, employing reflection anisotropy spectroscopy. Via the analysis of core level photoelectron intensities, we reveal core level shifts of the P 2p and Si 2p peaks near the interface as well as core level shifts in the Ga 3d peaks near the surface. We suggest an Inter-Diffused Layer (IDL) model of the GaP/Si(0 0 1) interfacial structure with Si-P bonds at the heterointerface and residual P atoms in the Si substrate. Using a newly developed Parametrized Polynomial Function (PPF) approach, we derive a non-monotonic band bending profile in the heterostructures, correct experimental valence band offsets implying interfacial electronic barriers, and determine valence band discontinuities of Delta EV = 1.1 +/- 0.2 eV (SD samples) and Delta EV = 0.8 +/- 0.2 eV (TD samples) at GaP/Si(0 0 1) interfaces.

The study of the chemical composition of buried interfaces by X-ray photoelectron spectroscopy (XPS) is limited by the inelastic mean free path of emitted photoelectrons (PE). Soft X-ray sources (AlKa) are generally suitable for careful probing of surfaces or very thin films. Here we applied gas cluster ion beam sputtering in combination with in-situ XPS (GCIB-XPS) to analyze buried GaP/Si(0 0 1) heterointerfaces. The GCIB method was used to dig a crater into the 20 nm thick GaP(0 0 1) film. We found optimal parameters for GCIB sputtering and achieved interface layers without severe damage. Destructive effects, i.e. broadening of core level peaks, could not be completely avoided, however, and the formation of metallic Ga on the GaP surface was observed. PE spectra of the sputtered heterostructures were compared with corresponding reference spectra of sputtered bulk crystals. Interface contributions to the intensity of phosphorus and sillicon core level peaks were revealed: interface components are shifted to high binding energies of P 2p Si 2p core levels. Similar results were obtained on 4 nm thick GaP/Si(0 0 1) by XPS. Finally, a top-to-bottom concept for buried semiconductor interfaces studies by GCIB-XPS is demonstrated.

We present a study of buried GaP/Si(001) heterointerfaces by hard X‐ray photoelectron spectroscopy. Well‐defined thin (4–50 nm) GaP films were grown on Si(001) substrates with 2° miscut orientations by metalorganic vapor phase epitaxy. Core level photoelectron intensities and valence band spectra were measured on heterostructures as well as on the corresponding reference (bulk) substrates. Detailed analysis of core level peaks revealed line broadening and energetic shifts. Valence band offsets were derived for the films with different thickness. Based on the observed variation of the valence band offsets with the GaP film thickness and on the experimental evidence of line broadening, the existence of charge displacement at the GaP/Si(001) interface is suggested.

The electron affinity of polycrystalline undoped and boron-doped diamond films was investigated by means of X-ray photoelectron spectroscopy and ultraviolet photoelectron spectroscopy. It was demonstrated that both as-grown and hydrogenated polycrystalline diamond films exhibited true negative electron affinity (NEA). Positive electron affinity (PEA) was observed after diamond surface oxidation in plasma. NEA reduction was observed after hydrogenation of the polycrystalline film surface in plasma. This effect was related to depolarization of the surface dipoles, an increase in the density of surface defects, and an increase in contaminations on the surface. Boron doping of the diamond film had a negligible effect on electron affinity. Downward surface band bending was found for all the samples that were studied, and there was larger band bending for polycrystalline diamond surfaces modified in plasma than for as-grown polycrystalline diamond surfaces.

The role of electron band bending and surface sensitivity in determining the core level binding energies by X-ray photoelectron spectroscopy is investigated. A dominating contribution of surface atomic layers to photoemission intensity is confirmed for normal photoemission. The energy of the photoelectron core level peak does not deviate from core level peak energies of electrons photoemitted from the surface atomic layers of the crystal. The higher surface sensitivity regime, achieved e.g. at off-normal photoelectron detection angle, can be used to study the surface potential barrier in just a few topmost atomic layers. In addition, it is demonstrated that core level binding energy measured by angle-resolved X-ray photoelectron spectroscopy reflect the electron attenuation anisotropy. In particular, core level binding energy changes with emission angle and correlates with the forward focusing directions in a crystal. This effect is demonstrated by measuring the polar angle dependence of Ga 3d core levels on clean GaN(0001) and GaN(000I) surfaces with a higher and a lower band bending, respectively. The effect is explained by variation of emission depth in a crystal for normal and off -normal photoelectron emission angles. (C) 2017 Elsevier B.V. All rights reserved.

Growth of GaN quantum dots (QDs) on polar and semipolar GaN substrates is a promising technology for efficient nitride-based light emitting diodes (LED). The QDs crystal orientation typically repeats the polarity of the substrate. In case of non-polar or semipolar substrates, the polarity of QDs is not obvious. In this article, the polarity of GaN QDs and of underlying layers was investigated nondestructively by X-ray photoelectron diffraction (XPD). Polar and semipolar GaN/Al0.5Ga0.5N heterostructures were grown on the sapphire substrates with (0001) and (1 (2) over bar 00) orientations by molecular beam epitaxy (MBE). Polar angle dependence of N 1s core-level photoelectron intensities were measured from GaN QDs and compared with the corresponding experimental curves from free-standing GaN crystals. It is confirmed experimentally, that the crystalline orientation of polar (0001) GaN QDs follows the orientation of the (0001) sapphire substrate. In case of semipolar GaN QDs grown on (1 (1) over bar 00) sapphire substrate, the (11 (2) over bar2) polarity of QDs was determined. (C) 2016 Elsevier B.V. All rights reserved.

The magnitudes of the surface band bending have been determined by X-ray photoelectron spectroscopy for polar, semipolar, and non-polar surfaces of wurtzite GaN crystals. All surfaces have been prepared from crystalline GaN samples grown by the hydride-vapour phase epitaxy and separated from sapphire substrates. The Ga 3d core level peak shifts have been used for band bending determination. Small band bending magnitudes and also relatively small difference between the band bendings of the surfaces with opposite polarity have been found. These results point to the presence of electron surface states of different amounts and types on surfaces of different polarity and confirm the important role of the electron surface states in compensation of the bound surface polarity charges in wurtzite GaN crystals.

We investigate GaN nanowire ensembles spontaneously formed in plasma-assisted molecular beam epitaxy by non-destructive low-energy electron diffraction (LEED) and x-ray photoelectron diffraction (XPD). We show that GaN nanowire ensembles prepared on AlN-buffered 6H-SiC(0001¯) substrates with well-defined N polarity exhibit similar LEED intensity-voltage curves and angular distribution of photo-emitted electrons as N-polar free-standing GaN layers. Therefore, as in the case of GaN layers, LEED and XPD are found to be suitable techniques to assess the polarity of GaN nanowire ensembles on a macroscopic scale. The analysis of GaN nanowire ensembles prepared on bare Si(111) allows us to conclude that, on this non-polar substrate, the majority of nanowires is also N-polar.

A fast and nondestructive method for polarity determination of wurtzite GaN crystals based on x-ray photoelectron diffraction (XPD) has been demonstrated. Photoelectron emission from N 1s core level excited by Mg K_α source was found sufficient for the polarity determination of GaN crystals. XPD polar plots from polar GaN {0001} and semipolar GaN $\left\{{10\bar 11} \right\},\;\left\{{20\bar 21} \right\},\;\left\{{11\bar 22} \right\}$ { 10 1 ¯ 1 } , { 20 2 ¯ 1 } , { 11 2 ¯ 2 } crystals have been analyzed. Due to dominant electron forward scattering along N–Ga directions, photoelectron intensities either increase or decrease within a relatively narrow emission polar angle range. The slopes of polar plots are found noticeably different in the polar angle range of 20°–25° for (0001) or $\left({000\bar 1} \right)$ ( 000 1 ¯ ) crystals, respectively. The semipolar GaN substrates can be divided into two groups, depending on whether m -plane or a -plane is perpendicular to the semipolar surface. It was found that the slopes of the polar plots are different in the angular range of 20°–27° for semipolar GaN $\left\{{10\bar 11} \right\}$ { 10 1 ¯ 1 } , 10°–22° for GaN $\left\{{20\bar 21} \right\}$ { 20 2 ¯ 1 } substrates, while for the GaN $\left\{{11\bar 22} \right\}$ { 11 2 ¯ 2 } semipolar planes, the slopes are different in the range of 0°–15° with respect to the surface normal.

fast and nondestructive method for polarity determination of wurtzite GaN crystals based on x-ray photoelectron diffraction (XPD) has been demonstrated. Photoelectron emission from N 1s core level excited by Mg K α source was found sufficient for the polarity determination of GaN crystals. XPD polar plots from polar GaN 0001 and semipolar GaN {101̅1}, {202̅1}, {112̅2} crystals have been analyzed. Due to dominant electron forward scattering along N–Ga directions, photoelectron intensities either increase or decrease within a relatively narrow emission polar angle range. The slopes of polar plots are found noticeably different in the polar angle range of 20°–25° for (0001) or (0001̅) crystals, respectively. The semipolar GaN substrates can be divided into two groups, depending on whether m -plane or a -plane is perpendicular to the semipolar surface. It was found that the slopes of the polar plots are different in the angular range of 20°–27° for semipolar GaN {101̅1} , 10°–22° for GaN {202̅1} substrates, while for the GaN {112̅2} semipolar planes, the slopes are different in the range of 0°–15° with respect to the surface normal.

A fast and nondestructive method for polarity determination of wurtzite GaN crystals based on x-ray photoelectron diffraction (XPD) has been demonstrated. Photoelectron emission from N 1s core level excited by Mg K-alpha source was found sufficient for the polarity determination of GaN crystals. XPD polar plots from polar GaN {0001} and semipolar GaN {10 (1) over bar1}, {20 (2) over bar1}, {11 (2) over bar2} crystals have been analyzed. Due to dominant electron forward scattering along N-Ga directions, photoelectron intensities either increase or decrease within a relatively narrow emission polar angle range. The slopes of polar plots are found noticeably different in the polar angle range of 20 degrees-25 degrees for (0001) or (000 (1) over bar) crystals, respectively. The semipolar GaN substrates can be divided into two groups, depending on whether m-plane or a-plane is perpendicular to the semipolar surface. It was found that the slopes of the polar plots are different in the angular range of 20 degrees-27 degrees for semipolar GaN {10 (1) over bar1}, 10 degrees-22 degrees for GaN {20 (2) over bar1} substrates, while for the GaN {11 (2) over bar2} semipolar planes, the slopes are different in the range of 0 degrees-15 degrees with respect to the surface normal.

A fast and nondestructive method for polarity determination of wurtzite GaN crystals based on x-ray photoelectron diffraction (XPD) has been demonstrated. Photoelectron emission from N 1s core level excited by Mg Kα source was found sufficient for the polarity determination of GaN crystals. XPD polar plots from polar GaN {0001} and semipolar GaN $\left\{{10\bar 11} \right\},\;\left\{{20\bar 21} \right\},\;\left\{{11\bar 22} \right\}$ crystals have been analyzed. Due to dominant electron forward scattering along N–Ga directions, photoelectron intensities either increase or decrease within a relatively narrow emission polar angle range. The slopes of polar plots are found noticeably different in the polar angle range of 20°–25° for (0001) or $\left({000\bar 1} \right)$ crystals, respectively. The semipolar GaN substrates can be divided into two groups, depending on whether m-plane or a-plane is perpendicular to the semipolar surface. It was found that the slopes of the polar plots are different in the angular range of 20°–27° for semipolar GaN $\left\{{10\bar 11} \right\}$ , 10°–22° for GaN $\left\{{20\bar 21} \right\}$ substrates, while for the GaN $\left\{{11\bar 22} \right\}$ semipolar planes, the slopes are different in the range of 0°–15° with respect to the surface normal.

Photoelectron diffraction is proposed to determine nondestructively the polarity of wurtzite crystals with polar surfaces (c-plane). Small segments of polar plots of photoemission from anion core levels in the (101¯0) azimuthal plane are qualitatively different for two polarities around the polar angle 20°. The magnitude of the ratio of electron photoemission intensities at two polar angles I20/I23 can be utilized as a simple criterion determining the crystal polarity.

Polarity of semipolar GaN(101¯1) (101¯1¯) and GaN(202¯1) (202¯1¯) surfaces was determined with X-ray photoelectron diffraction (XPD) using a standard MgKα source. The photoelectron emission from N 1s core level measured in the a-plane of the crystals shows significant differences for the two crystal orientations within the polar angle range of 80–100° from the 〈0001〉 normal. It was demonstrated that XPD polar plots recorded in the a-plane are similar for each polarity of the GaN{101¯1} and GaN{202¯1} crystals if referred to 〈0001〉 crystal axes. For polarity determinations of all important GaN{h0h¯l} semipolar surfaces, the above given polar angle range is suitable.

Polarity of semipolar GaN(10 (1) over bar1)(10 (1) over bar(1) over bar) and GaN(20 (2) over bar1)(20 (2) over bar(2) over bar) surfaces was determined with X-ray photoelectron diffraction (XPD) using a standard MgK alpha source. The photoelectron emission from N 1s core level measured in the a-plane of the crystals shows significant differences for the two crystal orientations within the polar angle range of 80-100 degrees from the < 0001 > normal. It was demonstrated that XPD polar plots recorded in the a-plane are similar for each polarity of the GaN{10 (1) over bar1} and GaN{20 (2) over bar1} crystals if referred to < 0001 > crystal axes. For polarity determinations of all important GaN{h0 (h) over barl} semipolar surfaces, the above given polar angle range is suitable. (C) 2014 AIP Publishing LLC.

Surface structure of the free-standing GaN substrates with polar (000-1), non-polar (1-100), (11-20), and semipolar (20-21) surface plane were investigated. Clean polar and non-polar GaN surfaces were prepared by annealing under NH3 atmosphere. (1x1) diffraction patterns were observed by low-energy electron diffraction (LEED) for both polar and non-polar GaN surfaces. The polar GaN surface was found well-ordered, while the non-polar GaN surfaces were found less ordered with atomic steps on the surface. Polar angle dependences of the photoelecton diffraction (PED) intensities exited by MgK alpha radiation from N 1s level were analyzed for all the GaN surfaces, aiming to determine the polarities of the GaN surfaces with polar and semipolar crystal orientations.