Reflection High-energy Electron Diffraction Data Research Articles

Rotation Error Detection of Gallium Nitride (GaN) Substrate in MBE Utilizing Ensemble Learning

A rotating substrate is essential for achieving high-quality and uniform growth of gallium nitride (GaN) thin films in molecular beam epitaxy (MBE). By rotating the substrate during growth, the deposited material can be distributed more uniformly across the surface, resulting in a more uniform and defect-free film. However, the substrate may stop rotating during fabrication, and it is a challenge for engineers to detect the rotation of the substrate using information from reflection high-energy electron diffraction (RHEED). The current method for analyzing rotation using RHEED data is manual, time-consuming, and error-prone. Therefore, this research proposes an automated machine learning-based rotation error detection (RED) model to determine wafer rotation and then alert the engineers in case of an anomaly. The proposed approach consists of (1) ensemble learning models such as random forest classifier and extreme gradient boosting (XGBoost) for rotation error detection, (2) prepared novel intensity data from RHEED videos, (3) comparison to select the best ensemble learning model at a small scale, (4) training XGBoost at a large scale, (5) deployment in production using MLflow, (6) development of a REST application programming interface (API) for inference, and (7) integrated KakaoTalk API to generate and send a detailed alert message to engineers so that they can take appropriate actions in time. The experimental results achieved F1 scores for normal and rotational errors of 98.9 and 83.1%, respectively. The proposed method can make a valuable contribution to epitaxial wafer growth by enhancing production efficiency, improving product quality, and reducing costs.

Distinct thin film growth characteristics determined through comparative dimension reduction techniques

Reflection high energy electron diffraction (RHEED) information is critical for the growth of thin films; however, only a small percentage of the data from RHEED videos is typically used. The use of full videos in machine learning can require dimension reduction techniques. In this paper, three dimension reduction techniques, principal component analysis (PCA), non-negative matrix factorization (NMF), and kmeans clustering, are compared to investigate their benefits to the analysis of RHEED data. Three different heterostructures with different growth modes, all deposited on Ti-terminated strontium titanate by pulsed laser deposition, were used for the analysis: lanthanum aluminate with layer-by-layer growth, lithium cobalt oxide with island growth, and strontium ruthenate with a transition from layer-by-layer to step-flow growth. A phase shift in intensity fluctuations of different RHEED spots was discovered and discussed in terms of their sensitivity to the film growth characterization. The diffraction spots that were more sensitive to the growth were differentiated from the spots that are affected by the substrate as a function of film thickness. It was concluded that NMF provides the analysis that is easiest to interpret without the loss of detailed physical information due to its non-negativity constraint and lack of forced orthogonality such as in PCA. Analysis of the full RHEED videos enables a more detailed understanding of growth characteristics and control of growth processes as aided by dimension reduction.

Machine learning analysis of perovskite oxides grown by molecular beam epitaxy

Reflection high-energy electron diffraction (RHEED) is a ubiquitous in situ molecular beam epitaxial (MBE) characterization tool. Although RHEED can be a powerful means for crystal surface structure determination, it is often used as a static qualitative surface characterization method at discrete intervals during a growth. A full analysis of RHEED data collected during the entirety of MBE growths is made possible using principle component analysis (PCA) and k-means clustering to examine significant boundaries that occur in the temporal clusters grouped from RHEED data and identify statistically significant patterns. This process is applied to data from homoepitaxial SrTiO$_{3}$ growths, heteroepitaxial SrTiO$_{3}$ grown on scandate substrates, BaSnO$_{3}$ films grown on SrTiO$_{3}$ substrates, and LaNiO$_{3}$ films grown on LaAlO$_{3}$ substrates. This analysis may provide additional insights into the surface evolution and transitions in growth modes at precise times and depths during growth, and that video archival of an entire RHEED image sequence may be able to provide more insight and control over growth processes and film quality.

On the Dynamic Annealing of Ion-Induced Radiation Damage in Diamond under Irradiation at Elevated Temperatures

The structure and properties of the near surface layer of diamond after high-fluence irradiation at elevated temperatures with 30-keV Ar+, Ne+, N+, $${\text{N}}_{{\text{2}}}^{ + }$$ and C+ ions are studied. Dynamic annealing at temperatures above 500°С results in recrystallization of the diamond phase only in the case of irradiation with carbon ions, and irradiation with impurity ions causes graphitization of the ion-modified layer of diamond. According to Raman spectroscopy and reflection high-energy electron diffraction data, a nanographite structure is formed in the modified layer when irradiated with noble-gas ions. When irradiated with carbon ions, diamond grows and recrystallizes with a thin (~1 nm) graphite-like layer on the surface. In the case of nitrogen ions, a graphite layer is observed up to 720°С. Graphite-like layers on the surface of diamond are thermostable and, according to X-ray photoelectron spectroscopy, are layers of sp2 carbon. The Raman spectroscopy of visible light shows the different optical transparency of ion-induced structures on the surface of diamond, which correlates with the resistivities. The diamond structure disordered during ion irradiation at room temperature and the nanographite structure are more transparent than the graphite structure.

Growth Kinetics and Microstructure of PbTe Films Produced on Si and BaF2 Substrates by a Modified Hot-Wall Method

Lead telluride films have been grown on Si (100) and BaF2 (100) substrates by a modified hot-wall method using a graphite reaction chamber. According to X-ray diffraction, X-ray microanalysis, and scanning electron microscopy characterization results, the average growth rate of PbTe films having compositions within the homogeneity range of lead telluride increases with increasing lead vapor partial pressure and decreases with increasing tellurium vapor partial pressure, independent of the nature of the substrate. The rate of PbTe film growth has been shown to be maximal in the initial stage of the process and decrease monotonically over time, independent of the nature of the substrate. Independent of the growth time, the average growth rate of the PbTe films on the Si (100) substrates is considerably higher than that on the BaF2 (100) substrates. Reflection high-energy electron diffraction data indicate that the texture of the PbTe films on Si (100) corresponds to the substrate orientation and that the misorientation angle of the mosaic blocks does not exceed 20°. On the BaF2 (100) substrates, we observe epitaxial PbTe film growth with the orientation relationship (100), [011] PbTe ║ (100), [011] BaF2.

Manganese 3×3 and3×3-R30∘structures and structural phase transition onw-GaN(0001¯) studied by scanning tunneling microscopy and first-principles theory

Manganese deposited on the N-polar face of wurtzite gallium nitride [GaN (000$\overline{1}$)] results in two unique surface reconstructions, depending on the deposition temperature. At lower temperature (less than 105${\phantom{\rule{0.16em}{0ex}}}^{\ensuremath{\circ}}$C), it is found that a metastable $3\ifmmode\times\else\texttimes\fi{}3$ structure forms. Mild annealing of this Mn $3\ifmmode\times\else\texttimes\fi{}3$ structure leads to an irreversible phase transition to a different, much more stable $\sqrt{3}\ifmmode\times\else\texttimes\fi{}\sqrt{3}\ensuremath{-}R{30}^{\ensuremath{\circ}}$ structure which can withstand high-temperature annealing. Scanning tunneling microscopy (STM) and reflection high-energy electron diffraction data are compared with results from first-principles theoretical calculations. Theory finds a lowest-energy model for the $3\ifmmode\times\else\texttimes\fi{}3$ structure consisting of Mn trimers bonded to the Ga adlayer atoms but not with N atoms. The lowest-energy model for the more stable $\sqrt{3}\ifmmode\times\else\texttimes\fi{}\sqrt{3}\ensuremath{-}R{30}^{\ensuremath{\circ}}$ structure involves Mn atoms substituting for Ga within the Ga adlayer and thus bonding with N atoms. Tersoff-Hamman simulations of the resulting lowest-energy structural models are found to be in very good agreement with the experimental STM images.

Initial stage growth of GexSi1-x layers and Ge quantum dot formation on GexSi1-x surface by MBE.

Critical thicknesses of two-dimensional to three-dimensional growth in GexSi1−x layers were measured as a function of composition for different growth temperatures. In addition to the (2 × 1) superstructure for a Ge film grown on Si(100), the GexSi1−x layers are characterized by the formation of (2 × n) reconstruction. We measured n for all layers of Ge/GexSi1−x/Ge heterosystem using our software with respect to the video recording of reflection high-energy electron diffraction (RHEED) pattern during growth. The n reaches a minimum value of about 8 for clear Ge layer, whereas for GexSi1−x films, n is increased from 8 to 14. The presence of a thin strained film of the GexSi1−x caused not only the changes in critical thicknesses of the transitions, but also affected the properties of the germanium nanocluster array for the top Ge layer. Based on the RHEED data, the hut-like island form, which has not been previously observed by us between the hut and dome islands, has been detected. Data on the growth of Ge/GexSi1−x/Ge heterostructures with the uniform array of islands in the second layer of the Ge film have been received.

RHEED monitoring of elastic stresses during MBE growth of group III nitride heterostructures

Results of statistical analysis of the patterns of reflection high-energy electron diffraction (RHEED) measured during molecular beam epitaxy (MBE) of heterostructures of wide-bandgap (Al, Ga, In)N compounds are presented, which can be used for determining the lateral lattice constant (a) of the epitaxial layers. It is established that the error of determination can vary from a minimum of 0.17% up to several percent, depending on the contrast of RHEED patterns, which is determined by the stoichiometry of MBE growth. It is demonstrated that the RHEED data can reveal relaxation of elastic stresses during the growth of short-period {GaN(4 nm)/AlN(6nm)}30/AlN superlattices.

Structure, morphology, and magnetic properties of Fe nanoparticles deposited onto single-crystalline surfaces.

Background: Magnetic nanostructures and nanoparticles often show novel magnetic phenomena not known from the respective bulk materials. In the past, several methods to prepare such structures have been developed – ranging from wet chemistry-based to physical-based methods such as self-organization or cluster growth. The preparation method has a significant influence on the resulting properties of the generated nanostructures. Taking chemical approaches, this influence may arise from the chemical environment, reaction kinetics and the preparation route. Taking physical approaches, the thermodynamics and the kinetics of the growth mode or – when depositing preformed clusters/nanoparticles on a surface – the landing kinetics and subsequent relaxation processes have a strong impact and thus need to be considered when attempting to control magnetic and structural properties of supported clusters or nanoparticles.Results: In this contribution we focus on mass-filtered Fe nanoparticles in a size range from 4 nm to 10 nm that are generated in a cluster source and subsequently deposited onto two single crystalline substrates: fcc Ni(111)/W(110) and bcc W(110). We use a combined approach of X-ray magnetic circular dichroism (XMCD), reflection high energy electron diffraction (RHEED) and scanning tunneling microscopy (STM) to shed light on the complex and size-dependent relation between magnetic properties, crystallographic structure, orientation and morphology. In particular XMCD reveals that Fe particles on Ni(111)/W(110) have a significantly lower (higher) magnetic spin (orbital) moment compared to bulk iron. The reduced spin moments are attributed to the random particle orientation being confirmed by RHEED together with a competition of magnetic exchange energy at the interface and magnetic anisotropy energy in the particles. The RHEED data also show that the Fe particles on W(110) – despite of the large lattice mismatch between iron and tungsten – are not strained. Thus, strain is most likely not the origin of the enhanced orbital moments as supposed before. Moreover, RHEED uncovers the existence of a spontaneous process for epitaxial alignment of particles below a critical size of about 4 nm. STM basically confirms the shape conservation of the larger particles but shows first indications for an unexpected reshaping occurring at the onset of self-alignment.Conclusion: The magnetic and structural properties of nanoparticles are strongly affected by the deposition kinetics even when soft landing conditions are provided. The orientation of the deposited particles and thus their interface with the substrate strongly depend on the particle size with consequences regarding particularly the magnetic behavior. Spontaneous and epitaxial self-alignment can occur below a certain critical size. This may enable the obtainment of samples with controlled, uniform interfaces and crystallographic orientations even in a random deposition process. However, such a reorientation process might be accompanied by a complex reshaping of the particles.

Growth mechanism, microstructure and transport properties of Sr1−x La x CuO2 (x=0.10−0.15) thin films

Sr1−xLaxCuO2 (x=0.10−0.15) thin films with an infinite-layer type structure were grown on BaTiO3 buffered (001) SrTiO3 substrates by pulsed laser deposition (PLD). The evolution of the growth front was monitored, in-situ, by high-pressure reflection high-energy electron diffraction (RHEED), while the surface morphology was analyzed by means of atomic force microscopy (AFM), ex-situ. X-ray diffraction (XRD) was used to determine the evolution of the film structure with deposition and cooling parameters, as well as to study the type and level of epitaxial strain in the Sr1−xLaxCuO2 films. The RHEED data showed that the Sr1−xLaxCuO2 films grow on BaTiO3/SrTiO3 following a 2D or Stranski-Krastanov mechanism, depending on the La doping level. The transition point (critical thickness dc) from layer-by-layer like (2D) to island (3D) growth depends on the film stoichiometry: decreasing the La doping concentration x from 0.15 to 0.10, the critical thickness dc increases from ∼45 nm to ∼75 nm. In order to induce superconductivity, the Sr1−xLaxCuO2 films were cooled down under reduction conditions. The as-deposited films showed semiconducting or metallic behavior, the resistivity decreasing with increasing La concentration. Post-deposition vacuum annealing resulted in a superconducting transition onset (but no zero resistance down to 4.2 K) only for some of the x=0.15 Sr1−xLaxCuO2 films.

Surface segregation of Sb atoms during molecular-beam epitaxy of InSb quantum dots in an InAs(Sb) matrix

Surface segregation of Sb atoms during molecular-beam epitaxy (MBE) of the InSb/InAs quantum dot nanostructures has been investigated using the reflection high-energy electron diffraction (RHEED). It has been shown that an increase of the average specular spot intensity (SSI) during growth of a cap InAs layer immediately after the surface exposure to the Sb flux is due to the presence of residual population of Sb atoms at the growth surface. The Sb surface segregation is responsible for formation of a wetting layer in the InSb QD structures obtained without intentionally deposition of InSb, when the Stranski–Krastanow growth mode is impossible. Temperature dependences of the length of Sb segregation and the Sb segregation coefficient are deduced from the RHEED SSI. The RHEED data are in a good agreement with both X-ray diffraction and potoluminescence data which confirm the presence of the InAsSb wetting layer surrounding InSb QDs.

Long-range disorder effects on the GaAs(0 0 1) β2(2 × 4) surface

We first summarise previous work on long-range disorder effects occurring on the GaAs(0 0 1) β2(2 × 4) surface, as determined by reflection high energy electron diffraction (RHEED), grazing incidence X-ray diffraction and scanning tunnelling microscopy (STM). Models which have been deduced are related to the assembly of β2(2 × 4) and c(2 × 8) unit cells and involve the presence of antiphase domains and one-dimensional disorder boundaries. We ignore kinks produced by cooperative shifts of whole (2 × 4) units in the [1 1 0] direction, since they are only present in significant numbers on surfaces formed in the presence of a large excess of arsenic. We show from a combination of multi-azimuthal RHEED data and structure factor calculations that none of the existing models is consistent with all of the diffraction information. In particular, STM images show that the surface has a non-ideal c(2 × 8) structure made up of a distribution of β2(2 × 4) cells having equal but irregular occupancy of the two possible phase positions. The equal occupancy means that certain reflections should be forbidden, but in practice some of these appear strongly. We have suggested possible structural effects which might produce these results, such as a change in atom positions in the boundary region where two cells conjoin, so that they are no longer identical entities. We have been unable to obtain confirmation, however, since the precise atom locations are not known.

Origin of surface metallization of Si(001) at high temperatures

Surface metallization of Si(001) has been observed at high temperatures, but its explanation is controversial. Such a temperature-induced metallization can be explained in terms of a symmetric-dimer model where symmetric dimers are alternately displaced up and down along the dimer rows. We find that this symmetric-dimer model has an electronlike Fermi surface around the $\ensuremath{\Gamma}$ point and a hole pocket centered at the ${K}_{1∕2}$ point, in good agreement with angle-resolved photoemission spectroscopy data. Since the $p(2\ifmmode\times\else\texttimes\fi{}2)$ symmetric-dimer structure is more favored over the conventional $p(2\ifmmode\times\else\texttimes\fi{}1)$ symmetric-dimer structure, its presence may be confirmed by a more detailed analysis of recent reflection high-energy electron-diffraction data which showed the phase transition from a buckled- to a symmetric-dimer structure at high temperatures.

Structure and growth of titanium buffer layers on Al2O3(0 0 0 1)

The structure of titanium films on α-Al2O3(0001) surfaces at room temperature was investigated through in situ reflection high energy electron diffraction (RHEED). The α-phase of titanium was observed to grow with the Ti(0001)∥Al2O3(0001), Ti[11̄00]∥Al2O3[21̄1̄0] and Ti[101̄0]∥Al2O3[11̄00] epitaxy. For up to 6 nm thick films, an other structure was found to co-exist with α-Ti. Its presence has dramatic consequences for the wetting of silver, which partly explains the non-trivial buffer effect of titanium at the silver/alumina interface. From the RHEED data, the extra structure is assigned to the high-pressure hexagonal ω-Ti phase. This is supported by tight-binding total energy calculations that demonstrate that the ω phase could actually be stabilized by the α-Al2O3(0001) substrate.

Structure and growth of titanium buffer layers on Al2O3(0001)

The structure of titanium films on α-Al2O3(0 0 0 1) surfaces at room temperature was investigated through in situ reflection high energy electron diffraction (RHEED). The α-phase of titanium was observed to grow with the Ti(0 0 0 1)∥Al2O3(0 0 0 1), Ti[1 1̄ 0 0]∥Al2O3[2 1̄ 1̄ 0] and Ti[1 0 1̄ 0]∥Al2O3[1 1̄ 0 0] epitaxy. For up to 6 nm thick films, an other structure was found to co-exist with α-Ti. Its presence has dramatic consequences for the wetting of silver, which partly explains the non-trivial buffer effect of titanium at the silver/alumina interface. From the RHEED data, the extra structure is assigned to the high-pressure hexagonal ω-Ti phase. This is supported by tight-binding total energy calculations that demonstrate that the ω phase could actually be stabilized by the α-Al2O3(0 0 0 1) substrate.

Indium surface segregation in AlSb and GaSb

The effect of indium surface segregation during the molecular beam epitaxial growth of AlSb and GaSb on In(As,Sb) is investigated using various techniques including reflection high-energy electron diffraction (RHEED), high-resolution transmission electron microscopy (HRTEM), and high-resolution X-ray diffraction (HRXRD). The peculiar RHEED intensity variation observed during the growth of AlSb and GaSb on In(As,Sb) is interpreted as the consequence of indium surface segregation. This is confirmed by HRTEM investigation. The interface width associated with In segregation is found to increase with the growth temperature, as expected for a kinetically limited surface segregation phenomenon. In the investigated growth temperature range (400–540°C), the In segregation length is always higher for GaSb than for AlSb. Finally, a very good agreement between simulated and experimental HRXRD data corresponding to a GaSb/InAs superlattice is obtained by taking into account the In segregation profile deduced from the RHEED data.

Modeling electron mobility in MBE-grown InAs/AlSb thin films for HEMT applications using neural networks

This paper presents experiments that examined the formation of the InAs/AlSb interface in high electron mobility transistor (HEMT) devices grown by molecular beam epitaxy (MBE). At the interface, indium barrier thickness and substrate temperature were varied. Reflection-high-energy-electron-diffraction (RHEED) intensity oscillations of the specular spot were recorded during formation of the interface. Electron mobility models based on growth conditions and RHEED data were developed using neural networks. Our results demonstrate a correlation of the electron mobility with growth conditions and RHEED data.

Infrared and Ultraviolet Raman Spectra of AlN Thin Films Grown on Si(111)

AbstractInfrared (IR) spectroscopy and ultraviolet near-resonance enhanced Raman scattering have been used as alternative techniques to characterize the AlN films grown on Si(111). Epitaxial AlN films have been prepared by plasma source molecular beam epitaxy (PSMBE) at growth temperatures of 400 and 650 oC. The AlN/Si grown at 650°C shows distinct reflection high-energy electron diffraction (RHEED) patterns with the following epitaxial relations: AlN [0001] ‖ Si [111], and AlN < 011 0 > ‖ Si < 112 >. This is in sharp contrast with the observation of a spotty RHEED pattern for AlN/Si sample grown at 400°C, which appeared the same when viewed along any Si azimuth. Furthermore, the X-ray diffraction (XRD) rocking curve width showed a significant reduction from 5.7ofor sample grown at 400°C to 2° for the one grown at 650°C. The observation of Raman active A1(LO) and E2, and IR active E1(TO) zone center phonons, in both the films, is in accordance with the c-axis orientation of the films. However, the observed phonon mode frequencies and line widths indicate that the AlN samples grown at 400°C have less strain but more disorder compared to the ones grown at 650°C in agreement with RHEED and XRD data.

Using neural networks to construct models of the molecular beam epitaxy process

This paper presents the systematic characterization of the molecular beam epitaxy (MBE) process to quantitatively model the effects of process conditions on film qualities. A five-layer, undoped AlGaAs and InGaAs single quantum well structure grown on a GaAs substrate is designed and fabricated. Six input factors (time and temperature for oxide removal, substrate temperatures for AlGaAs and InGaAs layer growth, beam equivalent pressure of the As source and quantum well interrupt time) are examined by means of a fractional factorial experiment. Defect density, X-ray diffraction, and photoluminescence are characterized by a static response model developed by training back-propagation neural networks. In addition, two novel approaches for characterized reflection high-energy electron diffraction (RHEED) signals used in the real-time monitoring of MBE are developed. In the first technique, principal component analysis is used to reduce the dimensionality of the RHEED data set, and the reduced RHEED data set is used to train neural nets to model the process responses. A second technique uses neural nets to model RHEED intensity signals as time series, and matches specific RHEED patterns to ambient process conditions. In each case, the neural process models exhibit good agreement with experimental results.

Kinetic energy dependence of TiSi 2 film growth from low energy Ti + ion beams

The growth of TiSi 2 thin films by using low energy (10–500 eV) Ti + ion beams on Si(1 1 1) has been studied by in situ Auger electron spectroscopy (AES) and reflection high energy electron diffraction (RHEED) as well as ex situ depth profile X-ray photoelectron spectroscopy ( d-XPS). Stoichiometric TiSi 2 films were grown with Ti + energies of 50 eV at a temperature of 600°C. The in situ RHEED and AES data reveal the existence of some elemental Si in the surface region, whose concentration is dependent on the kinetic energy of the incoming Ti + ions. Resolution of the chemical shift between the Si(2p) lines of the silicide and elemental silicon allows the ex situ d-XPS profiles to reveal the film composition and thickness. The effect of energetic ions in silicide film formation is discussed accordingly. The data show consistent trends that demonstrate that irradiation with Ti + ions suppresses the diffusion of Si as the ion kinetic energy increases. It is proposed that this radiation inhibited diffusion process is due to void formation resulting from the high density of defects in the subsurface region which inhibits diffusion of Si atoms through the silicide film to react with incoming Ti atoms.