No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Preparation of 4-inch Ir/YSZ/Si(001) substrates for the large-area deposition of single-crystal diamond
Abstract
Diamond/Ir/YSZ/Si(001) is currently the most promising multilayer structure for the future realisation of large-area diamond single crystals. A decisive key is the preparation of the iridium layers on silicon. It is shown in this work that high quality iridium films with mosaic spread below 0.2° can be grown on oxide buffer layers with a mosaic spread higher than 1°. An averaging process during the coalescence of the iridium islands provides a plausible mechanism for this phenomenon. The oxide buffer and the iridium overlayers can be grown homogeneously on 4-inch wafers in a similar quality as for 1 × 1 cm2 samples. Bias enhanced nucleation followed by 40 h growth on the large-area Ir/YSZ/Si(001) wafers yields diamond films with a mosaicity of 0.16° (tilt) and 0.34° (twist). For a further increase of the area of heteroepitaxial diamond nucleation the homogeneity of the plasma discharge has to be improved.
... Diamond grown on an iridium (Ir) buffer layer has been proved good crystallinity [7][8][9]. However, due to the difference of lattice constants between diamond and Ir, the crystal lattice mismatch, strains and defects are generated from diamond/Ir interface [10][11][12][13], leading to the higher defect density of heteroepitaxial diamond compared with homoepitaxial samples [14][15][16]. Therefore, it is essential to understand the origin and propagation of defects during the different growth stages of heteroepitaxial diamond, which could make great significance for the preparation and practicability of large-area, high-quality diamond substrate. ...
... The results showed that in the area A near diamond/Ir interface, there were regular spots representing cubic crystal systems in the electron diffraction spots. After the calculation of SADP images, the crystal plane spacing of diamond (2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20) plane was 1.274 Å, increased by 1% compared to standard diamond card (1.261 Å), which was related to the epitaxial diamond growth from Ir/Al 2 O 3 substrate. In addition, the crystal-plane spacing of Ir (2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20) was larger than that of diamond (1.357 Å), and two azimuthally broadened diffraction spots can be observed at the SADP of area A, indicating the in-plane disorientation of the initial diamond islands grown on the Ir layer. ...
... After the calculation of SADP images, the crystal plane spacing of diamond (2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20) plane was 1.274 Å, increased by 1% compared to standard diamond card (1.261 Å), which was related to the epitaxial diamond growth from Ir/Al 2 O 3 substrate. In addition, the crystal-plane spacing of Ir (2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20) was larger than that of diamond (1.357 Å), and two azimuthally broadened diffraction spots can be observed at the SADP of area A, indicating the in-plane disorientation of the initial diamond islands grown on the Ir layer. This means, because of the small disorientation of the Ir buffer, the diamond islands at the very beginning of the growth had a tendency to be textured [21]. ...
This paper investigates the formation and propagation of defects in the heteroepitaxial growth of single-crystal diamond with a thick film achieving 500 µm on Ir (001)/Al2O3 substrate. The growth of diamond follows the Volmer–Weber mode, i.e., initially shows the islands and subsequently coalesces to closed films. The films’ strain imposed by the substrate gradually relaxed as the film thickness increased. It was found that defects are mainly located at the diamond/Ir interface and are then mainly propagated along the [001] direction from the nucleation region. Etching pits along the [001] direction formed by H2/O2 plasma treatment were used to show defect distribution at the diamond/Ir/Al2O3 interface and in the diamond bulk, which revealed the reduction of etching pit density in diamond thick-film surface. These results show the evident impact of the thickness on the heteroepitaxially grown diamond films, which is of importance for various device applications.
... There has been a lot of effort devoted to finding a suitable material for diamond heteroepitaxy, such as Si, SiC, TiC, Ni, Pt, or Co. [7][8][9][10][11][12][13] As a result, epitaxial Ir films deposited on a foreign substrate (owing to the lack of a bulk single-crystalline Ir substrate) are at this moment considered to be the best material for diamond heteroepitaxy. [14][15][16][17][18][19][20][21][22] However, the quality of heteroepitaxial diamond is still an issue as it remains far from that of natural or HPHTsynthesized diamond. To the best of our knowledge, the best recorded crystal quality of heteroepitaxial diamond grown on Ir thin films exhibited a 0.16°full width at half maximum (FWHM) X-ray rocking curve (XRC) for diamond (004) reflections; 20) a reduction of at least one or more orders of magnitude in this FWHM value is expected. ...
... [14][15][16][17][18][19][20][21][22] However, the quality of heteroepitaxial diamond is still an issue as it remains far from that of natural or HPHTsynthesized diamond. To the best of our knowledge, the best recorded crystal quality of heteroepitaxial diamond grown on Ir thin films exhibited a 0.16°full width at half maximum (FWHM) X-ray rocking curve (XRC) for diamond (004) reflections; 20) a reduction of at least one or more orders of magnitude in this FWHM value is expected. Toward this task, in addition to the selection of a substrate material for Ir deposition and the optimization of the Ir deposition, the epitaxial lateral overgrowth (ELO) of diamond, which is a well-known method of significantly reducing dislocations in other semiconductor heteroepitaxial materials such as GaN, 23) is now increasing in technological importance. ...
... Consequently, the Si substrate, which offers a near-fit CTE with diamond, has been the focus of much attention in recent research. 16,19,20) However, since Si is not suitable for Ir deposition, heteroepitaxy with Si is still not perfect, and recent discussions have explored potential intermediate buffer layers grown on Si, such as yttria-stabilized zirconia (YSZ), on which highquality Ir can be grown. 19,20) The optimization of the growth conditions for the buffer layers is also under discussion. ...
The fabrication of a high-quality freestanding diamond substrate was successfully demonstrated via heteroepitaxy by introducing diamond micropatterns and microneedles in the early stage of growth. Micropatterns contributed to a marked reduction in the number of dislocations induced by epitaxial lateral overgrowth, and microneedles relaxed heteroepitaxial strain. Raman spectroscopy indicated the absence of nondiamond carbon inclusions in the obtained freestanding substrate. The full width at half maximum of the X-ray rocking curve for diamond (004) reflections was 0.07°, the lowest value for heteroepitaxial diamond that has been reported so far. The results provide novel insights toward realizing large-diameter single-crystalline diamond substrates.
... Diamant/Si 55,2 % 30 % 1˚4˚Sarrieu et al. [44] Diamant/β-SiC 22,2 % <50 % 0,5˚2,5˚Stoner et al. [12] Diamant/Ir 7,6 % >90 % -0,1-0,3˚Sawabe et al. [45] 0,2˚-Golding et al. [46] 0,16˚0,3˚Schreck et al. [47] Tableau 1.4 -Caractéristiques des cristaux de diamant épitaxiés et désorientations des films en fonction des hétérosubstrats utilisés. ...
... Ce type de substrat sera utilisé lors de cette thèse. Hörmann et al. [53] Golding et al. [71] Fischer et al. [47] 0,57 Bauer et al. [72] Tableau 1.5 -Caractéristiques des différents hétérosubstrats disponibles pour l'épitaxie de l'iridium. ...
... Ir/SrT iO 3 0,17 0,19 150 Schreck et al. [67] Ir/Al 2 O 3 0,2 / 150 Dai et al. [46] Ir/Y SZ/Si 0,2 0,28 150 Fischer et al. [47] Ir/SrT iO 3 0,23 0, 27 200 Thèse Chavanne [13] Tableau 2.2 -Désorientation des films d'iridium des substrats présents dans la littérature. ...
This thesis aims to master the MPCVD synthesis of heteroepitaxial diamond films of high crystalline quality on iridium substrate for radiotherapy dosimeters. This objective has led us to develop the epitaxial iridium layer grown on SrtiO3 substrates (001). A vacuum frame equipped with an electron gun has been developed and calibrated. The obtained layers characterized by XRD, possess a structural quality equivalent to the state of the art/in literature. Bias Enhanced Nucleation (BEN)- MPCVD induces nucleation of " domains " on the iridium surface, according a unique nucleation pathway. Significant work has been conducted on (BEN)-MPCVD optimization to obtain a reliable and reproducible method for generating homogeneous " domains " on a surface of 5x5mm2. Combined characterizations (SEM, XPS, AES) of " domains " surface enabled us to establish the identity card of their chemical and morphological properties. We demonstrate that they contain diamond nuclei. In addition, the temporal expansion of these " domains " seems to follow preferential directions <110> of iridium lattice during the (BEN)-MPCVD stage. From these results, self-supported heteroepitaxial diamond films 100μ-m thick have been grown. The correlation between their crystalline quality and their detection response was conducted with the LCD dosimeter team. The inhomogeneities in the crystal structure due to structural defects have been identified. To study more locally these samples, a measurement campaign was carried out by microbeam X on the DIFFABS line at Soleil Synchrotron. The combination of the different knowledges acquired during this thesis has allowed the fabrication and characterization of the first detector based on heteropitaxial diamond at the LCD laboratory.
... Therefore, diamond heteroepitaxial growth has been extensively studied for several decades. [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28][29] Compared to other materials, Ir is considered the best substrate for diamond heteroepitaxial growth because it offers the highest diamond nucleation density (>1 × 10 8 cm −2 ). 17) However, Ir single crystals are expensive and not commercially available. ...
... 17) However, Ir single crystals are expensive and not commercially available. Therefore, Ir/YSZ (yttria-stabilized zirconia)/Si [20][21][22] and Ir/MgO [23][24][25] have been extensively studied. Schreck et al. 22 25) They utilized the microneedle (Kenzan) process to delaminate the grown diamond layer from the substrate without cracking. ...
Two-inch-diameter high-quality free-standing (001) diamond layers were grown on misoriented (11 2¯ 0) sapphire. The substrate misorientation allows step-flow growth, and tensile stress is released in the diamond layer. Consequently, the diamond layer delaminates naturally from the substrate without cracking. For the diamond grown on the sapphire misoriented by 7° toward the [1 1¯ 00] direction, the widths of the (004) and (311) X-ray rocking curves were 98.35 and 175.3 arcsec, respectively, the lowest ever reported. The curvature radius of the diamond was 99.64 cm in the [1 1¯ 00] direction and 260.21 cm in the [0001] direction of the substrate, the highest ever reported.
... It possesses a high melting point, low oxygen permeability, good chemical resistivity and very high oxidation resistance, which makes it a very attractive material for e.g. electrochemical electrode [1][2][3][4], as a suitable substrate for the growth of crystalline diamond films [5][6][7][8][9][10], or as protective coating of carbon materials [11,12]. As for many other noble metal catalysts, the electronic structure and catalytic activity of the iridium surface thereby strongly depend on its crystallographic orientation. ...
... The same orientational relationship was also found by Dai et al. [35]. Furthermore he evaluated the mismatch along these directions and found a 1.2% mismatch along the [110]Ir//[1-100]Al2O3 direction and in case of a 5:3 ratio (iridium to oxygen nearest-neighbor distance) a possible net mismatch of 4.5% along the [1][2][3][4][5][6][7][8][9][10]Ir//[0001]Al2O3 direction [35]. ...
We present results on the investigation of the substrate bias effect on the growth behavior of iridium films deposited on A-plane sapphire by radio frequency (rf) sputtering at low substrate temperatures. Films deposited without substrate bias were compared to films deposited with simultaneous application of a second rf-plasma on the substrate. Resulting films were characterized by scanning electron microscopy, X-ray diffraction, and electron backscattering diffraction. We find that the application of an additional substrate bias has a strong effect on the growth behavior of Ir in such a way that preferential growth of iridium (001) on sapphire (11−20) at high deposition rates and at substrate temperatures as low as 350 °C becomes feasible.
... The polar and azimuthal mosaicities of the iridium films grown on SrTiO 3 films of different thicknesses also lead to the same conclusion. It is reported that the mosaic spread of the iridium is by an order of magnitude lower than the buffer layer because of the liquid-like coalescence mechanism of iridium film: with 1.7°(both polar and azimuthal mosaicities) for YSZ film on Si, 0.2°(polar) and 0.11°(azimuthal) for iridium film on YSZ/Si [21,36]. In the case of our study, iridium films also had lower mosaicities than SrTiO 3 films although this effect was limited as compared to YSZ on Si since the mosaicities of the SrTiO 3 films were already low. ...
... In the case of our study, iridium films also had lower mosaicities than SrTiO 3 films although this effect was limited as compared to YSZ on Si since the mosaicities of the SrTiO 3 films were already low. As iridium cannot be directly epitaxially grown in contact with Si [36], SrTiO 3 /Si appears as a good system for epitaxial iridium films with low mosaicity. The investigation of the effect of the compositions of Sr and Ti in SrTiO 3 thin films on iridium deposition seems to be important to conduct, although a similar investigation (the effect of YO 1.5 content) in the case of YSZ films for diamond heteroepitaxy concluded that all YSZ films resulted in epitaxial film [37]. ...
... This problem was tackled by a research group at the University of Augsburg that has succeeded in producing mosaic diamond crystals using plasma chemical vapour deposition (MWPCVD). Heteroepitaxial diamond layers are grown in the <001> direction on a 150 nm thick iridium (001) film coated onto a thin layer of yttria-stabilised zirconia deposited on a silicon (001) wafer [3,4]. The surface of crystal plates is nearly parallel to (100) crystallographic planes, typically a few degrees off while both edges are parallel to (110)-type directions. ...
... In the meantime it is possible to obtain good quality crystal plates that are up to 0.18 cm thick with lateral dimensions of 1.5 x 1.5 cm 2 . Neutron double-crystal diffraction studies of a great number of samples has shown a high diffraction efficiency yielding values of the peak reflectivity close to the theoretical predictions [4,5]. The potential of heteroepitaxial diamond crystals suitable for neutron monochromators has thus been demonstrated unequivocally. ...
We report on the performance of the first diamond neutron monochromator built at the ILL. It has been designed for the hot neutron diffractometer D9 with the aim of improving significantly the instrument performance in particular for short wavelengths in the 0.3-0.9 Å wavelength range. Diamond crystal plates with dimensions of 1.5 x 1.5 x 0.18 cm3 an average mosaic spread of 0.15° have been synthesized at the University of Augsburg. They exhibited excellent neutron diffraction properties when examined on a neutron double-crystal test setup. Sufficiently thick diamond elements with a controlled mosaic spread of 0.25° have been obtained by stacking several of these crystals. First tests runs carried out at the ILL confirmed the predicted high reflectivity of the diamond stacks. The diamond prototype monochromator uses the (220) reflection in transmission geometry replacing the Cu (220) monochromator on D9 that has the same d-spacing. The final performance studies on D9 showed that the diamond device did not perform better than the original copper crystal. This unexpected result could be explained by significant optical aberrations caused by non- uniformities of both the angular and spatial mosaic distribution in the individual diamond crystals, as revealed by a detailed characterisation study using high-energy X-ray diffraction.
... The deposition of cubic single crystal Iridium is challenging as it requires complex substrate combinations. Single-crystal iridium films can be grown on a variety of oxides like Al 2 O 3 , SrTiO 3 and MgO [49]. The adhesion of diamond/Ir layers on these oxides turned out as major problem, due to large misfits in thermal expansion coefficients. ...
In the future, electronic parts will penetrate everything, generating a new and fast-growing pollution problem. Future devices therefore need to be environmentally friendly with strong recycling options. A paradigm change in semiconductor technology is predicted based on applications of better suited materials which can fulfil these criteria. Carbon based materials and here especially diamond are promising candidates. Bulk and surface properties of diamond are introduced in combination with applications in power electronics, quantum technology, bio-and electrochemistry and MEMS. Large amounts of diamond seeds and wafers will be required to approach commercial markets. Their availability in combination with quality and size as well as required energies for production are introduced. The production of CVD diamond is currently about 100–250 times more intense with respect to energy than Silicon. A problem which is addressed by use of new solid-sates microwave sources. The definition of “green diamond” is given taking into account requirements with respect to energy and methane/hydrogen production. A brief discussion and comparison of diamond global markets and related potentials in comparison to SiC and GaN is given.
... Yttrium-stabilized zirconia (YSZ) has several remarkable physical properties, such as excellent corrosion resistance, extraordinary thermal stability, and high ionic conductivity. Based on these properties, epitaxial YSZ films on silicon wafers have been widely used as buffer layers for growing superconducting films (e.g., YBa 2 Cu 3 O 7− x ) [1], ferromagnetic and ferroelectric films (e.g., Hf 0.5 Zr 0.5 O 2 , CoFe 2 O 4 ) [2,3], and heteroepitaxial diamond [4]. The YSZ thin film acts as a buffer layer to effectively prevent the interdiffusion and chemical reaction between the functional films and the silicon substrate. ...
... The growth of YSZ on (001)-Si has been optimized in previous works and can be found in the literature. 23,24 The YSZ layer has a thickness of 100 nm for all samples. To avoid the formation of Eu 3þ during the ETO growth, the oxygen flow was stopped after the YSZ growth leading to a pressure of 8 × 10 -6 mbar. ...
In this work, we show the epitaxial growth of (111)-oriented EuTiO[Formula: see text] thin films on (001)-oriented silicon with an in situ grown yttria-stabilized zirconia (YSZ) buffer layer by pulsed laser deposition. X-ray diffraction measurements revealed a homogeneously strained EuTiO[Formula: see text] thin film with a strain dependency on the laser fluence during the film growth. From magnetization vs temperature measurements, we confirmed that the strained EuTiO[Formula: see text] films have an antiferromagnetic to ferromagnetic transition at 3.7 K, which disappeared for unstrained films. Furthermore, we used electron backscatter diffraction to analyze the columnar growth of EuTiO[Formula: see text] on YSZ, which showed four in-plane orientations.
... The diamond film on Ir has a special growth mode, the diamond with an orientation arrangement has very high density and orientation consistency, 14,15 and Ir has become the buffer material for preparing single crystal diamond wafers. 16 Several groups have deposited Ir on other substrates to grow a single crystal diamond such as Ir/MgO, 17 Ir/ YSZ/Si, [18][19][20][21] and Ir/SrTiO 3 /Si. 22,23 The full width at half maximum (FWHM) of the x-ray diffraction measurement of the (004) diamond is 230 arc sec, which was grown on an Ir/YSZ/Si substrate by Schreck et al. 24 In addition, the heteroepitaxy of a single crystal diamond has a FWHM of 113 arc sec using the needle type lateral epitaxial technique on the Ir/Al 2 O 3 substrate, which has been published by Kim et al. 25 Several factors have been considered to affect the quality of a heteroepitaxy single crystal diamond crystal: (i) the lattice mismatch between an Ir (001) film and a substrate material, (ii) the lattice mismatch between the Ir film and a diamond film, and (iii) the thermal mismatch between the diamond film and the substrate. ...
The heteroepitaxy of a single crystal diamond has been carried out in the KTaO3 substrate using Ir as a buffer layer. KTaO3 has a perovskite lattice structure and displays a face-centered cubic structure. Its lattice constant is 3.98 Å, which is only 3% mismatched with the lattice constant of Ir of 3.84 Å, and also, its thermal expansion coefficient is 4.031 × 10⁻⁶/K, which is nearly close to that of diamond 3.85 × 10⁻⁶/K (at 1223 K), making it to be regarded as an alternate substrate of the heteroepitaxy of a single crystal diamond. The magnetron sputtering technique was used to deposit Ir thin films with a high orientation in the (001) direction on a KTaO3 (001) substrate. Thereafter, bias enhanced nucleation on Ir surface was grown by direct current chemical vapor deposition (CVD) methods. At last, a single crystal diamond with a size of 10 × 10 × 0.78 mm³ has been grown, whose (004) rocking curve FWHM is 183 arc sec, which testifies to the excellent crystalline quality of the heteroepitaxial diamond film.
... 7 However, the area close to the junction between the initial substrates remained defective, thus limiting the use of this material in electronic or optical applications. 8,9 Heteroepitaxial growth of diamond on Ir/MgO (100), [10][11][12][13] Ir/YSZ/Si (001) [Iridium/ Yttria-Stabilized-Zirconia/Silicon(001)], [14][15][16][17] and Ir/SrTiO 3 /Si (001) [18][19][20] substrates was also developed, leading to sizes of up to several inches. However, the crystalline quality lacked significantly behind that obtained for homoepitaxial films because of the difference in the lattice parameter between Ir and diamond (7.2%). ...
The growth of large-area diamond films with low dislocation density is a landmark in the fabrication of diamond-based power electronic devices or high-energy particle detectors. Here, we report the development of a growth strategy based on the use of micrometric laser pierced hole arrays to reduce dislocation densities in heteroepitaxial chemical vapor deposition diamond. We show that, under optimal growth conditions, this strategy leads to a reduction in dislocation density by two orders of magnitude to reach an average value of 6x10^5 cm-2 in the region where lateral growth occurred, which is equivalent to that typically measured for commercial type Ib single crystal diamonds.
... At present, Ir is the most promising substrate for diamond heteroepitaxy because the diamond nucleation density on Ir (>1 Â 10 8 cm À2 ) is the highest among all. 22 However, because of the lack of the single-crystal Ir substrate, Ir/MgO, [22][23][24] Ir/YSZ (yttria-stabilized zirconia)/Si, [25][26][27][28] and Ir/ STO (strontium titanate)/Si 29,30 have been studied. On the Ir/MgO substrate, Aida and Kim et al. reported that the full-width at half maximum (FWHM) of the x-ray rocking curve (XRC) (004) is 252 arc sec. ...
One-inch free-standing (001) diamond layers on a (11 2 ¯0) (a-plane) sapphire substrate with an Ir buffer layer (Kenzan Diamond®) were grown. The full-width at half maximum values of (004) and (311) x-ray rocking curves were 113.4 and 234.0 arc sec, respectively. The dislocation density of the substrates was 1.4 × 10⁷ cm⁻², determined by plan-view transmission electron microscopy observation. These values are much lower than the reported values among heteroepitaxial diamonds. Furthermore, x-ray pole figure measurements showed four symmetry of the crystal, showing single crystallinity without any twinning. The curvature radius of diamond was measured to be 90.6 cm, which is much larger than previous values, ca. 20 cm. Surprisingly, a cubic-lattice (001) diamond crystal was epitaxially grown on a trigonal-lattice (11 2 ¯0) sapphire substrate. However, we found that the epitaxial relation is diamond (001) [110]//Ir (001) [110]//sapphire (11 2 ¯0) [0001]. Now, high-quality one-inch diamond wafers will be available as a substrate used for diamond electronic devices.
... There are several candidate technologies under development to achieve large size diamond substrates [10,11]. First, hetero-epitaxial growth techniques use large hetero-substrates such as MgO [12,13], SrTiO 3 [14], Al 2 O 3 [15] and YSZ/Si [16]. Assuming these substrates can be used for upscaling the epitaxial crystal growth process, available wafer sizes are at least 2 in. ...
A mosaic substrate is a promising candidate to create large size single crystal diamonds for various types of applications. In this study, the crystal orientations of the joint areas of four single crystal plates of a mosaic substrate were measured using the high resolution electron backscatter diffraction method. The lattice rotation of [110] to [–110] (X-Y direction) was less than 0.5° and the lattice rotation mappings of [001] to [110] (Z-X direction) and [1–10] (Z-Y direction) were less than 0.2°. Considering that plate alignment is not particularly considered during the fabrication process, further improvement can be expected to realize a well- aligned mosaic substrate.
... Furthermore, diamond is a host to quantum emitters such as the nitrogen vacancy (NV) and silicon vacancy (SiV) colour centres 23,24 that can be coupled to mechanical [25][26][27][28][29] and optical [30][31][32][33] resonators and used to generate single photons for quantum networking applications [34][35][36] and store quantum information. [37][38][39] SCD is currently not commercially available in heteroepitaxially grown thin film form, and although efforts to grow high quality diamond films on substrates such as Ir/MgO, 40 Ir/YSZ/Si, 41,42 and SiC/Si 43 are underway, this material is not yet readily available. Integrated optical and optomechanical devices have been demonstrated in polycrystalline diamond (PCD), [44][45][46] which is commercially available in thin film form but is not an ideal host for highly coherent quantum emitters. ...
Nanophotonic structures in single–crystal diamond (SCD) that simultaneously confine and
co-localize photons and phonons are highly desirable for applications in quantum
information science and optomechanics. Here we describe an optimized process for etching
SCD microdisk structures designed for optomechanics applications. This process allows the
optical quality factor, Q, of these devices to be enhanced by a factor of
4 over previous demonstrations to Q ∼ 335 000, which is sufficient to
enable sideband resolved coherent cavity optomechanical experiments. Through analysis of
optical loss and backscattering rates, we find that Q remains limited by
surface imperfections. We also describe a technique for altering microdisk pedestal
geometry which could enable reductions in mechanical dissipation.
... So gelang es Schreck et al. nach vorangegangener gleichspannungsunterstützter Bekeimung (Bias Enhanced Nucleation, BEN) [Yug91] über das CVD Verfahren erstmals einkristalline Diamantschichten auf dem Metall abzuscheiden [Sch01a], deren Qualität auf keinem anderen untersuchten Material erreicht wurde. Die Fläche und Kristallqualität des dabei verwendeten Schichtsystems Ir/YSZ/Si wurden in der Diamantgruppe der Universität Augsburg stetig erhöht [Gse07] [Fis08b], was die Synthese der mit einem Durchmesser von 92 mm weltweit größten künstlich hergestellten einkristallinen Diamantscheibe ermöglichte. Im Zuge dieser Forschungsarbeiten konnte darüber hinaus ein schlüssiges Modell zur Erklärung des Nukleationsvorgangs von Diamant auf Iridium gefunden werden [Sch17]. ...
In der vorliegenden Arbeit wurden (111)-orientierte Diamantschichten heteroepitaktisch auf einem Schichtsystem aus Ir/YSZ/Si hergestellt. Hierfür wurden zunächst Prozessparameter identifiziert, die die Ausbildung intrinsischer Spannungen verhindern. Dazu wurde mit Hilfe von Röntgenbeugung der vollständige Spannungstensor für hunderte Diamantschichten ermittelt. Als allgemeiner Trend wurde ein Übergang der intrinsischen Spannungen von Druckspannungen bei niedriger Depositionstemperatur zu Zugspannungen bei hoher Depositionstemperatur beobachtet. Dabei ergaben sich für auf vizinalen (111) Ebenen gewachsenen Schichten in der Regel biaxial anisotrope Spannungszustände, wobei die Spannungsanisotropie mit steigendem off-axis Winkel größer wird. Durch präzise Wahl der Prozessparameter konnten Bedingungen gefunden werden, die spannungsfreies Wachstum ermöglichen, wodurch die Abscheidung mehrere hundert µm dicker Schichten erreicht werden konnte. Zur Erklärung des Mechanismus der entstehenden intrinsischen Spannungen wurde im Rahmen der Doktorarbeit von Dr. Martin Fischer in der Augsburger Diamantgruppe ein Modell vorgeschlagen, das auf dem Prozess des „effektiven Versetzungskletterns“ basiert. In diesem Modell wird der Aufbau von intrinsischen Spannungen durch eine veränderte Propagationsrichtung von Versetzungen erklärt. In der vorliegenden Arbeit konnte mittels TEM Aufnahmen erstmalig eine direkte Korrelation zwischen dem Propagationsverhalten von Versetzungen und dem Aufbau intrinsischer Spannungen nachgewiesen werden. Hierfür wurden gezielt Doppelschichten synthetisiert, bei denen durch eine Veränderung der Prozessparameter in beiden Teilschichten stark unterschiedliche Spannungszustände vorlagen. Das Modell des „effektiven Versetzungskletterns“ wurde darüber hinaus noch um einen step-flow induzierten Beitrag erweitert, mit dessen Hilfe die in dieser Arbeit erstmalig beobachteten anisotropen Spannungszustände erklärt werden konnten. Für die Topographie der Diamantschichten ergab sich eine deutliche Abhängigkeit von den Prozessparametern und den off-axis Eigenschaften. Während die Schichten bei hoher Methankonzentration im Bereich von 4,5% von pyramidalen Erhebungen geprägt sind, ergibt sich bei niedrigem Methangehalt von 2%-2,5% speziell bei off-axis Schichten eine alternierende Abfolge von Terrassen- und Riserstrukturen. Erstere zeigen dabei glatte Oberflächen, wobei die Pole der Flächen zur [111] Richtung hin geneigt sind, so dass das Wachstum auf ihnen unter quasi on-axis Bedingungen stattfindet. Letztere weisen dagegen Oberflächen mit erhöhter Rauheit auf und sind um deutlich größere Winkel in die entgegengesetzte Richtung verkippt. Anhand der für die off-axis Schichten analysierten Oberflächenstrukturen konnte darüber hinaus ein Mechanismus gefunden werden, der die beobachtete Invertierung der Spannungsanisotropie erklären kann. Hierfür wurden die Spannungszustände, die sich an den lokalen Wachstumsfronten bilden zu einem makroskopischen Spannungszustand aufsummiert. Die Kristallqualität der (111)-orientierten Diamantschichten wurde über die Veränderung der Mosaikbreite sowie die der Versetzungsdichte bestimmt. Bei ersterer wurde insbesondere für die azimutale Komponente eine sehr schnelle Verkleinerung der Werte festgestellt, woraus geschlossen werden konnte, dass für (111)-orientierten Diamant die Bildung von Disklinationen hier der dominante Mechanismus ist. Für den Verlauf der Versetzungsdichte über der Schichtdicke ergaben sich für (111)-orientierten Diamant Werte die um einen Faktor von 2-3 unter denen liegen, die für (001)-orientierten Diamant im Rahmen früherer Arbeiten gefunden wurden. Bei der mit der Versetzungsdichte eng verbundenen Breite der Ramanlinie ergab sich ein Minimalwert von 1,8/cm, der unter allen Werten liegt, die bisher für (001)- und (111)-orientierten, heteroepitaktisch gewachsenen Diamant gemessen wurden. Für die in der Literatur bei (111)-orientiertem Diamantwachstum als besonders kritisch beschriebene Ausbildung von Zwillingen wurden systematische Abhängigkeiten gefunden. So kam es bei einer Erhöhung der Methankonzentration, der Zugabe von Stickstoff zur Gasphase sowie bei einer Reduzierung der Depositionstemperatur und einer Erhöhung des off-axis Winkels zu einer verstärkten Entstehung von Durchdringungszwillingen. Durch die präzise Wahl der Prozessparameter konnte deren Auftreten sowohl bei auf on-axis als auch bei auf off-axis Substraten hergestellten Schichten komplett verhindert werden. Einige dicke (111)-orientierte Diamantschichten wurden am Neutronenstrahl des ILL in Grenoble auf ihre absolute Neutronenreflektivität vermessen. Dabei erreichte insbesondere ein Plättchen, das aus (001) gewachsenem Diamant herausgeschnitten wurde eine Absolutreflektivität von nahezu 70%. Somit konnte bei einer Neutronenwellenlänge von 1,7A das Potenzial von Diamant als Konkurrenz zu den etablierten HOPG Neutronenmonochromatoren gezeigt werden.
... SCD is not currently commercially available in heteroepitaxially grown thin film form, and although efforts to grow high quality diamond films on substrates such as Ir/MgO 40 , Ir/YSZ/Si 41,42 , and SiC/Si 43 are underway, this material is not yet readily available. Integrated optical and optomechanical devices have been demonstrated in polycrystalline diamond (PCD) [44][45][46] , which is commercially available in thin film form but is not an ideal host for highly coherent quantum emitters. ...
Nanophotonic structures in single--crystal diamond (SCD) that simultaneously confine and co-localize photons and phonons are highly desirable for applications in quantum information science and optomechanics. Here we describe an optimized process for etching SCD microdisk structures designed for optomechanics applications. This process allows the optical quality factor, $Q$, of these devices to be enhanced by a factor of 4 over previous demonstrations to $Q \sim 335,000$, which is sufficient to enable sideband resolved coherent cavity optomechanical experiments. Through analysis of optical loss and backscattering rates we find that $Q$ remains limited by surface imperfections. We also describe a technique for altering microdisk pedestal geometry which could enable reductions in mechanical dissipation.
... Recent development of diamond wafers has been summarized in a special issue of MRS journal [17]. Hetero-epitaxial growth techniques using large hetero-substrates such as MgO [18], SrTiO3 [19], Al2O3 [20] and YSZ/Si [21] are hopeful for its original substrate size. A drawback of hetero-epitaxial growth is the high dislocation density, far beyond that of single crystals, that originates from the coalescence of crystal plates. ...
Two films grown by fast deposition of Mg on Si(111) surfaces
with temperature distributions within 365–395C and 545–
609C were studied by Raman spectroscopy, X-ray diffraction
(XRD), X-ray photoelectron spectroscopy (XPS) and atom force
microscopy (AFM). The film grown at 365–395C contains crystalline textured Mg2Si. The mechanism of Mg2Si formation at
high temperature is considered. The role of a Mg atom lifetime
before evaporation from the Si surface at high temperatures is
emphasized. The crucial requirement for the Mg2Si growth at
high temperatures is a sufficiently high deposition rate providing condensation of Mg on the surface and further formation of
the silicide.
... As a further alternative, we have recently reported an innovation in heteroepitaxial diamond substrate [11]. Heteroepitaxy has the theoretical possibility of producing SCD the same size as the basal substrates [12][13][14][15][16], and the most promising structure for producing SCD by heteroepitaxy is believed to be diamond/Ir/MgO [17,18]. However, the large mismatch in the coefficient of thermal expansion (CTE) of diamond and the MgO substrate actively prevents the growth of diamond layers thick enough for producing freestanding SCD substrate, as serious cracks are generated in the structure due to the bowing of the structure after the growth of diamond during the temperature reduction from growth to room temperature [15]. ...
A detailed study of diamond growth on diamond microneedles was conducted using micro-Raman spectroscopy of the microneedle and coalescence regions to approach the production of freestanding diamond substrate by heteroepitaxy with the microneedle growth method. The high-density non-diamond-phase carbon is contaminated in the initial stage of overgrowth of diamond on diamond microneedles, but this completely disappears through the quick recovery of the crystallinity of the overgrown diamond layers during the coalescence with the lateral direction growth. We also point out the possibility that a strong driving force is applied to the dislocations generated at the regrowth point and at the coalescence front to enhance the mutual annihilation of dislocations. In addition, we reveal that the stress state changes from compressive stress in the initial diamond layers to a nearly stress-free state in the bulk layers on the microneedles through a momentary tensile stress state at the regrowth point at the tip of the microneedle. Overall results indicate the strong feasibility of producing freestanding, stress-free, single-crystal diamond substrate by heteroepitaxy.
... For a 320 μm thick diamond, we found a strong reduction of the in-plane mosaicity: from 0.74°averaged on the whole crystal (conventional XRD, Fig. 3(d)) to 0.37°at the crystal surface ( Fig. 6(b)). Here again, these values are close to the state-of-the art for heteroepitaxial diamond films grown on iridium [19], which confirms the high structural quality of our samples. More generally, a mosaicity reduction for thick films have also been reported for heteroepitaxial diamond in both out-of-plane and in-plane directions [11]. ...
The present study provides a multi-scale investigation of the crystalline quality and the structural defects present in heteroepitaxial diamond films grown on iridium/SrTiO3 (001) substrates by microwave plasma assisted chemical vapor deposition. X-ray diffraction, Raman spectroscopy and low temperature cathodoluminescence are combined to accurately characterize the mosaicity, the density of dislocations and the residual strain within the films. X-ray diffraction and Raman results confirm a structural quality at the state-of-the-art according to the epitaxial relationship 〈100〉diamond(001) // 〈100 >〉iridium(001) // 〈100〉SrTiO3 (001). In addition, Raman and cathodoluminescence observations on cross-sections reveal the presence of local strain.
... Recent development of diamond wafers has been summarized in a special issue of MRS journal [17]. Hetero-epitaxial growth techniques using large hetero-substrates such as MgO [18], SrTiO3 [19], Al2O3 [20] and YSZ/Si [21] are hopeful for its original substrate size. A drawback of hetero-epitaxial growth is the high dislocation density, far beyond that of single crystals, that originates from the coalescence of crystal plates. ...
According to international energy proposal, about 25% of the total CO2 reduction should come from “end use efficiency”. Hence, low loss power devices are an important technology for the 21st century. Diamond-based devices have the potential, but this would require fast development in order to contribute to the CO2 reduction plan early in this century. Here, we present a clear target for the R&D of diamond wafer. According to the expected applications of diamond devices with a vertical structure, the required target properties for first stage diamond wafers are; a killer defect density less than 0.1cm−2, resistivity less than 0.005Ωcm and a size of 4in. For the final commercialization stage, targets of zero killer defects, resistivity of 0.001Ωcm and a size of 6in. are proposed. The challenges and proposal solutions are reviewed for each technology.
... Finally, a plasma etching step was carried out to produce etch pits on the diamond surface. The preparation of the Ir/YSZ/Si(001) substrate and diamond nucleation by bias-enhanced nucleation (BEN) is described in detail elsewhere 15 . In short, a several 10s of nms thick layer of yttria-stabilized zirconia (YSZ) was deposited epitaxially on the Si(001) substrate by pulsed laser deposition (PLD) 16 , followed by an e-beam evaporation of Ir 17 . ...
A thin film of heavily B-doped diamond has been grown epitaxially by microwave plasma chemical vapor deposition on an undoped diamond layer, on top of a Ir/YSZ/Si(001) substrate stack, to study the boron segregation and boron environment at the dislocations present in the film. The density and nature of the dislocations were investigated by conventional and weak-beam dark-field transmission electron microscopy techniques, revealing the presence of two types of dislocations: edge and mixed-type 45° dislocations. The presence and distribution of B in the sample was studied using annular dark-field scanning transmission electron microscopy and spatially resolved electron energy-loss spectroscopy. Using these techniques, a segregation of B at the dislocations in the film is evidenced, which is shown to be intermittent along the dislocation. A single edge-type dislocation was selected to study the distribution of the boron surrounding the dislocation core. By imaging this defect at atomic resolution, the boron is revealed to segregate towards the tensile strain field surrounding the edge-type dislocations. An investigation of the fine structure of the B-K edge at the dislocation core shows that the boron is partially substitutionally incorporated into the diamond lattice and partially present in a lower coordination (sp(2)-like hybridization).
... The mosaicity of 45 mm thick (100)-oriented diamond film deposited on Ir/YSZ/Si(001) substrate was rather low, 0.27 for the tilt (FWHM) as determined from rocking curves [158]. This approach promises the epitaxial growth on large area (w10 cm 2 ) wafers [159]. Recently [160], a heteroepitaxial diamond film on multilayer Ir/YSZ/Si substrate as thick as 300 mm for use as high energy particle detector has been reported. ...
In this chapter we have presented a review of the state-of-the-art of diamond crystal growth using high pressure high temperature (HPHT) and chemical vapor deposition (CVD) approaches. We considered the main methods, techniques, and equipment used for single crystal diamond growth. Special attention is given to discussing the effects of growth conditions on the growth and properties of diamond. Current achievements in HPHT and CVD growth of single crystal diamond and its applications are considered.
... Untersucht man jedoch sehr perfekte Einkristalle, muss die Mehrfachstreuung berücksichtigt werden, da es sonst zu größeren Abweichungen und Fehlern kommt. Diese Mehrfachstreuung (dargestellt in Abb.3.2) wird in der Dynamischen Beugungstheorie beschrieben[War90].Abbildung 3.2 Einfache Darstellung der Reflektion des reflektierten Strahls, wie sie im Rahmen der Dynamischen Beugungstheorie berücksichtig wird [War90]. Die zweimal gestreute Strahlung ist in blau dargestellt. ...
Diamant weist eine Vielzahl extremer Materialeigenschaften auf und nimmt diesbezüglich eine herausragende Stellung unter allen Festkörpern ein. Besonders hervorzuheben sind die höchste Wärmeleitfähigkeit aller Festkörper bei Raumtemperatur, die höchste mechanische Härte und die breitbandige optische Transparenz. Darüber hinaus sind durch die sehr hohen Ladungsträgerbeweglichkeiten in extrem reinen und strukturell perfekten Kristallen ausgezeichnete elektronische Eigenschaften gegeben. Diese Kombination herausragender Parameter macht Diamant für ein breites Feld von Anwendungen zum ultimativen Basismaterial.
Natürlich vorkommende Diamanten entstanden im Erdinneren unter sehr hohen Drücken und Temperaturen und wurden durch Vulkanismus an die Erdoberfläche bzw. in oberflächennahe Schichten befördert. Für die künstliche Herstellung von Diamant gibt es prinzipiell zwei unterschiedliche Ansätze, die Hochdruck-Hochtemperatur-Synthese (High Pressure High Temperature, HPHT) und die chemische Gasphasenabscheidung (Chemical Vapour Deposition, CVD).
Durch die letztgenannte Methode gelang es bereits, polykristalline Diamantscheiben mit einem Durchmesser von bis zu 15 cm und einer Dicke von mehreren Millimetern herzustellen. Dieses polykristalline Material wird dank seiner großflächigen Verfügbarkeit in der Zwischenzeit industriell genutzt.
Bei einkristallinem Material ist die großflächige Verfügbarkeit bislang nicht gegeben. Die strukturell sehr hochwertigen HPHT-Kristalle sind in ihrer Größe auf ca. 1 cm Kantenlänge begrenzt. Es gibt allerdings Ansätze, durch homoepitaktisches Wachstum mittels CVD auf einzelnen oder auf einer kachelförmigen Anordnung von HPHT-Kristallen großflächige Einkristalle zu wachsen.
Eine alternative Möglichkeit stellt die Heteroepitaxie von Diamant dar. Seit ca. 1990 wurden viele unterschiedliche Materialien als mögliche Substrate getestet. Jedoch gelang es einzig auf Iridium, dünne Diamantschichten mit einkristalliner Struktur zu deponieren. Iridium lässt sich als einkristalline Schicht auf Silizium, das thermisch sehr gut zu Diamant passt, abscheiden. Es wird dabei eine Oxidpufferschicht (z.B. Yttriumoxid-stabilisiertes-Zirkondioxid, YSZ) auf dem Silizium verwendet, um eine Iridiumsilizidbildung zu vermeiden. Für die epitaktische Keimbildung von Diamant auf Iridium wird darüber hinaus ein gleichspannungsunterstützter Nukleationsschritt (Bias Enhanced Nucleation, BEN) benötigt.
Die vorliegende Arbeit beschäftigt sich mit der Synthese heteroepitaktischer Diamantschichten auf Ir/YSZ/Si-Substraten durch chemische Gasphasenabscheidung im Mikrowellenplasma.
Im ersten Teil wird die gleichspannungsunterstützte Nukleation von Diamant auf Iridium (111) Substraten untersucht und mit den früheren Ergebnissen auf Iridium (001) verglichen. Es stellt sich heraus, dass für diese Kristallorientierung analoge Phänomene wie auf (001)-Flächen auftreten, was nahelegt, dass es sich um einen universellen Prozess für die Diamantnukleation auf Iridium handelt. Bei der Nukleation auf Iridium (111) tritt nahezu ausschließlich die epitaktische Orientierungsvariante ohne nennenswerte Zwillingsbeiträge auf.
Als zweites wird das Wachstum von Diamantschichten beider Orientierungen untersucht. Man beobachtet, dass Eigenspannungen beim Wachstum heteroepitaktischer Diamantschichten eine zentrale Rolle für die Abscheidung spielen. Sie beeinflussen die Eigenschaften des abgeschiedenen Materials und die Stabilität des Schichtsystems. Für die Wachstumsspannungen zeigt sich eine charakteristische Temperaturabhängigkeit. Durch Vergleichsexperimente zum homoepitaktischen Wachstum können systematische Abhängigkeiten für Wachstumseigenspannungen beim epitaktischen Diamantwachstum beobachtet werden, woraus ein Modell für deren Entstehung entwickelt wird. Damit kann erstmals eine Vielzahl von früheren Berichten anderer Autoren konsistent erklärt werden. Darüber hinaus sind diese Erkenntnisse von enormer technologischer Bedeutung für die Entwicklung stabiler Wachstumsprozesse.
Beim Wachstum (001)-orientierter Diamantschichten werden erstmals Dicken bis in den Millimeterbereich realisiert. Technologische Probleme wie Rissbildung, Verbiegung und Delamination vom Substrat können auf die Wachstumseigenspannungen bei der Abscheidung zurückgeführt und minimiert werden. (111)-orientierte Schichten erweisen sich als interessantes Wachstumssubstrat für die Bildung von Heterostrukturen mit Wurtzittyp-Halbleitern wie Zinkoxid oder Aluminiumnitrid.
Einen Schwerpunkt der vorliegenden Arbeit stellt die Synthese und Charakterisierung von Diamantmosaikkristallen hinsichtlich ihrer Verwendung als Monochromator für Neutronen dar. Theoretische Berechnungen sagen eine Erhöhung des Neutronenflusses um einen Faktor 2-4 voraus. Diamantmosaikkristalle werden deshalb hinsichtlich ihrer Struktur mittels Röntgen-, Synchrotron- und Neutronenstrahlung eingehend charakterisiert. Die Beugungsexperimente mit Neutronen an den Forschungsreaktoren ILL und FRM II bestätigen weitestgehend die theoretischen Vorhersagen und damit das Potential der auf Ir/YSZ/Si gewachsenen Mosaikkristalle als ultimatives Material für die Monochromatisierung von thermischen und heißen Neutronen.
Abschließend wird kurz eine Reihe von weiterführenden Experimenten beschrieben, bei denen die heteroepitaktischen Diamantschichten als Wachstumssubstrate für ZnO und als Matrix für photonische Kristalle sowie die Iridiumschichten als Substrate für Einzelphotonenemitter genutzt werden. Auch werden mechanische und thermische Eigenschaften der heteroepitaktischen (001)-orientierten Schichten charakterisiert.
... Bulk Ir single crystals are not a technologically relevant option due to prohibitive price and lack of availability. Instead, heteroepitaxial Ir fi lms on MgO (001), 74 SrTiO 3 (001), 60 Al 2 O 3 (11)(12)(13)(14)(15)(16)(17)(18)(19)(20), 75 or YSZ/Si (001) 76,77 have been used for nucleation and growth studies of diamond on Ir. In principle, all these substrates can be used for upscaling the epitaxial crystal growth process: available wafer sizes are at least 2 inches for SrTiO 3 and MgO, 8 inches for Al 2 O 3 , and 12 inches for Si. ...
Diamond offers a unique combination of extreme physical properties. For many technological applications, diamond samples of the highest crystal quality are required to utilize the ultimate potential of the material. Specifically, grain boundaries, as in polycrystalline films, have to be avoided. In this article, the two major current approaches of synthesizing single crystal diamond by chemical vapor deposition are described. In homoepitaxy, high gas pressure and high power density microwave discharges facilitating growth rates above 50 µm/h form the basis for the deposition of mm-thick single crystal samples. Cloning and tiling followed by homoepitaxial overgrowth are promising novel concepts aimed at an increase in the lateral dimensions. Heteroepitaxial deposition on large-area single crystals of a foreign material represents a second alternative approach. The state of the art for both concepts is summarized, and current as well as potential future applications are discussed.
... However, crystalline bulk Ir(1 1 1) substrates are costly and it would be much more convenient to use, and later eventually dissolve, only a metal monocrystalline thin film. High crystal quality and low-cost thin film of Ir(1 1 1) can be grown on Si(1 1 1) with an yttria-stabilized zirconia (YSZ) buffer layers [19][20][21]. Recently, graphene of high structural quality has been prepared on this new substrate and characterized by scanning tunnelling microscopy (STM) and low energy electron diffraction (LEED) [22,23]. ...
This work focuses on the epitaxial growth process and kinetics of silicon-based Ir/yttria-stabilized zirconia (YSZ) bilayer films prepared by PLD and magnetron sputtering, including the sequential growth of a YSZ film, an Ir seed layer, and an Ir film on a silicon substrate. The results show that for growing YSZ films by PLD at the target-to-substrate distance of 60 mm, the higher the laser energy density and the oxygen pressure, the lower the laser frequency, the higher the epitaxy quality. In addition, for the preparation of the Ir-seed layer by PLD, background gas decelerates the plasma and suppresses self-sputtering and implantation, which is beneficial to obtaining a high-quality seed layer. In particular, introducing a small amount of oxygen is able to improve the epitaxial quality of the films during the growth of Ir-seed and Ir films, and the mechanism of oxygen is proposed. For the growth of Ir films by magnetron sputtering, the growth rate and the energy of argon ions increased with the increase of RF power, while the epitaxial quality is the opposite. The kinetic theory and simulation results demonstrate that this trend is mainly caused by the increase in the number of particles reaching the substrate without significantly increasing the distribution of high-energy particles. Finally, high-quality Ir (100) films on the YSZ (100) with a fresh surface are prepared. It is shown that increasing the epitaxy quality of the YSZ (100) film and the growth temperature of the Ir film can improve the quality of Ir film. Based on our results, the growth of Ir films on YSZ films with fresh surfaces resulting from in-situ argon plasma cleaning provides a new idea for preparing Ir (100)/YSZ (100) bilayer epitaxial film.
Iridium (Ir) films were deposited on yttrium stabilized zirconia (YSZ) substrates by bias-assisted RF magnetron sputtering. The orientation of Ir films can be orientally modulated by negative substrate bias, and then used for diamond nucleation and growth by microwave plasma chemical vapor deposition technology combined with the bias-enhanced nucleation process. The morphology, microstructure and composition of Ir and diamond films were investigated by scanning electron microscope (SEM), atomic force microscope, X-ray diffraction (XRD), Raman and X-ray photoelectron spectroscopy. The morphological characterizations of Ir films showed that the grain shape was more uniform and regular symmetry with the increase of substrate negative bias voltage. XRD showed that the crystal orientation of the Ir films changed from (1 1 1) texture to (2 0 0) texture, and the optimal (2 0 0) preferred orientation is achieved under a substrate bias of −150 V. SEM images of diamond films exhibited that the diamond islands gradually merge into continuous diamond films and present a high degree of epitaxy. The crystalline quality of the central region of the epitaxial diamond films is higher than that of the edge region due to the substrate temperature gradient. The Raman full width at half maximum of the central region was 5.75 cm⁻¹.
We demonstrate the growth of the highest-quality 2-in-diameter (001) diamond on a (112¯0) sapphire substrate. To elucidate the mechanism of the formation of this diamond, we observe the surfaces of the Ir buffer layer, bias-enhanced nucleation (BEN)-treated Ir surfaces, and diamond surfaces grown at the initial stage with different diamond growth times on both the (112¯0) sapphire and (001) MgO substrates. The (001) Ir buffer surface exhibits atomically flat terraces with atomic steps. However, after the BEN treatment, a ridge surface appears and diamond island growth begins from the bottom of the ridges during the diamond chemical vapor deposition growth. On Ir/sapphire, quadrangular-pyramid diamond three-dimensional (3D) islands with {111} sidewalls grow first. Subsequently, the sidewall facet changes from {111} to {011} and preferentially coalesced in the 〈010〉 direction. In contrast, on Ir/MgO, quadrangular-pyramid diamond 3D islands with {111} sidewalls grow first. Subsequently, the columnar growth proceeds. Consequently, the coalescence of diamond 3D islands on Ir/MgO requires a longer time than that on Ir/sapphire. We detect Ir on the top and sidewalls of the diamond 3D islands. This suggests that the Ir buffer can promote diamond growth at the initial stage as a catalytic effect. The X-ray rocking curve full width at half maximum for the 311 Bragg reflection is roughly proportional to the etch pit density. This is in accord with the transmission electron microscopy observation that the etch pits are of point-bottom-type and originated from a pure edge-type dislocation.
This chapter presents a direct low-temperature bonding technology for the fabrication of AlGaN/GaN diodes and transistors on poly- and single crystalline diamond substrates. We explain how an aluminum nitride layer reacts with water to form a 30 nm thick electrically insulating and mechanically strong bond with excellent thermal properties. Based on a dissolution and crystallization reaction of various aluminum compounds, an adaptive aluminum hydroxide bond layer without voids is created at the bonding interface. An experimental analysis of AlGaN/GaN Schottky diodes on silicon, poly-, and single crystalline diamond demonstrates a large increase of the saturation currents on diamond, and a thermal resistance of ~ 10–100 m²K/GW is calculated from thermal simulations. A thermal resistance of ~ 15–30 m²K/GW is expected from theoretical considerations based on the 30 nm thickness and an expected thermal conductivity of 1–2 W/m K. 3 GHz load-pull measurements demonstrate a 15% higher Pout on single crystalline diamond as compared to silicon at similar power added efficiency. Additionally, similar performance is measured for a 2- and a 6-finger transistor on diamond, which shows that thermal crosstalk between device fingers is mitigated. A disadvantage in our current technology is identified in the thermally poor stress relief layers grown on Si to accommodate mechanical stress and improve the electrical breakdown voltage.
Die Motivation dieser Arbeit ist es einen Beitrag zur sauberen und effizienten Wasserstoffherstellung mittels photoelektrochemischer Wasserspaltung durch TiO2-basierte Systeme zu leisten. Mit dem Konzept photoelektrochemisch aktiver Mehrschichtsysteme in Nano-Plättchen-Geometrie wurde ein neuer Ansatz entwickelt. Dieses Konzept führt die Vorteile von großen 2-Elektroden-Systemen und von Nanopartikeln in Suspension zusammen. Die Nano-Plättchen haben ein sehr hohes Aspektverhältnis von Schichtdicke zu lateraler Plättchengröße. Sie können wie Nanopartikel in die Suspension gebracht werden und biete damit eine sehr große Oberfläche bei minimalem Materialeinsatz. Nano-Plättchen in Suspension sind weiterhin für eine potentielle industrielle Umsetzung leicht skalierbar. Verschiedene Herstellungsverfahren von Nano-Plättchen-Systemen und deren Eigenschaften wurden in dieser Arbeit untersucht. Hierzu wurden Schichtsysteme aus den Halbleitern TiO2 und Si, sowie Ir als Edelmetallschicht und eine Opferschicht (z.B. CuO oder ZnO) mittels Magnetronsputtern, Laserdeposition und Elektronenstrahlverdampfen hergestellt. Es konnte gezeigt werden, dass sich Nano-Plättchen aus TiO2 bzw. Nano-Plättchen-Systeme aus TiO2, Si und Ir mittels verschiedenen Opferschichten von einem Siliziumsubstrat ablösen und herstellen lassen. Die Schicht zerbricht dabei, so dass beispielsweise 100 nm dicke TiO2-Schichten laterale Durchmesser zwischen 50 bis 500 µm aufweisen. Weiterhin wurde hydriertes TiO2 in einem in-situ RF-Sputterprozess hergestellt und der Wasserstoffgehalt mittels elastischer Rückstreudetektionsanalyse (ERDA) bestimmt. Hydrierte TiO2 Schichten weißen dabei geringere Bandlücken auf, um einen größeren Bereich des solaren Spektrums für die photokatalytische Reaktion zu nutzen. Außerdem wurden plasmonische Au- und Ag-Nanopartikeln mittels sputtern und nachträglicher Wärmebehandlung dünner Schichten hergestellt. Die unterschiedlichen Nano-Plättchen-Systeme wurden untersucht bzgl. Größe, Morphologie, Kristallstruktur, sowie Diffusion innerhalb der Mehrschichtsysteme. Hierzu wurden Untersuchungen mittels Profilometrie, Rutherford-Rückstreu-Spektrometrie (RBS), Röntgendiffraktometrie (XRD), Transmissionselektronenmikroskopie (TEM), Rasterelektronenmikroskopie mit energiedispersiver Röntgenspektroskopie (REM EDX), elektrochemischer Analyse, Ellipsometrie, und Sekundär-Ionen-Massenspektrometrie (SIMS) durchgeführt. Die photokatalytischen Eigenschaften wurden u.a. durch die Zersetzung von Methylenblau untersucht.
Recently, a 2-inch-diameter high-quality (001) diamond wafer was prepared that was heteroepitaxially grown on a (001) Ir buffer layer/(112¯0) A-plane sapphire substrate. Herein, we investigate residual strain and curvature of diamond in a relation with Ir/sapphire substrate and Ir/(001) MgO substrate. Diamond layers grown on Ir substrates have tensile strain, whereas Ir buffer layers on sapphire or MgO substrates have compressive strain. However, the magnitude of these strains in diamond and Ir layers grown on sapphire is considerably lower than that of the same layers grown on MgO. The lower tensile strain in the diamond on Ir/sapphire leads to the growth of a 2-inch-diameter layer without cracking. The convex curvature radius of the diamond layers grown on sapphire is larger than that observed on the MgO substrate. The difference in the coefficients of thermal expansion between the diamond and sapphire is considerably smaller than that of the MgO substrate. The low tensile strain of diamond is caused by the difference in the lattice constants of the diamond and Ir/sapphire substrate.
Due to its unique combination of superior material properties, diamond is often referred to as the ultimate semiconductor material for high‐power electronics. Lack of wafer‐size electronic‐grade single crystals was always considered a crucial bottleneck for the device development and the subsequent transfer to industrial processes. Over the last years, significant progress has been made for the classical high‐pressure high‐temperature (HPHT) method and in particular for the alternative chemical vapor deposition (CVD) technique. This review first briefly describes the HPHT technique which copies the natural formation process working under conditions under which diamond is thermodynamically the stable phase of carbon. It is capable to produce small crystals virtually free of dislocations. The maximum size of available substrates is currently ≈15 × 15 mm 2 . In contrast, CVD growth takes place at more moderate temperatures, at pressures below ambient conditions and far from equilibrium. The general aspects of diamond CVD comprising the gas phase chemistry, different technical approaches for gas phase activation and reactor design are summarized. There are two competing approaches toward single crystal diamond wafers required for electronic applications. Homoepitaxy is performed on highest quality HPHT seed crystals, and various concepts are explored to increase sample dimensions during CVD processes. In contrast, heteroepitaxy involves nucleation and growth on foreign substrates. While homoepitaxial crystals excel with minimum dislocation densities, they are outperformed in terms of size by 3.5‐in.‐diameter diamond wafers synthesized by heteroepitaxy on Ir/YSZ/Si(001). Classical and novel concepts for further defect reduction during CVD growth are discussed.
The effects of deposition time of Ir buffer layer on MgO(100) support layer were investigated during the fabrication of the Ir/MgO compliant substrates with the use of the magnetron sputtering technique. It was observed that the XRD FWHM of Ir(200) decreases rapidly at first and then slowly with increasing deposition time, and the XRD FWHM data versus the deposition time can be fitted into the Logistic curve. For the surface roughness Ra of the Ir buffer layer, it increases firstly and then declines with increasing deposition time; the experimental data of Ra versus the deposition time can be fitted to the Lorentz curve. And the fitting functions corresponding to the two curves can be used for the determination of the deposition time of desired Ir buffer layer quality easily and conveniently, which provides a solid foundation for subsequent heteroepitaxy of high-quality diamond films on the Ir/MgO compliant substrate.
The exceptional mechanical, optical, surface and biocompatibility properties of nanodiamond have gained it much interest. Exhibiting the outstanding bulk properties of diamond at the nanoscale in the form of a film or small particle makes it an inexpensive alternative for many applications.
Nanodiamond is the first comprehensive book on the subject. The book reviews the state of the art of nanodiamond films and particles covering the fundamentals of growth, purification and spectroscopy and some of its diverse applications such as MEMS, drug delivery and biomarkers and biosensing. Specific chapters include the theory of nanodiamond, diamond nucleation, low temperature growth, diamond nanowires, electrochemistry of nanodiamond, nanodiamond flexible implants, and cell labelling with nanodiamond particles.
Edited by a leading expert in nanodiamonds, this is the perfect resource for those new to, and active in, nanodiamond research and those interested in its applications.
For the integration of the thermoelectric and magneto resistive active material (Ca2CoO3)0.62CoO2 (Ca3Co4O9) into silicon technology it is mandatory to provide, apart from diffusion barriers and epitaxial growth, electrical contacts with minimal electrical resistance between this oxide and the metal contact. Epitaxial Ca3Co4O9 thin films were grown by pulsed laser deposition on yttria stabilized zirconia (Y2O3)0.09(ZrO2)0.91 (YSZ) and Ir metallized YSZ buffered single crystalline (001)-Si substrates. X-ray diffraction pole figures of these multilayer systems reveal that the Ca3Co4O9 thin film exhibit a 12-fold in-plane rotational symmetry grown on a (001)-Ir and (001)-YSZ surface, but mutually rotated by 15∘. Ir-Ca3Co4O9 thin film electrical contacts were fabricated by lateral structuring and overgrowth to allow 4-wire measurements for determining the pure Ir-Ca3Co4O9 contact current-voltage characteristic and the area specific contact resistance (ASR). We demonstrate, that the Ca3Co4O9/Ir/YSZ and Ca3Co4O9/YSZ layer system allows to realize low ASR, low in-plane electrical resistivity of the Ca3Co4O9 thin film while maintaining a high Seebeck coefficient.
Diamond is known as the best candidate for power electronics application. Currently the most advance component is the Schottky diode. This device still is already showing promising results but some improvement are still needed. In this thesis, some of them will be proposed and presented from the diamond growth and fabrication steps and after characterized to see if it is suitable. I will first discuss about diamond growth and present new improvements about it that concern the heavily doped layer as well as the non-intentionally doped one. After both part of the characteristics curve (ON and OFF states) have been under investigation for improvement. The ON-state, to increase the current passing through the device by working on the design and on the component parallelization. The OFF-state, by proposing different solution to increase the reverse blocking capabilities. At the same time a transversal ( i.e in the diamond depth) study will be presented to understand how is the electric field inside the diode. Finally, the first heteroepitaxially grown diode on diamond/Ir/SrTiO3/Si will be described and characterized.
Diamond is a wide bandgap material with exceptional physical properties adapted to high power electronic devices. Nevertheless, the fabrication of such material by microwave plasma assisted chemical vapor deposition (MPACVD) is still limited by the small size and high density of structural defects present in the available substrates. The objective of this thesis work is to develop a growth process for heteroepitaxial “electronic-grade” diamond wafer deposition on a multilayer stack of Ir/SrTiO3 /Si (100).
To this aim, optimization of the growth parameters for the heteroepitaxy of intrinsic and boron-doped diamond allowed producing freestanding plates that are at the state-of-the-art. Thus, 200 µm, 490 µm-thick intrinsic films and 200 µm-thick boron doped film ([B]=1018 cm-3) with respectively 9 mm2, 56 mm2, and 9 mm2 areas, were obtained.
In parallel, to limit dislocations propagation in heteroepitaxial diamond, an overgrowth method on macrometric hole arrays was developed, which led to a decrease by an order of magnitude to reach values of 6 × 105 cm-2, the lowest ever reported.
Diamond structuring by the use of silica nanospheres as a mask for the inductively coupled plasma etching was also developed to produce diamond nanopillars with heights ranging from 500 nm to 20 μm. Using this nanopillars incorporation of color centers such as SiV and GeV was locally enhanced.
L’objectif principal de ce travail de thèse concerne la réalisation de films de diamant mosaïque épitaxié sur iridium en mettant en œuvre la nucléation assistée par polarisation. Ce travail s’est donc concentré sur trois axes: (i) Le développement de conditions conduisant à l’hétéroépitaxie du diamant sur l’iridium à partir d'un porte-échantillon spécifique. Deux modes de nucléation ont alors été obtenus, le premier par cristaux isolés, le deuxième par formation de domaines. Ce deuxième mode de nucléation spécifique à l’iridium se différencie par une croissance de diamant non observable par les méthodes conventionnelles au cours de l’étape de nucléation et un taux d’épitaxie des cristaux proche de 100%. (ii) L'étude des mécanismes impliqués dans la formation de l'interface diamant / iridium par une approche séquentielle sous ultra-vide en utilisant des techniques de spectroscopies électroniques (XPS et Auger). Celles-ci ont permis de mettre en évidence la solubilisation du carbone dans l’iridium, l’implantation du carbone dans les premiers nanomètres de l’iridium lors de la nucléation assistée par polarisation et la formation de diamant dès les premiers instants de la croissance. (iii) L'évaluation de la qualité des films de diamant mosaïque épais (≈100µm), réaliser avec un protocole de croissance spécifique se basant sur l’anisotropie des vitesses de croissance des faces (001) et (111). Les caractérisations des films de diamant mosaïque par diffraction des rayons X, par spectroscopie Raman et par cathodoluminescence ont montré que la nucléation par domaines permet d’obtenir des films de diamant mosaïque au niveau de l’état de l’art.
Heteroepitaxial diamond on Si substrates can offer large-area, low-cost diamond wafers for many practical applications. Bias-enhanced nucleation (BEN) has been shown to be one of the best process methods for diamond heteroepitaxy. However, the precise control of the nucleation is difficult because the optimized nucleation time is sensitive to both plasma and substrate conditions. In this study, we performed in situ monitoring for epitaxial diamond nucleation on 3C-SiC/Si substrates that detected the bias current during the BEN process. The bias current was found to decrease at the beginning of the BEN process and then increase during diamond formation. An increase in the bias current of 10% led to epitaxial diamond nucleation with a nucleation density of 10¹⁰ cm⁻². We found from an analytical model that the nucleation occurs at a constant rate in this region. Similar characteristics were obtained on both (001)- and (111)-oriented substrates. This technique was based on the difference between the secondary electron emission coefficient of diamond and that of the material underneath. Thus, this method would be applicable to optimizing the epitaxial diamond nucleation on other substrates or buffer layers in addition to the 3C-SiC/Si studies here.
В книге изложены современное состояние, возможности и перспективы использования алмаза в
различных областях электроники, оптики и техники. Представлено обобщение опыта существующих и
перспективных технических решений по созданию алмазных полупроводниковых пленок, технологии
выращивания высококачественных, высокочистых алмазных монокристаллических пленок, их легирования и получения наноструктур. Описаны электрофизические и фотоэлектрические свойства таких структур. В связи с тем, что основное внимание уделено получению алмазных полупроводниковых пленок методом химического осаждения из газовой фазы (CVD-метод), приведены сведения об основных типах СВЧ реакторов и об организациях (научных группах), работающих на переднем крае технологии и исследований CVD-алмаза. Отмечается уникальность свойств алмаза по сравнению с другими полупроводниковыми материалами для разработки и создания перспективных электронных приборов. Обсуждается основная проблема внедрения алмаза в современные электронику и оптику – отсутствие монокристаллического алмаза в виде пластин большого размера, пригодных для технологии фотолитографии, а также пути ее решения. Книга предназначена для ученых, технологов и инженеров, работающих над созданием оптоэлектронных приборов и устройств полупроводниковой электроники на основе алмаза. Может быть полезна преподавателям, аспирантам и студентам технических вузов, исследователям, инженерам и разработчикам.
Diamond offers a range of unique properties, including wide band of optical transmission, highest thermal conductivity, stiffness, wear resistance and superior electronic properties. Such high-end properties are not found in any other material, so theoretically it can be used in many technological applications. But the shortcoming has been the synthesis of the diamond material in the laboratory for any meaningful use. Although microwave plasma chemical vapour deposited (MPCVD) has been in practice since 1980s for the diamond growth but it is in the past 7-8 years that its potential has been realised by the industry due to capability of MPCVD to deposit diamond, pure and fast, for commercial uses. There are many CVD techniques for growing diamond but among them MPCVD can only make single crystal diamond (SCD) effectively. SCD grown by MPCVD is also superior to other forms of diamond produced in the laboratory. For example, SCD is necessary for the best electronic properties - often outperforming the polycrystalline diamond (PCD), the high pressure high temperature (HPHT) diamond and the natural diamond. In many applications the short lateral dimensions of the lab-grown diamond available is a substantial limitation. Polycrystalline CVD diamond layers grown by hot filament CVD solved this problem of large area growth, but the presence of grain boundaries are not appropriate for many uses. On the other hand, there is still limitation in the area over which SCDs are grown by MPCVD, only upto 10-15 mm lateral sizes could have been achieved so far, while there are recipes which rapidly grow several mm thick bulk SCDs. This lateral size limitation of SCDs is primarily because of the small seed substrate dimension. Although natural and HPHT diamonds may not be suitable for the intended application, still they are routinely used as substrates on which SCD is deposited. But the problem lies in the availability of large area natural SCD seeds which are extremely rare and expensive. Moreover, large diamond substrate plates suitable for CVD diamond growth have not been demonstrated by HPHT because of the associated high economic risk in their fabrication and use. Other than lateral dimension, purity of SCD is also very important for technological use. Natural diamond is often strained and defective, and this causes twins and other problems in the CVD overgrowth or fracture during synthesis. In addition, dislocations which are prevalent in the natural diamond substrate are replicated in the CVD layer, also degrading its electronic properties. HPHT synthetic diamond is also limited in size, and generally is of poorer quality in the larger stones, with inclusions being a major problem. There will be much research interest in the next 10 years for the MPCVD growth of SCD. Purer and bigger SCDs will be tried to grow with faster and reproducible MPCVD recipes. Here the MPCVD growth of SCD is being reviewed keeping in mind its huge technological significance in the next decades or so. Discoveries of the commercial productions of silicon, steel, cement different materials have built modern societies but higher scales will be achieved with the advent of lab-grown diamond.
A new concept for producing freestanding diamond substrate by heteroepitaxy is proposed. Thick diamond growth by heteroepitaxy is often prevented by heteroepitaxial-strain-related substrate bowing as it leads to substrate cracking. The possibility of using diamond microneedles as a mechanism to neglect and/or exploit substrate bowing is discussed in this new concept; the self-breaking effect of the microneedles is proposed as an application to prevent the main bulk diamond layers from cracking. With an aim toward the realization of this concept, the present study shows the first two important experimental verifications through a homoepitaxial experiment with a high-pressure high-temperature diamond substrate: (1) fabrication of high-aspect-ratio diamond microneedles (2 and 100 μm in diameter and length, respectively) by a thermo-chemical etching reaction and (2) overgrowth of diamond on the microneedles with air gaps remaining between microneedles after continuous diamond film overgrowth. We also detail a growth scheme that illustrates how continuous diamond films are created from the microneedles. The strong feasibility for applying the concept in actual heteroepitaxy is suggested through the present study.
CVD diamond film is potentially a very promising material in a series of high technology application fields because of the combination of its excellent electrical, thermal, optical, mechanical, acoustic and electrochemical properties, by which the well-known "Diamond Fever" was induced all over the world in the mid 80s of the 20th century. Although great progress in R&D of CVD diamond films was been achieved in the past 30 years, however, the market size are still not so big as expected. In the present paper the progress in the R&D as well as the commercialization of CVD diamond films is reviewed in detail, in the hope that more and more people will understand, and so as to push forward the process to realize the large scale market applications.
We demonstrate the controlled preparation of heteroepitaxial diamond nano-
and microstructures on silicon wafer based iridium films as hosts for single
color centers. Our approach uses electron beam lithography followed by reactive
ion etching to pattern the carbon layer formed by bias enhanced nucleation on
the iridium surface. In the subsequent chemical vapor deposition process, the
patterned areas evolve into regular arrays of (001) oriented diamond
nano-islands with diameters of <500nm and a height of approx. 60 nm. In the
islands, we identify single SiV color centers with narrow zero phonon lines
down to 1 nm at room temperature.
Diamant besitzt eine Vielzahl extremer Eigenschaften wie hohe mechanische Härte, Bruchfestigkeit, thermische Leitfähigkeit oder breitbandige optische Transparenz. Speziell für elektronische Anwendungen ist Diamant in einkristalliner Qualität erforderlich, wobei das für diesen Einsatz geeignete Material nur in begrenzter Größe (< 1 cm2) künstlich herstellbar ist. Letzteres ist aber nur von technologischer Relevanz, wenn es in Wafergröße zur Verfügung steht. In dieser Arbeit wurde deshalb ein Verfahren entwickelt einkristalline Diamantschichten auf großer Fläche (mehrere 10 cm2) mittels chemischer Gasphasenabscheidung auf einem Fremdsubstrat abzuscheiden. In den letzten Jahren stellte sich das Edelmetall Iridium (Ir) als vielversprechenste Wachstumsoberfläche für die heteroepitaktische Diamantabscheidung heraus. Zu Beginn dieser Arbeit wurde das Iridium als dünne Schicht auf SrTiO3-Einkristallen (1 cm2) mittels Elektronenstrahlverdampfen abgeschieden. Aufgrund des schlechten thermischen Fits zu Diamant (Problem des Abplatzens) ist ein Wachstum von Diamantschichten von mehreren 100 Mikrometern Dicke auf SrTiO3 und anderen Oxideinkristallen nur schwer kontrollierbar. Bezüglich dieser Anforderungen stellt Silizium ein ideales Substrat dar. Die erste Herausforderung bestand nun darin, die Ir(001)-Wachstumsoberfläche auf Siliziumsubstraten zu integrieren. Dazu wurde eine weitere oxidische Pufferschicht benützt um eine Iridium-Silizidbildung zu verhindern. Das Kapitel 5.1 beschäftigt sich zuerst mit der Entwicklung von oxidischen Pufferschichtsystemen. Mittels Laserablation konnten epitaktische YSZ-Filme (yttriumoxid-stabilisiertes Zirkonoxid) auf Silizium mit bislang unerreichter Kristallqualität deponiert werden. Im Kapitel 5.2 wird auf die Deposition der Ir-Filme auf den YSZ-Pufferschichten mittels Elektronenstrahlverdampfen eingegangen. Auch hier wurde ein neues Konzept entwickelt, um einkristalline Wachstumsoberflächen für die nachfolgende Diamantabscheidung bereitzustellen. Durch einen Zweistufenprozess konnten einkristalline Ir-Schichten auf verschiedenen oxidischen Unterlagen (YSZ, SrTiO3, MgO), die typischerweise eine hohe Fehlorientierung (> 1°) aufweisen, präpariert werden. Es wird ein Modell zur Texturverbesserung vorgestellt, das auf der gegenseitigen Ausrichtung der Ir-Kristallite im Anfangsstadium des Wachstums beruht. Die Abscheidung einkristalliner Schichten konnte auch auf andere Metalle (Rhodium, Ruthenium und Platin) übertragen werden. Der kritischste Prozess bei der heteroepitaktischen Diamantabscheidung auf Iridium ist der Nukleationsschritt, der in dieser Arbeit mit dem sog. BEN-Prozess sowohl in einem Mikrowellenplasma als auch mit einer reinen Gleichspannungs-Entladung durchgeführt wurde. Die Phänomene, die man bei der Diamantnukleation auf Iridium beobachtet, widersprechen den Vorstellungen der klassischen Keimbildungsmodelle. So bilden sich die Diamantkeime innerhalb einer ultradünnen Kohlenstoffschicht auf der Ir-Oberfläche unter Bedingungen, unter denen die Volumenphase, d.h. makroskopische Diamantkristallite, durch den Ionenbeschuss bei angelegter Biasspannung geätzt werden. Zudem stellt man direkt nach der Nukleation eine charakteristische Musterbildung und nach einem kurzen Wachstumsschritt eine selbstorganisierte Struktur der Diamantkeime fest. In Kapitel 5.3 werden deshalb die Prozesse bei der Diamantnukleation eingehend untersucht und die Struktur und Verteilung der Diamantkeime aufgeklärt. Auf der Grundlage all dieser Ergebnisse konnte ein Keimbildungsmodell für die Diamantnukleation auf Ir(001) formuliert werden, das die außergewöhnlichen Phänomene bei der Nukleation schlüssig erklärt. Abschließend wird gezeigt, dass eine großflächige Diamantabscheidung auf dem Schichtpaket Ir/YSZ/Si möglich ist.
Growth, electronic properties, and magnetic properties of an Fe monolayer (ML) on an Ir/YSZ/Si(111) multilayer system have been studied using spin-polarized scanning tunneling microscopy. Our experiments reveal a magnetic nano-skyrmion lattice, which is fully equivalent to the magnetic ground state that has previously been observed for the Fe ML on Ir(111) bulk single crystals. In addition, the experiments indicate that the interface-stabilized skyrmion lattice is robust against local atomic lattice distortions induced by multilayer preparation.
The unique properties of diamond have stimulated the study of and search for its applications in many fields, including optics, optoelectronics, electronics, biology, and electrochemistry. Whereas chemical vapor deposition allows the growth of polycrystalline diamond plates more than 200 mm in diameter, most current diamond application technologies require large-size (25 mm and more) single-crystal diamond substrates or films suitable for the photolithography process. This is quite a challenge, because the largest diamond crystals currently available are 10 mm or less in size. This review examines three promising approaches to fabricating large-size diamond single crystals: growing large-size single crystals, the deposition of heteroepitaxial diamond films on single-crystal substrates, and the preparation of composite diamond substrates.
The unique properties of diamond have stimulated the study of and search for its applications in many fields, including optics, optoelectronics, electronics, biology, and electrochemistry. Whereas chemical vapor deposition allows the growth of polycrystalline diamond plates more than 200 mm in diameter, most current diamond application technologies require large-size (25 mm and more) single-crystal diamond substrates or films suitable for the photolithography process. This is quite a challenge, because the largest diamond crystals currently available are 10 mm or less in size. This review examines three promising approaches to fabricating large-size diamond single crystals: growing large-size single crystals, the deposition of heteroepitaxial diamond films on single-crystal substrates, and the preparation of composite diamond substrates.
Using epitaxial SrTiO3 and yttria-stabilized zirconia (YSZ) buffer layers deposited on silicon as a starting point, epitaxial iridium layers were grown by electron-beam evaporation using a two-step growth process with an extremely low initial deposition rate. The iridium layers had in-plane (twist) and out-of-plane (tilt) full widths at half maximum as narrow as 0.08° and 0.15°, respectively, up to an order of magnitude narrower than the underlying SrTiO3 and YSZ layers. SrTiO3 and ZnO films grown on the iridium showed significantly narrower twist and tilt values than without the iridium interlayer, demonstrating a route to improved oxide heteroepitaxy on silicon.
Room-temperature drift mobilities of 4500 square centimeters per volt second for electrons and 3800 square centimeters per volt second for holes have been measured in high-purity single-crystal diamond grown using a chemical vapor deposition process. The low-field drift mobility values were determined by using the time-of-flight technique on thick, intrinsic, freestanding diamond plates and were verified by current-voltage measurements on p-i junction diodes. The improvement of the electronic properties of single-crystal diamond and the reproducibility of those properties are encouraging for research on, and development of, high-performance diamond electronics.
Epitaxial growth of diamond on iridium thin films was performed by direct-current plasma chemical vapor deposition with ion irradiation pretreatment of the substrate. Pyramidal epitaxial diamond particles with a number density of ∼ 10 8cm -2 were grown on the iridium film. The epitaxial relation is written as (100) diamond//(100)iridium and [001] diamond//[001]iridium. Tilting of the epitaxial relation, as occasionally observed for diamond on silicon or beta silicon carbide, is scarcely observed. Erosion,as observed for diamond on nickel substrates, is not observed. The effect of the ion irradiation of the substrate is discussed briefly.
Chemical vapour deposition techniques for fabricating large single-crystal diamond at very high growth rates (to >100 μm/h) have been developed. Single crystals >1 cm in thickness and >10 carats in size can now be produced having a range of optical and mechanical properties for different applications. This diamond can be tailored to possess high fracture toughness, and high-pressure /high-temperature annealing can significantly increase its hardness. The size of the diamonds can be further enlarged to by successive growth on different faces. The process has also been optimized to produce diamond with improved optical properties in the ultraviolet to infrared range, opening prospects for new technological and scientific applications.
For the heteroepitaxial deposition of diamond on silicon using the bias-enhanced nucleation procedure, several different processes contributing to the final misalignment of the layers can be identified: (i) The interface of Si/diamond or Si/SiC and SiC/diamond, respectively. (ii) The growth of individual grains during the biasing process. (iii) The growth competition between differently oriented grains and their coalescence during the growth of thick films. X-ray-diffraction texture studies revealed that the azimuthal alignment is essentially determined by the nucleation step. Oriented nucleation is only possible within a defined time window. Within this time window the azimuthal misalignment shows a characteristic variation depending on the absolute value of the bias voltage. The alignment of the SiC interlayer as measured by synchrotron radiation cannot explain the observed variation. In contrast, texture measurements of thick oriented films after exposure to the bias conditions suggest that the limitation of the process time window for oriented nucleation as well as the variation of misorientation with biasing time can be traced back to the detrimental effect of bias-assisted growth. Based on this mechanism, a model is proposed which allows one (a) to describe the temporal development of the azimuthal misorientation within the process time window, and (b) to estimate the contribution of bias-assisted growth on the misorientation. Finally, some epitaxial diamond films have been deposited on high-quality β-SiC layers. A minimum value of 2.9° for the width of the azimuthal distribution has been found.
A multilayer structure is presented which allows the deposition of high-quality heteroepitaxial diamond films on silicon. After pulsed-laser deposition of a thin yttria-stabilized zirconia (YSZ) layer on silicon, iridium was deposited by e-beam evaporation. Subsequently, diamond nucleation and growth was performed in a chemical vapor deposition setup. The epitaxial orientation relationship measured by x-ray diffraction is diamond(001)[110]∥Ir(001)[110]∥YSZ(001) [110]∥Si(001)[110]. The mosaicity of the diamond films is about an order of magnitude lower than for deposition directly on silicon without buffer layers and nearly reaches the values reported for single-crystal diamond on Ir/SrTiO3. In the effort towards single-crystal diamond wafers, the present solution offers advantages over alternative growth substrates like large-area oxide single crystals due to the low thermal expansion mismatch. © 2004 American Institute of Physics.
The bias‐enhanced nucleation (BEN) technique has been applied to TiC(111) substrates and resulted in deposition of oriented diamond particles. The orientation was observed via scanning electron microscopy. A dense region of oriented particles was not observed on the samples, presumably due to the excessive twinning of the diamond. However, micrographs taken throughout the substrate showed diamond particles having similar orientation with respect to each other. Some of the diamond particles showed evidence of azimuthal twist and tilting, resulting most likely from the ∼21% lattice mismatch. Raman spectra of the diamond crystals show a strong feature at 1332 cm−1, which is indicative of diamond, and smaller features at 1480 and 1602 cm−1 due to sp2‐bonded carbon. © 1995 American Institute of Physics.
In comparison with polycrystalline diamond, single crystal diamond (SC CVD) offers important advantages for applications to electronics and detection. This is because SC CVD diamond exhibits improved charge transport characteristics, specifically a high drift mobility and an increased charge carrier lifetime. So far, only limited studies have dealt with the measurements of electrical transport parameters in SC diamond and majority of these studies were based on industrial material only. In our study we have investigated the electrical transport properties of PE CVD SC diamond grown using relatively high pressure growth conditions and high deposition growth rates (typically 20 μm/hour) by two independent methods: the laser Time-of-Flight (TOF) and the alpha particle TOF techniques. Raman spectroscopy, defect spectroscopy were used to characterize the quality of the samples prepared. (© 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)
Large-area, single-crystalline iridium films are desired for the heteroepitaxial deposition of diamond. In the present work, we studied the potential of SrTiO3 buffer layers for the epitaxial deposition of iridium on silicon. Molecular beam epitaxy (MBE) was used to deposit a 100-nm-thick SrTiO3 layer. On top of this, iridium films were grown by e-beam evaporation. Subsequently, diamond was nucleated by the bias-enhanced nucleation procedure. The epitaxial orientation relationship of the resulting multilayer structure is diamond(001)[110]‖Ir(001)[110]‖SrTiO3(001)[110]‖Si(001)[100]. The Ir/SrTiO3 buffer layers lower the misorientation of the epitaxial diamond films by nearly an order of magnitude as compared to deposition directly on silicon. Oxides like yttria-stabilized zirconia (YSZ) or CeO2/YSZ prepared by pulsed laser deposition (PLD) provide a viable alternative to the MBE-grown SrTiO3. The crystalline quality of the diamond films and their good adhesion on the silicon substrate suggest diamond/Ir/metal-oxide/Si as a promising means to a large-area, single-crystal diamond technology.
Repetition of high rate homoepitaxial growth of diamond by microwave plasma CVD has been successfully applied to the growth of single-crystal diamonds with the thickness as large as 10 mm. By optimizing the shape of the substrate holder, smooth and flat surface morphology suitable for regrowth has been obtained. Using this condition, a 4.65 ct single-crystal diamond with the thickness of 1 cm has been grown on a HPHT synthetic 5 × 5 × 0.7 mm3 seed by 24 times repetition of high rate growth with the average growth rate of 68 μm/h. Also a method to enlarge the size of diamond three-dimensionally by growing on side {100} surface of a thick diamond prepared by this technique has been demonstrated.
Lifetimes at Ga 1-x In x P/GaAs heterojunction interfaces determined by photoluminescence power dependence measurements and diffusion model calculations have been correlated to dislocation densities derived from high‐resolution x‐ray diffraction measurements. The diffusion model calculations are used to determine lifetimes in the region of misfit dislocations by fitting experimental power dependencies of buried‐layer photoluminescence. High‐resolution x‐ray diffraction reveals dislocation densities through the broadening of diffraction peaks due to slight lattice tilts introduced by the dislocations. Lifetimes and dislocation density per dislocation length are correlated to show the functional relationship between the dislocation density and the density of the lifetime‐limiting recombination center at the interface.
Heteroepitaxial diamond films can be grown by bias enhanced nucleation on iridium buffer layers followed by an appropriate textured-growth step. Unlike epitaxial diamond films on silicon, the mosaicity reduction during textured growth includes tilt as well as twist. We conclude that different mechanisms causing the grain coarsening are working in the two cases. It is shown that the principle of evolutionary selection can be excluded as a decisive mechanism in the present films. Merging of neighboring grains by disclination formation yields an alternative explanation, that can convincingly substantiate the differences between the textured growth on iridium and silicon. From a Monte–Carlo type simulation describing the texture evolution due to merging of grains, a simple functional correlation between grain coarsening and mosaicity reduction is deduced. Comparison between simulation and experiment allows one to estimate the contributions of different processes. Finally, the general significance of the present findings for other materials is discussed. © 2002 American Institute of Physics.
It is shown that diamond nucleation on iridium buffer layers followed by an appropriate textured-growth step offers a viable way to realize single-crystal diamond films. Bias-enhanced nucleation on iridium layers results in heteroepitaxial diamond films with highly improved alignment. By a subsequent textured-growth step, the mosaicity can be further reduced for tilt as well as for twist in sharp contrast to former experiments using silicon substrates. Minimum values of 0.17° and 0.38° have been measured for tilt and twist, respectively. Plan view transmission electron microscopy of these films shows that, for low thicknesses (0.6 μm and 8 μm), the films are polycrystalline, consisting of a closed network of grain boundaries. In contrast, at the highest thickness (34 μm) most of the remaining structural defects are concentrated in bands of limited extension. The absence of an interconnected network of grain boundaries shows that the latter films are no longer polycrystalline. © 2001 American Institute of Physics.
Yittria‐stabilized zirconia films have been deposited on Si(100) substrates by ArF (193 nm) and KrF (248 nm) laser ablation. By using KrF radiation, high‐quality (h00) epitaxial films are obtained on bare silicon. Epitaxy, although with slightly worse properties, is also obtained on substrates covered with a native oxide. Films have been deposited by ArF laser ablation over a wide range of substrate temperature and oxygen partial pressure and with variation in substrate surface cleaning. These films are polycrystalline and texture or epitaxy have not been obtained in any case. These results reveal that laser wavelength is a crucial factor in determining film properties. © 1996 American Institute of Physics.
The preparation parameters of epitaxially grown buffer layers on silicon (100) wafers were investigated. We found that an in situ removal of the native amorphous SiO 2 layer from the Si surface is possible, avoiding the etching of the wafer prior to the deposition. YSZ and Y 2 O 3 were chosen as buffer layers for subsequent YBa 2 Cu 3 O 7-x thin‐film deposition. The orientation of the thin films during the deposition process was analyzed by RHEED. Different orientations on the substrates are obtained depending on the evaporation parameters. TEM studies of the interfaces, x‐ray diffraction analysis, and measurements of the superconducting properties were made after the deposition of the films.
Textured diamond films have been deposited on β‐SiC via microwave plasma chemical vapor deposition preceded by an in situ bias pretreatment that enhances nucleation. Approximately 50% of the initial diamond nuclei appear to be aligned with the C(001) planes parallel to the SiC(001), and C[110] directions parallel to the SiC[110] within 3°. The diamond was characterized by Raman spectroscopy and scanning electron microscopy.
Large-scale heteroepitaxial growth of diamond depends critically on the development of a suitable lattice-matched buffer layer and substrate system. Epitaxial (100) iridium films have been grown on terraced, vicinal a -plane (112¯0) α- Al <sub>2</sub> O <sub>3</sub> (sapphire) by electron-beam evaporation. The epitaxial relationship, Ir (100)// Al <sub>2</sub> O <sub>3</sub>(112¯0) with Ir [011]// Al <sub>2</sub> O <sub>3</sub>[11¯00], was determined by x-ray diffraction and electron backscattering diffraction analysis. For a 300-nm thickness of Ir, a (200) rocking curve yielded a linewidth of 0.21°, and the film exhibited a macrostepped surface with low pinhole density. This Ir/sapphire system provides a basis for large-area growth of (100) heteroepitaxial diamond. © 2003 American Institute of Physics.
Diamond thin films have been grown epitaxially on high‐pressure synthesized cubic boron nitride (c‐BN) particles by using dc plasma chemical vapor deposition. At the early growth stage of the film on c‐BN{111} surfaces, the island structure is observed and the number density of islands is about 10<sup>11</sup> cm<sup>-2</sup>. The growth and the coalescence of islands are also found by scanning electron microscopy observation. The continuous film is obtained at the thickness of about 2000 Å and the surface of the film is rather smooth. The Raman peak of the epitaxial diamond film shows the shift toward the lower wave number due to the tensile stress involved in the film.
Zinc oxide nanopillars were grown by a self-catalyzed growth process on an epitaxial Ir/yttria-stabilized zirconia/Si(111) multilayer structure in an optically heated tube furnace. The pillars obtained stand upright parallel to each other with their c -axis perpendicular to the sample surface. Problems due to alloying of Zn with Si are completely avoided, and no irregularities of the pillars in the initial growth phase are found. Cathodoluminescence measurements show narrow linewidths below 700 μ eV due to the excellent crystal quality. Termination of ZnO pillars with a flat metallic iridium layer is an attractive issue towards an optical cavity for laser action.
Highly oriented diamond films have been deposited on mirror-polished single-crystal (100) silicon by microwave plasma chemical vapour deposition. The crystallites are grown with their (001) planes parallel to silicon (001) and their [110] directions parallel to silicon [110], as shown by scanning electron microscopy and X-ray diffraction analysis.
Diamond possesses extraordinary material properties, a result that has given rise to a broad range of scientific and technological applications. This study reports the successful production of high-quality single-crystal diamond with microwave plasma chemical vapor deposition (MPCVD) techniques. The diamond single crystals have smooth, transparent surfaces and other characteristics identical to that of high-pressure, high-temperature synthetic diamond. In addition, the crystals can be produced at growth rates from 50 to 150 mum/h, which is up to 2 orders of magnitude higher than standard processes for making polycrystalline MPCVD diamond. This high-quality single-crystal MPCVD diamond may find numerous applications in electronic devices as high-strength windows and in a new generation of high-pressure instruments requiring large single-crystal anvils.
(Figure Presented) A nanotemplate surface that functions as a regular array of traps for molecules such as naphthalocyanine (see picture) is provided by a nanomesh of hexagonal BN on Rh(111), which has now been identified as a single, complete monolayer. The 2-nm-sized pores form at regions where the layer binds strongly to the underlying metal, while the regular network of mesh wires corresponds to regions where the layer is not bonded to the substrate.
- G Prinz
- A Reiser
- T Röder
- M Schirra
- M Feneberg
- U Röder
- R Sauer
- K Thonke
- S Gsell
- M Schreck
- B Stritzker
G. Prinz, A. Reiser, T. Röder, M. Schirra, M. Feneberg, U. Röder, R. Sauer, K. Thonke, S.
Gsell, M. Schreck, B. Stritzker, Appl. Phys. Lett. 90 (2007) 233115.
- J Isberg
- J Hammersberg
- E Johansson
- T Wikström
- D J Twitchen
- A J Whitehead
- S E Coe
- G A Scarsbrook
J. Isberg, J. Hammersberg, E. Johansson, T. Wikström, D.J. Twitchen, A.J. Whitehead,
S.E. Coe, G.A. Scarsbrook, Science 297 (2002) 1670.
- S Berner
- M Corso
- R Widmer
- O Groening
- R Laskowski
- P Blaha
- K Schwarz
- A Goriachko
- H Over
- S Gsell
- M Schreck
- H Sachdev
- T Greber
- J Osterwalder
S. Berner, M. Corso, R. Widmer, O. Groening, R. Laskowski, P. Blaha, K. Schwarz, A.
Goriachko, H. Over, S. Gsell, M. Schreck, H. Sachdev, T. Greber, J. Osterwalder,
Angew. Chem., Int. Ed. Engl. 46 (2007) 5115.
- K Ohtsuka
- K Suzuki
- A Sawabe
- T Inuzuka
K. Ohtsuka, K. Suzuki, A. Sawabe, T. Inuzuka, Jpn. J. Appl. Phys. 35 (1996) L1072.
Iridium epitaxy on high mosaicity oxides
- Gsell
Gsell et al. Iridium epitaxy on high mosaicity oxides (in preparation).
- M Schreck
- F Hörmann
- H Roll
- J K N Lindner
- B Stritzker
M. Schreck, F. Hörmann, H. Roll, J.K.N. Lindner, B. Stritzker, Appl. Phys. Lett. 78 (2001)
192.
- Th
- J Matthée
- H Wecker
- G Behner
- O Friedl
- K Eibl
- Samwer
Th. Matthée, J. Wecker, H. Behner, G. Friedl, O. Eibl, K. Samwer, Appl. Phys. Lett.
61 (1992) 1240.
- Y Mokuno
- A Chayahara
- Y Soda
- Y Horino
- N Fujimori
Y. Mokuno, A. Chayahara, Y. Soda, Y. Horino, N. Fujimori, Diamond Relat. Mater.
14 (2005) 1743.
- N Tranchant
- M Nesladek
- D Tromson
- Z Remes
- A Bogdan
- P Bergonzo
N. Tranchant, M. Nesladek, D. Tromson, Z. Remes, A. Bogdan, P. Bergonzo, Phys.
Status Solidi, A Appl. Res. 204 (2007) 3023.
- S S Ho
- C S Yan
- Z Liu
- H K Mao
- R J Hemley
S.S. Ho, C.S. Yan, Z. Liu, H.K. Mao, R.J. Hemley, Ind. Diamond Rev. 1 (2006) 28;
- C.-S Yan
- Y K Vohra
- H.-K Mao
- R J Hemley
C.-S. Yan, Y.K. Vohra, H.-K. Mao, R.J. Hemley, Proc. Natl. Acad. Sci. U. S. A. 99 (2002)
12523.
- X Jiang
- C.-P Klages
X. Jiang, C.-P. Klages, Diamond Relat. Mater. 2 (1993) 1112.
- T Bauer
- S Gsell
- M Schreck
- J Goldfuß
- J Lettieri
- D G Schlom
- B Stritzker
T. Bauer, S. Gsell, M. Schreck, J. Goldfuß, J. Lettieri, D.G. Schlom, B. Stritzker,
Diamond Relat. Mater. 14 (2005) 314.
- K.-H Thürer
- M Schreck
- Stritzker
K.-H. Thürer, M. Schreck, Stritzker, Phys. Rev., B 57 (1998) 15454.
- S Gsell
- M Fischer
- R Brescia
- M Schreck
- P Huber
- F Bayer
- B Stritzker
- D G Schlom
S. Gsell, M. Fischer, R. Brescia, M. Schreck, P. Huber, F. Bayer, B. Stritzker, D.G.
Schlom, Appl. Phys. Lett. 91 (2007) 061501.