FIG 1 - uploaded by Harry A. Atwater
Content may be subject to copyright.
RHEED images of MgO. ͑ a ͒ Simulated RHEED pattern of 20 keV electrons at 1.2° grazing incidence along ͓ 100 ͔ from well-textured polycrystalline MgO with effective lateral grain size L ϭ 4 nm, electron penetration depth h ϭ 1 nm, out-of-plane grain orientation distribution ⌬ ϭ 7°, and in-plane orientation distribution ⌬ ϭ 14°. The qualitative effects of these parameters upon the RHEED spot shapes and relative intensities are indi- cated. ͑ b ͒ Experimental RHEED image from IBAD MgO showing the diffraction spots that are simultaneously analyzed to measure effective grain size and out-of-plane orientation distribution. The diffraction pattern is characterized by measuring the diffraction spot widths along directions that are dependent on effective grain size, electron penetration depth, and out-of- plane orientation distribution. As an example, these directions are marked on the ͑ 024 ͒ diffraction spot.
Source publication
Analysis of biaxial texture of MgO films grown by ion-beam-assisted deposition (IBAD) has been performed using a quantitative reflection high-energy electron diffraction (RHEED) based method. MgO biaxial texture is determined by analysis of diffraction spot shapes from single RHEED images, and by measuring the width of RHEED in-plane rocking curves...
Contexts in source publication
Context 1
... of biaxially textured MgO thin films is techno- logically interesting because it may provide a suitable path for silicon integration of many important single crystal-like perovskite oxide thin films on top of the amorphous dielec- tric thin films present at the back end of a typical integrated circuit fabrication process. This is accomplished by using ion beam-assisted deposition ͑ IBAD ͒ to create biaxially textured films ͑ i.e., polycrystalline films with a preferred in-plane and out-of-plane grain orientation ͒ on amorphous substrates. Film functionality often depends on control of both the out- of-plane grain orientation distribution ͓ the full width at half maximum ͑ FWHM ͒ is designated as ⌬͔ and in-plane grain orientation distribution ͑ FWHM is designated as ⌬͒ . Some highly aligned biaxially textured oxide materials can exhibit similar functionality to single crystalline films. For example, biaxially textured superconductors such as YBa 2 Cu 3 O 7 Ϫ x have been reported to have critical current densities ap- proaching those of single crystalline films, while randomly oriented polycrystalline films exhibit much lower critical current densities. 1 Biaxially textured piezoelectric films with 90° domain rotations are also expected to have flexing char- acteristics similar to those of single crystalline piezoelectric films, while randomly oriented polycrystalline piezoelectric films experience significant degradation of translational range of motion. Incorporation of biaxially textured piezoelectric films with silicon integrated circuits would enable new types of ferroelectric actuators for microelectrical me- chanical systems ͑ MEMS ͒ . Previous work has shown that piezoelectric materials like BaTiO 3 and Pb ͑ Zr,Ti ͒ O 3 can be deposited heteroepitaxially onto single crystal MgO ͑ 001 ͒ 2,3 and even Si ͑ 001 ͒ . 4 However, conventional silicon integrated circuit processing employs extensive hydrogen passivation, which degrades ferroelectrics like Pb ͑ Zr,Ti ͒ O 3 and BaTiO 3 . It is therefore desirable to monolithically integrate piezoelectric materials following integrated circuit fabrication. Wang et al. demonstrated that IBAD MgO grown on amorphous Si 3 N 4 develops narrow biaxial texture in films only 11 nm thick. 5 By eliminating the requirement for a pre-existing het- eroepitaxial template, IBAD provides an opportunity to in- corporate piezoelectric materials on top of amorphous dielec- tric films in silicon integrated circuits during the back-end processing. The performance of piezoelectric MEMS is likely to de- pend on the biaxial texture inherited from the MgO substrate. Previous efforts to optimize the biaxial texture of IBAD MgO have been impeded by the ex situ nature of conventional biaxial texture analysis techniques ͓ transmission electron microscopy ͑ TEM ͒ or x-ray diffraction ͑ XRD ͔͒ . Be- cause the biaxial texture develops within 11 nm of growth, XRD cannot resolve crystallographic texture unless the x-ray source has synchrotron brightness. For these same reasons, the IBAD biaxial texturing mechanisms are difficult to inves- tigate. To circumvent these obstacles, we have developed a reflection high-energy electron diffraction ͑ RHEED ͒ based method for quantitative in situ biaxial texture analysis of MgO. RHEED has been previously used to analyze the out- of-plane texture for CoCr alloys, assuming the grains were not large enough to affect the RHEED pattern. 6 The small grain size of IBAD MgO films ͑ as small as 10 nm ͒ necessi- tate that we deconvolute the effects of grain size from the effects of out-of-plane orientation distribution for accurate texture distribution measurements. We also demonstrate the ability to measure the in-plane orientation distribution. Be- cause RHEED is sensitive to films as thin as 3 nm thick, we have the capability of analyzing the biaxial texture development during the earliest stages of film growth. 7 This analysis capability enables greater understanding of IBAD MgO biaxial texture development and how to optimize the MgO biaxial texture. Our RHEED-based biaxial texture analysis employs a previously reported kinematical electron scattering model. 8 These calculations predict that spot shapes are sensitive to the film microstructure, as shown in Fig. 1 ͑ a ͒ . Diffraction spot width and height are inversely proportional to the effective grain size and electron penetration depth, respectively. The width of the diffraction spot in the direction perpendicular to the location of the through spot, the nondiffracted electron beam, is directly proportional to the out-of-plane grain orientation distribution ͑⌬͒ . We therefore characterize RHEED patterns, whether calculated using a computer simulation or from an experiment, by cutting across the diffraction spots along the previously mentioned directions, and measuring the FWHM of these cuts, as shown in Fig. 1 ͑ b ͒ . We call this method ‘‘single image analysis.’’ All diffraction spots shown in Fig. 1 ͑ b ͒ are analyzed simultaneously, and then compared to calculated RHEED pattern measurements using a lookup table. The lookup table was generated by calculating the RHEED pattern for all relevant combinations of effective grains size ͑ 4 –25 nm ͒ , electron penetration depth ͑ 2.5–10 nm ͒ , and out-of-plane orientation distribution ͑ 0°–20° FWHM ͒ , and measuring the FWHM of cuts across the six diffraction spots shown in Fig. 1 ͑ b ͒ in the directions where the RHEED pattern is sensitive to electron penetration depth, grain size, and out-of-plane orientation distribution. The effects of the in-plane orientation distribution on diffraction spot shapes are negligible, so the in-plane orientation distribution FWHM was set to 10° for lookup table calculations, which is a typical value for in-plane orientation distributions observed in IBAD MgO. Lookup table entries exist for all combinations of effective grain size, electron penetration depth, and out-of-plane orientation distribution, and contain the measurements of the FWHM of the cuts across each previously specified spot along the previously specified directions. The film effective grain size, electron penetration depth, and out-of-plane orientation distribution are determined by comparing the FWHM of experimental RHEED pattern diffraction spot cuts with the FWHM of the spot cuts in the lookup table. For each lookup table entry, the experimentally measured FWHM of each spot cut is sub- tracted from the lookup table FWHM of the same spot in the same direction. The differences between the experimental and lookup table measurements are then individually squared before being added together to yield a total sum of the square errors measurement for that lookup table entry. The sum of the squared errors is calculated for every lookup table entry, and the microstructural parameters are determined as the simulated combination of electron penetration depth, effective grain size, and out-of-plane orientation distribution that yields the smallest sum of the squared errors. Even though the kinematical electron scattering calculations predict that the relative intensities of diffraction spots along the ͑ 00 ͒ , ͑ 02 ͒ , and ͑ 04 ͒ Bragg rods are correlated to the in-plane orientation distribution, it is not a very sensitive measurement. Therefore, RHEED in-plane rocking curves are used to measure the in-plane orientation distribution. RHEED in-plane rocking curves are constructed by rotating the sample around the surface normal and recording the maximum intensity for each diffraction spot, minus the average background intensity, for each angle ͑ the angle between the nominal 100 zone axis and the projection of the incident electron beam on the sample surface ͒ . The resulting intensity distributions are characterized by the FWHM. To experimentally measure in-plane grain orientation distribution ͑⌬͒ , the FWHM of RHEED in-plane rocking curves 9 from the ͑ 024 ͒ , ͑ 026 ͒ , ͑ 044 ͒ , and ͑ 046 ͒ diffraction spots are measured simultaneously and compared to the FWHM of calculated in-plane rocking curves in another lookup table. As for the single image analysis, a lookup table was generated by calculating the FWHM of diffraction spot in-plane RHEED rocking curves for all relevant film param- eter combinations, i.e., effective grain size ͑ 4 –25 nm ͒ , out- of-plane orientation distribution ͑ 0°–20° FWHM ͒ , and in- plane orientation distribution ͑ 0°–30° FWHM ͒ . In-plane rocking curve FWHM was calculated to be in- dependent of the electron penetration depth, so it was set to 5 nm, the value most often measured using single image analysis at this electron energy and incidence angle. Each lookup table entry was indexed by its unique combination of the relevant film parameters ͑ grain size, out-of-plane orientation distribution, and in-plane orientation distribution ͒ and con- tained the FWHM of the rocking curves from the ͑ 024 ͒ , ͑ 026 ͒ , ͑ 044 ͒ , and ͑ 046 ͒ diffraction spots. The in-plane orientation distribution is determined by searching the lookup table for the simulation that has RHEED in-plane rocking curves that most closely match the experimental rocking curves for all four diffraction spots. The FWHM of the in- plane rocking curves are highly correlated with the in-plane orientation distribution, however, the rocking curve FWHM is also convoluted with the effective grain size and out-of- plane orientation distribution. Therefore, to accurately measure in-plane orientation distribution using in-plane rocking curves, the effective grain size and out-of-plane orientation distribution is first measured using single image analysis as described above. The subsequent comparison between the experimental and simulated FWHM of the RHEED in-plane rocking curves in the lookup tables are restricted to simulations with the effective grain size and out-of-plane ...
Context 2
... of biaxially textured MgO thin films is techno- logically interesting because it may provide a suitable path for silicon integration of many important single crystal-like perovskite oxide thin films on top of the amorphous dielec- tric thin films present at the back end of a typical integrated circuit fabrication process. This is accomplished by using ion beam-assisted deposition ͑ IBAD ͒ to create biaxially textured films ͑ i.e., polycrystalline films with a preferred in-plane and out-of-plane grain orientation ͒ on amorphous substrates. Film functionality often depends on control of both the out- of-plane grain orientation distribution ͓ the full width at half maximum ͑ FWHM ͒ is designated as ⌬͔ and in-plane grain orientation distribution ͑ FWHM is designated as ⌬͒ . Some highly aligned biaxially textured oxide materials can exhibit similar functionality to single crystalline films. For example, biaxially textured superconductors such as YBa 2 Cu 3 O 7 Ϫ x have been reported to have critical current densities ap- proaching those of single crystalline films, while randomly oriented polycrystalline films exhibit much lower critical current densities. 1 Biaxially textured piezoelectric films with 90° domain rotations are also expected to have flexing char- acteristics similar to those of single crystalline piezoelectric films, while randomly oriented polycrystalline piezoelectric films experience significant degradation of translational range of motion. Incorporation of biaxially textured piezoelectric films with silicon integrated circuits would enable new types of ferroelectric actuators for microelectrical me- chanical systems ͑ MEMS ͒ . Previous work has shown that piezoelectric materials like BaTiO 3 and Pb ͑ Zr,Ti ͒ O 3 can be deposited heteroepitaxially onto single crystal MgO ͑ 001 ͒ 2,3 and even Si ͑ 001 ͒ . 4 However, conventional silicon integrated circuit processing employs extensive hydrogen passivation, which degrades ferroelectrics like Pb ͑ Zr,Ti ͒ O 3 and BaTiO 3 . It is therefore desirable to monolithically integrate piezoelectric materials following integrated circuit fabrication. Wang et al. demonstrated that IBAD MgO grown on amorphous Si 3 N 4 develops narrow biaxial texture in films only 11 nm thick. 5 By eliminating the requirement for a pre-existing het- eroepitaxial template, IBAD provides an opportunity to in- corporate piezoelectric materials on top of amorphous dielec- tric films in silicon integrated circuits during the back-end processing. The performance of piezoelectric MEMS is likely to de- pend on the biaxial texture inherited from the MgO substrate. Previous efforts to optimize the biaxial texture of IBAD MgO have been impeded by the ex situ nature of conventional biaxial texture analysis techniques ͓ transmission electron microscopy ͑ TEM ͒ or x-ray diffraction ͑ XRD ͔͒ . Be- cause the biaxial texture develops within 11 nm of growth, XRD cannot resolve crystallographic texture unless the x-ray source has synchrotron brightness. For these same reasons, the IBAD biaxial texturing mechanisms are difficult to inves- tigate. To circumvent these obstacles, we have developed a reflection high-energy electron diffraction ͑ RHEED ͒ based method for quantitative in situ biaxial texture analysis of MgO. RHEED has been previously used to analyze the out- of-plane texture for CoCr alloys, assuming the grains were not large enough to affect the RHEED pattern. 6 The small grain size of IBAD MgO films ͑ as small as 10 nm ͒ necessi- tate that we deconvolute the effects of grain size from the effects of out-of-plane orientation distribution for accurate texture distribution measurements. We also demonstrate the ability to measure the in-plane orientation distribution. Be- cause RHEED is sensitive to films as thin as 3 nm thick, we have the capability of analyzing the biaxial texture development during the earliest stages of film growth. 7 This analysis capability enables greater understanding of IBAD MgO biaxial texture development and how to optimize the MgO biaxial texture. Our RHEED-based biaxial texture analysis employs a previously reported kinematical electron scattering model. 8 These calculations predict that spot shapes are sensitive to the film microstructure, as shown in Fig. 1 ͑ a ͒ . Diffraction spot width and height are inversely proportional to the effective grain size and electron penetration depth, respectively. The width of the diffraction spot in the direction perpendicular to the location of the through spot, the nondiffracted electron beam, is directly proportional to the out-of-plane grain orientation distribution ͑⌬͒ . We therefore characterize RHEED patterns, whether calculated using a computer simulation or from an experiment, by cutting across the diffraction spots along the previously mentioned directions, and measuring the FWHM of these cuts, as shown in Fig. 1 ͑ b ͒ . We call this method ‘‘single image analysis.’’ All diffraction spots shown in Fig. 1 ͑ b ͒ are analyzed simultaneously, and then compared to calculated RHEED pattern measurements using a lookup table. The lookup table was generated by calculating the RHEED pattern for all relevant combinations of effective grains size ͑ 4 –25 nm ͒ , electron penetration depth ͑ 2.5–10 nm ͒ , and out-of-plane orientation distribution ͑ 0°–20° FWHM ͒ , and measuring the FWHM of cuts across the six diffraction spots shown in Fig. 1 ͑ b ͒ in the directions where the RHEED pattern is sensitive to electron penetration depth, grain size, and out-of-plane orientation distribution. The effects of the in-plane orientation distribution on diffraction spot shapes are negligible, so the in-plane orientation distribution FWHM was set to 10° for lookup table calculations, which is a typical value for in-plane orientation distributions observed in IBAD MgO. Lookup table entries exist for all combinations of effective grain size, electron penetration depth, and out-of-plane orientation distribution, and contain the measurements of the FWHM of the cuts across each previously specified spot along the previously specified directions. The film effective grain size, electron penetration depth, and out-of-plane orientation distribution are determined by comparing the FWHM of experimental RHEED pattern diffraction spot cuts with the FWHM of the spot cuts in the lookup table. For each lookup table entry, the experimentally measured FWHM of each spot cut is sub- tracted from the lookup table FWHM of the same spot in the same direction. The differences between the experimental and lookup table measurements are then individually squared before being added together to yield a total sum of the square errors measurement for that lookup table entry. The sum of the squared errors is calculated for every lookup table entry, and the microstructural parameters are determined as the simulated combination of electron penetration depth, effective grain size, and out-of-plane orientation distribution that yields the smallest sum of the squared errors. Even though the kinematical electron scattering calculations predict that the relative intensities of diffraction spots along the ͑ 00 ͒ , ͑ 02 ͒ , and ͑ 04 ͒ Bragg rods are correlated to the in-plane orientation distribution, it is not a very sensitive measurement. Therefore, RHEED in-plane rocking curves are used to measure the in-plane orientation distribution. RHEED in-plane rocking curves are constructed by rotating the sample around the surface normal and recording the maximum intensity for each diffraction spot, minus the average background intensity, for each angle ͑ the angle between the nominal 100 zone axis and the projection of the incident electron beam on the sample surface ͒ . The resulting intensity distributions are characterized by the FWHM. To experimentally measure in-plane grain orientation distribution ͑⌬͒ , the FWHM of RHEED in-plane rocking curves 9 from the ͑ 024 ͒ , ͑ 026 ͒ , ͑ 044 ͒ , and ͑ 046 ͒ diffraction spots are measured simultaneously and compared to the FWHM of calculated in-plane rocking curves in another lookup table. As for the single image analysis, a lookup table was generated by calculating the FWHM of diffraction spot in-plane RHEED rocking curves for all relevant film param- eter combinations, i.e., effective grain size ͑ 4 –25 nm ͒ , out- of-plane orientation distribution ͑ 0°–20° FWHM ͒ , and in- plane orientation distribution ͑ 0°–30° FWHM ͒ . In-plane rocking curve FWHM was calculated to be in- dependent of the electron penetration depth, so it was set to 5 nm, the value most often measured using single image analysis at this electron energy and incidence angle. Each lookup table entry was indexed by its unique combination of the relevant film parameters ͑ grain size, out-of-plane orientation distribution, and in-plane orientation distribution ͒ and con- tained the FWHM of the rocking curves from the ͑ 024 ͒ , ͑ 026 ͒ , ͑ 044 ͒ , and ͑ 046 ͒ diffraction spots. The in-plane orientation distribution is determined by searching the lookup table for the simulation that has RHEED in-plane rocking curves that most closely match the experimental rocking curves for all four diffraction spots. The FWHM of the in- plane rocking curves are highly correlated with the in-plane orientation distribution, however, the rocking curve FWHM is also convoluted with the effective grain size and out-of- plane orientation distribution. Therefore, to accurately measure in-plane orientation distribution using in-plane rocking curves, the effective grain size and out-of-plane orientation distribution is first measured using single image analysis as described above. The subsequent comparison between the experimental and simulated FWHM of the RHEED in-plane rocking curves in the lookup tables are restricted to simulations with the effective grain size and out-of-plane orientation distribution measured using single image analysis. Experimental RHEED in-plane rocking ...
Context 3
... ͑ i.e., polycrystalline films with a preferred in-plane and out-of-plane grain orientation ͒ on amorphous substrates. Film functionality often depends on control of both the out- of-plane grain orientation distribution ͓ the full width at half maximum ͑ FWHM ͒ is designated as ⌬͔ and in-plane grain orientation distribution ͑ FWHM is designated as ⌬͒ . Some highly aligned biaxially textured oxide materials can exhibit similar functionality to single crystalline films. For example, biaxially textured superconductors such as YBa 2 Cu 3 O 7 Ϫ x have been reported to have critical current densities ap- proaching those of single crystalline films, while randomly oriented polycrystalline films exhibit much lower critical current densities. 1 Biaxially textured piezoelectric films with 90° domain rotations are also expected to have flexing char- acteristics similar to those of single crystalline piezoelectric films, while randomly oriented polycrystalline piezoelectric films experience significant degradation of translational range of motion. Incorporation of biaxially textured piezoelectric films with silicon integrated circuits would enable new types of ferroelectric actuators for microelectrical me- chanical systems ͑ MEMS ͒ . Previous work has shown that piezoelectric materials like BaTiO 3 and Pb ͑ Zr,Ti ͒ O 3 can be deposited heteroepitaxially onto single crystal MgO ͑ 001 ͒ 2,3 and even Si ͑ 001 ͒ . 4 However, conventional silicon integrated circuit processing employs extensive hydrogen passivation, which degrades ferroelectrics like Pb ͑ Zr,Ti ͒ O 3 and BaTiO 3 . It is therefore desirable to monolithically integrate piezoelectric materials following integrated circuit fabrication. Wang et al. demonstrated that IBAD MgO grown on amorphous Si 3 N 4 develops narrow biaxial texture in films only 11 nm thick. 5 By eliminating the requirement for a pre-existing het- eroepitaxial template, IBAD provides an opportunity to in- corporate piezoelectric materials on top of amorphous dielec- tric films in silicon integrated circuits during the back-end processing. The performance of piezoelectric MEMS is likely to de- pend on the biaxial texture inherited from the MgO substrate. Previous efforts to optimize the biaxial texture of IBAD MgO have been impeded by the ex situ nature of conventional biaxial texture analysis techniques ͓ transmission electron microscopy ͑ TEM ͒ or x-ray diffraction ͑ XRD ͔͒ . Be- cause the biaxial texture develops within 11 nm of growth, XRD cannot resolve crystallographic texture unless the x-ray source has synchrotron brightness. For these same reasons, the IBAD biaxial texturing mechanisms are difficult to inves- tigate. To circumvent these obstacles, we have developed a reflection high-energy electron diffraction ͑ RHEED ͒ based method for quantitative in situ biaxial texture analysis of MgO. RHEED has been previously used to analyze the out- of-plane texture for CoCr alloys, assuming the grains were not large enough to affect the RHEED pattern. 6 The small grain size of IBAD MgO films ͑ as small as 10 nm ͒ necessi- tate that we deconvolute the effects of grain size from the effects of out-of-plane orientation distribution for accurate texture distribution measurements. We also demonstrate the ability to measure the in-plane orientation distribution. Be- cause RHEED is sensitive to films as thin as 3 nm thick, we have the capability of analyzing the biaxial texture development during the earliest stages of film growth. 7 This analysis capability enables greater understanding of IBAD MgO biaxial texture development and how to optimize the MgO biaxial texture. Our RHEED-based biaxial texture analysis employs a previously reported kinematical electron scattering model. 8 These calculations predict that spot shapes are sensitive to the film microstructure, as shown in Fig. 1 ͑ a ͒ . Diffraction spot width and height are inversely proportional to the effective grain size and electron penetration depth, respectively. The width of the diffraction spot in the direction perpendicular to the location of the through spot, the nondiffracted electron beam, is directly proportional to the out-of-plane grain orientation distribution ͑⌬͒ . We therefore characterize RHEED patterns, whether calculated using a computer simulation or from an experiment, by cutting across the diffraction spots along the previously mentioned directions, and measuring the FWHM of these cuts, as shown in Fig. 1 ͑ b ͒ . We call this method ‘‘single image analysis.’’ All diffraction spots shown in Fig. 1 ͑ b ͒ are analyzed simultaneously, and then compared to calculated RHEED pattern measurements using a lookup table. The lookup table was generated by calculating the RHEED pattern for all relevant combinations of effective grains size ͑ 4 –25 nm ͒ , electron penetration depth ͑ 2.5–10 nm ͒ , and out-of-plane orientation distribution ͑ 0°–20° FWHM ͒ , and measuring the FWHM of cuts across the six diffraction spots shown in Fig. 1 ͑ b ͒ in the directions where the RHEED pattern is sensitive to electron penetration depth, grain size, and out-of-plane orientation distribution. The effects of the in-plane orientation distribution on diffraction spot shapes are negligible, so the in-plane orientation distribution FWHM was set to 10° for lookup table calculations, which is a typical value for in-plane orientation distributions observed in IBAD MgO. Lookup table entries exist for all combinations of effective grain size, electron penetration depth, and out-of-plane orientation distribution, and contain the measurements of the FWHM of the cuts across each previously specified spot along the previously specified directions. The film effective grain size, electron penetration depth, and out-of-plane orientation distribution are determined by comparing the FWHM of experimental RHEED pattern diffraction spot cuts with the FWHM of the spot cuts in the lookup table. For each lookup table entry, the experimentally measured FWHM of each spot cut is sub- tracted from the lookup table FWHM of the same spot in the same direction. The differences between the experimental and lookup table measurements are then individually squared before being added together to yield a total sum of the square errors measurement for that lookup table entry. The sum of the squared errors is calculated for every lookup table entry, and the microstructural parameters are determined as the simulated combination of electron penetration depth, effective grain size, and out-of-plane orientation distribution that yields the smallest sum of the squared errors. Even though the kinematical electron scattering calculations predict that the relative intensities of diffraction spots along the ͑ 00 ͒ , ͑ 02 ͒ , and ͑ 04 ͒ Bragg rods are correlated to the in-plane orientation distribution, it is not a very sensitive measurement. Therefore, RHEED in-plane rocking curves are used to measure the in-plane orientation distribution. RHEED in-plane rocking curves are constructed by rotating the sample around the surface normal and recording the maximum intensity for each diffraction spot, minus the average background intensity, for each angle ͑ the angle between the nominal 100 zone axis and the projection of the incident electron beam on the sample surface ͒ . The resulting intensity distributions are characterized by the FWHM. To experimentally measure in-plane grain orientation distribution ͑⌬͒ , the FWHM of RHEED in-plane rocking curves 9 from the ͑ 024 ͒ , ͑ 026 ͒ , ͑ 044 ͒ , and ͑ 046 ͒ diffraction spots are measured simultaneously and compared to the FWHM of calculated in-plane rocking curves in another lookup table. As for the single image analysis, a lookup table was generated by calculating the FWHM of diffraction spot in-plane RHEED rocking curves for all relevant film param- eter combinations, i.e., effective grain size ͑ 4 –25 nm ͒ , out- of-plane orientation distribution ͑ 0°–20° FWHM ͒ , and in- plane orientation distribution ͑ 0°–30° FWHM ͒ . In-plane rocking curve FWHM was calculated to be in- dependent of the electron penetration depth, so it was set to 5 nm, the value most often measured using single image analysis at this electron energy and incidence angle. Each lookup table entry was indexed by its unique combination of the relevant film parameters ͑ grain size, out-of-plane orientation distribution, and in-plane orientation distribution ͒ and con- tained the FWHM of the rocking curves from the ͑ 024 ͒ , ͑ 026 ͒ , ͑ 044 ͒ , and ͑ 046 ͒ diffraction spots. The in-plane orientation distribution is determined by searching the lookup table for the simulation that has RHEED in-plane rocking curves that most closely match the experimental rocking curves for all four diffraction spots. The FWHM of the in- plane rocking curves are highly correlated with the in-plane orientation distribution, however, the rocking curve FWHM is also convoluted with the effective grain size and out-of- plane orientation distribution. Therefore, to accurately measure in-plane orientation distribution using in-plane rocking curves, the effective grain size and out-of-plane orientation distribution is first measured using single image analysis as described above. The subsequent comparison between the experimental and simulated FWHM of the RHEED in-plane rocking curves in the lookup tables are restricted to simulations with the effective grain size and out-of-plane orientation distribution measured using single image analysis. Experimental RHEED in-plane rocking curves and single image analyses were performed on 5- to 11-nm-thick IBAD MgO films. MgO was deposited, at room temperature, on amorphous Si 3 N 4 by electron beam evaporation at deposition rates ranging from 1.7 to 3.1 A/s, as measured by a quartz crystal monitor. Ion irradiation during MgO growth was carried out with 750 eV Ar ϩ ions at 45° incidence angle. Ion/MgO flux ratios were varied between 0.33 and 0.58. A single crystal ...
Context 4
... distribution FWHM was set to 10° for lookup table calculations, which is a typical value for in-plane orientation distributions observed in IBAD MgO. Lookup table entries exist for all combinations of effective grain size, electron penetration depth, and out-of-plane orientation distribution, and contain the measurements of the FWHM of the cuts across each previously specified spot along the previously specified directions. The film effective grain size, electron penetration depth, and out-of-plane orientation distribution are determined by comparing the FWHM of experimental RHEED pattern diffraction spot cuts with the FWHM of the spot cuts in the lookup table. For each lookup table entry, the experimentally measured FWHM of each spot cut is sub- tracted from the lookup table FWHM of the same spot in the same direction. The differences between the experimental and lookup table measurements are then individually squared before being added together to yield a total sum of the square errors measurement for that lookup table entry. The sum of the squared errors is calculated for every lookup table entry, and the microstructural parameters are determined as the simulated combination of electron penetration depth, effective grain size, and out-of-plane orientation distribution that yields the smallest sum of the squared errors. Even though the kinematical electron scattering calculations predict that the relative intensities of diffraction spots along the ͑ 00 ͒ , ͑ 02 ͒ , and ͑ 04 ͒ Bragg rods are correlated to the in-plane orientation distribution, it is not a very sensitive measurement. Therefore, RHEED in-plane rocking curves are used to measure the in-plane orientation distribution. RHEED in-plane rocking curves are constructed by rotating the sample around the surface normal and recording the maximum intensity for each diffraction spot, minus the average background intensity, for each angle ͑ the angle between the nominal 100 zone axis and the projection of the incident electron beam on the sample surface ͒ . The resulting intensity distributions are characterized by the FWHM. To experimentally measure in-plane grain orientation distribution ͑⌬͒ , the FWHM of RHEED in-plane rocking curves 9 from the ͑ 024 ͒ , ͑ 026 ͒ , ͑ 044 ͒ , and ͑ 046 ͒ diffraction spots are measured simultaneously and compared to the FWHM of calculated in-plane rocking curves in another lookup table. As for the single image analysis, a lookup table was generated by calculating the FWHM of diffraction spot in-plane RHEED rocking curves for all relevant film param- eter combinations, i.e., effective grain size ͑ 4 –25 nm ͒ , out- of-plane orientation distribution ͑ 0°–20° FWHM ͒ , and in- plane orientation distribution ͑ 0°–30° FWHM ͒ . In-plane rocking curve FWHM was calculated to be in- dependent of the electron penetration depth, so it was set to 5 nm, the value most often measured using single image analysis at this electron energy and incidence angle. Each lookup table entry was indexed by its unique combination of the relevant film parameters ͑ grain size, out-of-plane orientation distribution, and in-plane orientation distribution ͒ and con- tained the FWHM of the rocking curves from the ͑ 024 ͒ , ͑ 026 ͒ , ͑ 044 ͒ , and ͑ 046 ͒ diffraction spots. The in-plane orientation distribution is determined by searching the lookup table for the simulation that has RHEED in-plane rocking curves that most closely match the experimental rocking curves for all four diffraction spots. The FWHM of the in- plane rocking curves are highly correlated with the in-plane orientation distribution, however, the rocking curve FWHM is also convoluted with the effective grain size and out-of- plane orientation distribution. Therefore, to accurately measure in-plane orientation distribution using in-plane rocking curves, the effective grain size and out-of-plane orientation distribution is first measured using single image analysis as described above. The subsequent comparison between the experimental and simulated FWHM of the RHEED in-plane rocking curves in the lookup tables are restricted to simulations with the effective grain size and out-of-plane orientation distribution measured using single image analysis. Experimental RHEED in-plane rocking curves and single image analyses were performed on 5- to 11-nm-thick IBAD MgO films. MgO was deposited, at room temperature, on amorphous Si 3 N 4 by electron beam evaporation at deposition rates ranging from 1.7 to 3.1 A/s, as measured by a quartz crystal monitor. Ion irradiation during MgO growth was carried out with 750 eV Ar ϩ ions at 45° incidence angle. Ion/MgO flux ratios were varied between 0.33 and 0.58. A single crystal of MgO was also analyzed for reference. Optimal film thickness was determined by monitoring the ͑ 004 ͒ diffraction peak intensity. 10 RHEED measurements were done at 25 kV and 2.6° incidence angle. Bragg spots along the ͑ 00 ͒ , ͑ 02 ͒ , and ͑ 04 ͒ Bragg rods, as shown in Fig. 1 ͑ b ͒ were used in the RHEED analysis. A 16 bit, 1024 ϫ 1024 pixels CCD camera provided adequate dynamic range to simultaneously observe all necessary spots and to spatially resolve spot shapes for single image analysis. Before attempt- ing single image analysis, the diffuse background was reduced by subtracting the Si 3 N 4 substrate RHEED image from the IBAD MgO RHEED pattern. This procedure was necessary to resolve weak diffraction spots and to reduce shape distortion caused by the diffuse background. Biaxial texture was also measured with either TEM or XRD to evaluate the accuracy of RHEED based measurements. The grazing incidence geometry of in-plane rocking curves enabled the use of either a rotating anode source or the Advanced Photon Source ͑ APS ͒ synchrotron to measure in-plane orientation distributions. However, out-of-plane orientation measurements of IBAD MgO layers required synchrotron radiation for the out-of-plane x-ray rocking curves. Even with the APS synchrotron radiation, the x-ray rocking curves did not provide reliable out-of-plane orientation distribution measurements for 8-nm-thick MgO samples with the broadest out-of-plane distributions ͑ Ͼ 11° ͒ . The in-plane orientation distributions measured using RHEED-based analysis is compared to measurements from TEM or x-ray scattering in Fig. 2. The data are well represented by a linear fit, demonstrating that the RHEED-based method successfully measures the in-plane orientation distribution. There are many possible sources of deviation from the straight line. The RHEED measurements require the deconvolution of the effective grain size and out-of-plane orientation distribution from the in-plane distribution. Errors in measurements of the effective grain size and out-of-plane orientation distribution therefore produce errors in the in- plane orientation distribution measurement. There is also a convolution between the measurement of effective grain size and out-of-plane orientation distribution such that an error in one measurement is compensated by an error in the other measurement. Reasonable errors for measurements of effective grain size and out-of-plane orientation distribution ͑ Ϯ 1 nm and Ϯ 1°, respectively ͒ yield a total in-plane measure- ment error of 1.5°, represented by the error bars in Fig. 2. Additional deviations from liner dependence originate in different sample-to-sample growth conditions that were used to produce films with the wide range of in-plane orientation distributions observed in Fig. 2. Measurements of IBAD MgO in-plane orientation distribution as a function of film thickness demonstrate that the in-plane orientation distribution decreases with increased film thickness, as illustrated in Fig. 3. We have also observed that the rate at which the in-plane distribution decreases depends on the ion/MgO flux ratio. TEM and x-ray scattering techniques probe the biaxial texture in a scattering volume that spans the entire thin film, measuring the film’s average orientation distribution, while RHEED measurements are more surface sensitive. Therefore, the in-plane orientation distribution measured by RHEED, a surface sensitive measurement, is not expected to directly correspond to the x-ray measurement, which probes the entire film thickness. With 750 eV Ar ϩ ion bombardment, the first 3 nm of the IBAD MgO film are amorphous. However, this layer yields a biaxially textured film out of the amorphous matrix through solid phase crystallization. 7 The first measurable RHEED patterns reveal that the initial in- plane orientation distributions are very broad, but they narrow as the film thickens until reaching an optimal alignment. The difference between the initial and optimal in-plane orientation distribution measurements is typically on the order of 10° under these growth conditions. Depending on the thickness of the final film, the difference between the average and surface in-plane orientation distribution will be different, causing another possible source of deviation from the linear fit in Fig. 2. Optimal biaxial texture under specific growth conditions is achieved by growing the film until the ͑ 004 ͒ diffraction spot reaches its maximum intensity. 10 Integrating the measured in-plane orientation distribution in Fig. 3 over the entire film thickness yields an average film in-plane orientation distribution about 2.5° broader than the surface in-plane orientation distribution. This is consistent with the offset between the in-plane orientation distribution measurements based on RHEED analysis and the x-ray or TEM analysis. The magnitude of this offset depends on the thickness of the film when growth was stopped, as well as on growth conditions such as ion/MgO flux ratio. Despite the expected differences between the surface sensitive and bulk measurement methods, as well as the in- herent limitations of the RHEED measurements because of convolution with effective grain size and out-of-plane distribution ...
Context 5
... of biaxially textured MgO thin films is techno- logically interesting because it may provide a suitable path for silicon integration of many important single crystal-like perovskite oxide thin films on top of the amorphous dielec- tric thin films present at the back end of a typical integrated circuit fabrication process. This is accomplished by using ion beam-assisted deposition ͑ IBAD ͒ to create biaxially textured films ͑ i.e., polycrystalline films with a preferred in-plane and out-of-plane grain orientation ͒ on amorphous substrates. Film functionality often depends on control of both the out- of-plane grain orientation distribution ͓ the full width at half maximum ͑ FWHM ͒ is designated as ⌬͔ and in-plane grain orientation distribution ͑ FWHM is designated as ⌬͒ . Some highly aligned biaxially textured oxide materials can exhibit similar functionality to single crystalline films. For example, biaxially textured superconductors such as YBa 2 Cu 3 O 7 Ϫ x have been reported to have critical current densities ap- proaching those of single crystalline films, while randomly oriented polycrystalline films exhibit much lower critical current densities. 1 Biaxially textured piezoelectric films with 90° domain rotations are also expected to have flexing char- acteristics similar to those of single crystalline piezoelectric films, while randomly oriented polycrystalline piezoelectric films experience significant degradation of translational range of motion. Incorporation of biaxially textured piezoelectric films with silicon integrated circuits would enable new types of ferroelectric actuators for microelectrical me- chanical systems ͑ MEMS ͒ . Previous work has shown that piezoelectric materials like BaTiO 3 and Pb ͑ Zr,Ti ͒ O 3 can be deposited heteroepitaxially onto single crystal MgO ͑ 001 ͒ 2,3 and even Si ͑ 001 ͒ . 4 However, conventional silicon integrated circuit processing employs extensive hydrogen passivation, which degrades ferroelectrics like Pb ͑ Zr,Ti ͒ O 3 and BaTiO 3 . It is therefore desirable to monolithically integrate piezoelectric materials following integrated circuit fabrication. Wang et al. demonstrated that IBAD MgO grown on amorphous Si 3 N 4 develops narrow biaxial texture in films only 11 nm thick. 5 By eliminating the requirement for a pre-existing het- eroepitaxial template, IBAD provides an opportunity to in- corporate piezoelectric materials on top of amorphous dielec- tric films in silicon integrated circuits during the back-end processing. The performance of piezoelectric MEMS is likely to de- pend on the biaxial texture inherited from the MgO substrate. Previous efforts to optimize the biaxial texture of IBAD MgO have been impeded by the ex situ nature of conventional biaxial texture analysis techniques ͓ transmission electron microscopy ͑ TEM ͒ or x-ray diffraction ͑ XRD ͔͒ . Be- cause the biaxial texture develops within 11 nm of growth, XRD cannot resolve crystallographic texture unless the x-ray source has synchrotron brightness. For these same reasons, the IBAD biaxial texturing mechanisms are difficult to inves- tigate. To circumvent these obstacles, we have developed a reflection high-energy electron diffraction ͑ RHEED ͒ based method for quantitative in situ biaxial texture analysis of MgO. RHEED has been previously used to analyze the out- of-plane texture for CoCr alloys, assuming the grains were not large enough to affect the RHEED pattern. 6 The small grain size of IBAD MgO films ͑ as small as 10 nm ͒ necessi- tate that we deconvolute the effects of grain size from the effects of out-of-plane orientation distribution for accurate texture distribution measurements. We also demonstrate the ability to measure the in-plane orientation distribution. Be- cause RHEED is sensitive to films as thin as 3 nm thick, we have the capability of analyzing the biaxial texture development during the earliest stages of film growth. 7 This analysis capability enables greater understanding of IBAD MgO biaxial texture development and how to optimize the MgO biaxial texture. Our RHEED-based biaxial texture analysis employs a previously reported kinematical electron scattering model. 8 These calculations predict that spot shapes are sensitive to the film microstructure, as shown in Fig. 1 ͑ a ͒ . Diffraction spot width and height are inversely proportional to the effective grain size and electron penetration depth, respectively. The width of the diffraction spot in the direction perpendicular to the location of the through spot, the nondiffracted electron beam, is directly proportional to the out-of-plane grain orientation distribution ͑⌬͒ . We therefore characterize RHEED patterns, whether calculated using a computer simulation or from an experiment, by cutting across the diffraction spots along the previously mentioned directions, and measuring the FWHM of these cuts, as shown in Fig. 1 ͑ b ͒ . We call this method ‘‘single image analysis.’’ All diffraction spots shown in Fig. 1 ͑ b ͒ are analyzed simultaneously, and then compared to calculated RHEED pattern measurements using a lookup table. The lookup table was generated by calculating the RHEED pattern for all relevant combinations of effective grains size ͑ 4 –25 nm ͒ , electron penetration depth ͑ 2.5–10 nm ͒ , and out-of-plane orientation distribution ͑ 0°–20° FWHM ͒ , and measuring the FWHM of cuts across the six diffraction spots shown in Fig. 1 ͑ b ͒ in the directions where the RHEED pattern is sensitive to electron penetration depth, grain size, and out-of-plane orientation distribution. The effects of the in-plane orientation distribution on diffraction spot shapes are negligible, so the in-plane orientation distribution FWHM was set to 10° for lookup table calculations, which is a typical value for in-plane orientation distributions observed in IBAD MgO. Lookup table entries exist for all combinations of effective grain size, electron penetration depth, and out-of-plane orientation distribution, and contain the measurements of the FWHM of the cuts across each previously specified spot along the previously specified directions. The film effective grain size, electron penetration depth, and out-of-plane orientation distribution are determined by comparing the FWHM of experimental RHEED pattern diffraction spot cuts with the FWHM of the spot cuts in the lookup table. For each lookup table entry, the experimentally measured FWHM of each spot cut is sub- tracted from the lookup table FWHM of the same spot in the same direction. The differences between the experimental and lookup table measurements are then individually squared before being added together to yield a total sum of the square errors measurement for that lookup table entry. The sum of the squared errors is calculated for every lookup table entry, and the microstructural parameters are determined as the simulated combination of electron penetration depth, effective grain size, and out-of-plane orientation distribution that yields the smallest sum of the squared errors. Even though the kinematical electron scattering calculations predict that the relative intensities of diffraction spots along the ͑ 00 ͒ , ͑ 02 ͒ , and ͑ 04 ͒ Bragg rods are correlated to the in-plane orientation distribution, it is not a very sensitive measurement. Therefore, RHEED in-plane rocking curves are used to measure the in-plane orientation distribution. RHEED in-plane rocking curves are constructed by rotating the sample around the surface normal and recording the maximum intensity for each diffraction spot, minus the average background intensity, for each angle ͑ the angle between the nominal 100 zone axis and the projection of the incident electron beam on the sample surface ͒ . The resulting intensity distributions are characterized by the FWHM. To experimentally measure in-plane grain orientation distribution ͑⌬͒ , the FWHM of RHEED in-plane rocking curves 9 from the ͑ 024 ͒ , ͑ 026 ͒ , ͑ 044 ͒ , and ͑ 046 ͒ diffraction spots are measured simultaneously and compared to the FWHM of calculated in-plane rocking curves in another lookup table. As for the single image analysis, a lookup table was generated by calculating the FWHM of diffraction spot in-plane RHEED rocking curves for all relevant film param- eter combinations, i.e., effective grain size ͑ 4 –25 nm ͒ , out- of-plane orientation distribution ͑ 0°–20° FWHM ͒ , and in- plane orientation distribution ͑ 0°–30° FWHM ͒ . In-plane rocking curve FWHM was calculated to be in- dependent of the electron penetration depth, so it was set to 5 nm, the value most often measured using single image analysis at this electron energy and incidence angle. Each lookup table entry was indexed by its unique combination of the relevant film parameters ͑ grain size, out-of-plane orientation distribution, and in-plane orientation distribution ͒ and con- tained the FWHM of the rocking curves from the ͑ 024 ͒ , ͑ 026 ͒ , ͑ 044 ͒ , and ͑ 046 ͒ diffraction spots. The in-plane orientation distribution is determined by searching the lookup table for the simulation that has RHEED in-plane rocking curves that most closely match the experimental rocking curves for all four diffraction spots. The FWHM of the in- plane rocking curves are highly correlated ...
Citations
... For a film thickness of 0 nm, of course the in-plane mica "a-lattice parameter" is observed, which is much larger than the one of Bi 2 Te 3 . When Bi 2 Te 3 is grown on mica, the exponentially decaying penetration depth of the RHEED analysis, which is of the order of 1 nm, 35 will for the first atomic layers of Bi 2 Te 3 result in a weighted average of the lattice parameters of mica and Bi 2 Te 3 . The quickly decaying results indicate that Bi 2 Te 3 does not form a lattice matched film on mica and quickly relaxes. ...
Strain engineering as a method to control functional properties has seen in the last decades a surge of interest. Heterostructures comprising 2D-materials and containing van der Waals(-like) gaps were considered unsuitable for strain engineering. However, recent work on heterostructures based on Bi2Te3, Sb2Te3, and GeTe showed the potential of a different type of strain engineering due to long-range mutual straining. Still, a comprehensive understanding of the strain relaxation mechanism in these telluride heterostructures is lacking due to limitations of the earlier analyses performed. Here, we present a detailed study of strain in two-dimensional (2D/2D) and mixed dimensional (2D/3D) systems derived from mica/Bi2Te3, Sb2Te3/Bi2Te3, and Bi2Te3/GeTe heterostructures, respectively. We first clearly show the fast relaxation process in the mica/Bi2Te3 system where the strain was generally transferred and confined up to the second or third van der Waals block and then abruptly relaxed. Then we show, using three independent techniques, that the long-range exponentially decaying strain in GeTe and Sb2Te3 grown on the relaxed Bi2Te3 and Bi2Te3 on relaxed Sb2Te3 as directly observed at the growth surface is still present within these three different top layers a long time after growth. The observed behavior points at immediate strain relaxation by plastic deformation without any later relaxation and rules out an elastic (energy minimization) model as was proposed recently. Our work advances the understanding of strain tuning in textured heterostructures or superlattices governed by anisotropic bonding.
... 2(d)-2(f) is due to the oriented growth of CoO layer. 34 The presence of diffraction arc at h ¼ 0 for (222) or (111) CoO planes confirms the preferential orientation of body diagonal normal to the film surface 35 and remains unchanged with the thermal annealing up-to 20 min with the partial pressure of oxygen. In Figure 3, the diffraction intensity of as-prepared and annealed Co film at 573 K for 30 min is plotted as a function of K. ...
... The plot shows a peak for each ring, where broadening (FWHM) of the rings in the radial direction in reciprocal space depends on the grain size in the real-space. [33][34][35] It may be noted that after annealing, integrated peaks do not show any significant change in peak width and relative intensities. It suggests that no appreciable change in the grain size of the Co film takes place during thermal annealing process at 573 K. ...
The effect of interface roughness on exchange-bias (EB) properties of polycrystalline Co/CoO bilayer structure has been studied in-situ. Isothermal annealing of a 135 thick Co layer under the partial pressure of pure oxygen at 573 K results in the formation of a 35 thick CoO layer, the surface roughness of which increases with the increasing annealing time. Bilayers were characterized in-situ using magneto-optic Kerr effect, reflection high energy electron diffraction, and x-ray reflectivity for their magnetic and structural properties during each stage of bilayer growth. Combined analysis revealed that the increase in the roughness from 7 to 13 causes the exchange bias field (HEB) to decrease from 171 Oe to 81 Oe, whereas coercivity (HC) increases up to 616 Oe. In contrast to some earlier studies on polycrystalline films, where HEB increased with roughness due to the increase in the uncompensated spins at ferromagnetic-antiferromagnetic (AFM) layer interface, in the present case, dependence of HEB and HC on the roughness is attributed to the disorder at the interface of AFM layer, which leads to a decrease in HEB due to weakening of the effective spin coupling at the interface. Present in-situ experiments make it possible to study the variations in EB properties with the interface roughness in a single sample, and thus avoiding the possibility of the sample to sample variation in the morphological properties along with the change in the interface roughness.
... In particular, a longer deposition time (i.e. more laser pulses) was necessary for temperatures below 100 @BULLET C to achieve a biaxial textured layer (figure 1(a)). The circular elongation of the RHEED spots indicates at the same time that the texture spread is still broad [18]. A better alignment was obtained using substrate temperatures of about 250 @BULLET C ± 100 @BULLET C, where the textured nucleation is complete in an about 4 nm thick layer. ...
The incorporation of nanoscaled pinning centers in superconducting YBa2Cu3O7-δ (YBCO) films is one of the core topics to enhance the critical current density Jc(B, Θ) of coated conductors. The mixed double-perovskite Ba2Y(Nb/Ta)O6 (BYNTO) can be grown in nanosized columns parallel to the YBCO c-axis and in steplike patterns, making it customizable to meet specific working conditions (T, B, Θ). We compare a 1.6 μm thick film of pure YBCO and a similar film with additional 5 mol% of BYNTO, grown by pulsed laser deposition with a growth rate of 1.6 nm/s on chemically buffered biaxially textured Ni-5at.% W tape. Our doped sample shows nanosized BYNTO columns parallel to cYBCO and plates in the ab-plane containing Y, Nb, and Ta. An improved homogeneity of the critical current density Jc over the sample was evaluated from trapped field profiles measured with a scanning Hall probe microscope. The mean Jc in the rolling direction of the tape is 1.8 MA/cm2 (77 K, self-field) and doubles the value of the undoped sample. Angular-dependent measurements of the critical current density, Jc(Θ), show a decreased anisotropy of the doped film for various magnetic fields at 77 K and 64 K.
... Then IBAD-YSZ was abandoned by LANL, because the choice of MgO could significantly shorten the fabrication time [37]. The optimization of IBAD-MgO deposition parameters was conducted in LANL [38,39], and reflection high-energy electron diffraction (RHEED) used to monitor the texturing of MgO was studied in the cooperation of California Institute of Technology and LANL [40,41]. ...
Ion beam assisted deposition (IBAD) is an important technique to fabricate the second generation high temperature superconducting (2G HTS) wires. Among the fabrication routes of 2G HTS long wires, IBAD route achieved the best performance in recent years. IBAD was adopted in this field in 1991 to obtain biaxially textured buffer layers, which helped to deposit high quality YBCO superconducting films on metallic substrates for the first time. Series of experimental and industrial researches on IBAD were carried out by many groups worldwide. And in the researches lasting for over two decades, the focused material for IBAD was changed from Yttria-Stabilized Zirconia (YSZ), Gd2Zr2O7 (GZO) to MgO. In this paper, the research progresses and the main achievements were briefly reviewed.
... When annealing at 350 • C [ Fig. 7(b)], 3D spots are still present but interestingly they are distributed along concentric rings, a behavior similar to that observed from electron diffraction of a polycrystalline structure. [52][53][54][55][56][57] This is in line with STM measurement, for which it is, though, difficult to have a good statistics over all orientation of droplets. This is also in good agreement with the above XPD analyses depicted in the curve c of Fig. 5, which suggests that when Au droplets are formed for annealing at temperatures higher than T E , they exhibit a random distribution. ...
We investigate the chemical and morphological structure of the Au nanodots on Ge(111), which serve as catalysts for the formation of epitaxial Ge nanowires. We show that dewetting of an Au film on Ge(111) gives rise to a thin Au-Ge wetting layer and Au-Ge dots. These dots are crystallized but not with a single crystallographic orientation. Thanks to the spatially resolved x-ray and transmission electron microscopy measurements, a chemical characterization of both binary Au-Ge catalysts and wetting layer is obtained at the nanoscale. We show that Ge vertical growth is achieved even without an external Ge supply.
... For rough polycrystalline or nanocrystalline films, conventional RHEED diffraction patterns give only partial information about the surface texture. Brewer et al. have developed the RHEED in-plane rocking curve to obtain additional information on the azimuthal angle orientation [31]. By recording multiple diffraction patterns around the surface normal, we have shown that it is possible to construct a RHEED surface pole figure that contains complete information about the growth front texture [22][23][24]. ...
The most frequently used characterization technique for biaxial texture formation in thin films is x-ray pole figure analysis. However, x-rays interact weakly with matter and can penetrate a few microns deep into the film. The texture obtained by x-rays is therefore an average texture from the entire thickness of the film. As the texture of a film often changes during growth, information on the basic mechanisms that control the final texture is often lost. In contrast electrons interact strongly with matter and they have very limited penetration and escape depths of a few nm. In this paper we will show how we can use our newly developed reflection high energy electron diffraction (RHEED) surface pole figure technique to probe the surface texture evolution of the growth front from the initial stage (nm thick) to the later stage. The RHEED pole figure technique is a surface-sensitive technique that allows us to obtain information on the dynamic behavior of texture evolution of the growth front during film deposition. We shall explain the principle, measurement, and construction of such RHEED surface pole figures. An example of the biaxial texture evolution of CaF2 due to the atomic shadowing effect during oblique angle deposition is described.
... This confirms that the Ni NCs were embedded in BaTiO3 matrix, by the alternating of Ni NCs self-organization process and BaTiO3 epitaxial growth. Since the RHEED pattern is very sensitive to the surface microstructure [15-17], the microstructure and the period of such quantum dot superlattice can be engineered with the in situ monitoring of RHEED. ...
Ni nanocrystals (NCs) were embedded in BaTiO3 epitaxial films using the laser molecular beam epitaxy. The processes involving the self-organization of Ni NCs and the epitaxial growth of BaTiO3 were discussed. With the in situ monitoring of reflection high-energy electron diffraction, the nanocomposite films were engineered controllably by the fine alternation of the self-organization of Ni NCs and the epitaxial growth of BaTiO3. The transmission electron microscopy and the X-ray diffraction characterization confirmed that the composite film consists of the Ni NCs layers alternating with the (001)/(100)-oriented epitaxial BaTiO3 separation layers.
... In particular, a longer deposition time (i.e. more laser pulses) was necessary for temperatures below 100 • C to achieve a biaxial textured layer ( figure 1(a)). The circular elongation of the RHEED spots indicates at the same time that the texture spread is still broad [18]. A better alignment was obtained using substrate temperatures of about 250 • C ± 100 • C, where the textured nucleation is complete in an about 4 nm thick layer. ...
Ion-beam-assisted deposition (IBAD) offers the opportunity to prepare thin textured films on non-textured substrates. The approach was used to study if superconducting NbN can be textured in this way. Therefore, a reactive process using pulsed laser deposition of pure Nb in combination with a nitrogen-containing ion beam was utilized for the preparation of textured NbN on amorphous seed layers. It is shown that NbN reveals a textured nucleation similar to IBAD-MgO or IBAD-TiN. The biaxial texture was stabilized in thicker layers using homoepitaxial growth leading to highly textured NbN layers with an in-plane alignment below 5°. The dependence of the lattice parameter as well as of the superconducting transition temperature Tc on the deposition conditions, such as, for example, the N2 partial pressure during the homoepitaxial growth, was studied in detail. A clear correlation between structural and superconducting properties was found for the deposited NbN thin films. As a result, highly textured NbN layers were prepared showing a Tc up to 14 K.
... In particular, the preparation of cube textured films using IBAD is of major interest for the application of high-temperature superconductors [6]. It was reported within the last decade that materials with a rocksalt structure, as for example MgO or TiN, can be textured during nucleation showing strong cube textures in films with a thickness below 10 nm [7][8][9][10]. The 1 1 0 axis of the grown film is aligned parallel to the ion beam in these materials, if an ion incidence angle of 45 • towards the substrate normal is used. ...
Ion-beam assisted deposition (IBAD) is a promising technique for preparing highly textured templates on various substrates. A major requirement for the textured nucleation of materials with a rocksalt structure using the IBAD approach is an amorphous seed layer. Metallic Ta0.75Ni0.25 films were used as amorphous layers on Si/Si3N4 as well as on electropolished Hastelloy tapes for the ion-beam assisted pulsed laser deposition of transition metal nitrides. Highly biaxially textured TiN and NbN films have been grown on these layers even at room temperature in a reactive deposition process using an ion beam with an energy of 800 eV under an angle of 45° relative to the substrate normal. The cube texture was preserved to higher thicknesses using homoepitaxial growth. Finally, 10 nm thick Au films were used to study the surface texture quantitatively using x-ray diffraction showing in-plane orientations down to 8°. The results offer the possibility to realize a fully conductive buffer layer architecture for high-temperature superconductors based on the IBAD approach.
... One method of comparison between IBAD and DIBAD processing is the use of RHEED. In previous works, we demonstrated that monitoring of the RHEED spot intensity as a function of time can be used to control the in-plane texture of growing films [17], [18]. The in-plane texture of IBAD MgO films has been shown to degrade after 10 nm of film has been deposited. ...
Ion beam assisted deposition (IBAD) of magnesium oxide (MgO) has been shown to be a viable route for producing biaxially textured template films on flexible polycrystalline metal substrates for high-performance coated conductor development. We have refined the technique of IBAD by using a dual ion assist approach. Dual ion assist beam deposition (DIBAD) of MgO reduces the requirements for substrate surface finishing while maintaining comparable film quality (phi scan full-width at half-maximum values between 7 and 8deg). Furthermore, this adaptation of the IBAD process eliminates the degradation of MgO texture observed in IBAD MgO films deposited on silicon nitride. We have deposited films up to 50 nanometers thick without degradation of the in-plane texture. Increasing the MgO thickness increases the chemical stability of template layer and can eliminate the necessity for subsequent buffer layers or the application of the homoepitaxial MgO layer needed to stabilize the thin, conventional IBAD MgO layer. The initial performance of coated conductors made with DIBAD templates is quite promising.
