Heteroepitaxial structures of SrTiO3/TiN on Si(100) by in situ pulsed laser deposition

Abstract

High-quality ceramics based heteroepitaxial structures of oxide-nitride-semiconductors, i.e., SrTiO3/TiN/Si(100) have been fabricated by in situ pulsed laser deposition. The dependence of substrate temperature and oxygen partial pressure on the crystalline quality of the SrTiO3 films on Si with epitaxial TiN template has been examined. We found that epitaxial growth occurs on TiN/Si(100) above 500 °C, initially at a reduced O2 pressure (10−6 Torr), and followed by a deposition in the range of 5–10×10−4 Torr. X-ray diffraction (Θ, ω, and Φ scans) and transmission electron microscope (TEM) results revealed an excellent alignment of SrTiO3 and TiN films on Si(100) with a cube-on-cube epitaxy. Rutherford backscattering and ion channeling results show a channeling minimum yield (χmin) of ∼13% for the SrTiO3 films. High-resolution TEM results on the SrTiO3/TiN interface show that the epitaxial SrTiO3 film is separated from the TiN by an uniform 80–90 Å crystalline interposing layer presumably of TiNxO1−x (oxy-nitride). The SrTiO3 film fabricated at 700 °C showed a high relative dielectric constant of 312 at the frequency of 1 MHz. The electrical resistivity and the breakdown field of the SrTiO3 films were more than 5×1012 Ω cm and 6×105 V cm−1, respectively. An estimated leakage current density measured at an electric field of 5×105 V/cm−1 was less than 10−7 A/cm2.

Integration of artificial SrTiO3/BaTiO3 superlattices on Si substrates using a TiN buffer layer by pulsed laser deposition method

“…It is generally observed in perovskite oxide films on Si substrates that epitaxial perovskite oxide unit cell rotates 451 on Si [1 0 0] with small lattice mismatch because the diagonal length ( ffiffi ffi 2 p a; a ¼ 0:3820:40 nm) of lattice of perovskite oxides agrees well with the lattice constant of Si (0.543 nm) and then the [0 1 0] axis of perovskite oxide films tends to have 451 rotation with respect to that of Si substrate [24,25]. However, the cube-on-cube in-plane epitaxy nature, observed in this study, has been previously reported by others where a non-oxide TiN buffer layer was introduced between oxide films and Si substrates as explained in the BTO/TiN/Si [11] and the STO/TiN/Si system [20,26]. Our experimental results of the epitaxy nature of oxide films on Si substrates are well consistent with these previous reports, where the TiN buffer layer grown with cube-oncube epitaxy nature plays an important role in the formation of cube-on-cube epitaxy of oxide thin films on Si substrates.…”

“…Various kinds of buffer layers, such as, TiN [12], SiC [16], AlN [17], and CaF 2 [18] have been used to prepare high-quality epitaxial oxide thin films on Si substrates. Among them, TiN is the most promising candidate because it has very strong corrosion and erosion resistance, good diffusion resistance, low resistivity, and good adhesion property to most oxide materials [19][20][21][22]. In addition, the epitaxy nature of the TiN thin film on a Si substrate has been well known by domain matching epitaxy theorem even though they have very large lattice mismatch of 25% [19].…”

New frontiers in thin film growth and nanomaterials

“…If the interatomic potentials are too different across the interface, such as diamond film on copper, film grows polycrystalline despite lattice matching between the two materials. Recently, we have grown single-crystal thin films of MoO 2Ϫx (monoclinic, space group P21/n, [14] ␤ ϭ 119.37 deg) and ␤-Ga 2 O 3 (monoclinic, space A2/m, [12] ␤ ϭ 103.7 deg) on (0001) sapphire substrates, where thin film epitaxy occurs by matching of integral multiples of lattice planes, which are inclined with respect to the interface. [47] Thus, domain epitaxy provides a mechanism to grow epitaxial films on substrates with large misfits, and opens a new frontier in next-generation solid-state technology.…”

“…[11] Since the initial DME report, many systems with a large misfit have been grown with remarkable success. [12][13][14][15][16] Further progress in this field of critical importance will depend on the atomistic details of the relaxation process, and the nature of dislocations in such systems.…”

Solid-phase epitaxial growth of (111)-oriented Si film on InGaO3(ZnO)5 buffer layer

“…ionic crystals such as oxides. Many systems with a large misfit have been grown by DME [22][23][24][25][26]. This gives possibility that Si can grown epitaxially on a substrate even with a large lattice misfit [8%.…”