Skip to main content
eScholarship
Open Access Publications from the University of California

Bandgap Engineering of NiO-CdO

  • Author(s): Francis, Christopher
  • Advisor(s): Dubón, Oscar D
  • et al.
Abstract

Understanding and controlling the electronic structure of materials are longstanding endeavors in semiconductor physics and technology, especially in the transparent conducting oxide community. In thin films of CdO and NiO, such control has been attempted on these individual materials through alloying with other binary oxides. No such studies, however, exist on the electronic structure evolution of the combined CdO-NiO alloy system. This dissertation addresses this issue by conducting and reporting a bandgap engineering study on NixCd1-xO alloys.

This dissertation first reviews previous bandgap engineering studies performed on other systems, initially focusing on alloying and later focusing on other advanced bandgap engineering methods. A discussion of the earliest point defect and ADM theory studies then justifies the dissertation’s selection of CdO—it’s strong electrical properties are suitable for bandgap engineering. Investigations of previous literature involving the Group II-oxides of ZnO, MgO, and CdO lead to two key conclusions. First, few studies exist in which CdO is a primary alloying material. Second, the properties of CdO are tunable with better complementary materials such as NiO, a transition metal (TM) oxide, instead of commonly used ZnO and MgO. Given their unique properties, there is an opportunity to investigate the structure, properties, and behavior of CdO system after alloying it with NiO. Hall effect, optical reflectance and transmittance and X-ray diffraction measurements are utilized first and the structural, electrical transport and optical properties of NixCd1-xO films sputtered in argon (Ar) with radio frequency (RF) magnetron are then reported.

This initial study shows that NixCd1-xO alloys are rocksalt-structured and exhibit a monotonic shift of the (220) diffraction peak to higher 2θ angles with increasing Ni concentration. The electron mobility and electron concentration decrease with increasing Ni—becoming highly resistive for Ni content greater than 43.4% Ni. This decrease in n-type conductivity is consistent with the movement expected from a virtual crystal approximation (VCA) of the conduction band minimum (CBM) from below to above the Fermi stabilization energy (EFS). The optical absorption edge of the alloys is tunable from CdO to NiO. An intrinsic, carrier-free bandgap of the alloys, Eg, was calculated from the electrical and optical measurements, accounting for Burstein-Moss carrier filling and carrier-induced bandgap correlation effects. An unusual super linear composition dependence of the intrinsic bandgap is revealed when accounting for these effects. The super linear behavior was initially attributed to an interaction between the conduction-band extended states and localized donor and acceptor d-states of Ni.

To probe the mechanisms behind the anomalous electrical transport and optical behaviors of the Ar sputtered alloys a collection of experimental and modeling investigations via ion irradiation, band anticrossing (BAC) simulations and X-Ray Photoelectron Spectroscopy-Ultraviolet Photoelectron Spectroscopy (XPS-UPS) was then used. This study discovered that the introduction of a TM with two impurity levels leads to interactions that reconstruct both the conduction and valence bands of the alloy with increasing Ni. Irradiation of the films leads to a saturation of the electron concentration associated with the pinning of the Fermi level at EFS. The composition dependence of the pinned EF enables determination of the CBM energy relative to the vacuum level. There is an unusually strong deviation of this CBM energy observed from the VCA which is then explained by a BAC interaction between localized 3d states of Ni and the extended states of the NixCd1-xO alloy host. The resulting band structure is responsible for the dependence on composition of the electrical and optical properties of the alloys, the rapid reduction of the electron mobility, and previously observed positive bandgap bowing parameter. XPS-UPS studies confirm that the Γ- and L-point valence band maxima in the Cd-rich alloys are unaffected by interactions with Ni d-states. The results from this study provide much-needed context to the previously reported, but unexplained, electrical transport and optical behavior found in NixCd1-xO, NixMg1-xO and Ni1-xZnxO alloys—the interactions govern their measured electrical and optical properties. These breakthroughs are also applicable to metal-oxide-based semiconducting alloys with TM acting as the dopant or alloying agents—such as V-doped ZnO.

With an understanding of the structure, properties, and behavior of n-type, Ar sputtered NixCd1-xO, exploratory work for applications was then conducted. First, the electrochemical modification of these alloys for electrochromic windows was completed and the resulting electrical transport and optical properties were reported. Second, separate studies on the modification of NixCd1-xO with the percent of O2 sputtering gas were implemented to evoke p-type conductivity for p-n junctions and hole emitter applications. Following this growth method, rapid thermal annealing studies under N2 and O2-rich environments were conducted. These studies probed the defect mechanisms and discussed the optimal processing conditions that encourage the growth of reproducible and measurable p-type conductivity in NixCd1-xO. By altering the percent of O2 in the growth ambient, NixCd1-xO films with tunable electrical transport properties and charge type are realized—the first such result of its kind. This dissertation concludes with a proposal for the future studies that can provide additional information on NixCd1-xO and other metal-oxides as a result.

Overall, this dissertation makes exciting contributions to the general area of semiconductor science while shedding light on fundamental processes at the intersection of chemistry, materials science and processing, and solid-state physics. With this greater understanding, we can now proceed with tuning NixCd1-xO for transparent electronic, photovoltaic, and photoelectrochemical applications, which require its components to have tailored electrical transport and optical properties for effective use.

Main Content
Current View