Research Programs
IRG 2: Oxides as Semiconductors
OXIDES:

IRG-2: Oxides as Semiconductors

Semiconductors are a critically important area of materials research with great societal benefits. We focus on the exploration and development of semiconducting binary oxides (ZnO, Ga2O3, In2O3, SnO2, TiO2) as new wide-band-gap electronic materials. Even with their current high level of impurities and poor microstructures, these oxides already have significant industrial importance. The technical goal is to enable improved control over physical properties, the scientific goal is to both uncover new physics and realize dramatic improvements in materials properties, thereby spawning novel and innovative applications. We are employing high-quality growth techniques such as Molecular Beam Epitaxy and Metal-Organic Chemical Vapor Deposition to produce high purity epitaxial films with controlled doping. The experimental efforts are closely linked to computational research based on cutting-edge first-principles techniques.

IRG MEMBERS

Jim Speck (Co-leader)

Materials

MBE growth

Chris Van de Walle (Co-leader)

Materials

Theory/Doping

Steve DenBaars

Materials and Electrical & Computer Engineering

MOCVD growth

Umesh Mishra

Electrical & Computer Engineering

Devices

Ram Seshadri

Materials

Magnetism

Nicola Spaldin

Materials

Theory/Magnetism

Susanne Stemmer

Materials

Microscopy

IRG AFFILIATES

Shuji Nakamura

Materials

Growth/devices

Fred Lange

Materials and Chemical Engineering

Synthesis

David Awschalom

Physics

Spintronics



Zinc oxide homoepitaxy

The availability of reasonably priced high-quality single-crystal ZnO wafers stands out among the many interesting features of ZnO. Such substrates allow for homoepitaxial growth, which has many advantages over heteroepitaxy, including strain-free films and very low defect densities. Surprisingly, homoepitaxy of ZnO had proven to be a major challenge; at UCSB, we have successfully used metal-organic chemical vapor deposition to obtain very smooth ZnO thin films grown on high-quality ZnO substrates. No interface could be detected with transmission electron microscopy, indicating a perfect continuation of the substrate.

Zinc oxide as a transparent conducting oxide

Nitride-based light-emitting diodes are reaching unprecedented efficiencies, but extracting the light that is generated inside the diode is still a major challenge. This problem can be addressed by combining the LEDs with a transparent conducting oxide. A thin, highly doped ZnO layer is deposited on the surface of the diode and used as a transparent contact. ZnO is much less expensive than gold, titanium, or indium-based oxides, which are the materials that are commonly used as transparent contacts today. In addition, ZnO lends itself to being shaped into a millimeter-sized cone that can be bonded to the device. This geometry allows light extraction with unprecedented efficiency. We are also investigating the properties of In-doped ZnO layers grown by MOCVD and find that In is an effective dopant for controlling the n-type conductivity of ZnO.

T. Ive, T. Ben-Yaacov, C. G. Van de Walle, U. K. Mishra, S. P. DenBaars, and J. S. Speck, J. Cryst. Growth 310, 3407 (2008).

Molecular beam epitaxy of SnO2

Tin oxide alone or together with indium oxide (ITO) forms a transparent contact. SnO2 is also used for gas sensors. Due to these two major applications, the majority of research has focused on either increasing its electrical conductivity or optimizing the response to gas molecules, but neglecting the fundamental properties of tin oxide untainted by impurities or crystalline defects. We have used MBE to produce extremely high quality phase pure single crystalline SnO2 thin films. In the course of this work we observed a surprising decrease of the film growth rate when Sn flux was increased, an observation which we explain by formation of sub-oxides.

M.Y. Tsai, M.E. White, and J. S. Speck, J. Cryst. Growth 310, 4256 (2008).
M. E. White, M. Y. Tsai, F. Wu, and J. S. Speck, J. Vac. Sci. Technol. A
26, 1300 (2008).

Molecular beam epitaxy of Ga2O3

Ga2O3, with a band gap of 4.8 eV, is transparent throughout the visible and into the ultraviolet. Very little is known about its fundamental material properties. We have grown different Ga2O3 polymorphs on various substrates including c- and r-plane sapphire, and β-Ga2O3.

Novel MBE methods for high-quality TiO2 and SrTiO3 thin films

The low vapor pressure of Ti poses major challenges for MBE of Ti-containing oxides. This problem has been overcome by a hybrid growth scheme, using metal-organic sources for Ti. Extremely high-quality homoepitaxial films of SrTiO3 have been obtained.

B. Jalan, R. Engel-Herbert, J. Cagnon, S. Stemmer, J. Vac. Sci. Technol. A (March 2009, accepted).

Evolution of crystal structure and polarization in wurtzite ZnMgO alloys

Polycrystalline samples of ZnMgO with Mg content up to 15% have been prepared by an oxalate-based coprecipitation method. The effect of Mg substitution on the crystal structure of wurtzite ZnO was studied synchrotron x-ray diffraction, Raman spectroscopy, and NMR. Significant changes in the ionic polarization have been observed despite only subtle increments in the cell volume (~0.03%) and the ab-area dimension (~0.1%).

Y.-I. Kim, K. Page, R. Seshadri, Appl. Phys. Lett. 90, 101904 (2007).

Polarization, piezoelectric and elastic constants of ZnO, MgO and CdO

First-principles calculations have addressed equilibrium polarizations, piezoelectric stress constants, and elastic constants of these binary oxides in the wurtzite structure. The large polarization differences between the end-point compounds in the MgO-ZnO-CdO system augur well for the generation of large internal fields in ZnO-based field effect transistors.

P. Gopal and N. A. Spaldin, J. Electron. Mater. 35, 538 (2006).

Fundamental studies of transport

Our transport investigations are aimed at determining the conductivity type (p-type or n-type), exploring the conductivity of the oxide surface, and identifying the cause of the unintentional n-type conductivity. Conductivity measurements are commonly based on van der Pauw-Hall measurements and Capacitance-Voltage measurements. We have found that under certain conditions these techniques produce the wrong conductivity type (e.g., p-type instead of n-type). Procedures to retrieve the correct conductivity type have been outlined.

O. Bierwagen, T. Ive, C. G. Van de Walle, and J. S. Speck, Appl. Phys. Lett. 93, 242108 (2008).

First-principles investigations of conductivity in oxides

Many as-grown oxides are unintentionally conductive. E.g., ZnO and SnO2 often exhibit unintentional n-type conductivity. Because of its correlation with oxygen partial pressure, this unintentional n-type conductivity has long been attributed to oxygen vacancies. We have performed first-principles calculations that show that oxygen vacancies in these materials are high-energy defects, and do not act as shallow donors. Our results for the calculated configuration coordinate diagrams elegantly explain recent optically detected paramagnetic-resonance measurements of oxygen vacancies in ZnO.

A. Janotti and C. G. Van de Walle, Phys. Rev. B 76, 165202 (2007).

Hydrogen multicenter bonds

We propose that unintentionally incorporated impurities, such as hydrogen, are responsible for the observed conductivity.

Both interstitial and substitutional hydrogen act as shallow donors. Hydrogen favorably substitutes on the oxygen site, forming a multicenter bond with the neighboring Zn atoms and contributing to the observed conductivity. So far, substitutional hydrogen has been computationally identified to be a shallow donor in ZnO, SnO, TiO2, SrTiO3, and In2O3. We are also currently investigating other unintentional dopants with the aim of achieving greater control over the conductivity of ZnO.

A. Janotti and C.G. Van de Walle, Nature Materials 6, 44 (2007)

First-principles investigations of conductivity in oxides

p-type doping of the oxides is still a major challenge. For SnO2, we have proposed that doping with group-III acceptors (Al, Ga, In) should produce p-type conductivity. We have examined the effects of hydrogen, finding that the presence of hydrogen during growth will enhance acceptor solubility, and that it should be possible to remove hydrogen in a post-growth anneal. In addition, we have found anomalous frequencies for the hydrogen vibrational modes.

A. K. Singh, A. Janotti, M. Scheffler, and C. G. Van de Walle, Phys. Rev. Lett. 101, 055502 (2008).