An important aspect of the large expansion in the development and production of solid-state devices has been the demand for more sophisticated techniques for determining the electrical properties of semiconductors, especially silicon and the 111-V compounds. A very wide range of measurement techniques now exists and it is the purpose of this article to review those techniques which are in widespread use or which show promise for future application, and at a time of continuing innovation in this area, to indicate present trends and material problems which may arise in the near future. T h e emphasis of the review is on the physics of the methods: detailed discussion of results on specific materials is avoided and it is assumed that the intrinsic material properties are well known. The review is therefore concerned with the assessment of the electrical effects of impurities and defects.Routine characterisation using the resistivity, Hall coefficient and magnetoresistance effects is reviewed and methods for determining the donor and acceptor impurity concentrations from the 77 K mobility or from the temperature dependence of the free carrier density are described. Current knowledge of the Hall scattering factor is summarised. A review is given of rapid resistivity measurement by four-point probe, spreading resistance and contactless methods. Depth profiling by spreading resistance probe on bevelled structures, and by layer removal using chemical etching, anodic oxidation and ion beam etching are also considered. A major development in the last 10 years has been the use of capacitance methods for material characterisation. T h e principal C-V methods for measuring dopant profiles are compared, and the limitations are examined in detail. Capacitance and thermally stimulated current methods for studying deep traps are reviewed and the interpretation of their results is discussed. There has been increasing concern in recent years with the minority carrier diffusion length and lifetime. Measurement of these properties by photoluminescence, cathodoluminescence, and diode current collection methods is reviewed and the MOS capacitance method for lifetime measurement is also described. The review concludes with a section on optical methods in which the use of luminescence for chemical identification of electrically active impurities is emphasised and direct measurement of properties such as carrier density and epitaxial layer thicknesses is described.

No abstract

This article is a personal review of the principles, capabilities, limitations and potential of the technique of electrochemical capacitance-voltage (C-V) carrier concentration profiling of compound semiconductors and the associated technique of photovoltage absorption spectroscopy. The profiling technique was developed by Ambridge and co-workers to overcome the depth limitation in depletion C-Vprofiling by using an electrolyte barrier to measure the carrier density and to etch the material in a controlled electrolytic process. The electrolyte also provides a transparent barrier which facilitates observation of absorption spectra, hence providing the added capability of band-gap profiling. In this article the basic principles of C-V profiling are summarised, we analyse the balance between measurement accuracy and instrumental depth resolution, and consider the effect of series resistance. In reviewing the principles of electrochemical C-V profiling, we pay particular attention to the electrolyte (Helmholtz) capacitance, the high electrolyte resistance and the definition of contact area. In considering these problems, and those of depth resolution and the influence of deep states, we take account of the use of a fixed low reverse bias in electrochemical C-V profiling compared with an increasing bias in depletion profiling. The interpretation of photovoltage spectra from single layers and heterostructures is described and examples are given of band-gap profiling of laser structures. The article concludes with examples of the characterisation of multiple quantum-well structures including carrier density profiles and photovoltage spectra on structures with periods less than 200 A.

Optical absorption in semiconductor quantum dots coupling to dispersive phonons of infinite modes J. Appl. Phys. 112, 074324 (2012) Electronic structures of single-layer boron pnictides Appl. Phys. Lett. 101, 153109 (2012) The direct and indirect bandgaps of unstrained SixGe1−x−ySny and their photonic device applications J. Appl. Phys. 112, 073106 (2012) Structural, elastic, and polarization parameters and band structures of wurtzite ZnO and MgO J. Appl. Phys. 112, 073503 (2012) Full-zone k.p model for the electronic structure of unstrained GaAs1−xPx and strained AlxIn1−xAs alloys

In many studies, the value of the experimentally determined internal piezoelectric field has been reported to be significantly smaller than theoretical values. We believe this is due to an inappropriate approximation for the electric field within the depletion region, which is used in the analysis of experimental data, and we propose an alternative method. Using this alternative, we have measured the strength of the internal field of InGaN p-i-n structures, using reverse bias photocurrent absorption spectroscopy and by fitting the bias dependent peak energy using microscopic theory based on the screened Hartree-Fock approximation. The results agree with those using material constants interpolated from binary values. © 2005 American Institute of Physics. ͓DOI: 10.1063/1.1896446͔The internal field in GaN based quantum wells plays an important role in the operation of nitride-based light emitting diodes and lasers, affecting the emission wavelength, 1 the oscillator strength, 2 and the recombination lifetime, 3 hence an accurate value of the internal field is essential in understanding the properties of these devices. The internal field skews and breaks the symmetry of the well, causing spatial separation of the electron and hole wave functions and hence reduces the electron-hole overlap function. Reported values of the internal field 4,5 for the same nominal indium content vary by more than a factor of two, which is far greater than the expected error due to unintended variations in the indium content. Also there are large reported differences between theoretical and experimental results. 6 The majority of approaches to determine the internal field, have relied upon counteracting the quantum-confined Stark effect with an externally applied reverse bias and measuring properties of the quantum well as a function of this applied reverse bias. The reverse bias acts to oppose the internal field reducing the effect of the induced quantum confined Stark effect. At low bias, the well is skewed due to the internal field. At a critical bias, the contributions from the applied bias and the internal field are equal and opposite. In this case, the overlap of electron and hole wave functions and the ground state electron to heavy hole transition energy are maximized.The value of the externally applied bias to achieve flat band ͑"square-up" the quantum well͒ can then be used to obtain the internal field. The net internal field E in the well ͑E = 0 when the well is square͒ is related to the applied bias V using: 4where E int , L w , N, 0 , d d , and d u are the internal field, the quantum well width, the number of quantum wells, the built-in potential and the depletion and intrinsic widths, respectively. The width of the intrinsic region d u is given by the sum of the multiple well and barrier widths. The internal field E int , is the sum of the fields due to the piezoelectric effect and the spontaneous polarization. The first term of Eq. ͑1͒ is the background field written as the total voltage drop divided by the distance, over ...

No abstract

Pulsed light-current characteristics of InGaN/GaN quantum well light-emitting diodes have been measured as a function of temperature, with sublinear behavior observed over the whole temperature range, 130-330 K. A distinctive temperature dependence is also noted where the light output, at a fixed current, initially increases with temperature, before reaching a maximum at 250 K and then decreases with subsequent increases in temperature. On the basis of a drift diffusion model, we can explain the sublinear light-current characteristics and the temperature dependence by the influence of the large acceptor ionization energy in Mg-doped GaN together with a triangular density of states function characteristic of localized states. Without the incorporation of localization effects, we are unable to reproduce the temperature dependence whilst maintaining emission at the observed wavelength. This highlights the importance of localization effects on device performance.

We describe a technique for the measurement of optical gain and loss in semiconductor lasers using a single, multisection device. The method provides a complete description of the gain spectrum in absolute units and over a wide current range. Comparison of the transverse electric and transverse magnetic polarized spectra also provides the quasi-Fermi-level energy separation. Measurements on AlGaInP quantum well laser structures with emission wavelengths close to 670 nm show an internal loss of 10 cm Ϫ1 and peak gain values up to 4000 cm Ϫ1 for current densities up to 4 kA cm Ϫ2 .