Deep level measurements

In this Datareview, we concentrate on deep levels measured by capacitance and admittance techniques those measured by other techniques are detailed in Datareview 4.1. For completeness, trap parameters for major defects and impurities obtained from all techniques are listed. Capacitance techniques have proven useful for the characterisation of deep states in semiconductor devices. In particular, states which are non-radiative can be analysed by this technique. If the state under study is one which principally determines the conductivity of the crystal, the techniques of admittance spectroscopy are used. The set-up for doing capacitance and admittance spectroscopy on SiC is identical to that used for other semiconductors with the exception of the necessity to operate the system at higher temperatures in order to access potentially deeper levels in the energy gap. The data are summarised in TABLE 1. 

An argument against the defect mediated diffusion model is the same one used earlier that is, there are not enough defects as determined by ESR (Brodsky and Title, 1969, 1976) or Deep Level Transient Spectroscopy measurements (Johnson, 1983) to account for the motion of all of the bonded hydrogen in a-Si H. This objection is removed if the floating bonds are 104-106 times more mobile than the hydrogen atoms. However, such highly mobile defects would rapidly self-annihilate via the process. 

The lifetime of the core-ionized atom is measured from the moment it emits a photoelectron until it decays by Auger processes or X-ray fluorescence. As the number of decay possibilities for an ion with a core hole in a deep level (e.g. the 3s level) is greater than that for an ion with a core hole in a shallow level (e.g. the 3d level), a 3s peak is broader than a 3d peak. 

The resistivity of the SI wafers has been measured to be in excess of 10" Q-cm at room temperature [35]. As mentioned earlier, a deep level defect was found to have an activation energy of 1.4 eV. MESFETs manufactured on these wafers show an increased performance in the sense of reduced trapping, which the authors explain as being primarily due to a reduction of the shallow impurities in the material. 

Deep level experiments can be divided into thermal and optical categories, depending on which constants are being measured. Both categories can be further subdivided into steady state and transient techniques. An example... 

Deep state experiments measure carrier capture or emission rates, processes that are not sensitive to the microscopic structure (such as chemical composition, symmetry, or spin) of the defect. Therefore, the various techniques for analysis of deep states can at best only show a correlation with a particular impurity when used in conjunction with doping experiments. A definitive, unambiguous assignment is impossible without the aid of other experiments, such as high-resolution absorption or luminescence spectroscopy, or electron paramagnetic resonance (EPR). Unfortunately, these techniques are usually inapplicable to most deep levels. However, when absorption or luminescence lines are detectable and sharp, the symmetry of a defect can be deduced from Zeeman or stress experiments (see, for example, Ozeki et al. 1979b). In certain cases the energy of a transition is sensitive to the isotopic mass of an impurity, and use of isotopically enriched dopants can yield a positive chemical identification of a level. 

We now discuss some of the experimental aspects of temperature spectroscopy. Lang (1974) called his original method deep level transient spectroscopy (DLTS), and he measured capacitance transients produced by voltage pulses in diodes made from conductive materials. However, in SI materials, this method is not feasible and an alternate method, involving current transients produced by light pulses in bulk material (or Schottky structures), was... 

---