Capacitance transient

Fig. 3. Capacitance transient spectra from Au-diffused p-type Si showing passivation of the Au donor level (Ev + 0.35eV) after exposure to a hydrogen plasma.

Fig. 8. Capacitance transient spectra recorded under the same conditions for Ar+ sputter etched n-type Ge. (a) substrate temperature 25°C during sputtering, (b) substrate temperature 100°C during sputtering and (c) after sputter etching and hydrogenation at 180 C for 20 min. (Pearton et al., 1983).

Fig. 9. Capacitance transient spectra for (a) sputter-etched p-type Si and (b) after hydrogenation of the sputter-damaged Si at 180°C for 20 min.

Fig. 11. (a) Capacitance transient spectra from Co-60 - irradiated, n-type Si samples, one of which had been pretreated in an H plasma. Note the reduced defect state density in this sample, (b) Concentration profile of the O-V centers induced in these samples. There is a reduced defect concentration only in the region in which atomic hydrogen was incorporated. 

Fig. 11. Capacitive transient spectra of defect states associated with dislocations in ultra-pure germanium (crystal 281 grown along [100] under 1 atmosphere of H2) The micrographs show the etch pits produced by dislocations on a (100) surface. DLTS peak b has an activation energy of Ev + 20 meV. The net-shallow acceptor concentration is 1010 cm-3.

The electrical activity of a defect is characterized in part by its electrical level position, which can be determined by capacitance transient methods. When the capacitance transient spectra are monitored before and after exposure to atomic hydrogen, it is found in many systems that these levels disappear. This phenomenon has been associated with the formation of electrically inactive hydrogen-impurity complexes as summarized by Pear-ton et al. (1987) and in Chapter 5 of this volume. 

Thus measurement of the time constant of either the current or capacitance transient gives e if JVT JVD. Note from Eqs. (18) and (21) that for a current transient i(t) is proportional to e nT(t), whereas the capacitance signal is linear in tijit). This means that if e is large (i.e., the time constant of the transient is short), the current measurement is more sensitive. On the other hand, the capacitance transient is independent of e and therefore more sensitive for slower transients. 

As can be seen from Eqs. (18) and (21), the initial amplitude of the current or capacitance transient is proportional to nT(0). Therefore, if the bias has been kept at zero for a time sufficiently long to fill all traps with electrons, then nT(0) = NT and the trap concentration can be determined from the initial amplitude of the transient. The thermal capture rate is measured by restoring the reverse bias before the traps are completely filled by electrons. Adjusting conditions (low temperature) such that inequality (15) holds, then for a width tf of the filling pulse, the initial amplitude of the trap-emptying transient is given by... 

For a p+-n junction where the edge region is negligible, x, = 0 and x2 = W so that D = 2 and Eq. (25) reduces to Eq. (18). Analogous corrections must also be considered for capacitance transients results for the Schottky diode case have, for instance, been given by Noras and Szawelska (1982). 

Petroff and Lang (1977) combined DLTS with scanning electron microscopy in a method that allows spatial imaging of deep states in the plane of a junction. The scanned electron beam is pulsed on and off and the resulting thermally stimulated current or capacitance transient is analyzed using the usual DLTS methods. 

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... 

Two deficiencies of the current-transient spectroscopy, as compared with the capacitance-transient spectroscopy, involve the determinations of the trap... 

In Table IV we present Eai and Ei0 data on two important deep centers in GaAs, Cr, and O (EL2). The results from three different laboratories are included, but no attempt was made to show everything available in the literature. It is clear that neither the Eai results nor the Ei0 results agree well for Cr, but are not too bad for O. In contrast, the TDH measurements of El0, shown in Table II, are much more consistent. It should be noted that the TDH samples (Table II) were semi-insulating, whereas the emission-spectroscopy samples (Table IV) were conducting in order that capacitance transient (DLTS) experiments could be performed. The PITS and OTCS techniques applied to these samples would have been unable to clearly distinguish between hole and electron traps. 

A3.5 Time-resolved photoluminescence studies of GaN A3.6 Persistent photoconductivity in GaN A3.7 Electrical transport in wurtzite and zincblende GaN A3.8 Characterisation of III-V nitrides by capacitance transient spectroscopy... 

Capacitance transient spectroscopy encompasses a powerful set of techniques to detect and characterise deep levels in semiconductors. The list of techniques applied for III-V nitrides includes deep level transient spectroscopy (DLTS) [1,2], double correlation DLTS (DDLTS) [3], isothermal capacitance transient spectroscopy (ICTS) [2], photoemission capacitance transient spectroscopy (ODLTS) [4] and optical ICTS (OICTS) [5], This Datareview presents the current status of deep level studies by capacitance transient techniques for III-V nitrides. A brief introduction to the techniques is given, followed by an example that demonstrates the application of DLTS and DDLTS for Si-doped... 

The measurements utilise one-sided pn junction (p+n or pn ) and Schottky diodes, which possess a voltage modulated space charge region near the p/n and metal/semiconductor interface, respectively [8], The common feature of capacitance transient techniques is that the electronic properties of deep levels are determined by monitoring the transient high-frequency differential capacitance response of the diode as the electron occupancy of metastably charged deep levels located within the space charge... 

A3.8 Characterisation of111-Vnitrides by capacitance transient spectroscopy D2 p-Type Mg-Doped GaN... 

---