Photoconductivity decay time

The question whether amorphous semiconductors fulfil the condition for relaxation-case semiconductors was raised (Fagen (1972)) because the photoconductive decay time is found to be much longer than the lifetime assumed by van Roosbroeck and coworkers (about 10 sec). However, Ryvkin (1964) and also van Roosbroeck pointed out that in the relaxation regime the photoconductive decay is not governed by the lifetime but by the much longer relaxation time. 

Fig. 4. Energy below the conduction band of levels reported in the literature for GaP. States are arranged from top to bottom chronologically, then by author. At the left is an indication of the method of sample growth or preparation liquid phase epitaxy (LPE), liquid encapsulated Czochralski (LEC), irradiated with 1-MeV electrons (1-MeV e), and vapor phase epitaxy (VPE). Next to this the experimental method is listed photoluminescence (PL), photoluminescence decay time (PLD), junction photocurrent (PCUR), photocapacitance (PCAP), transient capacitance (TCAP), thermally stimulated current (TSC), transient junction dark current (TC), deep level transient spectroscopy (DLTS), photoconductivity (PC), and optical absorption (OA).

The time-resolved photoconductivity measurements shown in Fig. 15 give further support for a difference in the photoinduced charge transport in the polymerized samples versus the unpolymerized samples. For the incident laser of 100 mW/cm2 and a spot size of 2.5 mm, the decay time of the photoconductivity for the unpolymerized samples is 7.4 sec, whereas the photoconductivity of the polymerized samples does not significantly drop over a 30 sec period. Also, the photoconductivity of the polymerized sample is nearly twice that of the unpolymerized samples even at the peak of the unpolymerized photoconductive response. The unnormalized values for the dark conductivity in both samples is 1.7 x 10-10 S cm-1. The photoconductivity is 5.8 x 10-11 S cm-1 for the unpolymerized sample and 1.1 x 10-10 S cm-1 for the PSLC at an optical intensity of 2 W cm-2. 

The beam coupling, four-wave mixing, and photoconductivity data indicate that the mechanism for charge transport within the PSLCs is dramatically different than for the unpolymerized samples. The unpolymerized samples should exhibit a grating decay time that decreases quadratically with fringe spacing. This is supported by the data in Fig. 14b [47,80], However, the data for the... 

The photoconductivity response of a-Si H nipi structures has an extremely long recombination lifetime. A brief exposure to illumination causes an increase in the conductivity which persists almost indefinitely at room temperature (Kakalios and Fritzsche 1984). An example of this persistent photoconductivity is shown in Fig. 9.32. The decay time exceeds 10 s at room temperature and decreases as the temperature is raised, with an activation energy of about 0.5 eV. 

Since it is important to address this issue at the earliest times following photoexcitation, measurements of transient photoconductivity in the picosecond to nanosecond regime were carried out [145,146,201,202], In response to an ultrafast light pulse (duration 25 ps), there is an initial fast photocurrent response with decay time of about 100 ps followed by a slower component with... 

The photoconductive decay (PCD) lifetime measuring technique was one of the first lifetime characterization methods to be used. [70] As the name implies it uses optical excitation of e-h pairs. The carrier decay is monitored as a function of time following the termination of the optical pulse. Traditionally the sample is provided with contacts and the current is measured as a function of time. More recently, non-contacting techniques have been developed that make the method attractive because it is fast and non-destructive. 

The abihty to accept and hold the electrostatic charge in the darkness. The photoconductive layer should support a surface charge density of approximately 0.5-2 x 10 C/cm. The charge also has to be uniformly distributed along the surface, otherwise nonuniformities can print out as spot defects. The appHed surface potential should be retained on the photoreceptor until the time when the latent electrostatic image is developed and transferred to paper or, if needed, to an intermediate belt or dmm. In other words, the "dark decay" or conductivity in the dark must be very low. The photoconductor materials must be insulators in the dark. 

When a semiconductor is illuminated, electrons may be excited into the conduction band and/or holes into the valence band, producing photoconductivity. This excited condition is not generally permanent, and when the illumination ceases, the excess current carriers will decay, or recombine. The average time which a photoelectron remains in the conduction band is termed the lifetime. As the lifetime increases, the photocurrent, for a given intensity of illumination, increases. 

Photoconductive response (the rate of creation, or the rise, of the photocurrent, and the rate of decay of the photocurrent) appears to be divided into fast and slow responses. The fast responses, with time constants for rise and decay of the order of a second or less, have been adequately interpreted by Mollwo, et al. (53-55), Weiss (56), and Heiland (47,57) as bulk processes. These authors have concluded that the fast response processes are associated with the double ionization of interstitial zinc, and have proposed that the photon excites electrons from the valence band, and that the hole immediately recombines with the electron from an interstitial Zn+, producing double-ionized zinc ions. 

A few of his experimental results will be briefly described. A typical photoconductive rise at room temperature was found to reach half of the maximum conductivity change in a time of the order of an hour after the light was switched on. A similar period was required for half the decay to occur after the light was switched off. The rates of the rise and decay decreased with increasing time, the conductance approaching equilibrium. ... 

It was observed that the increase of photoconductivity under illumination was retarded by high oxygen pressure that is, the photoconductivity, at a given time and intensity of illumination, has a lower value if the oxygen pressure is high. On the other hand, decay of the induced conductivity was hastened by increased oxygen pressure. Nitrogen, water vapor, or carbon dioxide had no effect. 

Studying the effect of temperature changes, Melnick observed that the photoconductivity could be frozen in. For example, he found, while studying the decay of the photoconductivity at room temperature, that if the temperature was suddenly lowered to 130°K, the slow decay ceased. The photoconductivity was frozen in. If, at some later time, the temperature was raised again to room temperature, the decay recom-... 

From an analysis of the rate of decay of the photoconductance as a function of temperature, Melnick concluded that the barrier height was 0.43 e.v., as compared to Morrison s (31) estimate on a different sample of about 0.5 e.v., the latter being based on measurements of the reversible portion of the time-dependent conductivity. 

In this expression, q is the surface concentration of the adsorbed gas, ci and C2 are constants, and t is the time. By relating the conductance to the amount of adsorption, he derived an expression for the decay of the photoconductance due to adsorption... 

The photoconductivity increases when the a-Si H is lightly doped with phosphorus (Anderson and Spear, 1977). However, phosphorus doping causes very slow decay of photoresponse. The photoresponse characteristic for the phototconductive sensor using undoped a-Si H is shown in Fig. 3. The illumination is the modulated light from a GaP LED. The modulation ratio is defined as M = (it — i2)/i2, where is the peak photocurrent and i2 is the bottom current just prior to the next pulse. Figure 4 shows the modulation ratio of a-Si H versus the pulse width T, compared to that of the CdS-CdSe photoconductive sensor. The CdS-CdSe sensor modulation ratio decreases as the repetition time becomes shorter. On the other hand, in the a-Si H photoconductive sensor, the modulation ratio does not decrease... 

Recent measurements of fast transient photoconductivity (11) in trans-fCm have demonstrated that the photogenerated soiitons are mobile and contribute to the electrical conductivity. Figure 2 shows the transient photoconductivity following a 1 pJ pulse at 2.1 eV with a bias voltage of 300 V. The charge carriers are produced within picoseconds of optical excitation. The fast rise is foiiowed by (approximateiy exponentiai) decay with a time constant of - 300 ps. The magnitude and time decay of Oph(t) are temperature independent... 

They also observed that the species absorbing at 720 nm which decays on the microsecond time-scale by first-order kinetics is a long-lived SSIP. Evolution of the SSIP to free-radical ions was ruled out, because free-radical ions should decay by second-order kinetics. Later observations by Mataga [159, 160] and by Haselbach [161], on the basis of photoconductivity measurements and transient absorption spectroscopy of the benzophenone-DABCO system, showed that the species absorbing at 710 nm decays not by first-order kinetics but by second-order kinetics this is consistent with the formation of free-radical ions. 

Many kinetic studies of the thermal decomposition of silver oxalate have been reported. Some ar-time data have been satisfactorily described by the cube law during the acceleratory period ascribed to the three-dimensional growth of nuclei. Other results were fitted by the exponential law which was taken as evidence of a chain-branching reaction. Results of both types are mentioned in a report [64] which attempted to resolve some of the differences through consideration of the ionic and photoconductivities of silver oxalate. Conductivity measurements ruled out the growth of discrete silver nuclei by a cationic transport mechanism and this was accepted as evidence that the interface reaction is the more probable. A mobile exciton in the crystal is trapped at an anion vacancy (see barium azide. Chapter 11) and if this is further excited by light absorption before decay, then decomposition yields two molecules of carbon dioxide ... 

Figure 8 shows the growth and decay curves (the kinetics) of these three phenomena—photoconductivity, light induced polarization and the electron spin resonance signal. All of them have the same unimole-cular time constants at 25°C. When the system is cooled to —100°, which has been done for the photoconductivity and the spin signal, they decay faster at the lower temperature but again they are parallel they have the same kinetic behavior. 

---