Semiconductor Lifetime Research Articles

Detailed interpretation of photoconductivity decay response for lifetime determination

Photoconductive decay provides a method for estimating the recombination lifetime in semiconductors. It often exhibits a single characteristics time. For samples having large densities of recombination centers, previous experiments indicate that this characteristic time can considerably exceed the steady state lifetime. We present apparently the first explanation of these results. The explanation involves the effect on the overall photoconductive decay of the multiple characteristic times that describe trapping and recombination through bound states for variations in the electrochemical potentials that are small relative to the thermal voltage.

Radiative lifetime in semiconductors for infrared detection

The radiative lifetime of excess carriers calculated by van Roosbroeck and Shockley is re-examined. Due to reabsorption effects, this lifetime is an underestimate of what would be measured in the bulk. It is shown that this extension of the lifetime can be considered as effectively noiseless. The consequences of this for infrared detector operating temperatures are discusssed, with the conclusion that the state of the art is much further from the fundamental limit than had been supposed and substantial further improvements are possible.

Modification of the transient capacitance analysis of GaAs MIS structures for minority carrier lifetime determination

The insulators of GaAs MIS structures, because of their low temperature deposition, have relatively low resistivities compared to SiO2. This is responsible for finite leakage through the insulator when a voltage is applied across the MIS structure. The present work reports on the implications of leakage on the transient capacitance plot which is widely used for the determination of minority carrier lifetime in semiconductors.

The concept of generation and recombination lifetimes in semiconductors

The purpose of this brief is to discuss the concept of generation and recombination lifetimes, a concept frequently confused in the literature. The regimes of device operation where they apply is discussed and it is shown experimentally for the first time that the two can be very different in magnitude.

A systematic study of positron lifetimes in semiconductors

Positron lifetimes for the ten semiconductors Si, Ge, alpha -Sn, GaP, GaAs, GaSb, InP, InAs, InSb, and CdTe as well as for diamond and SiC have been measured. The bulk-annihilation rates range from 8.7 ns-1 for diamond to 3.44 ns-1 for CdTe and agree well with theoretical calculations. The effect of doping of GaAs was investigated with four different concentrations of Si and of Cd and Cr. In contrast to Si, significant doping effects are present in GaAs.

A method of determining minority carrier lifetime and doping level profiles of low-doped metal-oxide-semiconductor structures using pulsed photoinjection

This method. aimed at determining doping level and minority carrier lifetime of low-doped semiconductors (<1014/cm3 in silicon) is based on pulsing the metal-oxide-semiconductor device into deep depletion. A delayed photopulse is then applied to the sample, which causes partial collapse of the depletion region. The changes in fill time and in capacitance versus collected photocharge are measured. The minority carrier lifetime is computed through the dependence of fill time on the magnitude of the photoinjected charge. The doping level is determined by the change in capacitance following the photoinjection. The method is advantageous in (a) independently supplying the doping level and lifetime, (b) being insensitive to edge injection, and (c) enabling the determination of the diffusion length.

A quick method for the determination of bulk generation lifetime in semiconductors from pulsed MOS capacitance measurements

The high frequency capacitance transient that is observed when an MOS (Metal-Oxide-Semiconductor) capacitor is pulsed from accumulation to deep depletion condition is usually analysed graphically to determine the values of bulk generation rates or of minority carrier lifetimes in the semiconductor concerned. However, this analytical process is a long and laborious one and in this paper a quicker method for the determination of lifetime is suggested. The extracted lifetime value is typically within 10% of the true value and is sufficient for applications such as on-line process control.

Diffusion-limited lifetime in semiconductors

At high recombination rates, minority carrier recombination at defects in crystals ceases to obey Shockley-Read theory, and diffusion theory should be applied. These theories for neutral plane, line, and point defects are compared and used to interpret experimental results on hole recombination in $n$-type GaP. The steady-state kinetics and spatial distribution of the recombining carriers are briefly discussed but most attention is placed on the transient lifetime of the carriers for comparison with experiment. With no recourse to disposable parameters, the concentration and temperature dependences for plane sinks (surfaces) and line sinks (dislocations) in $n\ensuremath{-}\mathrm{G}\mathrm{a}\mathrm{P}$ are in excellent agreement with the diffusion theory. The resulting lifetimes are approximately independent of the sink strength (capture distance) and represent a lower limit to the observable lifetime for a fixed (low) defect concentration. For point sinks such limitation is not predicted and the experimental lifetime data do not agree with the diffusion theory. However, extending the diffusion theory to the case of a negatively charged defect, using a modified "Lax" treatment, produces concentration and temperature dependences in good agreement with experiment, and again a similar lifetime limitation is predicted providing the effective capture radius does not exceed 2\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}5}$ cm ($\ensuremath{\sigma}\ensuremath{\sim}{10}^{\ensuremath{-}11}$ ${\mathrm{cm}}^{2}$).

Lifetime measurement in Hg0.7Cd0.3Te by population modulation

A new contactless optical modulation technique for the determination of photogenerated carrier lifetimes in semiconductors is presented. The technique consists of measuring the modulation in the transmitted intensity of a dc probe beam (h/ω<Eg) due to a modulated pump beam (h/ω≳Eg). The fractional change in the probe beam transmission is proportional to the excess carrier lifetime. Data are presented for lifetimes of 0.1–4 μ sec measured by this technique in p-type and near-intrinsic Hg0.7Cd0.3Te samples at 300 K.

The influence of surface properties on minority carrier lifetime and sheet conductance in semiconductors. (GB)

The influence of surface properties on minority carrier lifetime and sheet conductance in semiconductors. (GB)

The concept of screening length in lifetime and relaxation semiconductors

The concept of screening length in lifetime and relaxation semiconductors

The concept of screening length in lifetime and relaxation semiconductors

It is shown that, contrary to normal practice, the most appropriate criterion for distinguishing between lifetime and relaxation semiconductors in the presence of traps is the ratio of the screening length L s , to the ambipolar diffusion length L Da ,. L s , is calculated. Its significance is not limited to zero current, even though it reduces to the conventional Debye length L D when the trap concentration is zero. (With traps, we always have L s < L D .) The dielectric relaxation time itself is unaffected by traps, but in steady state situations, a material behaves as if it had an effective lifetime τ 0s = τ 0η , where η depends on the concentration and energetic position of the traps. τ o, may be orders of magnitude greater than τ 0, the conventional diffusion length lifetime. Typical values of L s , are presented as a function of trapping parameters. L s > L Da leads to relaxation behavior; L s < L Da , to lifetime behavior.

Lifetime determination of MOS diodes with very thin oxide by their photocurrent spectra

A new method for measuring the minority-carrier lifetime in semiconductors is presented. The bulk lifetime in a wide range, 10−11–10−4 sec, can be determined from a simple analysis of photocurrent spectra of MOS tunnel diodes or of Schottky barrier diodes under deep depletion condition.

Bulk boundary conditions for injection and extraction in trap-free lifetime and relaxation semiconductors

Small-carrier perturbations close to the unperturbed bulk are analyzed which represent boundary conditions for the perturbations produced by injection and extraction in thick enough semiconductor samples. On this basis concentration contours for the strong-perturbation range can be qualitatively drawn and even numerically evaluated. The results allow for a clear distinction to be made between lifetime and relaxation behavior. There is also an intermediary regime between these, and within it a new kind of steady state with oscillations in space is proved to be possible for appropriately chosen currents. Two new conditions have also been evidenced for the high current conduction in the asymptotic range: "equality recombination" in the deep-lifetime case, and "unperturbed conductivity" in the deep-relaxation case. Necessary conditions for obtaining an increase in resistance through injection have also been outlined.

Hall effect in finite specimens of arbitrary lifetime and trap content

This paper presents a linearized Hall-effect theory, applicable to specimens of arbitrary size, carrier concentration, lifetime, dielectric relaxation time, concentration of (single level) recombination centers, and surface properties. Along the lines of Van Roosbroeck and Casey, the distinction is made between the behavior of lifetime and relaxation semiconductors. By analogy with Popescu and Henisch, who dealt with injection, the Hall effect also involves "lifetime regimes" and "relaxation regime," depending on whether carriers are augmented or depleted in certain regions. The present analysis, which avoids the a priori assumption of space-charge neutrality, is one-dimensional and limited to small magnetic fields. The boundary conditions at the Hall electrodes are assumed to depend on effective surface recombination velocities for electrons and holes of arbitrary value. The results show that the corrective terms involving surface properties and sample size can be of either sign and of magnitude comparable to the value, of the Hall voltage calculated by conventional method. As expected, this is most important when the specimen dimensions in the direction of the Hall field are of the same order as the ambipolar diffusion length. The typical examples calculated include high-lifetime germanium.

New results from linearised charge transport theory

PROBLEMS involving the simultaneous transport of electrons and holes by drift and diffusion arise in connection with a great variety of solid-state phenomena and devices. Their analysis is handicapped because transport equations defy explicit solution for the general case. The choice must be made between computer-derived numerical solutions of the complete equations for particular parameters and boundary conditions, or explicit treatment of the equations with sufficient ad hoc assumptions (for example, bulk neutrality, zero recombination, freedom from traps) to permit their solution. The former is cumbersome and excessively specific for many purposes; the latter is often misleading, because the precise consequences of ad hoc assumptions cannot be easily foreseen. It is hoped that they affect only the accuracy of the results, but there are many instances when ‘simplifying’ assumptions distort the entire physical picture. We report here a third possibility, namely explicit solutions of the equations in linearised form, but without any other simplifying assumptions. The significance of these results is limited to low currents (‘small signal theory’) but the conclusions are otherwise general, and permit the convenient examination of the various processes as a function of the transport parameters. Moreover, the same set of equations can be applied to various problems by suitable adjustment of the boundary conditions. Such procedures have been used with two distinct problems: (1) the injection of minority carriers into lifetime and relaxation semiconductors, with and without traps1,2, and (2) the Hall effect3 in specimens of finite size. The analysis even in linearised form is complex, and the details are published elsewhere, but explicit solutions can be obtained. They lead to new conclusions on the nature of the process involved.

Minority-carrier injection into semiconductors containing traps

The paper presents an analysis of minority-carrier injection into lifetime and relaxation semiconductors containing traps, linearized within the framework of a small-signal theory. In that sense it parallels a previous paper on the trap-free case. The restrictive assumptions customarily made in the literature as regards injection ratio, mobility ratio, carrier lifetime, dielectric relaxation time, doping levels, and specific characteristics of the trapping centers, zero recombination rate $np=n_{i}^{2}$, and neglect of diffusion current, are avoided here. Explicit solutions are obtained of the carrier concentration and field profiles, and plotted for a series of interesting cases, some designed to illustrate the nature of the phenomena, some to facilitate experimental verification. The results show that the establishment of lifetime and relaxation regimes depends in a complex manner on the parameters of the system. Minority-carrier injection can result in the appearance of a field maximum and a total resistance increase, not only in the trap-free case as previously reported, but to a greatly augmented extent in the presence of traps in suitable concentrations and energetic positions. The results have a potential bearing on the interpretation of many types of electrical measurements on semiconductors and semi-insulators. The equations themselves are general (except for the restriction to small currents) and can be extended to a variety of other nonequilibrium transport effects in solids.

Photoconductivity null apparatus for the determination of minority carrier lifetime

A variable time constant exponential pulse generator used to null exponential photoconductivity decay signals is described. Circuitry using this generator for the direct readout of minority carrier lifetime of semiconductors is also presented.

Cathodoluminescence measurements of the minority-carrier lifetime in semiconductors

Detailed calculations of the cathodoluminescence decay after an excitation pulse are developed; the generation function, the carrier diffusion, the surface recombination, the self-absorption, and the excitation-pulse duration are taken into account. It is demonstrated that the surface recombination leads to a nonexponential region at the beginning of the decay when short excitation pulses are used. This effect disappears with longer pulses. It can be used in order to determine separately the surface recombination velocity and the true bulk lifetime of the carriers. Nomograms are given which permit this determination without further calculation. The influence of the pulse duration on the form of the cathodoluminescence decay has been experimentally verified in GaP. The surface recombination velocities and the lifetimes in many samples have been deduced. Finally, it appears that this method provides a simple means of determining the minority-carrier lifetime in semiconductors.

Minority-carrier injection into semi-insulators

A previous paper was concerned with the injection of minority carriers into relaxation semiconductors without traps. The present paper considers traps in their dual role as recombination centers and charge-storage localities. It is shown that the presence of traps changes the boundary between majority-carrier depletion ($\ensuremath{\Delta}n &lt; 0$) and the majority-carrier augmentation ($\ensuremath{\Delta}n &gt; 0$), moving the system more towards the latter, compared with expectations based on the trap-free case. Trapping of injected minority carriers also leads to an increase of majority-carrier diffusion in opposition to the current, and the containment of this diffusion calls for higher-than-normal electric fields. As a consequence, the injection region can have a higher-than-normal resistance, qualitatively as predicted by Van Roosbroeck, but for quite different reasons. Computed contours show the corresponding charge concentration and field profiles. The resistance enhancement effect is expected to be most prominent in lifetime semiconductors of high trap density; conditions for its appearance (if any) in relaxation semiconductors are discussed. Questions relating to the definition of relaxation semiconductors are also discussed.