Pulsed MOS Capacitor Research Articles

Electrically driven light emission from single quantum dots using pulsed MOS capacitors

Robust, nanoscale light-emitting devices are attractive for emerging photonic and quantum engineering applications. However, conventional approaches suffer from fabrication challenges or lack the potential to address individual nanostructures, such as quantum dots. In this paper, we present a device that can produce electrically driven light emission from a single quantum dot using a single carbon nanotube as the charge injection contact. The device has a metal-oxide-semiconductor capacitor structure and operates based on an unconventional mechanism of electroluminescence, which relies on the use of bipolar voltage pulses. The proposed device can be fabricated in a simple manner using conventional lithographic processes, offering a scalable approach toward the development of optoelectronic devices at the single dot level.

Carrier Recombination Lifetime Measurement in Silicon Epitaxial Layers Using Optically Excited MOS Capacitor Technique

Carrier recombination lifetime is a key semiconductor parameter that not only has a major role in how a variety of solid-state devices operate, but one that can also be used as a process cleanliness monitoring tool. Lifetime measurement on epitaxial wafers, where the epilayer thickness is smaller than the minority carrier's diffusion length, has always been a challenging task. Although the pulsed MOS capacitor has been shown to be an eminently suitable technique for measuring the generation lifetime (τ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">g</sub> ) on these wafers, measuring the recombination lifetime (τ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">r</sub> ) yet has remained complicated and difficult to implement. In this paper, a new technique for accurate measurement of τ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">r</sub> will be presented. Lifetime will be extracted from the capacitance transient (C-t) of an inverted MOS device while being excited by an optical pulse. In addition to its easy implementation, the recombination lifetime extracted this way is least affected by Si/SiO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> surface and epi/substrate interface effects when compared with older techniques such as the photoconductance decay. TCAD simulations and experimental results will be presented to demonstrate the promising application of the optically excited MOS to the characterization of p/p <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">+</sup> silicon epitaxial layers.

Pulse Width Modulation for Reducing Pulsed MOS Capacitor Measurement Time

Carrier lifetime in silicon is a direct measure of its deep-level defect density. This parameter is therefore widely used for process screening purposes to check for unwanted contaminations such as metallic impurities. The dynamics of semiconductor industry requires the development of novel approaches to reduce the measurement time of commonly established carrier lifetime measurement techniques for more efficient screening in process lines. The pulsed MOS capacitor (MOS-C) is the most popular technique for measuring the generation lifetime (τg ) where τg is being extracted from the capacitance transient response (C-t) of a MOS capacitor device when pulsed from accumulation into deep-depletion and as it moves from the non-equilibrium deep-depletion into the inversion. Although accurate and easy to implement, the pulsed MOS capacitor takes a long time to accomplish. For clean wafers the time for a single measurement can easily reach several hours. This becomes specifically problematic when the goal is to map the whole wafer area. Many attempts have been made to reduce the lifetime measurement times. The challenge, however, is to maintain the accuracy while trying to reduce the test time. It will showed that by using a dynamic pulse train and exploiting the pulse width, the pulsed MOS-C technique can be implemented not only to reduce the measurement time. If the accumulation pulse in the pulsed MOS technique is applied at a high frequency the device response will become substantially short resulting in a reduced measurement time. Interestingly, this approach will not cause any compromise in the accuracy with which the lifetime is measured. TCAD simulations and experimental results will be presented to show the effect of the accumulating pulse width on the C-t transient of a typical MOS device.

Pulse Width Modulation for Reducing Pulsed MOS Capacitor Measurement Time

Semiconductor manufacturing economics necessitate the development of novel approaches to reduce the measurement time for commonly established lifetime measurement techniques. Although accurate and easy to implement, the pulsed MOS-Capacitor technique takes a long time to accomplish. This becomes problematic when the goal is to map the whole wafer area. We show that by using a dynamic pulse train and with manipulating the accumulation pulse width, the pulsed MOS-C technique can be implemented in much faster time frames while its accuracy is being kept constant.

Carrier Generation Lifetimes in 4H-SiC MOS Capacitors

The field of SiC electronics has progressed rapidly in recent years, but certain electronic properties remain poorly understood. For example, a consensus has not been reached as to the specific point defects which limit minority carrier recombination, and little is known about defects which limit generation lifetimes. This paper investigates generation lifetimes using the pulsed MOS capacitor technique and compares the results with defect densities, recombination lifetimes, and Schottky diode characteristics in the same material for the first time. Carrier generation lifetimes in 4H-SiC epilayers range from less than 1 ns to approximately 1 μs and depend strongly on measurement conditions and data interpretation. They are limited by dislocations only at densities higher than 106 cm-2.The only point defect that is theoretically capable of limiting generation lifetime to the levels currently observed in 4H-SiC is EH 6/7. However, this defect cannot account for the case where generation lifetimes are lower than recombination lifetimes in the same area. This is not seen in silicon and seems to be inconsistent with theory. Possible reasons for these perplexing results are discussed, and it is attempted to form a framework with which further understanding of the significance of carrier generation lifetime measurements in SiC can be achieved.

Characterization of Strained Si/SiGe With Pulsed MOS Capacitor and Gate Oxide Integrity Measurements

not Available.

Determining the generation lifetime in a MOS capacitor using linear sweep techniques

A comparison between two linear sweep techniques for generation lifetime profiling is reviewed here. These semiconductor characterization techniques find widespread application because of their availability in commercial equipment and ability to reduce the measurement time. It is shown experimentally that parameters such as generation lifetime and surface generation velocity determined by linear sweep techniques agree well with those obtained from pulsed MOS capacitor measurement.

Modeling of drain current overshoot and recombination lifetime extraction in floating-body submicron SOI MOSFETs

This work reports on a new general modeling of recombination-based mechanisms related to electrically floating-body partially-depleted (PD) SOI MOSFETs. The model describes drain current overshoots induced when turning on the transistor gate and suggests a novel extraction method for the recombination lifetime in the silicon film. We show that the recombination process associated with drain current overshoots in PD silicon-on-insulator (SOI) MOSFETs takes place mainly in the depletion region and not in the neutral region as in case of pulsed MOS capacitors. Associated with existing techniques for generation lifetime extraction, our model offers, for the first time, the possibility of complete and rapid characterization for both generation and recombination lifetime using drain current transients in floating-body SOI MOSFETs. The model is used in order to characterize submicron SOI devices, allowing a thorough investigation of technological parameters impact on floating-body-induced transient mechanisms.

Frequency domain lifetime characterization

Time-based measurements are commonly used for lifetime characterization of semiconductors. We have developed the theory, verified by experiment, of frequency-based lifetime characterization as an alternative to time-based measurements for MOS devices biased in inversion. One consideration during lifetime/diffusion length measurements, is whether the near-surface space-charge region or the bulk or quasineutral region is characterized. To characterize the near-surface space-charge region of the device, one usually makes room temperature pulsed MOS capacitor or diode leakage current measurements. We show that room-temperature, frequency-domain capacitance, conductance, or resistance measurements characterize the quasineutral bulk, not the space charge region, in contrast to room-temperature pulsed MOS-C or diode leakage current measurements which characterize the space-charge region.

Iron contamination in silicon wafers measured with the pulsed MOS capacitor generation lifetime technique

The pulsed MOS capacitor generation lifetime technique is used to determine the iron density in boron-doped silicon wafers. Effective generation lifetimes (/spl tau//sub g,eff/) are extracted from the Zerbst plots obtained from the measured capacitance-time (C-t) data. Upon thermal heating at 200/spl deg/C for 5 minutes and quenching to 23/spl deg/C, iron-boron (Fe-B) pairs dissociate into interstitial iron (Fe/sub i/) and substitutional boron (B). The post-heated /spl tau//sub g,eff/ decreases immediately after heating. As time elapses (pairing time t/sub p/ increases) after Fe-B dissociation, T/sub g,eff/ increases because Fe/sub i/ reforms into Fe-B pairs. It takes about four times the time constant (i.e., t/sub p//spl ap/4/spl tau/) of Fe-B pairing reaction before the post-heated /spl tau//sub g,eff/ recovers to the pre-heated /spl tau//sub g,eff/. An expression is developed to determine the iron density. The iron density obtained from this expression shows good agreement with that measured by deep-level transient spectroscopy.

Measurement time reduction for generation lifetimes

The defect properties of thin semiconductor layers (epitaxial and denuded zones as well as silicon on insulator) are not as easily characterized as those of bulk wafers. Recombination lifetime or diffusion length measurements, routinely used for bulk wafers, are unreliable when the layer thickness is significantly less than the diffusion length. Generation lifetime measurements are eminently suitable for such characterization because the characterized thickness of the sample is determined by the space-charge region width. However, the measurement times become inordinately long for high quality layers when pulsed MOS capacitors are used as test structures. Times of hundreds or even thousands of seconds per measurement are not uncommon. We present two methods to reduce this measurement time: optical excitation and elevated temperature. Optical excitation can only be used if the light is pulsed and the lifetime is measured from the remaining "dark" curve. Steady-state light leads to erroneous lifetimes and should not be used. Elevated temperature measurements lead to temperature-varying lifetimes, because the generation lifetime may be temperature dependent.

Effect of RPT/furnace processing on the minority carrier lifetime in very thin oxide MOS capacitors

In this paper the effects of oxidation and annealing processes on the minority carrier lifetime in silicon have been investigated. Using the pulsed MOS capacitor method a comparison of two methods used to grow very thin SiO 2 layers, conventional furnace processing and rapid thermal processing (RPT), has been carried out. It is shown that those substrates which underwent RT high temperature steps sustained less damage, probably due to the lower contamination that results from this method.

An explanation of field-enhanced non-equilibrium transients of pulsed MOS capacitors

A method has been formulated to explain field-enhanced non-equilibrium transients of pulsed MOS capacitors. In the author's theory the Coulombic emission of trap centres is assumed to be field dependent. The one-dimensional Poole-Frenkel model is used to describe the Coulombic emission probability of trap centres, which is dependent on the applied electric field. On the other hand, unlike previous theories, the contribution of carrier generation from non-Coulombic emission of trap centres is also considered. A different equation is founded to describe the change of the width of the space charge region with time, and then the MOS capacitance and gate current are related to the width of the space charge region. In such a manner the non-equilibrium capacitance-time and current-time transients are given. The comparison between theory and experiment has been made and shows that the author's theory can well explain the experimental results.

Capacitance—time transient characteristics of pulsed MOS capacitor application in measurement of semiconductor parameters

Using Rabbani's model for generation width, a differential equation, which describes the capacitance–time (C–t) transient characteristics of pulsed MOS capacitors was obtained. The theoretical C–t transient characteristics can be obtained by integrating this differential equation. It has also been shown that the minority generation lifetime of semiconductors can be determined by matching an experimental C–t transient characteristic with the theoretical one.

A method of determining generation lifetime suitable for computer-aided measurement

A new method of determining generation lifetime is suggested. By reading out several values of high-frequency MOS capacitance with equal time intervals from a capacitance-time transient curve of a pulsed MOS capacitor, the generation lifetime of minority carriers can be determined. This method involves straight line regression and is suitable for computer-aided measurement.

Description of non-equilibrium phenomena in pulsed MOS capacitors

A theoretical investigation of the non-equilibrium phenomena of pulsed MOS capacitors is presented. Unlike treated previously, a time-dependent minority carrier generation rate of bulk traps is considered and an improved model for the generation width is applied. A differential equation is found, which describes the change of the space-charge region in the non-equilibrium transients. The dependence of MOS capacitance and current on the width of the space-charge region is ascertained. In such a manner we can describe the non-equilibrium transients of pulsed MOS capacitors more accurately than in previous works.

A method for the determination of the generation sources, bulk lifetime, and surface generation velocity using MOS capacitor

Various pulsed MOS capacitance techniques have been developed for the determination of the generation lifetime and the surface generation velocity in semiconductors. A method that uses a new approach to surface generation velocity evaluation is described. The pulsed MOS capacitor is routinely used to measure generation lifetime. The high-frequency capacitance transient observed after the MOS capacitor has been pulsed from accumulation into deep depletion, is usually analyzed to determine the value of bulk generation lifetime in the semiconductor concerned. The method proposed is based upon a precise MOS structure model incorporating lateral effects, the diffusion current from the bulk, and the decrease of the surface generation velocity during the measurement process. The method uses a single measurement of the nonequilibrium C-t characteristic in MOS capacitors subjected to voltage stress and is particularly useful when applied to small gate diameter MOS samples with long relaxation times.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Temperature Dependent Analysis of the Pulsed MOS Capacitor for Semiconductor Material Characterization

An understanding of the physical mechanisms which govern carrier generation in a pulsed MOS capacitor is very important for analyzing the dynamic parameters such as carrier lifetime and diffusion length. In this paper we present a general equation which describes the decay of a MOS capacitor at different temperatures. The equation is a generalization of prior analyses of Heiman, Zerbst, and Schroder. The generalized temperature dependent MOS C‐t analysis was experimentally verified using internally gettered and nongettered samples.

Recombination lifetime of short-base-width devices using the pulsed MOS capacitor technique

The authors point out that the technique of D.K. Schroder, J.D. Whitfield, and C.J. Varker (see ibid., vol.ED-31, no.4, p.462, 1984) for the determination of recombination lifetime using pulsed MOS capacitors at elevated temperatures does not consider lateral quasi-neutral bulk generation and the time dependence of the width of the space-charge region in short-base-width devices (i.e. epitaxial wafers). Consequently, calculations using this technique indicate that the recombination lifetime is a function of device diameter. A simple one-dimensional approach is proposed in which bulk generation in the lateral area of the device is taken into consideration resulting in a fairly uniform recombination lifetime that is independent of the device diameter for short-base-width devices. >

A fast extrapolation technique for measuring minority-carrier generation lifetime

The transient response of a pulsed MOS capacitor is used to calculate the minority-carrier generation lifetime. For Si substrate with a long lifetime, traditional approaches suffer from the inefficiency of long measurement times. An extrapolation technique presented here significantly reduces the measurement time while still maintaining good accuracy. This method applies to capacitors with constant bulk-generation rates. The linear correlation coefficient of the 1n W- t plot is monitored to ensure accuracy of the measurement technique.