We calculated the steady-state recombination rate for two coupled defect levels and implemented the model into a device simulator. This model generalizes the familiar single-level Shockley–Read–Hall (SRH) formula. If the intercenter transition probability vanishes, it reduces to the sum of two individual SRH rates, which are only linked via the band occupancies. The cases, where one of the levels or even both behave like traps in carrier capture, and the case of a rate-limiting intercenter transition are derived from the general expression. The important feature of the model is a possible increased field effect which might lead to large excess currents. The field effect is discussed in terms of tunnel-assisted multiphonon capture or direct tunneling into the levels, respectively. We show by means of numerical simulation that the large ideality factors found for liquid phase epitaxy grown diodes with weak intrinsic fields can be the result of a rapid direct charge transfer between donors and acceptors and the high probability of tunneling into the hydrogenic states.

The ‘‘passivated emitter and rear locally diffused’’ (PERL) silicon solar cell structure presently demonstrates the highest terrestrial performance of any silicon-based solar cell. This paper presents a detailed investigation of the limiting loss mechanisms in PERL cells exhibiting independently confirmed 1-sun efficiencies of up to 23.0%. Optical, resistive, and recombinative losses are all analyzed under the full range of solar cell operating conditions with the aid of two-dimensional (2D) device simulations. The analysis is based on measurements of the reflectance, quantum efficiency, dark and illuminated current–voltage (I–V) characteristics, and properties of the Si–SiO2 interfaces employed on these cells for surface passivation. Through the use of the 2D simulations, particular attention has been paid to the magnitudes of the spatially resolved recombination losses in these cells. It is shown that approximately 50% of the recombination losses at the 1-sun maximum power point occur in the base of the cells, followed by recombination losses at the rear and front oxidized surfaces (25% and <25%, respectively). The relatively low fill factors of PERL cells are principally a result of resistive losses; however, the recombination behavior in the base and at the rear surface also contributes. This work predicts that the efficiency of 23% PERL cells could be increased by about 0.7% absolute if ohmic losses were eliminated, a further 1.1% absolute if there were no reflection losses at the nonmetallized front surface regions, about 2.0% by introducing ideal light trapping and eliminating shading losses due to the front metallization, and by about 3.7% absolute if the device had no defect-related recombination losses. New design rules for future efficiency improvements, evident from this analysis, are also presented.

This paper reports on the state of the art silicon micromachined microphone utilizing a dual poly silicon membrane system. MEMS chips from 1.4mm down to 1.0mm side length are applied for mobile communication. Design aspects related with key performance parameters such as sensitivity, signal to noise ration and distortion are discussed. Sensitivity of -38BV/Pa is achieved for different microphone membrane diameters. A maximum signal to noise ration of 66dB(A) for the largest system could be achieved. The perfect fit of simulation versus measurements enables deeper analysis and balancing of noise contributors. Environmental noise suppression of 5dB by acoustical high pass design is demonstrated.

No abstract

We present a generalized Ramo-Shockley theorem (GRST) for the calculation of time-dependent terminal currents in multidimensional charge transport calculations and simulations. While analytically equivalent to existing boundary integration methods, this new domain integration technique is less sensitive to numerical error introduced by calculations of finite precision. Most significantly, we derive entirely new optimized formulas for the ensemble Monte Carlo estimation of steady-state terminal currents from the time-independent form of our GRST, which are in general not equivalent to the time-average of the true time-dependent terminal currents. We then demonstrate, both analytically and by means of example, how our new varianceminimizing terminal current estimators may be exploited to improve estimator accuracy in comparison to existing methods.