Important as they are, these uses are not the subject of this chapter. Compact expressions for hand analysis or for automatic synthesis, valid in all operating regions, are presented throughout the book. The designers rely on the simulation of their design before building a prototype. Far from the metallurgical junction, neutrality prevails as shown in Fig. The judicious use of electrical simulators allows the quick evaluation of the circuit performance without the burden of a costly integrated prototype. Since lattice vibrations phonons depend on the temperature, so does the mobility.
A second metal layer along the back provides an electrical contact with the silicon substrate. However, there is a lot of confusion too. To simulate a circuit, simulators make use of element models, which provide a mathematical description of the element behavior in the circuit. A close comparison between 8. In accumulation, the minority carriers are irrelevant, thus the equivalent circuit is the same as that for the two-terminal capacitors. In order to write the concentrations under non-equilibrium conditions, preserving the form of the Boltzmann equilibrium expressions, one simply replaces 33 Chapter 2.
Given that the applied reverse forward voltage appears as an increase decrease in the potential barrier inside the semiconductor, the quasi-Fermi levels must split by the amount of the applied reverse forward bias, to keep the majority carrier concentrations in the neutral regions on both sides of the junction constant as shown in Fig. Clearly, the drain-bulk and source-bulk diodes contribute with their reverse currents, but here we will restrict our analysis to the intrinsic transistor. Due to the exponential dependence of αn on electric fields, impactionization dominates at the region where the electric field is at its maximum. Phonon scattering is caused by the interaction of carriers with lattice vibrations. A unified approach for both space matching and time noise fluctuations is introduced.
Another simplification considers the effective field to be constant along the channel and equal to its value at pinch-off. Treatment of multidimensional effects like barrier lowering is rudimentary, and gives an optimistic appraisal of simple models. A unified approach for both space matching and time noise fluctuations is introduced. The double integration of 8. The purpose of this development was mainly to test new circuits made available by the nascent field of integrated circuits. For a high quality SiSiO2 interface, the interface trap density is so low that its effect can be disregarded and the idealized model presented in this section is appropriate. The analysis for p-channel devices with a p+ poly-gate is similar.
For the n-channel device, pure electron current initiates the multiplication, which implies that the hole current at the drain end of the channel is zero. We will also consider edge effects and the electrostatic coupling between drain and source. Thesis, Universidad de la Republica, Uruguay, 2004. We now apply the concept in expression 9. In strong inversion Cgb is negligible because the channel charge screens the bulk from the variation in the gate voltage. The Ward-Dutton method, presented in Section 5. The flat-band voltage is composed of two terms, the work function difference between the poly-gate and the silicon substrate plus the equivalent oxide charge.
The most severe modification of the intrinsic characteristic occurs in the linear region where the channel resistance is minimal. In other words, the contact between the n+ region and the channel implies the equilibrium of electrons in these two regions, the electron density varying according to the value of the electrostatic potential. The contents, views, and opinions of the authors expressed herein do not necessarily state or reflect those of the Publisher, its Series Editor, and their employees. The most common metal plate materials are aluminum and heavily doped polycrystalline silicon polysilicon or poly. Deep in the Chapter 2. Presents practical guidance on device development and circuit implementation.
Typical thermal coefficients of the threshold voltage for nchannel devices lie in the range -0. As a consequence, electron-hole pairs are generated, the electrons being collected by the drain and the holes by the substrate, as shown in Fig. Since the hole current is carried by a high hole concentration in the p-region and by a low hole concentration in the n-region, the gradient of φ p must be much greater in the n- than in the p-region, as shown in Fig. We also assumed the mobility to be independent of the electric field. A course emphasizing general principles should be based on Chaps. It includes the effects of both the transverse and parallel electric fields.
Experimental and simulation results for the substrate impact ionization current are given in Fig. We conclude the chapter with some benchmark tests concerning compact transistor modeling. Enz, François Krummenacher and Eric A. A considerable effort in academia and companies has been made in the last few years to provide the design community with better compact models that can lead to more accurate simulation results. Even though the modeled threshold voltage fits the experimental results very well, the same cannot be said of the current dependence on the voltages, due to mobility differences in the center and edge regions as well as relative differences in the weights of the regions in weak and strong inversion. The constancy of the electron quasi-Fermi potential from the n+ to the nchannel region represents the fact that the inverted n-channel is connected to the n+ source region. Also, the partial derivatives of the virtual charges each have an additional term related to the saturation velocity as shown in expressions 5.