Simultaneous over relaxation

CGC = coarse grid correction CSD = critical slowing down DH = Debye-Huckel FAS = full approximation scheme FD = finite difference LFT = lattice field theory Ihs = left-hand side MG = multigrid PB = Poi.s.son-Boltzmann PBC = periodic boundary conditioas rhs = right-hand side SOR = successive (or simultaneous) over-relaxation. 

Simonsen s group has performed some elegant work over the years on NO release characteristics from rat superior mesenteric artery. Initially, Simonsen s group simultaneously monitored artery relaxation and NO concentration in the artery using a NO microsensor in response to various drugs [120], NO concentration was monitored via an ISONOP30 electrode, purchased from WPI and inserted into the artery lumen using... 

Simultaneously td is also the characteristic time of migrational relaxation of a spatially uniform space charge. Indeed, assume for simplicity that diffusivities of all ionic species are equal to >o. Assume further that concentrations Ci(x,t) are spatially constant, dependent on time only. Then from (1.1), (1.3), after multiplication by Z F, summation over all i, and use of (1.4), we get... 

Such a short spin-equilibrium relaxation time raises the question of whether discrete spin state isomers exist. Their existence is affirmed by two observations. One is the persistence of electronic spectral bands typical of the low-spin 2E state over a wide temperature range in solid samples (98). The other is the observation of EPR signals characteristic of the 2E state in both solids and solutions between 4 and 293 K (98,139). At very low temperatures EPR signals of both spin states can be observed simultaneously (98). At low temperatures hyperfine splitting into eight lines is observed from coupling with the 1 = 7/2 Co nucleus. As the temperature is raised the spectral features broaden and the hyperfine resolution is lost. This implies a relaxation process on the EPR time scale of 1010 sec-1, or a relaxation time of the order 0.1 nsec, consistent with the upper limit set by the ultrasonic experiments. 

In order to visualize the effects of water exchange, rotation and electronic relaxation as well as of magnetic field on proton relaxivity, we have calculated proton relaxivities as a function of these parameters (Fig. 2). The relaxivity maximum is attained when the correlation time, tc1, equals the inverse proton Lar-mor frequency (l/rcl = l/rR + l/rm + l/Tle = a>j). The most important message of Fig. 2 is that the rotational correlation time, proton exchange and electronic relaxation rates have to be optimized simultaneously in order to attain maximum relaxivities. If one or two of them have already an optimal value, the remaining parameter starts to become more limitative. The marketed contrast agents have relaxivities around 4-5 mM1 s 1 contrary to the theoretically attainable values over 100 mM 1 s1, which is mainly due to their fast rotation and slow water exchange. 

This means that Mod pq in Eq. (4.11) now changes over time according to (4.14). It is of considerable importance here that the precession (4.11) and relaxation (4.14) are independent, in the sense that the decay does not have any effect on the change of phase ip. Simultaneous actions of turn and relaxation are shown by the broken fine in Fig. 4.1(c). This corresponds to rotation of the figure p(0,[Pg.109]

The excess electron, relevant for the estimation of the electron affinity, and the hole — relevant for the ionization potential — are never simultaneously present in the system and do not interact. If one neglects electron relaxation effects, G lumo homO but this energy difference does not give an accurate estimate of G there is a systematic under-estimation in the DFT-LSDA, and a systematic over-estimation in UHF. However, part of the variation of G with the conformation, size and dimensionality of the system results from one-electron processes, and is reasonably described by the variations of twMO — homo- As discussed above, the ionic contribution ec — is smaller at the surface than in the bulk. This acts towards a narrowing of the gap, and the effect is stronger on more open surfaces. 

For generating models of about 512 atoms or more, the geometrical region over which local relaxation is made will be small compared with the model size. It would then be possible to simultaneously generate bond switching in several well-separated regions and thus speed up the entire process. This lends itself to parallel computation. We have not done this, but O Mard has implemented such a parallel modeling system [5]. 

---