Fitness modeling-directed evolution

HER as a function of overpotential clearly fit models [5, 150] in which the coverage of the intermediate is low at potentials above H2 evolution, and increases only slowly as the potential is moved into the hydrogen evolution region. This led to a search for the intermediate, the identity of which was first directly addressed using in situ FTIR again by Bewick [151] and then more recently, by Kitamura, Tokuda and coworkers [152]. Both groups started from a model in which the HER intermediate was beheved to be hydrogen atoms adsorbed on an on-top site on Pt, and employed the SNIFTIRS approach, but different data collection times, and obtained somewhat different results. 

Spin-lattice relaxation time increased continuously as a fimction of time, passing from 4000 ms to 8500 ms. It was possible to see two different evolutions. Ti increased rapidly during the two first days and slowed down after this time. Such a curve can be fitted with a power law model. The evolution was the same as for crystal size during Ostwald Ripening. The power law exponent determined for this system was 0.098, which is lower than the 0.2 to 0.3 which can be found for the evolution of crystal size followed by polarized microscopy. The actual deviation was due to the fact that we measured spin-lattiee relaxation times and not crystal size directly. However, the fact that we retained the power law model led us to expect a relationship between crystal size and spin-lattice relaxation time. 

In concomitance with the displacement observed by i.r., an evolution of the catalytic activity has been observed while studying the liquid-phase epoxidation of cyclohexene in the presence of (EGDA)- Mo(VI), freshly prepared or after four months of conditioning at room temperature under inert atmosphere. As usual, the appearance of epoxide was followed by gas chromatographic analyses or by direct titration of oxirane oxygen and the disappearance of hydroperoxide was monitored by iodometric titration. In figure we report concentration-time for typical runs in ethylbenzene at 80°C obtained with the experimental procedure already described (ref. 9). It may be seen that with a freshly prepared catalyst an induction period is observed which lowers the initial catalytic activity. Our modified Michaelis-Menten type model equation (ref. 9) cannot adequately fit the kinetic curves obtained due to the absence of kinetic parameters which account for the apparent initial induction period (see Figure). 

Lamarckian is modeled on Lamarck s theory of inheritance of acquired characteristics, in much the same way that the basic GA is modeled on Darwinian evolution. The term usually means that the evaluation function takes the initial parameters decoded from the chromosome and carries out a local function minimization. The parameter values at the local minimum are recoded into the chromosome and passed back to the main GA routine along with the value of the fitness function at that point. The environment has caused a direct and (locally) beneficial change in the individual s genes that can be passed on to its offspring. 

In particular, it must be noticed that classic dynamical calculations, directly applied to polymers, lead very often to unreliable assignments for vibrational IR or Raman modes and that the use of oligomers or model compounds is, in most cases, considerably useful. Apart from the fact that many more experimental frequencies are compared to calculated ones, the method has the advantage of following the evolution of vibrational modes from one con und to the next one in a series. As a consequence, the fitting procedure is much easier and assigments are made in a more reliable way. 

---