Practice5.PDF

Practice 5 Many physical chemical problems can be solved by repeated experiments or calculations that involve the element of randomness. For example, Buffon has shown that if a needle of length “a” is thrown at random onto a set of equally spaced parallel lines of width d, the probability that the needle will cross a line is 2a/πd. As a result, a random drop of needles on the floor can be used to estimate the value of π. Monte Carlo method is one of the methods that use the random numbers to solve the physical chemical problems. It has been applied extensively to the study of the reaction dynamics. By investigating the collisions between two reactants with random orientations and momenta (specified by the experimental condition), the potential energy surface describing the molecular interactions may be deduced, which can be used to determine the dynamics for similar reactions. Another important application of Monte Carlo methods is to evaluate multidimensional integrals that cannot be handled by the simple quadrature methods. The known function is a hypersurface in n-dimensional space. Over the range to be integrated it is enclosed in a hypervolume of n+1 dimensions, the value of which is easily calculated. Points in this space are chosen randomly by generating values for each of the independent variables and one for the dependent variable (the function of all the independent variables). The function is then evaluated to see whether it is greater than or less than the randomly chosen value (i.e., does it lie above or below the curve in hyperspace?). The value of the integral is given by the fraction of all random points, which lie below the curve, multiplied by the volume of the enclosing hyperspace. A similar proble