Higher order interaction

A fundamental approach by Steele [8] treats monolayer adsorption in terms of interatomic potential functions, and includes pair and higher order interactions. Young and Crowell [11] and Honig [20] give additional details on the general subject a recent treatment is by Rybolt [21]. 

The correlation functions provide an alternate route to the equilibrium properties of classical fluids. In particular, the two-particle correlation fimction of a system with a pairwise additive potential detemrines all of its themiodynamic properties. It also detemrines the compressibility of systems witir even more complex tliree-body and higher-order interactions. The pair correlation fiinctions are easier to approximate than the PFs to which they are related they can also be obtained, in principle, from x-ray or neutron diffraction experiments. This provides a useful perspective of fluid stmcture, and enables Hamiltonian models and approximations for the equilibrium stmcture of fluids and solutions to be tested by direct comparison with the experimentally detennined correlation fiinctions. We discuss the basic relations for the correlation fiinctions in the canonical and grand canonical ensembles before considering applications to model systems. 

Statistical mechanics provides physical significance to the virial coefficients (18). For the expansion in 1/ the term BjV arises because of interactions between pairs of molecules (eq. 11), the term C/ k, because of three-molecule interactions, etc. Because two-body interactions are much more common than higher order interactions, tmncated forms of the virial expansion are typically used. If no interactions existed, the virial coefficients would be 2ero and the virial expansion would reduce to the ideal gas law Z = 1). 

Of these 128 model parameters, the majority of the higher order interactions would most likely be statistically insignificant. To decrease the number of experiments required, FFDs can be used. In these designs, the ability to predict the highest order interactions is sacrificed in order to reduce the number of experiments. This has the effect of reducing... 

Screening designs are mainly used in the intial exploratory phase to identify the most important variables governing the system performance. Once all the important parameters have been identified and it is anticipated that the linear model in Eqn (2) is inadequate to model the experimental data, then second-order polynomials are commonly used to extend the linear model. These models take the form of Eqn (3), where (3j are the coefficients for the squared terms in the model and 3-way and higher-order interactions are excluded. 

The key step is to determine the errors associated with the effect of each variable and each interaction so that the significance can be determined. Thus, standard errors need to be assigned. This can be done by repeating the experiments, but it can also be done by using higher-order interactions (such as 123 interactions in a 24 factorial design). These are assumed negligible in their effect on the mean but can be used to estimate the standard error. Then, calculated... 

For simplicity we consider first the case where only pair interactions contribute to F the extension to include higher-order interactions will be very simple. In the wth semi-invariant occurs... 

Equation (5.21) assumes ternary interactions are small in comparison to those which arise from the binary terms. This may not always be the case and where evidence for higher-order interactions is evident these can be taken into account by a further term of the type Gijit = x< xj Xk Lijk, where Lijk is an excess ternary interaction parameter. There is little evidence for the need for interaction terms of any higher order than this and prediction of the thermodynamic properties of substitutional solution phases in multi-component alloys is usually based on an assessment of binary and ternary terms. Various other polynomial expressions for the excess term have been considered, see for example the reviews by Ansara (1979) and Hillert (1980). All are, however, based on predicting the properties of... 

For this arrangement, higher-order interactions are assumed to be negligible and their sums of squares are pooled to give an estimate of error. From the ANOVA table it appears that in the sub-plot analysis there... 

As was discussed with arrangement (I), it is possible to split the two degrees of freedom for Temperature and Humidity into linear and quadratic contrasts and to construct a normal probability plot for the environmental variable contrasts. This would reveal important effects due to the linear components of both Temperature and Humidity. A normal plot for the design contrasts would indicate that there appears to be a real effect due to A. The analysis of the design x environment interactions is obtained by pooling together higher-order interactions to obtain an... 

An alternative design would be to use a half-fraction of the design variables for each run of the chamber. Such a design, before randomization, is shown in Table 2.22. With this design the ABCDE five-factor interaction is confounded with the TxH whole-plot contrast. Under the assumption of negligible three-factor and higher-order interactions all main effects and two-factor interactions can be estimated as well as interactions between the design and the environmental variables. 

Let us now consider three-factor interactions (e.g. ABC in Table 3.5) to give a general idea how these and higher-order interaction effects (four-, five-factor interaction effects, etc.) are derived. A three-factor interaction means that a two-factor interaction effect is different at the two levels of the third factor. Two estimates for the AB interaction are available from the experiments, one for each level of factor C. The AB interaction effect is estimated once with C at level (+) (represented by Eab,C(+)) nd once... 

In Table 3.16 the columns of contrast coefficients for the two-factor interactions are given. They were obtained using the above stated rules. The contrast coefficients for three- and higher-order interactions can be... 

Using the effects of multiple-factor interactions from full or fractional factorial designs. Multiple-factor interactions (e.g. three- and four-factor interactions) are considered to have a negligible effect. It is then considered that these higher-order interaction effects measure differences arising from experimental error [31]. 

---