Selection of test items

In a situation where we wish to select a series of test compounds, e.g. for determining the scope and limitations of a reaction, we are facing a problem which involves a discrete variation between the test objects. To cover the possible variation of interest in order to determine as broad a scope as possible, we have to consider all relevant properties of the reaction system. Due to possible interaction effects, all these properties must be taken into account simultaneously. This is not a trivial 

In the following sections, the methods of principal components analysis (PCA) and factor anafysis (FA) are described. Principal components analysis is a special case of factor analysis. It is discussed how these methods can be applied to the above problem. Fortunately, these methods are not difficult to understand. A great advantage is that the results obtained by these methods cases can usually be evaluated graphically by visual interpretation of various plots. The systematic variation of the properties for a set of compounds is graphically displayed and it is therefore easy to see how a selection which spreads the properties should be made. 

In a situation where test compounds are selected, e.g. for studies on the scope of a reaction, or when the objective is to determine a suitable reagent for a given transformation, it is likely that the possible test candidates are in some way similar to each other. This will have several consquences  

The problem of selecting suitable test objects which span a sufficient range of variation in the intrinsic properties can therefore be stated as foUows From the set of all potential test candidates, a subset should be selected in such a way that a sufficient spread in aU pertinent properties is obtained in the test set These properties are specified by the descriptors. 

---

Selection of test items so that they are as dissimilar a possible to each other is accomplished by choosing items which are projected far from each other and on the periphery of the plot. Such designs are useful for answering the question of whether or not the properties have an influence. A maximum spread design wa used to determine whether modification of the solvent would increase the endo/exo stereoselectivity in the reduction of an enamine from camphor. The answer was negative [17 a],... 

When the objective is either to make a carful screening for finding a good candidate for future development, or to study whether a gradual change in the performance of the reaction could be traced to the properties of the reaction system, a design which affords a selection of test items which are uniformly spread in the score plot should be employed. An example of this principle is given in the study of the Fisher indole reaction below. 

The selection of test items by such a design can be accomplished as follows For each constituent of the reaction system, two principal property axes should be considered. The columns of a two-level fractional factorial design matrix contain an equal number of minus and plus signs. If we let the columns pairwise define the selection of test systems, four combinations of signs are possible [(—),(—)], [(-), (+)]. [(+), (—)]. and [( + ), (+)]. These combinations of signs correspond to different quadrants in the score plots. Hence we can use the sign combinations of two columns to define from which quadrant in the score plot a test item should... 

Fig.17.6 Selection of test items in the Fischer indole synthesis (a)...

The number of combinations of possible carbonyl substrates, substituted phenylhydrazines, add catalysts, and solvents is overwhelmingly large. The present study was limited to include dissymmetric ketone substrates with a and a methylene groups, phenylhydrazine, Lewis acids, and common solvents. The selection of test systems was based on the principal properties of the ketone, the Lewis acids, and the solvents. The selection was made to achieve approximately uniform distributions of the selected items in the score plots, see Fig.17.6. 

The above example shows a selection of test systems by a design affording a uniform spread in the principal properties. The objective was to establish whether there was a gradual change in the performance of the reaction which could be related to the properties of the reaction system. The aim was to detennine those properties which have an influence on the selectivity so that these properties could be controlled. Both these objectives were attained. The results would have been very difficult to achieve without the PLS method and without using a multivariate design for selecting test items. 

ISO 9000. The ISO 9000 series of standards was first issued in 1987 and then updated in 1994. The ISO 9000 standard describes the selection criteria for the four standards, ISO 9001—9004. ISO 9001 is composed of 20 items covering manufactured products from design and development in R D to commercial production and after-sale service or technical support. ISO 9002 has 19 of the items in 9001 describing the requirements of a quahty system for the manufacture of an item, such as chemicals, to a specification only the requirement relating to product development is omitted. Chemical companies seldom certify to ISO 9003, which describes the quahty system for a laboratory involved in final inspection and test. Finally, ISO 9004 presents guidelines for total quahty management. 

Engineering specification and purchasing procedures are essential to ensure that all items are to the design specification and to comply with company or national standards. During installation the features to consider are foundations, selection of materials, fabrication, assembly, supports, pressure testing etc. 

Raw materials and auxiliary products used in a process as well as materials of construction for equipment items can be the eause of scale-up effects . Pure raw and auxiliary materials must be used in laboratory studies to eliminate the influence of impurities on the ehoice of the process route, catalyst selection, and search for satisfactory process conditions. However, pure chemicals are usually too expensive to use for manufacture on a commercial scale. It is common practice to use raw materials of technical grade in a full-scale plant. These materials contain impurities, which can act as catalysts or inhibitors. They can react with reactants or intermediates, thereby decreasing yields and selectivities of desired produets. Therefore, raw materials of technical grade, even from different suppliers must first be tested on laboratory scale. 

For the selection of the test item all characteristics that could affect the integrity ofthe interlaboratory comparison should be considered... 

The selection of foods to eliminate should be based on a variety of items, including history of illness, age of patient, results of diagnostic tests, epidemiological considerations, adherence to the diet, and elimination of additional triggers which may cause symptoms (Sampson, 1999a Sicherer and Sampson, 1999). 

It is sometimes fairly well known which type of reagent or solvent should be used. To further improve the result, it might be profitable to investigate other candidates with similar properties. For this purpose, test items which are projected close to the good candidate should be chosen. We have found this principle to be useful for the selection of solvents for fractional crystallization. 

Table 12. Assignments of design settings to selected test items...

If we let columns in the design matrix define the constituents as follows 1, 2 define the substrate, columns 3, 4 define the amine co-substrate, and columns 5, 6 define the solvent, the first row in the design matrix in Table 11 would thus correspond to a selection of a substrate projected in the [( — ),( — )] quadrant, an amine from the [(—),( + )] quadrant, and a solvent from the [( + ),( + )] quadrant. The other rows define other combinations. The test items selected accordingly are shown in Table 13. To permit fair comparisons as to the performance of the reaction, it is necessary to adjust the experimental conditions for each system to yield an optimum result. The danger of using standardized conditions has been emphasized [1] and the arguments against such a technique are not repeated here. The conditions which afforded a maximum yield were determined by response surface techniques and these results are also shown in Table 13. 

Measurements of airblast overpressure and impulse were made at 12 gage locations along a double blast line (Fig k). The gages were spaced at selected scaled distances ranging from approximately 2-20 ft/lbsl/3. The pressure transducers were installed flushed with the top surface of a concrete slab in mechanically isolated steel plates. The test item was placed on a steel witness plate located on the surface of the slab. Fastax motion pictures were taken of all tests. 

---