Journal of research of the National Bureau of Standards (1977)最新文献

Ruggedness Test (RT) experiments were performed to assess the significance of the various main factors which affect pH measurements in low ionic strength aqueous solutions, as well as to establish the presence of interactions between the main factors. Stirring has an adverse effect on the measurement of pH, since it not only increases the random noise but also biases the measured value. Temperature control to the nearest 0.5 °C is sufficient for maintaining measurements accurate to 0.01 pH. Addition of NaNO₃ or KCl can not be tolerated in accurate pH measurements. Three small two-factor interactions were also revealed.

A straightforward explanation of the statistical technique of ruggedness testing is presented. Efficient Plackett-Burman designs are used in ruggedness tests. These designs involve the simultaneous change of levels of a number of variables. The designs allow the ruggedness test user to determine the effect of the separated variables on the measurement process. This paper (Part I) deals with the common situation where two-factor and higher order interactions can be safely ignored. A method is presented for evaluating the experimental uncertainties. A detailed example of glass electrode measurements of pH of dilute HCl solutions is used to illustrate ruggedness testing procedures.

This paper describes the development of Standard Reference Material, SRM 2694, "Simulated Rainwater," intended to aid in the analysis of acidic rainfall. Details of the formulation and preparation of the two levels of solutions (2694-I and 2694-II) are given. The 10 analytical techniques used to measure the 12 components in the solutions are described in brief. The data used in the statistical evaluation of the results are summarized and the recommended values for pH, specific conductance, acidity, fluoride, chloride, nitrate, sulfate, sodium, potassium, ammonium, calcium, and magnesium are tabulated. The instability of ammonium ion in acidic solutions is discussed. Recommendations for the use of SRM 2694, particularly with regard to the measurement of pH, are given.

Feasibility studies on the application of multivariate statistical and mathematical algorithms to chemical problems have proliferated over the past 15 years. In contrast to this, most commercially available computerized analytical instruments have used in the data systems only those algorithms which acquire, display, or massage raw data. These techniques would fall into the "preprocessing stage" of sophisticated data analysis studies. An exception to this is, of course, are the efforts of instrumental manufacturers in the area of spectral library search. Recent firsthand experiences with several groups designing instruments and analytical procedures for which rudimentary statistical techniques were inadequate have focused efforts on the question of multivariate data systems for instrumentation. That a sophisticated and versatile mathematical data system must also be intelligent (not just a number cruncher) is an overriding consideration in our current development. For example, consider a system set up to perform pattern recognition. Either all users need to understand the interaction of data structures with algorithm type and assumptions or the data system must possess such an understanding. It would seem, in such cases, that the algorithm driver should include an expert systems specifically geared to mimic a chemometrician as well as one to aid interpretation in terms of the chemistry of a result. Three areas of modem analysts will be discussed: 1) developments in the area of preprocessing and pattern recognition systems for pyrolysis gas chromatography and pyrolysis mass spectrometry; 2) methods projected for the cross interpretation of several analysis techniques such as several spectroscopies on single samples; and 3) the advantages of having well defined chemical problems for expert systems/pattern recognition automation.

Although real polymers involve the sequential addition of monomers having fixed bond lengths, fixed bond angles and some freedom of rotation about single bond, the properties of polymers aver large length scales can be modeled by treating the polymer configuration as that of a random walk formed by the monomer units. Serious complications arise in the theoretical description of these polymers because of excluded volume constraints which prohibit different monomers from occupying the same position in space. This polymer excluded volume problem has been modeled in terms of a simple continuous random walk with short range repulsive interactions. The expansion of polymer properties in this repulsive interaction can readily be shown by dimensional analysis to involve an expansion in a large parameter, in the limit of long polymers. The renormalization group method is utilized as a systematic means for resuming this divergent perturbation expansion. The theory proceeds by analytically continuing the treatment to continuous range of spatial dimensionalities to expose and regularize the singularities in the analytically continued theory. The renormalization group approach is described from a heuristic physical standpoint and extensive comparisons are provided to show how it quantitatively reproduces vast amounts of dilute solution polymer properties with no adjustable parameters.

Kinetic models described by systems of linear differential equations can be fitted to data quickly and easily by taking advantage of the special properties of such systems. The estimation situation can be greatly improved when multiresponse data are available, since one can then automatically determine starting values and better discriminate between rival models.