Reference data, calibration

Another aspect of traceability of the results is the linkage of data from the homogeneity study, the stability study, and the characterization study of the reference material. In order to establish this link, the coordinator must be in the position to demonstrate that the results of these three studies have a common reference. Such a reference can be a calibrant, reference material, or possibly some realization by means of a suitable method. If such a common reference is not available, it is impossible to link the data sets, and therefore it is impossible to translate the results from the homogeneity and stability studies to the characterization of the material. 

For the extraction of rubber and rubber compounds a wide variety of solvents (ethyl acetate, acetone, toluene, chloroform, carbon tetrachloride, hexane) have been used [149]. Soxtec extraction has also been used for HDPE/(Tinuvin 770, Chimassorb 944) [114] and has been compared to ultrasonic extraction, room temperature diffusion, dissolution/precipitation and reflux extraction. The relatively poor performance of the Soxtec extraction (50% after 4h in DCM) as compared with the reflux extraction (95% after 2-4 h in toluene at 60 °C) was described to the large difference in temperature between the boiling solvents. Soxtec was also used to extract oil finish from synthetic polymer yam (calibration set range of 0.18-0.33 %, standard error 0.015 %) as reference data for NIRS method development [150]. 

This chapter deals with the necessity of representative sampling in the context of PAT. All PAT sensors need to be calibrated with respect to relevant, reliable reference data (Y data). This presupposes that representative samples are at hand for this characterization - but sampled howl Additionally, X signals (X measurements) need to be qualified as representative of the same volume as was extracted for Y characterization, or at least a sufficiently well-matching volume. How does one demonstrate this in a quantitative manner If the quality of both X and Y data involved is suspect, how can a multivariate calibration be expected to be trustworthy This also includes the issue regarding proper validation of the chemometric multivariate calibration(s) involved, which can only be resolved based on proper understanding of the phenomenon of heterogeneity. The TOS delivers answers to all these issues. The TOS constitutes the missing link in PAT. 

For inverse calibration methods, the fact that reference data (y) is never noise-free in practice allows irrelevant variation in the x variables to find its way into the calibration model. 

Figure 2 serves to illustrate the remarkable reproducibility and precision of the spectra that are obtained. The upper trace shows the 950-1150 cm-1 region of the absorbance spectrum of 35 x 10 3 Torr (buffered to 700 Torr total pressure with ultra pure air and run at forty meters path length) of supposedly 90% pure methanol, as received from a supplier. The middle trace represents 12 x 10 3 Torr of methanol, as determined more than a year previously in the short auxiliary cell. The lower trace shows the difference. Note that the and ISq spectra are completely distinctive, that the 1 0 enrichment is in fact only 65%, and that in the region of spectral overlap perfect separation of the two spectra can be achieved in spite of their great complexity. Further, the lower trace, obtained using archival reference data, represents a calibrated reference spectrum for methanol, even though no purified sample has ever existed. 

The laboratory shall have instructions on the use and operation of all relevant equipment, and on the handling and preparation of items for testing and/or calibration, or both, where the absence of such instractions could jeopardize the results of tests and/or calibrations. All instractions, standards, manuals and reference data relevant to the work of the laboratory shall be kept up to date and shall be made readily available to persoimel (see also 4.3 in ISO/IEC 17025). 

The use of more calibration data can result in improved model performance, especially if the additional calibration data result in an improved representation of the sample states that need to be covered for calibration. Referring to the earlier plot of interference error and estimation error versus model complexity (Figure 8.18), an improved representation of the calibration space corresponds to a general drop in the level of the estimation error curve. This results in lower overall prediction error. 

Analysts usually have two principal objectives, the first of these being an exploratory method which enables spectra to be classified into pre-defined families of compounds. Such a tool enables the sample to be identified as a must, a must in fermentation, a dry wine, a liqueur wine or a naturally sweet wine. The second objective involves a quantitative approach which enables the attribution of analytical values or indices particular to the wine or must on the basis of the previously acquired reference data (calibration). It is above all this second approach which is used by analytical laboratories where is possible to replace classical analytical techniques by FTIR. 

Any prerequisites such as test equipment, calibration, test data, reference SOPs, user manuals, training, and sequences between different test scripts should be defined in the preapproved Test Specifications. 

For an estimate of an input quantity that has not been obtained from repeated observations, the associated estimated variance or standard uncertainty is evaluated by scientific judgement based on all available information on the possible of its variability. This is the case of Type B standard uncertainty. The pool of information may include previous data, experience with general knowledge of behaviour of relevant materials and instruments, manufacture s specifications, or data provided in calibration and other certificates or uncertainties assigned to references data taken from handbooks. 

Standard dictionaries define semiempirical as involving assumptions, approximations, or generalizations designed to simplify calculation or to yield a result in accord with observation [1]. In this spirit, the semiempirical methods of quantum chemistry start out from the ab initio formalism and then introduce rather drastic assumptions to speed up the calculations, typically by neglecting many of the less important terms in the ab initio equations. In order to compensate for the errors caused by these approximations, empirical parameters are incorporated into the formalism and calibrated against reliable experimental or theoretical reference data. If the chosen semiempirical model retains the essential physics to describe the properties of interest, the parametrization may... 

The parametrization of a given implementation serves to determine optimum parameter values by calibrating against suitable reference data. The most widely used methods (see Section 21.2) adhere to the semiempirical philosophy and attempt to reproduce experiment. However, if reliable experimental reference data are not available, accurate theoretical data (e.g. from high-level ab initio calculations) are now generally considered acceptable as substitutes for experimental data. The quality of semiempirical results is strongly influenced by the effort put into the parametrization. 

One of the more difficult experimental aspects of Mossbauer spectroscopy is the accurate determination of the absolute velocity of the drive. The calibration is comparatively easy for constant-velocity instruments, but most spectrometers now use constant-acceleration drives. The least expensive method, and therefore that commonly used, is to utilise the spectrum of a compound which has been calibrated as a reference. Unfortunately, suitable international standards and criteria for calibration have yet to be decided. As a result, major discrepancies sometimes appear in the results from different laboratories. The problem is accentuated by having figures quoted with respect to several different standards, necessitating conversion of data before comparison can be made. However, calibration of data from an arbitrary standard spectrum will at least give self-consistency within each laboratory. 

---