Instrument bias

A PHENOMENOLOGICAL DESCRIPTION OF THE MASS-DEPENDENT INSTRUMENTAL BIAS... 

Figure 9. Sketch of the double spike Zn- Zn method. The surface is constructed by drawing an infinite number of straight-lines through the point representing the spike composition (supposed to be known with no error) and each point of the mass fractionation line going through the point representing the measured mixture. One of these straightlines, which is to be determined from the calculations, is the sample-spike mixing line (stippled line). Each determination of the Zn isotope composition of a sample involves only one run for the mixture of the sample with the spike. Since all natural samples plot on the same mass fractionation line, any reference composition will adequately determine isotope composition of the sample, note that, since the instrumental bias is not linear with mass, the mass discrimination lines are curved.

CORRECTION OF THE NON-MASS DEPENDENT INSTRUMENTAL BIAS IN STATIC MODE... 

A double spike technique is essential for TIMS analyses of Se and Cr, and may also be useful in MC-ICP-MS analysis. Briefly, two spike isotopes with a known ratio are added to each sample, and the measured ratio of the spike isotopes is used to determine and correct for instrumental bias. Examples of Se and Cr double spikes currently in use are given in Table 1. The fact that small amounts of the spike isotopes are present in the samples and small amormts of nominally unspiked isotopes are found in the spikes is not a problem, as the measurements allow highly precise mathematical separation of spike from samples. Algorithms for such calculations are described by Albarede and Beard (2004) and, specifically for Se, by Johnson etal. (1999). 

NIST SRM-979 is a high purity Cr(N03)3 9H20 crystalline sohd sold as an isotopic reference material. It was prepared from a commercial material and its isotopic composition was analyzed by TIMS, along with gravimetric mixtures of purified isotopes to examine instrumental bias (Sheilds et al. 1966). The absolute ratio is reported as 0.11339, with... 

Reproducibility encompasses the variation in analytical results between laboratories and provides a second level of method ruggedness. This is becoming an increasingly important part of method validation as the pharmaceutical industry becomes more specialized and diversified. Major manufacturers may develop and validate a method in a corporate research center for use in a foreign manufacturing site or at a contract testing laboratory. It is therefore critical that the validation demonstrates that the method is free of analyst or instrument bias. 

The natural isotope ratio of 88se/ 2se is 5.42. Correcting for the contributions of the ligand isotopes, the ratio should be 5.36. This ratio is actually observed, indicating no instrumental bias in the measurements. Frew et al. (8) have discussed these corrections in detail. So, Equation 2 must be corrected for this small effect of the ratio differences of a factor of 1.01 (i.e., 5.42/5.36), namely ... 

Instrument bias errors are often seen at high frequencies, especially for systems exhibiting a small impedance. 

Check the results The presence of instrument bias errors can be difficult to discern. The Kramers-Kronig relations may provide a suitable guide, but as discussed in Chapter 22, some instrument-imposed bias errors are Kramers-Kronig transformable. If possible, high-frequency asymptotic values should... 

If a single model were to be regressed to all spectra showing nonstationary behavior, the resulting residual error would include contributions from the drift between scans as well as the lack of fit of the model, instrumental bias errors, and stochastic errors, i.e.,... 

Accuracy Characterizes the difference between the result of a measurement and the true value of the measured data. It is difficult to estimate properly because the true value is most generally unknown. Computer experiments permit an easy estimate of the errors due to the use of incorrect models, but are unable to accoimt for instrumental bias. 

QCPs are intended to make available to analysts a standardized approach for minimizing analytical errors and to provide the assurance of the generation of good data with the best possible precision and accuracy. This is achieved by integrating into normal laboratory practices steps which would ensure freedom from somces of error such as contamination, matrix effects, human and instrumental bias and random errors, fluctuating instrumental sensitivity, and discrepancies in analytical standards. 

Because three of the four isotopes of the Pb isotope system are radiogenic, there is no stable reference isotopic ratio. Thus, there is no stable isotope pair for internal checking of the instrumental bias, which results in a precision of about one order of magnitude less than that obtained for other isotope systems used in geochemistry (Sr, Nd, etc.). As a consequence, it is very difficult to refer to absolute reference values even for a standard material after careful double-spike (DS) TIMS measurements. 

The MC-ICP-MS measurements were performed at a power of about 1200 W, which correspond to so-called hot plasma conditions. The stable introduction system (SIS) used consists of a quartz dual-spray chamber with a low-flow perfluoroalkoxy (PFA) microconcentric nebulizer, which produces a flow rate of approximately 50 pL/min. A conventional H cone system was used. Typically, a ° Pb signal of 4 V per 100 ng/mL is obtained with Faraday cup detectors (10 fl). The instrument is located in a clean laboratory (ISO 7, NF ISO 14644-1). The T1 spike technique, which allows in-run instrumental bias correction, was used. The required set of Faraday cups simultaneously collects the following masses Hg or Hg, Tl, Pb -h Hg, =T1, ° Pb, Pb, and Pb (Thble 31.1). A gain calibration of all the Faraday cups was carried out to determine the correction coefficients between all the Faraday cups, after the most efficient instrumental settings were determined by using a pure Pb standard solution (SRM-981) at 50 n mL. Then, the peak shape at all masses of interest was evaluated. For a given analysis, 5 blocks of 10 cycles with 8-s integration time per peak were recorded in static mode for common samples over 7 min [9]. 

No sample standard normalization was applied since T1 normalization appears to be efficient to correct for instrumental bias. A blank solution of 3% v/v HNO3 added to a-Q water (18 Mil) similar to that used to adjust the sample solutions was prepared. The blank solution was systematically measured between each sample after careful washing out of the instrument. No electronic baseline or half-mass baseline was measured. The blank subtraction involved measurement of a real blank solution. In fact, the measurement included both the electronic blank from the instrument and the chemical blank from the water used to dilute samples. 

The selection of a zircon standard is required for instrumental bias of the MC-ICP-MS and to adjust the parameters of the laser. A careful tuning of the MC-ICP-MS... 

Begley, I.S., Sharp, B.L. (1997) Characterisation and correction of instrumental bias in indnctively coupled plasma quadrupole mass spectrometry for accurate measurement of lead isotope ratios. Journal of Analytical Atomic Spectrometry, 12,395-402. 

Precision is a measure of the ability of a method to generate the same result for multiple analyses of the same sample. Conversely, from the above case with accuracy, a precise method need not be an accurate one. Instrument bias, loss of analyte in the sample preparation, or analyte instability in solvents, may all lead to inaccuracies. However, if the error is constant and reproducible, a very precise result is obtained. 

By the use of the ratio and because the labelled analogue and analyte of interest are detected at very close to the same time, IDMS also allows for correction of instrumental bias that can affect other kinds of quantitative approaches used in mass spectrometry. 

---