Analytical procedure recovery deviations

It is absolutely essential that the accuracy be validated with the same quantitation method that is used in the control test procedure. Recovery deviations from the theoretical values while performing a calibration with a drug substance alone may indicate interferences between the analyte and placebo components. In such a case, the calibration should be done with a synthetic mixture of placebo and drug substance standard. Such interferences may also be detected by the separate determination of linearity for dilutions of the drug substance and for a spiked placebo. 

In another study, PLE with methanol was compared with Soxhlet with methanol [49], Recoveries and relative standard deviations were comparable for both the methods, confirming that Soxhlet can easily be substituted by PLE in already developed methods, without other substantial changes in the analytical procedure. 

Determination of LAQL. Tests were performed with the recommended analytical procedures and sorbent materials to establish the LAQL for each analyte. As stated previously, the LAQL is the smallest amount of a compound that can be determined with a recovery from the sorbent greater than 80% and a relative standard deviation less than 10% (9). 

The limit of quantification is numerically equivalent to six times the standard deviation of the measured unit when applying the analytical procedure to 20 representative blank samples. For recoveries less than 100% the limit of quantification must be corrected for the average recovery of the analyte. 

Procedure Approx 60 mg of the mixt is dissolved in 40 ml of a methyl isobutyl ketone-isopropanol mixt (4 1) and titrated with sodium methoxide added from a microburet. After each ad dition of the titrant, the soln is thoroughly mixed by a magnetic stirrer for 1 min and the potentials recorded after allowing the soln to settle for another min. The titration is cote ducted in an atm of nitrogen. An av erage recovery of 99 75% with a standard deviation. of 0.38 is to to be expected. See also Analytical Procedures under HMX in this Volume... 

Cahill et al. [241] have developed a simple and sensitive analytical procedure for determining the concentration of trifluoroacetic acid in plant, soil, and water samples. The analysis involves extraction of trifluoroacetic acid by sulfuric acid and methanol followed by derivatisation to the methyl ester of trifluoroacetic acid. This is accomplished within a single vial without complex extraction procedures. The highly volatile methyl ester is then analysed using headspace gas chromatography. The spike recovery trials from all media ranged from a low of 86.7% to a high of 121.4%. The relative standard deviations were typically below 10%. The minimum detectable limit for the method was 34 ng/g for dry plant material, 0.20 ng/g for soil and 6.5 ng/1 for water. 

An example of a recovery control chart is shown in Figure 4.7. The mean recovery of individual measurements is represented by the centreline. The upper warning limit (UWL) and the lower warning limit (LWL) are calculated as plus/minus two standard deviations (mean recovery + 2s) and correspond to a statistical confidence interval of 95 percent. The upper control limit (UCL) and the lower control limit (LCL) are calculated as plus/minus three standard deviations (mean recovery 3s), and represent a statistical confidence interval of 99 percent. Control limits vary from laboratory to laboratory as they depend on the analytical procedure and the skill of the analysts. 

In the method described, the procedural blank determination is essential for the analytical procedure since it allows the estimation of all kinds of contaminations. To assess the matrix effect for each sample, recovery rates were determined. The checking of the half range of the calibration graph for each 10 runs of samples allowed the evaluation of calibration graph deviation. 

The choice of an internal standard for an analytical procedure is often made in a too cavalier a fashion and may actually provide lower precision than external calibration. The successful use of an internal standard depends on the existence of a high correlation between the peak areas (heights) of the analytes and the internal standard for the complete analytical procedure and their being lower variability in the internal standard area (height) compared to those of the analytes [286-288]. An external standard will generally provide higher precision than an internal standard when the variation of the recovery for analytes and standards is sufficiently different and the standard deviation in the mean for repeated analyses of the internal standard is larger than that for the analytes of interest... 

To test for systematic deviations, explained by influences due to different steps of an analytical procedure or by the sample matrix, the recovery rate is calculated (c Eq. (4.6)). To investigate an analytical procedure for systematic deviations, the recovery function is applied. 

Method Performance. A blank sample, prepared using the same procedure as for the samples, was included with every five samples. PCB 28 and y-HCH were the only compounds detected in the blanks. Detection limits, calculated as mean blank +3 SD, were typically 2.3-13.3 pg/pF = 0.02-0.12 ng/g soil. Results were not blank corrected. Replicate analysis (the same soil sample extracted three times) was done for several samples. The relative standard deviation (RSD) for replicate analysis was always less than 20% (n = 3). Analytical recoveries were monitored with the aid of two recovery standards mirex for FI and 5-HCH for F2. The mean recovery for mirex was 100 ... 

Tanaka et al. [ 16] have described a spectrophotometric method for the determination of nitrate in vegetable products. This procedure is based on the quantitative reaction of nitrate and 2-sec-butylphenol in sulfuric acid (5 + 7), and the subsequent extraction and measurement of the yellow complex formed in alkaline medium. The column reaction is sensitive and stable and absorbances measured at 418 nm obey Beer s law for concentrations of nitrate-nitrogen between 0.13 and 2.5 xg/ml. In this procedure, the vegetable matter is digested at 80 °C with a sodium hydroxide silver sulfate solution, concentrated sulfuric acid and 2-sec-butylphenol are added, and after 15 minutes of standing time the nitrated phenol is extracted with toluene. Finally, the toluene layer is back-extracted with aqueous sodium hydroxide and evaluated spectrophotometrically at 418 nm. The standard deviation of the whole procedure was 1.4%, and analytical recoveries ranged between 91 and 98%. 

Second, work in the sample matrix. Choose a sample matrix for automating the manual SPE method that is free of analyte. Thus, all the interferences and other problems associated with the matrix will be addressed immediately. Third, spike the clean sample matrix to a known concentration so that recoveries may be calculated. Samples should be spiked to a midlevel concentration range. Finally, run samples in triplicate. Triplicates will give the percent relative standard deviation and are some indication of the reproducibility of your automated procedure. Samples that show a low relative standard deviation typically prove that the method is a well-written automated procedure. High relative standard deviations suggest problems with the matrix and recovery of the sample. 

Spectrophotometric analysis of cyclohexylsulphamic acid (46) and its salts has been carried out. The procedure involves conversion of the cyclamate to N,N-dichloro-cyclohexylamine (123) using excess hypochlorite. 123 is determined by measuring its UV absorption at 314 nm. Two collaborative studies have been reported using this analytical technique for the determination of cyclamate in soft drinks, desserts and jams162,163. The first study reports the results of nine laboratories assaying 3 soft drinks with cyclamate levels of 0.36-0.37 g kg"1 and 3 jams with cyclamate levels of 1.23-1.50 g kg -1. Average recoveries of cyclamate were 99.7% in the soft drinks and 103.8% in the jams with reproducibility coefficients of variation of 6.7% and 4.4%, respectively. The second study involved determination of cyclamates at much lower concentration levels, namely 90-311 mg 1 1 and 202-526 mg kg-1. The results from 15 collaborators gave cyclamate recoveries of 97.5% in soft drinks with relative standard deviations from 4.7% to 6.5%. The recovery from desserts was 98.6% with relative standard deviations of 6.9% to 8.5%. 

In some cases SFE, MAE, or PFE cannot be used because of the weakness of the targeted compounds. An analytical method based on a soft extraction procedure was applied to the extraction and quantification of the OXA and ESA acid metabolites of the acetochlor and metolachlor herbicides in soil samples. The extractions were performed by using 50 or 100 g of soil and a solvent extraction method with a mixture of acetonitrile/water (60/40) in acid medium. Each of the four different soil matrixes was spiked in triplicate with standards of each degradation compound, at three concentration levels between 2 and 80 /rg/kg. The average recoveries range from 90 to 120% for both metabolites, with relative standard deviations lower than 15%. The limits of detection were about 0.5 and 2 /rg/kg for the ESA and the OXA metabolites, respectively. Sample preparation and chromatographic analysis of acidic herbicides in soils and sediments were recently reviewed by Macutkiewicz et al. ... 

The detailed residue analysis method must be reported including the sample extraction procedure and clean-up. The method of detection must be simple and feasible, and the instrumentation and its operating conditions are to be acceptable for residue laboratories. Secondly, the recovery, RSD (relative standard deviation), and sensitivity (limit of detection and minimum concentration of detection) of the analytical method should meet the requirements of the Residues Experiment Guideline of Pesticide Registration. 

---