Data Validation of Graphite Furnace Analysis
The analyst follows the proper documented methodology found in the Standard Operating Procedures to analyze a
batch of samples. The analysis is documented in a bound notebook.
For each run, a record of the following information is kept: date, time, analyst
initials, analysis name, method number, PMT voltage, lamp current, standard calibration curve, QC sample, sample identification numbers, dilution
factors, readings, slope and intercept of standard calibration curve, calculations, and comments/concerns of the analyst.
At the start of each
analysis the following information is recorded on the chart: analysis name, method number, date, analyst initials, PMT voltage, and any other notes by the
analyst. The absorbance received from the injection is recorded on the chart next to its peak.
For each standard calibration curve obtained, the database generates a least-square analysis determining the best fit
curve as well as the actual curve. The slope and intercept is calculated and used to determine the concentration in ppm for the sample tested.
[Result(ppm) = m x +b, where m = slope, x = absorbance determined the analytical analysis, and b = intercept. ]
When the analysis is complete, the analyst reviews the calculations and enters the data into the database.
The data is then printed and reviewed. The printout of the run includes the following information: sample identification number, result, percent
relative error for the duplicates, percent spike recovery, date, and analyst initials.
The analyst reviews the control chart *** Did the above process produce a control chart?*** for duplicates and the mean control chart for spike recoveries for the last 20 samples
entered into the database for that particular analysis. The analyst checks to ensure that the samples were in control and that the run was stable.
The analyst the reviews the printout of the run and signs and dates it. The printout is attached to the strip chart, and with the notebook is given to another analyst for review.
When completed, the other analyst signs and dates the printout and returns it back to the original analyst. The results remain in the database for
the final report assembly. The final report is reviewed, signed, and dated by the laboratory director.
