Guidelines for Use

These standards are manufactured specifically for DC-arc calibration. They are designed to be used by themselves to create a complete calibration curve for the ten ASTM impurities. It is strongly recommended that no other “standards” be used with the CopperSpec™ series.

The CopperSpec™ standards are manufactured in such a manner as to produce extremely accurate analyte impurity levels. Introducing other data points risks diluting the accuracy of these standards. The stated certified value should always be used. The estimated uncertainty reflects the greatest cumulative error which might possibly affect each analyte at the given concentration. However, extensive analysis and statistical evaluation confirm that the certified values of these standards are extremely close to the true value.

If your results fail to create r2 correlations approaching 1.0000 using the certified values, then error in method or technique is to be suspected. The most common error is the inclusion of other “standards” with the CopperSpec™ series. This desire to put “everything in the drawer” into the curve and draw a line through the cloud is understandable considering what has previously been available for DC-arc calibration. It is our hope that spectroscopists will allow a change in mindset to reorder the way they calibrate their instruments.

Figure 1. Bi Calibration Curve (see above) using CopperSpec standards.

A. Gating: Please consider these timing gates as a suggestion or starting point only. Experienced spectroscopists will not need any guidance in this area. For a method using 250 mg standard and sample size @ 12 amps we suggest the following gates for a 90 second burn: 1 sec → 90 sec Ag, Bi, Pb, Sb, Se, Sn, Te 1 sec → 50 sec As, Zn 1 sec → 20 sec Ni 1 sec → 15 sec Fe

B. Gap and Arc technique: The gate settings in the previous section assume a stable 4 mm gap is achieved within 1 second of the arc strike. This can be achieved by starting with a 0.5 mm gap between the sample and the counter electrode and then rapidly drawing the electrodes apart. A preferred technique is to start the counter electrode at + 2mm. Stand the standard pellet on edge. The top of the pellet is moved to -0.5mm. This 2.5 mm gap is easily bridged by the high voltage AC leader. The electrode controls are not adjusted when the initial arc is formed. The standard or sample is allowed to “melt down” to form a globule whose top is approximately -2 mm. Note: the bottom electrode may need to be adjusted very slightly to move the bead to the -2 mm mark. It has been our experience with the Thermo Elemental® instruments that this strike and gap method rarely fails to strike the sample and also gives very low detection limits for Fe and Ni

C. Calibration: New electrodes should be used for intensity collection. Start with the lowest standard and work up. At least four shots for each standard should be collected. Repetitions with intensities that look like outliers should not be discarded unless there is a reason to suspect the data was influenced by an errant initial strike or other problem. The Thermo Elemental® CID (charge injection device) instruments are susceptible to small drift over even short periods of time. Collection of ten to twelve reps of the same standard will reveal this wandering oscillation. The best calibration curves are generated by including all reps. Curve fit the data using full fit. Select “None” for weighting. The ThermoSpec® software default is 1/concentration. A more accurate fit is achieved by selecting “None.”

D. Method Validation: Good laboratory practice dictates that the new calibration should be validated prior to use. High quality certified reference materials (CRMs) such as the BCR 074 and 075 and the NIST Copper benchmark series (457, 494, 495, 498) or the CopperSpec™ CS-075 should be analyzed as unknowns. This will ensure that the new calibration curves were properly created.

E. Daily Control: For CID or CCD instruments it is suggested that the array maps be checked daily with a 10 ppm standard and re-center wavelengths as needed. Re-standardize using the method high and low standard if needed. The low should always be the 1.0 ppm and the high should be the highest standard in the curve or 10 ppm, whichever is lower. A CRM should be run at the beginning and end of each set of unknowns. Recommended CRMs are the CopperSpec CS-075, NIST Copper benchmark CRMs (457, 494, 495, 498), or the BCR 074 and 075 CRMs. Other “CRMs” may be used but should be evaluated against the BCR 075 which is the “gold standard” of copper CRMs. Not all “CRMs” are created equal. There are some of very poor quality which should never have been released.

F. Guidelines for Leeman Prodigy® are similar with the exception that the Prodigy® software does not include the option to weight lower concentration standards. For this reason it is suggested that higher concentration standards (higher than normal daily unknowns) NOT be included in your calibration curves.  The normal noise inherent in DC-arc integration will cause these high standards to unduly dilute the accuracy of the low end of your calibration curve.  For example, there is no reason for an SXEW producer to use anything higher than 10ppm (or even 5ppm) as their highest standard.

These Guidelines for Use are offered as a guide only. Contact us with your input at support@copperspec.com.

Teledyne Leeman Prodigy® is a registered trademark of Teledyne Leeman.  ThermoSpec® and Thermo Elemental® are registered trademarks of Thermo Electron®.