Antares Analytical Tech Bulletin

July, 2003

Subject: Standards Selection

Relevant Models: All Baird Spectrometers

Relevant Matrices: All Matrices

Background Information: This Tech Bulletin will cover the selection of Calibration and Standardization Standards for any particular metal matrix. Unless specified, all the following information will cover all standards (metallurgical, ICP and oil / fuel) but selection procedures for Oil, Fuel and ICP / AA standards will be covered separately. BAIRDAS software calibration procedures will be covered in a future series of Tech Bulletins covering Calibration & Standardization software procedures.

With the extensive experience Antares Analytical Tech people have with Standards & Applications selection, our knowledge of  Standards manufacturers / suppliers worldwide and our unbiased advice, we are well qualified to help you with any problem you may have with your Calibration or Standardization requirements. We try to meet your specific requirements, within a reasonable budget, and can advise on alternatives when necessary.

Types of Standards: There are generally 2 accepted types of standards which will be used for calibration and standardization (OES, ICP & Oil Analysis). One or both types can be used for standardization techniques such as “type standardization” “drift correction” or “check samples.” These are normally a software / calibration subject and will be covered in a later Tech Bulletin.

·         Certified Calibration Standards (known as CRM or Certified Reference Material) are standards used for spectrometer calibration only, though can be used for standardization. These standards  have been analyzed by a large number (usually 10 or more) of different laboratories using a number of different techniques. The results from all labs are averaged (high and low analysis not used) and that average is certified by the standard manufacturer. A variation for the certified result may or may not be included on the Certificate of Analysis

·         Secondary Reference Standards (known as RM or Setup Standards – SUS) are samples which have been produced strictly for periodic standardization of calibrated curve sets. These are normally analyzed by only several labs and the results issued are not certified. Cost is a major consideration when producing an RM / SUS standard. By using a smaller number of labs for analysis, and producing a larger number of samples per batch, this helps to keep the cost reasonable.

General Information for Standards: All commercially produced / available standards all have several things in common:

• Standards are produced by a number of different standards manufacturers, both as CRM & RM / SUS materials. There are more than 60 different Standards manufacturers, covering all metallurgical alloys, worldwide.
• Generally the more common the alloy type, the more standard sets are available for it. Example: For Low Alloy Steel (and it’s alloy types within the LAS designation) over 150 different CRM sets from 20 different manufacturers / suppliers are available. Differences are in elements analyzed, LAS type (PCS, Low C Steel etc) and manufacture (rolled, wrought etc)
• Standards (especially RM / SUS type) should be available in quantities over a period of years. As time goes on, new SUS stds will be replaced as old ones are used up. While it is possible to go to a different standardization sample, it’s advisable to stay with the RM / SUS used during calibration … even if a different batch number.

Choosing Certified Calibration Samples (CRM):

·         CRMs should be selected for the basic Alloys type you are analyzing, with sub alloys types in mind (for example: choosing a broad range LAS set for analysis of low C steel and plain C steel).

·         Standards MUST be certified by the manufacturer and preferably have information on the types of analysis used, the number of Labs used in the Certification process, and statistical data for the various analysis. While this information is not always provided, it is preferable to have it.

·         It is NOT advisable to “mix & match” various alloy types within a given curve set, except when using a point as an “anchor” for the calibration curve. An example would be using a Low Alloy Steel Standard to “anchor” the Ni and Cr curve in a Stainless Steel calibration set.

·         It must be remembered sometimes CRM standards are not available for every possible alloy type in the world. While there may be Company specific standards for a unique alloy, this unique alloy may not have exact CRM’s available from one of the Commercial Suppliers. In this case it is best to get Applications advice from knowledgeable people, such as Antares Analytical or your spectrometer manufacturer.

·         Generally CRM’s will be more expensive than RM / SUS stds, but will only be used for calibration and not day to day standardization, however cost should be considered when choosing CRM’s. More expensive doesn’t always mean a better standard. Again Antares’s extensive experience can be useful to you when knowing which CRM’s are most cost effective.

Choosing Secondary Reference Materials (Setup or SUS samples): These are the samples normally used for periodic standardization of the spectrometer. Characteristics of setup standards are:

• Any samples chosen for periodic standardization should be homogenous, available from the manufacturer / supplier over a long period of time (years) and have a reasonable cost.

• Normally these samples will come with a Certificate of Analysis, however will  usually not be certified. Since the manufacturers goal is to have a good, low cost supply available, their samples will usually be tested by only 2-3 outside Labs, making the analysis less accurate than a CRM.
• Setup standards should not be used, by themselves, for calibration. They can be run along with CRM standards and included as part of the calibration curve, but should never be relied on, by themselves, for calibration.
• Setup standards do not have to plot on the calibration curve to be accurate. A wrong idea many customers have is that an SUS needs to plot on the calibration curve, for accurate standardization
• Almost all manufacturers of spectrometers standardize calibration curves using “Intensity or Intensity ratio” and not concentration percents (the exception will be a Type Standardization”  which usually does use actual concentration).
• Baird spectrometers (all models) use the element’s Intensity Ratio (I/R) result for periodic standardization. To standardize a curve set, the BAIRDAS software compares the current I/R to the stored I/R and calculates the factor & offset (N & O on the MOA) by comparing these 2 results. This “factor & offset are then applied to all unknown samples run, until the next  standardization is done.
• Setup standards of different alloy types (LAS samples in a Stainless Steel alloy for example) can be used to standardize an alloy. Standards of different base materials are never mixed to standardize alloys (ie: an Aluminum sample can not be used to standardize Copper alloys, Copper can’t standardize Steels etc)
• Some companies will manufacture inhouse setup standards and this practice is perfectly acceptable. When this is done, the company should manufacture a large amount of the sample specifically for this Standardization. The company manufactured sample should be as homogenous as possible, then representative samples should be sent to an outside lab for independent analysis.
• A good  number of each sample should be manufactured and then stored for replacement as current samples get used up. Depending on number of times per year the spectrometer is Standardized, a minimum of 10-12 discs of each Standardization sample should be produced & stored.
• It is bad practice to grab a sample off the floor and call that your setup standard …. There is no guarantee of the rod or bar homogeneity and if there is a difference in chemistry at different spots on the bar (very common), it will affect your standardization later, when you use different sample from that bar.

Liquid and Oil Standards: Most of the above information is applicable to ICP / AA / Rotrode (Rotating Disc Electrode as in the Baird MOA or FAS spectrometers) spectrometers, but there are major differences.

• Liquid / Oil standards are almost always made at standard PPM levels because of the calibration techniques of a “liquid” spectrometer (10 PPM, 100 PPM, 1 PPB etc). Vendors will make special standards to a customer specific needs. Manufacturing a liquid standard to a customer specific application is much easier, and less costly, than with metallurgical standards.
• There are fewer manufacturers / suppliers of ICP and Oil / Fuel standards worldwide.
• There is no “CRM” or “RM” with liquid standards.
• Liquid standards have a “shelf life” after which they are no longer certified to be accurate, similar to expiration dates on milk.
• Many labs buy their own “base” ICP standard, usually 100 or 1000 PPM, and then “mix down” their own standards. Using good lab techniques this practice is acceptable, but because of the nature of Oil / Fuel standards trying to “mix down” your own can lead to poor results.
• Liquid Standards are always made to a specific customer Purchase Order, rather than” stocking” standards.

__________________________________________________________________________­­­­­­­­­­­________

This Technical Bulletin is edited by Gerry Smith, Technical Services Manager of Antares Analytical. Having spent 25 years working with Baird spectrometers (22 years with Baird Analytical in various capacities, including many years as Final Test / Calibration Manager and Senior Service Engineer, International Operations) he or any of our Technical Support Staff, are well qualified to answer any Technical or Applications problems you may have.

For further information or access to past monthly Technical Bulletins covering a variety of subjects, please  visit our website (still under development) at www.antaresanalytical.com .For more information, to be added to this list,  or help with other subjects including Tech Support, spare & replacement parts, or on-site Service by qualified Antares Analytical Service Engineers, please contact us at info@antaresanalytical.com 

This bulletin is meant for informational purposes only and it not meant as a substitute for Service performed by a qualified Service Engineer. Any use of this information is done strictly at the discretion of the user. Antares Analytical, nor any of it’s employees or Partners, is not responsible for any misfortune which results as the result of use of these Technical Bulletins, in place of service by a Qualified Service Engineer.