NATA


Dr Stephen Davies
Team Leader, Chemical Reference Materials

Comparability and accuracy are critical components of chemical measurement.  Metrological traceability is the important concept ensuring comparability of measurement results both nationally and internationally, while the accuracy of measurement results will be delivered by accurate instrument calibration and appropriate method validation.  Laboratories performing chemical analysis will often report their measurement results as a mass fraction of analyte in a given matrix (e.g. µg per kg).  Implicit in the reporting of results as a mass fraction is that metrological traceability to the SI unit for mass (kg) has been established. 

ISO/IEC Standard 17025 outlines the responsibility of chemical testing laboratories for establishing and maintaining metrological traceability of their measurement results.  NATA’s policy document on metrological traceability, titled Policy Circular 11, is aligned with ILAC Policy P10, and describes how metrological traceability may be established through use of reference materials, placing the onus on the analyst to assess the material and associated documentation.  For chemical reference materials a wide range of issues need to be considered. 

A common issue is the quality of the supporting documentation provided by reference material producers (RMPs), which should clearly distinguish whether the material is a reference material (RM) or a certified reference material (CRM).  By definition an RM must possess property values that are homogeneous and stable.  The property values of CRMs have the same basic characteristics but in addition are reported with an uncertainty and a statement of metrological traceability.  The problem faced by many laboratories is that while so-called certificates of analysis imply CRM status the material does not comply with the definition listed in ISO Guide 31:2015(E).  Ultimately, the variations seen in the quality of certificates issued by accredited RMPs places a responsibility on the end user to check that the documentation provides all of the evidence needed to support the assigned status. 

Another problem is that many commercially available chemicals are provided with an indicative purity, a typical statement being ‘98% by HPLC’, with little or no supporting information.  For a laboratory intending to use this material for instrument calibration the analyst will need to ask some rudimentary questions of the quality and metrological traceability of the assigned property value.  What does ‘98% by HPLC’ actually mean?  The implication is clearly that the material is 98% pure i.e. every milligram of material contains 980 g of the analyte of interest.  This, of course, is only true if the sample is anhydrous, contains no solvent and no non-volatile residue such as inorganic salts.  This is an unlikely scenario and in all likelihood, chemicals which have their purity values assigned by a limited number of techniques will suffer from bias in the purity estimates. 

In a proficiency testing scheme for the quantification of folic acid in flour, organised by National Measurement Institute Australia (NMI), it was noted that a significant number of participants had used a commercial sample of folic acid as their calibration standard, quoting the statement of purity (>97%) listed on the bottle label.  What they had all failed to take into account of was the 8% water content listed in the associated documentation.  This oversight lead to significantly biased calibration curves and related results, and highlighted the need to review the documentation to ensure that typical impurities have been considered and taken into account. 

Producing high quality organic calibration standards is not a trivial exercise, the Chemical Reference Materials team at NMI having dedicated the past 20 years to improving the accuracy and establishing metrological traceability of assigned purity values.  Participation in international comparisons with other NMIs and in-house experience has led to significant improvements in all facets of purity assignment.  For example, it is no longer considered sufficient to use chromatographic peak areas as a direct ‘measure’ of organic purity.  Further work is required to assess the response of the main analyte and related impurities in the detector of choice to facilitate conversion of chromatographic peak areas into an accurate realisation of the relative mass fraction of the main analyte.  This is particularly important with HPLC-UV analysis where small changes in structure, especially conjugation, can have a dramatic impact on the molar response.  The RMP would also need to consider whether impurities, not possessing a suitable chromophore have gone un-detected, thereby introducing bias into the purity estimate. 

Another key improvement has been the routine assessment of water content by Karl Fischer titration.  This is crucial for a number of reasons.  Firstly, with very few exceptions, all organic calibration standards in our collection (over 500 steroids, forensic drugs and agrichemicals) contain moisture.  Secondly, water absorption and/or desorption has been shown to be the most common source of instability and needs regular monitoring. 

Proton nuclear magnetic resonance spectroscopy not only provides highly specific structural information, but also assists in determining levels of solvents etc.  When operating under specific ‘quantitative’ parameters it provides an independent direct measure of the analyte of interest making it a critical tool in modern day purity assignment.  Agreement between purity values assigned by the traditional mass balance approach, which quantifies all impurities and subtracts from 100%, and quantitative NMR provides the highest level of assurance for the accuracy of the assigned purity value. 

Over the years it has become increasingly evident that purity assignment with established traceability is far from routine, requiring highly qualified chemists with backgrounds in analytical and organic chemistry.  From the end-users’ perspective sourcing high quality certified reference materials from appropriate accredited Reference Material Producers and being ever vigilant of the quality of the certification will go a long way to ensuring the quality of their chemical measurements. 

Upcoming training dates in Perth
16 May                 Verifying Non-automatic Weighing Instruments                               
17 May                 Verifying Weighbridges                                                                                
24 May                 Introduction to Estimating Measurement Uncertainty               
    
For any enquiries regarding NMI training courses, email This email address is being protected from spambots. You need JavaScript enabled to view it.

 

ANNOUNCEMENTS
Veterinary Testing Classes of Test and Application Document updated Veterinary Testing Classes of Test and Application Document updated
Veterinary Testing Classes of Test and Application Document updated has been recently updated on the website. These documents can be downloaded from the 'Accreditation Publications' section of the website:

Read More...

 
Latest Edition of NATA News Latest Edition of NATA News
NATA News - Aug 2017In This Issue- ISO/IEC 17025 Revision- Assessment if hydrocarbon pollution- Preparing for the new kilogram and More

Read More...

 
ISO/IEC 17025 Review updates ISO/IEC 17025 Review updates
August 2017 update on ISO/IEC 17025 Review has been recently posted on the website

Read More...

 
Nominations invited for the NATA Non-destructive Testing Accreditation Advisory Committee Nominations invited for the NATA Non-destructive Testing Accreditation Advisory Committee
The NDT Accreditation Advisory Committee (AAC) is established to provide input to NATA’s NDT field of accreditation in a way that represents the interests of accredited NDT facilities and the broader NDT industry

Read More...

 
NATA Young Scientists Of the Year Awards NATA Young Scientists Of the Year Awards
The annual NATA Young Scientists of the Year Awards 2017 competition is now open

Read More...