Combined Alpha (3^rd) and Gamma-Ray (5^th) Spectrometry Users’ Forum

Globe Room, Bushy House, NPL – 3 December 2002

Participants:		Ian Adsley	RWE Nukem
		Arzu Arinc	NPL
		Roger Benzing	Imperial College
		Chris Carter	High Technology Sources
		Sean Collins	NPL
		Matthew Curtis	VLA
		Julian Dean	NPL
		Mike Dolan	AWE
		Mike Froggatt	BNFL
		George Ham	NRPB
		Arvic Harms	NPL
		Trevor Hatt	Ametek
		Hilary Hayward	AWE
		Janice Holding	DRPS
		Elizabeth Howie	British Energy
		Simon Jerome	NPL
		Michael John	BNFL
		Gareth Johnson	AWE
		Laurence Jones	DRPS
		Steven Judge	NPL (Chairman)
		Arthur Lally	RTC Ltd
		Richard Lukey	VLA
		John McQuirk	BNFL
		Ann Mair	UKAEA
		Shona Mansfield	DRPS
		Desmond McMahon	NPL
		Fionna Miller	UKAEA
		Susan Parry	Imperial College
		Graeme Parkin	NPL
		Dan Parvin	BNFL Instruments
		Andy Pearce	NPL
		Antony Rigby	BNFL
		Bill Rogerson	BNFL
		Martin Rushby	AWE
		Ben Sanderson	BNFL Instruments
		Bud Sielaf	Canberra
		John Simpson	RWE Nukem
		Barbara Tawton	BNFL
		Pam Thompson	AWE
		Tim Twomey	Ametek
		Stephanie Vallet	Canberra Harwell
		Sylvie Ward	Canberra Harwell
		Tony Ware	Environmental Energy Consultancy
		Phil Warwick	Southampton Oceanography Centre
		Nigel Watkins	NPL
		Denise Woods	NPL
		Mike Woods	NPL
		Simon Woods	NPL (Secretary)
		Larraine Young	NRPB
		Mike Youngman	NRPB
Apologies:		David Copplestone	University of Liverpool
		Ian Croudace	Southampton Oceanography Centre
		Ron Crawford	UKAEA
		John Hemingway	University of Liverpool
		Phil Hurst	British Energy
		Alan O'Brien	Raigmore Hospital
		Steve Waters	IRS Ltd
		Dave Wickenden	AEA Technology

1. Welcome and Apologies for Absence

1.1 Steven Judge introduced himself as the new Chairman of the Gamma Spectrometry Users Forum and the Alpha Spectrometry Users Forum and explained that it was likely that the two fora would be combined in future. This matter would be addressed later.

1.2 Apologies for absence were given as above.

2. Approval of Minutes

2.1 2^nd ASUF 28 November 2000
Approved nem com

2.2 4^th GSUF 20 November 2001
Approved nem com

3. Matters Arising

3.1 GSUF Action 4.1: NPL (Simon Woods) to arrange a meeting on QA/Test Spectra.
These matters would be addressed during this meeting.

3.2 GSUF Action 4.2: Manufacturers to address the 5^th GSUF on the QA and validation of their software.
This meeting.

3.3 GSUF Action 4.3: NPL (Simon Woods) to arrange for someone from their Software Validation Group to give their views on software QA and validation at the 5^th GSUF.
This Meeting.

3.4 GSUF Action 4.4: Martin Phillips to send measurement data to NPL (Simon Woods).
Matter is not urgent and may be processed between Berkeley and NPL independently.

3.5 GSUF Action 4.5: Typographical errors in nuclear data in The Radiochemical Manual: NPL (Simon Woods) to inform GSUF members and enquire as to any further errors spotted.
Three errors have been identified - zero’s being dropped or added to printed gamma-ray emission probabilities. These errors have already have been communicated to users. No further mistakes have been spotted.

3.6 GSUF Action 4.6: NPL (Mike Woods) to contact DTI regarding ownership of The Radiochemical Manual.
The legal situation is still unclear (authors v Crown Copyright). It was confirmed that Harwell Analytics are still marketing the document.

3.7 GSUF Action 4.7: NPL (Simon Woods) to arrange discussions (regarding Field Measurements) with Martin Phillips and Matt Simpson.
No progress as yet, but will be progressed outside of this meeting.

3.8 GSUF Action 4.8: Manufacturers to provide updates on detectors other than HPGe at the 5^th GSUF.
This meeting.

3.9 GSUF Action 4.9: All to advise Simon Woods if there are other topics that they feel should be addressed.
No direct comments received, but normally picked up by the invitation for users to request matters to be raised.

3.10 GSUF Action 4.10: All to return questionnaires (if desired) to Simon Woods by 31/12/01.
Very few questionnaires have been returned. The questionnaires have been reissued with the meeting information pack in the hope of a better response.

3.11 GSUF Action 4.11: NPL to produce summary report of returned questionnaires.
Given the low response to GSUF Action 4.10, this action has been dropped.

3.12 GSUF Action 4.12: All to provide feedback to Simon Jerome on suitable calibration sources.
No responses were received.

3.13 ASUF Action 2.1: NPL (Simon Woods) to issue a proforma asking for available details of UK services and facilities.
Issued, but as with the GSUF there was a very low response.

3.14 ASUF Action 2.2: NPL (John Makepeace) to establish AMUG.
Completed with two meetings held.

3.15 ASUF Action 2.3: NPL (Simon Woods) to arrange uncertainties subgroup meeting.
This was deferred pending the outcome of the GSUF WG with a view to possible amalgamation.

4. New Forum Format and Terms of Reference

4.1 Attendees agreed that an amalgamation of the ASUF and GSUF would be beneficial as many aspects of concern are common to both fora, e.g. calculation of uncertainties, spectral deconvolution.

4.2 A single forum, called the Nuclear Spectrometry Users’ Forum (NSUF), would be held in future. This would have the remit to include any other forms of nuclear spectrometry that the users felt appropriate.

4.3 Regarding new Terms of Reference, as the ASUF and GSUF ToRs are identical in all respects apart from the reference to technique, it was agreed to simply replace “Alpha Spectrometry” and “Gamma-ray Spectrometry” with “Nuclear Spectrometry”. These new terms of reference are attached.

5. Invited Talk: Validation of Measurement Software
SSfM Best Practice Guide No 1
Graeme Parkin (NPL)

5.1 A summary of the talk will be made available on the website.

5.2 Extensive discussions arose:

5.2.1 Mike Woods:

• Queried how assessors would be affected.
• Ideal would be certification of software.
• Need international status for SSfM document.
• Only official stamp at present is being developed at NPL.

5.2.2 Ian Adsley:

• Queried the links with TickIt. The NPL guide appears to be more specific.

5.2.3 Mike Woods:

• For assessment, software needs documentation and version control. The use of spreadsheets (e.g. MS Excel) requires particular care. The NPL website does have guides on how to develop programs (see http://www.npl.co.uk/ssfm/download/ ).

5.2.4 Bill Rogerson:

• Need to distinguish between software being developed and high level software.
• The NPL guide does explain the security that’s required.

5.2.5 Steven Judge:

• Suggested a Good Practice Guide might be required longer term for validation of software for nuclear spectrometry.
• Action 1.1: NPL to report back on how the various guides were implemented during the installation of their new gamma spectrometry software.

6. Invited Talk: Environmental Comparisons
Julian Dean (NPL)

6.1 These slides will be available on the NSUF website..

6.1.1 Comments raised:

• It is not too late to book onto the current comparison, though returns are required by the end of February 2003.
• Comparison Workshop to discuss the results will be held 8^th May 2003.
• There is as yet no explanation as to the low return of ⁹⁹Tc results.

7. Invited talk: Mass Spectrometry Techniques
Phil Warwick (Southampton Oceanography Centre)

7.1 A summary of the talk will be made available on the website if possible.

7.2 Comments raised:

• Ian Adsley queried whether MS would overtake alpha spectrometry. Phil Warwick replied that for some nuclides it has already done so, particularly ⁹⁹Tc, ²³⁷Np and ²³²Th. Certain radionuclides are still best done by alpha spectrometry. ²³⁸Pu can be measured by MS, but a special detector is required. For ^239/240Pu MS is currently the preferred option.
• System is calibrated by a ²⁴²Pu standard.
• Cost differential? ICP-MS is cheaper than alpha-spec, but if an M-Collector is required then the cost can rise to £30-50 per sample due to set-up costs.
• Laser ablation is possible; though efficiency will be reduced.
• Cost of a typical machine: Between £500k and £1M.

8. Manufacturers’ Presentations on QA

8.1 Bud Seilaf (Canberra [UK: Canberra-Harwell])

• Presented an overview of Software Quality Assurance, Verification and Validation at Canberra.
• Processes are compliant with IEE730 and NQA (Waste acceptance and waste assay). All are certified to ISO-9001; “per-site” certifications are also provided. Note: ISO-9001 covers both Design and Manufacturing; ISO-9002 Manufacturing only.
• Software quality resides in design…
  Concepts ® Requirements ® Design ® Implementation ® Test ® Release ® Operations and Maintenance
• Use quality feedback through software performance reports, with a significant focus on those aspects adverse to quality.
  1. Data analysis and verification:
  2. Verification of calculations
  3. Test spectra available covering a wide range of applications
  4. Tests are performed constantly
  5. All changes are documented
  6. V&V documentation is available summarising QA results
  7. QA testing is involved from the very beginning
• New product testing:
  1. All new methods are compared with traditional methods or other known references
  2. Numerous 3^rd party/customer tests are employed
  3. Verification on individual detector performance is maintained
• Test spectra:
  1. In-house spectra and those accumulated from customers are available
  2. It is recommended that site operators use test spectra
  3. Test spectra should be run whenever there is reason to believe any change to the software has occurred.
  4. It was queried whether, when changing software, does the user need to keep an old copy of the software. Mike Woods replied that this is incumbent on the Manufacturers, who confirmed that this practice was done.

8.2 Tim Twomey (EG&G Ortec)

• Presented a description of Ortec’s processes and procedures for QA
• Again working to ISO 9001 and moving to ISO 2000 accreditation.
• Specification of software is very important; the process is very similar to that given above with everything being done on a project basis.
• Product challenges/Goals:
  1. Enhancements
  2. Operating system changes
  3. Literature changes
• Process overview:
  1. Design - “Forward Focus”
  2. Development
  3. Testing/Release
• QA:
  1. Any new product should be functionally useful to what the user needs
  2. It should do it in a useful manner
  3. It should do what it says it does
• Alpha testing performed by R&D Development Team.
• Beta testing subject to completion tests and appropriateness to application by “Scientist Review”.
• Final testing: This is the third complete test and is subject to Senior Scientist and Product Mangers review. Decision on (no-)release is made by the Product Manager.
• For testing examples, see Keyser (1996).
• Looking ahead:
  1. Planning QA resource additions
  2. Incorporating IEEE testing standards

8.3 Questions arising:

Q: Both presentations dealt with software testing: Is any associated documentation (e.g. the user manual) also validated?

A: Both manufacturers replied “Yes”, particularly as the manuals themselves are used as testing benchmarks. Also users are informed through regular users’ bulletins as to any changes.

9. User Topics

9.1 Portable High Resolution Gamma-Ray Spectrometry Measurements
Ian Adsley (RWE Nukem)

• The aim of the presentation was to obtain feedback from members of NSUF on possible solutions to the measurement of radioactive particles washed up on the beach at Dounreay
• Media is pushing for high sensitivity measurements
• The original measurement system used Geiger counters
• Newer system use NaI detectors
• Dounreay particles are very small (similar in size to a sand particle)
• Downscatter causes a problem, the background is 3 to 4 times the photopeak hence the particle must be less than 1 m from the detector to be “seen”
• In practice, a typical NaI detector has the following count rates in different parts of the spectrum:
  • Whole spectrum: Background 200cps; Count rate due to particle 50cps
  • Cs-137 window: Background 15 cps; Count rate due to particle 15cps
• Problem:
  Signal from a radioactive particle is close to the limit of detection
• Possible solutions:
  1) Bigger NaI crystals, slower measurements
  2) High resolution Ge detectors (a 100% Ge spectrometer detects 15 cps in Cs-137 peak due to a typical radioactive particle but the background is much lower than a NaI detector)
• However, the Ge detectors have disadvantages for this application:
  • Have to operate in a harsh environment
  • Detector cooling problems
  • Complex spectra – expert interpretation needed
  • Expensive
  • Robustness
  • Ease of replacement of components
• Desmond McMahon commented that with high resolution the peak/Compton ratio would be able to identify particle depth
• Ian Adsley replied that a better solution may be to cover as large an area with a NaI system as possible

9.2 Quality Assurance
Susan Parry (Imperial College)

• Some years ago found fault in software (decay during counting- on/off reversed!)
• She is pleased to see that software verification is now better
• The weakest link was the documentation
• Manufacturers need to ensure their documentation is accurate
• Tim Twomey suggested the use of accredited test spectra but went on to point out that the NPL test spectra were designed to test analysis algorithms rather than test functions i.e. performance testing for QA need simpler/routine spectra

Action 1.2 NPL programme formulation focus groups to follow up

• Mike Woods commented that ISO/IEC world standards exist as well as American standards aimed at manufacturers based in the USA
• Bud Seilaf observed that these US standards were derived from European standards

9.3 QA of Commercially Produced Software
George Ham (NRPB)

Already discussed

9.4 Uncertainty Budgets
Susan Parry (Imperial College)

Mike Woods observed that the next GSUF-Working Group meeting is due and that the next step is to address magnitudes of the possible uncertainties. Advice is also given in GUM/M3003 and also a NPL GPG. Examples in these guides are being extended into the radiological area e.g. GPG 49 soon to be published. A draft of this document is available on the NPL website following the IRMF link.

10 Any Other Business

• Simon Woods gave a short description of the UKNSF which is now being managed by the HSE with the secretariat being provided by NPL.
• It was noticed that an IAEA TecDoc update was being undertaken, to be published shortly. The manufacturers would insert these data as a separate file in their systems in order to maintain traceability

11 Manufacturers’ Updates

• The manufacturers were asked to give information on alternative detectors
• EGG (Tim Twomey) briefly described the Radiant 200 portable CdTe detector (marketed through Ortec)
• Canberra (Bud Seilaf) also described APEX (a new software package for controlling / sequencing the measurement of samples)

12 Date of Next Meeting

Provisionally Thursday 18^th November 2003

 

Simon Woods

Secretary NSUF