24^th January 2006

National Physical Laboratory

Attendees:			Simon Jerome	NPL (Chairman)
			Arzu Arinc	NPL (Secretary)
			Matthew Ashworth	DRPS Labs
			Trevor Birkett	AWE
			Andrew Boschier	Veterinary Labs Agency
			François Briot	PROCORAD
			Pete Burgess	UKAEA
			Darryl Campling	UKAEA
			Sally Chapman	AWE
			Andrew Clarke	PerkinElmer
			Anthony Curran	Beckman Coulter
			Julian Dean	NPL
			Laurent Exmelin	PROCORAD
			Chris Gilligan	NPL
			George Ham	HPA
			Arvic Harms	NPL
			Eileen Hayden	RPI
			Robert Huggett	Zinsser Analytic
			Kevin Kelleher	RPI
			David Lambert	Veterinary Labs Agency
			Ian McCain	Beckman Coulter
			John McNamara	British Energy
			Jenny Morris	AWE
			Andy Pearce	NPL
			Jill Walker	RCD Lockinge
			Phil Warwick	National Oceanography Centre
			Claire Watts	AWE
			Dave Wickenden	AEAT
			Paul Wright	RPA
Apologies:			Phil Harrison	Cronus Technologies Ltd
			Steven Judge	NPL
			Lena Johansson	NPL
			Pieter Kwakman	RIVM

 

1. Chairman’s Welcome, Previous Minutes and Actions Arising

• Simon Jerome welcomed the members to the fourth meeting and explained briefly the aim of the forum. The delegates new to the forum were invited to introduce themselves.
• The minutes of the third meeting were accepted as a true record.
• Actions:

A3.1: Andrea Woodman and Steven Judge to contact Quantulus. Quantulus were contacted but unfortunately were not able to send a specialist to the meeting.

A3.2: Franz to let Andrea Woodman have relevant information about next LSC conference. Done. The meeting was held in Poland last October.

A3.3: Delegates to let Andrea Woodman have information about their problems with PerkinElmer. Done.

A3.4: Andrea Woodman and Steven Judge to write to PerkinElmer. PerkinElmer acknowledged the problem encountered by their customers and are trying to improve their service.

2. Invited talk: "LSC for surface contamination" – Dave Wickenden, AEAT

• Dave began his talk by describing the analytical services provided by Waste Management Technology Limited (WMTL).
• Dave explained that contaminated surfaces are measured essentially either by direct monitoring or by swabbing of the surfaces (followed by liquid scintillation counting in the case of low-energy b-emitters). He explained that swabbing of surfaces leads to very inaccurate results due to the very large uncertainties on the pick-up factors. Dave presented a range of factors to illustrate how much the amount of radioactivity picked up on swabs can vary.
• Dave gave a historical summary of recommended pick-up factors and concluded that, as a rule of thumb, 10% seems to be reasonably pessimistic and applicable to most situations.
• Dave explained the approach used for measuring ³H surface contamination at WMTL, including sampling, source preparation and method testing. Tests revealed different pick-up factors for different materials, including water damped swabs (~ 45% pick-up), methanol damped swabs (~ 16%) and dry swabs (~ 5%).
• Dave concluded by saying that this technique was appropriate for monitoring relative levels of contamination and to get a reasonable estimate of the degree of contamination and that pessimism was the safest approach (i.e. erring on the side of underestimating pick-up factors).

• Following the talk, Dave was asked if he had any experience of monitoring oil on surfaces. Dave replied that he had done some work on it but that it is difficult due to the high levels of quenching encountered. He was asked if water-damped swabs would be the best option for oil and replied that he would use dry filters.
• The need for obtaining the maximum information possible on samples prior to analysis was discussed. This can include the history of the material itself and the working area it came from. The need for good communication between the analytical laboratory and the end user was stressed.
• Simon Jerome asked if the technique described by Dave would be applicable if contaminants other than Tritium were present. Dave replied that the Tritium data provide a good starting point, but one would need to carry out further work for other radionuclides to identify the best approach.
• Peter Burgess commented that one needs to be careful when wiping the outside layer as the diffusion of one radionuclide can be different from another; for example, for elements which tend to diffuse such as Cs, Sr and Y, the fingerprint varies with depth.

3. NPL talk: "Shewhart chart Controls – ISO 8258:1991" – Simon Jerome, NPL

• Simon Jerome gave a brief presentation on the IOS standard dealing with Shewhart Control Charts and briefly touched upon the use of Cumulative Summation (CuSum) Charts. ISO 8258:1991 extends the use of control charts beyond simple violations of ‘2 ω – warning, 3 ω – action limits’. By the use of trend analysis techniques, it is possible to identify operational problems long before they become acute. This was illustrated with measurements obtained from liquid scintillation counters in use at NPL, where a deterioration in the performance of one of the photomultiplier tubes was identified over three months prior to the warning and action limits for the counter being breached.

• Following the talk, a question was raised regarding the frequency of the QA tests. Simon replied that one had to use common sense and that measurements done too frequently or too infrequently would not provide useful data.
• Phil Warwick asked if the QA software NWA has the capability of plotting CuSum charts. Simon replied that he thought it did, but he needed to check.
• George Ham commented that a balance had to be struck between the amount of time spent on QA measurements and on sample measurements. This is especially important when samples need long measurement times.

4. Invited talk: "Use of proficiency tests organised by PROCORAD to improve quality of tritium in urine measurement"
    – Laurent Exmelin, PROCORAD, France

• Laurent began his talk by describing the PROCORAD organisation. Then he explained the goals of PROCORAD, which are mainly to promote quality management in the laboratories and to facilitate exchange of knowledge.
• The various sample types and radionuclides made available in these exercises were described with a particular emphasis on the Tritium-inurine samples. Laurent pointed out the quality management involved in the preparation and dispatching of samples and the processing of the results.
• Laurent explained how the submitted data were processed with the PROCOSTAT software, which uses various statistical tests in order to reject outliers.
• Laurent showed some typical graphs, such as: (i) bias as a function of the laboratory number, (ii) reported and true activities and their uncertainties as a function of the laboratory number, and (iii) repartition of the laboratories as a function of activity intervals.
• Laurent pointed out that the quality of the results for the tritium in urine exercise had been remarkable and that a decrease with time of the number of outlier results was evident. He mentioned that the activities of OBT samples had been overestimated by the participants and that some of the details regarding the methods used were missing.

• Following the talk, one delegate asked what improvements the participants had made to improve the quality of their data. Laurent replied that they’ve observed a general improvement in the participants’ results, but that he wasn’t sure what the reason for this was. He suggested it might have been due to improved calibration protocols or use of new reagents, but added that it was difficult to find the reason due to lack of details regarding the protocols used.
• Arvic Harms mentioned that NPL sells certified OBT standards that are 97% pure and he asked which procedures had been used to identify outliers. Laurent replied that a result needed to fail three out of the four tests applied to be considered as an outlier.
• George Ham mentioned that the organisation of PROCORAD exercises was very good and it did run very smoothly.
• A participant raised the need for standardised samples in a milk matrix. Simon Jerome explained that NPL was aware of the demand but that it would not be possible to prepare such samples under the current NMS programme.

5. Invited talk: "Liquid scintillation counting in decommissioning issues" – Phil Warwick, National Oceanography Centre

• Phil started his talk by describing the work carried out at the National Oceanography Centre (NOC). The work is mainly radioanalytical chemistry, most of the work being associated with decommissioning and waste characterisation studies.
• Phil explained the advantages of working with liquid scintillation techniques and presented four case studies for his talk: Gross screening techniques, ³H measurement, ⁵⁵Fe measurement and Cerenkov counting.
• Phil explained that for α/β discrimination on the Quantulus counter, the misclassification of the β- and α-particles reached a minimum at a Pulse Shape Analysis (PSA) of 70; however, he added that the optimum PSA value shifts when the level of quenching in the vial is changed.
• Phil pointed out that liquid scintillation counting is a potential method for the screening of samples for the presence of β and α-emitters. He gave as an example the screening of zinc bromide which had revealed unexpected ³H contamination.
• Both leaching and combustion had been used as sample preparation techniques for ³H analyses of various sample types, and very good agreement had been observed between the two methods. A further example was given for ³H in metal profiling.
• Phil explained the technique used at NOC to separate ⁵⁵Fe from other activation products and the use of phosphoric acid to produce a colourless Fe solution when in chloride form.
• An example was given of the use of Cerenkov counting for simultaneous measurement of ⁸⁹Sr and ⁹⁰Sr. Good agreement had been observed with standardised values when using this technique.
• Phil pointed out “Air luminescence” as an interesting alternative technique for α-emitters, being the equivalent of Cerenkov counting for β-emitters.

• Following the talk, one delegate asked how soon the first samples were measured when dual measurements of ⁸⁹Sr and ⁹⁰Sr were undertaken. Phil replied that the samples were measured one day after separation and again 10 days later. Simon Jerome commented that the integrated decay growth needed to be calculated for the first measurement otherwise deviation from the true value would be observed.
• Andy Pearce mentioned that, in principle, the CIEMAT/NIST technique could be used to translate a ³H quench curve into a quench curve for any β-emitting radionuclide, but that he would recommend building up some experience of the technique before using it to calibrate the counters. Phil added that the efficiencies used for the ⁴¹Ca were derived from the CIEMAT/NIST technique, and that the agreement between experimental values and theoretical values had been very good.

6. Manufacturers’ presentations:

6.1 John Davies (PerkinElmer)

• John started his talk by describing the key features of the Ultra Low Level Liquid Scintillation Spectrometer 1220 Quantulus. These included passive and active rejection of cosmic and other environmental radiation, low noise, low-radioactivity phototubes, cross-talk rejection, EM noise rejection, removal of static electricity, 4 programmable Multi-Channel Analyzers to boost performance, cooling, stabilization, α/β separation, and quench level detection using a low-activity γ-source.
• John described the benefits of a and β-particle measurements using liquid scintillation counting, and the various areas of application.
• John pointed out that the Quantulus software was now able to run under Windows XP. He described the advantages of the new 1220-307 WinQ user interface in Windows and the Easy View Spectrum Analysis Programme.

• Following the talk, a question was raised regarding the commitment of PerkinElmer towards the Wallac counters. John explained that after the acquisition of Wallac and Packard by PerkinElmer, it had been decided to use the Packard range for general scintillation counting and the Wallac Wizard for γ-counting.
• A question was asked about the length of time for which the spares would be guaranteed after a model had been taken off the market. John replied that the normal industry standard was to provide support up to seven years after the last instruments had been produced.
• One delegate asked if detectors other than Photomultiplier Tubes were available for light detection. John replied that he was not aware of any such techniques.

6.2 Ian McCain (Beckman & Coulter)

• Ian started his talk by describing the concepts behind the LS6500 counter. He explained in detail the various steps necessary to achieve them, namely, electronic and hardware design, sample handling facilities, instrument set-up and editing, and data storage/PC connection.
• Ian explained what the Horrocks’ number is and how it is used to monitor quenching in a precise, reliable and reproducible way. The following features of the LS6500 were mentioned: DPM determination, colour detection and correction, two-phase monitoring, and the ‘Radioactive waste manager’.
• Ian pointed out that the LS 6500 was a general counter with some low level capability.
• The calculation of the Minimum Detectable Activity (MDA) on the LS6500 was described. A typical MDA for 3H with optimised windows was approximately 3.3 Bq L^-1.
• Ian finished his talk by showing the a/b discrimination capability of the LS6500.

• Following the talk, one delegate asked how the background correction was applied to quenched and unquenched samples. Ian replied that the background was counted on an open window and that it was then adjusted as a function of the sample window and the quench level.

6.3 Phil Warwick (Raddec Ltd)

• Phil started his talk by describing the features of the Raddec Pyrolyser. He pointed out that the Pyrolyser allowed the simultaneous decomposition of up to 6 samples.
• Phil then explained how the Raddec Pyrolyser works and the various sample types it covers. Examples include soil, water, grass, milk, wood, concrete, brick, metal, plastic, and graphite.
• Typical counting efficiency for ³H is about 18 to 20% and the furnace recovery is approximately 95% ± 5%, which gives a detection limit for ³H of 0.020 Bq g^-1.
• Phil finished his talk by describing LSC-Plus, this is “home made” software that takes the raw data from the LS counter and calculates the activities of the samples directly.

• Following the talk, a delegate asked if it was possible that ³H might stick to the platinum catalyst. Phil replied that they had carried out some tests but that no evidence of such a problem has been found.
• A question was asked regarding testing of the furnace with thymidine - was any residue observed on the platinum? Phil explained that an experiment had been done to determine the ³H evolution profile and that after a long run (4 hours) the ³H counts return to background levels. George Ham commented that he had noticed a small amount of ‘tailing’ when running a blank just after a tritiated thymidine sample, but that this disappeared when compressed air was passed through the pyrolyser overnight.
• A question was raised regarding how often the catalyst needed changing. Phil replied that the catalyst was changed every 20 runs.
• A question regarding the lifetime of the tubes was asked. Phil answered that the lifetime depended on the sample types which had been run and that some of his tubes had been running for years.

7. Any Other Business

• Arzu Arinc presented a brief talk on the non-linearity observed when measuring low-energy emitters with a normal DBB setting using a TriCarb 2700TR counter. A step function is observed at approximately 2000 counts per second; for example, the count-rate falls by 7% for ^99mTc. Arzu pointed out that this could be a major issue if the quench curve was made with points above 2000 counts per second when the measured samples were below this level.

Forthcoming events:

• NPL will be holding the Fourth Nuclear Spectrometry Users’ Forum (NSUF) on 23^rd May 2006 (http://www.npl.co.uk/nsuf/). Please contact Arzu Arinc for further information.
• NPL will be holding the next meeting of the Ionising Radiation Metrology Forum (IRMF) on 17^th May 2006 (http://www.npl.co.uk/irmf/). Please contact Dagmara Tyler for further information.

8. Date of Next Meeting

This is provisionally scheduled for Tuesday, 28^th November 2006.

Arzu Arinc

Secretary, LSUF