Attachment_IV-_Tables_from_QSM.pdf

DoD/DOE QSM, July 2013 Appendix A, Page 148

Table – 16. Alpha Spectrometry

QC Check Minimum Frequency Acceptance Criteria Corrective Action Flagging Criteria Comments

Initial Calibration

(ICAL)

(Energy, efficiency and FWHM peak resolution)

Prior to initial use, following repair or loss of control and upon incorporation of new or changed instrument settings. (MARLAP 18.5.6.3)

Verify manufacturer’s specifications for point source efficiency

(MARLAP);

and

Two calibration peaks that are: 1) ≥700 keV apart; or 2) that bracket all peaks to be determined.

Energy vs. channel slope equation <15 keV per channel.

Full Width –Half Maximum (FWHM) <100 keV for each peak used for calibration.

Minimum of 3,000 net counts in each peak.

Correct problem, then repeat ICAL.

Flagging criteria are not appropriate.

Use traceable calibration source (CS) that matches sample test source (STS) configuration (type, size and position relative to the detector).

May use same count for initial efficiency calibration.

No samples may be run until energy and FWHM calibration criteria are met.

DoD/DOE QSM, July 2013 Appendix A, Page 149

Initial Calibration Verification (ICV)

After initial calibration.

Determine peak location, resolution, and ROI/alpha peak efficiency (where counting efficiency is an analytical requirement) using at least two alpha peaks (MARLAP 18.5.6.3).

or Observed peak centroid falls within ±20 keV from reference energy for each peak used in the initial energy calibration.

FWHM ≤100 keV and within ±20 keV of corresponding calibration peaks in initial energy calibration.

Repeat ICV to check for error.

If that fails, identify and correct problem and repeat ICV or ICAL and ICV, as appropriate.

Flagging criteria are not appropriate.

Use a second-source standard that matches STS configuration (type, size and position relative to the detector) or pulsar for energy check only.

Bracketing peaks may also be used that are >1000 keV apart.

No samples may be run until calibration has been verified with a second source.

Continuing Calibration Verification (CCV) (Pulsar check)

Pulsar energy verification weekly, prior to analysis of samples.

Use either Pulsar check or Check source.

Energy response check shall have a tolerance limit set at ± 3% or control chart set at ± 3σ

(MARLAP 18.5.6.3).

or Observed peak centroid falls ≤20 keV from reference energy.

Recount and check control chart for trends.

Determine cause, correct problem, and repeat CCV and all associated samples since last successful CCV.

Flagging criteria are not appropriate.

Pulsar check can be used to verify energy calibration when using radiotracers during analysis.

No samples may be run until calibration has been verified.

DoD/DOE QSM, July 2013 Appendix A, Page 150

Continuing Calibration Verification (CCV) (Check source)

Weekly source check verification prior to analysis of samples.

Use either Pulsar check or Check source.

Response checks shall have a tolerance limit or control chart set at ± 3% or 3σ. (MARLAP 18.5.6.3) or Observed peak centroid falls within 20 keV from reference energy for each peak used in the initial energy calibration.

FWHM ≤100 keV and within 30 keV of corresponding calibration peaks in initial energy calibration.

Recount and check control chart for trends.

Determine cause, correct problem, and repeat CCV and all associated samples since last successful CCV.

Flagging criteria are not appropriate.

Source check can be used to verify energy, FWHM and efficiency.

No samples may be run until calibration has been verified.

Background Subtraction Count (BSC) Measurement

Prior to initial use or after initial calibration and monthly. (MARLAP 18.5.6.3)

Within ±3σ of mean activity of recent BSCs for total ROI for all isotopes of interest (minimum of 3 BSC values).

Check control chart for trends and recount.

Determine cause, correct problem, re-establish

BSC.

If background activity has changed, re-establish BSC and reanalyze all impacted samples since last acceptable BSC.

If reanalysis cannot be performed, apply B-flag where count rate <10 times that in the affected ROI(s) in the

BSC.

BSC test source matches STS configuration (type, size and position relative to the detector).

Activity must meet project objectives.

DoD/DOE QSM, July 2013 Appendix A, Page 151

Instrument Contamination Check

(ICC)

Performed weekly, at minimum, and after counting high activity samples.

Count duration ≥ longest STS count.

ZBlank ≤3 for blank subtracted (net) activity in all ROIs. (MARLAP 18.4.1)

Check control chart for trends and recount.

Determine cause and correct problem.

If background activity has changed, re-establish BSC and reanalyze all infected samples.

If reanalysis cannot be performed, apply Q-flag to all affected result since last acceptable ICC where the STS count rate in the impacted ROI is ≤5 times that of STS.

Explain in the case narrative.

Method Blank (MB) One per preparatory batch. (MARLAP 18.4.1)

|ZBlank |≤ 3. Investigate recurrent results with |ZBlank| ≥ 2. (MARLAP 18.4.1) or In-house control limits of ±3 σ of the mean.

Recount the blank to confirm results. Inspect MB control chart for indication of significant bias If required, reprep and reanalyze method blank and all samples processed with the contaminated blank.

If reanalysis cannot be performed, data must be qualified and explained in the case narrative.

Apply B-flag to all results for the specific analyte(s) in all samples in the associated preparatory batch.

Results may not be reported without a valid method blank. Flagging is only appropriate in cases where the samples cannot be reanalyzed.

Blank matrices must be the same as the associated samples (i.e.

radon free distilled or deionized water, representative solid material, physically and chemically identical filter media.

With project approval and appropriate qualification and narration, report results with a count rate >5 times that of the affected ROI in the MB.

DoD/DOE QSM, July 2013 Appendix A, Page 152

Laboratory Control Sample (LCS)

One per preparatory batch.

|ZLCS |≤ 3. Investigate

|ZLCS|≥ 2. (MARLAP

18.4.3) or Use in-house control limits of LCS ± 3 σ of the mean. In-house control limits may not fall more than 25% from the known LCS value.

Recount the LCS to confirm results. Inspect LCS control chart for indication of significant bias.

Reprep and reanalyze the LCS and all associated samples.

If reanalysis cannot be performed, data must be qualified and explained in the case narrative.

Apply Q-flag to specific nuclide(s) in all samples in the associated preparatory batch.

Results may not be reported without a valid LCS. Qualification is only appropriate in cases where the samples cannot be reanalyzed.

LCS matrices must be the same as the associated samples.

LCS must be counted for a sufficient time to meet the required project minimum activity.

Acceptance criteria for LCS recovery may be specified by the project.

Matrix Spike (MS) One per preparatory batch.

(MS not required when chemical yield tracers or carriers are employed).

If activity of the MS > 5 times the unspiked sample, then |ZMS |≤ 3.

(MARLAP 18.4.3)

or Within 60-140% recovery.

Examine the project-specific requirements.

Contact the client as to additional measures to be taken.

For the specific nuclide(s) in the parent sample, apply J-flag if acceptance criteria are not met.

The data shall be evaluated to determine the source of difference.

DoD/DOE QSM, July 2013 Appendix A, Page 153

Sample Duplicate One per preparatory batch per matrix.

|ZDup | ≤ 3. Investigate recurrent results with |ZDup|≥ 2. (MARLAP 18.4.1) or The duplicate error ratio (DER) between the sample and the duplicate is <3; or the relative percent difference (RPD) is <25%.

Check for lab error.

Examine the project-specific requirements.

Contact the client as to additional measures to be taken.

For the specific nuclide(s) in the parent sample, apply J-flag if acceptance criteria are not met.

The data shall be evaluated to determine the source of difference.

Tracers (if used) Added to each sample as isotopic yield monitor.

Isotopic yield within 30- 110%.

FWHM <100 keV and peak energy within ±40 keV of known peak energy.

Reanalysis of sample, including sample preparation.

For the specific nuclide(s) in the parent sample, apply J-flag if acceptance criteria are not met.

The data shall be the source of difference.

Flagging is only appropriate in cases where the samples cannot be reanalyzed.

Carriers (if used) Added to each sample as chemical yield monitor.

Chemical yield within 30-110%.

Reanalysis of sample, including sample preparation.

For the specific nuclide(s) in the parent sample, apply J-flag if acceptance criteria are not met.

The data shall be evaluated to determine the source of difference.

Flagging is only appropriate in cases where the samples cannot be reanalyzed.

DoD/DOE QSM, July 2013 Appendix A, Page 154

Table -17. Gamma Spectrometry

Initial Calibration

(ICAL)

(Energy, efficiency and FWHM peak resolution)

Prior to initial use, following repair or loss of control and upon incorporation of new or changed instrument

18.5.6.2) specifications for gamma peak resolution.

(MARLAP 18.5.6.2)

Efficiency vs. energy for each geometry/matrix.

95% confidence limit of the fitted function: ≤8% over energy range.

(MARLAP 18.5.6.2)

or Peak energy difference is within 0.1 keV of reference energy for all points.

Peak Full Width at Half Maximum (FWHM) <

2.5 keV at 1332 keV.

Energy vs. channel slope equation shall be linear and accurate to

0.5 keV.

Correct problem, then repeat ICAL.

Flagging criteria are not appropriate.

Traceable calibration source (CS) matches sample test source (STS) configuration (type, size, geometry and position relative to the detector).

Minimum of 10,000 net counts in each peak in at least six calibration peaks that bracket the range of use.

No samples may be run until all calibration criteria are met.

DoD/DOE QSM, July 2013 Appendix A, Page 155

Initial Calibration Verification (ICV)

After ICAL for energy/efficiency and prior to analysis of samples.

Observed peaks of second source standard fall within ± 10% of initial calibration value relative to energy, FWHM, and efficiency.

Verify second source standard and repeat ICV to check for errors.

If that fails, identify and correct problem and repeat ICV or ICAL and ICV as appropriate.

Flagging criteria are not appropriate.

Traceable second-source standard matches STS configuration (type, size, geometry and position relative to the detector).

Minimum of 10,000 net counts in each peak in at least six calibration verification peaks that bracket the range of use.

No samples may be run

DoD/DOE QSM, July 2013 Appendix A, Page 156

Continuing Calibration Verification (CCV) (Daily Check)

Daily or prior to use.

When working with long count times or batch sequences that run more than a day, CCV is performed at the beginning and end of each analytical batch as long as it not longer than a week.

Verify peak shift within tolerance limit; verify efficiency within control parameters; verify resolution in tolerance limit. Response checks shall have a tolerance limit or control chart set at ± 3% or 3σ of the mean. (MARLAP 18.5.6.2) or Peak Energy/Efficiency:

low, mid, and high energies within 10% of the initial calibration value;

FWHM: low, mid, and high energies within 10% of initial FWHM value.

Correct problem, rerun CCV. If that fails, then repeat ICAL.

Reanalyze all samples since the last successful calibration verification.

If reanalysis cannot be performed, data must be qualified and explained in the case narrative.

Apply Q-flag to all results for the specific nuclide(s) in all samples since the last acceptable calibration verification.

Results may not be reported without a valid

CCV.

Flagging is only appropriate in cases

DoD/DOE QSM, July 2013 Appendix A, Page 157

Background Subtraction Count Measurement (BSC) (Long count for subtracting background from blanks or test sources)

Immediately after ICAL and then performed on at least a monthly basis.

(MARLAP 18.5.6.2)

Statistical test of successive counts and count rates for identified background peaks show no significant difference.

(MARLAP 18.5.6.2)

Recount and check control chart for trends.

Determine cause, correct problem, re-establish

BSC.

If background activity has changed, re-establish BSC and reanalyze or qualify all impacted samples since last acceptable BSC.

Apply B-flag to all results for specific nuclide(s) in all samples associated with the blank.

A detector’s background should be determined immediately after calibration, with or without a counting container, depending on the inherent radionuclide activity levels in the counting container.

The counting interval for the long count shall be between one and four times the nominal counting interval of the test sources.

(MARLAP 18.5.6.2)

Instrument Contamination Check

(ICC)

(Short count for controlling gross contamination)

Daily or when working with long count times before and after each analytical batch.

Check after counting high activity samples.

No extraneous peaks identified (i.e., no new peaks in the short background spectrum compared to previous spectra); The tolerance limit or control chart: ± 3% or 3σ of the mean activity.

Recount the background.

If still out of control, locate and correct problem; reanalyze or qualify all impacted samples since last acceptable ICC.

If background activity has changed, re-establish BSC and reanalyze samples.

If corrective action fails, apply Q-flag to all results for specific nuclide(s) in all samples associated with the

BSC.

Integrate spectrum from ~50 - 2,000 keV to check for gross contamination.

DoD/DOE QSM, July 2013 Appendix A, Page 158

Method Blank (MB)

|ZBlank |≤3 for blank in all ROIs. (MARLAP 18.4.1) or No analytes detected > 2 times the blank Combined Standard Uncertainty (CSU). Blank result must not otherwise affect sample results.

Recount the blank to confirm results, unless all sample results are >5 times the blank activity.

Inspect MB control chart for indication of significant bias.

If required, reprep and reanalyze method blank and all samples processed with the contaminated blank.

If reanalysis cannot be performed, data must be qualified and explained in the case narrative. Apply B-flag to all results for the specific analyte(s) in all samples in the associated preparatory batch.

The results of method blanks typically are not used to correct sample activities, but only to monitor for contamination.

(MARLAP 18.4.1)

Blank matrices must be the same as the associated samples (i.e., radon free distilled or deionized water, representative solid material, physically and chemically identical filter media.)

Results may not be reported without a valid method blank. Flagging is only appropriate in cases where the samples cannot be reanalyzed.

DoD/DOE QSM, July 2013 Appendix A, Page 159

Laboratory Control Sample (LCS)

|ZLCS |≤ 3 . Investigate

|ZLCS|≥ 2. (MARLAP

18.4.3) or Use in-house control chart limits of ± 3σ of the mean. In-house control limits may not fall more than 25% from the known LCS value.

Acceptance criteria for LCS recovery may be specified by the project.

Recount the LCS to confirm results. Inspect LCS control chart for indication of significant bias.

If required, reprep and reanalyze the LCS and all associated samples.

If reanalysis cannot be performed, data must be qualified and explained in the case narrative.

Apply Q-flag to specific nuclide(s) in all samples in the associated preparatory batch.

LCS matrices must be the same as the associated samples and shall contain nuclides within the energy ranges of all those nuclides to be reported.

LCS must be counted for a sufficient time to meet the required project minimum activity.

Results may not be reported without a valid LCS. Qualification is only appropriate in cases where the samples cannot be reanalyzed.

Sample Duplicate One per preparatory batch per matrix.

|ZDup | ≤ 3. Investigate recurrent results with |ZDup|≥ 2. (MARLAP 18.4.1) or The duplicate error ratio (DER) between the sample and the duplicate is <3; or the relative percent difference (RPD) is <25%.

Check for lab error.

Examine the project-specific requirements.

Contact the client as to additional measures to be taken.

For the specific nuclide(s) in the parent sample, apply J-flag if acceptance criteria are not met.

The data shall be

DoD/DOE QSM, July 2013 Appendix A, Page 160

Table – 18. Gas Flow Proportional Counting

Initial Calibration - Voltage Plateau

(ICALV)

(separate plateaus determined for alpha and beta activity)

Prior to initial use and after loss of control.

(MARLAP 18.5.6.1)

specifications.

(MARLAP 18.5.6.1)

Plot voltage vs. count rate to determine proper operating voltages.

(MARLAP 18.5.6.1)

or Slope of the plateau less than 5% over a range of 100V.

Correct problem, then repeat ICALV.

Flagging criteria are not appropriate.

Series of 1-minute counts in <50V steps from ~300V to ~1500V.

No samples may be run until plateau calibration criteria are met.

Initial Calibration - Efficiency (ICALE)

Prior to initial use, after loss of control, and upon incorporation of new or changed instrument

18.5.6.1) detector efficiency for both alpha and beta counting modes using electroplated sources.

(MARLAP 18.5.6.1)

A 1σ counting uncertainty of ≤1% shall be achieved for all detector efficiency determinations.

(MARLAP 18.5.6.1)

Correct problem, then repeat ICALE.

Flagging criteria are not appropriate.

Detector’s counting efficiency, using traceable calibration sources, shall be determined for each radionuclide used to analyze test sources.

No samples may be run until efficiency

DoD/DOE QSM, July 2013 Appendix A, Page 161

Initial Calibration – Cross-talk Factors

(ICALCT)

Prior to initial use, after loss of control, and upon incorporation of new or changed instrument specifications for cross talk in alpha and beta channels. (MARLAP

Correct problem, then repeat ICALCT.

Flagging criteria are not appropriate.

Determine crosstalk factors for each nuclide, matrix and method.

For mass loaded test sources, determine crosstalk factors for the nuclide as a function of test source mass.

No samples may be run until cross talk calibration criteria are met.

Initial Calibration – Self-Absorption Curve

(ICALSA)

Prior to initial use, after loss of control, and upon incorporation of new or changed instrument

For each radionuclide of interest, establish mathematical function (curve) of detector efficiency vs. source mass loading. 95% confidence limit of the fitted function (curve) over the calibration range to ≤10% and ≤5% uncertainty for alpha and beta, respectively.

(MARLAP 18.5.6.1)

or Best fit of data with correlation coefficient closest to 1.00 and the smallest standard error.

Correct problem, then repeat ICALSA.

Flagging criteria are not appropriate.

Minimum of seven mass attenuated standards.

No samples may be run until mass attenuation

DoD/DOE QSM, July 2013 Appendix A, Page 162

Efficiency Calibration Verification (IECV)

After ICALE for alpha and beta and prior to analysis of samples.

A tolerance limit or control chart shall be established immediately after the initial counting efficiency calibration, and after instrument loss of control. A tolerance limit or control chart shall be set at ± 3% or 3σ of the mean.

(MARLAP 18.5.6.1)

or Value of second source calibration for each isotope within ±10% of initial calibration value.

Correct problem and verify second source standard. Rerun IECV.

If that fails, correct problem and repeat

ICALE.

Flagging criteria are not appropriate.

Use traceable second source standard that matches sample test source configuration (type, size, and position relative to the detector).

No samples may be run until calibration has been verified.

Continuing Calibration Verification (CCV)

After a counting gas change and daily for short test-source counting intervals.

For longer test-source counting times, a detector response check for a multi-sample shelf unit shall be conducted prior to test source counting, while a detector response check for a sequential sample counter shall be performed before and after the sample batch.

(MARLAP 18.5.6.1)

Within tolerance or control chart limits ± 3% or 3σ of the mean.

Correct problem, rerun calibration verification.

If that fails, then repeat ICALE. Reanalyze all samples since the last successful calibration verification.

If reanalysis cannot be performed, data must be qualified and explained in the case narrative.

Apply Q-flag to all results for the specific nuclide(s) in all samples since the last acceptable calibration verification.

Minimum of 2,000 net counts for each energy level.

Results may not be reported without a valid

CCV.

Flagging is only appropriate in cases

DoD/DOE QSM, July 2013 Appendix A, Page 163

|ZBlank |≤3 for blank in all ROIs. (MARLAP 18.4.1) or No analytes detected > 2 times the blank Combined Standard Uncertainty (CSU). Blank result must not otherwise affect sample results.

Recount the blank to confirm results, unless all sample results are >5 times the blank activity.

Inspect MB control chart for indication of significant bias.

If required, reprep and reanalyze method blank and all samples processed with the contaminated blank.

If reanalysis cannot be performed, data must be qualified and explained in the case narrative.

Apply B-flag to all results for the specific analyte(s) in all samples in the associated preparatory batch.

Results may not be reported without a valid method blank.

Flagging is only appropriate in cases where the samples cannot be reanalyzed.

Blank matrices must be the same as the associated samples (i.e.

radon free distilled or deionized water, representative solid material, physically and chemically identical filter media.

Background Subtraction Count (BSC) Measurement (Long count for subtracting background from blanks or test sources)

Performed at least on a monthly basis.

Determine alpha and beta background initially and after efficiency calibration. (MARLAP

Use a statistical test to determine a change in the background count rate value. (MARLAP 18.5.6.4) or Within ±3σ of mean activity of recent BSCs (minimum of 3 BSCs).

Check control chart for trend and recount.

Determine cause and correct problem.

If background activity has changed, re-establish

BSC.

All samples following the last acceptable background measurement must be reanalyzed.

If reanalysis of samples is not possible, apply B-flag to all results in all samples associated with the failed blank.

Detector background measured using a contamination-free source mount.

Activity must meet project requirements.

DoD/DOE QSM, July 2013 Appendix A, Page 164

Instrument Contamination Check

(ICC)

(Short count for controlling gross contamination)

Daily or when working with long count times, before and after each analytical batch.

Check after counting high activity samples.

the background count rate value. (MARLAP 18.5.6.4) or Within ±3σ of mean activity of recent BSCs (minimum of 3 BSCs).

Recount the background.

If still out of control, locate and correct problem; reanalyze or qualify all impacted samples since last acceptable ICC.

If background activity has changed, re-establish BSC and reanalyze samples.

If corrective action fails, apply Q-flag to all results for specific nuclide(s) in all samples associated with the

BSC.

Develop detector response control chart immediately after calibration and loss of control.

Laboratory Control Sample (LCS)

|ZLCS|≥ 2. (MARLAP

18.4.3) or Use in-house control chart limits of ± 3σ of the mean. In-house control limits may not fall more than 25% from the known LCS value.

Acceptance criteria for LCS recovery may be specified by the project.

Recount the LCS to confirm results. Inspect LCS control chart for indication of significant bias.

If required, reprep and reanalyze the LCS and all associated samples.

If reanalysis cannot be performed, data must be qualified and explained in the case narrative.

Apply Q-flag to specific nuclide(s) in all samples in the associated preparatory batch.

LCS matrices must be the same as the associated samples and shall contain nuclides within the energy ranges of all those nuclides to be reported.

LCS must be counted for a sufficient time to meet the required project minimum activity.

Results may not be reported without a valid

LCS.

Flagging is only appropriate in cases

DoD/DOE QSM, July 2013 Appendix A, Page 165

Matrix Spike (MS) One per preparatory batch.

(MS not required when yield tracers are employed)

If activity of the MS > 5 times the unspiked sample, |ZMS |≤ 3.

(MARLAP 18.4.3)

or Within 60-140% recovery.

Examine the project-specific requirements.

Contact the client as to additional measures to be taken.

For the specific nuclide(s) in the parent sample, apply J-flag if acceptance criteria are not met.

The data shall be evaluated to determine the source of difference.

Sample Duplicate One per preparatory batch per matrix.

|ZDup | ≤ 3. Investigate recurrent results with |ZDup|≥ 2. (MARLAP 18.4.1) or The duplicate error ratio (DER) between the sample and the duplicate is <3; or the relative percent difference (RPD) is <25%.

Check for lab error.

Examine the project-specific requirements.

Contact the client as to additional measures to be taken.

For the specific nuclide(s) in the parent sample, apply J-flag if acceptance criteria are not met.

The data shall be evaluated to determine the source of difference.

Tracers (if used) Added to each sample. Recovery (isotopic yield) within 30-110%.

Reanalysis of sample, including sample preparation.

For the specific nuclide(s) in the parent sample, apply J-flag if acceptance criteria are not met.

The data shall be evaluated to determine the source of difference.

Flagging is only appropriate in cases

DoD/DOE QSM, July 2013 Appendix A, Page 166

Carriers (if used) Added to each sample as chemical yield monitor.

Chemical yield within 30-110%.

Reanalysis of sample, including sample preparation.

For the specific nuclide(s) in the parent sample, apply J-flag if acceptance criteria are not met.

The data shall be evaluated to determine the source of difference.

Flagging is only appropriate in cases

DoD/DOE QSM, July 2013 Appendix A, Page 167

Table - 19. Liquid Scintillation Counter Analysis

Initial Calibration

(ICAL)

(Efficiency, ROI)

Prior to initial use, following repair or loss of control and upon incorporation of new or changed instrument

18.5.6.4) counting efficiency.

(MARLAP 18.5.6.4)

Establish energy ROIs for nuclides of interest.

(MARLAP 18.5.6.4)

repeat ICAL.

Flagging criteria are not appropriate.

Use appropriate reference radionuclide sources, typically unquenched LS cocktails tagged with 3H and/or 14C.

No samples may be run until efficiency calibration criteria are met and energy ROIs are established for radionuclides of interest.

Method Calibration

(QCAL)

(Quench curve)

Prior to method application, matrix, and cocktail changes or if control of system cannot be re-established or demonstrated.

(MARLAP 18.5.6.4)

A mathematical function and quench curve shall be developed so that the 95 percent confidence limit of the function is ≤5% over the expected quench range of the sources.

Individual calibration sources shall be counted to achieve ROI measurement uncertainty of ≤1%.

(MARLAP 18.5.6.4)

or Minimum 10,000 counts for each data point.

Correlation coefficient for quench curve is > 0.995.

Flagging criteria are not appropriate.

When establishing a quench curve, a minimum of five calibration sources of different quench factors shall be used.

No samples may be run until calibration criteria are passed.

DoD/DOE QSM, July 2013 Appendix A, Page 168

Method Calibration

(QCAL)

(Standard Addition)

Once after each ICAL. Statistically evaluate replicate test-source analyses. (MARLAP 18.5.6.4) or The duplicate error ratio (DER) between the sample and the duplicate is <3; or the relative percent difference (RPD) is <25%.

Flagging criteria are not appropriate.

Add a spike to a duplicate processed sample or add a spike to a sample that has been counted and then recount.

No samples may be run until calibration (Standard Addition) has been verified.

Initial Calibration Verification (ICV)

Once after each ICAL.

Value of each second source nuclide ± 10% of initial calibration value.

Correct problem and verify second source standard. Rerun ICV.

If that fails, correct problem and repeat

ICAL.

Flagging criteria are not appropriate.

Use a second source standard for each nuclide.

No samples may be run

DoD/DOE QSM, July 2013 Appendix A, Page 169

Continuing Calibration Verification (CCV)

Counting efficiency performance check performed on day-of-use basis.

Prior to use for short counting intervals.

Before and after a test source batch for longer counting intervals.

(MARLAP 18.5.6.4)

For batch sequences that run more than a day, performance check is performed at the beginning and end of the batch, as long as it is not longer than a week.

Response checks should have a tolerance limit or control chart set at ± 3% or 3σ of the mean. (MARLAP

Correct problem, rerun calibration verification. If that fails, then repeat

ICAL.

Reanalyze all samples since the last successful calibration verification.

If reanalysis cannot be performed, data must be qualified and explained in the case narrative.

Apply Q-flag to all results for the specific nuclide(s) in all samples since the last acceptable calibration verification.

ROI for unquenched reference standards (typically 3H and/or 14C).

Results may not be reported without a valid

CCV.

Flagging is only appropriate in cases where the samples cannot be reanalyzed.

|ZBlank |≤ 3. Investigate

|ZBlank| ≥ 2. (MARLAP 18.4.1) or In-house control limits of ±3 σ of the mean.

With project approval and appropriate qualification and narration, report results with a count rate >5 times that of the affected ROI in the MB.

Recount the blank to confirm results, unless all sample results are >5 times the blank activity.

If required, reprep and reanalyze method blank and all samples processed with the contaminated blank.

All samples following the last acceptable background measurement must be reanalyzed.

If reanalysis cannot be performed, data must be qualified and explained in the case narrative.

Apply B-flag to all results for the specific analyte(s) in all samples in the associated preparatory batch.

Results may not be reported without a valid method blank.

Flagging is only appropriate in cases where the samples cannot be reanalyzed.

Blank matrices must be the same as the associated samples (i.e.

radon free distilled or deionized water, or representative of the material media.

DoD/DOE QSM, July 2013 Appendix A, Page 170

Background Subtraction Count (BSC) Measurement (Unquenched blank;

applicable when MSA is used)

Prior to initial use and monthly. (MARLAP the unquenched background ROI count rate value. (MARLAP 18.5.6.4) or Within ±3σ of mean activity of recent BSCs for each ROI to be determined. (minimum of 3 BSCs)

Check control chart for trend and recount.

Determine cause and correct problem.

If background activity has changed, re-establish BSC and reanalyze all impacted samples since last acceptable BSC.

If reanalysis cannot be performed, apply B-flag to all results for specific nuclide(s) in all samples associated with the blank.

Unquenched sealed background vial not used for background subtraction.

Activity must meet project objectives.

Method Background Measurement (MBM) (Quenched blank)

Each batch. (MARLAP the quenched background ROI count rate value. (MARLAP 18.5.6.4) or Within ±3σ of mean activity of recent MBMs for each ROI to be determined. (minimum of 3 MBMs)

Check control chart for trends and recount.

Determine cause, correct problem.

If background activity has changed, re-establish MBM and reanalyze all impacted samples since last acceptable MBM.

If reanalysis cannot be performed, apply B-flag where count rate <10 times that in the affected ROI(s) in the

MBM.

MBM test source matches STS configuration (type, size and position relative to the detector).

DoD/DOE QSM, July 2013 Appendix A, Page 171

Laboratory Control Sample (LCS)

|ZLCS|≥ 2. (MARLAP

18.4.3) or Use in-house control chart limits of ± 3σ of the mean. In-house control limits may not fall more than 25% from the known LCS value.

Acceptance criteria for LCS recovery may be specified by the project.

Recount the LCS to confirm results. Inspect LCS control chart for indication of significant bias.

Reprep and reanalyze method LCS and all associated samples.

If reanalysis cannot be performed, data must be qualified and explained in the case narrative.

Apply Q-flag to specific nuclide(s) in all samples in the associated preparatory batch.

LCS matrices must be the same as the associated samples and shall contain nuclides within the energy ranges of all those nuclides to be reported.

LCS must be counted for a sufficient time to meet the required project minimum activity.

Results may not be reported without a valid

LCS.

Flagging is only appropriate in cases where the samples cannot be reanalyzed.

Matrix Spike (MS) One per preparatory batch.

(MS not required when yield tracers or carriers are employed)

If activity of the MS > 5 times the unspiked sample, |ZMS |≤ 3.

(MARLAP 18.4.3)

or Within 60-140% recovery.

Examine the project-specific requirements.

Contact the client as to additional measures to be taken.

For the specific nuclide(s) in the parent sample, apply J-flag if acceptance criteria are not met.

The data shall be

DoD/DOE QSM, July 2013 Appendix A, Page 172

Sample Duplicate One per preparatory batch per matrix.

|ZDup | ≤ 3. Investigate recurrent results with |ZDup|≥ 2. (MARLAP 18.4.1) or The duplicate error ratio (DER) between the sample and the duplicate is <3; or the relative percent difference (RPD) is <25%.

Examine the project-specific requirements.

Contact the client as to additional measures to be taken.

For the specific nuclide(s) in the parent sample, apply J-flag if acceptance criteria are not met.

The data shall be evaluated to determine the source of difference.

Tracers Added to each sample as yield monitor.

Yield within 30-110%.

preparation.

For the specific nuclide(s) in the parent sample, apply J-flag if acceptance criteria are not met.

The data shall be evaluated to determine the source of difference.

Flagging is only appropriate in cases where the samples cannot be reanalyzed.

Carriers Added to each sample as yield monitor.

Chemical yield within 30-110%.

preparation.

For the specific nuclide(s) in the parent sample, apply J-flag if acceptance criteria are not met.

The data shall be evaluated to determine the source of difference.

Flagging is only appropriate in cases