vvEPA
United States
Environmental Protection
Agency
Method 903.0, Revision 1.0: Alpha-Emitting Radium
Isotopes in Drinking Water
-------
Questions concerning this document should be addressed to:
Glvnda A. Smith. Ph.D.
U.S. EPA, Office of Ground Water and Drinking Water, Standards and Risk Management Division,
Technical Support Center, 26 W. Martin Luther King Dr., Cincinnati, OH 45268
Phone:(513)569-7652
smith.glvndaffiepa.gov
Office of Water (MS-140)
EPA 815-B-21-002
January 2021
Authors
Glynda A. Smith, Ph.D., U.S. EPA (Cincinnati, OH)
Steven C. Wendelken, Ph.D., U.S. EPA (Cincinnati, OH)
Acknowledgements
Katie Adams, Inorganic Chemistry Technical Lead, US EPA Region 10 Laboratory, Port Orchard,
Washington.
John Griggs, Ph.D., National Analytical Radiation Environmental Laboratory (NAREL), Office of Radiation
and Indoor Air (ORIA), USEPA.
Keith McCroan, Ph.D., National Analytical Radiation Environmental Laboratory (NAREL), Office of
Radiation and Indoor Air (ORIA), USEPA.
Bahman Parsa, Ph.D., Environmental and Chemical Laboratory Services, PHEL/PHILEP, New Jersey
Department of Health.
Bob Read, Ph.D., Environmental Chemistry Laboratory, Division of Laboratory Services, Tennessee
Department of Health.
Robert L. Rosson, CSRA, LLC, Alexandria, Virginia.
Bob Shannon, Quality Radioanalytical Support, LLC, Grand Marais, Minnesota.
Larry F. Umbaugh, CSRA, LLC, Alexandria, Virginia.
Disclaimer
Publication of the method, in and of itself, does not establish a requirement, although the use of this
method may be specified by the EPA or a state through independent actions. Terms such as "must" or
"required," as used in this document, refer to procedures that are to be followed to conform with the
method.
i
-------
Contents
1 Scope and Application 1
2 Summary of the Method 1
3 Definitions 2
4 Interferences 3
5 Safety 3
6 Equipment and Supplies 4
7 Reagents and Standards 4
8 Sample Collection, Preservation, and Storage 7
9 Quality Control 8
10 Calibration 11
11 Procedure 12
12 Data Analysis and Calculations 14
13 Pollution Prevention 17
14 References 18
ii
-------
1 Scope and Application
1.1 Background
Method 903.0, Revision 1.0 is a method for the determination of alpha emitting isotopes of radium in
drinking water. When other radium alpha emitters (such as radium-223 or radium-224) are present, this
method will not provide an accurate measurement of the radium-226 content of a sample. But, even
when Ra-223 or Ra-224 are present, this method can be used to screen drinking water samples for Ra-
226.
1.2 Drinking Water Regulatory Requirements
The Code of Federal Regulations (CFR) at 40 CFR 141.66(b) specifies a maximum contaminant level
(MCL) for combined Ra-226 and Ra-228 of 5 pCi/L in finished drinking water. If the sum of Ra-228 (a beta
emitter measured using other techniques) activity and the total radium alpha activity (measured using
this method as an estimate of Ra-226 activity) exceeds 5 pCi/L for a drinking water sample, analysis
using a Ra-226 specific method should be considered prior to concluding that the combined Ra-226 and
Ra-228 activity is unacceptably high. Laboratories using this method as a screening technique for Ra-226
must be able to achieve the required radium-226 detection limit of 1 pCi/L as specified at 40 CFR
141.25(c)(1).
1.3 Sensitivity
The sensitivity of the method is a function of sample size, instrument background, counting efficiency,
chemical yield and counting time.
1.4 Timing
Radium-224 and radium-223 are short-lived isotopes with half-lives of 3.6 days and 11.4 days,
respectively. If the primary interest in using this method is to screen for radium-226, samples should
either be held 2-3 weeks before preparing and counting them (thus, removing the contribution of
radium-224 and radium-223 activity) or, alternately, prepare the samples and then hold them 2-3 weeks
before counting. Holding prepared samples 2-3 weeks will reduce the contribution of Ra-224 and Ra-223
by allowing them to decay while improving detection sensitivity by allowing ingrowth of the alpha-
emitting Ra-226 progeny. If, on the other hand, the intent is to also capture the alpha particle activity
contributed by Ra-224 and Ra-223 in a drinking water sample, sample preparation and counting should
be completed as soon as possible.
2 Summary of the Method
Lead and barium carriers, along with citric acid, are added to a preserved drinking water sample. Sulfuric
acid is added to precipitate radium, barium and lead as sulfates. The precipitate is washed with dilute
sulfuric acid and dissolved in basic ethylenediamine tetraacetic acid (EDTA). Ammonium sulfate is added
followed by dropwise addition of acetic acid to adjust the pH to approximately 4.5. Under these
conditions, radium-barium sulfate precipitates while the lead carrier and other naturally occurring alpha
emitters remain in solution. The final BaS04 precipitate, which includes the carried radium, is alpha
counted to determine the total activity of the alpha-emitting radium isotopes. Counting efficiencies for
radium alpha particles, with barium sulfate carrier present, are determined using a known standard
radium activity and the same amount of barium carrier as used for sample analysis. Since ingrowth of
1
-------
radium-226 progeny can occur before the separated radium is counted, it is necessary to make
appropriate activity corrections.
3 Definitions
3.1 Activity
Rate of nuclear decay occurring in a body of material, equal to the number of nuclear disintegrations per
unit time.
3.2 Batch, Preparation
A set of up to 20 environmental field samples of the same matrix that are prepared and/or analyzed
together with the same instrumentation and personnel, using the same lot(s) of reagents, with a
maximum time between the start of preparation of the first and last sample in the batch being 24 hours.
3.3 Detection Limit (DL)
The DL for radionuclides in drinking water is defined in 40 CFR 141.25(c) as the radionuclide
concentration that can be counted with a precision of plus or minus 100% at the 95% confidence level
(1.96o, where o is the standard deviation of the net counting rate of the sample).
3.4 Duplicate (DUP)
A second aliquot of a field sample that is processed in the same manner as the samples in the
preparation batch. Analysis of the DUP provides a measure of the precision associated with batch
preparation.
3.5 Field Blank
Samples preserved with reagents that are not provided by the laboratory should be accompanied by a
field blank sample that is preserved in the same manner as the submitted samples. The field blank is a
volume of blank matrix that is placed in a clean sample container, preserved in the field, shipped along
with the samples and subjected to the same analytical procedures as the samples. A sample of the
preservative should also accompany the field samples to determine whether it contributes any
contamination.
3.6 Laboratory Fortified Blank (LFB)
The LFB consists of a volume of a blank matrix to which a known activity of a radioisotope has been
added. The LFB is processed in the same manner as the samples in the preparation batch, and its
purpose is to determine whether the methodology is in control, and whether the laboratory is capable
of making accurate measurements.
3.7 Laboratory Reagent Blank (LRB)
The LRB consists of an aliquot of a blank matrix that is processed in the same manner as the samples in
the preparation batch, including exposure to all glassware and equipment that are used in the
preparation batch. The LRB is used to assess the process of handling, preparation and analysis for cross-
contamination and for low-level analytical bias.
3.8 Laboratory Fortified Sample Matrix (LFSM)
An aliquot of a field sample to which a known activity of the radionuclide(s) being measured has been
added. The LFSM is processed in the same manner as the samples in the preparation batch. Its purpose
is to determine whether the sample matrix contributes bias to the results. The native level of the
2
-------
radionuclide(s) must also be determined in an unspiked field sample aliquot in order to correct for levels
already present in a sample that could contribute to the LFSM response.
3.9 Laboratory Fortified Sample Matrix Duplicate (LFSMD)
An additional aliquot of a field sample that has the same quantity of radionuclide(s) added to it as the
LFSM. The LFSMD is processed in the same manner as the samples in the preparation batch. It may be
used in place of the DUP to assess preparation batch precision.
3.10 Picocurie (pCi)
The pCi is the quantity of radioactive material producing 2.22 nuclear disintegrations per minute.
3.11 Uncertainty, Counting
The component of measurement uncertainty attributable to the random nature of radioactive decay
and radiation counting.
3.12 Uncertainty, Standard
An estimate of the measurement uncertainty expressed as one standard deviation.
3.13 Uncertainty, Expanded
An estimate of the uncertainty Uof a measurement result y that provides a high level of confidence that
the interval y ± U includes the actual value of the quantity being measured. The expanded uncertainty is
typically obtained by multiplying the standard uncertainty by a coverage factor. A coverage factor of
1.96 is routinely used for drinking water analysis and the coverage factor (or confidence level) should be
specified on the sample report.
4 Interferences
4.1 Naturally Occurring Barium
The radiochemical yield of the radium activity is based on the chemical yield of the BaS04 precipitate; a
significant natural barium content in a sample may bias chemical yield high resulting in low sample
results.
4.2 Short-lived Radium Isotopes
Radium-224 and radium-223 can contribute to the alpha particle activity. If the interest in this method is
screening for radium-226, interference from these short-lived radium isotopes can be minimized by
delaying analysis for 2-3 weeks after sampling to allow for decay.
5 Safety
The specific toxicity of each reagent used in this method has not been precisely defined. Each chemical
should be treated as a potential health hazard, and exposure to these chemicals should be minimized.
Each laboratory is responsible for maintaining a chemical hygiene plan (CHP) with an awareness of the
appropriate regulations regarding safe handling of chemicals used in this method. A reference file of
safety data sheets (SDSs) should be made available to all personnel involved in the preparation of
samples and their analyses.
3
-------
6 Equipment and Supplies
6.1 Gas-flow Proportional Counter
The detector may be either a windowless (internal proportional counter) or a thin window type. The
system should be capable of accommodating counting planchets and be sufficiently free from
background so that required detection levels can be met within a reasonable counting time.
6.2 Stainless Steel Counting Planchets
A planchet should be flat-bottomed, or with concentric rings, with a raised wall to contain the sample.
NOTE: Always use the same type planchet (to maintain the same geometric configuration) for
calibration and sample determinations.
6.3 Electric Hot Plate
6.4 Drying Oven/Heat Lamp
Drying oven capable of maintaining a temperature of 105 °C + 2 °C. Alternately, heat lamps may be used
if a drying oven is not available.
6.5 Desiccator
6.6 Centrifuge
6.7 Beakers
Beakers of various sizes as appropriate for sample preparation.
6.8 Pipettes
Pipettes of various sizes.
6.9 Analytical Balance
The analytical balance with a readability of 0.1 mg.
6.10 Optional: As Appropriate for Procedural Options
6.10.1 pH Meter
Potentiometer with glass electrode, reference electrode, and temperature compensation capability.
6.10.2 pH Paper
Short range (pH 3.0 - 5.5) and wide range (pH 1 -14).
6.10.3 Membrane Filters and Filter Funnel Assemblies
Vacuum filter funnel assemblies and membrane filters with 0.45 pim pore diameter with size appropriate
for use in gamma spectrometer detectors.
7 Reagents and Standards
Analytical reagent grade or better chemicals should be used. Commercial reagents are often not tested
for trace radioactivity. Therefore, analysts need to carefully monitor their laboratory reagent blank (LRB)
control charts to identify situations where levels of radioactivity may be present in reagents that could
compromise results.
NOTE: Laboratories can adjust reagent and solution volumes as appropriate to meet testing needs
provided the molar ratios are maintained.
4
-------
7.1 DistiIled/Deionized Water
Distilled or deionized water meeting the requirements of ASTM Type 1, 2 or 3 reagent water.
7.2 Acetic Acid
Acetic acid, 17.4 M: glacial CH3COOH (conc.), sp.gr. 1.05, 99.8%.
7.3 Ammonium Sulfate (200 mg/mL)
Prepare by dissolving 20 grams (NhUhSCU in a minimum of water and dilute to 100 mL.
7.4 Barium Carrier (16 mg Ba2+/mL)
Prepare carrier by dissolving about 2.8 g BaCl2*2H20 in water, adding 0.5 mL 16 M HN03, and diluting to
100 mL with distilled/deionized water. Alternately, purchase a high purity commercial barium standard
solution and dilute appropriately. Standardize in triplicate using one of the options described below.
Note: Use the appropriate calculations in Section 12 to determine yield.
7.4.1 Standardize According to Approach used to Determine Yield
7.4.1.1 Standardize Barium for Chemical Yield
Pipette 2.00 mL carrier solution into a centrifuge tube containing 15 mL water. Add 1 mL 9 M H2S04 with
stirring and digest precipitate in a hot (near boiling) water bath for 10 minutes. Cool, centrifuge, and
decant the supernatant. Wash precipitate with 15 mL water. Centrifuge and decant supernatant.
Transfer the precipitate to a tared stainless steel planchet with a minimum of water. Dry in a 105 °C + 2
°C oven for two hours (or alternately, dry under a heat lamp). Cool and store in a desiccator, and weigh
as BaS04. Verify the average mass is within + 5% of the expected level with a relative standard deviation
between the replicates of < 5%. Calculate the barium content:
(m,Ba504)x("^7
-y "I" ™ ™ ! ~~~ T ^
Ba in mg/mL
Vnlmnc Ba2+
Substituting (137.34 g/mole Ba2+/233.404 g/mole BaS04) = 0.5884, and 2.00 mL carrier volume yields:
(mg BaS04) x (0.5884)
Ba ^ in mq mL =
w/ 2.00 mL
NOTE: If barium-133 will be used as a tracer for radiochemical yield analysis, the barium carrier
prepared in Section 7.4 does not need to be separately standardized.
7.4.1.2 Standardize Barium-133 Tracer for Radiochemical Yield
Add barium-133 tracer solution to the prepared barium carrier (Section 7.4) at a level that will yield at
least 6000 - 8000 pCi/100 mL. Pipette 2.00 mL of the barium carrier containing the Ba-133 tracer into a
centrifuge tube containing 15 mL water. Add 1 mL 9 M H2S04 with stirring and digest precipitate in a hot
water bath (near boiling) for 10 minutes. Cool, centrifuge and decant supernatant. Wash precipitate
with 15 mL water. Centrifuge and decant supernatant. Place a 0.45 pim membrane filter in a vacuum
filter assembly, prewet with distilled/deionized water and start vacuum. Transfer the precipitate to the
filter using distilled/deionized water to ensure the transfer is quantitative. Add a small volume of
ethanol to the precipitate and maintain suction to fully dry the precipitate. The filter containing the
precipitate should be mounted and counted in a geometry for which the gamma spectrometer has been
properly calibrated. Ba-133 is counted long enough so about 10,000 counts above background are
accumulated. Verify that the average Ba-133 activity is within + 5% of the expected level with a relative
standard deviation between the replicates of < 5%. This method does not discuss specifications related
5
-------
to analytical requirements for gamma spectrometers; however, laboratories incorporating barium-133
as a tracer will be expected to document energy and efficiency calibration as well as performance checks
related to verifying detection efficiency, energy calibration, background, peak resolution, etc.
Calculation of yield using the tracer is discussed in Section 12.1.2. In addition, the three prepared
standardization replicates should be counted on the gas flow proportional counter to verify there is no
impact of Ba-133 on alpha channel counts.
7.5 Citric Acid (1 M)
Prepare by dissolving 21 g C6H807*H20 in water and dilute to 100 mL.
7.6 EDTA, Basic Reagent (0.25 M)
Prepare by dissolving 20 g NaOH in 750 mL water, heat and slowly add 93 g ethylenediamine tetraacetic
acid disodium salt dihydrate (CioHi4N2Na208*2H20). Heat and stir until dissolved, filter through coarse
filter paper and dilute to 1 L.
7.7 Ethanol (optional)
Ethanol, 95%.
7.8 Lead Carrier (15 mg/mL)
Prepare by dissolving 2.4 g Pb(N03)2 in water, add 0.5 mL 16 N HN03 and dilute to 100 mL with water.
NOTE: Only carriers employed for yield determinations need to be standardized. Lead carrier does not
require standardization.
7.9 Sodium Hydroxide (6 M)
Prepare by dissolving 24 g NaOH in 80 mL water and dilute to 100 mL.
7.10 Sulfuric Acid (9 M)
Cautiously mix 1 volume 18 M H2S04 (conc.) with 1 volume of water.
7.11 Sulfuric Acid (0.05 M)
Mix 1 volume 9 M H2S04 with 179 volumes of water.
7.12 Nitric Acid
Nitric acid, HN03, concentrated (16 M)
7.13 Indicator Solutions (optional)
The following commercially available pH indicator solutions are optional: phenolphthalein indicator
solution and bromocresol green indicator solution
7.14 Calibrant
Radium-226 standard traceable to a national metrology institute (such as the National Institute of
Standards and Technology [N 1ST]).
NOTE: After making a working solution from the standard, verify the concentration by conducting at
least three separate verification measurements to confirm that each individual measurement is within
5% of the expected value.
NOTE: Prepared standards and working solutions should be assigned unique identification numbers
traceable to the original standard and labeled with expiration dates.
6
-------
8 Sample Collection, Preservation, and Storage
8.1 Containers
Collect samples in glass or plastic containers. A sample volume of 1 gallon (approximately 4 liters) is
recommended for drinking water samples, but collection volume is left to the discretion of the
laboratory.
8.2 Sample Collection for Drinking Water
Open the cold water tap and allow the system to flush until the water temperature has stabilized (about
3 to 5 minutes). Collect samples from the flowing system. If the samples are preserved with reagents
(i.e. nitric acid) that are not provided by the laboratory, the samples should be accompanied by a Field
Blank (Section 3.5) that is preserved in the same manner as the samples. A sample of the preservative
used in the field should also accompany the samples and Field Blank to the laboratory to determine the
contribution, if any, from the addition of the preservative.
NOTE: This section describes collection of drinking water samples at the entrance to the distribution
system. Other guidance may apply for collection of water samples for other programs.
8.3 Preservation
Samples should generally be preserved at the time of collection by adding enough nitric acid to the
sample to bring it to a pH < 2 (e.g. 1-2 mL 16 M HN03 per liter of sample will generally yield a pH < 2).
Alternate concentrations of HN03 are permitted, although the volume added to achieve a pH < 2 should
not exceed 30 mL/L. If the samples are preserved in the field using acid that has not been supplied by
the laboratory, add the same level of acid to a Field Blank (Section 3.5). The pH of all samples must be
verified upon receipt in the laboratory. If the pH is > 2, treat the sample as if it were collected without
preservation as described in Section 8.4.
NOTE: Drinking water samples do not require thermal preservation.
8.4 Handling Unpreserved Samples and Storage
If samples are collected without preservation, they must be received by the laboratory within 5 days,
then preserved with 16 M HN03 to a pH < 2 and held in their original containers for a minimum of 16
hours. The 16-hour waiting period helps solubilize finely suspended materials or surface-adsorbed
materials. Verify that the pH is < 2 before preparing the samples for analysis.
NOTE: Screen sample preservatives used by the laboratory (or by field samplers) for radioactive content
by lot number prior to use if possible, to verify that preservatives will not introduce radioactive
contamination.
NOTE: Drinking water compliance samples must be prepared 'as received' (i.e. not filtered). Other
programs/projects may have alternate requirements. If such programs/projects specify filtration, it must
be performed prior to preservation.
NOTE: The requirement for pH < 2 adequately addresses radionuclides of concern for drinking water
compliance samples; however, that may not be adequate for programs evaluating other radionuclides in
non-drinking water matrices. Other programs should optimize sample preservation as appropriate to
address such radionuclides of concern.
NOTE: The laboratory must have appropriate segregation procedures in place to prevent cross
contamination of samples.
7
-------
9 Quality Control
9.1 QC Requirements
QC requirements include an Initial Demonstration of Capability (IDC), ongoing Demonstration of
Capability, and ongoing QC requirements that must be met when preparing and analyzing drinking
water compliance samples. This section describes QC parameters, their required frequencies, and
performance criteria that must be met in order to meet EPA quality objectives for drinking water
analyses. These QC requirements are considered the minimum acceptable QC criteria. Laboratories are
encouraged to institute additional QC practices to meet their specific needs.
9.1.1 Initial Demonstration of Capability
A successful IDC must be performed by each analyst prior to analyzing any field samples. Before
conducting the IDC, set up and calibrate the instrument as described in Section 10. The IDC consists of
the following: demonstration of low system contamination (Section 9.1.1.1). demonstration of accuracy
(Section 9.1.1.2) and confirmation of method sensitivity through a DL study (Section 9.1.1.3).
NOTE: The primary analyst is responsible for conducting a full DL study along with the demonstration of
low system contamination and demonstration of accuracy. Other analysts/technicians that may also be
responsible for performing the method can document their IDCs by either conducting a DL study or by
evaluating four LRBs (Section 9.1.1.1) and four LFBs (Section 9.1.1.2) and successfully meeting the
specified acceptance criteria.
9.1.1.1 Demonstration of Low System Contamination
Contamination due to sample processing is assessed by preparing and counting at least four LRBs
(Section 3.7). For drinking water, the LRB volume should be the same as the volume used in establishing
detection capability (Section 9.1.1.3) and representative of typical drinking water sample volumes. LRBs
are prepared and handled like samples following the procedure described in Section 11. Each LRB result
must be below the radium-226 required detection limit of 1 pCi/L.
9.1.1.2 Demonstration of Accuracy
Initial demonstration of accuracy is verified by preparing and counting at least four LFBs (Section 3.6)
following the procedure in Section 11. The recommended LFB fortification with radium-226 standard is
about 2.5 - 5 pCi/L. The average recovery for the LFBs must be within + 10% of the known amount of
added radium-226 activity.
9.1.1.3 Detection Limit Study
Laboratories testing drinking water samples for Safe Drinking Water Act compliance monitoring need to
confirm detection capability by performing a Detection Limit (DL) study. After calibrating the instrument
as described in Section 10. fortify seven LFBs with the radium-226 standard at activity concentrations
near the 1 pCi/L required detection limit. Prepare and analyze the LFBs following the procedure
described in Section 11. Calculate the DLs as described in Section 12.5.
NOTE: Further discussion of the drinking water DL procedure along with derivation of the final DL
equation from the 40 CFR 141.25(c) definition can be found in Procedure for Safe Drinking Water Act
Program Detection Limits for Radionuclides, USEPA 815-B-17-003, April 2017.
NOTE: Programs that do not analyze drinking water compliance samples may have alternate detection
capability requirements.
8
-------
9.1.1.4 Exception for Experienced Analysts
If an analyst has at least one year of experience preparing and analyzing LRBs and LFBs for a
coprecipitation Ra-226 method (such as a previous version of EPA Method 903.0), and there have been
no changes to the instrumentation, previously documented data may be used to fulfill the IDC low
system contamination and accuracy requirements. Ongoing demonstrations of capability (Section 9.1.2)
will verify analyst conformance to the criteria described in this revised method.
9.1.2 Ongoing Demonstration of Capability
Ongoing demonstrations of capability may be fulfilled by repeating the IDC studies described in Sections
9.1.1.1 - 9.1.1.2 annually or by documenting batch QC LRBs and LFBs that an analyst has processed
during the year since the last demonstration of capability. The data for at least four LRBs processed in
different batches can be used to assess sample processing contamination and the data for at least four
LFBs processed in different batches can be used to assess accuracy. The amount of the Ra-226 standard
added to the sample batch LFBs should follow the guidance described in Section 9.1.1.2.
9.2 Ongoing QC Criteria
This section summarizes ongoing QC criteria that must be followed when processing and analyzing
drinking water compliance samples.
9.2.1 Calibration Stability and Background Checks
The calibration stability and background of each detector used to count analytical samples or QC
samples is checked and recorded on control charts each day prior to use to verify the instrument
response has not changed since it was calibrated. During periods when gas proportional counters are
idle, check the detector calibration stability and the background weekly to confirm the ready status of
the instrument for sample measurements.
9.2.1.1 Calibration Stability Check
To verify detector calibration stability, each day prior to sample counting, run an alpha QC check source.
The check source does not have to be NIST-traceable but must have a documented count rate. Count
the check source long enough to obtain about 10,000 counts (1% counting uncertainty). Record and
monitor measurements on a control chart.
9.2.1.2 Background Check
Each day prior to sample counting, run a background check with only a clean planchet in the detector.
Record and monitor background measurements on a control chart.
9.2.1.3 Post-run Calibration Stability and Background Checks
After completion of a sample counting batch, run the alpha check source and a background check to
verify that the calibration and background did not change significantly while the samples were being
counted.
9.2.2 Control Charts
Analysts are responsible for preparing and maintaining control charts. After a sufficient number of
checks have been obtained (usually at least 20 measurements), calculate the mean values on the control
charts. Establish warning limits at ±2 standard deviations and control limits at ±3 standard deviations
relative to the mean values. Monitor control charts to ensure measurements remain in statistical control
relative to the control limits. Also, monitor control charts to ensure instrument performance does not
change significantly (i.e., drifting or trending of responses) relative to the time of the initial calibration.
9
-------
NOTE: As opposed to manually plotting data and calculating control limits, most instrument software
provides this capability. Analysts still have a responsibility to monitor the control charts as described
above to ensure that measurements remain in statistical control.
9.2.3 Corrective Action
If instrument control measurements exceed their control limits or exhibit a significant change in
performance, the proportional counter is placed out of service until stability of the system relative to
the initial calibration can be demonstrated.
9.2.4 QC Options for Automated Systems
If a gas proportional counter measures samples sequentially with an automatic sample changer, it is
possible that the batch size combined with a lengthy count time will create a sample counting set that
requires several days to complete. In such cases, it is permissible to arrange the batch order so that a
batch QC sample fortified with a known activity (e.g., an LFB, Section 9.2.6 or an LFSM, Section 9.2.7)
and a QC sample that is not expected to have any measurable activity (e.g., an LRB, Section 9.2.5) are
counted at the end of each 24-hour period. The LFB or LFSM would thus substitute as the calibration
stability check source and the LRB would substitute as the instrument background check. If these QC
samples meet their acceptance criteria, then it may be assumed that little or no change in the
calibration or background have occurred while samples were being measured.
9.2.5 Laboratory Reagent Blank (LRB)
An LRB (Section 3.7) must be prepared with each preparation batch to confirm there is no significant
contamination introduced in processing the batch which would contribute bias to the analytical results.
Ensure that LRB activity does not exceed the required DL as described in Section 9.1.1.1. Record LRB
activities on control charts and monitor for trends that could indicate the need for corrective action.
9.2.6 Laboratory Fortified Blank (LFB)
An LFB (Section 3.6) must be prepared with each preparation batch to assess batch accuracy
independent of compliance sample matrix effects. Fortify with radium-226 standard at a level between
the required detection limit and the MCL - it is recommended to keep the fortification level between 2.5
and 5.0 pCi/L since uncertainty at the DL is higher. Accuracy as percent recovery must be within ±10% of
the amount of activity added. If the LFB fails to meet the recovery criterion, the batch is considered
compromised which may be due to contamination, poor precipitation/preparation technique, etc. Re-
prepare the sample batch with new QC checks provided sufficient volume is available. Otherwise, a new
set of samples should be collected. Record and monitor LFB recoveries on control charts.
9.2.7 Laboratory Fortified Sample Matrix (LFSM)
Prepare one LFSM (Section 3.8) per preparation batch. Spike the LFSM with a known activity of radium-
226 standard that is approximately 10 times the anticipated level in the samples or at least 10 times the
DL (i.e. 10 pCi/L). Accuracy as percent recovery must be within ±20% of the amount of activity added. If
the LFSM fails to meet the recovery criterion, recount the samples, if practical or re-prepare the sample
batch with new QC checks provided sufficient volume is available. Otherwise, flag sample results in the
batch as possibly biased low or high (as the LFSM result indicates) due to matrix effects. Record and
monitor LFSM recoveries on control charts.
NOTE: If Ba-133 tracer is incorporated in each sample, a LFSM is not required with each batch. The
tracer yield will indicate if there are matrix interference issues.
10
-------
9.2.8 Duplicate (DUP) or Laboratory Fortified Sample Matrix Duplicate (LFSMD)
Batch precision is assessed through preparation of either a DUP (Section 3.4) or, if non-detects are
frequent, a LFSMD (Section 3.9) with every preparation batch. The LFSMD is spiked at the same level as
the LFSM. Precision is assessed by calculating the relative percent difference (RPD). RPD must be < 20%.
If the RPD exceeds 20%, and a duplicate sample measurement is < 5X the DL, calculate the normalized
absolute difference (NAD). Calculations for the RPD and NAD are provided in Section 12.6. A
sample/DUP or LFSM/LFSMD that fail the batch precision requirement need to be recounted. If the
reanalysis fails, it may be an indication of a lack of sample homogeneity and all sample results in the
preparation batch should be reported with a qualifier indicating the measurement has questionable
precision. If a client requires unqualified results, prepare a new sample batch with new QC checks
provided sufficient volume is available. Otherwise, a new set of samples should be collected. Record and
monitor RPD and NAD on control charts.
9.2.9 Field Blank (if needed)
If a Field Blank (Section 3.5) is provided with the samples, prepare and analyze it to confirm that field-
supplied preservative does not contribute radioactive contamination to the analytical results.
10 Calibration
10.1 Instrument Setup
Establish voltage plateaus and appropriate operating conditions as recommended by the instrument
manufacturer. Perform a calibration stability check and background check as described in Section 9.2.1
following instrument set-up, anytime operating voltage is changed, following instrument repairs, and
after gas bottle changes. If QC checks fail, take corrective action (which may entail re-establishing
instrumental operating conditions and recalibration).
10.2 Detector Background
Detector chamber background levels must be determined to provide for background subtraction in
activity calculations and verification that the instrument is free of contamination. A clean, empty
planchet is counted (for each detector in the counting system) for at least the same length of time that
typical samples are expected to be counted.
10.2.1 Background Subtraction
For drinking water compliance samples, background subtraction measurements should be performed
with each batch of samples; however, a weekly long background count as described in Section 10.2.2 is
also acceptable. If desired, the background subtraction measurement can substitute as either the pre- or
post-background check as described in Section 9.2.1.
NOTE: Other programs that do not analyze drinking water compliance samples may have alternate
background subtraction counting frequencies, as appropriate.
10.2.2 Multiple Detector Systems
There are counting systems that have multiple single detectors and the batch is counted on several
different detectors. For such systems, the laboratory may establish a control chart with a weekly long
background count. On a daily basis, count a clean planchet before and after analyzing a drinking water
compliance batch for a shorter time (at least ten minutes). Plot the short background measurements on
the longer background control chart. If the short background measurements are in statistical control of
11
-------
the long background (i.e., within + 3 standard deviations of the mean long background measurements),
then the weekly long background count can be used for background subtraction.
10.3 Geometry Considerations
Initial calibration involves preparation of a mass-efficiency (alternately, a self-absorption) curve.
Standard radium-226 is carried with a precipitate of barium sulfate as described in the analytical
procedure. The mass of barium sulfate determines the degree of self-absorption which affects the
detection efficiency of alpha particles. The geometry of the calibration sources prepared for the mass-
efficiency curve needs to be the same as that of the prepared sample and QC planchets (i.e., prepare
and mount calibration sources in the same manner as samples).
10.4 Calibration and Development of Mass-Efficiency (Self-Absorption) Curve
Use standardized barium carrier for preparing the efficiency standards. Determine the range of carrier
that can be added based on the allowable range for the yield. In this method, the yield must fall
between 70-110%. Therefore, prepare a set of calibration sources as described in Section 10.5
containing standardized barium carrier at levels bracketing the minimum (1.4 mL barium carrier) and
maximum (2.2 mL barium carrier) allowable yields.
10.5 Preparation of Standards
Acidify a set (at least three) of 1-L samples of distilled/deionized water with a few mL 16 M HN03. Fortify
each sample with approximately 250 dpm radium-226 standard. Follow the procedure described in
Section 11. but add a variable aliquot of barium carrier (1.4 - 2.2 mL) to each water sample. Be sure to
note the time of the final barium sulfate precipitation (Section 11.6) because it is the reference time for
ingrowth of decay products which must be accounted for in the calculations.
10.6 Counting
Count the prepared efficiency standards until at least 10,000 total counts greater than background
(where the background should be less than 1-2% of the total counts above background) have been
accumulated.
10.7 Generate Curve
Prepare the mass-efficiency curve and generate the best curve fit by plotting the efficiency of the
radionuclide standard as calculated in Section 12.3 along the y-axis vs. the measured precipitate mass
along the x-axis. This curve is used to generate efficiencies for sample calculations in Section 12.
10.8 Annual Verification
Annual verification of mass-efficiency (self-absorption) calibration is required. The accuracy of the curve
is verified by preparing and measuring efficiency standards as described in Section 10.5. For the re-
verification measurement to be acceptable, the original measurements of each efficiency standard
should lie within the range defined by the uncertainty of the newly prepared efficiency standard's
measurement calculated at the 95% confidence level. If the curve verification fails, establish a new
calibration curve.
11 Procedure
11.1 Add Carriers
To a 1-L acid-preserved drinking water sample, add 5 mL 1 M citric acid (Section 7.5), 1 mL lead carrier
(15 mg, Section 7.8), and 2.0 mL standardized barium carrier (32 mg, Section 7.4), and heat to boiling.
12
-------
NOTE: If barium-133 is used for yield determination, remember to use the standardized barium carrier
described in Section 7.4.I.2.
11.2 Coprecipitate Radium as Ba/Pb/RaPbSC>4
Cautiously, with vigorous stirring, add 20 mL concentrated (18 M) H2S04. Digest at a low boil for 5 to 10
minutes, turn off the heat, and allow the mixed BaS04-PbS04 precipitate to settle overnight.
11.3 Isolate and Wash Precipitate
Decant and discard the supernatant. Transfer the precipitate to a centrifuge tube with a minimum
amount of 0.05 M H2S04. Centrifuge and discard the supernatant. Wash the precipitate with
approximately 10 mL 0.05 M H2S04. Centrifuge and discard the supernatant. Wash the precipitate again
with approximately 10 mL 0.05 M H2S04. Centrifuge and discard the supernatant.
11.4 Dissolve Precipitate with Basic EDTA Solution
Dissolve the precipitate by adding 15 mL basic EDTA reagent (Section 7.6). Heat in a hot water bath and
add a few drops 6 M NaOH until dissolution is complete.
NOTE: Use of indicators to monitor pH is an acceptable method option, if desired, add a few drops of
phenolphthalein indicator to verify that the solution is alkaline (at pH > 8, the indicator will turn the
solution a deep reddish purple).
11.5 Precipitate Ba/RaSC>4
Add 1 mL (NH4)2S04 (200 mg, Section 7.3) and stir thoroughly. Add 17.4 M CH3COOH dropwise until
precipitation begins. If phenolphthalein was added as an optional pH monitor, when the red/purple
color disappears upon addition of the acetic acid, add a few drops of bromocresol green indicator.
Continue to add acid dropwise to ensure complete precipitation at a pH of 4.5. This step is pH-
dependent and monitoring the pH is essential, either by the use of an indicator or pH measurement.
Acidification releases barium/radium from the EDTA complex, but if the pH drops below about 4.5, the
lead that is bound within the EDTA complex will begin to be released. Digest in a boiling water bath for 5
to 10 minutes.
NOTE: If pH is monitored using the bromocresol green indicator, the appropriate endpoint is green. It is
recommended that analysts become familiar with recognizing the appropriate color of a pH 4.5 solution
by adjusting a solution of similar composition to pH 4.5 using a pH meter/pH paper.
11.6 Isolate and Wash Precipitate
Centrifuge and discard supernatant. Wash the BaS04 precipitate with 15 mL water, centrifuge, and
discard wash. Separation of BaS04 is completed and ingrowth of radon begins. Record the time.
11.7 Prepare Test Source, Determine Yield, and Count
11.7.1 Procedure Based on Gravimetric Chemical Yield Determination
Transfer the precipitate to a tared planchet with a minimum of water. Dry under a heat lamp or in a 105
°C + 2 °C oven until a constant weight is obtained. The geometry of samples, calibration sources, and QC
checks must be consistent (i.e. prepare planchets in the same manner for all sources). Cool in a
desiccator and weigh. The calculated yield (Section 12.1) must be within 70 - 110%. If the yield exceeds
110%, the sample may contain a high level of natural barium causing interference. Evaluation of an
aliquot of the original sample by an atomic absorption-based method (including ICP-AES and ICP-MS)
can be used to verify native barium levels. Count in a gas-flow proportional counter to determine alpha
activity.
13
-------
11.7.2 Procedure Based on Radiochemical Yield using Ba-133 Tracer
Place a 0.45 pirn membrane filter in a vacuum filter assembly, prewet with distilled/deionized water and
start vacuum. Transfer the precipitate to the filter using distilled/deionized water to ensure transfer is
quantitative. Add a small volume of ethanol to the precipitate and maintain suction to fully dry the
precipitate. Place the filter containing the precipitate in a planchet and count in a gas-flow proportional
counter to determine alpha activity of the sample. After counting, place the planchet containing the
precipitate in a gamma spectrometer detector and count to assess Ba-133 yield. The radiochemical yield
must be within 70 - 110% of the standardized barium (Section 7.4.1.2).
12 Data Analysis and Calculations
12.1 Yield Calculations
12.1.1 Chemical Yield (Precipitate)
The chemical yield for the barium carrier is calculated as follows:
(mg BaS04)(0.5884)
Standardized Value of Ba + carrier in mg/mL = ——
2.0 mL
Yield of barium sulfate is determined based on the standardized concentration of the barium carrier
(Section 7.4.1.1). If the standardized carrier concentration is 16 mg/mL, then 100% yield of barium
sulfate would be 54.38 mg.
The fractional yield would thus be:
_ (jns - mp)
54.38
Where
ms= mass of planchet with the dried barium sulfate precipitate, mg
mp= mass of planchet, mg
54.38 = mass of barium sulfate precipitate if all the added barium carrier (32 mg) is recovered. If the
standardized concentration of the barium carrier differs from the 16 mg/mL prepared in Section 7.4,
adjust the theoretical Ba2+ concentration accordingly.
12.1.2 Radiochemical Yield (Ba-133 Tracer)
Alternately to the chemical yield calculated in Section 12.1.1. radiochemical yield based on addition of
barium-133 as a tracer is calculated using the following equation:
Y = —
As
Where
Am= Activity of Ba-133 measured in the sample, pCi/mL
As= Activity of Ba-133 in standardized carrier, pCi/mL
12.2 Count Rate
The net count rate, Ra, for any single source (sample, background, standard) is generically calculated as
^ _ Na Ba
where
Ra = Net count rate in counts per minute
Na = Number of counts observed over the sample counting period
ts = Duration of the sample counting period (i.e., live time) in minutes
14
-------
Ba = Number of counts observed over the background counting period
tB = Duration of the background counting period (i.e., live time) in minutes
The standard uncertainty ("one-sigma") of Ra is given by:
rn \ — a i a
-2 4-2
^ ls lB
The square of the standard uncertainty is denoted by u2(Ra):
2ru \ _ a j. a
u C^a) — .2 4-2
ls lB
12.3 Efficiency
The alpha counting mass-efficiency (self-absorption) calibration is established as described in Section 10.
Calculate efficiencies for each standard with a mass, m, of BaS04 as follows:
where
£am = Alpha Efficiency determined for mass m
Ra = Net count rate of the Ra-226 standard, counts per minute
^std(a) = Activity of the Ra-226 standard at the time of count, disintegrations per minute
Y = Yield (as calculated in Section 12.1)
A = Decay constant for radon-222, in days (0.1813 day1)
(t2 — ti) = (Midpoint of sample count - time of final BaS04 precipitation, Section 11.6). in days
The factor [1 + 3(l — e-^2-4^)] represents the ingrowth of the radium-226 daughters between the
time of formation of the final barium sulfate precipitate and the midpoint of the counting time.
Fit the data to develop a model equation for sam vs. m and use this to calculate the efficiency, sa, in the
following equations of sample results.
where
Ra = Net alpha count rate
sa = Alpha efficiency, determined from efficiency curve equation
V = Volume of sample aliquot, in liters
2.22 = Conversion factor from dpm to pCi
Y = Yield (as calculated in Section 12.1)
The factor [1 + 3(l — represents the ingrowth of the radium-226 daughters between the
time of separation and the midpoint of the counting time.
The radium standard counting uncertainty and expanded counting uncertainty (95 % confidence) are
calculated as:
12.4 Radium Alpha Activity
Calculate the radium alpha activity concentration for each sample.
Radium Alpha Activity pCi/L =
saxVx 2.22 xfX[l + 3(l -
ju2(Ra) ^ 1
x V x 2.22 xfXll + 3(l - e-Afe-tJ)
and U95o/o = 1.96 x u(ca)
where
15
-------
u(ca) = Radium standard counting uncertainty ("one-sigma") in pCi/L
U95o/0 = Expanded counting uncertainty (95 % confidence) in pCi/L
Ra = Net alpha count rate
u2(Ra) = Standard uncertainty of Ra, calculated as in Section 12.2
sa = Alpha efficiency, determined from efficiency curve equation
V = Volume of sample aliquot, in liters
2.22 = Conversion factor from dpm to pCi
Y = Yield (as calculated in Section 12.1)
The factor [1 + 3(l — represents the ingrowth of the radium-226 daughters between the
time of separation and the midpoint of the counting time.
1.96 = Coverage factor for 95 % level of confidence
12.5 Safe Drinking Water Act Detection Limit
The detection limit (DL) requirement for drinking water compliance samples is defined in Section 3.3 and
determination of method detection capability is described in Section 9.1.1.3. From the definition, the
equation in Section 12.5.1 can be derived (see reference 7 in Section 14).
12.5.1 DL Equation
The single-sample drinking water detection limit is calculated as:
where
ts = Sample count time, in minutes
tB = Background count time, in minutes
B = Number of background counts
s = Alpha efficiency, determined from efficiency curve equation
V = Volume of sample aliquot, in liters
2.22 = conversion factor from dpm to pCi
Y = Yield (as calculated in Section 12.1)
The factor [1 + 3(l — represents the ingrowth of the radium-226 daughters between the
time of separation and the midpoint of the counting time.
12.5.2 DL Study
The DL study described in Section 9.1.1.3 consists of seven replicate laboratory fortified blanks that are
prepared and counted as specified in the method. The replicate results are assessed using a chi-square
statistic to test whether the relative standard deviation of the results exceeds the maximum value
allowed at the required DL.
Calculate the mean of the measured values and the chi-square statistic as follows:
DL =
£ x V x 2.22 x Y x [1 + 3(1 - g-^-ti))]
n
i
j=i
And
16
-------
Where
n = Number of replicate measurements
= Activity fortified in sample replicates
To be deemed acceptable, the value of x2 should be less than or equal to the 99th percentile of the^2
distribution with (n-1) degrees of freedom.
12.6 Relative Percent Difference Calculation
As described in Section 9.2.8, relative percent difference (RPD) is used to evaluate precision of duplicate
measurements. The RPD is calculated as
where
As = Net activity of the first aliquot of sample
Adup = Net activity of the measurement obtained from a second aliquot of the same sample
If a duplicate sample measurement has an activity < 5x the detection limit and the calculated RPD > 20%,
calculate the normalized absolute difference (NAD). The NAD of the two measurements made from the
same sample assesses whether they are within 2 standard deviations of their aggregate measurement
uncertainty of each other. Calculate the NAD as
where
As = Net activity of the first aliquot of sample
Adup = Net activity of the measurement obtained from a second aliquot of the same sample
u2(As) = Square of the standard counting uncertainty ("one-sigma") associated with As
u2(Adup) = Square of the standard counting uncertainty ("one-sigma") associated with Adup
If the NAD is less than or equal to 2, then the two measurements are within 2 standard deviations of
each other and therefore acceptable. If, however, the NAD exceeds 2, recount the sample and duplicate.
If RPD/NAD evaluation still fails, a new sample and duplicate should be prepared.
The procedures described in this method generate relatively small amount of waste since only small
amounts of reagents are used. The matrices of concern are finished drinking water or source water.
Laboratory waste practices must be conducted consistent with the laboratory's radioactive materials
license and all applicable rules and regulations, and that laboratories protect the air, water, and land by
minimizing and controlling all releases from fume hoods and bench operations. Also, compliance is
required with any sewage discharge permits and regulations, particularly the hazardous waste
identification rules and land disposal restrictions.
RPD = )As ^1x100%
(y4s + Adupjj
13 Pollution Prevention
17
-------
14 References
1. Standard Methods for the Examination of Water and Wastewater, 22nd Ed., American Public
Health Association, Washington, D.C. (2011).
2. Manual for the Certification of Laboratories Analyzing Drinking Water, Criteria and Procedures
Quality Assurance, 5th Ed., USEPA 815-R-05-004, January 2005.
3. Multi-Agency Radiological Laboratory Analytical Protocols Manual (MARLAP), NTIS PB2004-
105421, July 2004.
4. Goldin, A. S., Determination of Dissolved Radium. Anal. Chem. 33, 406-409 (March 1961).
5. Kirby, H. W., Decay and Growth Tables for the Naturally Occurring Radioactive Series, Anal.
Chem., 26, 1063-1071 (1954).
6. Sill, C. W., Determination of Radium-226 in Ores, Nuclear Wastes and Environmental Samples by
High-Resolution Alpha Spectrometry, Nuclear and Chemical Waste Management, 7, 239-256
(1987).
7. Procedure for Safe Drinking Water Act Program Detection Limits for Radionuclides, USEPA 815-B-
17-003. April 2017.
18
------- |