Technical Notes for EPA Method 900.0 —
Gross Alpha and Gross Beta Radioactivity in Drinking Water
1. Scope and Application
1.1	The current regulation that stipulates acceptable methods is 40 CFR 141.25 and the limit on
gross alpha and gross beta radioactivity is 40 CFR 141.26.
1.2	The method is intended to act as a general screening tool to assess if further investigation
into activity of specific radionuclides is warranted.
The minimum energy limits of the alpha and beta particles relate to self-absorption of
lower energy emissions by the sample itself. Although the method is approved for alpha
emitters and mid-to-high energy beta emitters, there are some limitations. Alpha emitters
and mid-range energy beta emitters will be significantly attenuated in samples containing
dissolved or suspended solids. Thus sample self-absorption calibrations are necessary to
minimize bias for these radionuclides. Samples containing radionuclides of maximum beta
particle energy less than 100 keV, like 63Ni, 210Pb, 228Ra and 241Pu, cannot be effectively
screened using this method. For these low energy beta emitters a screening technique
employing liquid scintillation would be much more effective if these radionuclides are of
potential concern
1.2 Note that the NIPDWR requirements referenced in the method scope have been superseded
by recent regulations. Current requirements for drinking water compliance testing are
found in the most recent revision of 40 CFR 141. At the time of this writing, the Maximum
Contaminant Level (MCL) for Gross Alpha is 15pCi/L and for gross beta is 50 pCi/L with
specific cautions for each that are administrative in nature and are outside the scope of this
document. The Required Detection Limit (RDL) for gross alpha is 3 pCi/L and for gross
beta is 4 pCi/L. The technical derivation of this detection limit is provided in section 9 of
this document.
1.3-1.7 The limit on the total mass of solids that can be deposited on a 2 inch planchet is
necessary to minimize corrections for the effects of sample self-absorption of the alpha
and beta particles emitted from radionuclides in the sample. This value is 100 mg for
gross alpha and 200 mg for gross beta and assumes that the mass is evenly distributed on
a 2-inch planchet.
Laboratories should make efforts to assess the maximum volume of water that can be
used for this analysis prior to making any radiochemical measurements. This will prevent
exceeding the maximum allowable masses saving time and effort by the laboratory staff.
No guidance is provided in the method on how to perform this estimate of sample
volume. However, two straightforward and easily performed methods are identified here:
• Evaporate an aliquant of the sample into any vessel that can be easily dried and weighed
(similarly to how the final drying is performed in the actual analysis). Based on the dried
weight calculate a maximum volume that will not exceed 100 mg for gross alpha
Revision 0
1
04/01/2009

-------
Technical Notes for EPA Method 900.0, Gross Alpha and Gross Beta Radioactivity in Drinking Water
analysis. As an example, if 5 mL of water were evaporated yielding 2.5 mg of solids, the
maximum volume of water that could be used is (100 mg/2.5 mg)(5 mL) = 200 mL of
sample.
• Using a conductivity meter and a conversion scale for micromhos/cm to ppm (mg/L)
CaC03 estimate the total solids loading for a given volume in an unpreserved portion of
the sample. For example, a certain conductivity meter is calibrated at 1 micromho/cm per
3 ppm of calcium carbonate (0.33 micromho/cm/ppm). A sample having a conductivity
of 300 micromho/cm would have a solids loading of 900 mg/L. Thus the maximum
volume of sample that could be used would be estimated at 111 mL.
It is important to note that these would only be estimates and that the final determination of
the mass would still need to be performed on the aliquant of sample that was evaporated and
counted.
It has been established that a volume of water that yields no more than 100 mg of solids may
be used to perform gross alpha analysis and for gross beta analysis the limit is 200 mg. The
absorption of alpha particles is more significant than for that of beta particles and the value of
100 mg is often the limiting factor in determining how much volume can be used when
simultaneous gross alpha beta is performed.
1.8	The evaporation technique used in the method does not accommodate those radionuclides
that potentially are volatile as nitrates at 105 °C, or those radionuclides that may be volatile
(such as tritium, iodine, carbon and technetium).
Some metals form hygroscopic salts with nitric acid that cannot be easily dried prior to
their being cooled and counted. The term "hygroscopic" means that the material can readily
absorb moisture from the atmosphere. This will be noted when attempting to bring the
planchet to constant weight and the weight of the sample continues to increase. In these
instances, the planchets may be flamed until a dull cherry red color is imparted to the
planchet for a few minutes (this corresponds to a temperature range of 500 °C to about
800 °C). At this temperature, nitrates will usually be converted to oxides that are not
hygroscopic. After flaming of the planchet, the final mass on the planchet needs to be re-
determined (to correct for sample self absorption) and should not exceed 100 mg.
At 500-800 °C, it is possible to lose some radionuclides regardless of their chemical state
or that imparted to the radionuclide by its matrix. Samples containing polonium, lead, and
cesium may sustain losses at these temperatures.
1.9	Drinking water samples with high levels of solids will prove to be challenging for this
technique as the solids will contribute significantly to self-absorption of the alpha and beta
particles prior to reaching the detector. An alternate method, Gross Alpha Screening (EPA
Method 900.1), first precipitates BaS04in an attempt to reduce the amount of soluble salts,
and limiting the mass of precipitate. This alternate method however, assumes that the bulk
of the alpha emitters are radium related and will coprecipitate with barium. This may not
always be the case, and the results of the Methods 900.0 and 900.1 may not give
Revision 0
2
04/01/2009

-------
Technical Notes for EPA Method 900.0, Gross Alpha and Gross Beta Radioactivity in Drinking Water
comparable values. The tables in 40 CFR 141 suggest alternate screening methods for high
solids samples that are acceptable (such as SM 7110C).
The alternate method for Gross Alpha Screening, 900.1, is meant for total radium isotopes
in water with high dissolved solids. Method 900.1 (and others identified in 40 CFR 141) is
recommended if the sample has solid loading of >500 ppm.
2 Summary of the Method
The method is very straightforward and does not require any chemical separations. An
appropriately sized aliquant of the sample is verified to have a pH less than 2.0, or have its pH
adjusted to less than 2.0 using nitric acid. The nitric acid is added to volatilize chlorides during
sample evaporation. The sample is reduced in volume to the point where it can be transferred to a
stainless steel planchet and evaporated to a dry solid. After verifying that the sample is dry (i.e.,
at constant weight), it is counted 72 hours after evaporation on a gas proportional counter for
gross alpha and gross beta analysis. However, gross beta count can be made any time after
planchet preparation. The flow of the overall sample processing is depicted in Figure 1.
Figure la. Sample Processing Flow for Gross Alpha and Gross Beta Analysis
Revision 0	3	04/01/2009

-------
Technical Notes for EPA Method 900.0, Gross Alpha and Gross Beta Radioactivity in Drinking Water
Figure lb. Sample Processing Flow for Gross Alpha and Gross Beta Analysis (Continued)
The term "constant weight" means that the residue on the planchet must be dried and weighed at
least twice. The temperature of the oven is at 105 °C in order to drive off waters adsorbed onto
the surface of the solid residue. Drying at 105 °C neither removes waters of hydration, nor
changes the chemical form of potentially hygroscopic salts so they do not readily absorb
atmospheric moisture. In between the first and second drying and weighing cycle, the planchet
should be placed in a desiccator to cool and prevent re-absorption of water. For many solids, the
slope of the self-absorption curve is on the order of 0.2% per milligram of solids. Water will
have a smaller self absorption as it has a lower average Z (atomic number) value that any solids.
Thus a good estimate of constant weight would be when the change in the mass measured
between successive weighings is no greater than 2 mg (less than about 0.4% of self absorption).
Should the first two cycles not yield values that are near this weight loss, additional drying and
weighing cycles should continue until the value of 2 mg is approached. It is important to note
that "constant weight" is determined after samples have been stored in a desiccator. The samples
may quickly reabsorb ambient moisture, however, when they are removed from the desiccator
for counting. For this reason, if constant weight cannot be achieved by heating at 105 °C, more
aggressive heating may be required so that the chemical form of the salt causing the inconstant
weight can be altered. This may require that the sample be heated in a flame to about 900 °C
(usually considered "cherry red").
Efficiency curves for alpha and beta activity must be determined based on varying the mass of
solids for a given radionuclide activity. This plot of efficiency vs. mass of deposited solids is
then used to determine the efficiency for each sample based on its mass of deposited solids.
For these efficiency curves it is important to note the following:
•	The areal density and mass distribution of the sample should match those of the calibration
standards.
•	The planchets used for the calibration standards and the samples are identical in size, shape
and characteristic form (e.g., flat versus ridged bottom).
Revision 0
4
04/01/2009

-------
Technical Notes for EPA Method 900.0, Gross Alpha and Gross Beta Radioactivity in Drinking Water
3	Sample Handling and Preservation
Sample preservation and the verification that this has been successfully achieved are an
important requirement to this method. The preservation technique for this method uses enough
nitric acid to bring the pH of the sample to less than 2.0. This prevents precipitation of not only
the radionuclides but also other metals that may be in the sample that may coprecipitate the
radionuclides.
A sample that is received more than 5 days after sampling and does not have a pH less than 2.0
will not be acceptable for analysis. This is because hydroxides of trace metals in the sample,
including the radionuclides of interest, can precipitate and irreversibly adhere to the container
walls. This would be a loss of analyte prior to analysis leading to a non-conservative estimate of
the gross activity.
Simply adding 15 mL of concentrated nitric acid may be insufficient to lower the pH to less than
2.0 for drinking waters that have high concentrations of calcium and magnesium.1 This volume
of added nitric acid also does not account for the sample size; it is assumed the sample size is 1
liter. If your sample size is different the added nitric acid should be proportionally larger or
smaller. Even though acid has been added, it is necessary to check the sample pH upon receipt. If
not less than 2.0, add acid in 15 mL increments, equilibrate the sample by shaking and check pH
again. Once the pH is less than 2.0 the sample must be allowed to equilibrate for 16 hours prior
to analysis, and then the pH checked again ensuring it is less than 2.0. The reason for this waiting
period is to ensure that all finely suspended materials or surface adsorbed materials have been
solubilized. Should more than 30 mL of acid be added to reduce the pH to the appropriate value,
consideration should be made for applying a volume correction factor.
The certification manual allows acidification with either nitric or hydrochloric acid. However if
HC1 is used, nitric acid should be added during the process of sample evaporation to ensure that
all chlorides are removed via volatilization. Chlorides if present will rapidly corrode stainless
steel planchets causing flowing solids of iron hydroxide, increasing the mass of the residue, and
creating significant sample self absorption.
4	Interferences
Moisture associated by the sample residue may interfere with this method. The salts in the dried
sample residues are frequently hygroscopic. When these salts absorb ambient moisture, the areal
density of the residue, that is the mass of residue per cm2 of source surface area, will change.
Since the areal density affects the number of alpha or beta particles that reach the detector
sensitive area, the accuracy with which the mass is known will limit the accuracy of self-
absorption corrections applied to the counting results. Thus, it is important to consider whether
the mass of residue as weighed corresponds to that present at the time of counting (see discussion
of constant weight in Section 2 above).
1 15 mL of 16 M HN03 will reduce the pH of 1 liter of unbuffered, demineralized water from 7.0 to 0.6. Thus any
chemicals in the water which react with acid will not achieve this low of a pH. Thus it is necessary to check the pH
following the acid addition.
Revision 0
5
04/01/2009

-------
Technical Notes for EPA Method 900.0, Gross Alpha and Gross Beta Radioactivity in Drinking Water
The counterpoint, dry samples that have obtained a static charge, may ionize the carrier gas in
the gas proportional counting system yielding increased counts above the level of radioactivity
that is actually present. Accumulation of static charge may occur when conditions of low
humidity (<60% relative humidity) exist in the laboratory. If such conditions exist, the static
charge may be successfully removed by connecting a small wire to a grounded metal structure
and touching the planchet with the wire before counting.
Modern instruments can perform "simultaneous" alpha and beta counting (also termed
"simultaneous counting" or "counting at the beta voltage"). These instruments separate counts
into distinct "alpha" and "beta" channels by discriminating between the typical pulse sizes
created by alpha and beta particles. Crosstalk2 occurs when alpha decays are misclassified as
beta counts and stored in the beta channel or when beta decays are misclassified as alpha counts
and stored in the alpha channel. Alpha-to-beta crosstalk is generally ranges from one to two
orders of magnitude greater than beta-to-alpha crosstalk. Although crosstalk cannot be
eliminated when counting on the beta plateau, beta-to-alpha crosstalk can be minimized during
set-up of the instrument. The intrinsic crosstalk of the detector can be minimized during
instrument set-up by adjusting the alpha and beta discriminators.3 While minimizing crosstalk
will also minimize corrections needed for most samples, keep in mind that if there is a significant
difference in alpha and beta activities significant crosstalk may still be present. For example a
gross alpha measurement of 5 pCi/L and a gross beta of 35 pCi/L may have significant crosstalk
of beta into alpha that should be accounted for using crosstalk equations. In cases where
unusually high ratios of beta activity to alpha activity are present in samples (e.g., effluents from
a nuclear power plant or high levels of beta-emitting contaminants), determining the alpha
activity of samples in a separate measurement from the beta by counting on the alpha plateau
provides the most effective discrimination against beta-to-alpha crosstalk and the most accurate
measurement of alpha activity in samples. Alpha-to-beta and beta-to-alpha crosstalk are
accounted and corrected for by calibrating the instrument with standards of the pure alpha or
pure beta emitters. The method and calculations for determining crosstalk are further explained
in Section 9 of this Technical Note.
5. Apparatus
Ensure that the planchets used for sample preparation are identical to those used for efficiency
calibration and background counts.
2	In section 9.2.2 of Method 900.0, "alpha amplification factor" is used instead of crosstalk. The term "alpha
amplification factor" is no longer used and has been replaced by the term crosstalk. If the beta count rate is
significant, a beta-to-alpha channel crosstalk also may occur and may need to be accounted for by similar
calculations.
3	Minimizing beta-to-alpha crosstalk to a fraction of a percent (e.g., to ~0.1%) during instrument set-up will
minimize the absolute contributions to count activity to the alpha channel from beta activity in the sample. This
minimizes the size of the crosstalk correction that is applied and optimizes the accuracy and uncertainty of the
measurement. Note that access to discriminators varies by manufacturer and model of the instrument. Consult the
operation manual and instrument manufacturer regarding specific capabilities and discriminator set-up of
instruments.
Revision 0
6
04/01/2009

-------
Technical Notes for EPA Method 900.0, Gross Alpha and Gross Beta Radioactivity in Drinking Water
6. Reagents
No comments
7 Calibration
The standard traditionally used for alpha activity was 241Am. This standard had several
advantages:
•	It has a long half-life (432 years) and thus provides a very constant activity source.
•	Its first progeny is 237Np also has a long half-life (2.14x 106 years). This is important
because a 5% ingrowth of this alpha-emitting radionuclide will take about 200,000 years.
Thus, it will not contribute to the total alpha activity.
•	The 241 Am has an energy that is in between the naturally occurring radionuclides that are
likely contaminants. Thus it provides a response that will be adequately representing
either end of the alpha energy range.
Unfortunately 241Am has the distinct disadvantage of having a 59 keV gamma-ray that creates a
response in the beta channel.
However in 40 CFR 141, 230xh4 is now recommended for this function since it is not only a long-
lived alpha emitter but its energy better mimics the alpha particle energy of naturally occurring
radionuclides, such as radium isotopes and their progeny. It also has the disadvantage of a low
energy beta component associated with its internal conversion electron at 68 keV, plus
miscellaneous low-energy gamma rays.
241	230
Because neither Am nor Th is a pure alpha emitter they both have their disadvantages.
Although not permitted for calibration, 210Po, a pure alpha emitter would be the best way to
assess the potential crosstalk in the instrument just resulting from alpha.
The required calibration standard for beta channel is 90Sr/90Y that also has several advantages.
•	Both have long half-lives (90Sr is 29 years)
•	90Sr is in secular equilibrium with 90Y (half-life ~ 2.67 days) in the source and thus the
total activity is actually twice that of the documented source activity. The decay product
for 90Y is stable 90Zr, thus adding no further activity.
•	137Cs, which has been used at some laboratories as a beta reference, is in secular
equilibrium with 137mBa. The 137mBa is a gamma emitter and decays to its stable ground
state. Thus there is no additional contribution to the activity.
For 90Sr, it will be necessary to decay correct the standard to the date of the efficiency
calibration.
4 Except for California, whose regulations require their laboratories to use natural uranium
Revision 0	7
04/01/2009

-------
Technical Notes for EPA Method 900.0, Gross Alpha and Gross Beta Radioactivity in Drinking Water
There are several suppliers of National Institute of Standards and Technology5 (NIST) traceable
radioactivity standards. It is no longer required (or recommended) to purchase these standards
directly from NIST or any other government agency.
The laboratory's tap water is the recommended matrix for use in preparing the efficiency
standards so as to provide a residue matrix which is comparable to that of other drinking water
samples to be tested. However, many labs now receive samples from a variety of regional areas
where the water composition may be significantly different from their own. This means that the
residue matrix and the calibration matrix may not necessarily match too well. A surrogate water
matrix similar to one used for commercially available performance test samples may provide a
more universal matrix that will be comparable among several laboratories. Another possibility
would be to attempt to matrix match the calibration standards to the average concentrations of
the common ions contributing to the water samples analyzed (e.g., calcium, magnesium,
aluminum, carbonate, chloride, sulfate).
For the alpha self-attenuation curve the method has 241 Am as the standard. Remember that this is
now changed to 230Th.
Residues for beta analysis can be as high as 200 mg because beta particles are more penetrating
than alpha particles. Although beta counting can be performed with solid residues up to 200 mg
of solids on a 2-inch planchet, alpha counting must be based on a sample with less than 100 mg
of solids. This means that it may be necessary to prepare separate samples for gross alpha and
gross beta analyses in order to achieve the required detection limit.
8 Procedure
The actual technique of reducing the volume can have a significant effect on the recovery of
material from the samples. Nitric acid forms an azeotropic mixture with water and boils of as the
water volume is reduced. Thus, during the process of volume reduction small aliquants of nitric
acid should be rinsed down the sides of the evaporation vessel. This ensures three things will
occur:
•	The acidity of the sample will not decrease during the evaporation process;
•	The radionuclides will not dry out on the container walls and be irreversibly adsorbed; and
•	Any chlorides contained in the sample will be evaporated during the volume reduction
process, because HC1 is more volatile than nitric acid.
The method indicates that removal of chloride salts may take place when the sample is at the
residue stage. However, removal of chlorides before they form salts on the planchet will prevent
considerable problems later on when the final solution is evaporated on the planchet. An
indication that not all chlorides have been removed will be a voluminous brown residue that does
not appear to belong to the sample. In such cases, the sample should be re-prepared to remove
residual chlorides prior to evaporation on the planchet. Evaporated sample residues should be
deposited as uniformly on the planchet as possible.
5 NIST was formerly the National Bureau of Standards (NBS).
Revision 0
8
04/01/2009

-------
Technical Notes for EPA Method 900.0, Gross Alpha and Gross Beta Radioactivity in Drinking Water
Step 8.4 of the method indicates that the beta counts can be made immediately but that 72 hours
of in-growth is required for the alpha count after evaporation onto the planchet. The 72-hour
waiting period does not have a very good technical basis. If the presence of 224Ra is not
considered, 72 hours (3 days) allows two things to occur:
•	Decay of any unsupported progeny of 222Rn, although it is likely that within 8 hours these
should be decayed completely.
ooo	226
•	Partial ingrowth of Rn (and its supported progeny) from Ra.
The actual ingrowth for full equilibrium of 222Rn progeny occurs in about 21 days. In this respect
the method does not provide the most conservative estimate of the true gross alpha or the gross
beta activity.
If 224Ra is present, the issue of ingrowth and decay is significantly more complicated, and is not
easily accounted for in the 72-hour timeframe. However laboratories should be aware of the fact
that some states impose specific regulations (beyond 40 CFR 141) with regards to 224Ra and its
progeny in drinking water.
No requirements exist for when gross beta counting may be performed. The limit for gross beta
concentration is based on whether or not a water supply is "deemed vulnerable." If a water
supply has a gross beta value that exceeds the maximum contaminant level (MCL) it may be
necessary to perform other analyses (e.g., 40K by gamma spectrometry). However this decision
lies with the water supply regulating authority. Thus values for gross beta of greater than the
MCL should be communicated to the client so that if speciation of beta emitters is required the
laboratory can be told how to proceed.
9 Calculations
The equations for activity calculation are relatively straight forward. The determination of the
cross talk factor (alpha amplification factor) and its application to the gross beta activity must be
done to accurately determine the final beta activity. It may be possible that high beta activity
could lead to beta into alpha crosstalk that would affect the final alpha result. Thus a cross talk
correction factor for this should also be determined. The best time to establish these crosstalk
factors would be at the time of efficiency calibration and measurement of self-absorption
correction factors.
If the alpha activity results in a significant contribution to beta activity due to cross talk (greater
than 4% of the total beta activity, or 2 pCi if the sample's beta activity was as large as 50 pCi)
the appropriate correction should be made and this process would be identified in the
1 ab oratory' s procedure.
The method as written does not provide an equation to calculate the counting uncertainty
associated with each sample measurement. Generally speaking the counting uncertainty is the
principal contributor to the total uncertainty of the analysis for an individual sample where the
gross alpha or beta concentration is below about 5 pCi/L. The following equation is based on the
Revision 0
9
04/01/2009

-------
Technical Notes for EPA Method 900.0, Gross Alpha and Gross Beta Radioactivity in Drinking Water
generalized formula for counting uncertainty ("error") shown in Appendix B of the 900 Series
method manual and can be used to calculate the counting uncertainty for the result:
U, at 95% confidence =

Ro Rr
1 96

1
ts tB
2.22 x V x 8
The requirement for detection limit is stated in 40 CFR 141:
For the purpose of monitoring radioactivity concentrations in drinking water, the required
sensitivity of the radioanalysis is defined in terms of a detection limit. The detection limit
shall be that concentration which can be counted with a precision of plus or minus 100
percent at the 95 percent confidence level (1.96[sigma] where [sigma] is the standard
deviation of the net counting rate of the sample.
This definition of the detection limit (DL) in the current version of 40 CFR 141.26 translates into
the following equation:
Where:
4
fa
Rb
V
£
DL =
2.22 xVxs
time of the measurement used to accumulate the sample count, minutes
time of the measurement used to accumulate the background count, minutes
mean background count rate, cpm
sample volume used, L
efficiency and the self absorption correction
The equation for the DL cited above will be used here in an example for determination of gross
alpha activity.
Assume a sample volume of 125 mL (V) yields a sample mass of 100 mg. The sample and
background count times are both 240 minutes (7S and fa). The alpha background count rate is
0.07 cpm (Rb), and the detection efficiency is 0.072 (e). Inserting these values into the equation
we get a DL of 2.8 pCi/L. This would meet the required detection limit (RDL) of 3 pCi/L for
gross alpha analysis.
In Appendix C of the EPA 900 series procedures manual there is a note that states the maximum
recommended count time is 1,000 minutes. If the DL cannot be achieved in that time period a
different instrument (i.e., higher efficiency), a larger sample or a different method should be
used.
Revision 0
10
04/01/2009

-------