United States
                    Environmental Protection
                    Agency
Environmental Monitoring Systems -
Laboratory
Las Vegas  NV 89114
                    Research and Development
EPA-600/S2-83-039  July 1983
&EPA         Project  Summary
                    Test Procedure for  lodine-131  in
                     Drinking Water:  Interlaboratory
                     Collaborative  Study
                    W. H. Yanko, C. T. Bishop, and A A. Glosby
                      An interlaboratory collaborative study
                    was conducted on a test procedure for
                    measuring the iodine-131 concentra-
                    tion in drinking water. The purpose of
                    the study was to determine the esti-
                    mated precision and accuracy of test
                    results from participating laboratories
                    using this test procedure and analyzing
                    drinking water samples.
                      Drinking water samples containing
                    iodine-131 at concentrations on the
                    designated collection date of 7.8,25.9,
                    and 78.3 picocuries per liter were an-
                    alyzed in triplicate by 11 collaborators
                    (10 laboratories) using the test pro-
                    cedure.
                      A statistical analysis of the test re-
                    sults showed coefficients of variation
                    for repeatability (within-laboratory pre-
                    cision) of 8.7, 10.5, and 7.7  percent
                    respectively for the three samples for
                    an average repeatability precision of
                    9.0 percent  The analysis also gave
                    coefficients of variation for reproduci-
                    bility (combined within-and between-
                    laboratory precision) of 17.4,15.5, and
                    15.7 percent for the respective samples
                    for an average reproducibility precision
                    of 16.2 percent
                      The average carrier iodine recoveries
                    were 76.4,  79.3, and 79.8  percent
                    respectively for the three samples, giving
                    an overall average of 78.5 percent. A
                    comparison of the grand average test
                    results for the three samples with the
                    known values for those samples shows
                    accuracy indexes of 97.4, 97.3, and
                    84.8 percent respectively, for an aver-
                    age accuracy for the  test method of
                    93.2 percent.
                      This Project Summary was developed
                    by EPA's Environmental Monitoring
                    Systems Laboratory, Las Vegas, NV, to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).

Introduction
  This report is submitted as partial fulfill-
ment of an Interagency Agreement, EPA-
IAG-79-D-X0736, between the  Environ-
mental Protection Agency and the Depart-
ment of Energy (DOE). Work was uone at
the Mound Facility of the Monaanto Re-
search Corporation under DOE  Contract
Number  DE-AC04-76-DP00053.   The
work covered the period of September 1,
1979 to June 15, 1980.
  The National Interim Primary  Drinking
Water Regulations (NIPDWR) require the
use of approved test procedures for an-
alyzing public drinking water supplies for
contaminants.  The Nl PDWR contains pro-
visions for the use of alternate test pro-
cedures  with precision  and  accuracy
equivalent to or better than the approved
test procedure. This  report describes a
multilaboratory test of a method selected
for the analysis of drinking water samples
for iodine-1 31 concentrations.  The pur-
pose of the study was to determine what
precision and accuracy could be expected
in the test results from laboratories using
the method for analyzing drinking  water
supplies for iodine-131 contamination.
  The NIPDWR requires a sensitivity (lower
detection  limit) of  1  picocurie  per liter
(pCi/l) for measuring iodine-131 concen-
trations in drinking water supplies, lodine-
131 decays by emitting a beta particle and
a gamma photon.  Gamma spectroscopy
has good specificity for gamma emitters,
but its sensitivity will meet the  required
sensitivity (1 pCi /I) only if the iodine-131

-------
is first separated from multiliter size sam-
ples. The  method tested, and described
here, requires separating the iodine-1 31,
using carrier iodine, from  one  or more
liters of sample and then counting the beta
activity produced by the separated iodine,
in a low beta background counting system.

Procedures

/.   Analytical Test Procedure
  The analytical procedure used  in this
study consists of detailed steps  in which
the iodine-1 31 with added iodate carrier is
separated  from the  sample as palladium
iodide.  The iodine-131 (plus carrier) is
reduced to the  iodide state with sodium
sulfite, precipitated  as silver iodide, and
purified  with zinc powder and  by two
precipitations as palladium iodide.  The
final palladium iodide is collected on a filter
and then counted for beta  activity.  The
recovery of the added carrier is determined
for each sample by weighing the palladium
iodide precipitate.  The counting efficiency
is determined using the same amount of
iodate carrier as is used in the procedure
and standard iodme-131 activity, precipi-
tated as palladium iodide, and beta counted.

2.   Collaborative Test
     Procedure
  A copy of the test procedure was sent to
potential participant laboratories; those
laboratories stating their willingness to
participate were used in this study.
  Three reference sample solutions were
prepared using tap water,  sodium thio-
sulfate and sodium carbonate preservatives,
known quantities of iodine-131  activity,
and lithium nitrate (for batch homogeneity
testing).   The three reference solutions
contained iodine-131 concentrations of
7.8,  25.9, and 78.3  pCi/l respectively
(activity calculated for the collection date
to which  analytical results were to be
normalized).  The first,  middle, and last
3.5-liter fractions of each sample batch
were used for homogeneity tests.
  Each participant was sent an instruction
sheet, a data report sheet, and iodme-1 31
standard solution for calibrations, and  a
3.5-hter  portion  of each  of the three
sample solutions.

3.   Data Processing Procedures
  A statistical evaluation of the test results
was done by the procedures described in
E-691,  E-177,  and E-178  of the ASTM
Standard Part 41, 1 980, to determine the
repeatability precision  (within-laboratory
variation);  the  reproducibility precision
(combined within- and between-laboratory
variation); and  the  accuracy of the test
procedure.  The standard deviations and
equations for their calculations are listed
below.
  Standard deviation of replicate test re-
sults within Lab i, for sample j, S,j
ra
l_h-i
                              •r
                                   (1)
where:   Xl)h  = the result reported for
                the h replicate of the j
         	     sample material by Lab i
         X,j   = the mean of the individual
                results of sample j for
                Labi
         n,j   = the number of replicates
                of sample j reported by
                Labi

Repeatability (within-laboratory) standard
deviation for sample j, Sr

  Since the number of replicates is the
same (3) for all participants for all three
samples, the  equation  can be given as
follows:
                    P
       Sr =  [  1/P2S,2
         'l    I         M
                                   (2)

where:   P  = the number of participants
              in the study.

  Standard deviation of individual labora-
tory average from grand average for the j
sample material, S*
           i= 1
          _
where:    X,  = the average of the test re-
            '    suits for sample material j
          	    by Lab i
          Xj  = the grand  average for
                sample material j

  Standard  deviation of between-labora-
tories for the j sample material,  SL.
      SL=
                                   (4)
Reproducibility (combined within- and be-
tween-laboratory) standard deviation for
the ] sample material, SR
      SR)=     Srj2
                                   The  percent  coefficient of variation for
                                   repeatability (within- laboratory precision)
                                   (also called repeatability index) for sample
                                   J. Vr%
                                                   Vr%= 100Sr/X~
                                                     J            '
                                                                            (6)
                                         The percent coefficient of variation for
                                         between- laboratory precision for sample j,
                                         V, %
                                                   VL%= 100 SL/X,
                                                                            (7)
                                         The percent coefficient  of variation for
                                         reproducibility (combined within- and be-
                                         tween-laboratory precision)  (also  called
                                         reproducibility index) for sample j, VR%
                                                   VR%=100SR)/X,
                                                                            (8)
                                         Accuracy index, a percent relationship of
                                         the grand average to the known value for
                                         the j sample material, Aj%
                                                    •i-  =  100^-
                                                                      O)
                                   where:    Y,   = the known value for the
                                                    j sample material (pCi/l)
                                   t-test to determine significant differences
                                   or systematic error for sample j, t
                                          t.=
                                   (5)
                                        	  , (P-1) degrees of freedom
                                        Sxj/(P)1/2                   (10)
                                   where:    P   = number of participants
                                             Y,   = known value of the sam-
                                                   ple j iodine-131 concen-
                                                   tration
                                             tc   = 2.23, critical  value for
                                                   11  participants, values
                                                   for t greater than 2.23
                                                   are significantly differ-
                                                   ent and show a system-
                                                   atic error.
                                   Results and Discussion
                                     A summary of the statistical evaluation
                                   of the test results for the three samples is
                                   given in Table 1, which lists the following
                                   statistical  parameters  and values for the
                                   three iodine-131 concentrations:

                                     1. The known value, Y,, for each of the
                                        iodine-131 concentrations^in pCi/I.
                                     2. The grand average value, X,, for each
                                        iodine-131  concentration (from 11
                                        participants)  in pCi/l.

-------
Table 1.    Summary of Collaborative Study Results - Precision and Accuracy

                                 lodine-131 (pCi/l)
Parameter e
Yj (pCi/l)
X, fpd/lj
A°/0
S, (pCi/l)
Sr(pCi/l)
Si (pCi/l)
c ' /*r;/n
&p \pL,l/l/
vr<%
v'%
Vfr%
1
1
7.8
7.6
97.4
1.19
0.66
1.14
1.32
8.7
15.0
17.4

2
25.9
25.2
97.3
3.25
2.64
2.87
3.90
10.5
11.4
15.5

3
78.3
66.4
84.8
9.54
5.14
9.07
10.43
7.7
13.7
15.7

Average


93.2




9.0
13.4
16.2

a Terms are defined in the text.

   3.  The accuracy index, A,%  (from  11
      participants) for each  iodine-131
      concentration, and  the  average  ac-
      curacy index over the concentration
      range of 7.8 to 78.3  pCi/l.
   4.  The standard deviation of the grand
      average values for the three iodine-
      1 31 concentrations, S ^..
   5.  The repeatability (withirl-laboratory)
      standard deviation,  Sr,  for  each
      iodine-131 concentration.
   6.  The between-laboratories standard
      deviation, SL, for each iodine-131
      concentrator!.
   7.  The reproducibility (combined with-
      in- and between-laboratory) standard
      deviation, SR, for each iodine-131
      concentration.
   8.  The coefficients of variation for re-
      peatability, Vr%; for between-labora-
      tory precision, VL %;  and  for repro-
      ducibility, VR%, fdreach iodine-1 31
      concentrator1! and the  averages of
      each for the  concentration range of
      7.8 to 78.3  pCi/l.
   The 66.4 pCi /I grand average compared
 to the 78.3  pCi/l known value showed a
 significant difference or systematic  error.
 A t-test gave a t( value of4.13 compared to
 the critical value (tc) of 2.23 for 11 partici-
 pants Therefore, 66.4 pCi/l is significantly
 different than 78.3 pCi/l and shows a low
 bias for the method for that level of iodine-
 1 31  concentration.

Conclusions
  The repeatability precision (within-la-
boratory  precision), reproducibility preci-
sion (the combined within- and between-
laboratory precision), and accuracy have
been determined by a multilaboratory test
of the method. Criteria by which to evaluate
an alternate test procedure for equivalency
have been established.
  Accuracy indexes of 97.4, 97.3, and
84.8  percent for the 7.8, 25.9, and 78.3
pCi/l iodine-131  concentration samples
respectively, and an average accuracy index
of 93.2 percent were obtained with this
test procedure. The study did not reveal an
explanation  for the lower 84.8 percent
accuracy for the sample with the highest
iodine-131 concentration.
  The estimated repeatability (within-lab-
oratory), single-operator, single-machine,
same-day, relative precision of the test
procedure for the determination of iodine-
131 concentrations in drinking water at
the 7.8 pCi /I level (averaged over the 11
participants) is ±1.3 pCi/l(17.4 percent
2Spercent); atthe25.9 pCi/l level(aver-
aged  over the  11  participants) is ±5.3
pCi/l (20.9 percent, 2S percent); and at
the 78.3 pCi/I level (averaged over the 11
participants) is±10.3 pCi/l(1 5.5 percent,
2S percent), for an average 2S percent of
18.0  percent for the range of iodine-131
concentrations of 7.8 to 78.3 pCi/l.
  The estimated reproducibility (combined
within- and between-laboratory), multiop-
erator (multilaboratory), single-machine,
same-day, relative precision of the test
procedure in the determination  of iodine-
131 concentrations at the 7.8 pCi/llevel
(averaged over  the  11  participants)  is
±2.7 pCi/l (35.8 percent, 2S percent); at
the 2 5.9 pCi/I level (averaged over the 11
participants) is ±8.2 pCi/l(32.6 percent,
2S percent); and at the 78.3 pCi/l level
(averaged over  the  11  participants)  is
±21.5 pCi/l (32.3 percent 2S percent),
for an average 2S percent of 33.6 percent
for the range of iodine-131  concentra-
tions  of 7.8  to 78.3 pCi/l.
  The variations observed  in the iodine
carrier recoveries by the 11 participants
for the three samples showed  a  depen-
dence on analyst technique for the method.
Also,  the  similar average iodine carrier
recoveries for the  three samples (76.4,
79.3, and 79.8 percent respectively) does
not offer any possible explanation for the
low bias observed for the sample with the
78.3  pCi/l iodine-131 concentration.
  The sensitivity of the test procedure is
not limited as much by the chemistry of
the procedure (iodine recovery) as it is by
the background and the counting efficiency
of the counting system used. With a beta
background of <5  cpm  and  a counting
efficiency of >20 percent, the sensitivity
of measurement  (lower  detection limit)
required in the NIPDWR (1  pCi/l)  can
easily be met  All participants whose test
results were used  in the  precision  and
accuracy analysis of this study used count-
ing instruments with beta backgrounds
<5 cpm and  counting efficiencies  >20
percent.

Recommendations
  It  is  recommended  that the  method
tested  in this study be  investigated for
procedure steps that result in analyst tech-
nique dependence (as demonstrated by
the variations in the iodine carrier  re-
coveries ranging from 48 to 100 percent).
One possible weakness in the procedure is
in the instructions for the  reaction of solid
zinc powder with solid silver iodide in an
aqueous system.
  The  procedure  should  be modified to
include a more detailed procedure for the
preparation  of a palladium iodide-131
counting standard and for the determina-
tion of the counting efficiency.  A revision
of section 6.5, the standardization of po-
tassium iodate, is recommended.
  The method should be re-tested to de-
termine if the low bias is real for the 78.3
pCi/l iodine-131 concentration and higher
levels.

-------
    W. H. Yanko, C. T. Bishop, and A. A. Glosby are with Monsanto Research Corp.,
      Miamisburg, OH 45342.
    Earl L. Whittaker is the EPA Project Officer (see below).
    The complete report, entitled "Test Procedure for Iodine-131 in Drinking Water:
      Interlaboratory Collaborative Study,"(Order No. PB 83-207589; Cost: $10.00,
      subject to change) will be available only from:
           National Technical Information Service
           5285 Port Royal Road
           Springfield,  VA 22161
           Telephone: 703-487-4650
    The EPA Project Officer can be contacted at:
           Environmental Monitoring Systems Laboratory
           U.S. Environmental Protection Agency
           P.O. Box 15027
           Las Vegas, NV 89114
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
                PS   «>0?912PROTECTION
                saaRfcs-

-------