Tennessee
Valley
Authority
Office of Natural
Resources
Chattanooga TN 37401
TVA/ONR/NRO-82/4
United States
Environmental Protection
Agency
Research and Development
Office of Environmental
Processes and Effects Research
Washington DC 20460
EPA-600 7-82-005
March 1982
Microprocessor-
Controlled Ion
Selective Electrode
Determination of
Total Chlorine
Interagency
Energy/Environment
R&D Program
Report
600782005
-------
-------
EPA-600/7-82-005
TVA/ONR/NRO-82/4
XXXXX XXXX
MICROPROCESSOR-CONTROLLED ION SELECTIVE ELECTRODE
DETERMINATION OF TOTAL CHLORINE
by
Lyman H. Howe, Reginald E. Hadley, and Gary A. Fischer
Office of Natural Resources
Tennessee Valley Authority
Chattanooga, Tennessee 37401
Interagency Agreement No. D5-E721
Project No. E-AP 81BDH
Program Element No. 1NE 833
Project Officer
James T. Stemmle
Office of Environmental Processes and Effects Research
U.S. Environmental Protection Agency
Washington, DC 20460
Prepared for
Office of Environmental Processes and Effects Research
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC 20460
-------
DISCLAIMER
This report was prepared by the Tennessee Valley Authority and has been
reviewed by the Office of Energy and Air, U.S. Environmental Protection
Agency, and approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the Tennessee
Valley Authority or the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
-------
ABSTRACT
A microprocessor-controlled ion selective electrode (MC) method was
evaluated and compared to the forward amperometric titration (AT)
method for determining total chlorine in condenser cooling river from
coal-burning electric plants. The effective range for quantification
by the MC method is from the minimum detection limit (MNDL) of 6.5
chlorine for unspiked condenser water .and 13-3 pg/1 chlorine for spiked
condenser water to 100 [Jg/1 chlorine. Interferences by zinc(Il), cop-
per(II), iron(III), arsenic(III), and manganese(VII) are discussed. The
pH, chromium(VI), mercury(II), bromide, and arsenic(V) do not interfere
with measurement of total chlorine. For both unspiked and spiked con-
denser water, the overall pooled standard deviation and overall mean per-
centage relative standard deviation for concentrations from 20 to 200 (Jg/1
chlorine are lower for the MC method than the AT method. Standard devia-
tions are discussed for the MC method for concentrations from 2 to 20 (Jg/1
chlorine.
This report was submitted by the Tennessee Valley authority, Office of
Natural Resources, in partial fulfillment of Energy Accomplishment Plan
81BDH under terms of Interagency Energy Agreement D5-E721 with the
Environmental Protection Agency. Work was completed in September 1981.
111
-------
-------
CONTENTS
Page
Abstract iii
Figures vi
Tables xvii
Abbreviations and Symbols xxi
Acknowledgements xxiii
1. Introduction 1
2. Conclusions 5
3. Recommendations 7
4. Experimental 8
Sample Preparation 8
Equipment 9
Preparation of Solutions for Conducting Tests 9
Determination of Total Chlorine by the MC Method 11
Determination of Total Chlorine by the AT Method 12
5. Results and Discussion 14
Deionized Water 14
River Water 25
Condenser Cooling River Water 28
References 43
Glossary 46
-------
FIGURES
Number
Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentration
in micrograms per liter for experimental recovery
by MC method of total chlorine from deionized water
spiked with chloramine-T 49
Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by MC method
of total chlorine from deionized water spiked
with chloramine-T 50
Percentage mean bias against total chlorine concentra-
tion added in micrograms per liter for experimental
recovery by MC method of total chlorine from deionized
water spiked with chloramine-T 51
Total chlorine standard deviation in raicrograms per
liter against mean total chlorine concentration in
micrograms per liter for experimental recovery by
AT method of total chlorine from deionized water
spiked with chloramine-T 51
Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by AT method
of total chlorine from deionized water spiked
with chloramine-T 52
Percentage mean bias against total chlorine
concentration added in micrcgrams per liter for
experimental recovery by AT method of total
chlorine from deionized water spiked with
chloramine-T 53
Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by MC method of total chlorine from
deionized water spiked with calcium hypochlorite ... 54
VI
-------
FIGURES
(Continued)
Number Pagt
8 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by MC
method of total chlorine from deionized water
spiked with calcium hypochlorite 55
9 Percentage mean bias against total chlorine
concentration added in micrograms per liter
for experimental recovery by MC method of total
chlorine from deionized water spiked with calcium
hypochlorite 56
10 Total chlorine standard deviation in micrograms per
liter against mean total chlorine concentration
in micrograms per liter for experimental recovery
by AT method of total chlorine from deionized water
spiked with calcium hypochlorite 57
11 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by AT method
of total chlorine from deionized water spiked
with calcium hypochlorite 58
12 Percentage mean bias against total chlorine
concentration added in micrograms per liter
for experimental recovery by AT method of
total chlorine from deionized water spiked
with calcium hypochlorite 59
13 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by MC method of total chlorine from
deionized water spiked with 50% (w/w) mixture
of calcium hypochlorite and chloramine-T as
chlorine 60
14 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by MC method
of total chlorine from deionized water spiked
with 50% (w/w) mixture of calcium hypochlorite
and chloramine-T as chlorine 60
VI1
-------
FIGURES
(Continued)
Number Page
15 Percentage mean bias against total chlorine con-
centration added in micrograms per liter for
experimental recovery by MC method of total
chlorine from deionized water spiked with
50% (w/w) mixture of calcium hypochlorite
and chloramine-T as chlorine 61
16 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion for experimental recovery by AT method of
total chlorine from deionized water spiked with
50% (w/w) mixture of calcium hypochlorite
and chloramine-T as chlorine 62
17 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by AT method
of total chlorine from deionized water spiked
with 50% (w/w) mixture of calcium hypochlorite
and chloramine-T as chlorine 62
18 Percentage mean bias against total chlorine
concentration added in micrograms per liter
for experimental recovery by AT method of total
chlorine from deionized water spiked with
50% (w/w) mixture of calciuirt hypochlorite
and chloramine-T as chlorine 63
19 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by MC method of total chlorine from
raw Ross Landing surface river water sample
collected October 3, 1980, analyzed October
7-12, 1980, spiked with chloramine-T 63
20 Percentage relative standard deviation in
micrograms per liter against mean total
chlorine concentration in micrograms per
liter for experimental recovery by MC method
of total chlorine from raw Ross Landing surface
river water sample collected October 3, 1980,
analyzed October 7-12, 1980, spiked with
chloramine-T 64
VI11
-------
FIGURES
(Continued)
Number
21 Total chlorine standard deviation against
mean total chlorine concentration in micro-
grams per liter for experimental recovery by
AT method of total chlorine from raw Ross
Landing surface river water sample collected
October 3, 1980, analyzed October 7-12, 1980,
spiked with chloramine-T 65
22 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by AT method
of total chlorine from raw Ross Landing surface
river water sample collected October 3, 1980,
analyzed October 7-12, 1980, spiked with
chloramine-T 66
23 Total chlorine standard deviation in micro-
grams per liter against mean total chlorine
concentration in micrograms per liter for
experimental recovery by MC method of total
Chlorine from raw Ross Landing surface river
water sample collected October 3, 1980,
analyzed October 12, 1980, spiked with
calcium hypochlorite 67
24 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by MC method
of total chlorine from raw Ross Landing surface
river water sample collected October 3, 1980,
analyzed October 12, 1980, spiked with calcium
hypochlorite 68
25 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion for experimental recovery by AT method of
total chlorine from raw Ross Landing surface
river water sample collected October 3, 1980,
analyzed October 12, 1980, spiked with calcium
hypochlorite 69
-------
FIGURES
(Continued)
Number
26 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by AT method
of total chlorine from raw Ross Landing surface
river water sample collected October 3, 1980,
analyzed October 12, 1980, spiked with calcium
hypochlorite 69
27 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by MC method of total chlorine from
raw unchlorinated, chlorinated, and mixtures
of unchlorinated and chlorinated Kingston
Steam Plant condenser cooling river water
samples analyzed October 15-16, 1980 70
28 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by MC method
of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated
and chlorinated Kingston Steam Plant condenser
cooling river water samples analyzed
October 15-16, 1980 71
29 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by AT method of total chlorine from
raw unchlorinated, chlorinated, and mixtures
of unchlorinated and chlorinated Kingston Steam
Plant condenser cooling river water samples
analyzed October 15-16, 1980 72
30 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by AT method
of total chlorine from raw unchlorinated, chlori-
nated, and mixtures of unchlorinated and chlorinated
Kingston Steam Plant condenser cooling river water
samples analyzed October 15-16, 1980 72
-------
FIGURES
(Continued)
Number
31 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by MC method of total chlorine from
raw unchlorinated, chlorinated, and mixtures
of unchlorinated and chlorinated Kingston
Steam Plant condenser cooling river water
samples analyzed October 15-16, 1980, spiked
with calcium hypochlorite 73
32 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by MC method
of total chlorine from raw unchlorinated, chlori-
nated, and mixtures of unchlorinated and chlorinated
Kingston Steam Plant condenser cooling river water
samples analyzed October 15-16, 1980, spiked with
calcium hypochlorite 74
33 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by AT method of total chlorine from
raw unchlorinated, chlorinated, and mixtures
of unchlorinated and chlorinated Kingston Steam
Plant condenser cooling river water samples
analyzed October 15-16, 1980, spiked with
calcium hypochlorite 75
34 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by AT method
of total chlorine from raw unchlorinated, chlori-
nated, and mixtures of unchlorinated and chlorinated
Kingston Steam Plant condenser cooling river water
samples analyzed October 15-16, 1980, spiked with
calcium hypochlorite 76
XI
-------
FIGURES
(Continued)
Number
35 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by MC method of total chlorine from
raw unchlorinated, chlorinated, and mixtures
of unchlorinated and chlorinated Shawnee Steam
Plant condenser cooling river water samples
analysed October 21-22, 1980 77
36 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per 1Lter for experimental recovery by MC method
of total chlorine from raw UMchlorinated, chlori-
nated, and mixtures of unchlorinated and chlorinated
Shawnee Steam Plant condenser cooling river water
samples analyzed October 21-22, 1980 78
37 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by AT method of total chlorine from
raw unchlorinated, chlorinated, and mixtures
of unchlorinated and chlorinated Shawnee Steam
Plant condenser cooling river water samples
analyzed October 21-22, 1980 79
38 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by AT method
of total chlorine from raw unchlorinated, chlori-
nated, and mixtures of unchlorinated and chlorinated
Shawnee Steam Plant condenser cooling river water
samples analyzed October 21-22, 1980 80
39 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by MC method of total chlorine from
raw unchlorinated, chlorinated, and mixtures
of unchlorinated and chlorinated Shawnee Steam
Plant condenser cooling river water samples
analyzed October 21-22, 1980, spiked with
calcium hypochlorite 81
-------
FIGURES
(Continued)
Number
4-0 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by MC method
of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and
chlorinated Shawnee Steam Plant condenser cooling
river water samples analyzed October 21-22, 1980,
spiked with calcium hypochlorite 82
41 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by AT method of total chlorine from
raw unchlorinated, chlorinated, and mixtures
of unchlorinated and chlorinated Shawnee Steam
Plant condenser cooling river water samples
analyzed October 21-22, 1980, spiked with
calcium hypochlorite 83
42 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by AT method
of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and
chlorinated Shawnee Steam Plant condenser cooling
river water samples analyzed October 21-22, 1980,
spiked with calcium hypochlorite 84
43 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by MC method of total chlorine from
raw unchlorinated, chlorinated, and mixtures
of unchlorinated and chlorinated Allen Steam
Plant condenser cooling river water samples
analyzed October 28-29, 1980 85
44 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by MC method
of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and
chlorinated Allen Steam Plant condenser cooling
river water samples analyzed October 28-29, 1980 ... 86
x i :L i
-------
FIGURES
(Continued)
Number
45 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by AT method of total chlorine from
raw unchlorinated, chlorinated, and mixtures of
unchlorinated and chlorinated Allen Steam Plant
condenser cooling river water samples analyzed
October 28-29, 1980 87
46 Percentage relative standard deviation against
mean concentration for experimental recovery
by AT method of total chlorine from raw unchlori-
nated, chlorinated, and mixtures of unchlorinated
and chlorinated Alien Steam Plant condenser cooling
river water samples analyzed October 28-29, 1980 ... 88
47 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by MC method of total chlorine from
raw unchlorinated, chlorinated, and mixtures of
unchlorinated and chlorinated Allen Steam Plant
condenser cooling river water samples analyzed
October 28-29, 1980, spiked with calcium
hypochlorite 89
48 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by MC method
of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and
chlorinated Allen Steam Plant condenser cooling
river water samples analyzed October 28-29, 1980,
spiked with calcium hypochlorite 90
49 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by AT method of total chlorine from
raw unchlorinated, chlorinated, and mixtures of
unchlorinated and chlorinated Allen Steam Plant
condenser cooling river water samples analyzed
October 28-29, 1980, spiked with calcium
hypochlorite 91
xiv
-------
FIGURES
(Continued)
Number Pagt
50 Percentage relative standard deviation against
mean total chlorine concentration in micrograms
per liter for experimental recovery by AT method
of total chlorine from raw utichlorinated,
chlorinated, and mixtures of unchlorinated and
chlorinated Allen Steam Plant condenser cooling
river water samples analyzed October 28-29, 1980,
spiked with calcium hypochlorite 92
51 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by MC method of total chlorine from
raw unchlorinated, chlorinated, and mixtures
of unchlorinated and chlorinated John Sevier
Steam Plant condenser cooling river water
samples analyzed November 4-5, 1980 93
52 Percentage relative standard deviation against mean
total chlorine concentration in micrograms per liter
for experimental recovery by MC method of total
chlorine from raw unchlorinated, chlorinated, and
mixtures of unchlorinated and chlorinated John Sevier
Steam Plant condenser cooling river water samples
analyzed November 4-5, 1980 94
53 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by AT method of total chlorine from
raw unchlorinated, chlorinated, and mixtures of
unchlorinated and chlorinated John Sevier Steam
Plant condenser cooling river water samples
analyzed November 4-5, 1980 95
54 Percentage relative standard deviation against mean
total chlorine concentration in micrograms per liter
for experimental recovery by AT method of total
chlorine from raw unchlorinated, chlorinated, and
mixtures of unchlorinated and chlorinated John
Sevier Steam Plant condenser cooling river water
samples analyzed November 4-5, 1980 96
xv
-------
FIGURES
(Continued)
Number
55 Total chlorine standard deviation in tnicroorams
per liter against mean total chlorine concentra-
tion Ln micrograms per liter for experimental
recovery by MC method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of
unchlorinated and chlorinated John Sevier Steam
Plant condenser cooling river water samples
analyzed November 4-5, 1980, spiked with calcium
hypochlorite 97
56 Percentage relative standard deviation against mean
total chlorine concentration in micrograms per liter
for experimental recovery by MC method of total
chlorine from raw unchlorinated, chlorinated, and
mixtures of unchlorinated and chlorinated John Sevier
Steam Plant condenser cooling river water samples
analyzed November 4-5, 1980, spiked with calcium
hypochlorite 98
57 Total chlorine standard deviation in micrograms
per liter against mean total chlorine concentra-
tion in micrograms per liter for experimental
recovery by AT method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of
unchlorinated and chlorinated John Sevier Steam
Plant condenser cooling river water samples
analyzed November 4-5, 1980, spiked with calcium
hypochlorite 99
58 Percentage relative standard deviation against mean
total chlorine concentration in micrgrams per liter
for experimental recovery by AT method of total
chlorine from raw unchlorinated, chlorinated, and
mixtures of unchlorinated and chlorinated John Sevier
Steam Plant condenser cooling river water samples
analyzed November 4-5, 1980, spiked with calcium
hypochlorite 100
XV]
-------
TABLES
Number
Spiking experiments with chloramine-T or calcium
hypochlorite for precision and accuracy tests of
the MC method 102
Spiking experiments with chloramine-T or calcium
hypochlorite for precision and accuracy tests of the
AT method t • • ' 103
Spiking experiments with 50% (w/w) mixture of
chloramine-T and calcium hypochlorite as chlorine for
precision and accuracy tests of the MC method .... 104
Spiking experiments with 50% (w/w) mixture of
chloramine-T and calcium hypochlorite as chlorine for
precision and accuracy tests of the AT method .... 105
Preparation of solutions for tests of chemical
interference with the MC method 106
AT method recovery of total chlorine from deionized
water spiked with 249 MS/1 °f chloramine-T as chlorine
with 0.00564-N and 0.001128-N phenylarsine oxide . . . 107
Experimental recovery by MC and AT methods of
total chlorine from deionized water spiked with
chloramine-T 108
Effects of pH, chemical interference, and
pyrophosphate (for overcoming chemical inter-
ference) on experimental recovery by MC method
of total chlorine from deionized water spiked
with 50 |Jg/l of chloramine-T as chlorine 115
Effect of 2% (w/v) sodium pyrophosphate for
overcoming chemical interference on experi-
mental recovery by MC method of total chlorine
from deionized water spiked with chloramine-T
as chlorine 124
xv 11
-------
TABLES
(Continued)
Number
10 Effect of 2% (w/v) sodium pyrophosphate for
overcoming chemical interference on experi-
mental recovery by MC method of total chlorine
from deionized water spiked with calcium
hypochlorite as chlorine 127
11 Experimental recovery by MC and AT methods of
total chlorine from deionized water spiked with
calcium hypochlorite as chlorine 130
12 Experimental recovery by MC and AT methods of
total chlorine from deionized water spiked with
50% (w/w) mixture of calcium hypochlorite and
chloramine-T as chlorine 135
13 Chlorine demand of raw Ross Landing surface
river water sample, collected October 3, 1980 .... 140
14 Inorganic species concentrations, sanitary
chemical characteristics, an.d physical properties
of Tennessee river water collected at Ross Landing,
Chattanooga, Tennessee, October 3, 1980 141
15 Experimental recovery by MC and AT methods of
total chlorine from raw Ross Landing surface
river water sample collected October 3, 1980,
spiked with chloramine-T 142
16 Experimental recovery by MC and AT methods of
total chlorine from raw Ross Landing surface
river water sample collected October 3, 1980,
spiked with calcium hypochlorite 146
17 Chlorine demand of raw Kings;ton Steam Plant sample
composed of 100% (v/v) unchlorinated raw river water
intake at water supply plant, collected October 15,
1980 149
18 Inorganic species concentrations, sanitary chemical
characteristics, and physical properties of raw
Kingston Steam Plant sample composed of 100% (v/v)
unchlorinated raw river water intake at water supply
plant, collected October 15, 1980 150
xvi 11
-------
TABLES
(Continued)
Number
19 Experimental recovery by MC and AT methods of total
chlorine from unspiked, chlorinated Kingston
Steam Plant condenser cooling river water samples
and ones spiked with calcium hypochlorite 151
20 Chlorine demand of raw Shawnee Steam Plant sample
composed of 100% (v/v) unchlorinated unit Number 5
condenser discharge outlet water, collected
October 21, 1980 155
21 Inorganic species concentrations, sanitary
chemical characteristics, and physical pro-
perties of raw Shawnee Steam Plant sample
composed of 100% (v/v) unchlorinated unit
Number 5 condenser discharge outlet water,
collected October 21, 1980 156
22 Inorganic species concentrations of settled materials
sampled October 22, 1980, from raw Shawnee Steam Plant
sample composed of 100% (v/v) unchlorinated unit
Number 5 condenser discharge outlet water, collected
October 21, 1980 157
23 Experimental recovery by MC and AT methods of
total chlorine from unspiked, chlorinated Shawnee
Steam Plant condenser cooling river water samples
and ones spiked with calcium hypochlorite 158
24 Chlorine demand of raw Allen Steam Plant sample
composed of 100% (v/v) unchlorinated unit Number 3
condenser discharge inlet water collected
October 28, 1980 162
25 Inorganic species concentrations, sanitary
chemical characteristics, and physical pro-
perties of raw Allen Steam Plant sample
composed of 100% (v/v) unchlorinated unit
Number 3 condenser discharge inlet water,
collected October 28, 1980 163
xix
-------
TABLES
(Continued)
Number
26 Experimental recovery by MC and AT methods of
total chlorine from unspiked, chlorinated Allen
Steam Plant condenser cooling river water samples
and ones spiked with calcium hypochlorite 164
27 Chlorine demand of raw John Sevier Steam Plant
sample composed of 100% (v/v) unchlorinated unit
Number 1 condenser discharge inlet water collected
November 4, 1980 169
28 Inorganic species concentrations, sanitary chemical
characteristics, and physical properties of raw
John Sevier Steam Plant sample composed of 100% (v/v)
unchlorinated unit Number 1 condenser discharge inlet
water, collected November 4, 1980 170
29 Experimental recovery by MC and AT methods of total
chlorine from unspiked, chlorinated John Sevier
Steam Plant condenser cooling river water samples
and ones spiked with calcium hypochlorite 171
xx
-------
LIST OF ABBREVIATIONS AND SYMBOLS
ABBREVIATIONS
ASTM --American Society for Testing and Materials
AT —forward amperometric titration method
C —degrees Celsius (centigrade), equals degrees Kelvin minus 273
EMSL —Environmental Monitoring and Support Laboratory
EPA --U.S. Environmental Protection Agency
FAS —ferrous ammonium sulfate
g —gram
h --hour
3
1 --liter, equals 0.001 m
LCL —lower 95% confidence limit
m —meter
_o
m- --milli-, xlO (as a prefix, e.g., mm)
M --molar, mole per liter
MC --microprocessor-controlled ion selective electrode
min --minute
MNDL —minimum detection limit
N --normal, equivalent per liter
NBS --National Bureau of Standards
PAO —phenylarsine oxide
RSD --relative standard deviation
xxi
-------
LIST OF ABBREVIATIONS AND SYMBOLS
(Continued)
s --second
SD --standard deviation
t —student t statistic
TVA --Tennessee Valley Authority
UCL --upper 95% confidence limit
SYMBOLS
e —electron with negative charge of one
|J- --micro-, xlO (as a prefix, e.g., (Jm)
% --percent
xx 11
-------
ACKNOWLEDGEMENTS
This work was conducted as part of the Federal Interagency Energy/
Environment Research and Development Program with funds administered
through the Environmental Protection Agency (EPA Contract No. D5-E721,
TVA Contract No. TV-41967A).
The EPA Project Officer for this research is James T. Stemmle, Office
of Environmental Processes and Effects Research, U.S. Environmental
Protection Agency, Washington, DC, and the TVA Project Director is
C. Wayne Holley, Tennessee Valley Authority, Laboratory Branch,
Chattanooga, Tennessee. Their contributions to the direction of
the research and constructive review of the reported results are
gratefully acknowledged.
The authors also appreciate technical review and direction from
Hollis B. Flora, II, Energy Demonstrations and Technology Division,
and Carl Cain, Jr., Fossil and Hydro Power Division, Tennessee
Valley Authority, Chattanooga, Tennessee, and Gerald D. McKee and
Daniel F. Bender, Environmental Monitoring and Support Laboratory,
U.S. Environmental Protection Agency, Cincinnati, Ohio; technical
assistance and loan of the microprocessor-controlled ion selective
electrode equipment from Lester P. Rigdon, Lawrence Livermore
Laboratory, Livermore, California; loan of amperometric titrator
equipment from John W. Shipp, Jr., Water Resources Division,
Tennessee Valley Authority, Chattanooga, Tennessee.
XXlll
-------
-------
SECTION 1
INTRODUCTION
Because numerous studies have documented the toxicity of low levels
of chlorine and its byproducts for aquatic organisms and because current
technology makes it possible to reduce substantially the amount of chlorine
used to control biological organisms that foul once-through cooling water
systems, the U.S. Environmental Protection Agency (EPA) recently proposed
that total chlorine discharges to surface waters be limited to 0.14 mg/1
(1). A precise and accurate method of measuring very low concentrations
of total chlorine is needed both for routine monitoring of cooling-tower
operations and for assessing compliance with this proposed discharge limit.
Among the methods now recommended by EPA for monitoring total chlor-
ine in all types of waters and wastes that do not contain appreciable
quantities of organic matter is the forward amperometric titration (AT)
method (2), which is identical to the ASTM D1253-76-A direct amperometric
titration method (3). In recent collaborative tests, however, the aver-
age operator precision (standard deviation) of the AT method was about
0.03 mg/1 total chlorine (4,5). At concentrations of about 0.02 mg/1
total chlorine, determinations were hampered by lack of highly precise
conditions for the equipment used in the analysis (5).
One promising alternative to the AT method is the microprocessor-
controlled ion selective electrode (MC) method. During development of this
method (6,7), experimental determination of total chlorine in "spiked" stan-
dards with combined chlorine (deionized water with known additions of chlor-
ine using chloramine T), public water supply samples, and swimming pool
water samples indicated the minimum detection limit of the MC method to be
about 0.002 mg/1 (values of 0.003 and 0.004 mg/1 were actually reported),
at least for these applications. These studies also indicated that
0.1 mg/1 was the highest concentration that could practically be deter-
mined. These earlier experiments did not test the MC method with "spiked"
standards of calcium hypochlorite [a free chlorine source used as a biocide
(8)], compare the results of the AT and MC methods, or test the MC method
for determining total chlorine in untreated river water or chlorinated river
water used for condenser cooling, which could contain other chemicals that
might interfere with the accuracy and precision of measurements or degrade
the detection limit.
-------
However, if the MC method could provide the same quality of results
for samples of river water and cooling water, the 10-fold improvement in
sensitivity [compared to the estimated minimum detection limit of 0.02 mg/1
for the AT method that can be calculated from minimum amount (0.02 ml) of
0.00564-N phenylarsineoxide (PAO) titrant dispensed when titrating a 200 ml
sample by the standard method (2-4,9)] would make it a useful method for
monitoring cooling water treated with the minimum levels of chlorine (needed
to produce enough free chlorine to control biofouling) and for assessing
compliance with the proposed limit for chlorine discharges to surface waters.
The MC method indicates chlorine concentrations by measuring the
iodine that results from the complete reaction of chlorine with an excess
of potassium iodide in an acetate buffer at pH 4.0 (the same chemistry used
for the AT method). The Orion Research Model 97-70 total chlorine electrode
used for the MC method (10), contains a platinum element that develops an
electromotive force (potential) in proportion to the iodine that is produced
(the potential developed depends on relative levels of iodine and iodide in
solution) and an iodide-sensing element that develops a similar potential in
proportion to the excess iodide that remains in solution following the
reaction. By measuring the difference between these potentials, the MC
method determines the iodine concentration, which (because the reaction
proceeds to completion) equals the original concentration of chlorine.
This electrode used without a microprocessor, has a useful range up to
20 mg/1.
After the reagents have been added to the sample, the microprocessor
automatically
(a) controls a buret to make a series of additions of known, minute volumes
of a standard chlorine solution (the number of additions is determined
by the operator),
(b) computes the Gran function (G.) for each electromotive force value
sensed by the ion selective electrode after each volume addition (i)
of the standard chlorine solution,
(c) estimates (by algebraic solutions of the slope and intercept values for
least-squares analyses of the G. and i values) the most nearly linear
fit for a subset of 35 volume additions and calculates from this a
total chlorine concentration for that subset,
(d) drops the first five data points of the first subset and adds five new
data points to form a second subset,
(e) determines the difference between the chlorine concentration values
calculated for that pair of subsets,
(f) continues in this manner until all volume additions have been analyzed
in overlapping subsets and the differences between concentrations for
each pair of subsets have been determined,
-------
(g) determines which pair of subsets has the least difference between the
total chlorine concentrations calculated for its first and second
subsets, and
(h) displays the value calculated for the first subset of the selected
pair as the best indication of the total chlorine concentration of the
sample (6,7,1.1,12).
Because the microprocessor combined with the Orion 97-70 electrode
lowers the minimum detection limit from 0.01 to 0.002 mg/1 chlorine and
reduces the maximum detection limit from 20 to 0.1 mg/1, the effective
range for quantification by the MC method is 0.002 to 0.1 mg/1 (6,7,10).
The addition of a small amount of sodium hydroxide to the potassium iodide
reagent for the MC method eliminates need for a blank to obtain a linear
curve for low concentrations (6,10).
Chemical contaminants present in a sample could interfere with chlorine
determinations by the MC method. These include (a) those that alter the pH
of the solution; (b) those that oxidiz,e potassium iodide at pH 4.0, just as
total chlorine does, and produce an extraneous source of iodine (2,3,10);
(c) those that inhibit or poison the sensitivity of the platinum iodine-
sensing element of the electrode or otherwise affect its operation (10);
and (d) those $hat complex with iodine:, which is predominantly present as
tri-iodide, T3 , in solutions containing excess iodide (13), to form
chelated compounds that either cannot be sensed by the platinum element
or that alter the potential it develops. River water and condenser cooling
water (14) may contain oxidizing chemicals such as chromium(VI), iron(IIl),
and manganese(VII); platinum-sensing poisons such as mercury(II) and
arsenic(III); or chelating chemicals jiuch as zinc(II) and copper(II).
Although the same iodimetric chemistry at pH 4.0 is used for both the
MC and AT methods, there is little information available from investigations
of the AT method that can be used to assess the effects of chemical inter-
ference on the MC method. Manabe (15) found that the interference with the
AT method caused by iron(III) and copper(ll) could be eliminated by
pyrophosphate.
The purpose of this work, therefore, was to assess the accuracy, pre-
cision, and detection limit of the microprocessor-controlled ion selective
electrode (MC) method for use in routine measurements of microgram-per-liter
concentrations of total chlorine in the chlorinated river water used for
condenser cooling at coal-burning electric power plants. The research
included comparative laboratory and field tests of the MC and AT methods
for determining total chlorine concentrations in both simulated and actual
samples of cooling water and tests of the interference caused by chemicals
typically found in river water and cooling water.
The laboratory tests evaluated
(a) the maximum detection limit (the highest concentration of total chlo-
rine that could practically be determined by the MC method); the
-------
effects of interference by pH, cbromium(VI), mercury(II), bromide,
zinc(II), copper(ll), iron(III), arsenic(III), and manganese(VII) ;
and the effects of pyrophosphate in overcoming interference by
zinc(II), copper(II), and iron(III) in deionized water spiked by
adding accurately known quantities of chlorine from standard solu-
tions of chloramine-T or calcium hypochlorite;
(b) the precision, accuracy, and minimum detection limit (the lowest con-
centration that could practically be determined by the MC method) for
determining total chlorine in deionized water and river water spiked
by adding accurately known quantities of chlorine from standard solu-
tions of chloramine-T, calcium hypochlorite, or both; and
(c) the comparative precisions, accuracies, and minimum detection limits
of the MC and AT methods for determining total chlorine in deionized
water and river water spiked by adding accurately known quantities of
chlorine from standard solutions of chloramine-T, calcium hypochlorite,
or both.
The field tests compared the precisions, accuracies, and minimum detec-
tion limits of the MC and AT methods for determining total chlorine concen-
trations in
(a) samples of chlorinated condenser cooling water collected at four TVA
coal-burning electric power plants, and
(b) samples of chlorinated condenser cooling water collected at four TVA
coal-burning power plants and spiked by adding accurately known quanti-
ties of chlorine from standard solutions of calcium hypochlorite.
-------
SECTION 2
CONCLUSIONS
A microprocessor-controlled ion selective electrode (MC) method has
been evaluated and compared to the forward amperometric titration (AT)
method for determining total chlorine in spiked and unspiked condenser
cooling river water.
The highest concentration of total chlorine that can be practically
determined with suitable precision and accuracy by the MC method is
100 M8/1 chlorine. The pH in the 3.6-4.4 range, chromium(VI) at 500 |Jg/l,
mercury(II) at 10 M8/l> and bromide at 100 (JgA do not affect recovery of
total chlorine. A concentration of 990 pg/1 of zinc(Il), 500 Hg/1 of
copper(II), and 5 mg/1 of iron(IIl) do not appreciably affect recovery of
total chlorine. However, a concentration of 10 mg/1 of zinc(II), 5 mg/1
of copper(II),, and 9.9 mg/1 of iron(lll) do interfere. Concentrations of
500, 200, and 50 (Jg/1 of arsenic(III) interfere, whereas concentrations of
200 and 50 pg/1 of arsenic(V) do not. A concentration of 990 (Jg/1 of
manganese(VIl) also interferes.
For the MC method, for concentrations from 2 to 20 M8 A chlorine in
unspiked, chlorinated condenser water, the total chlorine overall pooled
standard deviation is 2.1 pg/l and the overall mean percentage relative
standard deviation is 20.8. For concentrations from 20 to 200 |Jg /I
chlorine in unspiked, chlorinated condenser water, the total chlorine
overall pooled standard deviation of 6.3 Mg/1 and overall mean percentage
relative standard deviation of 7.7 by the MC method is lower than the total
chlorine overall standard deviation of 8.7 M8/1 an(i overall mean percentage
relative standard deviation of 12.9 by the AT method. The overall, pooled
minimum detection limit (MNDL) of 6.5 M8 A chlorine by the MC method in
unspiked condenser water is lower than that of 17.8 M8 A chlorine by the
AT method.
The Student t value -0.091 with mean difference of -0.1 M8 A chlorine
and standard deviation of 13.0 Mg A chlorine for 112 split samples of
unspiked, chlorinated condenser water analyzed for total chlorine by the MC
and AT methods for concentrations from 20 to 200 MS A chlorine shows there
is no reason to conclude the mean of the differences between the MC and AT
methods differs from zero at 95 percent significance.
For the MC method, for concentrations from 2 to 20 Mg /I chlorine,
the total chlorine overall standard deviations of 1.9 and 2.1 M8A are
similar for spiked and unspiked condenser water. For concentrations from
-------
20 to 200 (Jg/1 chlorine in spiked, chlorinated condenser cooling river
water, the total chlorine overall pooled standard deviation of 8.7 pg/1
and the overall mean percentage relative standard deviation of 10.4 by
the MC method is lower than the total chlorine overall pooled standard
deviation of 10.2 |Jg/l and overall mean percentage relative standard
deviation of 19.6 by the AT method. For spiked condenser water, the
overall MNDL of 13.3 |.!g /I chlorine by the MC method is lower than that
of 22.2 |Jg/l chlorine by the AT method.
-------
SECTION 3
RECOMMENDATIONS
The microprocessor-controlled ion selective electrode (MC) method
described in this report has equivalent or better sensitivity, precision,
and accuracy than the forward amperometric titration (AT) method for deter-
mining total chlorine in condenser cooling river water at coal-burning
electric power plants. Therefore, the authors recommend its use for deter-
mining chlorine at concentrations from 7 to 100 |Jg/l in chlorinated con-
denser cooling river water. Higher concentrations of chlorine can be
determined by dilution.
In its present state of development, the MC method has three main
limitations which would impair the practical use of this instrument:
a. the size of equipment and power requirements restrict the mobility
required in power plant chlorine monitoring;
b. the time required to assay a sample (3-4 minutes typically)
restricts the number of sample analyses possible during the
chlorination periods (typically 20-30 minutes);
c. free chlorine, not total chlorine, is the main factor to maintain
condenser efficiency.
Because of these limitations, the amperometric titration (AT) method
is sufficient in both ease of use and sensitivity to meet power plant needs;
however, the MC method could be utilized in special studies of a nonroutine
nature for which the AT method is not suited.
-------
SECTION 4
EXPERIMENTAL
SAMPLE PREPARATION
Synthetic Samples
For tests of the precision and accuracy of the MC and AT methods,
aliquots of deionized water, river water, and chlorinated river water used
for condenser cooling at the coal-burning electric power plants tested were
spiked by chlorinating sequential portions with accurately known quantities
of chlorine from standard solutions of chloramine-T, calcium hypochlorite,
or both to yield the expected concentrations shown in Tables 1-4.
For tests of chemical interference with the MC method, deionized
water aliquots were spiked individually with chromium(Vl), mercury(II),
bromide, zinc(II), copper(II), iron(III), arsenic(III), arsenic(V), and
manganese(VII) to give the final concentrations shown in Table 5. The
10-mg/l concentration of zinc(II), the 5- and 500-mg/l concentrations of
copper(II), and the 5-mg/l concentration of iron(IH) for the interference
studies were prepared by diluting the 1000-mg/l standard solution of the
respective element with deionized water, while the 1000-mg/l concentrations
of zinc(II) and iron(III) were drawn from the 1000-mg/l standard solution
of the respective element. The 2 percent (w/v) solution of sodium pyro-
phosphate decahydate salt for overcoming interference by copper(Il), iron
(III), and zinc(II) was prepared by dosing a 50-ml aliquot of test solution
with 1 g of the salt dispensed by adding 2 scoops from a calibrated measur-
ing spoon (Catalogue Number 907, Hach Chemical Company, Ames, Iowa).
The pH 3.6, 3.8, 4.0, 4.2, and 4.4 buffer solutions for the inter-
ference tests were prepared by dispensing 10 ml of deionized water and
55.9 ml of glacial acetic acid in a series of five 150-ml pyrex beakers,
slowly adding 10-M sodium hydroxide, mixing the contents to adjust solu-
tions sequentially in the beakers to pH values of 3.6, 3.8, 4.0, 4.2, and
4.4, and diluting each solution to 100 ml. The 10 M sodium hydroxide used
for adjusting the pH was prepared by dissolving sodium hydroxide pellets in
deionized water and diluting the solution to 1000 ml (CAUTION! Wear gloves
and safety goggles. Sodium hydroxide attacks the skin and blinds).
The synthetic samples of mercury(II), zinc(II), and copper(ll) were
drawn from 1000-mg/l certified atomic emission standard solutions (Spex
Industries Incorporated, Metuchen, New Jersey) and diluted with deionized
-------
water to form the respective 50-, 100-, and 50-mg/l stock spiking solu-
tions shown in Table 5. The 50-mg/l concentration of chromium(Vl) shown
in Table 5 was prepared gravimetrically from potassium dichromate
(0.1414 g in 1000 ml of deionized water); weights were checked against
reference weights traceable to the National Bureau of Standards (NBS).
The 50-mg/l concentration of bromide was prepared by dissolving exactly
0.0745 g of potassium bromide (previously dried for 6 h at 150 C) in
deionized water and diluting the solution to 1000 ml in a volumetric flask.
The 1000-mg/l concentration of iron(III) was prepared by dissolving 8.634 g
of ferric ammonium sulfate dodecahydrate in deionized water and diluting
the solution to 1000 ml in a volumetric flask.
The 50-mg/l concentration of arsenic(IIl) stock spiking solution
shown in Table 5 was prepared by diluting the 1000-mg/l concentration of
arsenic(III). The latter concentration was prepared by weighing 4.0 g of
sodium hydroxide pellets in a 150-ml pyrex beaker (CAUTION! Wear safety
goggles. Sodium hydroxide attacks the skin and blinds), weighing 1.320 g
of arsenic(III) oxide in a metal pan, and quantitatively transferring the
arsenic(III) oxide to the beaker with about 25 ml of deionized water. When
the solid had dissolved completely with the aid of stirring, the solution
was transferred quantitatively to a one-liter volumetric flask containing
a few hundred milliliters of deionized water and diluted to 1000 ml. The
50-mg/l and 1000-g/l concentrations of arsenic(V) were prepared similarly
except that 1.534 g of arsenic(V) oxide were used.
The 100-mg/l concentration of manganese(VIl) stock spiking solution
shown in Table 5 was prepared by diluting potassium permanganate that had
been standardized against sodium oxalate (16). Because the equivalent
weight of manganese(VII) is 10.99, the normality (0.0525) of standard per-
manganate multiplied by 10990 gives the concentration of manganese(VII) in
the stock solution (577 mg/1) that was diluted to yield the stock spiking
solution shown in Table 5.
EQUIPMENT
Total chlorine was measured by the MC method (6,7) with equipment
designed and fabricated by Lawrence Livermore Laboratory, Livermore,
California (17) and by the forward amperometric titration method (2)
with a Model 17T1010 amperometric titrator obtained from Fisher and
Porter Company, Warminster, Pennsylvania (9).
PREPARATION OF SOLUTIONS FOR CONDUCTING TESTS
Reagent-grade chemicals were used to prepare all solutions.
The titer of the 0.0375-N phenylarsine oxide (PAO) reagent (Hach
Chemical Company, Ames, Iowa) was determined by standardization against
0.05000-N potassium biiodate, while that of the 0.00564-N PAO (Fisher and
-------
Porter Company, Warminster, Pennsylvania) was verified against 0.00500-N
potassium biiodate (2). Also the strength of the 0.001128-N PAO, which was
prepared by diluting the 0.00564-N PAO. was checked against 0.00500-N
potassium biiodate (2). Results found agreed with those certified.
The 0.0282-N ferrous ammonium sulfate (FAS) reagent was prepared by
dissolving 1.1,06 g of ferrous ammonium sulfate hexahydrate in deionized
water containing 1 ml of a mixture of one volume of sulfuric acid in three
volumes of water (CAUTION! Wear gloves and goggles. Sulfuric acid causes
severe burns. It must be added to water slowly. Heat is evolved during
this addition), diluting the solution to 1000 ml, and standardizing it
against 0.0705-N potassium dichromate that had been prepared by dissolving
3.4567 g in 1000 ml of deionized water (16). Results found agreed with
those expected.
The standard 0.0375-N PAO and 0.0282-N FAS reagents were used to
assay for total chlorine (2,16) in the 2000-mg/l chlorine solutions of
chloramine-T and calcium hypochlorite to establish the true concentration
of chlorine. Assay results for titration with 0.0282-N FAS were 5 to 15
percent lower than those for titration with 0.0375-N PAO. Because the
end point in the FAS titration (16) was obscured by turbidity, the assay
with FAS was abandoned in favor of assay with PAO (2).
The 2000-mg/l solution of chloramine-T as chlorine was prepared by
dissolving 7.945 g of sodium N-chloro-p-toluenesulfonamide trihydate
(chloramine-T trihydrate, Catalogue Number 6-E494, J. T. Baker Chemical
Company, Phillipsburg, New Jersey), which has an equivalent weight of
140.8, equal to one-half of its molecular weight of 281.69 as shown by
half-reaction in equation 1, in deionized water and diluting the
solution to 1000 ml in a volumetric flask:
CH0C,H,S00NCl"+ H++ 2 e" -> Cl~+ CH0C>H,S00NH~. (1)
J o q- 2 3642
This solution was stored in a borosilicate glass bottle that had been
sealed with a cone-shaped polyethylene-lined screw-cap and wrapped in
aluminum foil to protect from decomposition by sunlight. Assay results
with 0.0375-N PAO (2) verified that this solution contained 2000 mg/1
of total chlorine. In preparing a liter of the 2000-mg/l solution of
chloramine-T as chlorine, we used 7.945 g of sodium N-chloro-p-toluene-
sulfonamide trihydrate instead of the 15.882 g specified by Rigdon (6).
Rigdon actually used a 4000-mg/l solution of chloramine-T as chlorine
for dilution to 4000 pg/1 to calibrate for the MC method rather than the
2000-|jg/l solution he thought he was using.
The 2000-mg/l solution of calcium hypochlorite as chlorine was prepared
by dissolving 4.066 g of calcium hypochlorite (Catalogue Number C-100,
Fisher Scientific Company, Fair Lawn, New Jersey), which has an equivalent
weight of 35.74, equal to one-fourth its molecular weight of 142.98 as shown
by half-reaction in equation 2, in deionized water and diluting the solu-
tion to 1000 ml in a volumetric flask:
10
-------
2 OC1~+ 4 H++ 4 e" -> 2 Cl"+ 2 H20. (2)
This solution was filtered through a series of 0.45-|Jm membrane filters
(Millipore Corporation, Bedford, Massachusetts), and the filtrate was
stored in a borosilicate glass bottle that had been sealed with a cone-
shaped, polyethylene-lined screw-cap and wrapped in aluminum foil to
protect from decomposition by sunlight. Assay results with 0.0375-N PAD
(2) verified that this solution contained 2055 mg/1 of total chlorine.
Also, our total chlorine assay result of 49.6 percent (w/w) for calcium
hypochlorite in the filtrate by titration with 0.0375-N PAO (2) did not
agree with that of 74.3 percent (w/w) for the raw powder certified by
Fisher Scientific Company.
For the MC method (12,13), the 10 percent (w/v) potassium iodide
with 0.01-M sodium hydroxide was prepared by mixing 10 ml of 1-M sodium
hydroxide in about 750 ml of deionized water, mixing in 100 g of potassium
iodide until it dissolved, and diluting the solution to 1000 ml. The 1-M
sodium hydroxide was prepared by dissolving 4.0 g of the salt in deionized
water and diluting the solution to 100 ml (CAUTION! Wear gloves and safety
goggles. Sodium hydroxide attacks the skin and blinds). This reagent con-
tained 0.01-M sodium hydroxide, as recommended orally by Lester P. Rigdon,
Lawrence Livermore Laboratory, Livermore, California, instead of the 0.025-M
sodium hydroxide reported in the literature (6). The acetate buffer solu-
tion with pH 4.0 for total chlorine determinations by the MC method (6,7)
was prepared by dissolving 243 g of sodium acetate trihydrate in about
100 ml of deionized water, stirring in 480 g (455 ml) of glacial acetic
acid, and diluting the solution with water to 1000 ml.
For the AT method (2), the pH 4 buffer, the 5 percent (w/v) potassium
iodide solution, bottles with eye droppers for dispensing the buffer and
iodide solutions, and the 0.00564-N (0.2 mg/ml chlorine) PAO reagent were
obtained from Fisher and Porter Company, Warminster, Pennsylvania (9).
DETERMINATION OF TOTAL CHLORINE BY THE MC METHOD
All total chlorine determinations by the MC method were performed
automatically by the microprocessor, operated as directed in the pre-
liminary instruction manual from Lawrence Livermore Laboratory (17),
under the following conditions: (a) the concentration of the standard
solution was 4000 |Jg/l chloramine-T as chlorine; (b) the number of addi-
tions of the standard solution was 120; (c) the volume of each addition
of the standard solution was 0.005 ml; (d) the total time to add the stand-
ard was 2 min; (e) the temperature of the test solution being analyzed was
used; and (f) the volume of test solution was 50 ml (plus the volume of
spiking solution used for spiked samples).
A 4000-pg/l solution of chloramine-T as chlorine was used to calibrate
the MC instead of a 2000-(Jg/l solution, because (as explained earlier) the
directions published by Rigdon (6,17) for preparing this solution yield a
4000-pg/l solution rather than 2000-|Jg/l solution.
11
-------
Assays of samples for total chlorine by the MC method were performed
by (a) adding 50.0 ml of sample to a disposable 50-ml polystyrene beaker
(Catalogue Number B2718-50, Scientific Products, McGaw Park, Illinois);
(b) adding 1 ml of the pH 4 acetate buffer reagent; (c) adding 1 ml of the
10-percent (w/v) potassium iodide in 0.01-M sodium hydroxide reagent; and
(d) initiating the stirring, standard addition assay, and display of results
as given in the instruction manual (17).
Spiked samples for determination of total chlorine by the MC method
were prepared by chlorinating the 50.0-ml sample—after it had been added to
the disposable polystyrene beaker but before the reagents were added—with
accurately known quantities of chlorine dispensed by micropipettes from
standard 12.5-mg/l solutions of chloramine-T, calcium hypochlorite, or both
as chlorine to yield the expected concentrations shown in Tables 1 and 3.
Before the samples were analyzed, a spiked solution of deionized water
containing 50 (Jg/1 of chloramine-T or calcium hypochlorite as chlorine was
prepared as shown in Table 1 to verify that the MC method was functioning
as expected. These spikes verified the accuracy of calibration of the MC
method with the standard solution of 4,000 pg/1 chloramine-T as chlorine,
which was prepared daily by diluting the stock standard (2 g/1). The stock
solution was assayed frequently with PAO to determine if its true concen-
tration had changed. The titer of the 4,000 pg/1 chloramine-T was not
assayed daily before and after conducting experiments.
The 4000-|jg/l solution of chloramine-T as chlorine that was used
to calibrate the MC instrument was prepared daily by dispensing with
a micropipette the required volume of exactly 2 ml of 2000-mg/l stock
standard of chloramine-T as chlorine and then diluting to the mark
with deionized water in a 1000 ml volumetric flask. The solutions
of chloramine-T and calcium hypochlorite that were used for spiking
experiments were prepared daily by (a) delivering with micropipettes
the required volume of the stock standard (typically 2.50 ml for 2000
mg/1 chloramine-T as chlorine and 2.43 ml for 2055 mg/1 calcium hypo-
chlorite as chlorine) and then diluting to the mark with deionized water
in a 100 ml volumetric flask to yield a 50-mg/l chlorine solution and
(b) diluting 25 ml of the 50-mg/l chlorine solution to 100 ml to yield
the 12.5-mg/l chlorine solution.
DETERMINATION OF TOTAL CHLORINE BY THE AT METHOD
All determinations by the AT method were performed as directed in the
section of the instruction manual from Fisher and Porter (9) that described
the titration for total chlorine.
Spiked samples for determination of total chlorine by the AT method
were prepared by chlorinating the 200-ml sample--after it had been added
to the plastic sample jar but before the of reagents were added—with
accurately known quantities of chlorine dispensed by micropipettes from
12
-------
standard 50-mg/l solutions of chloramine-T, calcium hypochlorite, or both
as chlorine to yield the expected concentrations shown in Tables 2 and 4.
Before the samples were analyzed, a spiked solution of deionized water
containing 249 M8/1 of chloramine-T or calcium hypochlorite as chlorine
was prepared as shown in Table 2 to verify that the AT method was func-
tioning as expected.
Directions for preparing the 50-mg/l solutions of chloramine-T and
calcium hypochlorite as chlorine are given at the end of the section
describing the "Determination of Total Chlorine by the MC Method."
In addition to forward amperometric titration by 0.00564-N (0.20
mg/ml chlorine) PAO, as described in the instruction manual from Fisher
and Porter (9), titration by 0.001128-N (0.04 mg/ml chlorine) PAO, as
recommended by Gerald D. McKee, EMSL, EPA, Cincinnati, Ohio, was attempted
to determine whether the sensitivity of the method could be enhanced by
using a diluted titrant. Table 6 shows that results that are low by about
15 percent are obtained by titration with 0.001128-N (0.04 mg /ml chlorine)
PAO as compared to those obtained by titration with 0.00564-N (0.20 mg/ml
chlorine) PAO for assays of a deionized water solution spiked with 249 |Jg/l
of chloramine-T as chlorine. Therefore, titrations with 0.001128-N PAO
were abandoned in favor of those with 0.00564-N PAO.
13
-------
SECTION 5
RESULTS AND DISCUSSION
DEIONIZED WATER
Maximum Concentration Limit
Spiked standard total chlorine solutions were prepared as described
in Section 4, Tables 1-4. The concentrations of total chlorine in spiked
deionized water sample shown in Tables 7-12 for experimental recoveries by
MC and AT methods were the average of the final concentrations listed in
Tables 1-2 (for the MC and AT methods) and Tables 3-4 (for the MC and AT
methods). Definitions for mean, standard deviation, relative deviation,
bias (accuracy) (18), and one-sample t test (19) are given in the literature.
The one-sample t test values reported in Tables 7-12 are useful for
assessing the relevance of chemical recovery by applying a t test for deter-
mining the difference between the mean values for total chlorine recoveries
("mean recovered") and the amount of chlorine known to have been added (as
a "spike") for replicate determinations of total chlorine at 95 percent
confidence (t- /\or~ two-tailed for number of replicates minus one). If t
(calculated) is greater than t (from table), then the difference is signi-
ficant. For seven replicates, the t (from table) is 2.45, so t (calculated)
values greater than this are significant (19).
These 95 percent confidence limits were used throughout this report.
Experimental recovery of total chlorine from seven replicate analy-
ses of deionized water spiked with chloramine-T for determination of the
maximum detection limit of the MC method are shown in Table 7, Experi-
ments 7.4-7.10. For concentrations of 5-100 (Jg/1 total chlorine, shown
in Experiments 7.4-7.8, the agreement between spike and mean recovered
is acceptable, although about 2 (Jg/1 is lost for the 5- and 10 Mg/1 spikes
and about 10 |Jg/l is gained for a 100 (Jg/1 spike with a highly significant
calculated t value of 15. The slightly high bias of 9.0 percent by the
MC method from the seven replicate 100-(Jg/l spikes with chloramine-T as
chlorine (Experiment 7.8) compares favorably with that of 9.0 percent from
the seven replicate 100-|Jg/l spikes with calcium hypochlorite (Experiment
11.6) and with that of 2.0 percent from the seven replicate 100-|Jg/l spikes
with a 50 percent (w/w) mixture of calcium hypochlorite and chloramine-T as
chlorine (Experiment 12.5). The maximum concentration limit of 100 (J§/1
agrees with that of 100 |Jg/l reported by Rigdon in his original publication
(6) but is greater than the 75 M8/1 he reported in a subsequent study (7).
14
-------
This value is much smaller than the maximum detection limit of 20,000
|jg/l chlorine for the Orion Research Model 97-70 total chlorine elec-
trode (10) when used without microprocessor calculation of the best
value as is done by the MC method (8). For spikes of 247 and 481 [Jg/1
(Experiments 7.9-7.10), the biases of 23.1 and 48.6 percent are judged
unacceptable. But results of Experiments 7.11-7.13 showed that dilution
(see Table 1 for specific dilutions) of 149, 198, and 247 (JgA spikes,
before addition of iodide and acetate reagents and automated standard
addition assay (17), yielded truly favorable biases of 2.7, 1.0, and
0.8 percent, respectively.
Interferences
The pH of the test solution may affect the completeness of the reaction
between chlorine and iodide, as exemplified by the following reaction:
HOC1 + 2I~ + H+ -* I2 + Cl" + H2<3 (3)
The results of tests conducted at different pH values are shown in Experi-
ments 8.1-8.5 for experimental recovery, by the MC method, of total chlorine
from individual deionized water solutions. These solutions were spiked
sequentially with 50 pg/1 of chloramine-T as chlorine, and the total chlo-
rine was assayed following the addition of potassium iodide and a series,
respectively, of pH 3.6, 3.8, 4.0, 4.2, and 4.4 buffers. As shown by the
complete recovery of total chlorine from these buffers, chemical contami-
nants that affect sample pH, such as sulfuric acid in runoff from aban-
doned strip mines and metallurgical industry (8), will not interfere with
determinations. This assumes the pH of the sample solution falls between
pH 3.6 and 4.4 after the addition of the pH 4.0 buffer used to assay for
total chlorine.
Chromium(VI) may oxidize potassium iodide, just as total chlorine does,
and produce an extraneous source of iodine (2,3,10). Mercury may poison the
response of the platinum iodine-iodide sensing element (10). Bromide may
also affect the chemistry of the reaction between chlorine and iodide by
complex means. However, the results of Experiments 8.6 and 8.8, show that
the experimental recovery by the MC method of total chlorine from deionized
water spiked with chloramine-T as chlorine is unaffected by 500 (Jg/1 of
chromium(VI) (introduced as dichromate), 10 (Jg/1 of mercury(II), and 100
|Jg/l of bromide. These concentrations of chromium(VI) and mercury(II) are
much higher than the total concentrations (sum of all chemical forms) of
each metal expected in ash pond effluent waters from coal-burning electric
plants (14) and are also higher than the concentrations present in the con-
denser cooling river water samples shown in Tables 18, 21, 22, 25, and 28.
The concentration of bromide tested is about the highest found in sulfur
dioxide scrubber solutions, which contain bromide in higher concentrations
than river water. The absence of interference from chromium(VI) and mer-
cury(II) with the MC method confirmed the electrode characteristics of the
Orion Reserch Model 97-70 total chlorine electrode(lO).
15
-------
Experiments 8.10-8.11 show the effect of zinc(II) on experimental
recovery, by the MC method, of total chlorine from deionized water spiked
with 50 |Jg/l of chloramine-T as chlorine.
This total chlorine spike was used in subsequent experiments
unless specified otherwise.
While experimental recovery was unaffected by 990 [Jg/1 zinc(II), as
evidenced by the 2.6 percent bias in Experiment 8.9, 1000 mg/1 zinc(II)
(Experiment 8.10) produced very high results (164 percent bias) with a
highly significant t value of 13, as compared to the two-tailed t_ ,.„,.
test value of 2.447, and 10 mg/1 zinc(II) (Experiment 8.11) gave sligntly
high results (13.6 percent bias) with a highly significant t value of 13.6
compared to that of 2.6 without zinc(II), as shown in Experiment 7.7. Many
condenser cooling river water samples contain small amounts of zinc, as
shown by the results in Tables 18, 21, 22, 25, and 28, but should there
be a 10-fold increase in concentration in recirculated condenser cooling
river water blowdown, the concentration of 1 mg/1 of zinc(II) will not be
exceeded; thus, no interference will be caused by zinc. For samples with
higher concentrations of zinc(II), Experiment 8.12 shows that the spike of
50 pg/1 of chloramine-T as chlorine was completely recovered (0 percent
bias) when 2 percent (w/v) sodium pyrophosphate decahydrate was added to
the sample before the potassium iodide and acetate buffer (pH 4.0) rea-
gents were added. As shown by the stability constants (20), zinc(II)
forms a chelated compound with excess pyrophosphate to stop free zinc(II)
from reacting with tri-iodide, !„ , which predominates in iodine solutions
with excess iodide(13), to form unwanted chelated compounds with 2inc(II)
that cannot be sensed by the platinum iodine-sensing element of the elec-
trode for the MC method. With the AT method, which uses the same iodi-
metric chemistry as does the MC method, Manabe (15) used 0.1 percent
(w/v) pyrophosphate to overcome interference by copper(II) and iron(III),
which are chelating metals similar to zinc(II).
The 6.8 percent bias in Experiment 8.13 shows that 500 M8/1 of
copper(II) slightly affected recovery, by the MC method, of total
chlorine from deionized water (6.8 percent bias). With the possible
exception of the condenser cooling river water from the Kingston Steam
Plant, which contained 120 M8/1 copper (Table 18), most of the samples of
river water used for condenser cooling contained very small amounts of cop-
per (Tables 21, 22, 25, and 28). Thus, should there be a 10-fold increase
in the concentration of copper in recirculated condenser cooling river
water blowdown, the concentration of 500 |Jg/l of copper will not be
reached (with the possible exception of the river water at Kingston
Steam Plant), and no interference will be caused by copper. At con-
centrations of 500 and 5 mg/1 of copper(II) (Experiments 8.14 and
8.15), recovery was biased by 134 and 34 percent, respectively.
Experiment 8.16 shows that the addition of 2 percent (w/v) sodium
pyrophosphate decahydrate to a solution containing 5 mg/1 copper(II)
and 50 M8/1 °f chloramine-T as chlorine, before the iodide and acetate
buffer (pH 4.0) reagents are added to generate iodine, improved total
chlorine recovery to -2.6 percent bias. Our results corroborate those
16
-------
obtained by Manabe (15), who used 0.1 percent (w/v) pyrophosphate
to overcome interference by copper(II) with the AT method, and those
obtained by Orion (10), who warned that copper(II), a strong oxidizing
agent, should be excluded from iodimetric methods but did not specify
what concentration levels of copper(II) may be tolerated without
interference.
The results from Experiments 8.17, 8.19, and 8.21 show, respectively,
that 1000 mg/1, 9900 MgA, and 5 mg/1 of iron(III) cause biases of 2680,
24.6, and 8 percent in the experimental recovery, by the MC method, of total
chlorine from deionized water. Although the results from Experiment 8.21
show that 5 mg/1 of iron(III) produces a significant bias of 8 percent
[for example, compare the t value of 8.1 with that of 2.5 for Experiment
7.7 without iron(III)], the river water TVA uses for condenser cooling
contains only 0.38-1.6 mg/1 iron (Tables 18, 21, 25, and 28). However,
a slurry of settled materials from unchlorinated Shawnee condenser outlet
water, which contained 8 mg/1 of iron (Table 22), produced a small total
chlorine value of about 35 Hg/1 by the MC method (see Experiment 23.1
replicate numbers 6 and 7) as compared to none in the same sample—except
without settled materials (see replicate numbers l-5)--which contained
1.2 mg/1 iron (see Table 21).
The concentration of 5 mg/1 iron will not be exceeded except in slur-
ries remaining after decantation of supernatants from condenser water sam-
ples and in recirculated condenser cooling river water blowdown; therefore,
iron(III) will not cause inteference. For river water containing more
than 5 mg/1 of iron(III), 2 percent (w/v) pyrophosphate decahydrate can be
used to overcome most of the interference from 1000 mg/1 iron(III) and all
interference from 9900 M8/1 iron(III) as shown by biases of 11.8 and -1.8
percent in Experiments 8.18 and 8.20, respectively. It should be noted
that a bias of 3286 percent was noted when the MC method was used to
attempt to recover 50 |Jg/l of chloramine-T as chlorine from 1000 mg/1
iron(III) in deionized water containing 0.4 percent (w/v) sodium pyro-
phosphate decahydrate. With the exception of the 2 percent (w/v) pyro-
phosphate concentration that had to be used to eliminate interference
by iron(III), these results corroborate those obtained by Manabe (15),
who used 0.1 percent (w/v) pyrophosphate to overcome interference from
iron(III) with the AT method.
Pyrophosphate was shown to negate interference by concentrations of
10 mg/1 zinc(II) (Experiment 8.12), 5 mg/1 copper(II) (Experiment 8.16),
and 9.9 mg/1 iron(III) (Experiment 8.20). Further tests were conducted to
determine whether quantitative recovery of total chlorine was affected by
pyrophosphate itself. Experiments 10.7 and 10.8 indicate that pyrophos-
phate was not suitable for overcoming interference by 10 mg/1 zinc(II),
5 mg/1 copper(II), or 9.9 mg/1 iron(III) because it caused biases of -48
and -46 percent for the experimental recovery of total chlorine, by the
MC method, from 2 percent (w/v) pyrophosphate in deionized water spiked
with 149 and 198 (jg/1 of calcium hypochlorite as chlorine. The latter
two test concentrations are very important because EPA has proposed to
17
-------
limit total chlorine discharges to 140 |Jg/l (1). These biases (-48
and -46 percent) are lower than those obtained in the absence of pyro-
phosphate (-1.3 and -1.0 percent, respectively, as shown in Experiments
11.7 and 11.8). These results do not agree with those by Manabe (15),
who observed quantitative recovery by the AT method from 0.1 -percent
(w/v) pyrophosphate in deionized water spiked with 1000 |Jg/l sodium
hypochlorite as chlorine.
The loss of chlorine when 2 percent (w/v) pyrophosphate is spiked
with 149 and 198 pg/1 of calcium hypochlorite as chlorine may be caused by
break-point chlorination products derived from traces of ammonia in the
pyrophosphate reagent (6). This possibility was not investigated, but
this hypothesis is supported by the absence of grossly low biases for pyro-
phosphate spiking experiments as compared to those without pyrophosphate for
0, 5, 10, 25, 50, and 100 Mg/1 of calcium hypochlorite as chlorine (see
Experiments 10.1-10.6 and Experiments 11.1-11.6) and for 0, 5, 10, 25, 50,
100, 149, 198, and 247 MgA °f chloramine-T as chlorine (see Experiments
9.1-9.9 and Experiments 7.1, 7.4-7.8, and 7.11-7.13).
Therefore, pyrophosphate needs further study.
Experiment 8.22 shows that 500 (Jg/1 of arsenic(III) produced erra-
tic, generally high recoveries of total chlorine. (The standard devia-
tion was 114 |Jg/l and the bias was 138 percent by the MC method for
deionized water.) In Experiment 8.23, a consistently high bias of 50
percent was caused by 200 |Jg/l of arsenic(III) , and in Experiment 8.24
somewhat low results (-12.8 percent bias) were caused by 50 |Jg/l of
arsenic(III) . Apparently, arsenic(III) inhibits or poisons the response
of the platinum iodine-sensing element in the Orion Research Model 97-70
total chlorine electrode used for the MC method, although this is not
mentioned by Orion (10). In contrast, Experiments 8.25 and 8.26 show
no interference from 200 pg/1 and 50 |Jg/l arsenic(V) in the experimental
recovery, by the MC method, of total chlorine from deionized water.
Thus, arsenic(III) affected experimental recovery, but arsenic(V) had
no effect. From a table of standard oxidation-reduction potentials (21),
it can be shown, by subtracting appropriate half-reactions, that the
potential is +0.67 volt for the following reaction, and that, because
of the positive potential, the oxidation of arsenic(III) to arsenic(V)
by atmospheric oxygen, which is normally present, proceeds spontaneously:
2 HAs02 + 2 H20 + 02 -> 2 H3As04. (4)
This reaction is also independent of pH, so complete oxidation of arsenic
(III) to arsenic(V) will occur without effect from the pH of the sample.
Because all arsenic will exist as arsenic(V), its interference should be
negligible. Turner (22) found that oxidation of arsenic(III) to arsenic(V)
is rapid in aerated leachate solutions of fly ashes, although the oxidation
rate may be affected by catalysts, such as copper salts and manganese oxide
present in such solutions. Also, chemical analysis of condenser cooling
river water samples, shown in Tables 18, 21, 22, 25, and 28, revealed a
18
-------
maximum total arsenic concentration of only 6 |Jg/l, although samples of
ash pond water (which is not used for condenser cooling in TVA) , may con-
tain total arsenic concentrations as high as 290 |Jg/l (14).
As shown by Experiment 8.27, 990 MgA of manganese (VII) produced
a very erratic, generally low experimental recovery, by the MC method,
of total chlorine from deionized water. Because manganese (VII) is a
very strong oxidizing agent, its existence is unlikely in river water,
although some manganese dioxide could be present and serve as an oxidant
(just as chlorine does) to produce iodine in the iodimetric reaction (10).
The total manganese concentrations in the condenser cooling river water
samples analyzed for this study were 50-670 |Jg/l (Tables 18, 21, 22, 25,
and 28).
Precision, Accuracy, and Minimum Detection Limit
for Spikes with Chloramine-T
Chloramine-T is a source of combined chlorine. It slowly liberates
hypochlorous acid on contact with water (23). It simulates a combined
source of chlorine, such as monochloramine or dichloramine, that exists
in chlorinated condenser cooling river water containing ammonia at coal-
burning electric power plants (8).
Table 7, Experiments 7.1-7.8 and 7.11-7.13, gives the results of
experimental recoveries, by the MC and AT methods, of total chlorine from
seven replicate deionized water solutions spiked with 0, 1.25, 2.5, 5, 10,
25, 50, 100, 149, 198, and 247 pg/1 of chloramine-T as chlorine. No values
below 20 [JgA °f total chlorine could be determined by the AT method because
one drop (0.02 ml) of 0.00564-N PAO from the 1-ml buret (the minimum amount
that could be dispensed during an amperometric titration) represents 20 |Jg/l
of total chlorine when a 200 ml sample is titrated by the standard method
(2-4,9). For experimental recovery of total chlorine from spiked deion-
ized water, by the MC and AT methods, Table 7 shows the total chlorine
spike of chloramine-T as chlorine, the mean total chlorine, the total
chlorine standard deviation, the relative standard deviation, the bias
(accuracy), and a calculated t value to test for significance of total
chlorine recovered (18,19).
For the MC method, Figures 1-3 summarize different combinations of
paired values shown in Table 7: total chlorine standard deviation against
mean total chlorine, relative standard deviation against mean total chlo-
rine, mean bias against total chlorine concentration added. Figures 4-6
show analogous summaries for the AT method.
In chlorine discharge and minimization studies, the usual concentra-
tion range of interest is 0-200 JJgA °f total chlorine (4). In fact,
EPA recently proposed that total chlorine discharges to surface waters be
limited to 140 |Jg/l (1), and recent collaborative tests of the AT method
reported standard deviations that were obtained by averaging values over
19
-------
0-200 |Jg/l (4). In light of these considerations and the limiting measur-
able concentration of 20 |Ug/l of total chlorine by the AT method, standard
deviations were pooled (19), relative standard deviations were averaged,
and biases were averaged for the recoveries by the AT method (Table 7)
over the mean total chlorine concentration range of from 20" to 200 Mg/1.
For comparison with the AT method, the same averages for the MC method
were calculated over mean total chlorine concentrations of 20-200 pg/1;
and because of the reported minimum detection limit of about 2 JJg/1 of
total chlorine (values of 3 and 4 (Jg/1 were actually reported) by the MC
method (6,7), averages were also computed over mean total chlorine con-
centrations from 2 to (but not including) 20 M8/1-
This procedure for averaging values of 2-20 |Jg/l and 20-200 |Jg/l was
used throughout this report.
Experimental recoveries, by the MC method, of total chlorine from
deionized water spiked with chloramine-T as chlorine yielded a total
chlorine pooled standard deviation of 0.4 (Jg/l> a mean percent relative
standard deviation of 10.5 percent, and a mean bias of -21.2 percent for
concentrations of 2-20 pg/1 (Experiments 7.3-7.5) and a total chlorine
pooled standard deviation of 2.9 (Jg/1, a mean relative standard deviation
of 2.1 percent, and a mean bias of 2.2 percent for concentrations of 20-200
(Jg /I (Experiments 7.6-7.8 and 7.11-7.12). The statistical values deter-
mined by the MC method for 20-200 |Jg /I are similar to those reported by
Rigdon (6), but values for 2-20 jjg /I are several times larger. For the
AT method, for concentrations of 20-200 pg /I, the experimental recoveries
with chloramine-T showed a total chlorine pooled standard deviation of
4.0 (Jg/1, a mean relative standard deviation of 4.6 percent, and a mean
bias of 2.6 percent (Experiments 7.6-7.8 and 7.11-7.12). These values
compare favorably with those determined by the MC method (2.9 M8/l> 2.1
percent, and 2.2 percent).
A procedure being developed by EPA (24) allows the minimum detection
limit (MNDL) for the MC method to be calculated from the standard devia-
tions of the data for experimental recovery of very low concentrations of
total chlorine (for example, 1.25, 2.5, 5 and 10 |Jg /I, for the seven
replicate deionized water spikes with chloramine-T shown in Experiments
7.2-7.5). For example, from the standard deviation for the 2.5-|Jg /I
spikes (Experiment 7.3), the minimum detection limit (MNDL) can be
calculated by means of the following equation:
"^^(n-l, 0.99) (SD)'
where, t, qq,. = t value for 99 percent confidence level (one-tailed)
with n-1 degrees of freedom (the number of replicate analyses minus one),
and, SD = standard deviation of the replicate analyses.
For the seven replicate analyses shown in Experiment 7.3, the appro-
priate t value is t, = 3.143 (19,24). Substituting the SD value of
0.38 |Jg/l from Experiment 7.3 and t = 3.143 into equation 5 yields a MNDL
of 1.2 (Jg/1.
20
-------
The 95 percent confidence interval estimate for the 1.2 |Jg/l MNDL can
be estimated from the following equations:
LCL =0.64 MNDL (6)
and
UCL =2.20 MNDL, (7)
where LCL and UCL are the lower and upper 95 percent confidence limits,
respectively, based on seven replicate analyses.
The right-hand members of equations 6 and 7 were derived by multi-
plying the MNDL by the reciprocal of the result obtained by taking the
square-root of the chi-squared distribution values divided by the degrees
of freedom for the 0.975 and 0.025 probabilities, respectively,
MNDL/-V-^p and MNDL/^/i^ (18,25).
Substitution of the 1.2 Mg/1 MNDL into equations (6) and (7) yields
an LCL of 0.77 and a UCL of 2.64. Since the mean total chlorine of 2.23
Mg/1 recovered in Experiment 7.3 falls within the LCL and UCL, there is no
good reason to reject the MNDL of 1.2 MS/I as a good estimate, because when
this occurs, it is assumed that the variability comes from measurements
of total chlorine within the range of concentrations where it is first
detected. Thus, the mean recovery is consistent with the UCL and LCL
of the MNDL. The MNDL of 1.2 MgA for the MC method for deionized water
spiked with chloramine-T is lower than the 3- and 4~Mg/l values that
Rigdon estimated for the MC method by spiking deionized water with
chloramine-T (6,7).
Although an MNDL for the MC method can be calculated, too, for each set
of the seven 1.25, 5, and 10 M8/1 deionized water spikes with chloramine-T
as chlorine shown in Table 7, Experiments 7.2 and 7.4-7.5, none of the mean
total chlorine concentrations recovered falls within the LCL and UCL for
the MNDL. For example, the MNDL calculated from the 1.25 Mg/1 spikes in
Experiment 7.2 yields a mean total chlorine concentration recovered that
falls below the LCL. This means try a higher spike. The MNDL values
calculated from the 5- and 10~Mg/l chlorine spikes in Experiments 7.4
and 7.5 give mean total chlorine concentrations that are above the UCL.
This means try a lower spike. It turns out that the optimum spike for
determining the MNDL was 2.5 M8/1 chlorine (Experiment 7.3), because
this yielded a mean total chlorine concentration within the LCL and UCL.
This procedure for determining the MNDL was used throughout this
report.
It was not possible to make a good estimate of the MNDL for the AT
method from the 25 M8/1 spikes of chloramine-T as chlorine shown in Table 7,
Experiment 7.6, because the mean total chlorine was above the UCL.
21
-------
Precision Accuracy, and Minimum Detection Limit for Spikes
with Calcium Hypochlorite
Calcium hypochlorite, a chemical compound largely used as a source of
hypochlorite, represents the usual source of free chlorine employed for
disinfecting water (8); therefore, it simulates the chlorine source used
as a biocide in condenser cooling river water at coal-burning electric
plants (4).
Table 11, Experiments 11.1-11.9, gives experimental recoveries, by the
MC and AT methods, of total chlorine from seven replicate deionized water
solutions spiked with 0, 5, 10, 25, 50, 100, 149, 198, and 247 [Jg/1 of
calcium hypochlorite as chlorine.
For the MC method, Figures 7-9 summarize different combinations of
paired values shown in Table 11: total chlorine standard deviation against
mean total chlorine, relative standard deviation against mean total chlo-
rine, and mean bias against total chlorine concentration added. Figures
10-12 show similar summaries for the AT method.
The experimental recoveries, by the MC method, of total chlorine from
deionized water spiked with calcium hypochlorite as chlorine (Table 11)
yield a total chlorine pooled standard deviation of 0.5 (Jg/1, a mean rela-
tive standard deviation of 9.8 percent, and a mean bias of -23.8 percent
for concentrations of 2-20 pg/1 (Experiments 11.2-11.3) and a total chlo-
rine pooled standard deviation of 4.7 |Jg/l, a mean relative standard devia-
tion of 4.4 percent, and mean bias of -1.0 percent for concentrations of
20-200 |jg/l (Experiments 11.4-11.8). The statistical values determined
by the MC method for 20-200 |Jg/l chlorine in standards spiked with calcium
hypochlorite are similar to those with. chloramine-T by Rigdon (6), but
values for 2-20 pg/1 are several times larger. Over the concentration
range of 2-200 (Jg/1, the statistical values determined by the MC method
with calcium hypochlorite spikes are similar to those with the chloramine-T
spikes, although the standard and relative standard deviations for the
calcium hypochlorite spikes are slightly higher for concentrations of
20-200 H8/1-
For the AT method over concentrations of 20-200 pg/1, the experimental
recoveries with calcium hypochlorite give a total chlorine pooled standard
deviation of 4.3 pg/1, a mean relative standard deviation of 6.4 percent,
and mean percent bias of -1.0 percent (Experiments 11.4-11.8). These values
compare favorably with those determined by the MC method (4.7 (Jg/l> 4.4
percent, and -1.0 percent, respectively).
The MNDL can be calculated for the MC and AT methods from results
in Table 11 for experimental recovery from deionized water spiked with
calcium hypochlorite. A good estimate of MNDL cannot be calculated for
the MC method because the mean recovery of 3.66 (Jg/1 for the 5-(Jg/l spike
(Experiment 11.2) was above the UCL of the MNDL.
-------
However, two valid estimates of the MNDL for the AT method are 11.9
and 15.1 pg/1 which come from the results of Experiments 11.3 and 11.4.
The standard deviations from Experiments 11.3 and 11.4 can be combined (19)
to yield a pooled MNDL of 11.5 H§A for the AT method with a pooled mean
recovery of 18.9
Precision, Accuracy, and Minimum Detection Limit for Spikes with 50 Percent
(w/w) Mixture of Calcium Hypochlorite and Chloramine-T as Chlorine
Calcium hypochlorite simulates a source of free chlorine used as a
biocide in condenser cooling river water (4,8), and chloramine-T
simulates a source of combined chlorine, such as monochloramine, which may
be present in chlorinated condenser cooling water containing ammonia (8).
Therefore, the mixture of both calcium hypochlorite and chloramine-T used
for spiking provides a synthetic mixture of simulated free and combined
chlorine species found in some chlorinated condenser waters.
Table 12, Experiments 12.1-12.9, gives experimental recoveries by the
MC and AT methods of total chlorine from seven replicate deionized water
solutions spiked with 0, 10, 20, 50, ICO, 149, 198, 247, and 488 (jgA of
a 50 percent (w/w) mixture of calcium hypochlorite and chloramine-T as
chlorine.
For the MC method, Figures 13-15 summarize different combinations of
paired values shown in Table 12: total chlorine standard deviation against
mean total chlorine, relative standard deviation against mean total chlo-
rine, and mean bias against total chlorine concentration added. Figures
16-18 show similar summaries for the AT method.
The experimental recoveries by the MC method of total chlorine from
deionized water spiked with a 50 percent (w/w) mixture of calcium hypo-
chlorite and chloramine-T as chlorine (Table 12) yield a total chlorine
pooled standard deviation of 0.8 pg/1, a mean relative standard deviation
of 4.9 percent, and a mean bias of -5.5 percent for concentrations of
2-20 (Jg/1 chlorine (Experiments 12.2-12.3) and a total chlorine pooled
standard deviation of 10.0 |JgA, a mean relative standard deviation of
6.1 percent, and a mean bias of -2.7 percent for concentrations of 20-200
(jg chlorine/1 (Experiments 12.4-12.6). The statistical values determined
by the MC method for 20-200 (Jg/1 chlorine in standards spiked with a 50
percent (w/w) mixture of calcium hypochlorite and chloramine-T as chlorine
are slightly higher than those with chloramine-T by Rigdon (6), but values
for 2-20 JJg/1 chlorine are several times larger.
For the concentration range of 2-200 |Jg/l chlorine, the statistical
values of pooled standard deviation, mean relative standard deviation,
and mean bias by the MC method with the 50 percent (w/w) spikes of calcium
hypochlorite and chloramine-T as chlorine are similar to those with the
individual spikes of chloramine-T and calcium hypochlorite except that
23
-------
the standard and relative standard deviations for the spiked mixture of
calcium hypochlorite and chloramine-T are somewhat higher for concentra-
tions of 20 and 200 |Jg/l chlorine.
For the AT method, for concentrations of 20-200 (Jg/1 chlorine, the
experimental recoveries with a 50 percent (w/w) mixture of calcium hypo-
chTorite and rhloramine-T as chlorine (Table 12) give a total chlorine
pooled standard deviation of 3.4 Mg/1, a mean relative standard deviation
of 3.4 percent, and a mean bias of -0.8 percent (Experiments 12.4-12.6).
These values determined by the AT method are about one-half those deter-
mined by the MC method (10.0 (Jg/1, 6.1 percent, and -2.7 percent, respec-
tively). For concentrations of 20-200 (Jg/1 chlorine, the statistical
values of pooled standard deviation, mean relative standard deviation,
and mean bias by the AT method with the 50 percent (w/w) spikes of calcium
hypochlorite and chloramine-T as chlorine are about the same as those with
the individual chloramine-T and calcium hypochlorite spikes.
The MNDL for the MC and AT methods can be calculated from results in
Table 12 for experimental recovery from deionized water spiked with a 50
percent (w/w) mixture of calcium hypochlorite and chloramine-T as chlorine.
An accurate estimate of MNDL cannot be calculated for the MC method because
the mean recovery of 9.10 |Jg/l for the 10-|Jg/l spike (Experiment 12.2) was
above the UCL of the MNDL. But a valid estimate of 16.7 (Jg/1 can be calcu-
lated for the MNDL by the AT method from the results of Experiment 12.3,
which showed a mean recovery of 14.3 (Jg/1.
Overall Precision, Accuracy, and Minimum Detection Limit for Spikes
with Chloramine-T, Calcium Hypochlorite, and Both
The pooled standard deviations, mean relative standard deviations,
and mean biases for the MC and AT methods for each of the three spiking
experiments with chloramine-T, calcium hypochlorite, and both in deionized
water can be averaged to yield overall values.
For the MC method, the total chlorine overall pooled standard devia-
tion is 0.5 Hg/1, tne overall mean relative standard deviation is 8.7
percent, and the overall mean bias is -17.5 percent for concentrations
of 2-20 |Jg/l chlorine (Table 7, Experiments 7.3-7.5; Table 11, Experiments
11.2-11.3; and Table 12, Experiments 12.2-12.3), and the total chlorine
overall pooled standard deviation is 5.9 (Jg/l> the overall mean relative
standard deviation is 3.9 percent, and the overall mean bias is -0.2 per-
cent for concentrations of 20-200 (Jg/1 chlorine (Table 7, Experiments
7.6-7.8 and 7.11-7.12; Table 11, Experiments 11.4-11.8; and Table 12,
Experiments 12.4-12.6). The overall statistical values for the MC
method for 20-200 (Jg/1 chlorine in standards spiked with chloramine-T,
calcium hypochlorite, and both are slightly higher than those with
chloramine-T reported by Rigdon (6), but values for 2-20 (Jg/1 chlorine
are several times greater. The overall mean bias of -17.5 percent means
24
-------
that a true concentration of total chlorine cannot be established at con-
centrations of 2-20 pg/1 in deionized water because it is not a conserved
species (some of it is lost by decomposition to chemical species that are
not measured by the MC analytical procedure).
For the AT method, for concentrations of 20-200 |Jg/l, the total chlo-
rine overall pooled standard deviation is 4.0 (Jg/1, the overall mean rela-
tive standard deviation is 5.0 percent, and the overall mean bias is 0.5
percent (Table 7, Experiments 7.6-7.8 and 7.11-7.12; Table 11, Experiments
11.4-11.8; and Table 12, Experiments 12.4-12.6). These values are about
the same as those for the MC method (5.9 M8/l> 3.9 percent, and -0.2 per-
cent, respectively).
The overall pooled MNDL can be calculated for the MC and AT methods
by combining the respective pooled or individual MNDL values from each of
the three spiking experiments with chloramine-T, calcium hypochlorite, and
both. For the MC method, the overall MNDL is 1.2 Mg/1, which is the same
as the MNDL for spiking with chloramine-T in Experiment 7.3, in which the
mean total chlorine of 2.23 M8/1 was recovered. This was the only valid
MNDL generated for the MC method in deionized water media. This MNDL is
lower than the of 3- and 4-pg/l values estimated by Rigdon (6,7) from
spiking experiments with chloramine-T in deionized water. However, a
valid overall pooled MNDL of 12.0 |Jg/l can be calculated for the AT
method from the results of Experiments 11.3, 11.4, and 12.3, which
yield an overall, pooled mean recovery of 17.4 Hg/1.
RIVER WATER
Precision and Minimum Detection Limit
for Spikes with Chloramine-T
Table 13 shows the chlorine demand characteristics of a raw Ross
Landing surface river water sample (chlorinated at particular concen-
trations for specified periods of time) that was used for spiking
experiments with chloramine-T, and Table 14 shows its concentrations
of inorganic species, sanitary chemical characteristics, and physical
properties. Spiking with chloramine-T simulates a combined source of
chlorine, such as monochloramine, that exists in chlorinated condenser
cooling river water containing ammonia at coal-burning electric power
plants (8).
Table 15, Experiments 15.1-15.11, shows experimental recoveries, by
the MC and AT methods, of total chlorine from seven replicate river water
solutions that were spiked with 50, 55, 60, 75, 99, 148, 197, 245, and 295
(jg/1 chloramine-T as chlorine for the MC method and with 50, 55, 60, 75,
100, 150, 199, 249, and 299 pg/1 chloramine-T as chlorine for the AT method.
Comparison of the mean total chlorine recoveries by the MC and AT methods
from river water spiked with the total chlorine concentrations (Table 15)
shows that 10-30 percent of the chlorine in chloramine-T spikes is lost
25
-------
after spiking without delay before analysis. This loss, which is pro-
bably due to chlorine demand, is evident in river water even at concen-
trations between 20 and 200 pg/1 whereas this loss does not occur in
deionized water, as shown by Table 7, Experiments 7.6-7.8 and 7.11-7.12.
Apparently a true concentration of total chlorine in river water cannot
be established by spiking with accurately known amounts of total chlorine
in the form of chloramine-T, because some of it is lost by reaction with
substances in the water to form chemical species that are not measured by
the MC and AT analytical methodology. Therefore, the true recoveries of
the chloramine-T spikes were not evaluated.
For the MC method, Figure 19 summarizes total chlorine standard devia-
tion against mean total chlorine for data taken from Table 15, Experiments
15.2-15.11, and Figure 20 summarizes relative standard deviation against
mean total chlorine concentration. Figures 21-22 similarly summarize the
statistical results for the AT method.
The experimental recoveries, by the MC method, of total chlorine
from river water spiked with chloramine-T (Table 15) yield a total chlorine
pooled standard deviation of 2.3 pg/1 and mean relative standard deviation
of 2.5 percent for concentrations of 20-200 pg/1 (Experiments 15.2-15.8).
Although no data were obtained for concentrations of 2-20 M§A> the total
chlorine pooled standard deviation of 2.3 MS/I an<^ mean relative standard
deviation of 2.5 percent for concentrations of 20-200 jJg/1 in river water
are similar to those found by Rigdon (6,7) in deionized water.
For the AT method, for concentrations of 20-200 |Jg/l, the results in
Table 15 show a total chlorine pooled standard deviation of 4.1 |Jg/l and a
mean relative standard deviation of 7.8 percent (Experiments 15.5-15.8).
These values are about twice those for the MC method (2.3 MS/1 and 2.5
percent, respectively).
An accurate estimate of MNDL in river water cannot be calculated for
the MC method from the data in Table 15 because the mean recovery of 28.0
|Jg/l for the lowest chloramine-T spike of 50 |Jg/l (Experiment 15.2) was
greatly above the UCL of the MNDL. However, three valid estimates of the
MNDL in river water can be calculated for the AT method from the results
of Experiments 15.2, 15.4, and 15.5. The MNDL values are 11.9, 16.7, and
15.1 pg/1 f°r mean recoveries of 11.4, 15.7, and 28.6 |Jg/l, respectively.
These three MNDL values could not be pooled to obtain a valid pooled MNDL
because the pooled mean recovery value was larger than the UCL.
Precision, Accuracy, and Minimum Detection Limit
for Spikes with Calcium Hypochlorite
Table 16, Experiments 16.1-16.8, shows experimental recoveries, by
the MC and AT methods, of total chlorine from seven replicate river water
solutions, characterized by results shown in Tables 13 and 14, that were
26
-------
sequentially spiked with 15, 30, 55, 80, 129, 202, 344, and 383 [ig/l cal-
cium hypochlorite as chlorine for the MC method and with 25, 50, 90, 125,
180, 249, 304, and 434 |Jg/l calcium hypochlorite as chlorine for the AT
method. Spiking with calcium hypochlorite simulates free chlorine that
exists in chlorinated condenser cooling river at coal-burning electric
power plants (8). Because 30-60 percent of the chlorine in the calcium
hypochlorite spikes is lost in samples analyzed immediately after spiking
(see Table 16), the total chlorine is not conserved. Therefore, evalua-
tion of true chlorine recoveries was abandoned.
For the MC method, Figure 23 summarizes total chlorine standard devia-
tion against mean total chlorine for data taken from Table 16, Experiments
16.2-16.8, and Figure 24 summarizes relative standard deviation against
mean total chlorine concentration. Figures 25-26 similarly summarize the
statistical results for the AT method.
For the MC method, the experimental recoveries of total chlorine from
river water spiked with calcium hypochlorite (Table 16) yield a total chlo-
rine pooled standard deviation of 0.5 fJg/1 and mean a relative standard
deviation of 4.8 percent for concentrations of 2-20 pg/1 (Experiment 16.2)
and a total chlorine pooled standard deviation of 2.3 HgA anc* a mean rela-
tive standard deviation of 2.2 percent for concentrations of 20-200 pg/1
(Experiments 16.3-16.7). These values (2.3 pg/1 and 2.2 percent) for river
water spiked with calcium hypochlorite are nearly identical to those for
river water spiked with chloramine-T (2.3 H8/1 and 2.5 percent). These
results are similar to those obtained by Rigdon (6,7) for deionized
water spiked with chloramine-T.
For the AT method, for concentrations of 20-200 |Jg/l, the results in
Table 16 can be combined to yield a total chlorine pooled standard devia-
tion of 3.7 Mg/1 and a mean relative standard deviation of 5.2 percent
(Experiments 16.3-16.7). These values (3.7 Mg/1 and 5.2 percent) for
river water spiked with calcium hypochlorite are about the same as those
for river water spiked with chloramine-T (4.1 pg/1 and 7.8 percent).
A valid MNDL for the MC and AT methods could not be calculated from
the data shown in Table 16 for river water spiked with calcium hypochlorite
because the mean recoveries of 9.59 and 28.6 fJg/1 from Experiments 16.2 and
16.3, respectively, were greatly above the UCL of the MNDL. No spiking
tests were conducted at lower concentrations.
Overall Precision, and Minimum Detection Limit
for Spikes with Chloramine-T and Calcium Hypochlorite
The pooled standard deviations and mean relative standard deviations
for the MC and AT methods can be combined for the spiking experiments
with chloramine-T and calcium hypochlorite in river water to yield overall
values.
27
-------
For the MC method, the total chlorine overall pooled standard devia-
tion is 0.5 pg/1 and the overall mean relative standard deviation is 4.8
percent for concentrations of 2-20 |Jg/l (Table 16, Experiment 16.2). The
total chlorine overall pooled standard deviation is 2.3 iJg/1 and the over-
all mean relative standard deviation is 2.4 percent for concentrations! of
20-200 |Jg/l (Table 15, Experiments 15.2-15.8; Table 16, Experiments 16.3-
16.7). The overall statistical values for the MC method for spiked river
water are about the same as those for spiked deionized water.
For the AT method, for concentrations of 20-200 M8/1 in spiked river
water, the total chlorine overall pooled standard deviation is 3.9 pg/1
and the overall mean relative standard deviation is 6.3 percent (Table 15,
Experiments 15.5-15.8; and Table 16, Experiments 16.3-16.7). These over-
all statistical values for the AT method are about twice those for the MC
method.
No valid overall MNDL could be determined for either the MC or the
AT method from the experimental data in Tables 15 and 16 because the mean
recoveries were greater than the UCL. However, three individual MNDL
values of 11.9, 16.7, and 15.1 |Jg/l (which could not be combined) were
calculated for the AT method from replicate determinations of river water
spiked with chloramine-T. These were the only valid MNDL values that
could be computed from the experimental recoveries for river water spiked
with chloramine-T or calcium hypochlorite.
CONDENSER COOLING RIVER WATER
Precision and Minimum Detection Limit for Unspiked,
Chlorinated Kingston Steam Plant Water Samples
Table 17 shows the chlorine demand characteristics of unchlorinated
river water used for condenser cooling at Kingston Steam Plant chlorinated
at particular concentrations for specified periods of time and Table 18
shows its concentrations of inorganic species, sanitary chemical char-
acteristics, and physical properties. As shown in Table 18, copper (120
(jg/1) and iron (890 pg/1) are only metals present in a sufficiently high
concentration to warrant a check against possible interference. However,
there should be no interference from either copper or iron at these con-
centrations, as indicated by experimental recoveries from the spiked
deionized water solutions (Table 8, Experiments 8.13 and 8.21).
Table 19, Experiments 19.1, 19.2, 19.4, 19.6, 19.8, and 19.10, shows
recoveries by the MC and AT methods for seven replicate determinations of
total chlorine in a series of six split samples of unspiked, chlorinated
condenser cooling river water from the Kingston Steam Plant with total
chlorine concentrations of 0-200 pg/1. The specific composition of each
unspiked chlorinated sample is given in Table 19 for each of the experi-
ments. Each sample discussed here and herinafter was analyzed for the
seven split, replicate total chlorine determinations by the MC and AT
28
-------
methods when total chlorine determinations by the AT method agreed within
10 percent between analyses.
This procedure for determining when to begin the split, replicate
total chlorine determinations was used throughout this report.
For the MC method, Figure 27 summarizes total chlorine standard devia-
tion against mean total chlorine for values from Table 19, Experiments 19.1,
19.2, 19.4, 19.6, 19.8, and 19.10, and Figure 28 summarizes relative stand-
ard deviation against mean total chlorine concentration. Figures 29-30
show similar summaries of the statistical results for the AT method.
For the MC method, experimental recoveries of total chlorine from
unspiked, chlorinated condenser cooling river water from the Kingston
Steam Plant (Table 19) yield a total chlorine pooled standard deviation
of 0.8 [Jg/1 and a mean relative standard deviation of 10.8 percent for
concentrations of 2-20 |Jg/l chlorine (Experiments 19.1-19.2); and a total
chlorine pooled standard deviation of 4.9 Mg/1 and mean relative standard
deviation of 5.0 percent for concentrations of 20-200 |Jg/l chlorine
(Experiments 19.4, 19.6, 19.8, and 19.10). These values are about equal
to overall values for spiked, deionized water. They are about twice those
for spiked river water and those Rigdon (6,7) obtained using deionized
water.
For the AT method, the results in Table 19 yield a total chlorine
pooled standard deviation of 6.4 )Jg/l and a mean relative standard devia-
tion of 9.5 percent for concentrations of 20-200 jJg/1 chlorine (Experiments
19.4, 19.6, 19.8, and 19.10). These values are slightly higher than those
for the MC method (4.9 |Jg/l and 5.0 percent, respectively). The total
chlorine pooled standard deviation of 6.4 JJg/1 for the AT method for con-
centrations of 20-200 JJg/1 is smaller than that of 28 M8/1 f°r tne same
concentration range found by a collaborative test in which the sampling,
kinds of samples, and variability between operators were included (4).
The MNDL for the MC method for samples of river water used for con-
denser cooling at Kingston Steam Plant is 2.0 Mg/1 with a mean recovered
concentration of 4.27 |Jg/l (Table 19, Experiment 19.1), well within the
UCL and LCL. For the AT method, the MNDL is 23.6 |Jg/l with a mean
recovered concentration of 51.4 (Jg/1 (Table 19, Experiment 19.6), just
barely below the UCL. That means that a lower test concentration should
give a better, probably lower, estimate of the MNDL; however, a calcula-
tion of the MNDL from data showing a lower mean recovery of 27.9 pg/1
(Table 19, Experiment 19.4), failed to yield LCL and UCL values that
bracketed this recovery.
For the MC method, the MNDL of 2.0 pg/1 chlorine in condenser river
water at Kingston Steam Plant compares well with the MNDL of 1.2 MgA in
deionized water, but for the AT method, the MNDL of 23.6 |JgA is higher
than the overall pooled MNDL of 12.0 (Jg/1 in deionized water and the
individual values of 11.9, 16.7, and 15.1 Mg/1 for the MNDL in river
water.
29
-------
From the analytical results for split samples of condenser water
with concentrations of 20-200 pg/1 (Table 19, Experiments 19.4, 19.6,
19.8, and 19.10), a t value can be calculated from the mean for differ-
ences ("X" - "Y") between the paired values by the MC and AT methods
(the MC method is taken as "X" and the AT method is taken as "Y"). This
t value can be compared to the tabulated value of t at the 0.025 level
for the number of paired values less one, called the degree of freedom,
to check whether the mean differs from zero at 95 percent significance
(19,25). Because the absolute value for the calculated t of -1.515 is
less than that of 2.052 for 27 degrees of freedom, there is no reason to
conclude the mean differs from zero at 95 percent significance. Also,
the mean differences between the MC and AT methods was -3.4 [Jg/1 with
standard deviation of 12.0 |Jg/l.
Precision, Accuracy, and Minimum Detection Limit for Spiked,
Chlorinated, Kingston Steam Plant Water Samples
Table 19, Experiments 19.3, 19.5, 19.7, 19.9, and 19.11, shows the
total chlorine recoveries from the chlorinated Kingston Steam Plant sam-
ples shown in Experiments 19.2, 19.4, 19.6, 19.8, and 19.10 after they
had been spiked with calcium hypochlorite and the mean background con-
centration of the unspiked sample had been subtracted. Accuracy calcu-
lations were not attempted on recoveries from spiked condenser water from
this steam plant or any one discussed hereinafter for the same reasons
already given in the discussion of experiments with river water. Further-
more, results obtained by spiking water already chlorinated will probably
be biased due to the changes in chlorine damand, which would require cor-
rection for the changing chlorine demand of the water. Also, no paired-
sample t-tests were attempted for the spiked condenser water samples from
this steam plant or any one discussed hereinafter because the spiked con-
centration of calcium hypochlorite differed for the MC and AT methods.
(The spiked concentrations differed for each method in attempts to spike
at a concentration to double the background concentration of total chlorine
for determinations of the subsequent seven replicates.)
For the MC method, Figure 31 summarizes total chlorine standard
deviation against mean total chlorine in spiked condenser water (Table 19,
Experiments 19.3, 19.5, 19.7, 19.9, and 19.11), and Figure 32 summarizes
relative standard deviation against mean total chlorine concentration.
Figures 33-34 similarly summarize the statistical data for the AT method.
For the MC method, experimental recovery of total chlorine from spiked
condenser cooling water from Kingston Steam Plant (Table 19) yield a total
chlorine pooled standard deviation of 1.0 pg/1 and a mean percent relative
standard deviation of 8.8 percent for concentrations of 2-20 [Jg/1 chlorine
(Experiment 19.3) and a total chlorine pooled standard deviation of 9.2 H8/1
and a mean relative standard deviation of 10.1 percent for concentrations of
20-200 |Jg/l chlorine (Experiments 19.5, 19.7, 19.9, and 19.11).
30
-------
For the AT method, the results in Table 19 yield a total chlorine
pooled standard deviation of 5.5 Mg/1 and a mean relative standard devia-
tion of 7.8 percent (Experiments 19.5, 19.7, 19.9, and 19.11) for concen-
trations of 20-200 Mg/1 chlorine. These values are somewhat lower than
those for the MC method (9.2 MgA and 10.1 percent, respectively).
The MNDL of the MC and AT methods could not be calculated because
the mean recoveries of 11.3 and 22.9 Mg/1 chlorine, respectively, were
the lowest recoveries for spiked condenser water at Kingston Steam Plant
(Experiments 19.3 and 19.5) and were above the UCL for each respective
MNDL.
Precision and Minimum Detection Limit for Unspiked,
Chlorinated Shawnee Steam Plant Water Samples
Table 20 shows the chlorine demand characteristics of unchlorinated
river water used for condenser cooling at Shawnee Steam Plant chlorinated
at particular concentrations for specified periods of time, and Table 21
shows its concentrations of inorganic species, sanitary chemical charac-
teristics, and physical properties. The concentration of 1200 pg/1 for
iron (Table 21) is sufficiently high to warrant investigation of possible
interference. However, there should be no inteference from iron at this
concentration, as shown by experimental recoveries from the spiked deion-
ized water solutions (Table 8, Experiment 8.21). After the insoluble
materials in the sample shown in Table 21 settled and aliquots of the
supernant were withdrawn for diluting chlorinated samples, the settled
materials had been concentrated to yield a water slurry with the con-
centrations of inorganic species shown in Table 22. This solution
contained 8000 pg/1 of iron, which is sufficiently high to cause inter-
ference, as shown by recoveries from spiked deionized water solutions
(Experiment 8.19). This fact was confirmed by determinations of 32.3 and
37.7 Mg/1 chlorine by the MC method for the slurry of settled materials;
these concentrations could not be corroborated by the AT method (Table 23,
Experiment 23.1, Replicate Numbers 6 and 7).
Table 23, Experiments 23.1, 23.2, 23.4, 23.6, 23.8, and 23.10, shows
recoveries by the MC and AT methods for seven replicate determinations of
total chlorine in a series of six split samples of unspiked, chlorinated
condenser cooling river water from the Shawnee Steam Plant with total
chlorine concentrations of 0-200 pg/1. The specific composition of each
unspiked, chlorinated sample is given in Table 21.
For the MC method, Figure 35 summarizes total chlorine standard devia-
tion against mean total chlorine for values from Table 23, Experiments 23.2,
23.4, 23.6, 23.8, and 23.10, and Figure 36 summarizes relative standard
deviation against mean total chlorine concentration. Figures 37-38 simi-
larly summarize the statistical results for the AT method.
For the MC method, the experimental recoveries of total chlorine from
unspiked, chlorinated condenser cooling river water from Shawnee Steam Plant
31
-------
(Table 23) yield a total chlorine pooled standard deviation of 4.5 MS/1
mean relative standard deviation of 56.9 percent for concentrations of 2-20
(Jg/l chlorine (Experiment 23.2) and a total chlorine pooled standard devia-
tion of 7.9 MS/1 an respectively (Experiments 23.4 and
23.6). Pooling these values failed to yield a valid pooled MNDL for the
AT method because the pooled mean recovery of 40.4 MS/1 fell outside the
UCL. The MNDL values for the AT method for unspiked Shawnee condenser
cooling river water are similar to those for unspiked Kingston condenser
cooling river water.
A t value can be calculated from the split sample data by the MC and
AT methods covering concentrations of 20-200 pg/1 chlorine shown in Table
23, Experiments 23.4, 23.6, 23.8, and 23.10. Because the calculated t of
1.095 is less than that of 2.052 for t 0.025 with 27 degrees of freedom,
there is no reason to conclude the mean of the differences between the MC
and AT methods differs from zero at 95 percent significance. Also, the
mean difference was 3.7 MS/1 with a standard deviation of 17.8 pg/1.
Precision and Minimum Detection Limit for Spiked,
Chlorinated Shawnee Steam Plant Water Samples
Table 23, Experiments 23.3, 23.5, 23.7, 23.9, and 23.11, shows the
total chlorine recoveries from the chlorinated Shawnee Steam Plant samples
shown in Experiments 23.2, 23.4, 23.6, 23.8, and 23.10, respectively, after
they were spiked with calcium hypochlorite and the mean background of the
unspiked sample was subtracted.
For the MC method, Figure 39 summarizes total chlorine standard devia-
tion against mean total chlorine for spiked condenser water (Table 23,
32
-------
Experiments 23.3, 23.5, 23.7, 23.9, and 23.11), and Figure 40 summarizes
relative standard deviation against mean total chlorine concentration.
Figures 41-42 similarly summarize the statistical data for the AT method.
For the MC method, the results shown in Table 23 for spiked condenser
cooling water from the Shawnee Steam Plant yield a total chlorine pooled
standard deviation of 1.6 [Jg/1 and mean relative standard deviation of 13.8
percent for concentrations of 2-20 (Jg/1 chlorine (Experiment 23.3) and a
total chlorine pooled standard deviation of 7.9 Mg/1 and a mean relative
standard deviation of 11.5 percent for concentrations of 20-200 |Jg/l chlo-
rine (Experiments 23.5, 23.7, and 23.9).
For the AT method, the results shown in Table 23 yield a total chlo-
rine pooled standard deviation of 9.7 M8/1 and a mean relative standard
deviation of 28.8 percent (Experiments 23.3, 23.5, 23.7, and 23.9) for
concentrations of 20-200 [Jg/1 chlorine. These values are somewhat higher
than those for the MC method (7.9 Mg/1 and 11.9 percent, respectively),
the reverse of what was found for spiked condenser water from Kingston
Steam Plant.
The MNDL of the MC method could not be calculated, because the mean
recovery of 11.4 M8/1 chlorine, the lowest recovery from spiked condenser
water at the Shawnee Steam Plant (Experiment 23.3), was above the UCL for
the MNDL. However, three valid MNDL values can be calculated for the AT
method from recoveries shown by Experiments 23.3, 23.5, and 23.7. The
MNDL values are 38.3, 26.7, and 29.9 pg/1 for mean recoveries of 27.1,
21.4, and 47.1 pg/1, respectively. These values can be pooled to yield
a valid pooled MNDL of 26.0 |Jg/l and a pooled mean recovery of 31.9 Mg/1-
Precision and Minimum Detection Limit for Unspiked,
Chlorinated Allen Steam Plant Water Samples
Table 24 shows the chlorine demand characteristics of unchlorinated
river water used for condenser cooling at Allen Steam Plant chlorinated at
particular concentrations for specified periods of time, and Table 25 shows
its concentrations of inorganic species, sanitary chemical characteristics,
and physical properties. The iron concentration of 1600 |Jg/l (Table 25) can
be eliminated as a possible source of interference, as explained for the
Kingston and Shawnee water samples. However, the sample from Allen Steam
Plant contained 0.45 mg/1 of ammonia as nitrogen (Table 25), whereas the
samples from Kingston (Table 18) and Shawnee (Table 21) did not contain
more than 0.06 mg/1. The chlorine demand for the Allen sample was also
greater, and this was probably caused by the presence of ammonia.
Table 26, Experiments 26.1, 26.2, 26.4, 26.6, 26.8, and 26.10, shows
recoveries by the MC and AT methods for seven replicate determinations of
total chlorine in a series of six split samples of unspiked, chlorinated
condenser cooling river water from Allen Steam Plant with total chlorine
concentrations of 0-250 Mg/1. The specific composition of each unspiked,
chlorinated sample is given in Table 25.
33
-------
For the MC method, Figure 43 summarizes total chlorine standard devia-
tion against mean total chlorine in values from Table 26, Experiments 26.2,
26.4, 26.6, 26.8, and 26.10, and Figure 44 summarizes relative standard
deviation against mean total chlorine. Figures 45-46 similarly the
statistical results for the AT method.
The results obtained by the MC method in Experiment 26.10 were about
40 percent higher than those obtained by the AT method (mean recoveries of
247 and 177 M8/l> respectively). When 2 percent (w/v) sodium pyrophosphate
decahydrate was added to the unspiked sample (Experiment 26.11) and the
spiked sample (Experiment 26.12), there was closer agreement between the
two methods; the mean recoveries by the MC method were only about 10 and
20 percent higher than those by the AT method for these respective samples.
The high results obtained by the MC method may have been caused by the
0.45 mg/1 of ammonia as nitrogen in the unchlorinated sample (Table 25)
that continued to react with free chlorine to form combined chlorine (25)
in the freshly composited sample on October 28, 1980, and by the fact that
the MC method uses standard addition with chloramine-T (a combined source
of chlorine) to assay for total chlorine, whereas the AT method uses direct
titration with standard PAO. The large discrepancy between methods was
absent on the following day, October 29, 1980, for unspiked samples com-
posed of the same chlorinated and unchlorinated water, after its chlorine
reactions had equilibrated for about a day and after the sample had pro-
bably lost most of its free chlorine, the only oxidant in the sample known
to react with ammonia. Except in Experiment 26.7, however, no large dis-
crepancy was observed for samples spiked with calcium hypochlorite on
October 29, 1980. Since the same water sample was analyzed both days,
this automatically eliminates the 1600 pg/1 of iron and 670 (Jg/1 of manga-
nese in the sample (see Table 25) as possible sources for the high results.
Furthermore, the MC method did not produce high results for chlorinated
river water or for chlorinated condenser river water from Kingston, Shawnee,
and John Sevier Steam Plants, probably because the ammonia was less than
0.06 mg/1 as nitrogen (see results in Tables 14, 18, 21, and 28).
For the MC method, the experimental recoveries of total chlorine from
unspiked, chlorinated condenser cooling river water from Allen Steam Plant
(Table 26) yield a total chlorine pooled standard deviation of 7.4 (Jg/1 and
a mean relative standard deviation of 11.4 percent for concentrations of
20-200 Mg/1 chlorine (Experiments 26.2, 26.4, 26.6, and 26.8). These values
approximate those for unspiked condenser water from the Shawnee and Kingston
Steam Plants. No total chlorine data were obtained by the MC method for con-
centrations of 2-20 |Jg/l.
For the AT method, the results in Table 26 yield a total chlorine
pooled standard deviation of 10.0 (Jg/1 and a mean relative standard devia-
tion of 15.4 percent (Experiments 26.2, 26.4, 26.6, 26.8, and 26.10) for
concentrations of 20-200 [Jg/1 chlorine. As with unspiked condenser water
from Kingston and Shawnee, these values are slightly greater than those
obtained by the MC method (7.4 pg/1 and 11.4 percent, respectively).
34
-------
The MNDL for the MC method is 10.1 |Jg/l chlorine for the mean recovery
of 21.1 |Jg/l from unspiked, chlorinated condenser river water at Allen
Steam Plant (Experiment 26.2). The MNDL for the AT method is 26.7 |Jg/l for
the mean recovery of 21.1 (Jg/1 shown in Experiment 26.2. The MNDL for the
AT method obtained from unspiked sample results at Allen Steam Plant is
similar to that for Kingston and Shawnee. Although no MNDL was obtained
for the MC method for Shawnee, the one for Allen (10.1 |Jg/l) was several
times higher than that for Kingston (2.0 |Jg/l).
A t value can be calculated from the split sample data obtained by
the MC and AT methods for concentrations of 20-200 |jg/l shown in Table 26,
Experiments 26.2 (replicate numbers 2 and 4-7), 26.4, 26.6, and 26.8.
Because the calculated t of 0.596 is less than that of 2.060 for t ~ -.,.
with 25 degrees of freedom, there is no reason to conclude the mean of the
differences between the MC and AT methods differs from zero at 95 percent
significance. Also, the mean difference was 1.0 pg/1 with standard devia-
tion of 8.9 [Jg/1.
Precision and Minimum Detection Limit for Spiked,
Chlorinated Allen Steam Plant Water Samples
Table 26, Experiments 26.3, 26.5, 26.7, 26.9, and 26.12, gives the
total chlorine recoveries from the chlorinated Allen Steam Plant samples,
shown in Experiments 26.2, 26.4, 26.6, 26.8, and 26.10, respectively, after
they were spiked with calcium hypochlorite and the mean background of the
unspiked sample was subtracted.
For the MC method, Figure 47 summarizes total chlorine standard devia-
tion against mean total chlorine for spiked condenser water (Table 26,
Experiments 26.3, 26.5, 26.7, 26.9, and 26.12), and Figure 48 summarizes
relative standard deviation against mean total chlorine. Figures 49-50
similarly summarize the statistical results for the AT method.
For the MC method, the results shown in Table 26 for spiked condenser
cooling water from the Allen Steam Plant yield, a total chlorine pooled
standard deviation of 3.7 (Jg/1 and a mean relative standard deviation of
19.5 percent for concentrations of 2-20 pg/1 chlorine (Experiment 26.3)
and a total chlorine pooled standard deviation of 8.9 Mg/1 and mean relative
standard deviation of 12.7 percent for concentrations of 20-200 pg/1 chlorine
(Experiments 26.5, 26.7, 26.9, and 26.12).
For the AT method, the results in Table 26 yield a total chlorine
pooled standard deviation of 14.6 Mg/1 and a mean relative standard devia-
tion of 21.4 percent (Experiments 26.5, 26.7, 26.9, and 26.12) for concen-
trations of 20-200 (Jg.l. These values are nearly twice as large as those
for the MC method (8.9 Mg/1 and 12.7 percent, respectively).
A pooled MNDL of 17.2 |Jg/l for a pooled mean recovery of 23.6 |Jg/l can
be calculated for the MC method from the individually computed values of
35
-------
11.6 and 25.8 |Jg/l with mean recoveries of 19.0 and 28.3 (Jg/1, respectively,
for spiked condenser water at Allen Steam Plant (Experiments 26.3 and 26.5).
No valid MNDL values were calculated from spiked condenser water for the
Kingston and Shawnee Steam Plants.
For the AT method, two MNDL values can be calculated from the recov-
eries shown in Experiments 26.5 and 26.7. These MNDL values are 22.9 and
29.2 (Jg/1 for mean recoveries of 34.3 and 25.7 |Jg/l, respectively. These
values can be pooled to yield a valid pooled MNDL of 22.5 |Jg/l which has
a pooled mean recovery of 30.0 MS/1- This pooled MNDL value compares well
with the values of 26.0 and 31.9 (Jg/1 obtained for spiked condenser water
at Shawnee Steam Plant.
Precision and Minimum Detection Limit For Unspiked,
Chlorinated John Sevier Steam Plant Water Samples
Table 27 shows the chlorine demand characteristics of unchlorinated
river water used for condenser cooling at John Sevier Steam Plant chlori-
nated at particular concentrations for specified periods of time, and
Table 28 shows its concentrations of inorganic species, sanitary chemical
characteristics, and physical properties. None of the inorganic species
shown in Table 28 were present in sufficiently high concentration to cause
interference.
Table 29, Experiments 29.1, 29.2, 29.4, 29.6, 29.8, 29.10, and 29.12,
shows recoveries by the MC and AT methods for seven replicate determina-
tions of total chlorine from a series of seven split samples of unspiked,
chlorinated condenser cooling river water from Allen Steam Plant with
0-325 M8/1 total chlorine. The specific composition of the unspiked
samples is given in Table 28.
For the MC method, Figure 51 summarizes total chlorine standard devia-
tion against mean total chlorine for values from Table 29, Experiments 29.1,
29.2, 29.4, 29.6, 29.8, 29.10, and 29.12, and Figure 52 summarizes relative
standard deviation against mean total chlorine. Figures 53-54 similarly
summarize the statistical results for the AT method.
For the MC method, experimental recoveries of total chlorine from
unspiked, chlorinated condenser cooling river water from John Sevier Steam
Plant (Table 29) yield, a total chlorine pooled standard deviation of 0.5
[Jg/1 and a mean relative standard deviation of 12.8 percent for concentra-
tions of 2-20 |jg/l (Experiments 29.1 and 29.2) and a total chlorine pooled
standard deviation of 4.4 (Jg/1 an^ a mean relative standard deviation of
7.5 percent for concentrations of 20-200 (Jg/1 chlorine (Experiments 29.4,
29.6, 29.8, and 29.10). These values for total chlorine pooled standard
deviation and mean relative standard deviation for John Sevier are similar
to those for the MC method for concentrations of 20-200 (Jg/1 chlorine in
unspiked condenser water from Kingston and Allen Steam Plants (9.2 (Jg/1
and 10.1 percent, 7.9 (Jg/1 and 6.8 pecent, and 7.4 (Jg/1 and 11.4 percent,
36
-------
respectively). For the MC method, the total chlorine standard deviation
of 0.5 Mg/1 and mean relative standard deviation of 12.8 percent for
concentrations of 2-20 |Jg/l chlorine in unspiked condenser water from
John Sevier Steam Plant compare to total chlorine standard deviations
of 1.0 and 4.5 MS/I aQd mean relative standard deviations of 8.8 and
56.9 percent for the Kingston and Shawnee Steam Plants, respectively.
No comparable data were obtained at Allen Steam Plant.
For the AT method, results in Table 29 yield a total chlorine pooled
standard deviation of 7.5 and a mean relative standard deviation of 11.8
percent (Experiments 29.4, 29.6, 29.8, and 29.10) for concentrations of
20-200 MS/1 chlorine. As was the case for unspiked condenser cooling
water at Kingston, Shawnee, and Allen, these values for the AT method
are greater than those for the MC method (4.4 |Jg/l and 7.5 percent,
respectively). The values for John Sevier compare to the 6.4, 9.9,
and 10.0 pg/1 total chlorine pooled standard deviation values and 9.5,
14.3, and 15.4 percent mean relative standard deviation values for the
AT method for concentrations of 20-200 |Jg/l chlorine for unspiked con-
denser water at the Kingston, Shawnee, and Allen Steam Plants and to
those of 4.9, 7.9, and 7.4 |Jg/l and 5.0, 6.8, and 11.4 percent, respec-
tively, for the MC method for the same concentrations for the same steam
plants.
For the MC method, MNDL values of 1.3 and 12.6 pg/1 for mean recover-
ies of 2.10 and 25.9 M8/1 can be calculated from the data shown in Experi-
ments 29.1 and 29.4. These MNDL values could not be combined, because the
pooled mean recovery of 14.0 MS/1 was above the pooled UCL for the combined
MNDL. For the AT method, the pooled MNDL is 16.6 |jg/l for a pooled mean
recovery of 22.1 Mg/1 (from data for Experiments 29.2 and 29.4). The
individual MNDL values for the latter two experiments are 19.8 and 19.2
[jg/1 with mean recoveries of 13.5 and 30.7 H8/l> respectively.
For the MC method, MNDL values of 1.3 and 12.6 |Jg chlorine/1 for
unspiked condenser water at John Sevier Steam Plant compare with those of
2.0 and 10.1 pg/1 for unspiked condenser water at the Kingston and Allen
Steam Plants, respectively. No MNDL was obtained for the MC method for
the Shawnee Steam Plant.
For the AT method, the pooled MNDL of 16.6 Mg/1 for unspiked condenser
water at the John Sevier Steam Plant compares with that of 26.7, 22.0 or
25.5, and 23.6 MS/1 for unspiked condenser water at Allen, Shawnee, and
Kingston. For all steam plants, the MNDL for the AT method was higher
than that for the MC method.
As was the case for the Kingston, Shawnee, and Allen results, a t
value can be calculated from the split sample data obtained by the MC and
AT methods for concentrations of 20-200 MS/1 shown in Table 29, Experi-
ments 29.2 (Replicate Numbers 1 and 7), 29.4, 29.6, 29.8, and 29.10.
Because the absolute value of the calculated t of -0.779 is less than
that of 2.045 for t_ .„_ with 29 degrees of freedom, there is no reason
to conclude the mean"of the differences between the MC and AT methods
37
-------
differs from zero at 95 percent significance. Also, the mean differences
was -1.6 (JgA with standard deviation of 10.9
Precision and Minimum Detection Limit for Spiked,
Chlorinated John Sevier Steam Plant Water Samples
Table 29, Experiments 29.3, 29.5, 29.7, 29.9, 29.11, and 29.13,
shows the total chlorine recoveries from the chlorinated John Sevier
Steam Plant samples (Experiments 29.2, 29.4, 29.6, 29.8, 29.10, and 29.12,
respectively), after they had been spiked with calcium hypochlorite and
the mean background of the unspiked sample had been subtracted.
For the MC method, Figure 55 summarizes total chlorine standard devia-
tion against mean total chlorine for spiked condenser water (Table 29,
Experiments 29.3, 29.5, 29.7, 29.9, 29.11, and 29.13), and Figure 56 sum-
marizes relative standard deviation against mean total chlorine. Figures
57 and 58 similarly summarizes the statistical results for the AT method.
For the MC method, experimental recoveries from spiked condenser cool-
ing water from the John Sevier Steam Plant yield a total chlorine pooled
standard deviation of 0.9 pg/1 and mean relative standard deviation of 12.7
percent for concentrations of 2-20 |Jg/l chlorine (Experiments 29.3 and 29.5)
and a total chlorine pooled standard deviation of 8.6 MS/I and mean relative
standard deviation of 7.7 percent for concentrations of 20-200 |Jg/l chlo-
rine (Experiments 29.7, 29.9, 29.11, and 29.13).
For the AT method, the results in Table 29 yield a total chlorine
pooled standard deviation of 9.0 M8/1 an
-------
estimated for the MC method from data gathered for spiked condenser water
at the Kingston and Shawnee Plants.
For the AT method, the pooled MNDL of 23.1 |Jg/l for spiked condenser
water at the John Sevier Steam Plant is about the same as the pooled MNDL
of 26.0 MS/1 for spiked condenser water at the Shawnee Steam Plant and the
pooled MNDL of 22.5 Mg/1 for spiked water at Allen. A valid MNDL could not
be derived from data for spiked water at Kingston.
Overall Precision and Minimum Detection Limit for Unspiked, Chlorinated
Kingston, Shawnee, Allen, and John Sevier Steam Plant Water Samples
The pooled standard deviations and mean relative standard deviations
for the MC and AT methods for the recovery experiments with unspiked, chlo-
rinated condenser cooling river water samples from the Kingston, Shawnee,
Allen, and John Sevier Steam Plants can be combined to yield the overall
values.
For the MC method, the total chlorine overall pooled standard
deviation and the overall mean relative standard deviation are 2.1 |Jg/l
and 20.8 percent, respectively, for concentrations of 2-20 (Jg/1 chlorine
(Table 19, Experiments 19.1 and 19.2; Table 23, Experiment 23.2; Table 29,
Experiments 29.1 and 29.2) and the total chlorine overall pooled standard
deviation and the overall mean relative standard deviation are 6.3 Mg/1
and 7.7 percent for concentrations of 20-200 MS/1 chlorine (Table 19,
Experiments 19.4, 19.6, 19.8, and 19.10; Table 23, Experiments 23.4,
23.6, 23.8, and 23.10; Table 26, Experiments 26.2, 26.4, 26.6, and 26.8;
Table 29, Experiments 29.4, 29.6, 29.8, and 29.10). For the MC method,
the total chlorine overall pooled standard deviation (2.1 MS/1) and over-
all mean relative standard deviation (20.8 percent) for concentrations
of 2-20 Mg/1 chlorine in unspiked condenser water are greater than the
respective values for the same range of concentrations in spiked river
water (0.5 Mg/1 and 4.8 percent) or in spiked deionized water (0.5 MS/1
and 8.7 percent). For the MC method, the total chlorine overall pooled
standard deviation (6.3 MS/1) and overall mean relative standard devia-
tion (7.7 percent) for concentrations of 20-200 MS/1 chlorine in unspiked
condenser water are greater than the respective values for the same range
of concentrations in spiked river water (2.3 MS/I and 2.4 percent) or in
spiked deionized water (5.9 MS/1 ano< 3.9 percent).
For the AT method, the total chlorine overall pooled standard devia-
tion and overall mean relative standard deviation are 8.7 MS/1 and 12.9
percent, respectively, for concentrations of 20-200 MS/1 chlorine in
unspiked, chlorinated condenser cooling river water (Table 19, Experi-
ments 19.4, 19.6, 19.8, and 19.10; Table 23, Experiments 23.4, 23.6,
23.8, and 23.10; Table 26, Experiments 26.2, 26.4, 26.6, 26.8, and
26.10; Table 29, Experiments 29.4, 29.6, 29.8, and 29.10). For the
AT method, the total chlorine overall pooled standard deviation (8.7
Mg/1) and overall mean relative standard deviation (12.9 percent) for
concentrations of 20-200 MS/1 chlorine in unspiked condenser water are
39
-------
greater than the respective values for the same range of concentrations
in spiked river water (3.9 Mg/1 and 6.3 percent) or in spiked deionized
water (4.0 Mg/1 and 5.0 percent).
For concentrations of 20-200 MgA chlorine in unspiked, chlorinated
condenser cooling river water, the total chlorine overall pooled standard
deviation of 6.3 Mg/1 and overall mean relative standard deviation of 7.7
percent for the MC method are lower than the total chlorine overall pooled
standard deviation of 8.7 Mg/1 and overall mean relative standard devia-
tion of 12.9 percent for the AT method. The total chlorine overall pooled
standard deviation of 8.7 (Jg/1 obtained in this study for the AT method for
20-200 |Jg/l in unspiked chlorinated condenser water is lower than the 28
Mg/1 obtained in a collaborative test study with unspiked, chlorinated
condenser water for 0-200 Mg/1 (4). The collaborative test study (4)
included sampling variation and operator-to-operator variation in addition
to the instrumental, kind of sample, and single-operator variation involved
in this study.
A valid overall pooled MNDL could not be calculated for the MC method
by combining the individual valid MNDL values of 2.0 Mg/1 for unspiked,
chlorinated condenser water from the Kingston Steam Plant, 10.1 |Jg/l from
the Allen Plant; and 1.3 and 12.6 Mg/1 from the John Sevier Plant (Table 19,
Experiment 19.1; Table 26, Experiment 26.2; Table 29, Experiments 29.1 and
29.4) because the mean for the combined results (13.3 Mg/1) was greater than
the UCL (9.03 (JgA) for the overall MNDL (6.5 pg/1). However, the invalid
overall estimate of 6.5 Mg/1 for the MNDL as well as the individual valid
MNDL values of 2.0, 10.1, 1.3, and 12.6 |Jg/l are greater than the valid
MNDL of 1.2 Mg/1 for deionized water spiked with chloramine-T (Table 7,
Experiment 7.3). For the MC method, Rigdon estimated an MNDL of 3 or
4 |Jg/l in deionized water spiked with chloramine-T (6,7).
Although not subjected to the validation test, the overall MNDL for
the AT method is 17.8 Mg/1 for the combined results of individual valid
MNDL values of 23.6 pg/1 from the Kingston Steam Plant (Table 19, Experi-
ment 19.6); 22.0 and 25.5 (Jg/1 from the Shawnee Plant (Table 23, Experi-
ments 23.4 and 23.6); 26.7 |Jg/l from the Allen Plant (Table 26, Experiment
26.2); and 19.8 and 19.2 |jg/l from the John Sevier Plant (Table 29, Experi-
ments 29.2 and 29.4). The overall MNDL of 17.8 |jg/l is greater than the
individual valid MNDL values of 11.9, 16.7, and 15.1 Mg/1 for river water
spiked with chloramine-T (Table 15, Experiments 15.2, 15-4, and 15.5) and
the valid overall pooled MNDL of 12.0 (Jg/1 for deionized water spiked with
calcium hypochlorite (Table 11, Experiments 11.3 and 11.4) and 50 percent
(w/w) calcium hypochlorite and chloramine-T as chlorine (Table 12,
Experiment 12.3).
For unspiked, chlorinated condenser water from the Kingston, Shawnee,
Allen, and John Sevier Steam Plants, the overall MNDL of 6.5 Mg/1 f°r the
MC method is lower than that of 17.8 |jg/l for the AT method. The overall
MNDL of 17.8 Mg/1 in unspiked condenser water, which was estimated in this
study, is lower than the 85 Mg/1 obtained in a collaborative test study
40
-------
with unspiked, chlorinated condenser water (4). The collaborative test
study (4) included sampling and operator-to-operator variation as well
as the instrumental, kind of sample, and single-operator variation
involved in this study.
A t value can be calculated from the split sample data for the MC and
AT methods for concentrations of 20-200 Mg/1 f°r &H split sample results
for unspiked, chlorinated condenser water from Kingston, Shawnee, Allen,
and John Sevier, shown in Table 19, Experiments 19.4, 19.6, 19.8, and
19.10; Table 23, Experiments 23.4, 23.6, 23.8, and 23.10; Table 26,
Experiments 26.2 (Replicate Numbers 2 and 4-7), 26.4, 26.6, and 26.8;
and Table 29, Experiments 29.2 (Replicate Numbers 1 and 7), 29.4, 29.6,
29.8, and 29.10. Because the absolute value of the calculated t of -0.091
is less than that of 1.96 with 111 degrees of freedom, there is no reason
to conclude the mean of the differences between the MC and AT methods
differs from zero at 95 percent significance. Also, the mean difference
was -0.1 (Jg/1 with standard deviation of 13.0 (Jg/1.
Overall Precision and Minimum Detection Limit for Spiked, Chlorinated
Kingston, Shawnee, Allen, and John Sevier Steam Plant Water Samples
The pooled standard deviations and mean relative standard deviations
for the MC and AT methods for the recovery experiments with the spiked,
chlorinated condenser cooling river water samples from the Kingston,
Shawnee, Allen, and John Sevier Steam Plants can be combined to yield
overall values.
For the MC method, the total chlorine overall pooled standard devia-
tion and the overall mean relative standard deviation are 1.9 (Jg/1 and
13.5 percent, respectively, for concentrations of 2-20 (Jg/1 chlorine
(Table 19, Experiment 19.3; Table 23, Experiment 23.3; Table 26, Experi-
ment 26.3; Table 29, Experiments 29.3 and 29.5); and the total chlorine
overall pooled standard deviation and the overall mean relative standard
deviation are 8.7 (Jg/1 and 10.4 percent for concentrations of 20-200 |Jg/l
chlorine (Table 19, Experiments 19.5, 19.7, 19.9, 19.11; Table 23, Experi-
ments 23.5, 23.7, 23.9; Table 26, Experiments 26.5, 26.7, 26.9, 26.11;
Table 29, Experiments 29.7, 29.9, 29.11, and 29.13). For the MC method,
the total chlorine overall pooled standard deviations of 1.9 and 8.7 (Jg/1
for concentrations of 2-20 and 20-200 (Jg/1 chlorine, respectively, for
spiked, chlorinated condenser water from the Kingston, Shawnee, Allen, and
John Sevier Steam Plants can be compared with 2.1 and 6.3 (Jg/1 for concen-
trations of 2-20 and 20-200 (Jg/l> respectively, for unspiked, chlorinated
condenser water from the same four steam plants. For concentrations of 2-20
|jg/l, the total chlorine overall standard deviations of 1.9 and 2.1 |Jg/l are
similar for spiked and unspiked condenser water, but for concentrations of
20-200 (Jg/1, the total chlorine standard deviation of 8.7 (JgA f°r spiked
condenser water is greater by approximately
-------
For the AT method, for concentrations of 20-200 Mg/1 in spiked,
chlorinated condenser cooling river water, the total chlorine overall
pooled standard deviation and overall mean relative standard deviation
are 10.2 Mg/1 and 19.6 percent, respectively (Table 19, Experiments 19.5,
19.7, 19.9, 19.11; Table 23, Experiments 23.3, 23.5, 23.7, and 23.9;
Table 26, Experiments 26.5, 26.7, 26.9, and 26.12; Table 29, Experiments
29.5, 29.7, 29.9, and 29.11). For concentrations of 20-200 (Jg/1 chlorine,
the total chlorine overall standard deviation of 10.2 (Jg/1 f°r the AT
method for spiked, chlorinated condenser water from the Kingston, Shawnee,
Allen, and John Sevier Steam Plants is greater than that of 8.7 Mg/1 for
unspiked, chlorinated condenser water from the same four steam plants. It
is not greater by *J2, which would be expected ideally (26).
For concentrations of 20-200 |Jg/l in spiked, chlorinated condenser
cooling river water, the total chlorine overall pooled standard deviation
of 8.7 (Jg/1 and overall mean relative standard deviation of 10.4 percent
for the MC method is lower than the total chlorine overall pooled standard
deviation of 10.2 M8/1 and overall mean relative standard deviation of 19.6
percent for the AT method.
Although not subjected to the validation test, the overall MNBL of
the MC method for spiked, chlorinated condenser water is 13.3 Mg/1 for the
combined results of individual valid MNDL values of 11.6 and 25.8 Mg/1 from
Allen Steam Plant (Table 26, Experiments 26.3 and 26.5) and 3.2 Mg/1 from
the John Sevier Plant (Table 29, Experiment 29.3). This MNDL value of 13.3
Mg/1 for the MC method for spiked condenser water is about twice that of
6.5 Mg/1 f°r unspiked condenser water, for which the associated individual
MNDL values are 2.0, 10.1, 1.3, and 12.6 Mg/1-
Although not subjected to the validation test, the overall MNDL for
the AT method is 22.2 Mg/1 f°r spiked condenser water, for which the indi-
vidual valid MNDL values are 38.3, 26.7, and 29.9 MS/1 for Shawnee Steam
Plant (Table 23, Experiments 23.3, 23.5, and 23.7); 22.9 and 29.2 Mg/1 for
Allen Plant (Table 28, Experiments 26.5 and 26.7); and 23.6 and 29.9 M8/1
for the John Sevier Steam Plant (Table 29, Experiments 29.5 and 29.7).
This value of 22.2 M8/1 f°r tne AT method for spiked condenser water is
larger than that of 17.8 Mg/1 for unspiked condenser water, for which the
associated individual valid MNDL values are 23.6, 22.0, 25.5, 26.7, 19.8,
and 19.2 Mg/1-
For spiked, chlorinated condenser water from Kingston, Shawnee, Allen,
and John Sevier Steam Plants, the overall MNDL of 13.3 Mg/1 for the MC
method is lower than that of 22.2 M8/1 f°r the AT method.
42
-------
REFERENCES
1. Effluent Limitation Guidelines, Pretreatment Standards and New Source
Performance Standards Under Clean Water Act; Steam Electric Power
Generating Point Source Category. Proposed Rules. Fed. Regist.,
45 (200): 68352, October 14, 1980.
2. U.S. Environmental Protection Agency. Methods for Chemical Analysis
of Water and Wastes. Chlorine, Total Residual, Method 330.1
(Titrimetric, Amperometric) and Method 330.3 (Titrimetric, lodometric)
EPA-600/4-79-020, U.S. Environmental Protection Agency, Cincinnati,
Ohio, March 1979.
3. American Society for Testing and Materials. Standard Test Methods
for Residual Chlorine in Water, D1253-76, Method A - Direct Ampero-
metric Titration. 1979 Annual Book of ASTM Standards, Part 31.
American Society for Testing and Materials, Philadelphia,
Pennsylvania, 1979. pp 323-329.
4. NUS Corporation. Collaborative Test Results for Chlorine Analysis
by Amperometric Titration. NUS Corporation, Cyrus Wm. Rice Division,
Manor Oak Two, 1910 Cochran Road, Pittsburg, Pennsylvania 15220,
March 1979. 196 pp.
5. Amperometric Chlorine Methods Studied. Anal. Control, 4(5):
1-2, September 1979. NUS Corporation, Cyrus Wm. Rice Division, 15
Noble Avenue, Pittsburgh, Pennsylvania 15205.
6. Rigdon, L. P., G. J. Moody, and J. W. Frazer. Determination of
Residual Chlorine in Water with Computer Automation and a Residual-
Chlorine Electrode. Anal. Chem., 50(3): 465-469, 1978.
7. Rigdon, L. P., C. L. Pomernacki, D. J. Balaban, and J. W. Frazer.
Automated Potentiometric Analysis Using Ion-Selective Electrodes.
Preprint UCRL-82378. Lawrence Livermore Laboratory, Livermore,
California 94550, April 19, 1979. 18 pp.
8. Sawyer, C. N. and P. L. McCarty. Chemistry for Sanitary Engineers.
2nd Edition. McGraw-Hill Book Co., New York, 1967. pp 322, 366,
368-370.
9. Fischer and Porter Company. Instruction Bulletin for Model 17T1010
Amperometric Titrator. Instruction Bulletin 17T1010, Revision 1.
Fischer and Porter Company, P.O. Box C-5010, Warminster, Pennsylvania.
43
-------
10. Orion Research Incorporated. Instruction Manual, Residual Chlorine
Electrode, Model 97-70. IM97-70/776, Orion Research Incorporated,
380 Putnam Avenue, Cambridge, Mass 02139, 1977.
11. Frazer, J. W., S. Walter and L. P. Rigdon. Equivalence-Point
Determination for Automated Potentiometric Titrations with Ion-
Selective Electrodes. Anal. Chem., 49(8): 1250-1255, 1977.
12. Frazer, J. W., A. M. Kray, W. Selig, and R. Lim. Interactive-
Experimentation Employing Ion-Selective Electrodes. Anal. Chem.,
47(6): 869-875, 1975.
13. Gould, E. 3. Inorganic Reactions and Structure. Holt, Rinehart,
and Winston, Inc., New York, 1955. p 214.
14. Howe, L. H. Trace Analysis of Arsenic by Colorimetry, Atomic
Absorption, and Polarography. EPA-600/7-77-036, E-EP-77-3,
Tennessee Valley Authority, Chattanooga, Tennessee, 1977.
pp 34-35.
15. Manabe, R. M. Measurement of Residual Chlorine Levels in Cooling
Water—Amperometric Method. EPA-660/2-73-039, U.S. Environmental
Protection Agency, Corvallis, Oregon, 1974. pp 1, 15, 20-21, 42-43.
16. American Public Health Association. Standard Methods for the
Examination of Water and Wastewater. 14th Edition. American Public
Health Association, Washington, D.C., 1976. pp 187, 329-332, 550-554.
17. Rigdon, L. P. Operating Instructions for the EPA Microcomputer
Controlled Potentiometric Analyzer (Preliminary). LPR:78-79,
Lawrence Livermore Laboratory, Livermore, California, September 1978.
18. Guide for Use of Terms in Reporting Data in Analytical Chemistry.
Anal. Chem., 47(14): 2527, 1975.
19. Miller, I. and J. E. Freund. Probability and Statistics for Engineers.
Prentice-Hall, Inc., Englewood Cliffs, New Jersey 1965. pp 162-170,
177, 399.
20. Sillen, G. L., and A. E. Martell. Stability Constants of Metal-Ion
Complexes. Special Publication No. 17. The Chemical Society,
Burlington House, W. 1, London, 1964. pp 191-192, 338-340, 343.
21. Day, R. A., Jr., and A. L. Underwood. Quantitative Analysis.
Prentice-Hall, Inc., Englewood Cliffs, New Jersey, 1958. pp 147-150.
22. Turner, Ralph, R. Oxidation State of Arsenic in Coal Ash Leachate.
Environ. Sci. Technol., 15(9): 1067-1066, 1981.
44
-------
23. Campbell, M. M. and G. Johnson. Chloramine-T and Related N-Halogeno-N-
Metallo Reagents. Chem. Rev., 78(1): 65-80, 1978.
24. Glaser, J. A., D. L. Forest, G. D. McKee, S. A. Quave, and W. L.
Budde, "Trace Analysis for Wastewaters," Environ. Sci. Technol.,
15(12): 1426-1435, 1981.
25. Texas Instruments. Program Manual ST1 Statistics Library. 1220801-1,
Texas Instruments, Inc., Dallas, Texas, pp 26-33.
26. Youden, W. J., and E. H. Steiner. Statistical Manual of the Association
of Official Analytical Chemists. Association of Official Analytical
Chemists, Washington, DC, 1975. p 23.
45
-------
GLOSSARY
free chlorine (FC): Also known as free available chlorine. When
chlorine reacts with water, hypocalorous acid and hydrochloric
acid are formed:
C12 + H20 -* HOC1 + HC1
hypochlorus hydrochloric
acid acid
The hypochlorous acid further dissociates into hypochlorite ion
and hydrogen ion.
HOC1 -»• OCl" + H+
FC is the sum of the HOC1 and OCl fractions in the chlorinated
solutions.
combined chlorine (CC): When HOC1 combines with ammonia in the
water, mono-, di-, and trichloramine are formed. The sum of
these three fractions is called combined chlorine.
HOC1 + NH -* NH2C1 + H20
monochloramine
HOC1 + NH2C1 -»• NHC12 + ^0
dichloramine
HOC1 + NHC12 -> NCI + H20
trichloramine
total chlorine (TC): The sum of the FC and CC and fractions.
chlorine demand: The difference between the amount of chlorine
added to the water and the amount of TC remaining at the end of
a specified contact period.
chloramine-T: Chloramine-T trihydrate. Also known as sodium
N-chloro-p-toluenesulfonamide trihydrate.
-------
gran function (Gi) and volume addition (i): These terms are defined
in reference 7.
mean, standard deviation (SD), and relative standard deviation (RSD):
These terms are defined in reference 18.
% bias: Also known as percentage accuracy; which is defined in
reference 18.
maximum detection limit: A subjective judgment of the highest con-
centration that can be determined with useful accuracy and precision
for the intended application of the analytical method.
minimum detection limit: A subjective judgment of the lowest concen-
tration that can be determined with useful accuracy and precision for
the intended application of the analytical method.
pH: Negative logarithm of hydrogen ion concentration.
steam plant: Coal-burning electric power plant.
47
-------
Figures
-------
75 100 125 150 175
MEAN TOTAL CHLORINE CONCENTRATION, Jig/I
200
225
Figure 1. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by MC method of total chlorine from deion-
ized water spiked with chloramine-T.
49
-------
a
a
51 10
a
z
2.5
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION, jig/I
200
225
Figure 2. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by MC method of total chlorine from deionized water
spiked with chloramine-T.
-------
75 100 125 150 175
TOTAL CHLORINE CONCENTRATION ADDED, ,19/1
200
Figure 3.
Percentage mean bias against total chlorine concentration added
in micrograms per liter for experimental recovery by MC method of
total chlorine from deionized water spiked with chloramine-T.
1 8.0
z
o
P 7.0
a 5.0
a
3 3.0
< 2.0
Figure 4.
25
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION,
250
Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by AT method of total chlorine from deion-
ized water spiked with chloramine-T.
51
-------
25
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION, ng/l
225
250
Figure 5. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by AT method of total chlorine from deionized water
spiked with chloramine-T.
52
-------
+20
75 100 125 150 175
TOTAL CHLORINE CONCENTRATION ADDED, H9/I
Figure 6. Percentage mean bias against total chlorine concentration added
in micrograms per liter for experimental recovery by AT method of
total chlorine from deionized water spiked with chloramine-T.
53
-------
9.0
0
Q
3 6.0
0
I 5.0
to
UJ
z
5 4.0
3
*. 3.0
1.0
75 100 125 150 175
MEAN TOTAL CHLORINE CONCENTRATION, ng/l
250
Figure 7. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by MC method of total chlorine from deion-
ized water spiked with calcium hypochlorite.
-------
12.5
Q
Q
a
<
a
z
75 100 125 150 175
MEAN TOTAL CHLORINE CONCENTRATION, mg/1
Figure 8. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by MC method of total chlorine from deionized water
spiked with calcium hypochlorite.
55
-------
+20
+10
m
z
5 -10
•20
75 100 125 150
TOTAL CHLORINE CONCENTRATION ADDED, ji
250
Figure 9. Percentage mean bias against total chlorine concentration added
in micrograms per liter for experimental recovery by MC method
of total chlorine from deionized water spiked with calcium
hypochlorite.
-------
:s.o
EC
3
X
25
75 100 125 150 175
MEAN TOTAL CHLORINE CONCENTRATION, (is/I
Figure 10. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by AT method of total chlorine from deion-
ized water spiked with calcium hypochlorite.
57
-------
35
30
z
g
i-
10
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION,
Figure 11. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by AT method of total chlorine from deionized water
spiked with calcium hypochlorite.
58
-------
-10
75 100 125 150
TOTAL CHLORINE CONCENTRATION ADDED,
Figure 12. Percentage mean bias against total chlorine concentration added
in micrograms per liter for experimental recovery by AT method
of total chlorine from deionized water spiked with calcium
hypochlorite.
-------
20
18
Z
EC 8
150 200 250 300
MEAN TOTAL CHLORINE CONCENTRATION,
350
450
500
Figure 13.
Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by MC method of total chlorine from deion-
ized water spiked with 50% (w/w) mixture of calcium hypochlorite
and chloramine-T as chlorine.
z
o
Q
Q 7.5
2.5
150 200 250 300
MEAN TOTAL CHLORINE CONCENTRATION,
350
Figure 14.
Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by MC method of total chlorine from deionized water
spiked with 50% (w/w) mixture of calcium hypochlorite and
chloramine-T as chlorine.
60
-------
5 .2.5
Z
50
ISO 200 250 300 350
TOTAL CHLORINE CONCENTRATION ADDED. wg/l
500
Figure 15. Percentage mean bias against total chlorine concentration added
in micrograms per liter for experimental recovery by MC method of
total chlorine from deionized water spiked with 50% (w/w) mixture
of calcium hypochlorite and chloramine-T as chlorine.
61
-------
z
g
F-
Q 7
a
ct
I 6
z
ir
3
i
o
J
50
100
ISO 200 250 300
MEAN TOTAL CHLORINE CONCENTRATION, /
350
Figure 16. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration for experimental recovery by AT
method of total chlorine from deionized water spiked with 50%
(w/w) mixture of calcium hypochlorite and chloramine-T as
chlorine.
40 -
'.35
Z
o
< 30
a
Q 25
a.
> 15
§
£ 10
50
150 200 250 300 350
MEAN TOTAL CHLORINE CONCENTRATION, «g/l
500
Figure 17.
Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by AT method of total chlorine from deionized water
spiked with 50% (w/w) mixture of calcium hypochlorite and
chloramine-T as chlorine.
62
-------
-10
-20
50
150 200 250 300 350
TOTAL CHLORINE CONCENTRATION ADDED, pg/|
400
Figure 18. Percentage mean bias against total chlorine concentration added
in micrograms per liter for experimental recovery by AT method of
total chlorine from deionized water spiked with 50% (w/w) mixture
of calcium hypochlorite and chloramine-T as chlorine.
Figure 19,
75 100 125 150 175
MEAN TOTAL CHLORINE CONCENTRATION. jig/I
200
250
Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by MC method of total chlorine from raw
Ross Landing surface river water sample collected October 3,
1980, analyzed October 7-12, 1980, spiked with chloramine-T.
63
-------
50
75 100 125 150 175
MEAN TOTAL CHLORINE CONCENTRATION, /jg/l
200
250
Figure 20. Percentage relative standard deviation in micrograms per liter
against mean total chlorine concentration in micrograms per liter
for experimental recovery by MC method of total chlorine from raw
Ross Landing surface river water sample collected October 3,
1980, analyzed October 7-12, 1980, spiked with chloramine-T.
64
-------
75 100 125 150 17S
MEAN TOTAL CHLORINE CONCENTRATION, Hg/l
Figure 21. Total chlorine standard deviation against mean total chlorine
concentration in micrograms per liter for experimental recovery
by AT method of total chlorine from raw Ross Landing surface
river water sample collected October 3, 1980, analyzed
October 7-12, 1980, spiked with chloramine-T.
65
-------
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION,
Figure 22. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by AT method of total chlorine from raw Ross Landing
surface river water sample collected October 3, 1980, analyzed
October 7-12, 1980, spiked with chloramine-T.
66
-------
9.0
z 8.0
O
> 7-°
Q
Q
CC 6.0
I
< 5.0
l/>
= 4.0
I
-------
6.0
O 4.0
1
>
UJ
Q
O
$3.0
a
52.0
2
cr
1.0.
50
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION,
175
200
Figure 24. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by MC method of total chlorine from raw Ross Landing
surface river water sample collected October 3, 1980, analyzed
October 12, 1980, spiked with calcium hypochlorite.
68
-------
5.0
§ 4.0
3.0
2.0
1.0
25
50
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION
175
200
225
250
Figure 25.
Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration for experimental recovery by AT
method of total chlorine from raw Ross Landing surface river
water sample collected October 3, 1980, analyzed October 12,
1980, spiked with calcium hypochlorite.
25
75 100 125 150 175
MEAN TOTAL CHLORINE CONCENTRATION, pg/l
225
Figure 26.
Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by AT method of total chlorine from raw Ross Landing
surface river water sample collected October 3, 1980, analyzed
October 12, 1980, spiked with calcium hypochlorite.
69
-------
1 8.0
z"
o
f: 7.0
O 6.0
D
IX
Q 5.0
cr
3 3.0
1.0
25
50
75 100 125 150 175
MEAN TOTAL CHLORINE CONCENTRATION, /ag/l
Figure 27. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by MC method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of unchlorinated and
chlorinated Kingston Steam Plant condenser cooling river water
samples analyzed October 15-16, 1980.
70
-------
15.0
Q
O
< 10.0
Q
Z
5.0
50
75 100 125 150 175
MEAN TOTAL CHLORINE CONCENTRATION, W9/I
200
Figure 28. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by MC method of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and chlorinated
Kingston Steam Plant condenser cooling river water samples
analyzed October 15-16, 1980.
71
-------
10.0
9.0
z
o
6.0 -
(C
3 3.0
2.0 h
1.0
25 50 75 100 125 150 175 200 225 250
MEAN TOTAL CHLORINE CONCENTRATION, jij/l
Figure 29. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by AT method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of unchlorinated and
chlorinated Kingston Steam Plant condenser cooling river water
samples analyzed October 15-16, 1980.
25.0
UJ 15.0
O
O
1C
z
Si 10.0
Figure 30.
50
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION, j/g/l
200
225
250
Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by AT method of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and chlorinated
Kingston Steam Plant condenser cooling river water samples
analyzed October 15-16, 1980.
72
-------
1.0 -
25
SO
75 100 125 150 175
MEAN TOTAL CHLORINE CONCENTRATION. jig/I
225
250
Figure 31. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by MC method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of unchlorinated and
chlorinated Kingston Steam Plant condenser cooling river water
samples analyzed October 15-16, 1980, spiked with calcium
hypochlorite.
73
-------
17.5
12.5
UJ 10
O
55 7.5
UJ
P
UJ
a
5.0
50
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION,
175
225
Figure 32. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by MC method of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and chlorinated
Kingston Steam Plant condenser cooling river water samples
analyzed October 15-16, 1980, spiked with calcium hypochlorite.
74
-------
50
75 100 121 150 175
MEAN TOTAL CHLORINE CONCENTRATION, Ji9/l
225
Figure 33. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by AT method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of unchlorinated and
chlorinated Kingston Steam Plant condenser cooling river water
samples analyzed October 15-16, 1980, spiked with calcium
hypochlorite.
75
-------
5.0-
2.5 •
25
50
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION,
200
225
Figure 34. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by AT method of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and chlorinated
Kingston Steam Plant condenser cooling river water samples
analyzed October 15-16, 1980, spiked with calcium hypochlorite.
76
-------
13.0
8.0
UI 6.0
3.0
1.0
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION,
175
200
225
250
Figure 35. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by MC method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of unchlorinated and
chlorinated Shawnee Steam Plant condenser cooling river water
samples analyzed October 21-22, 1980.
77
-------
100
25
75 100 125 150 175
MEAN TOTAL CHLORINE CONCENTRATION, jig/I
Figure 36. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by MC method of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and chlorinated
Shawnee Steam Plant condenser cooling river water samples
analyzed October 21-22, 1980.
78
-------
11.0
£ 10.0
z
o
8.0
7.0
LLI 6.0
I ,.,
3.0
2.0 -
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION, ^
200
250
Figure 37. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by AT method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of unchlorinated and
chlorinated Shawnee Steam Plant condenser cooling river water
samples analyzed October 21-22, 1980.
79
-------
24
22
18
;10
25
50
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION,
225
250
Figure 38. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by AT method of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and chlorinated
Shawnee Steam Plant condenser cooling river water samples
analyzed October 21-22, 1980.
80
-------
17.5
IS
?12.5
<
>
LU
Q 10
a
a
a:
3
0
J
5.0 -
50
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION, fig/1
200
225
250
Figure 39. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by MC method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of unchlorinated and
chlorinated Shawnee Steam Plant condenser cooling river water
samples analyzed October 21-22, 1980, spiked with calcium
hypochlorite.
81
-------
17.5
12.5
5.0
2.5
25
50
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION,
200
225
Figure 40. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by MC method of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and chlorinated
Shawnee Steam Plant condenser cooling river water samples
analyzed October 21-22, 1980, spiked with calcium hypochlorite.
82
-------
17.5
Q 10
Q
-------
100 ,
25
75 100 125 ISO
MEAN TOTAL CHLORINE CONCENTRATION,
175
200
Figure 42. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by AT method of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and chlorinated
Shawnee Steam Plant condenser cooling river water samples
analyzed October 21-22, 1980, spiked with calcium hypochlorite.
84
-------
z"
o
9.0
O 8.0
a
5
D 7.0
6.0
3.0
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION,
175
200
225
250
Figure 43. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by MC method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of unchlorinated and
chlorinated Allen Steam Plant condenser cooling river water
samples analyzed October 28-29, 1980.
85
-------
17.5
15
7.5 •
75 100 125 150 175
MEAN TOTAL CHLORINE CONCENTRATION, jig/'
200
250
Figure 44. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by MC method of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and chlorinated Allen
Steam Plant condenser cooling river water samples analyzed
October 28-29, 1980.
86
-------
17.5r
J12.5-
Q 10
0
a
5.0-
50
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION,
Figure 45. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by AT method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of unchlorinated and
chlorinated Allen Steam Plant condenser cooling river water
samples analyzed October 28-29, 1980.
87
-------
100
z
g
P
<
>
UJ
O
Q
O
Z
75 100 125 150 175
MEAN TOTAL CHLORINE CONCENTRATION, jig/I
200
225
Figure 46. Percentage relative standard deviation against mean concentration
for experimental recovery by AT method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of unchlorinated and
chlorinated Allen Steam Plant condenser cooling river water
samples analyzed October 28-29, 1980.
88
-------
12.5-
Q
O
tt
5.0-
25
75 100 125 150 175
MEAN TOTAL CHLORINE CONCENTRATION. tig/I
200
250
Figure 47. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by MC method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of unchlorinated and
chlorinated Allen Steam Plant condenser cooling river water
samples analyzed October 28-29, 1980, spiked with calcium
hypochlorite.
89
-------
O 20
l»
ui
(E
5.0.
25
50
75 100 125 ISO
MEAN TOTAL CHLORINE CONCENTRATION,
200
225
250
Figure 48. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by MC method of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and chlorinated Allen
Steam Plant condenser cooling river water samples analyzed
October 28-29, 1980, spiked with calcium hypochlorite.
90
-------
30
Z
o
0
0
tr
2,5
CC
3 10 |-
O
-I
I
75 100 125 ISO
MEAN TOTAL CHLORINE CONCENTRATION,
175
200
225
2SO
Figure 49. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by AT method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of unchlorinated and
chlorinated Allen Steam Plant condenser cooling river water
samples analyzed October 28-29, 1980, spiked with calcium
hypochlorite.
91
-------
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION,
250
Figure 50. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by AT method of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and chlorinated Allen
Steam Plant condenser cooling river water samples analyzed
October 28-29, 1980, spiked with calcium hypochlorite.
92
-------
z
o
Q 10
O
a:
z
a.
3
u
< 5.0
O
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION, (jg/l
Figure 51. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by MC method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of unchlorinated and
chlorinated John Sevier Steam Plant condenser cooling river water
samples analyzed November 4-5, 1980.
93
-------
25
§20
a
a
I"
: 10
150 200 250 300 350
MEAN TOTAL CHLORINE CONCENTRATION, ,ig/l
Figure 52. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by MC method of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and chlorinated John
Sevier Steam Plant condenser cooling river water samples analyzed
November 4-5, 1980.
94
-------
12.5
z
o
1-10
z
SO
100
150 200 250 300 350
MEAN TOTAL CHLORINE CONCENTRATION, H9/I
Figure 53. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by AT method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of unchlorinated and
chlorinated John Sevier Steam Plant condenser cooling river water
samples analyzed November 4-5, 1980.
95
-------
100
z
o
Q
O
Q;
< 10
a
z
ISO 200 250 300 350
MEAN TOTAL CHLORINE CONCENTRATION, ng/1
Figure 54. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by AT method of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and chlorinated John
Sevier Steam Plant condenser cooling river water samples analyzed
November 4-5, 1980.
96
-------
12.5
z
o
pio
>
ul
O
Q
tt
7.5
EC
35.0
2.5'
50
100 125 150
MEAN TOTAL CHLORINE CONCENTRATION.
200
250
Figure 55. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by MC method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of unchlorinated and
chlorinated John Sevier Steam Plant condenser cooling river water
samples analyzed November 4-5, 1980, spiked with calcium
hypochlorite.
97
-------
100
z
o
UJ
0
Q
a.
< ,0
z
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION,
225
250
Figure 56. Percentage relative standard deviation against mean total chlo-
rine concentration in micrograms per liter for experimental
recovery by MC method of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and chlorinated John
Sevier Steam Plant condenser cooling river water samples analyzed
November 4-5, 1980, spiked with calcium hypochlorite.
98
-------
14
Q 8.0
a
a:
UJ 6.0
E
o
4.0
25
50
75 100 125 150
MEAN TOTAL CHLORINE CONCENTRATION,
175
200
225
250
Figure 57. Total chlorine standard deviation in micrograms per liter against
mean total chlorine concentration in micrograms per liter for
experimental recovery by AT method of total chlorine from raw
unchlorinated, chlorinated, and mixtures of unchlorinated and
chlorinated John Sevier Steam Plant condenser cooling river water
samples analyzed November 4-5, 1980, spiked with calcium
hypochlorite.
99
-------
50 -
Z
O 40
Q
Q
IT
< 30
Q
10
100
150
175
250
MEAN TOTAL CHLORINE CONCENTRATION,
Figure 58. Percentage relative standard deviation against mean total chlo-
rine concentration in micrgrams per liter for experimental
recovery by AT method of total chlorine from raw unchlorinated,
chlorinated, and mixtures of unchlorinated and chlorinated John
Sevier Steam Plant condenser cooling river water samples analyzed
November 4-5, 1980, spiked with calcium hypochlorite.
100
-------
Tables
101
-------
TABLE 1. SPIKING EXPERIMENTS WITH CHLORAMINE-T OR
CALCIUM HYPOCHLORITE FOR PRECISION AND ACCURACY TESTS
OF THE MC METHOD
o
10
o
Chlorine
spiking
solution
(mg/1)
12.5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
Volume of
solution
spiked
(ml)
50
50
50
50
50
50
50
50
50
Volume of Final
spiking concentration
solution, of chlorine
(yl) (ug/1)
0
20
40
100
200
400
600
800
1000
0
5
10
25
50
99
148
197
245
Volume of
Aliquot for deionized
dilution water added
(ml) (ml)
None None
None None
None None
None None
None None
None None
20 30
20 30
10 40
Expected
concentration
of chlorine Dilution
(pg/1) factor
Same
Same
Same
Same
Same
Same
59
79
49
None
None
None
None
None
None
2.5
2.5
5
Chloramine-T or calcium hypochlorite as chlorine.
-------
TABLE 2. SPIKING EXPERIMENTS WITH CHLORAMINE-T OR CALCIUM HYPOCHLORITE
FOR PRECISION AND ACCURACY TESTS OF THE AT METHOD
Chlorine
spiking
solution
(mg/1)
50
50
50
50
50
50
50
50
50
Volume of Volume of Final
solution spiking concentration
spiked solution, of chlorine
(ml)
-------
TABLE 3. SPIKING EXPERIMENTS WITH 50 % (w/w) MIXTURE
OF CHLORAMINE-T AND CALCIUM HYPOCHLORITE AS CHLORINE FOR PRECISION
AND ACCURACY TESTS OF THE MC METHOD
o
Chlorine
spiking
solution
(mg/1)
12.5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
Volume of
solution
spiked
(ml)
50
50
50
50
50
50
50
50
50
Volume of
spiking
solution
(yi)
0
20
40
100
200
300
400
500
1000
Final ,
o
concentration
of chlorine
(yg/D
0
10
20
50
99
148
197
245
481
Aliquot for
dilution
(ml)
None
None
None
None
None
20
20
10
5
Volume of
deionized
water added
(ml)
None
None
None
None
None
30
30
40
45
Expected
concentration
of chlorine
(Pg/1)
None
Same
Same
Same
Same
59
79
49
48
Dilution
factor
None
None
None
None
None
2.5
2.5
5
10
Chloramine-T or calcium hypochlorite as chlorine.
Equal volumes of chloramine-T and calcium hypochlorite spiking solutions were added to yield twice
the volumes listed. The sum of the two volumes was used in calculating the final concentration.
-------
o
Ui
TABLE 4. SPIKING EXPERIMENTS WITH 50 % (w/w) MIXTURE OF CHLORAMINE-T
AND CALCIUM HYPOCHLORITE AS CHLORINE FOR PRECISION AND ACCURACY
TESTS OF THE AT METHOD
*a
Chlorine
spiking
solution
(mg/1)
50
50
50
50
50
50
50
50
50
Volume of
solution
spiked
(ml)
200
200
200
200
200
200
200
200
200
Volume of
spiking
solution,
(vD
0
20
40
100
200
300
400
500
1000
Final
concentration
of chlorine
(yg/D
0
10
20
50
100
150
199
249
495
Aliquot for
dilution
(ml)
None
None
None
None
None
None
None
None
None
Volume of
deionized
water added
(ml)
None
None
None
None
None
None
None
None
None
Expected
concentration
of chlorine
(yg/i)
Same
Same
Same
Same
Same
Same
Same
Same
Same
Chloramine-T or calcium hypochlorite as chlorine.
Equal volumes of Chloramine-T and calcium hypochlorite spiking solutions were added to yield twice
the volumes listed. The sum of the two volumes was used in calculating the final concentration.
-------
TABLE 5. PREPARATION OF SOLUTIONS FOR TESTS OF CHEMICAL
INTERFERENCE WITH MC METHOD
Stock spiking
solution
Chromium(VI)
Mercury (II)
Bromide
Zinc(II)
Copper (II)
Iron(III)
Arsenic (III)
Arsenic (III)
Arsenic (III)
Arsenic (V)
Arsenic (V)
Manganese (VII)
Concentration
of spiking
solution
(mg/1)
50
50
50
100
50
1000
50
50
50
50
50
100
Volume of
spiking
solution
(Ml)
500
10
100
500
500
500
500
200
50
50
50
500
Volume
of solution
spiked
(ml)
50
50
50
50
50
50
50
50
50
50
50
50
Final
concentration
(yg/D
500
10
100
990
500
9900
500
200
50
200
50
990
106
-------
TABLE 6. AT METHOD3 RECOVERY OF TOTAL CHLORINE FROM DEIONIZED WATER
SPIKED WITH 249 yg/1 OF CHLORAMINE-T AS CHLORINE
WITH 0.00564N AND 0.001128N PHENYLARSINE OXIDE
Replicate
Number
1
2
3
1
2
3
Total
Chlorine
(yg/1) Mean SD RSD (%)° BIAS (%)
Experiment 6.1 Titration with 0.00564N (0.20 mg Cl^/ml) PAO
250
250 250 0 0 1.2
250
Experiment 6.2 Titration with 0.001128N (0.04 mg Cl0/ml) PAO
210
210 209 2.3 1.1 -15.4
206
Q
AT is forward amperometric titration method.
SD is standard deviation.
RSD is standard deviation expressed as percentage of mean.
-------
TABLE 7. EXPERIMENTAL RECOVERY BY MC AND AT METHODS OF TOTAL
CHLORINE FROM DEIONIZED WATER SPIKED WITH CHLORAMINE-T
a,b
Total
Chlorine
Replicate
Number
(yg/D
MC
AT
Mean SD RSD (%)
MC AT MC AT MC AT
Bias (%)
MC AT
t Value
MC AT
SAC
MC
95%?
AT
Experiment 7.1 Unspiked, Raw Water
c,d
o
oo
1
2
3
4
5
6
7
3.5
0.9
2.2
1.5
1.5
0.6
1.7
<20
<20
<20
<20
<20
<20
<20.
-1.70
0.95
56.9
Experiment 7.2 1.25 yg/1 Spike
1
2
3
4
5
6
7
-0.2
0.6
0.6
0.6
0.5
-0.1
0.6
0.37
0.36
97.3
-70.4
6.29
YES
continued
-------
TABLE 7. CONTINUED
o
vo
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine
(yg/1) Mean SD RSD (%) Bias (%) t Value SAC 95%?
MC AT MC AT MC AT MC AT MC AT MC AT MC AT
Experiment 7.3 2.5 yg/1 Spike
i.O
2.8
1.8
2.4
2.0
2.3
2-5>
) 2.23 0.38 17.0 -10.8 1.88 NO
Experiment 7.4 5 yg/1 Spike
3>1N
3.4
3.2
3.1
3.8
3.6
3.0
f 3.31 0.30 9.06 -33.8 14.9 YES
continued
-------
TABLE 7. CONTINUED
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine
(yg/1) Mean SD RSD (%) Bias (%) t Value SAC 95%?
MC
8.2
8.4
7.5
8.4
8.4
7.5
8.4
24
24
24
24
23
24
24
AT MC AT MC AT MC AT MC AT MC AT MC AT
Experiment 7.5 10 yg/1 Spike
•v
<20
<20
<20
<20
<20
<20
<20j
\ 8.11 0.43 5.30 -18.9 11.6 YES
Experiment 7.6 25 yg/1 Spike
30*1
30
30
20
30
30
30,
\ 23.9 28.6 0.4 3.8 1.7 13.3 -4.4 14.4 7.3 2.5 YES YES
continued
-------
TABLE 7. CONTINUED
Total
Chlorine
Replicate
Number
(yg/D
MC
AT
Mean
MC AT
SD RSD
MC AT MC
(%)
AT
Bias
MC
(%) t Value
AT MC AT
SAC 95%?
MC AT
1
2
3
4
5
6
7
49
51
53
52
52
52
50
50
50
50
50
50
50
50
Experiment 7.7 50 yg/1 Spike
51.3 50 1.4 0 2.7 0 2.6 0 2.5
YES
1
2
3
4
5
6
7
111
109
109
109
106
107
109
100
100
100
100
100
100
90^
Experiment 7.8 100 yg/1 Spike
109 99 1.6 3.8 1.5 3.8 9.0 -1.0 15 0.7 YES NO
continued
-------
TABLE 7. CONTINUED
Total
Chlorine
Replicate
Number
(yg/D
MC
AT
Mean SD RSD (%)
MC AT MC AT MC AT
Bias (%)
MC AT
t Value
MC AT
SAC
MC
95%?
AT
1
2
3
4
5
6
7
Experiment 7.9 247 yg/1 Spike
332
297
326
287
303
283
297
304
18.7
6.2
23.1
8.1
YES
1
2
3
4
5
6
7
Experiment 7.10 481 yig/1 Spike
671
654
903
632
715
126
17.6
48.6
3.7
YES
continued
-------
TABLE 7. CONTINUED
u>
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine
(yg/1) Mean SD RSD (%) Bias (%) t Value SAC 95%?
MC
152
160
155
152
152
151
149
198
208
204
198
200
194
196
AT MC AT MC AT MC AT MC AT MC AT MC AT
Experiment 7.11 149 yg/1 Spike6
•V
145
150
140
150
150
150
150J
153 148 3.6 3.9 2.4 2.6 2.7 -0.7 2.9 0.7 YES NO
Experiment 7.12 198 yg/1 Spike6
•v
190
200
200
200
210
200
195.
> 200 199 4.8 6.1 2.4 3.1 1.0 0.5 1.1 0.4 NO NO
continued
-------
TABLE 7. CONTINUED
Total
Chlorine
Replicate
Number
1
2
3
4
5
6
7
(yg/1) Mean SD RSD (%) Bias (%) t Value SAC 95%?
MC
246
250
248
247
251
250
248
AT MC AT MC AT MC AT MC AT MC AT MC AT
Experiment 7.13 247 yg/1 Spike6
230^
250
240
240
250
250
240J
> 249 243 1.8 7.6 0.7 3.1 0.8 -1.6 2.9 0.7 YES NO
and AT are microprocessor-controlled ion selective electrode and forward amperometric titration
methods, respectively. SD is standard deviation, % RSD is standard deviation expressed as percen-
tage of mean, and SAC 95%? is significant at 95% confidence.
Spiked solution was not diluted before analysis by MC to establish linear concentration range.
This solution was background for experiments 7.2-7.13.
An additional significant figure was retained for use in calculating a more precise estimate of
minimum detection limit by MC.
"Spiked solution was diluted before analysis by MC.
-------
TABLE 8. EFFECTS OF pH, CHEMICAL INTERFERENCE, AND
PYROPHOSPHATE (FOR OVERCOMING CHEMICAL INTERFERENCE) ON
EXPERIMENTAL RECOVERY BY MC METHOD OF TOTAL CHLORINE FROM DEIONIZED
WATER SPIKED WITH 50 yg/1 OF CHLORAMINE-T AS CHLORINE13
Replicate
Number
Total
Chlorine
(yg/1) Mean
SAC
SD RSD (%) Bias (%) t Value 95%?
1
2
3
4
5
6
7
52
53
51
51
54
52
49
Experiment 8.1. pH 3.6 Buffer
51.7
1.6
3.1
3.4
2.8
YES
1
2
3
4
5
6
7
52
52
52
53
52
53
54
Experiment 8.2 pH 3.8 Buffer
52.6
0.8
1.5
5.2
8.6
YES
1
2
3
4
5
6
7
49
52
55
51
51
55
50
Experiment 8.3 pH 4.0 Buffer
51.9
2.3
4.4
3.8
2.2
NO
continued
115
-------
TABLE 8. CONTINUED
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine SAC
(yg/1) Mean SD RSD (%) Bias (%) t Value 95%?
Experiment 8.4 pH 4.2 Buffer
50**
52
51
50
54
51
56 J
52.0 2.2 4.2 4.0 2.4 YES
Experiment 8.5 pH 4.4 Buffer
511
53
51
51
55
50
53j
52.0 1.7 3.3 4.0 3.1 YES
Experiment 8.6 500 pg/1 of Chromium (VI)
46^
48
53
47
47
48
44.
> 47.6 2.8 5.9 -4.8 2.3 NO
continued
116
-------
TABLE 8. CONTINUED
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine SAC
(yg/1) Mean SD RSD (%) Bias (%) t Value 95%?
Experiment 8.7 10 yg/1 of Mercury
48*1
52
49
52
50
48
5lJ
50.0 1.7 3.4 0 0 NO
Experiment 8.8 100 yg/1 of Bromide
48^
49
49
49
49
49
49j
48.9 0.4 0.8 -2.2 7.3 YES
Experiment 8.9 990 yg/1 Zinc (II)
521
50
51
52
50
49
55^
> 51.3 2.0 3.9 2.6 1.7 NO
continued
117
-------
TABLE 8. CONTINUED
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
7
1
2
3
4
5
6
7
Total
Chlorine SAC
(yg/1) Mean SD RSD (%) Bias (%) t Value 95%?
Experiment 8.10 1000 yg/1 of Zinc (II)
137"
154
111
113
144
140
124J
\ 132 16.2 12.3 164 13 YES
Experiment 8.11 10 mg/1 of Zinc (II)
54*"
59
57
57
'
) 56.8 2.1 3.7 13.6 8.6 YES
Experiment 8.12 10 mg/1 of Zinc (II) with 2 %
(w/v) Sodium Pyrophosphate Decahydrate
48>
52
50
50
51
49
50^
50.0 1.3 2.6 0 0 NO
continued
118
-------
TABLE 8. CONTINUED
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine SAC
(yg/1) Mean SD RSD (%) Bias (%) t Value 95%?
Experiment 8.13 500 yg/1 of Copper (II)
521
51
58
56
52
54
5lJ
• 53.4 2.7 5.1 6.8 3.3 YES
Experiment 8.14 500 mg/1 of Copper (II)
84*1
163
146
148
80
83
116>
117 35 30 134 5.1 YES
Experiment 8.15 5 mg/1 Copper (II)
•v
63
68
69
68
) 67.0 2.7 4.0 34 17 YES
continued
119
-------
TABLE 8. CONTINUED
Total
Replicate Chlorine SAC
Number (yg/1) Mean SD RSD (%) Bias (%) t Value 95%?
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Experiment 8.16 5 mg/1 of Copper (II) With 2 % (w/v)
Sodium Pyrophosphate Decahydrate
49^
48
49
49
49
49
48j
» 48.7 0.5 1.0 -2.6 6.9 YES
Experiment 8.17 1000 mg/1 of Iron(III)
987""
1203
1828
1370
1699
1568
1068^
1389 322 23.2 2680 11 YES
Experiment 8.18 1000 mg/1 of Iron(III) With 2 % (w/v)
Sodium Pyrophosphate Decahydrate
56*"
67
62
54
44
52
56j
55.9 7.3 13.1 11.8 2.1 NO
continued
120
-------
TABLE 8. CONTINUED
Total
Replicate Chlorine SAC
Number (yg/1) Mean SD RSD (%) Bias (%) t Value 95%?
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Experiment 8.19 9900 yg/1 of Iron(III)
68]
60
60
62
60
62
64j
62.3 2.9 4.7 24.6 11 YES
Experiment 8.20 9900 yg/1 of Iron(III) With 2 % (w/v)
Sodium Pyrophosphate Decahydrate
46^
50
48
50
50
49
5lJ
> 49.1 1.7 3.5 -1.8 1.4 NO
Experiment 8.21 5 mg/1 of Iron(III)
561
55
54
52
53
53
54j
> 54 1.3 2.4 8 8.1 YES
continued
121
-------
TABLE 8. CONTINUED
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine SAC
(pg/1) Mean SD RSD (%) Bias (%) t Value 95%?
Experiment 8.22 500 yg/1 of Arsenic (III)
196*"
246
-21
7
193
12
199,,
> 119 114 96 138 1.6 NO
Experiment 8.23 200 yg/1 of Arsenic (III)
741
65
77
74
78
76
82j
> 75.1 5.2 6.9 50 13 YES
Experiment 8.24 50 yg/1 of Arsenic (III)
40^1
41
43
48
48
44
41.
• 43.6 3.3 7.6 -12.8 5.1 YES
continued
122
-------
TABLE 8. CONTINUED
Total
Replicate Chlorine SAC
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
(yg/1) Mean SD RSD (%) Bias (%) t Value 95%?
Experiment 8.25 200 yg/1 of Arsenic (V)
5
-------
TABLE 9. EFFECT OF 2 % (w/v) SODIUM PYROPHOSPHATE
FOR OVERCOMING CHEMICAL INTERFERENCE ON EXPERIMENTAL RECOVERY BY MC METHOD
OF TOTAL CHLORINE FROM DEIONIZED WATER SPIKED WITH
CHLORAMINE-T AS CHLORINE3
Total
Replicate Chlorine
Number (ug/1)
Mean
SD
SAC
RSD (%) Bias (%) t Value 95%?
Experiment 9.1 Unspiked, Raw Water
1
2'
3
4
5
6
7
-5
-6
-6
-6
_y
-4
-4
-5.4
1.1
20.4
Experiment 9.2 5 yg/1 Spike
1
2
3
4
5
6
7
2
2
-2
-2
0
-3
-3
-0.9
2.2
244
-118
Experiment 9.3 10 yg/1 Spike
1
2
3
4
5
6
7
4.6
2.6
5 6'. 5
-54
5.5 YES
continued
124
-------
TABLE 9. CONTINUED
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine SAC
(yg/1) Mean SD RSD (%) Bias (%) t Value 95%?
Experiment 9.4 25 yg/1 Spike
211
22
22
20
24
21
2lJ
21.6 1.3 6.0 -13.6 6.9 YES
Experiment 9.5 50 yg/1 Spike
44^
45
46
45
44
44
46j
» 44.9 0.9 2.0 -10.2 15 YES
Experiment 9.6 100 yg/1 Spike
88^
91
86
105
93
92
94^
> 92.7 6.1 6.6 -7.3 3.2 YES
continued
125
-------
TABLE 9. CONTINUED
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine SAC
(yg/1) Mean SD RSD (%) Bias (%) t Value 95%?
Experiment 9.7 149 yg/1 Spike
124*"
123
137
154
135
127
154J
136 13.2 9.7 -8.7 2.6 YES
Experiment 9.8 198 yg/1 Spike
16(f|
174
158
179
172
174
179J
• 171 8.5 5.0 -13.6 8.4 YES
Experiment 9.9 247 yg/1 Spike
223^
240
213
212
227
211
228>
222 10.7 4.8 -10 6.2 YES
is microprocessor-controlled ion selective electrode method, SD is
standard deviation, % RSD is standard deviation expressed as percentage of
mean, and SAC 95%? is significant at 95% confidence.
This solution was background for experiments 9.2-9.9.
126
-------
TABLE 10. EFFECT OF 2 % (W/v) SODIUM PYROPHOSPHATE
FOR OVERCOMING CHEMICAL INTERFERENCE ON EXPERIMENTAL RECOVERY BY
MC METHOD OF TOTAL CHLORINE FROM DEIONIZED WATER SPIKED WITH
CALCIUM HYPOCHLORITE AS CHLORINE3
Total
Replicate Chlorine
Number (yg/1)
Mean
SD
SAC
RSD (%) Bias (%) t Value 95%?
Experiment 10.1 Unspiked, Raw Water
1
2
3
4
5
6
7
-4.3
1.0
23.3
Experiment 10.2 5 yg/1 Spike
1
2
3
4
5
6
7
-80
Experiment 10.3 10 yg/1 Spike
1
2
3
4
5
6
7
5.7
0.8
14.0
-43
14
YES
continued
127
-------
TABLE 10. CONTINUED
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine SAC
(yg/1) Mean SD RSD (%) Bias (%) t Value 95%?
Experiment 10.4 25 yg/1 Spike
20^
20
18
19
18
18
16^
> 18.4 1.4 7.6 -26 12 YES
Experiment 10.5 50 yg/1 Spike
421
41
40
41
49
48
48j
> 44.1 4.0 9.1 -11.8 3.9 YES
Experiment 10.6 100 yg/1 Spike
96^
91
98
92
91
90
92 ,
. 92.9 3.0 3.2 -7.1 6.3 YES
continued
128
-------
TABLE 10. CONTINUED
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine SAC
(yg/1) Mean SD RSD (%) Bias (%) t Value 95%?
Experiment 10.7 149 yg/1 Spike
8l"
68
107
63
87
71
66j
> 78 15.5 19.9 -48 12 YES
Experiment 10.8 198 yg/1 Spike
1171
100
100
109
128
108
89j
> 107 12.7 11.9 -46 19 YES
flIC is microprocessor-controlled ion selective electrode method, SD is
standard deviation, % RSD is standard deviation expressed as percentage
of mean, and SAC 95%? is significant at 95% confidence.
This solution was background for experiments 10.2-10.8.
129
-------
TABLE 11. EXPERIMENTAL RECOVERY BY MC AND AT METHODS OF TOTAL
CHLORINE FROM DEIONIZED WATER SPIKED WITH CALCIUM HYPOCHLORITE AS CHLORINE2
Replicate
Number
Total
Chlorine
(yg/D
Mean
SD
RSD
Bias
MC
AT
t Value
MC AT MC AT MC AT MC AT MC AT
SAC 95%?
MC
AT
Experiment 11.1 Unspiked, Raw Water *
u>
o
1
2
3
4
5
6
7
-0.7
0.0
-1.2
-1.4
-1.6
-3.9
-2.5
<20
<20
<20
<20
<20
<20
<20
-1.61
1.27
78.9
Experiment 11.2 5 yg/1 Spike
1
2
3
4
5
6
7
3.6
3.9
3.5
4.6
3.7
3.0
3.3
<20
<20
<20
<20
<20
<20
<20
3.66
0.51
13.9
-26.8
6.95
YES
continued
-------
TABLE 11. CONTINUED
u>
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine
(yg/1) Mean SD RSD (%) Bias (%) t Value SAC 95%?
MC
7.9
8.3
8.1
7.6
7.6
7.3
8.6
22
23
21
21
20
25
25
AT MC AT MC AT MC AT MC AT MC AT MC AT
c d
Experiment 11.3 10 yg/1 Spike '
<20>
20
<20
<20
<20
<20
<20J
1 7.91 11.4 0.45 3.8 5.69 33.3 -20.9 14.0 12.3 1.0 YES
Experiment 11.4 25 yg/1 Spike
•v
20
30
20
25
30
30
30^
22.4 26.4 2.0 4.8 8.9 18.2 -10.4 5.6 3.4 0.2 YES
continued
-------
TABLE 11. CONTINUED
to
to
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine
(yg/1) Mean SD RSD (%) Bias (%) t Value SAC 95%?
MC
49
50
50
47
48
51
50
109
111
110
109
104
109
108
AT MC AT MC AT MC AT MC AT MC AT MC AT
Experiment 11.5 50 yg/1 Spike
50^
50
50
50
45
50
50^
> 49.3 49.3 1.4 1.9 2.8 3.9 -1.4 -1.4 1.3 1.0 NO NO
Experiment 11.6 100 yg/1 Spike
90*1
90
90
90
90
100
90J
\ 109 91 2.2 3.8 2.0 4.2 9.0 -9.0 11 6.3 YES YES
continued
-------
TABLE 11. CONTINUED
u>
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine
(yg/1) Mean SD RSD ^%) Bias (%) t Value SAC 95%?
MC
162
150
149
131
144
144
146
201
199
200
193
192
193
197
AT MC AT MC AT MC AT MC AT MC AT MC AT
Experiment 11.7 149 yg/1 Spike
150*]
150
150
140
150
150
145J
> 147 148 9.2 3.9 6.3 2.6 -1.3 -0.7 0.6 0.7 NO NO
Experiment 11.8 198 yg/1 Spike
190^
200
200
210
200
200
195J
> 196 199 3.7 6.1 1.9 3.1 -1.0 0.5 1.4 0.4 NO NO
continued
-------
TABLE 11. CONTINUED
u>
Total
Chlorine
Replicate
Number
1
2
3
4
5
6
7
(yg/1) Mean SD RSD (%) Bias (%) t Value SAC 95%?
MC
244
234
232
232
224
231
233
AT MC AT MC AT MC AT MC AT MC AT MC AT
Experiment 11.9 247 yg/1 Spike
230^
245
250
240
240
250
245J
^ 230 243 4.2 7.0 1.8 2.9 -6.9 -1.6 28 1.5 YES NO
and AT are microprocessor-controlled ion selective electrode and forward amperometric titration
methods, respectively. SD is standard deviation, % RSD is standard deviation expressed as percen-
tage of mean, and SAC 95%? is significant at 95% confidence.
This solution was background for experiments 11.2-11.9.
An additional significant figure was retained for use in calculating a precise estimate of minimum
detection limit by MC.
For the AT, results less than 20 yg/1 were included in the statistical calculations as 10 yg/1.
-------
TABLE 12. EXPERIMENTAL RECOVERY BY MC AND AT METHODS OF TOTAL CHLORINE
FROM DEIONIZED WATER SPIKED WITH 50 % (w/w) MIXTURE OF
CALCIUM HYPOCHLORITE AND CHLORAMINE-T AS CHLORINE3
Total
Chlorine
Replicate
Number
(yg/D
MC
AT
Mean SD RSD (%)
MC AT MC AT MC AT
Bias (%)
MC AT
t Value
MC AT
SAC
MC
95%?
AT
Experiment 12.1 Unspiked, Raw Water
b,c
u>
1
2
3
4
5
6
7
1.6
5.3
1.8
2.3
1.6
2.3
2.7
<20
<20
<20
<20
<20
<20
<2()
-2.51
1.30
51.8
Experiment 12.2 10 yg/1 Spike
1
2
3
4
5
6
7
9.8
9.3
8.9
9.0
8.7
8.6
9.4
<20
<20
<20
<20
<20
<20
<20^
9.10
0.42
4.61
-9.00
5.67
YES
continued
-------
TABLE 12. CONTINUED
u>
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7 ,
Total
Chlorine
(yg/1) Mean SD RSD (%) Bias (%) t Value SAC 95%?
MC
21
20
18
19
20
20
19
50
50
51
49
48
49
48
AT MC AT MC AT MC AT MC AT MC AT MC AT
Experiment 12.3 20 yg/1 Spike
<20>
<20
<20
20
20
20
<20j
\ 19.6 14.3 1.0 5.3 5.1 37.1 -2.0 -28.5 2.6 2.9 YES YES
Experiment 12.4 50 yg/1 Spike
501
45
50
50
50
45
50J
\ 49.3 49.0 1.1 2.4 2.2 4.9 -1.4 -2.0 1.7 1.1 NO NO
continued
-------
TABLE 12. CONTINUED
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine
(yg/1) Mean SD RSD (%) Bias (%) t Value SAC 95%?
MC
105
105
105
105
99
99
96
148
152
131
153
129
106
131
AT MC AT MC AT MC AT MC AT MC AT MC AT
Experiment 12.5 100 yg/1 Spike
100*1
100
100
100
95
100
looj
> 102 99 3.9 1.9 3.8 1.9 2.0 -1.0 1.4 1.4 NO NO
Experiment 12.6 149 yg/1 Spike
150^
150
155
140
150
150
155^
> 136 150 16.8 5.0 12.4 3.3 -8.7 0.7 2.0 0.5 NO NO
continued
-------
TABLE 12. CONTINUED
U)
00
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine
(yg/1) Mean SD RSD (%) Bias (%) t Value SAC 95%?
MC
209
204
207
208
197
196
196
245
251
247
247
238
254
246
AT MC AT MC AT MC AT MC AT MC AT MC AT
Experiment 12.7 198 yg/1 Spike
210*"
210
200
210
195
190
200J
> 202 202 5.9 8.1 2.9 4.0 2.0 2.0 1.8 1.3 NO NO
Experiment 12.8 247 yg/1 Spike
250^
240
250
250
260
250
240^
> 247 249 5.0 6.9 2.0 2.8 0 0.8 0 0.8 NO NO
continued
-------
TABLE 12. CONTINUED
VO
Total
Chlorine
Replicate
Number
1
2
3
4
5
6
7
(yg/1) Mean SD RSD (%) Bias (%) t Value SAC 95%?
MC
459
443
482
491
442
469
481
AT MC AT MC AT MC AT MC AT MC AT MC AT
Experiment 12.9 488 yg/1 Spike
430*"]
450
450
460
450
455
460J
> 467 451 19.4 10.2 4.2 2.3 -4.3 -7.6 2.9 9.6 YES YES
nC and AT are microprocessor-controlled ion selective electrode and forward amperometric titration
methods, respectively. SD is standard deviation, % RSD is standard deviation expressed as percen-
tage of mean, and SAC 95%? is significant at 95% confidence.
This solution was background for experiments 12.2-12.9.
Q
An additional significant figure was retained for use in calculating a precise estimate of minimum
detection limit by MC.
For the AT, results less than 20 yg/1 were included in the statistical calculations as 10 yg/1.
-------
TABLE 13. CHLORINE DEMAND OF RAW ROSS LANDING
SURFACE RIVER WATER SAMPLE COLLECTED
OCTOBER 3, 1980a
Contact Time
(Min)
1
1
1
1
1
5
5
5
5
5
10
10
10
10
10
15
15
15
15
15
30
30
30
30
30
60
60
60
60
60
Added
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
1.2
0.6
1.0
3.0
5.0
Chlorine Concentration, mg/1
Found
0.22
0.52
0.92
2.5
4.4
0.22
0.58
0.68
1.6
2.6
0.22
0.50
0.70
1.3
2.7
0.22
0.50
0.65
1.4
2.2
0.20
0.48
0.65
1.2
2.8
0.20
0.44
0.72
1.4
3.2
j_
Demand"
-0.02
0.08
0.08
0.5
0.6
-0.02
0.02
0.32
1.4
2.4
-0.02
0.1
0.3
1.7
2.3
-0.02
0.1
0.35
1.6
2.8
0.0
0.12
0.35
1.8
2.2
0.0
0.16
0.28
1.6
1.8
ERoss Landing is located on the Tennessee River in Chattanooga, Tennessee.
Demand is difference between total chlorine added initially as NaOCl-Cl-
and that found after elapsed contact time.
140
-------
TABLE 14. INORGANIC SPECIES CONCENTRATIONS,
SANITARY CHEMICAL CHARACTERISTICS, AND PHYSICAL PROPERTIES
OF TENNESSEE RIVER WATER COLLECTED AT ROSS LANDING,
CHATTANOOGA, TENNESSEE, OCTOBER 3, 1980
Characteristics
Value
Inorganics, yg/1
Arsenic
Chromium
Copper
Iron
Mercury
Manganese
Zinc
Solids, mg/1
Suspended
Dissolved
Chemical oxygen demand, mg/1
Total organic carbon mg/1
Nitrogen, mg/1 as N
Organic
Nitrite plus nitrate
Ammonia
Alkalinity, mg/1 as Ca CO
Total
Phenol
Color, mg/1 of Ft in the chloroplatinate standard
True
Apparent
Turbidity, mg/1 of natural Si
PH
< 2
< 1
< 10
190
< 0.2
30
< 10
3
110
8
2.6
0.34
< 0.01
64
< 1
4
13
7.1
7
141
-------
TABLE 15. EXPERIMENTAL RECOVERY BY MC AND AT METHODS OF TOTAL CHLORINE
FROM RAW ROSS LANDING SURFACE RIVER WATER SAMPLE COLLECTED
OCTOBER 3, 1980, SPIKED WITH CHLORAMINE-Ta
Total
Chlorine
Replicate
Number
(yg/D
MC
AT
Mean SD RSD (%)
MC AT MC AT MC AT
Total Chlorine
Spike (yg/1)
MC AT
Experiment 15.1 Unspiked, Raw Water
1
2
3
4
5
6
7
-9
-8
-7
-8
-5
-9
-9
<20
<20
<20
<20
<20
<20
7.9
1.5
19.0
Experiment 15.2° Analyzed October 12, 1980°
1
2
3
4
5
6
7
26
28
27
28
30
28
29
<20
<20
<20
20
<20
<20
28.0 11.4 1.3
3.8 4.6 33.3
50
50
Experiment 15.3 Analyzed October 8, 1980
1
2
3
4
5
6
7
34
33
33
35
34
32
32
<20
<20
<20
<20
<20
33.3
1.1
3.3
55
55
continued
142
-------
TABLE 15. CONTINUED
Replicate
Number
Total
Chlorine
(yg/D
MC AT
Mean
SD
MC
AT
MC
AT
RSD (%)
MC AT
Total Chlorine
Spike (ug/1)
MC AT
1
2
3
4
5
c c
Experiment 15.4 Analyzed October 7, 1980
36
37
35
36
37
35
36
<20>
20
20
20
20
<20
<20,
36.0 15.7 0.8 5.3 2.2 33.8
60
60
Experiment 15.5 Analyzed October 7, 1980
1
2
3
4
5
6
7
48
49
49
48
49
48
49
30
30
20
30
25
35
30
48.6 28.6 0.5
4.8 1.0 16.8
75
75
Experiment 15.6 Analyzed October 7, 1980
1
2
3
4
5
6
7
70
72
71
71
70
70
71
60
50
50
60
60
50
55
70.7 55.0 0.8
5.0 1.1 9.1
99
100
continued
143
-------
TABLE 15. CONTINUED
Total
Chlorine Total Chlorine
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
(yg/1) Mean SD RSD (%) Spike (yg/1)
MC
120
121
112
118
118
119
120
178
178
168
168
167
170
168
208
208
212
210
200
196
210
AT MC AT MC AT MC AT MC AT
Experiment 15.7 Analyzed October 7, 1980
lOO"'
100
110
100
100
100
100^
> 118 101 3.0 3.8 2.5 3.8 148 150
Experiment 15.8 Analyzed October 7, 1980
150^
150
150
150
145
150
150^
1 171 149 4.9 1.9 2.9 1.3 197 199
Experiment 15.9 Analyzed October 3, 1980
200**
200
200
195
200
210
200^
206 201 5.9 4.5 2.9 2.2 245 249
continued
144
-------
TABLE 15. CONTINUED
Replicate
Number
Total
Chlorine
(yg/D
MC AT
Mean
SD
MC
AT
MC
AT
RSD (%)
MC AT
Total Chlorine
Spike (yg/1)
MC AT
I
2
3
4
5
6
7
Experiment 15.10 Analyzed October 7, 1980
203 200
221 200
221 200
205 195
202 200
212 200
206 195
210 199 8.2 2.4 3.9 1.2 245
249
1
2
3
4
5
6
7
Experiment 15.11 Analyzed October 7, 1980
240 240
264 250
258 250
252 240
259 260
260 250
252 245
255 248 7.9 7.0 3.1 2.8
295
299
Ross Landing is located on the Tennessee River in Chattanooga, Tennessee.
MC and AT are microprocessor-controlled ion selective electrode and forward
amperometric titration methods, respectively. SD is standard deviation,
% RSD is standard deviation expressed as percentage of mean, and SAC 95%?
is significant at 95% confidence.
This solution was background for experiments 15.2-15.11.
-i
"For the AT, results less than 20 yg/1 were included in the statistical
calculations as 10 yg/1.
145
-------
TABLE 16. EXPERIMENTAL RECOVERY BY MC AND AT METHODS OF TOTAL CHLORINE
FROM RAW ROSS LANDING SURFACE RIVER WATER SAMPLE COLLECTED
OCTOBER 3, 1980, SPIKED WITH CALCIUM HYPOCHLORITE3'
Total
Chlorine Total Chlorine
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
(yg/1) Mean SD RSD (%) Spike (yg/1)
MC
-1.
-0.
1.
1.
-0.
-0.
1.
10.
9.
9.
10.
9.
9.
9.
25
26
27
26
26
27
27
8
6
3
2
1
4
0
3
8
4
0
1
1
4
AT MC AT MC AT MC AT MC AT
Experiment 16.1 Analyzed October 12, 1980°
<20>
<20
<20
<20
<20
<20
<20J
f 0.0857 1.14 1330 15 25
Experiment 16.2 Analyzed October 12, 1980°
<2
-------
TABLE 16. CONTINUED
Replicate
Number
Total
Chlorine
(yg/D
MC AT
Mean
SD
MC
AT
MC
AT
RSD (%)
MC AT
Total Chlorine
Spike (yg/1)
MC AT
Experiment 16.4 Analyzed October 12, 1980
1 52 50~
2 55 45
3 50 50
4 52 40 ) 52.4 45.7 1.6 4.5 3.1 9.8 80 125
5 52 40
6 54 50
7 52 45
1
2
3
4
5
6
7
92
94
92
92
92
93
92
Experiment 16.5 Analyzed October 12, 1980
100
90
95
100
100
95
100
92.4 97.1 0.8 3.9 0.9 4.0 129
180
1
2
3
4
5
6
7
150
148
141
141
140
147
147
Experiment 16.6 Analyzed October 12, 1980
145 149
4.1
1.9 2.8 1.3
202
249
continued
147
-------
TABLE 16. CONTINUED
Replicate
Number
Total
Chlorine
(yg/D
MC AT
Mean
SD
MC
AT
MC
AT
RSD (%)
MC AT
Total Chlorine
Spike (yg/1)
MC AT
1
2
3
4
5
6
7
Experiment 16.7 Analyzed October 12, 1980
200 180
194 190
195 190
196 195
194 190
199 195
194 190
> 196 190 2.5 5.0 1.3 2.6 344
304
1
2
3
4
5
6
7
Experiment 16.8 Analyzed October 12, 1980
232 250
231 250
232 250
232 250
230 250
208 250
240 250
) 229 250 9.9
0 4.3 0
383
434
ERoss Landing is located in Chattanooga Tennessee. MC and AT are micro-
processor-controlled ion selective electrode and forward amperometric
titration methods, respectively. SD is standard deviation and % RSD
is standard deviation expressed as percentage of mean.
Background for experiments 16.1-16.6 is shown by experiment 15.1, Table 15,
"An additional significant figure was retained for use in calculating a
precise estimate of minimum detection limit by MC.
148
-------
TABLE 17. CHLORINE DEMAND OF RAW KINGSTON STEAM PLANT
SAMPLE COMPOSED OF 100 % (v/v) UNCHLORINATED RAW RIVER WATER
INTAKE AT WATER SUPPLY PLANT, COLLECTED OCTOBER 15, 1980&
Contact Time
(Min)
1
1
1
1
1
5
5
5
5
5
10
10
10
10
10
15
15
15
15
15
30
30
30
30
30
60
60
60
60
60
Added
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
Chlorine Concentration, mg/1
Found
0.20
0.70
1.0
2.6
4.9
0.2
0.6
0.96
2.4
3.5
0.2
0.5
0.84
2.4
3.4
0.2
0.46
0.8
2.3
3.5
0.2
0.42
0.78
2.0
3.4
0.2
0.4
0.6
1.9
3.4
i
Demand"
0.00
-0.1
0.00
0.4
0.1
0.00
0.00
0.04
0.6
1.5
0.00
0.1
0.16
0.6
1.6
0.00
0.14
0.2
0.7
1.5
0.00
0.18
0.22
1.0
1.6
0.00
0.2
0.4
1.1
1.6
Kingston Steam Plant is located near Kingston, Tennessee.
Demand is difference between total chlorine added initially as NaOCl-Cl- and
that found after elapsed contact time.
149
-------
TABLE 18. INORGANIC SPECIES CONCENTRATIONS,
SANITARY CHEMICAL CHARACTERISTICS, AND PHYSICAL PROPERTIES
OF RAW KINGSTON STEAM PLANT SAMPLE COMPOSED OF 100 % (v/v)
UNCHLORINATED RAW RIVER WATER INTAKE AT WATER SUPPLY PLANT
COLLECTED OCTOBER 15, 1980H
Characteristics
Nitrogen, mg/1 as N
Organic
Nitrite plus nitrate
Ammonia
Alkalinity, mg/1 as CaCO_
Total
Phenol
Color, mg/1 of Pt in the chloroplatinate standard
True
Apparent
Turbidity, mg/1 of natural Si
PH
Value
Inorganics, yg/1
Arsenic
Chromium
Copper
Iron
Mercury
Manganese
Zinc
Solids, mg/1
Suspended
Dissolved
Chemical oxygen demand, mg/1
Total organic carbon mg/1
6
< 1
120
890
< 0.2
50
10
9
140
3
1.8
0.16
0.30
< 0.01
85
0
4
15
6.0
7.0
aThe Kingston Steam Plant is located near Kingston, Tennessee,
150
-------
TABLE 19. EXPERIMENTAL RECOVERY BY MC AND AT METHODS OF TOTAL
CHLORINE FROM UNSPIKED, CHLORINATED KINGSTON STEAM PLANT
CONDENSER COOLING RIVER WATER SAMPLES AND ONES SPIKED WITH
CALCIUM HYPOCHLORITES
Replicate
Number
Total
Chlorine
(yg/D
MC AT
Mean SD RSD (%)
MC AT MC AT MC AT
Total Chlorine
Spike (yg/1)
MC AT
1
2
3
4
5
6
7
Experiment 19.1 Unspiked Sample Composed of 100 % (v/v)
Unchlorinated Raw River Water Intake at Water Supply Plant,
Collected and Analyzed October 15, 1980
5.3 <20"
4.0 <20
3.4 <20
4.3 <20 } 4.27 0.63 14.8
4.3 <20
4.8 <20
3.8 <20
Experiment 19.2 Unspiked Sample Composed of 75 % (v/v)
Unchlorinated Raw River Water Intake at Water Supply Plant
and 25 % (v/v) Chlorinated Unit Number 2 Condenser Discharge
at Plant Outlet, Collected, Composited, and Analyzed
October 15. 1980
1
2
3
4
5
6
7
15
15
15
17
15
14
14
<20
<20
<20
<20
<20
<20
<20
15.0
1.0
6.7
1
2
3
4
5
6
7
Experiment 19.3 Spiked Sample Shown in Experiment 19.2
11
11
11
10
12
13
11
<20
<20
<20
<20
<20
<20
<20
11.3
1.0
8.8
25
15
continued
151
-------
TABLE 19. CONTINUED
Total
Chlorine Total Chlorine
Replicate
Number
(yg/1) Mean SD RSD (%) Spike
MC AT MC AT MC AT MC AT MC
Experiment 19.4 Unspiked Sample Composed of 50 % (v/v)
Unchlorinated Raw River Water Intake at Water Supply Plant
(yg/D
AT
and 50 % (v/v) Chlorinated Unit Number 2 Condenser Discharge
1
2
3
4
5
6
7
1
2
3
4
5
6
7
at Plant Outlet Collected, Composited, and Analyzed
28 30*]
24 30
29 25
26 20
28 30
26 30
28 30J
October 15, 1980
} 27.0 27.9 1.7 3.9 6.3 14.0
Experiment 19.5 Spiked Sample Shown in Experiment 19.4
20 20^
24 25
20 20
20 25
23 25
20 25
22 20^
I 21.3 22.9 1.7 2.7 8.0 11.8 40
Experiment 19.6 Unspiked Sample Composed of 100 % (v/v)
50
Chlorinated Unit Number 2 Condenser Discharge at Plant Outlet,
1
2
3
4
5
6
7
Collected and Analyzed October 15, 1980
56 60 >
57 60
54 55
52 40
53 50
50 45
53 50 ,
\ 53.6 51.4 2.4 7.5 4.5 14.6
continued
152
-------
TABLE 19. CONTINUED
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine Total Chlorine
(yg/1) Mean SD RSD (%) Spike (yg/1)
MC AT MC AT MC AT MC AT MC AT
Q
Experiment 19 . 7 Spiked Sample Shown in Experiment 19 . 6
44 75*"
48 75
50 70
45 70
43 70
36 70
43 60^
> 44.1 70.0 4.5 5.0 10.2 7.1 60 115
Experiment 19.8 Unspiked Sample Composed of 100 % (v/v)
Chlorinated Unit Number 2 Condenser Discharge Outlet Water,
Collected and Analyzed October 16, 1980
107 120^
114 130
104 130
108 110
102 135
102 120
99 130^
> 105 125 5.0 8.7 4.8 7.0
Experiment 19.9 Spiked Sample Shown in Experiment 19.8
111 85*"
94 85
99 85
97 75
93 70
76 85
70 90^
91.4 82.1 14.0 7.0 15.3 8.5 99 100
continued
153
-------
TABLE 19. CONTINUED
Replicate
Number
Total
Chlorine
(pg/D
MC AT
Mean SD RSD (%)
MC AT MC AT MC AT
Total Chlorine
Spike (yg/1)
MC AT
1
2
3
4
5
6
7
Experiment 19.10 Unspiked Sample Composed of 100 % (v/v)
Chlorinated Unit Number 2 Condenser Discharge Outlet,
Collected and Analyzed October 16, 1980
176 170
180 170
186 175
177 170
174 175
164 170
187 180
178
173 7.8
3.9 4.4 2.3
Experiment 19.11 Spiked Sample Shown in Experiment 19.10
1 165 180*1
2 179 165
3 164 175
4 162 165 \ 160 171 10.9
5 148 175
6 148 175
7 156 165
6.3 6.8 3.7 197
199
The Kingston Steam Plant is located near Kingston, Tennessee; MC and AT are
microprocessor-controlled ion selective electrode and forward amperometric
titration methods, respectively; SD is standard deviation, and % RSD is
standard deviation expressed as percentage of mean.
An additional significant figure was retained for use in calculating a
precise estimate of minimum detection limit by the MC method.
°Total chlorine is the value obtained by subtracting the mean background
concentration of the unspiked sample.
154
-------
TABLE 20. CHLORINE DEMAND OF RAW SHAWNEE STEAM PLANT
SAMPLE COMPOSED OF 100 % (v/v) UNCHLORINATED UNIT NUMBER 5
CONDENSER DISCHARGE OUTLET WATER,
COLLECTED OCTOBER 21, 19803
Contact Time
(Mln)
1
1
1
1
1
5
5
5
5
5
10
10
10
10
10
15
15
15
15
15
30
30
30
30
30
60
60
60
60
60
Added
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
Chlorine Concentration, mg/1
Found
0.20
0.48
0.86
2.8
4.8
0.20
0.48
0.82
2.2
4.6
0.20
0.48
0.70
2.2
4.4
0.20
0.48
0.66
1.9
4.2
0.20
0.42
0.58
1.8
4.1
0.20
0.40
0.56
1.2
3.7
V
Demand
0.0
0.12
0.14
0.2
0.2
0.0
0.12
0.18
0.8
0.4
0.0
0.12
0.3
0.8
0.6
0.0
0.12
0.34
1.1
0.8
0.0
0.18
0.42
1.2
0.9
0.0
0.2
0.44
1.8
1.3
Shawnee Steam Plant is located near Paducah, Kentucky.
Demand is difference between total chlorine added initially as NaOCl-Cl9 and
that found after elapsed contact time.
155
-------
TABLE 21. INORGANIC SPECIES CONCENTRATIONS,
SANITARY CHEMICAL CHARACTERISTICS, AND PHYSICAL PROPERTIES
OF RAW SHAWNEE STEAM PLANT SAMPLE COMPOSED OF 100 % (v/v)
UNCHLORINATED UNIT NUMBER 5 CONDENSER DISCHARGE OUTLET WATER,
COLLECTED OCTOBER 21, 1980*
Characteristics
Value
Inorganics, yg/1
Arsenic
Chromium
Copper
Iron
Mercury
Manganese
Zinc
Solids, mg/1
Suspended
Dissolved
Chemical oxygen demand, mg/1
Total organic carbon mg/1
Nitrogen, mg/1 as N
Organic
Nitrite plus nitrate
Ammonia
Alkalinity, mg/1 as CaCO_
Total
Phenol
Color, mg/1 of Pt in the chloroplatinate standard
True
Apparent
Turbidity, mg/1 of natural Si
PH
2
< 1
10
1200
0.3
150
20
37
220
10
4.4
0.26
0.59
0.06
91
< 1
14
28
23
7.2
Shawnee Steam Plant is located near Paducah, Kentucky.
156
-------
TABLE 22. INORGANIC SPECIES CONCENTRATIONS OF SETTLED
MATERIALS SAMPLED OCTOBER 22, 1980,
FROM RAW SHAWNEE STEAM PLANT SAMPLE COMPOSED OF 100 % (v/v)
UNCHLORINATED UNIT NUMBER 5 CONDENSER DISCHARGE OUTLET WATER,
COLLECTED OCTOBER 21, 198Qa'
Characteristics Value
Inorganics, yg/1
Arsenic 6
Chromium < 1.
Copper 30
Iron 8000
Mercury <0.2
Manganese 650
Zinc 60
o
Shawnee Steam Plant is located near Paducah, Kentucky.
After the insoluble materials in the sample shown in Table 21 settled and
aliquots of the supernatant were withdrawn for diluting chlorinated
samples, the settled materials had been concentrated to yield a water
slurry with the given characteristics.
157
-------
TABLE 23. EXPERIMENTAL RECOVERY BY MC AND AT METHODS OF TOTAL CHLORINE
FROM UNSPIKED, CHLORINATED SHAWNEE STEAM PLANT CONDENSER
COOLING RIVER WATER SAMPLES AND ONES SPIKED
WITH CALCIUM HYPOCHLORITE3
Replicate
Number
Total
Chlorine
(pg/D
MC AT
Mean SD RSD (%)
MC AT MC AT MC AT
Total Chlorine
Spike (yg/1)
MC AT
Experiment 23.1 Unspiked Sample Composed of 100 % (v/v)
Unchlorinated Unit Number 5 Condenser Discharge Outlet Water
Collected October 21, 1980, and Analyzed October 22, 1980
1
2
3
4
5
6
7
-2.0 <20
0.3 <20
0.2 <20
1.4 <20
0.2 <20
32.3C<20C
37.7C<20C:
0.0200
1.24
6200
Experiment 23.2 Unspiked Sample Composed of 10 % (v/v)
Chlorinated Unit Number 4 Condenser Discharge Outlet Water
and 90 % (v/v) Unchlorinated Unit Number 5 Condenser Discharge
Outlet Water Collected October 21, 1980, and Analyzed October 22. 1980
1
2
3
4
5
6
7
5.2 <20
4.1 <20
8.7 <20
3.4 <20
6.0 <20
14.7 <20
13.3 <20
7.91
4.50
56.9
Experiment 23.3 Spiked Sample Shown in Experiment 23.2
d,e
1
2
3
4
5
6
7
11.0
12.7
9.6
13.8
11.4
9.5
11.9
35
30
30
40
<20
<20
11.4 27.1 1.57 12.2 13.8 45.0
25
50
continued
158
-------
TABLE 23. CONTINUED
Total
Chlorine Total
Replicate (ug/1) Mean SD RSD (%) Spike
Number MC AT MC AT MC AT MC AT MC
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Experiment 23.4 Unspiked Sample Composed of 25 % (v/v)
Chlorinated Unit Number 4 Condenser Discharge Outlet Water
and 75 % (v/v) Unchlorinated Unit Number 5 Condenser Discharge
Outlet Water Collected October 21, 1980, but Composited
and Analyzed October 22, 1980
21 20^
22 30
20 40
21 40 > 20.0 32.9 1.4 7.0 7.0 21.3
18 35
19 30
19 35J
d e
Experiment 23.5 Spiked Sample Shown in Experiment 23.4 '
40 30^
36 20
33 25
30 <20 ) 33.1 21.4 4.0 8.5 12.1 39.7 50
34 25
31 30
28 <20>
Experiment 23.6 Unspiked Sample Composed of 50 % (v/v)
Chlorinated Unit Number 4 Condenser Discharge Outlet Water
and 50 % (v/v) Unchlorinated Unit Number 5 Condenser Discharge
Outlet Water Collected October 21, 1980, but Composited
and Analyzed October 22, 1980
45 60^
44 40
45 50
45 40 } 46.6 47.9 2.8 8.1 6.0 16.9
51 40
46 50
50 55^
Chlorine
(yg/D
AT
100
continued
159
-------
TABLE 23. CONTINUED
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine Total Chlorine
(yg/1) Mean SD RSD (%) Spike (yg/1)
MC AT MC AT MC AT MC AT MC AT
Experiment 23.7 Spiked Sample Shown in Experiment 23.6
66 40N
56 40
71 40
78 60
72 60
69 40
76 50^
) 69.7 47.1 7.3 9.5 10.5 20.2 99 100
Experiment 23.8 Unspiked Sample Composed of 100 % (v/v)
Chlorinated Unit Number 4 Condenser Discharge Outlet Water
Collected October 21, 1980, and Analyzed October 22, 1980
114 100>
111 120
123 120
113 120
118 100
129 100
121 100^
118 109 6.4 10.7 5.4 9.8
Experiment 23.9 Spiked Sample Shown in Experiment 23.8
80 70"
92 90
99 90
91 80
85 80
111 80
82 70,
91.4 80.0 10.8 8.2 11.8 10.2 99 100
continued
160
-------
TABLE 23. CONTINUED
Replicate
Number
Total
Chlorine
(yg/D
MC AT
Mean
SD
MC
AT
MC
AT
RSD (%)
MC AT
Total Chlorine
Spike (yg/1)
MC AT
Experiment 23.10 Unspiked Sample Composed of 75 % (v/v)
Chlorinated Unit Number 4 Condenser Discharge Outlet Water
and 25 % (v/v) Unchlorinated Unit Number 5 Condenser Discharge
Outlet Water Collected and Analyzed October 21, 1980
1
2
3
4
5
6
7
171 150
169 120
154 140
131 140
161 120
152 150
145 130
155
136 14.0 12.7 9.0 9.3
1
2
3
4
5
6
7
Experiment 23.11 Spiked Sample Shown in Experiment 23.10
224 245^1
246 215
264 225
247 205
242 215
238 205
214 205
239
216 16.3 14.6 6.8 6.8 296
299
The Shawnee Steam Plant is located near Paducah, Kentucky. MC and AT are
microprocessor-controlled ion selective electrode and forward amperometric
titration methods, respectively. SD is standard deviation and % RSD is
standard deviation expressed as percentage of mean.
An additional significant figure was retained for use in calculating a
precise estimate of minimum detection limit by MC.
->
"These values were omitted in calculation of mean, SD, and % RSD by MC and
were obtained on settled materials sampled October 22, 1980 (See Table 22
for concentrations of inorganic species).
Total chlorine is value obtained after subtracting the mean background con-
centration of unspiked sample.
"For the AT, results less than 20 yg/1 were included in the statistical
calculations as 10 yg/1.
161
-------
TABLE 24. CHLORINE DEMAND OF RAW ALLEN STEAM PLANT
SAMPLE COMPOSED OF 100 % (v/v) UNCHLORINATED
UNIT NUMBER 3 CONDENSER DISCHARGE WATER INLET COLLECTED
OCTOBER 28, 198Qa
Contact Time
(Min)
1
1
1
1
1
5
5
5
5
5
10
10
10
10
10
15
15
15
15
15
30
30
30
30
30
60
60
60
60
60
Added
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
Chlorine Concentration, mg/1
Found
0.2
0.6
1.0
2.4
3.6
0.2
0.6
1.0
1.8
2.6
0.2
0.6
1.0
1.6
2.3
0.2
0.6
1.0
1.4
2.2
0.2
0.6
1.0
1.0
2.0
0.2
0.56
0.96
1.0
1.5
Y
Demand "
0.0
0.0
0.0
0.6
1.4
0.0
0.0
0.0
1.2
2.4
0.0
0.0
0.0
1.4
2.7
0.0
0.0
0.0
1.6
2.8
0.0
0.0
0.0
2.0
3.0
0.0
0.04
0.04
2.0
3.5
The Allen Steam Plant is located near Memphis, Tennessee.
DDemand is difference between total chlorine added initially as NaOCl-Cl,, and
that found after elapsed contact time.
162
-------
TABLE 25. INORGANIC SPECIES CONCENTRATIONS, SANITARY
CHEMICAL CHARACTERISTICS, AND PHYSICAL PROPERTIES OF RAW ALLEN
STEAM PLANT SAMPLE COMPOSED OF 100 % (v/v)
UNCHLORINATED UNIT NUMBER 3 CONDENSER DISCHARGE INLET WATER,
COLLECTED OCTOBER 28, 1980a
Characteristics
Alkalinity, mg/1 as Ca CO,
Total :
Phenol
Color, mg/1 of Pt in the chloroplatinate standard
True
Apparent
Turbidity, mg/1 of natural Si
PH
Value
Inorganics, vg/1
Arsenic
Chromium
Copper
Iron
Mercury
Manganese
Zinc
Solids, mg/1
Suspended
Dissolved
Chemical oxygen demand, mg/1
Total organic carbon mg/1
Nitrogen, mg/1 as N
Organic
Nitrite plus nitrate
Ammonia
3
< 1
20
1600
< 0.2
670
20
30
30
24
4.4
0.37
0.95
0.45
14
< 1
50
74
19
7.4
The Allen Steam Plant is located near Memphis, Tennessee.
163
-------
TABLE 26. EXPERIMENTAL RECOVERY BY MC AND AT METHODS OF TOTAL CHLORINE
FROM UNSPIKED, CHLORINATED ALLEN STEAM PLANT CONDENSER
COOLING RIVER WATER SAMPLES AND ONES SPIKED
WITH CALCIUM HYPOCHLORITE3
Replicate
Number
Total
Chlorine
(yg/D
MC AT
Total Chlorine
Mean
MC AT
SD RSD (%)
MC AT MC AT
Spike
MC
(yg/D
AT
Experiment 26.1 Unspiked Sample Composed of 100 % (v/v)
Unchlorinated Unit Number 3 Condenser Discharge Inlet Water.
Collected October 28, 1980, and Analyzed October 29, 1980
1
2
3
4
5
6
7
2.0 <20
2.4 <20
-3.3 <20
-2.6 <20
-2.1 <20
-1.3 <20
-0.9 <20
-0.83
2.22
267
Experiment 26.2 Unspiked Sample Composed of 5 % (v/v)
Chlorinated Unit Number 3 Condenser Discharge Inlet Water,
and 95 % (v/v) Unchlorinated Unit Number 3 Condenser Discharge
Inlet Water. Collected October 28, 1980, but Composited
and Analyzed October 29. 1980°
1
2
3
4
5
6
7
28
20
22
20
19
19
20
<20
30
<20
30
25
20
25
21.1 21.1
3.2
8.5 15.2 40.3
Experiment 26.3 Spiked Sample Shown in Experiment 26.2
c,d
1
2
3
4
5
6
7
16
14 25
22 35
22 <20
24 35
18 <20
17 <20
19.0 19.3 3.7 12.1 19.5 62.7 25
25
continued
164
-------
TABLE 26. CONTINUED
Replicate
Number
Total
Chlorine Total
(yg/1) Mean SD RSD (%) Spike
MC AT MC AT MC AT MC AT MC
Experiment 26.4 Unspiked Sample Composed of 10 % (v/v)
Chlorinated Unit Number 3 Condenser Discharge Inlet Water
Chlorine
(yg/D
AT
and 90 % (v/v) Unchlorinated Unit Number 3 Condenser Discharge
1
2
3
4
5
6
7
1
2
3
4
5
6
7
and
1
2
3
4
5
6
7
Inlet Water Collected October 28, 1980, but Composited
and Analyzed October 29, 1980
39 45^
52 45
44 45
43 50
50 40
48 40
52 50^
\ 46.9 45.0 5.0 4.1 10.7 9.1
Experiment 26.5 Spiked Sample Shown in Experiment 26.4
38 35^)
20 30
39 20
32 40
27 35
21 40
21 40>
i 28.3 34.3 8.2 7.3 29.0 21.3 50
Experiment 26.6 Unspiked Sample Composed of 15 % (v/v)
Chlorinated Unit Number 3 Condenser Discharge Inlet Water
85 % (v/v) Unchlorinated Unit Number 3 Condenser Discharge
Inlet Water Collected October 28, 1980, but Composited
and Analyzed October 29, 1980
80 85 "1
90 90
82 75
85 70
80 60
66 80
53 65^
50
> 76.6 75.0 12.7 10.8 16.6 14.4
continued
165
-------
TABLE 26. CONTINUED
Replicate
Number
Total
Chlorine Total Chlorine
(yg/1) Mean SD RSD (%) Spike (yg/1)
MC AT MC AT MC AT MC AT MC AT
c d
Experiment 26.7 Spiked Sample Shown in Experiment 26.6 '
1
2
3
4
5
6
7
in
1
2
3
4
5
6
7
1
2
3
4
5
6
7
38 20*)
44 30
42 30
40 20
43 35
43 35
38 <20
) 41.1 25.7 2.5 9.3 6.1 36.2 50 50
Experiment 26.8 Unspiked Sample Composed of 15 % (v/v)
Chlorinated Unit Number 3 Condenser Discharge Inlet Water
Collected October 28, 1980, and 85 % (v/v) of Mixture Shown
Experiment 26.2, but Composited and Analyzed October 29, 1980
143 145*1
155 155
152 135
145 140
155 150
149 145
151 145J
> 150 145 4.7 6.5 3.1 4.5
Experiment 26.9 Spiked Sample Shown in Experiment 26.8
122 125*"
134 105
123 125
104 115
126 105
118 125
113 100 J
> 120 114 9.6 11.0 8.0 9.6 148 150
continued
166
-------
TABLE 26. CONTINUED
Total
Chlorine Total Chlorine
Replicate (iWD Mean SD RSD (%) Spike (vg/1)
Number
75
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
MC AT MC AT MC AT MC AT MC AT
Experiment 26.10 Unspiked Sample Composed of 25 % (v/v)
Chlorinated Unit Number 3 Condenser Discharge Inlet Water and
% (v/v) Unchlorinated Unit Number 3 Condenser Discharge Inlet
Water Collected and Analyzed October 28, 1980
250 160^)
249 205
256 170
241 180
250 190
250 165
235 170^
) 247 177 7.0 15.8 2.8 8.9
Experiment 26.11 Effect of 2 % (v/v) Sodium Pyrophosphate
Decahydrate in Overcoming Chemical Interference for Total
Chlorine Determinations on Sample Shown in Experiment 26.10
168 140^
162 160
158 145
J
163 148 5.0 10.4 3.1 7.0
•
Experiment 26.12 Spiked Sample Shown in Experiment 26.10
180 165>
156 155
148 105
154 105
163 115
141 135
156 145
\ 157 132 12.3 24.3 7.8 18.4 245 199
continued
167
-------
TABLE 26. CONTINUED
Q
The Allen Steam Plant is located near Memphis, Tennessee. MC and AT are
microprocessor-controlled ion selective electrode and forward amperometric
titration methods, respectively. SD is standard deviation, and % RSD is
standard deviation expressed as percentage of mean.
An additional significant figure was retained for use in calculating a
precise estimate of minimum detection limit by MC.
For the AT, results less than 20 yg/1 were included in the statistical
calculation as 10 yg/1.
Total chlorine is value obtained after subtracting the mean background
concentration of unspiked sample.
168
-------
TABLE 27. CHLORINE DEMAND OF RAW JOHN SEVIER STEAM PLANT
SAMPLE COMPOSED OF 100 % (v/v) UNCHLORINATED
UNIT NUMBER 1 CONDENSER DISCHARGE INLET WATER
COLLECTED, NOVEMBER 4, 1980a
Contact Time
(Min)
1
1
1
1
1
5
5
5
5
5
10
10
10
10
10
15
15
15
15
15
30
30
30
30
30
60
60
60
60
60
Added
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
0.2
0.6
1.0
3.0
5.0
Chlorine Concentration, mg/1
Found
0.20
0.56
0.88
2.4
4.7
0.18
0.46
0.78
2.40
4.50
0.16
0.44
0.72
2.30
4.40
0.12
0.38
0.60
2.25
4.40
0.10
0.28
0.44
2.0
4.2
0.08
0.22
0.32
1.8
4.0
1
Demand"
0.0
0.04
0.12
0.6
0.3
0.02
0.14
0.22
0.6
0.5
0.04
0.16
0.28
0.7
0.6
0.08
0.22
0.4
0.75
0.6
0.1
0.32
0.56
1.0
0.8
0.12
0.38
0.68
1.2
1.0
The John Sevier Steam Plant is located near Rogersville, Tennessee.
Demand is the difference between total chlorine added initially as NaOCl-Cl,
and that found after elapsed time.
169
-------
TABLE 28. CONCENTRATIONS OF INORGANIC SPECIES, SANITARY
CHEMICAL CHARACTERISTICS, AND PHYSICAL PROPERTIES OF RAW JOHN SEVIER
STEAM PLANT SAMPLE COMPOSED OF 100 % (v/v)
UNCHLORINATED UNIT NUMBER 1 CONDENSER DISCHARGE INLET WATER
COLLECTED NOVEMBER 4, 1980
Characteristics Value
Inorganics, yg/1
Arsenic < 2
Chromium < 1
Copper < 10
Iron 380
Mercury 0.3
Manganese 50
Zinc < 10
Solids, mg/1
Suspended 7
Dissolved 150
Chemical oxygen demand, mg/1 8
Total organic carbon mg/1 2.5
Nitrogen, mg/1 as N
Organic 0.22
Nitrite plus nitrate 0.85
Ammonia 0.06
Alkalinity, mg/1 as Ca CO
Total J 72
Phenol 0
Color, mg/1 of Pt in the chloroplatinate standard
True 15
Apparent 21
Turbidity, mg/1 of natural Si 43
pH 6.6
aThe John Sevier Steam Plant is located near Rogersville, Tennessee.
170
-------
TABLE 29. EXPERIMENTAL RECOVERY BY MC AND AT METHODS OF TOTAL CHLORINE
FROM UNSPIKED, CHLORINATED JOHN SEVIER STEAM PLANT
CONDENSER COOLING RIVER WATER SAMPLES AND ONES SPIKED
WITH CALCIUM HYPOCHLORITE3
Replicate
Number
Total
Chlorine
(yg/D
MC AT
Mean SD RSD (%)
MC AT MC AT MC AT
Total Chlorine
Spike (yg/1)
MC AT
1
2
2
4
5
6
7
Experiment 29.1 Unspiked Sample Composed of 100 % (v/v)
Unchlorinated Unit Number 1 Condenser Discharge Inlet Water
Collected and Analyzed November 4, 1980
2.3 <20
2.2 <20
2.3 <20
2.3 <20
2.3 <20
1.2 <20
2.1 <20
S 2.10
0.40
19.0
1
2
3
4
5
6
7
Experiment 29.2 Unspiked Sample Composed of 10 % (v/v)
Chlorinated Unit Number 1 Condenser Discharge Inlet Water
and 90 % (v/v) Unchlorinated Unit Number 1 Condenser Discharge
Inlet Water Collected November 4, 1980, buit Composited
and Analyzed November 5, 1980 *
9.7 25
9.1 <20
8.8 <20
10.1 <20
8.3 <20
9.2 <20
8.8 20
> 9.14 13.5 0.60 6.3 6.6 46.7
1
2
3
4
5
6
7
Experiment 29.3 Spiked Sample Shown in Experiment 29.2
b,c,d
5.6
5.1
5.11 19.3 1.03 11.0 20.2 57.0
10
15
continued
171
-------
TABLE 29. CONTINUED
Total
Chlorine Total Chlorine
Replicate
Number
and
1
2
3
4
5
6
7
1
2
3
4
5
6
7
and
1
2
3
4
5
6
7
(yg/1) Mean SD RSD (%) Spike
MC AT MC AT MC AT MC AT MC
Experiment 29.4 Unspiked Sample Composed of 25 % (v/v)
Chlorinated Unit Number 1 Condenser Discharge Inlet Water
75 % (v/v) Unchlorinated Unit Number 1 Condenser Discharge
Inlet Water Collected November 4, 1980, but Composited
and Analyzed November 5, 1980
29 35*")
25 40
23 30
26 35
24 25
33 25
21 25J
> 25.9 30.7 4.0 6.1 15.4 19.9
Experiment 29.5 Spiked Sample Shown in Experiment 29.4
16 10*1
16 15
14 20
15 30
16 20
15 30
15 25^
\ 15.3 21.4 0.8 7.5 5.2 35.0 25
Experiment 29.6 Unspiked Sample Composed of 50 % (v/v)
Chlorinated Unit Number 1 Condenser Discharge Inlet Water
50 % (v/v) Unchlorinated Unit Number 1 Condenser Discharge
Inlet Water Collected November 4, 1980, but Composited
and Analyzed November 5, 1980
55 50*]
52 60
57 50
53 65
61 65
55 50
54 65,
(yg/D
AT
25
55.3 57.9 3.0 7.6 5.4 13.1
continued
172
-------
TABLE 29. CONTINUED
Replicate
Number
Total
Chlorine
(yg/D
MC AT
Total Chlorine
Mean SD
MC AT MC AT
RSD (%)
MC AT
Spike
MC
(yg/D
AT
1
2
3
4
5
6
7
Experiment 29.7 Spiked Sample Shown in Experiment 29.6
44
45
44
40
33
42
40
35
20
35
50
40
35
45
41.1 37.1 4.1
9.5 10.0 25.6 50
50
1
2
3
4
5
6
7
Experiment 29.8 Unspiked Sample Composed of 75 % (v/v)
Chlorinated Unit Number 1 Condenser Discharge Inlet Water
and 25 % (v/v) Unchlorinated Unit Number 1 Condenser Discharge
Inlet Water Collected November 4, 1980, but Composited
and Analyzed November 5, 1980
95 110"
93 100
105 100
103 85 } 95.7 99.3 6.0 7.9 6.3 8.0
90 105
90 100
94 95
1
2
3
4
5
6
7
Experiment 29.9 Spiked Sample Shown in Experiment 29.8C
62
85
73
83
69
70
78
60
74.3 73.6 8.2 10.3 11.0 14.0 99
100
continued
173
-------
EXPERIMENT 29. CONTINUED
Replicate
Number
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Total
Chlorine Total Chlorine
(yg/1) Mean SD RSD (%) Spike (yg/1)
MC AT MC AT MC AT MC AT MC AT
Experiment 29.10 Unspiked Sample Composed of 100 % (v/v)
Chlorinated Unit Number 1 Condenser Discharge Inlet Water
Collected November 4, 1980, and Analyzed November 5, 1980
137 130^
145 125
139 135
137 135
137 135
143 115
134 140J
} 139 131 3.9 8.4 2.8 6.4
Experiment 29.11 Spiked Sample Shown in Experiment 29.10
132 125*"
132 115
130 135
125 120
134 110
135 115
130 115 J
\ 131 119 3.3 8.4 2.5 7.1 148 150
Experiment 29.12 Unspiked Sample Composed of 100 % (v/v)
Chlorinated Unit Number 1 Condenser Discharge Inlet Water
Collected and Analyzed November 4, 1980
294 230^
313 270
326 250
301 255
338 260
316 260
323 240.
\ 316 252 15.0 13.5 4.7 5.4
continued
174
-------
TABLE 29. CONTINUED
Replicate
Number
Total
Chlorine
(yg/D
MC AT
Mean SD RSD (%)
MC AT MC AT MC AT
Total Chlorine
Spike (yg/1)
MC AT
Experiment 29.13 Spiked Sample Shown in Experiment 29.12
1
2
3
4
5
6
7
215
188
171
205
187
200
195
245^
235
220
215
215
235
220^
194 226 14.2 11.8 7.3 5.2 245
249
The John Sevier Steam Plant is located near Rogersville, Tennessee. MC and
AT are microprocessor-controlled ion selective electrode and forward
amperometric titration methods, respectively. SD is standard deviation,
and % RSD is standard deviation expressed as percentage of mean.
An additional significant figure was retained for use in calculating a
precise estimate of minimum detection limit by MC.
e-t
For the AT, results less than 20 yg/1 were included in the statistical
calculations as 10 yg/1.
Total chlorine is value obtained after subtracting the mean background
concentration of unspiked sample.
175
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-600/7-82-005
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Microprocessor-Controlled Ion Selective Electrode
Determination of Total Chlorine
5. REPORT DATE
March 1982
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Lyman H. Howe, Reginald E. Hadley, and Gary A. Fischer
8. PERFORMING ORGANIZATION REPORT NO.
TVA/ONR/NRO-82/4
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Office of Natural Resources
Tennessee Valley Authority
Chattanooga, Tennessee 37401
10. PROGRAM ELEMENT NO.
INF.
11. CONTRACT/GRANT NO.
81 BDH
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Environmental Processes and Effects Research
Washington, D.C. 20460
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
This project is part of the EPA-planned and coordinated Federal Interagency
Energy/Environment R&D Program
16. ABSTRACT
A microprocessor-controlled ion selective electrode (MC) method was evaluated and
compared to the forward amperometric titration (AT) method for determining total
chlorine in condenser cooling river water from coal-burning electric plants. The
effective range for quantification by the MC method is from the minimum detection
limit of 6.5 yg/1 chlorine for unspiked condenser water and 13.3 yg/1 chlorine for
spiked condenser water to 100 yg/1 chlorine. Interferences by zinc(II), copper(II),
iron(III), arsenic(III), and manganese(VII) are discussed. The pH, chromium(VI),
mercury(II), bromide, and arsenic(V) do not interfere with measurement of total
chlorine. For both unspiked and spiked condenser water, the overall pooled
standard deviation and overall mean percentage relative standard deviation over
concentrations of 20-200 yg/1 chlorine are lower for the MC method than for the
AT method. Standard deviations are discussed for the MC method over concentrations
of 2-20 yg/1 chlorine.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COS AT I Field/Group
Inorganic Chemistry
Chemical Engineering
Charac., Meas., and
Monit.
7B
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (ThisReport/
Unclassified
21. NO. OF PAGES
200
20. SECURITY CLASS {Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-------
-------
------- |