c/EPA
United States
Environmental Protection
Agency
Office of Acid Deposition,
Environmental Monitoring and
Quality Assurance
Washington DC 20460
EPA/600/8-87/038
August 1987
Research and Development
Western Lake Survey
Phase I
Analytical Methods
Manual
-------
Pacific
Northwest (4B)
California (4A)
Northern
Rockies (4C)
Central
Rockies (4D)
Southern
Rockies (4E)
Subregions of the Western Lake Survey - Phase I
-------
EPA 600/8-87/038
August 1987
Western Lake Survey
Phase I
Analytical Methods Manual
A Contribution to the
National Acid Precipitation Assessment Program
U.S. Environmental Protection Agency
Office of Research and Development
Washington, DC 20460
Environmental Monitoring Systems Laboratory - Las Vegas, NV 89119
Environmental Research Laboratory - Corvallls, OR 97333
U.S. Environmental Protection Agency
Region 5, Library (5PL-16)
230 S. Dearborn Street, Room 1670
Chicago, IL 60604
-------
Notice
The information in this document has been funded wholly or in part by the U.S.
Environmental Protection Agency under Contract Number 68-03-3249 to
Lockheed Engineering and Management Services Company, Inc. It has been
subject to the Agency's peer and administrative review, and it has been
approved for publication as an EPA document.
Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
This document is one volume of a set which fully describes the Western Lake
Survey - Phase I. The complete document set includes the major data report
(2 volumes), quality assurance plan, analytical methods manual, field
operations report, and quality assurance report. Similar sets are being
produced for each Aquatic Effects Research Program component project.
Colored covers, artwork, and use of the project name in the document title
serve to identify each companion document set.
Proper citation of this document is:
Kerfoot, H. B. and M. L. Faber. National Surface Water Survey, Western Lake
Survey (Phase I - Synoptic Chemistry) Analytical Methods Manual.
EPA/600/8-87/038. U.S. Environmental Protection Agency, Las Vegas,
Nevada.
-------
Abstract
The Analytical Methods Manual for the Western Lake Survey - Phase I is a
supplement to the Analytical Methods Manual for the Eastern Lake Survey -
Phase I. This supplement provides a general description of the analytical
methods that are used by the field laboratories and by the analytical
laboratories; a detailed description of the analytical methods appears in the
Analytical Methods Manual for the Eastern Lake Survey - Phase I. The
supplement also describes new and modified sample processing procedures
that were developed specifically for the West.
The Eastern Lake Survey - Phase I and the Western Lake Survey - Phase I
are parts of the National Surface Water Survey. The National Surface Water
Survey component of the National Acid Precipitation Assessment Program is
designed to evaluate the present water chemistry of lakes and streams, to
determine the status of certain biotic resources, and to select regionally
representative surface waters for a long-term monitoring program that will
study changes in aquatic resources.
To meet U.S. Environmental Protection Agency requirements, the data
collection for the survey must ensure that the resulting data are (1) of known
quality, (2) suitable for the intended purpose, and (3) consistent and
comparable. For these reasons, all analysts participating in the survey must
use the same reliable, detailed analytical methods. The analytical methods
used during the Western Lake Survey - Phase I are given below in the
context of the analytes and physical parameters to which they were applied.
Analvte or Physical Parameter Method
1. Acid neutralizing capacity Titration and Gran analysis
2. Aluminum, total Atomic absorption spectroscopy
(furnace)
3. Aluminum, extractable Extraction with 8-hydroxyquinoline
into methyl isobutyl ketone
followed by atomic absorption
spectroscopy (furnace)
4. Ammonium, dissolved Automated colorimetry (phenate)
5. Base neutralizing capacity Titration and Gran analysis
6. Calcium, dissolved Atomic absorption spectroscopy
(flame) or inductively coupled
plasma atomic emission
spectroscopy
7. Chloride, dissolved Ion chromatography
8. Conductance Conductivity cell and meter
-------
Abstract (continued)
Analyte or Physical Parameter
9. Fluoride, total dissolved
10. Inorganic carbon, dissolved
11. Iron, dissolved
12. Magnesium, dissolved
13. Manganese, dissolved
14. Nitrate, dissolved
15. Organic carbon, dissolved
16. pH
17. Phosphorus, total
18. Potassium, dissolved
19. Silica, dissolved
20. Sodium, dissolved
21. Sulfate, dissolved
Method
Ion-selective electrode and meter
Instrument (acidification, carbon
dioxide generation, infrared
radiation detection)
Atomic absorption spectroscopy (flame)
or inductively coupled plasma
atomic emissions spectroscopy
Atomic absorption spectroscopy (flame)
or inductively coupled plasma
atomic emission spectroscopy
Atomic absorption spectroscopy (flame)
or inductively coupled plasma
atomic emission spectroscopy
Ion chromatography
Instrument (ultraviolet-promoted
oxidation, carbon dioxide generation,
infrared radiation detection)
pH electrode and meter, pH indicator
strips
Automated colonmetry
(phosphomolybdate)
Atomic absorption spectroscopy (flame)
Atomated colonmetry
(molybdate blue)
Atomic absorption spectroscopy (flame)
Ion chromatography
22. True color
23. Turbidity
Comparison to platinum-cobalt
color standards
Instrument (nephelometer)
This manual was submitted in partial fulfillment of Contract Number 68-03-
3249 by Lockheed Engineering and Management Services Company, Inc.,
under the sponsorship of the U.S. Environmental Protection Agency. This
report covers a period from September 10, 1985, to November 4, 1985, and
work was completed as of May 14, 1986.
-------
Contents
Page
Abstract i"
Figures vii
Tables viii
Abbreviations ix
Acknowledgments xi
1.0 Introduction 1
1.1 Document Content 3
1.2 Analytes and Physical Parameters Measured 3
1.2.1 Acid Neutralizing Capacity 3
1.2.2 Aluminum, Total 3
1.2.3 Aluminum, Extractable 3
1.2.4 Base Neutralizing Capacity 3
1.2.5 Inorganic Carbon, Dissolved 3
1.2.6 Ions, Dissolved (Ca, CI", Fe, K, Mg, Mn,
Na, NH4 + ' NCV, SCV, and Total r) 5
1.2.7 Organic Carbon, Dissolved 5
1.2.8 pH 5
1.2.9 Phosphorus, Total 5
1.2.10 Silica, Dissolved 5
1.2.11 Conductance 5
1.2.12 True Color 5
1.2.13 Turbidity 5
1.3 References 6
2.0 Field Laboratory Operations 7
2.1 Daily Operations 8
2.1.1 General Daily Operations 8
2.1.2 Calibration Study Operations 9
2.2 Quality Control Activities 11
2.3 Field Laboratory Methods 11
-------
Contents (continued)
Page
2.3.1 Determination of Dissolved Inorganic
Carbon 11
2.3.2 Determination of pH 11
2.3.3 Determination of Turbidity 11
2.3.4 Determination of True Color 11
2.3.5 Aliquot Preparation 11
2.4 References 13
3.0 Analytical Laboratory Operations 15
3.1 Daily Operations 15
3.2 Quality Control Activities 16
3.2.1 Percent Relative Standard Deviation 16
3.2.2 Percent Matrix Spike Recovery 18
3.2.3 Percent Ion Balance Difference 19
3.2.4 Percent Conductance Difference 19
3.3 Analytical Laboratory Methods 19
3.3.1 Determination of ANC, BNC, and pH 19
3.3.2 Determination of Ammonia 20
3.3.3 Determination of Chloride, Nitrate, and
Sulfate by Ion Chromatography 20
3.3.4 Determination of Dissolved Organic Carbon
and Dissolved Inorganic Carbon 20
3.3.5 Determination of Total Dissolved
Fluoride by Ion-Selective Electrode 20
3.3.6 Determination of Total Phosphorus 21
3.3.7 Determination of Dissolved Silica 21
3.3.8 Determination of Conductance 21
3.3.9 Determination of Dissolved Metals (Al,
Ca, Fe, K, Mg, Mn, Na) by Atomic
Absorption Spectroscopy 21
3.3.10 Determination of Dissolved Metals (Ca,
Fe, Mg, and Mn) by Inductively Coupled
Plasma Atomic Emission Spectroscopy 22
3.4 References 23
Appendix
Calculation Procedures Required for ANC and BNC
Determinations 25
-------
Figures
Page
1-1 Timetable for field activities of the National
Surface Water Survey
2-1 Flow scheme of daily field laboratory activities,
Western Lake Survey - Phase I
2-2 Shipping and data form flow scheme,
Western Lake Survey -Phase I 9
2-3 Sample flow for the calibration study 10
2-4 National Surface Water Survey Form 2 -
Batch/QC Field Data 12
2-5 Preparation, identification, and shipment of
sample batches for the calibration study 13
-------
Tables
Page
1-1 Analytes, Physical Parameters, and Data Quality
Objectives, Western Lake Survey - Phase I 4
2-1 Sample Codes Used to Complete Lake Data Forms,
Western Lake Survey - Phase I 14
2-2 Aliquots, Containers, Preservatives, and Corresponding
Analyses, Western Lake Survey - Phase I 14
3-1 Maximum Holding Times for Samples,
Western Lake Survey -Phase I 15
3-2 Summary of Internal Quality Control Checks for
Analytical Methods, Western Lake Survey - Phase I 17
3-3 Maximum Control Limits for Quality Control Samples,
Western Lake Survey - Phase I 18
3-4 Conductance Factors of Ions 18
3-5 Chemical Reanalysis Criteria, Western Lake Survey -
Phase I 18
3-6 Atomic Absorption Concentration Ranges 22
VIII
-------
Abbreviations*
AA
ANC
APHA
ASTM
BG
BH
BNC
%CD
DG
DGC
DH
DHC
DIG
DOC
ELS-I
EMSL-LV
EPA
ERL-C
1C
ICP
ID
%IBD
IDL
IR
MIBK
NAPAP
NLS
NSWS
NTU
ORNL
PCU
QA
atomic absorption spectroscopy
acid neutralizing capacity
American Public Health Association
American Society for Testing and Materials
field blank sample - ground
field blank sample - helicopter
base neutralizing capacity
percent conductance difference
duplicate lake sample - ground
duplicate calibration sample - ground
duplicate lake sample - helicopter
duplicate calibration sample - helicopter
dissolved inorganic carbon
dissolved organic carbon
Eastern Lake Survey - Phase I
Environmental Monitoring Systems Laboratory - Las
Vegas
U.S. Environmental Protection Agency
Environmental Research Laboratory - Corvallis
ion chromatography
inductively coupled plasma atomic emission spectroscopy
identification
percent ion balance difference
instrument detection limit
infrared
methyl isobutyl ketone
National Acid Precipitation Assessment Program
National Lake Survey
National Surface Water Survey
nephelometric turbidity unit
Oak Ridge National Laboratory
platinum-cobalt unit
quality assurance
-------
Abbreviations* (continued)
QC quality control
QCCS quality control check sample
RF radio frequency
RG routine lake sample - ground
RGC routine calibration sample - ground
RH routine lake sample - helicopter
RHC routine calibration sample - helicopter
RPD relative percent difference
%RSD percent relative standard deviation
TB field laboratory (trailer) blank sample
TD field laboratory (trailer) duplicate sample
THC triplicate calibration sample - helicopter
TISAB total ionic strength buffer solution
WLS-I Western Lake Survey - Phase I
* This list does not include chemical symbols or units of measurement
-------
Acknowledgments
M. B. Bonoff, A. W. Groeger, D. C. Hillman, and D. V. Peck of Lockheed
Engineering and Management Services Company, Inc., Las Vegas, Nevada,
were responsible for the preparation of earlier National Surface Water Survey
documents that serve as the basis for this manual. The following individuals
provided helpful technical reviews: M. E. Peden of the Illinois State Water
Survey, Champaign, Illinois; F. S. Sanders of the Wyoming Water Research
Center at the University of Wyoming, Laramie, Wyoming; J. L. Engels, A. H.
Hall, K. M. Howe, J. M. Nicholson, D. V. Peck, M. E. Silverstem, and D. W.
Sutton of Lockheed Engineering and Management Services Company, Inc.,
Las Vegas, Nevada; L. W. Creelman of Radian Corporation, Austin, Texas; J.
W. Pfaff of the U.S. Environmental Protection Agency, Environmental
Monitoring and Support Laboratory, Cincinnati, Ohio; W. L. Kinney of the U.S.
Environmental Protection Agency, Environmental Monitoring Systems
Laboratory, Las Vegas, Nevada; and J. Messer of the U.S. Environmental
Protection Agency, Environmental Research Center, Corvallis, Oregon. The
word processing staff at Computer Sciences Corporation, Las Vegas, Nevada,
and the graphic arts staff at Donald Clark and Associates, Las Vegas, Nevada,
were essential to the timely production of this document. Recognition also is
due to R. D. Schonbrod of the U.S. Environmental Protection Agency,
Environmental Monitoring Systems Laboratory, Las Vegas, Nevada, who
served as Technical Monitor of this project.
-------
-------
7.0 Introduction
The National Acid Precipitation Assessment Program
(NAPAP) was initiated at the request of the
Administrator of the U.S. Environmental Protection
Agency (EPA) to evaluate the extent of the effects of
acidic deposition on aquatic resources within the
United States. When it became apparent that existing
data could not be used to assess quantitatively the
present chemical and biological status of surface
waters in the United States, the National Surface
Water Survey (NSWS) program was incorporated as
part of NAPAP to obtain that information. The
National Lake Survey (NLS) component of NSWS
comprises Phase I - Eastern Lake Survey (ELS-I),
Phase I - Western Lake Survey (WLS-I), and
Phase II - Temporal Variability (see Figure 1-1).
Data published in earlier studies are consistent with
the hypothesis that certain surface waters within the
United States have decreased in pH or alkalinity over
time. Acidic deposition has been suggested as a
contributor to such decreases. Also, numerous
studies have led to the conclusion that the effects of
acidic deposition on surface water chemistry are
influenced by variations in the characteristics of
lakes, streams, and their associated watersheds.
Existing data were compiled largely from numerous,
small-scale studies, and extrapolating these data to
the regional or national scale was difficult because
the studies often were biased in terms of site
selection. In addition, many previous studies were
incomplete with respect to the chemical variables of
interest, were inconsistent relative to sampling and
analytical methods, or were highly variable in terms
of data quality.
ELS-I, a synoptic survey of the chemistry of 1,612
lakes in the East, was conducted to obtain a regional
and national data base of water quality parameters
that are pertinent to evaluating the effects of acidic
deposition. To provide a base of information that is
complete and consistent in terms of the variables
measured and the sampling and analytical
procedures used, ELS-I was carried out on
representative lakes in the Southeast, Northeast, and
Upper Midwest. Detailed sampling procedures (Morris
et al., 1986), standardized analytical protocols
(Hillman et al., 1986), and a rigorous quality
assurance (QA) program (Drouse et al., 1986) were
implemented.
The WLS-I portion of NSWS, a synoptic survey of
757 lakes in the West, will be conducted during fall
overturn. During this overturn period chemical
variability within a lake is expected to be minimal as a
result of circulation within the water column. WLS-I
is designed to meet the following objectives for
designated regions of the West:
Determine the percentage (by number and area)
and location of lakes that are potentially
susceptible to change as a result of acidic
deposition and that have low acid neutralizing
capacity (ANC).
Investigate the relationships among water
chemistry, regional acidic deposition patterns,
land use, physiographic features, lake
morphology, and basin geometry within and
among regions.
Identify smaller subsets of representative lakes
for more intensive sampling in future surveys.
Of the lakes to be sampled during WLS-I, 455 lie
within designated wilderness areas. In order to
observe the guidelines and regulations set forth in
the Wilderness Act, almost all lakes located within
wilderness areas that have been selected for
sampling must be sampled by ground crews of the
U.S. Department of Agriculture - Forest Service.
The ground crews travel to lakes on foot or on
horseback. The lakes that are not in wilderness areas
are sampled by helicopter crews under the direction
of EPA.
Selected wilderness-area lakes that are inaccessible
to ground crews are sampled by helicopter crews
during periods when disturbance to wildlife and to
hikers is minimal. In addition, 45 wilderness area
lakes are sampled by ground crews and by helicopter
crews. The results for samples collected from these
45 calibration lakes will be used to evaluate the
comparability of ground crew and helicopter crew
sampling protocols for collecting and handling water
samples.
-------
Figure 1-1 Timetable for field activities of the National Surface Water Survey.
National Surface Water Survey (NSWS)
I
National Lake Survey (NLS)
Phase I - Synoptic Chemistry
Eastern Lakes (1984)
Western Lakes (1985)
National Stream Survey (NSS)
Phase I - Synoptic Survey
Pilot Survey (1985)
Synoptic Survey (1986)
Southeast Screening (1986)
Episodes Pilot (1986)
I
Temporal Variability (1986-87) I I Episodic Effects (1988) I
This calibration study is designed to meet three
goals:
Quantify the differences between the two
sampling methods (helicopter and ground
access).
Quantify the effects of holding samples for
different lengths of time prior to processing,
preservation, and analysis.
Quantify any significant mterlaboratory bias
between the two analytical laboratories that
analyze WLS-I samples.
Data derived from the chemical analyses conducted
during the calibration study will be used to establish
calibration factors that can be applied to analytical
values reported for all WLS-I samples. The
calibration factors are intended to eliminate value
differences that result from variations in sampling
protocol, sample holding time, or laboratory bias.
Two other studies are being conducted as part of
WLS-I. The purpose of one study, the nitrate/sulfate
stability study, is to compare sample preservation
methods and to study the effects of holding samples
for different lengths of time before preserving them.
The nitrate/sulfate sample is an extra aliquot that is
taken from the Van Dorn sampling unit, is preserved
with HgCI2, and is analyzed at the EPA
Environmental Monitoring Sytems Laboratory - Las
Vegas (EMSL-LV). The purpose of the second
study, the Corvallis study, is to compare results for
splits of the same sample when the splits have been
analyzed by different methods. The analytical
laboratories use NSWS protocols for atomic
absorption spectroscopy (AA), inductively coupled
plasma emission spectroscopy (ICP), automated
colonmetry, and ion chromatography, according to
the analyte; ERL-C uses ICP only and is not
restricted to the NSWS protocols and detection
limits. Another purpose of the Corvallis study is to
determine whether or not the ICP data can be
substituted in the data base if problems arise with the
standard analysis. Both studies can provide checks
on sampling, processing, and analytical performance.
To ensure that WLS-I procedures are performed
consistently and that the quality of the data generated
can be determined, the QA project plan for WLS-I
specifies the following measures:
Provide detailed written sampling methodology.
Simultaneously tram and test all personnel
participating in field activities.
Conduct on-site visits to each field operations
base and remote site soon after sampling begins,
and maintain daily telephone contact throughout
the sampling period to ensure that all methods
are being performed properly,
Perform extensive evaluation of analytical
laboratories throughout their participation.
Assess variability introduced at each level of
activity in field and analytical laboratories by
processing audit samples (synthetic and natural
-------
lake samples), duplicate samples, and blank
samples along with routine samples. (Field
laboratory refers to the on-site mobile labora-
tory that performs preliminary analyses and
aliquot preparation; analytical laboratory refers to
the off-site contract laboratory that performs the
more sophisticated analyses.)
Provide detailed, written analytical methodology.
Use internal quality control (QC) procedures at
the analytical laboratory to detect potential
contamination and to verify established detection
limits.
Enforce holding-time requirements.
Use protocols in the field and in the analytical
laboratory to confirm that reported data are
correct.
Enter data from standardized data reporting
forms into the data base twice, and scan for
outlying values to eliminate effects of transcrip-
tion errors.
Verify data by means of range checks, internal
consistency checks, and QA evaluations.
Validate verified data by identifying values that
are not typical of other sample values observed
for the group of lakes (e.g., stratum) from which
the sample or samples were drawn.
1.1 Document Content
This document is a supplement to the ELS-I
Analytical Methods Manual (Hillman et al., 1986).
This supplement discusses in detail the field
laboratory procedures that were not used during
ELS-I. It provides a general description of the
WLS-I field laboratory methods that were used
during ELS-I, and it gives a general description of
the analytical laboratory methods. Because the
analytical methods were identical for ELS-I and
WLS-I, this supplement provides only a general
discussion of the analytes and physical parameters
studied. Detailed descrtptions of the WLS-I sample
collection and processing methods are described in
Morris et al. (1985), in Peck et al. (1985), and in
Bonoff and Groeger (1987). ELS-I analytical
methods are given in Hillman et al. (1986). The
WLS-I QA plan is presented in Silverstem et al.
(1987), a document that is based on concepts
originally presented in Drouse et al. (1986).
1.2 Analytes and Physical Parameters
Measured
The analytes and physical parameters that will be
measured during WLS-I are described in sections
1.2.1 through 1.2.13. Table 1-1 lists the analytes,
physical parameters, and data quality objectives.
1.2.1 Acid Neutralizing Capacity
The acid neutralizing capacity (ANC, i.e., the quantity
of hydrogen ions reacted over a given pH range
during an acid titration) of a carbonate system is the
alkalinity of that system. The soluble species are
H2CO3, HCO3', and CO32. The calculations assume
that the lakes in this survey are represented by a
carbonate system; the ANC definition is in the
context of that system. Because of its inherent
relationship to buffering capacity, ANC is an
important variable in acidic deposition studies.
1.2.2 Aluminum, Total
Total aluminum is an estimate of the aluminum pool
that is potentially available to the biological
environment.
7.2.3 Aluminum, Exractable
Extractable aluminum is a component of dissolved
aluminum and includes most mononuclear aluminum
species. Aluminum is considered to be highly toxic,
especially to fish. Knowing its concentration is
important in assessing the biological environment of a
lake.
1.2.4 Base Neutralizing Capacity
The base neutralizing capacity (BNC) of a carbonate
system is the acidity of that system. The calculations
assume that the lakes in this survey are represented
by a carbonate system; the BNC definition is in the
context of that system.
7.2.5 Inorganic Carbon, Dissolved
The preliminary estimation of dissolved inorganic
carbon (DIG) is useful in determining whether or not
a lake is saturated with dissolved CO2. The
preliminary determinations and the analytical
laboratory determinations of DIG (when evaluated in
combination with pH measurements) are useful in QA
and QC calculations.
-------
Table 1-1. Analytes, Physical Parameters, and Data Quality Objectives, Western Lake Survey - Phase I (after Drouse et al.,
1986)
Site3
2, 3
2,3
3
3
3
3
1, 3
2, 3
3
3
3
3
3
3
3
3
3
1,2
3
3
3
Parameter*1
Al, Extractable
Al, Total
ANC
BNC
Ca
cr
Conductance
DIG
DOC
F", Total
dissolved
Fe
K
Mg
Mn
Na
NH/
NO3"
pH, Field
pH, Analytical
laboratory
P, Total
Si02
Method
Extraction with 8-
hydroxyquinoline into
methyl isobutyl ketone
followed by AAe (furnace)
AA8 (furnace)
Titration and Gran analysis
Titration and Gran analysis
AAe (flame) or ICP9
Ion chromatography
Conductivity cell and meter
Instrumental (acidification,
C02 generation, IR
detection)
Instrumental (UV-
promoted oxidation, CO2
generation, IR detection)
Ion-selective electrode
and meter
AAe (flame) or ICP9
AAe (flame)
AAe (flame) or \CPa
AA« (flame) or ICP9
AAe (flame)
Automated colonmetry
(phenate)
Ion chromatography
pH electrode and meter
pH electrode and meter
Automated colonmetry
(phosphomolybdate)
Automated colonmetry
Unit of
Measure
mg/L
mg/L
neq/L
peq/L
mg/L
mg/L
US/cm
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
pH units
pH units
mg/L
mg/L
Expected
Range0
0.005 - 1.0
0.005 - 1.0
-100 - 1,000
-10 - 150
0.5 - 20
0.2 - 10
5 - 1,000
0 05 - 1 5
0.1 - 50
0.01 - 02
0.01 - 5
01-1
0.1 - 7
0.01 - 5
0.5 - 7
001-2
0.01 - 5
3 - 8
3 -8
0 005 - 0.07
0 200 - 25
Required
Detec-
tion
Limits
0.005
0.005
l
I
0.01
0.01
h
0 05
0.1
0.005
001
0.01
0.01
0.01
001
001
0 005
-
0.002
005
Precision
(Relative Standard
Deviation [RSD] Upper
Limit [%])<'
10 (Al cone. >001 mg/L)
20 (Al cone. <0.01 mg/L)
10 (Al cone. >0 01 mg/L)
20 (Al cone. 5 mg/L)
10 (DOC cone. <5 mg/L)
5
10
5
5
10
5
5
10
±01'
±005'
10 (P cone. >0 01 mg/L)
20 (P cone. <0 01 mg/L)
5
Accur-
acy
(Max.
Absolute
Bias
[%])
10
20
10
20
10
10
10
10
5
10
10
10
10
10
10
10
10
10
10
±0 1'
±005'
10
20
10
(molybdate blue)
(continued)
-------
Table 1-1. (Continued)
Accur-
acy
Required Precision (Max
Detec- (Relative Standard Absolute
Site"
3
2
Parameter*"
SO/'
True color
Method
Ion chromatography
Comparison to platinum-
Unit of
Measure
mg/L
PCU<
Expected
Rangec
1 - 20
0 - 200
tion
Limits
005
0
Deviation [RSD] Upper
Limit [%]«
5
±5'
Bias
[%])
10
__
Turbidity
cobalt color standards
Instrument (nephelometer)
NTU*
2 - 15
10
10
a 1 = lake site, 2 = field laboratory, 3 = analytical laboratory.
b Dissolved ions and metals are being determined, except where noted.
c Ranges are for lake water
d Unless otherwise noted, this is the %RSD at concentrations greater than 10 times the required detection limit.
eAA = atomic absorption spectroscopy.
1 Absolute blank value must be < 10 u.eq/L.
a ICP = inductively coupled plasma atomic emission spectroscopy
h Blank must be <0.9 jiS/cm.
' Absolute precision goal in terms of applicable units
/ American Public Health Association platinum-cobalt unit.
* Nephelometnc turbidity unit
7.2.6 Ions, Dissolved (Ca, CI; Fe, K, Mg, Mn, Na,
NH4 + , NO3; SO42; and Total F-)
Dissolved ions are determined so that the lake can
be characterized chemically, especially for mass ion
balance and buffering capacity. Fluoride is also
important as a natural chelator of aluminum.
7.2.7 Organic Carbon,Dissolved
Dissolved organic carbon (DOC) determination is
necessary to establish the relation between the
organic carbon content and the true color of tha lake
water. DOC can be used to estimate the
concentration of organic acids. It is also important as
a natural chelator of aluminum.
7.2.8 pH
The pH is a general and direct indication of free
hydrogen ion concentration.
7.2.9 Phosphorusjotal
Total phosphorus is an indicator of overall trophic
status and of nutrients that are potentially available to
phytoplankton.
7.2.70 Silica,Dissolved
The absence or existence of dissolved silica (SiO2) is
an important factor controlling diatom blooms; the
determination also assists in identifying trophic
status. Dissolved silica is also an indication of mineral
weathering.
7.2.77 Conductance
The conductance of lake water is a general indicator
of its ionic strength and is related to buffering
capacity.
7.2.72 True Color
True color is an indicator of natural BNC and DOC.
Substances- that impart color also may be important
natural chelators of aluminum and of other trace
elements. Color is measured in platinum-cobalt
units (PCUs; APHA et al., 1985).
7.2.73 Turbidity
Turbidity is a measure of suspended material in lake
water and is measured in nephelometric turbidity
units (NTUs).
-------
1.3 References
American Public Health Association, American Water
Works Association, and Water Pollution Control
Federation, 1985. Standard Methods for the
Examination of Water and Wastewater, 16th Ed.
APHA, Washington, D.C.
Bonoff, M. B., and A. W. Groeger, 1987. National
Surface Water Survey, Phase I Western Lake
Survey Field Operations Report. U.S.
Environmental Protection Agency, Las Vegas,
Nevada.
Drouse, S. K., D. C. Hillman, L. W. Creelman, and S.
J. Simon, 1986. National Surface Water Survey,
Eastern Lake Survey (Phase I -- Synoptic
Chemistry) Quality Assurance Plan. EPA 600/4-
86-008. U.S. Environmental Protection Agency,
Las Vegas, Nevada.
Hillman, D. C., J. F. Potter, and S. J. Simon, 1986.
National Surface Water Survey, Eastern Lake
Survey (Phase I -- Synoptic Chemistry)
Analytical Methods Manual. EPA 600/4-86-
009. U.S. Environmental Protection Agency, Las
Vegas, Nevada.
Morris, F. A., D. V. Peck, D. C. Hillman, K. J. Cabbie,
S. L. Pierett, and W. L. Kinney, 1985. National
Surface Water Survey, Western Lake Survey
(Phase I) Field Training and Operations Manual,
EPA Internal Report. U.S. Environmental
Protection Agency, Las Vegas, Nevada.
Morris, F. A., D. V. Peck, M. B. Bonoff, and K. J.
Cabbie, 1986. National Surface Water Survey,
Eastern Lake Survey (Phase I - Synoptic
Chemistry) Field Operations Report. EPA 600/4-
86/010. U.S. Environmental Protection Agency,
Las Vegas, Nevada.
Peck, D. V., R. F. Cusimano, and W. L. Kinney,
1985. National Surface Water Survey, Western
Lake Survey (Phase I - Synoptic Chemistry)
Ground Sampling Training and Operations
Manual, EPA Internal Report. U.S. Environmental
Protection Agency, Las Vegas, Nevada.
Silverstem, M. E., S. K. Drouse, J. L. Engels, M. L.
Faber, and T. E. Mitchell-Hall, 1987. National
Surface Water Survey, Western Lake Survey
(Phase I - Synoptic Chemistry) Quality
Assurance Plan. U.S. Environmental Protection
Agency, Las Vegas, Nevada.
-------
2.0 Field Laboratory Operations
Field laboratory activities are described in detail in the
WLS-I Field Training and Operations Manual (Morris
et at., 1985) and the WLS-I Field Operations Report
(Bonoff and Groeger, 1987); field laboratory analytical
methods are presented in the ELS-I Analytical
Methods Manual (Hillman et al., 1986). Lists of the
equipment kept at each field laboratory are given in
the WLS-I Ground Sampling Training and
Operations Manual (Peck et al., 1985) and in Morris
et al. (1985). Daily activities conducted at each field
laboratory are shown in Figure 2-1 and are
described in Section 2.1.
Figure 2-1. Flow scheme of daily field laboratory activities, Western Lake Survey
-Phase I.
Before
Sample Arrival
1. Prepare reagents for analysis:
a) Total extractable Al
b) DIC
c)pH
2. Warm up and calibrate instruments:
a) Turbidimeter
b) Carbon analyzer
c) pH meter
^
1.
r
mple Arrival
Confirm with
of samples a
Following
Sample Arrival
1. Insert required audit samples, assign batch
and sample ID numbers, start batch form
2. Measure DIC
3. Measure pH
4. Measure turbidity
5. Measure true color
6. Complete batch and shipping forms
Shipment
of Samples
and Forms
-------
2.1 Daily Operations
2.7.7 General Daily Operations
Field laboratory operations are based at fully
equipped laboratory trailers. The five field laboratory
trailers, one stationed at each WLS-I field base,
provide facilities for sample receipt, analysis, and
preservation, and for aliquot preparation and
shipping. Lake samples, blank samples, and duplicate
samples, which are collected and labeled by the
sampling crews, are delivered to the field laboratory
for preliminary analysis and processing. Samples are
filtered, preserved, and divided into aliquots. Natural
and synthetic audit samples are added to the batches
at the field laboratory, and the completed batches are
shipped by overnight courier from the field laboratory
to analytical laboratories that perform detailed sample
analyses. Split samples (collected from the Van Dorn
sampler by the ground crews only) are preserved
with HgCI2 at the lake site before they are delivered
to the field laboratory. These unfiltered split samples
are labeled and are shipped to the EPA
Environmental Monitoring Systems Laboratory in Las
Vegas, Nevada (EMSL-LV), for use in the nitrate-
sulfate study. (This type of split sample was not
collected during ELS-I.) In addition, the field
laboratory prepares the split samples for shipment to
the EPA Environmental Research Laboratory in
Corvallis, Oregon (ERL-C). These splits are filtered,
then they are preserved with HN03. At ERL-C, the
splits are analyzed by ICP. The field laboratory also
serves as the distribution center for water used to
prepare blank samples, for conductivity standards
used to perform Hydrolab calibration, and for supplies
used at the lake sites. In general, WLS-I field
laboratory operations parallel ELS-I field laboratory
operations; however, WLS-I employs two sampling
methods (sampling from helicopters and sampling
from boats) that require some differences in sample
processing (see Section 2.1.2 and Morns et al.,
1985).
Overall administration of the field laboratory is the
responsibility of the field laboratory coordinator. The
field laboratory coordinator provides the helicopter
crews with water for blank samples and with
conductivity standards for Hydrolab calibration.
Supplies for the ground crew are provided through
the field laboratory coordinator by the logistics
coordinator. (The position of logistics coordinator was
created for WLS-I.) The laboratory coordinator
receives the lake samples, lake data forms, and
cham-of-custody forms (from the helicopter crews
directly and from the ground crews through the U.S.
Department of Agriculture - Forest Service field
manager). The coordinator also receives field audit
samples from the designated supplier. (Laboratory
audit samples, which were used during ELS-I, are
not used for WLS-I activities.) The laboratory
coordinator organizes the routine, duplicate, blank,
and audit samples into batches, labels the samples
with batch and sample identification (ID) numbers,
initiates sample batch forms, and maintains a tracking
form for all sample batches. After the sample batches
are organized, the laboratory coordinator refrigerates
the samples at 4°C until they are removed for
analysis or processing. The laboratory coordinator
also is responsible for maintaining and ordering all
supplies through the NSWS Communications Center
in Las Vegas, Nevada.
The laboratory supervisor, who is responsible for
managing daily laboratory operations, performs
dissolved inorganic carbon (DIG) and pH
determinations. The supervisor also designates one
sample in each batch as the trailer duplicate. This
sample is analyzed in duplicate for DIG, pH, turbidity,
and true color.
Three laboratory analysts share sample processing
and sample analysis duties. One analyst performs
aluminum extractions; the second analyst performs
sample filtration; the third analyst prepares and
preserves aliquots, prepares split samples for ERL-
C, and performs turbidity and true color
determinations.
After a set of aliquots is processed, each aliquot
bottle and aliquot centrifuge tube (for extractable
aluminum analysis) is sealed with electrical tape and
is placed in an individual plastic bag; the bag then is
tied with a twist-tie. Next, the set of aliquot bottles
is placed in a 1-gallon Ziploc bag; this bag is sealed
also. The aliquot bottles and the centrifuge tube are
refrigerated at 4°C until they are prepared for
shipment.
For transport to the analytical laboratories, each set
of aliquots and the corresponding shipping forms are
placed in an insulated shipping box that contains
enough frozen freeze-gel packs to maintain the
insulated compartment at 4°C. The boxes, which
may contain as many as eight samples, are shipped
to the analytical laboratories via overnight courier
service. The same packing and shipping procedures
are used to ship split samples to EMSL-LV and to
ERL-C, except that ERL-C split samples are not
refrigerated during shipping and are shipped in
multiple batches rather than in daily batches. (During
ELS-I, these samples were refrigerated and they
were shipped daily.)
All forms that are completed at the lake site (the lake
data form and chain-of-custody form) and at the
field laboratory (the batch/QC field data form, the
-------
shipping form, and the chain-of-custody form) are
reviewed by the laboratory coordinator. Copies of all
forms are sent to the QA staff at EMSL-LV where
they are reviewed for data accuracy and
completeness. Copies of the lake data and batch/QC
field data forms are sent to the data base manager at
Oak Ridge National Laboratory (ORNL) in Oak Ridge,
Tennessee, and are used for data entry. Figure 2-2
is a flow scheme of the shipping and data forms used
during WLS-I.
2.7.2 Calibration Study Operations
Most processing and analysis procedures for the
calibration study are identical to those used for other
WLS-I sample batches. Specific sampling and
analysis procedures for the calibration study are
described in detail in Morris et al. (1985); a general
discussion is presented here.
Each calibration lake is sampled by one helicopter
crew and by one ground crew. The two crews collect
samples from approximately the same location (the
deepest spot) on the lake. The ground crew samples
the lake first, and the helicopter crew samples the
lake as soon as possible thereafter (optimally, within
1 hour). The ground crew collects a routine sample
and a duplicate sample; the helicopter crew collects
a routine sample, a duplicate sample, and a triplicate
sample. Both types of sampling crews use sample
collection techniques standard for all WLS-I lakes.
Use of the ground-access method presents
logistical problems that are unique to the WLS-I
sampling design. Ground crews that sample lakes in
wilderness areas might deliver samples to the field
laboratory as long as 1 to 5 days after collection. As
a result, designers of WLS-I are interested in the
possible effects of delayed sample preservation (or
"holding time").
To evaluate the effects of holding time, the field
laboratory preserves the ground crew's samples on
the collection date and preserves two of the
helicopter crew's samples on the collection date. The
third sample collected by the helicopter crew is not
preserved on the collection date. Instead, this
sample, which is randomly selected from among the
three samples collected by the helicopter crew, is
refrigerated at the field laboratory for a specified
length of time before it is preserved. The holding time
for the withheld sample is equivalent to the time
difference between the receipt of the samples
collected by the ground crew and by the helicopter
crew.
The calibration study also is designed to provide data
that can be used to evaluate significant, systematic
differences in analytical results between analytical
laboratories. To meet this goal, the field laboratory
sends one sample from each routine-duplicate pair
(collected by a ground crew) and one sample from
Figure 2-2. Shipping and data form flow scheme, Western Lake Survey
Phase I.
Analytical
Laboratory
Shipping Form (2 Copies)
Shipping Form
(1 Copy) .
Sample
Management
Office
Data Base
(ORNL)
Field
Laboratory
(Keeps
1 Copy of
Each Form)
Shipping
Form
QA Manager U_
(EMSL-LV) r
All Forms
Lake Data and
Batch QC Field
Data Forms
-------
each routine-duplicate-triplicate set (collected by a
helicopter crew) to each analytical laboratory.
The remaining sample from each set collected by
helicopter is the one that is withheld at the field
laboratory for later processing and preserving; the
assignment of a sample (routine, duplicate, or tripli-
cate) to a particular analytical laboratory is made
randomly and individually for every pair or set of
samples. Figure 2-3 illustrates the sample
assignment procedure that is used.
Figure 2-3. Sample flow for the calibration study.
Ground Samples Helicopter Samples
(Forest Service) (Lockheed-EMSCO, EPA)
Routine Duplicate
1st 2nd
Sample Sample
Taken Taken
RGC DGC
Routine
1st
Sample
Taken
Duplicate Triplicate
2nd 3rd
Sample Sample
Taken Taken
1 RHC DHC THC
i , '
Field Laboratory
~^
WW
1 * * . J
RGC DGC
7 7
Aliquots Aliquots
RHC
7
Aliquots
DHC THC
7 7
Aliquots Aliquots
Randomly Selected T
|^ Sample Shipment |
/-
1
1
1 1
t *
Analytical
Laboratory
- ~~ -s
Withheld
Helicopter "
Sample
1 I
1 1
,J L*
-_-J
t t
Alternate
Analytical
Laboratory
RGC-Routme Ground Calibration
DGC-Duphcate Ground Calibration
RHC-Routme Helicopter Calibration
DHC-Duphcate Helicopter Calibration
THC-Triplicate Helicopter Calibration
10
-------
Specific sample codes are recorded on NSWS Form
2 - Batch/QC Field Data (see Figure 2-4) for these
reasons: (1) to distinguish the calibration lake
samples from the routine lake samples, (2) to
distinguish samples collected by the helicopter crews
from samples collected by the ground crews, and (3)
to distinguish samples collected by the helicopter
crews and withheld for later processing from samples
processed on the day of sample collection.
Samples collected by the ground crews are coded
RGC (routine ground calibration) and DGC (duplicate
ground calibration). The samples collected by the
helicopter crews are coded RHC (routine), DHC
(duplicate), and THC (triplicate). A "W" is added after
the last letter of one of the helicopter sample codes
to indicate that the sample is being withheld (e.g., if
the RHC sample is withheld, it is coded RHCW on
the batch form).
When the calibration lake samples collected by the
helicopter crews arrive at the field laboratory, they
are separated into batches, are assigned
identification (ID) numbers, and are shipped to the
analytical laboratory. This process is shown in Figure
2-5. The laboratory coordinator uses the random
selection procedure mentioned above to determine
(1) which sample is withheld at the field laboratory (at
4°C in the dark), (2) which samples are processed
with the corresponding samples collected by the
ground crew, and (3) which analytical laboratory
analyzes each sample.
Thus, the five samples collected from a single
calibration lake are processed in three different
batches: one sample collected by the helicopter crew
and one sample collected by the ground crew are
incorporated into the sample batch that is sent to the
"regular" analytical laboratory (the laboratory to
which the field base ships its usual, daily sample
batch). One sample collected by each crew is
incorporated into the sample batch that is sent to the
"alternate" analytical laboratory (the laboratory used
regularly by another field base). The last sample (the
withheld sample designated from among the samples
collected by the helicopter crew) is kept at the field
base for incorporation into a sample batch shipped
on a different day.
2.2 Quality Control Activities
In accordance with the QA plan for WLS-I
(Silverstein et al., 1987) adequate records (in the
form of logbooks, data forms, and shipping forms)
must be maintained, and a uniform sample-coding
procedure must be used. Table 2-1 shows the
sample codes that are used for routine, duplicate,
audit, and split samples. The data forms and shipping
forms are illustrated in Silverstein et al. (1987).
2.3 Field Laboratory Methods
2.3.1 Determination of Dissolved Inorganic
Carbon
At the lake site, samples for DIG determination are
collected and sealed in syringes. At the field
laboratory, a syringe filter is attached to the syringe,
and the sample is filtered into the sample loop of a
Dohrman DC-80 carbon analyzer. Subsequently, the
sample is injected into a reaction chamber that
contains 5 percent phosphoric acid. The carbonates
(DIG) in the sample react with the acid to form C02,
which is sparged from the reaction chamber with a
nitrogen gas carrier stream. The CO2 in the carrier
stream then is detected and quantified (in terms of
DIG) by an infrared (IR) C02 analyzer. See Hillman et
al. (1986) for a detailed description of the procedure.
2.3.2 Determination of pH
At the lake site, samples for pH determination are
collected and sealed m syringes. At the field
laboratory, the pH sample is injected into a closed
system to prevent exposure of the sample to the
atmosphere. A pH meter and an electrode are used
to determine pH. See Hillman et al. (1986) for a
detailed description of the procedure.
2.3.3 Determination of Turbidity
At the lake site, samples are collected in Cubitainers.
At the field laboratory, turbidity is determined using
an unfiltered sample and a Monitek Model 21
nephelometer. See Hillman et al. (1986) for a detailed
description of the procedure.
2.3.4 Determination of True Color
At the lake site, samples are collected in Cubitainers.
At the field laboratory, true color is determined on the
supernatant of an unfiltered, centrifuged sample. A
Hach Color Determination Kit is used for this
procedure. See Hillman et al. (1986) for a detailed
description of the procedure.
2.3.5 Aliquot Preparation
Table 2-2 lists the aliquots and split samples
prepared as well as the containers, preservaties, and
analytes for each aliquot.
11
-------
Figure 2-4. National Surface Water Survey Form 2 - Batch/QC Field Data
Date Received
National Surface Water Survey By Data Mgt
Form 2
Batch QC Field Data
Entered
Re-Entered
D,m> Shipped
Ldb Crew ID .
BHC (mg L)
OCCS Limits
UCL 2 2
LCL 1 8
Vak
Sl.iuon pH
OCCS Limits
UCL 4 1
LCL J 9
ValiK
Turbidity |NIU)
OCCS Lim.ts
UCL b b
LCL 4 b
Cclor
(APHA
Units)
Split
Codes
(E LI
21
TD
Comments
WhiteORNL Copy YellowField Copy Pink EMSL-LV Copy
12
-------
Figure 2-5. Preparation, identification, and shipment of sample batches for the calibration study.
Regular
Analytical Laboratory
I 1
I Normal Lake
Samples ' ~|
I 1 L.
Alternate
Analytical Laboratory
Shipment -from
Both Field
Laboralor les
Combined into
One Field Batch
and Analyzed
Normallv
Calibration Lake
Samples (3 of 5)
Randomly Separated
from Normal Sample
Batch
Reference Field
Laboratory
Coordinator Called
for ' Borrowed ' Batch
ID and Two Sample ID
Numbers
One Helicopter
Sample from Each
Calibration Lake
Withheld for Future
Processing
R eference
Field Laboratory
Reference Field
Laboratory in
Normal Sample
Processing Mode
Two Sample IDs from
Day's Batch "Lent" to
Principal Field
Laboratory
. Normal Lake
' Samples '
2.4 References
Bonoff, M. B., and A. W. Groeger, 1987. National
Surface Water Survey, Phase I Western Lake
Survey Field Operations Report. U.S.
Environmental Protection Agency, Las Vegas,
Nevada.
Hillman, D. C., J. F. Potter, and S. J. Simon, Ib86.
National Surface Water Survey, Eastern Lake
Survey (Phase I - Synoptic Chemistry)
Analytical Methods Manual. EPA 600/4-86-
009. U.S. Environmental Protection Agency, Las
Vegas, Nevada.
Morns, F. A., D. V. Peck, D. C. Hillman, K. J. Cabbie,
S. L. Pierett, and W. L. Kinney, 1985. National
Surface Water Survey, Western Lake Survey
(Phase I) Field Training and Operations Manual.
EPA Internal Report. U.S. Environmental
Protection Agency, Las Vegas, Nevada.
Peck, D. V., R. F. Cusimano, and W. L. Kinney,
1985. National Surface Water Survey, Western
Lake Survey (Phase I - Synoptic Chemistry)
Ground Sampling Training and Operations
Manual. EPA Internal Report. U.S. Environmental
Protection Agency, Las Vegas, Nevada.
Silverstein, M. E., S. K. Drouse, J. L. Engels, M. L.
Faber, and T. E. Mitchell-Hall, 1987. National
Surface Water Survey, Western Lake Survey
(Phase I - Synoptic Chemistry) Quality
Assurance Plan. U.S. Environmental Protection
Agency, Las Vegas, Nevada.
13
-------
Table 2-1. Sample Codes Used to Complete Lake Data Forms, Western Lake Survey - Phase I
Sample Type
Code
Description
Normal3
Calibration Study3
Audit
RH
RG
DH
DG
BH
BG
TB
TD
RHCb
DHC"
THCb
RGC
DGC
FL1-001
Routine lake sample - helicopter
Routine lake sample - ground
Duplicate lake sample - helicopter
Duplicate lake sample - ground
Field blank - helicopter
Field blank - ground
Field laboratory (trailer) blank
Field laboratory (trailer) duplicate
Routine calibration - helicopter
Duplicate calibration - helicopter
Triplicate calibration - helicopter
Routine calibration - ground
Duplicate calibration - ground
-Radian ID number
- Concentrate (or lake) lot number
Natural or synthetic (concentration level)
N = natural L = low synthetic
Type of audit sample
F = field (no other type used)
Split
. Shipping destination of split sample
E = ERL-C
L = EMSL-LV
A split sample is an additional aliquot from a routine, duplicate, audit, or blank
sample. It has the same ID number as the original sample; however, the
aliquot (split) has an additional sample code. For example, if the original
sample is assigned the ID number 4, the split sample also receives the ID
number 4, with the letter E (or L) recorded in the "split code"' column on the
batch/QC form
a Samples require a lake ID, except for TB.
0 A "W" is added to this sample code to indicate that the sample was withheld as part of the calibration study.
Table 2-2. Aliquots, Containers, Preservatives, and Corresponding Analyses, Western Lake Survey - Phase I
Aliquot3 Container Volume Preservative and Description Analyses
1
2
3
4
5
6
7
EMSL-LV Splits
ERL-C Splits
250 mL
10 mL
250 mL
125 mL
500 mL
125 mL
125 mL
125 mL
125 mL
Filtered, preserved with HNO3 to pH < 2
Filtered, preserved with methyl isobutyl
ketone-8-hydroxyqumoline extract
Filtered, no preservative
Filtered, preserved with H2SO4 to pH < 2
Unfiltered, no preservative
Unfiltered, preserved with H2SO4 to pH <2
Unfiltered, preserved with HNO3 to pH < 2
Unfiltered, preserved with 0.1 mL HgCI2
Filtered, preserved with HNO3 to pH < 2
Ca, Fe, K, Mg, Mn, Na
Extractable Al
Cr, NO3', SiO2, SO42",
total dissolved F"
DOC, NH/
ANC, BNC, DIC, pH, conductance
Total P
Total Al
NO3', SO42"
Ca, Fe, K, Mg, Mn, Na, SiO2, SO42" total
Al, total P
a Aliquots 2,3,4,5, and 6 and the EMSL-LV split samples must be stored at 4°C in the dark. Although it is not required that aliquots 1 and 7
be stored at 4°C in the dark, it is recommended that these aliquots be refrigerated after analysis to minimize evaporation. ERL-C split
samples are stored at room temperature.
14
-------
3.0 Analytical Laboratory Operations
Analytical laboratory procedures that are used for
WLS-I are identical to the ones that were used
during ELS-I. Procedures for sample receipt and
handling, sample analysis, QC, and data reporting are
detailed in the ELS-I Analytical Methods Manual
(Hillman et al., 1986). Hillman et. al. (1986) contains
appendices that describe the cleaning of laboratory
plasticware, provide facsimiles of data forms used in
the analytical laboratory, and give examples of
calculations required for ANC and BNC
determinations and for anion-cation balance and
conductivity balance determinations. The major steps
that must be followed to perform the ANC, BNC,
anion-cation balance, and conductivity balance
determinations also are presented in the appendix to
this WLS-I manual.
3.1 Daily Operations
Analytical laboratory personnel receive the samples
shipped to them by the field laboratory, inspect the
samples for damage, log in the sample batches,
analyze the samples according to procedures
described in Hillman et al. (1986), and prepare and
distribute data packages that document the analyses
performed (Silverstein et al., 1987).
When the samples arrive from the field laboratory,
the overnight courier service delivers the samples or
notifies the analytical laboratory that the samples
have arrived at the airport. The analytical laboratory
is responsible for picking up any shipment that is not
delivered to them by the courier within 24 hours of
notification. Analytical laboratory personnel inspect
the shipping boxes to ensure that (1) all aliquot
containers are present and intact, (2) the sample IDs
match the ones listed on the shipping forms, and (3)
samples have been maintained at 4°C or colder
during transport. Laboratory personnel note all
deviations from these conditions on the shipping
form, and a copy of the annotated form is sent to the
Sample Management Office. In addition, if the
samples have not been maintained at 4°C, the QA
manager is notified immediately.
After samples are logged in, they are analyzed
according to the analytical procedures and QA/QC
procedures specified in Hillman et al. (1986). Each
sample consists of seven aliquots that are prepared
at the field laboratory; each aliquot is processed
differently, depending on the analytes for which the
aliquot is to be analyzed. In addition, each analysis
must be conducted within a specified holding time.
The variables that are measured and the
measurement methods that are used are listed in
Table 1-1. Descriptions of each aliquot and the
corresponding variables to be measured are given in
Table 2-2, and the maximum holding time for each
variable is listed in Table 3-1. This holding time
refers to the number of days it takes to analyze the
sample in the analytical laboratory after the sample
has been processed and preserved in the field
laboratory, not the time it takes to transport the
sample from the lake site to the field laboratory for
preservation and processing.
Table 3-1. Maximum Holding Times for Samples, Western
Lake Survey Phase I
Holding Time
Variable
7 days
14 days
28 days
28 days3
Extractable Al, NO3', air-
equilibrated pH
ANC, BNC, conductance, DIG,
DOC
Cf, NH4 + , total P, S
-------
percent conductance difference calculation for
each sample (check required; submission of form
optional)
percent ion balance difference calculation for
each sample (check required; submission of form
optional)
ion chromatograph resolution specifications
instrument detection limits
sample holding times and date of sample analysis
calibration and reagent blank values; quality
control check sample (QCCS) values
matrix spike percent recovery calculations
internal duplicate precision as percent relative
standard deviation (%RSD)
standard additions analysis results (when
applicable)
The data package also includes a cover letter from
the manager of the analytical laboratory to the QA
manager. The letter identifies the batch number and
the number of samples analyzed, discusses problems
associated with the analyses, describes deviations
from protocol, and contains other information that the
laboratory manager considers to be pertinent to the
particular sample batch. Copies of the completed
data package are sent to the QA staff and to the data
base manager.
When the analyses have been completed and the
data packages have been distributed, the aliquots are
stored at the analytical laboratory according to
storage requirements specified in the Statement of
Work and in Hillman et al. (1986). The analytical
laboratory is required to store the samples for at least
6 months and must receive written notification from
the QA manager before the samples can be
discarded.
On the basis of the analytical results reported for
internal and external QA and QC samples, the QA
staff, with the approval of the QA manager, can
request that the analytical laboratory confirm reported
values or that they reanalyze selected samples or
sample batches. In such cases, sample reanalysis
usually is requested for a given variable on a per-
batch basis.
3.2 Quality Control Activities
The QC procedures required for WLS-I are
described in Silverstein et al. (1987) and are
discussed fully in Drouse et al. (1986). A summary is
presented below. Table 3-2 summarizes the QC
checks that are incorporated in the analytical
procedures, the control limits for the results, and the
corrective actions that are required if the results are
unacceptable. Table 3-3 lists the control limits for
analytical results that are derived from blind QC
samples sent to the analytical laboratories. The
control limits are expressed as the percent difference
from the theoretical concentration.
On the basis of analytical results, the analytical
laboratories must calculate the conductance
expected for a sample and must compare that value
to the measured value. If the difference between the
two values exceeds the maximum allowable
difference, suspect results are redetermined. Table
3-4 lists the conductance factors of the ions
determined, and Table 3-5 gives the chemical
reanalysis criteria.
The equations used to determine percent relative
standard deviation (%RSD) between duplicate
samples, matrix spike percent recovery for matrix
spike samples, percent ion balance difference
between cations and anions in a sample, and percent
conductance difference between the measured and
calculated conductance summed from the specific
conductance of each measured ion in a sample are
given below.
3.2.1 Percent Relative Standard Deviation
Prepare and analyze one sample per batch in
duplicate. Calculate the %RSD between duplicates.
(NOTE: For pH, the absolute standard deviation is
used instead of the %RSD.) The duplicate precision
%RSD must not exceed the value given in Table 1-
1.
%RSD = X 100
x
s =
S(x -x)'
1/2
n - 1
16
-------
Table 3-2. Summary of Internal Quality Control Checks for Analytical Methods, Western Lake Survey - Phase I
Parameter
QC Check
Control Limits
Corrective Action3
1 . Titrant standardization
cross-check
2. Electrode calibration
(Nernstian response
check)
ANC, BNC, pH 3. pH QCCS (pH 4 and 10)
analysis
4. Blank analysis (KCI
spike)
5. Duplicate analysis
6. Protolyte comparison
la. Initial QCCS analysis
(calibration and
verification)
1b Continuing QCCS
analysis (every 1 0
samples).
Ions (CI", total dissolved F", 2a. Detection limit
NH4 + , NO3". SO42" determination (weekly)
2b Detection limit QCCS
1 . Relative differences
2. Slope = 1.00 ± 0.05
3. pH 4 = 4.00 ± 0 05
pH 10 = 10.00 + 0.05
4. Blank < 10 neq/L (ANC
and BNC only)
5. RSD <10% (ANC and
BNC only)
± 0.05 pH unit (pH only)
6. See Hillman etal. (1986)
1 a.b. The lesser of the 99%
confidence interval or
value given in Table 3-3
2a. Detection limits given in
Table 1-1
2b. % Recovery = 100 ±
1 . Restandardize titrants.
2. Recalibrate or replace
electrode.
3 Recalibrate electrode
4. Prepare fresh KCI spike
solution.
5 Refine analytical
technique.
Analyze another
duplicate.
6 See Hillman etal (1986)
la. Prepare new standards
and recalibrate.
ib. Recalibrate. Reanalyze
associated samples.
2a,b. Optimize instrumentation
and technique.
Metals (total Al, extractable Al,
Ca, Fe, K, Mg, Mn, Na)
SiO2, total P, DIG, DOC,
conductance
analysis daily (metals and
total P only)
3. Blank analysis
4. Duplicate analysis
Matrix spike (except ext.
Al, DIG, and
conductance)
6. Resolution test (CI",
NO3', SO42' only)
20%
3a. Blank <2 x detection
limit (except
conductance)
3b. Blank <09 nS/cm
(conductance only)
4. Duplicate precision
(%RSD) limits given in
Table 1-1
5. % Recovery = 100 ±
15%
Resolution > 60%
3a.b. Determine and eliminate
contamination source
Prepare fresh blank
solution. Reanalyze
associated samples
4 Investigate and eliminate
source of imprecision.
Analyze another
duplicate.
5 Analyze 2 additional
spikes. If one or both are
outside control limits,
analyze all samples in
that batch by method of
standard additions
6. Clean or replace ion
chromatograph separator
column Recalibrate.
a To be followed when QC check is outside control limits.
17
-------
Table 3-3. Maximum Control Limits for Quality Control
Samples, Western Lake Survey - Phase I
Table 3-5. Chemical Reanalysis Criteria, Western Lake
Survey - Phase I
Variable
Maximum Control Limit for QC
Sample3
Al, extractable
Al, total
Ca
ci-
Conductance
DIC
DOC
F", total dissolved
Fe
K
Mg
Mn
Na
NH4 +
N03-
P, total
SiO2
S042-
± 20%
± 20%
±5%
±5%
±2%
±10%
± 1 0%
±5%
±10%
±5%
±5%
± 1 0%
±5%
± 1 0%
± 1 0%
±20%
±5%
±5%
A. Amon-Cation Balance
Total Ion Strength (neqlL)
<50
>50<100
>100
B. Conductance
Measured Conductance
(fiS/cm)
<5
>5<30
>30
Maximum
% Ion Balance Difference3
60
30
15
Maximum
% Conductance Difference3
50
30
20
a If the absolute value of the percent difference exceeds these
values, the sample is reanalyzed. When reanalysis is indicated,
the data for each variable are examined for possible analytical
error. Any suspect results then are redetermmed, and the percent
differences are recalculated (Peden, 1981). If the differences still
are unacceptable or if suspect data are not identified, the QA
manager should be consulted for guidance.
[H+]moles/L = 10~pH
pH = pH determined at V = 0 of the
acidity titration.
Kw
a This limit is the maximum allowable percent difference from the
theoretical concentration of the QC sample.
[H + ]
5.080[DIC(mg/L)][H + ] K,
HC03- =
K
Table 3-4. Conductance Factors of lonsa
2-
4.996 [DIC (mg/L) ] KjK2
Ion
Ca2 +
cr
C032-
HC03-
Mg-
Conductance
(p.S/cm at
25°C) per mg/L Ion
2.60 Na*
2.14 NH4 +
2.82 SO42"
3.5 x 105 NO3"
(per mole/L)
0.715 K +
3.82 OH"
Conductance
(jaS/cm at
25°C) per mg/L
2 13
4.13
1.54
1.15
1.84
1.92 x 105
(per mole/L)
[H+]2 + [H + ]K + K
Kt =4.4463 X 10~7
K2 = 4.6881 X 10~U
a APHA et al. (1985) and Weast (1972).
3.2.2 Percent Matrix Spike Recovery
Prepare one matrix spike with each batch
a portion of a sample with a known c
K2
by spikinc
luantitv o
18
-------
analyte. The spike concentration must be the larger
of two times the endogenous level or ten times the
required detection limit for the analyte. The volume of
the spike added must be negligible (less than or
equal to 0.001 of the aliquot volume). The spike
recovery must be 100 ± 15 percent and is
calculated as follows:
% Matrix Spike Recovery =
measured
concentration
of sample
plus spike
measured
concentration
of unspiked
sample
X 100
(actual concentration of spike added)
3.2.3 Percent Ion Balance Difference
Theoretically, the ANC of a sample equals the
difference between the concentration (in equivalents
per liter) of the cations and the anions in the sample
(Kramer, 1982). In practice, this is rarely true, owing
to the effects of analytical variability and of ions that
are present but that are not measured. The percent
ion balance difference (%IBD) is performed as an
internal QC check.
For each sample, calculate the %IBD as follows:
ANC + S anions S cations
% IBD = X 100
TI
where:
TI (total ion strength) = S anions + S cations +
ANC + 2 [H + ]
Z, anions = [CI'] + [F] + [NO3'] +
S cations = [Na + ] + [K + ] + [Ca2 + ] + [Mg2*] +
[NH4 + ]
[H + ] = (10-pH) x 106 ueq/L
The %IBD must not exceed the values given in Table
3-5.
3.2.4 Percent Conductance Difference
Estimate the conductance of a sample at 25 °C by
summing the equivalent conductances for each
measured ion. Calculate the equivalent conductance
for each ion by multiplying the ion concentration by
the appropriate factor in Table 3-4. (NOTE: Only
major ions are included in the calculation.) Calculate
the percent conductance difference (%CD) as
follows:
calculatedcond. measured cond.
%CD = x 100
measured conductance
The %CD must not exceed the limits listed in Table
3-5.
3.3 Analytical Laboratory Methods
General descriptions of the methods that are used by
the analytical laboratories are given here. The
methods are discussed in detail in Hillman et al.
(1986). Because ANC, BNC, and pH determinations
are central to the study, determination of these
variables is reviewed in detail in the appendix to this
manual. This appendix is a compilation of the
information given in Section 4.0 and Appendix C of
Hillman et al. (1986).
3.3.1 Determination of ANC, BNC, and pH
This procedure is applicable to the determination of
ANC, BNC, and pH in weakly buffered waters of low
ionic strength. For calculation purposes, it is
assumed that WLS-I lakes are represented by a
carbonate system; therefore, the ANC and BNC of
samples are in relation to equilibria of the carbonate
system (Butler, 1982; Kramer, 1982).
While the pH is monitored and recorded, samples are
titrated with standardized acid and base. Gran
analysis technique (Gran, 1952; Butler, 1982;
Kramer, 1982) is used to analyze the titration data for
the ANC and BNC determinations.
The Gran analysis technique defines the Gran
functions F, and F3 (see appendix) that are
calculated from sample volume, titrant (acid or base)
volume added, and constants. The Gran functions
are calculated for several data pairs (volume of titrant
added, resulting pH) from each titration. When the
Gran functions are plotted versus the volume of
titrant added, the linear portion of the curve can be
extrapolated to the end point. The titration results are
used to calculate ANC and BNC. The BNC titration
procedure is a subject of ongoing research; future
users of this method should review subsequent
literature before adopting the method as it stands.
The pH is determined before titrations are performed.
The same electrode that is used to measure the
initial pH is used for the pH determinations that are
made during the titration. (See NBS, 1982;
McQuaker et al., 1983; U.S. EPA, 1983.)
The air-equilibrated pH is determined similarly, after
the sample is equilibrated with 300 ppm CO2 in air for
20 minutes. Air equilibration is expected to normalize
19
-------
pH values by factoring out the day-to-day and
seasonal fluctuations associated with dissolved C02
concentrations.
The appendix provides examples of the ANC and
BNC calculations.
3.3.2 Determination of Ammonia
This method is used to determine ammonia over the
concentration range 0.01 to 2.6 mg/L NH4+ in
natural surface waters. This range is for photometric
measurements made at 630 to 660 nm in a 15-mm
or 50-mm tubular flow cell. Higher concentrations
can be determined by sample dilution.
Approximately 20 to 60 samples per hour can be
analyzed. Alkaline phenol and hypochlonte react with
ammonia to form an amount of indophenol blue that
is proportional to the ammonium concentration. The
blue color formed is intensified with sodium
nitroprusside (U.S. EPA, 1983).
3.3.3 Determination of Chloride, Nitrate, and
Sulfate by Ion Chromatography
This method is applicable to the determination of
chloride, nitrate, and sulfate in natural surface waters.
Samples are analyzed by ion Chromatography (1C). 1C
is a liquid chromatographic technique that combines
ion-exchange Chromatography, eluant suppression,
and conductimetnc detection.
A filtered sample portion is injected into an ion
chromatograph. The sample is pumped through a
precolumn, a separator column, a suppressor
column, and a conductivity detector. The precolumn
and the separator column are packed with a low-
capacity, anion-exchange resin. The sample anions
are separated into these two columns on the basis of
their affinity for the resin exchange sites.
The suppressor column reduces the conductance of
the eluant to a low level and converts the sample
anions to their acidic form. Typical reactions in the
suppressor column are:
Na* HC03"+ R - H
(high conductivity)
H2CO3 + R - Na
[low conductivity)
Na+ A" + R - H -» HA + R - Na
Three types of suppressor columns are available: the
packed-bed suppressor, the fiber suppressor, and
the micromembrane suppressor. The packed-bed
suppressor contains a high-capacity cation
exchange resin in the hydrogen form. The resin is
consumed during analysis and must be periodically
regenerated off-line. The fiber suppressors and the
micromembrane suppressors contain cation
exchange membranes. These suppressors are
regenerated continuously throughout the analysis.
Their dead volumes are substantially less than that of
a packed-bed suppressor. For these reasons, the
fiber suppressors and the micromembrane
suppressors are preferred.
A conductivity cell is used to measure the separated
anions in their acidic form. Anions are identified on
the basis of retention time. Quantification is
performed by comparing sample peak heights to a
calibration curve that is generated from known
standards (ASTM, 1984; O'Dell et al., 1984; Topol
and Ozdemir, 1982).
Operating conditions for the ion chromatograph will
depend upon the column and the system selected.
3.3.4 Determination of Dissolved Organic Carbon
and Dissolved Inorganic Carbon
This method is applicable to the determination of
DOC and DIG over the concentration range 0.1 to 30
mg/L DOC or DIG in natural waters. The method
detection limit is about 0.4 mg/L DOC and about 0.1
mg/L DIG, as determined from replicate analyses of a
blank sample. The method assumes that a
Dohrman-Xertex DC-80 Analyzer is used;
however, any instrumentation that has similar
operating characteristics may be substituted.
Two samples (aliquots 4 and 5) are sent to the
laboratory for DIG and DOC analysis. Aliquot 4 is
filtered and preserved in the field (acidified to pH < 2
with H2SO4). It is analyzed for DOC. Aliquot 5 is an
unfiltered sample. It is filtered in the laboratory when
it is analyzed for DIG.
DOC is determined (after the aliquot is subjected to
external sparging to remove DIG) by ultraviolet-
promoted persulfate oxidation followed by infrared
detection. DIG is determined directly by acidifying the
aliquot to generate C02; this process is followed by
infrared detection (U.S. EPA, 1983; Xertex-
Dohrman, 1984).
3.3.5 Determination of Total Dissolved Fluoride
by Ion-Selective Electrode
This method is applicable to the determination of total
dissolved fluoride in natural surface waters. It
requires a fluoride ion-selective electrode. The
applicable concentration range is 0.005 to 2 mg/L
fluoride.
20
-------
The total dissolved fluoride in a sample is determined
electrometrically. A fluoride ion-selective electrode
is used after a total ionic strength buffer solution
(TISAB) is added to the aliquot. The TISAB adjusts
sample ionic strength, adjusts pH, and breaks up
fluoride complexes.
The potential of the fluoride ion-selective electrode
varies logarithmically as a function of the fluoride
concentration. A calibration curve is prepared by
measuring the potential of known fluoride standards
(after TISAB addition), and by plotting the potential
versus the fluoride concentration (on a semi-log
scale). Sample concentrations are determined by
comparing the sample potential to the calibration
curve. This method is derived from existing methods
(Warner and Bressan, 1973; Barnard and Nordstrom,
1982; Kissa, 1983; U.S. EPA, 1983; LaZerte, 1984).
3.3.6 Determination of Total Phosphorus
This method can be used to determine
concentrations of total phosphorus over the range
0.001 to 0.200 mg/L phosphorus in natural surface
waters. This method should not be used to analyze
samples preserved with HgCI2.
All forms of phosphorus, including organic
phosphorus, are converted to orthophosphate by an
acid-persulfate digestion. Orthophosphate ions react
with ammonium molybdate in acidic solution to form
phosphomolybdic acid. Upon reduction with ascorbic
acid, this solution produces an intensely colored blue
complex. Antimony potassium tartrate is added to
increase the rate of reduction (Murphy and Riley,
1962; Gales et al., 1966; Skougstad et al., 1979).
3.3.7 Determination of Dissolved Silica
This method is applicable to the determination of
dissolved silica over the concentration range 0.1 to
10 mg/L silica in natural surface waters.
Silica reacts with molybdate reagent in acidic media
to form a yellow silicomolybdate complex. This
complex is reduced by ascorbic acid to form the
molybdate blue color. The silicomolybdate complex
may form as an alpha polymorph, as a beta
polymorph, or as a mixture of the two. Because the
absorbance maxima of the two polymorphic forms
are at different wavelengths, the pH of the mixture is
kept below 2.5. This condition favors the formation of
the beta polymorph (Mullen and Riley, 1955; Govett,
1961; Strickland, 1962).
A 1-hour digestion with 1.0 M NaOH is required to
ensure that all the silica is available for reaction with
the molybdate reagent.
The procedure specified utilizes automated
technology and is derived from existing methodology
(Skougstad et al., 1979).
3.3.8 Determination of Conductance
This method is applicable to natural surface waters of
low ionic strength. Most of the lakes sampled for
NSWS have conductance values in the range from
10 to 100 uS/cm.
The conductance in samples is measured with a
conductance meter and conductivity cell. Potassium
chloride standards of known conductance are used to
calibrate the meter and cell (U.S. EPA, 1983).
Samples are analyzed at 25°C.
3.3.9 Determination of Dissolved Metals (Al, Ca,
Fe, K, Mg, Mn, Na) by Atomic Absorption
Spectroscopy
Metals in solution may be determined readily by
atomic absorption spectroscopy. The method is rapid
and easy to perform. It is applicable to the
determination of Al, Ca, Fe, K, Mg, Mn, and Na in
natural surface waters.
In direct aspiration atomic absorption spectroscopy, a
sample is aspirated and atomized in a flame. A light
beam from a hollow cathode lamp that contains a
cathode made of the element to be determined is
directed through the flame, into a monochromator,
and onto a detector that measures the amount of
light absorbed. Absorption depends on the presence
of free, unexcited, ground-state atoms in the flame.
Because the wavelength of the light beam is
characteristic of only the metal being determined, the
light energy absorbed by the flame is a measure of
the concentration of that metal in the sample.
When the furnace technique is used in conjunction
with an atomic absorption spectrophotometer, a
representative aliquot of a sample is placed in the
graphite tube in the furnace, evaporated to dryness,
charred, and atomized. Because a larger percentage
of available analyte atoms is vaporized and
dissociated for absorption in the tube than in the
flame, the use of small sample volumes or the
detection of low concentrations of elements is
possible. The principle is essentially the same as that
for direct aspiration atomic absorption except that a
furnace, rather than a flame, is used to atomize the
sample. Radiation from a given excited element is
passed through the vapor that contains ground-state
21
-------
atoms of that element. The intensity of the
transmitted radiation decreases in proportion to the
amount of the ground-state element in the vapor.
The metal atoms to be measured are placed in the
beam of radiation by increasing the temperature of
the furnace, which causes the injected specimen to
be volatilized. A monochromator isolates the
characteristic radiation from the hollow cathode lamp,
and a photosensitive device measures the attenuated
transmitted radiation. Table 3-6 gives the
concentration ranges that should be obtained when a
satisfactory atomic absorption spectrophotometer is
used.
Dissolved metals (Ca, Fe, K, Mg, Mn, and Na) are
determined in a filtered sample (aliquot 1) by flame
atomic absorption spectroscopy (U.S. EPA, 1983).
Total Al is determined in an unfiltered sample (aliquot
7) by graphite furnace atomic absorption
spectroscopy (U.S. EPA, 1983) after acid digestion.
Extractable Al is determined by graphite furnace
atomic absorption spectroscopy (Barnes, 1975; May
et al., 1979; Driscoll, 1984) in a sample that has been
treated with 8-hydroxyquinoline and that has been
extracted into methyl isobutyl ketone (MIBK; aliquot
2). The MIBK procedure is a subject of ongoing
research; future users of this method should review
subsequent literature before adopting the method as
it stands.
This procedure assumes the following definitions for
major concepts:
Optimum Concentration RangeThis is a
range, defined by limits expressed in
concentration, below which scale expansion must
be used and above which curve correction
should be considered. This range will vary with
the sensitivity of the instrument and with the
operating conditions employed.
Sensitivity-Sensitivity is the concentration in
milligrams of metal per liter that produces an
absorption of 1 percent.
Dissolved Metals Dissolved metals are
defined for this study as those constituents
(metals) that can pass through a 0.45-um
membrane filter.
Total Metals-Following vigorous digestion, the
total concentration of metals is determined on an
unfiltered sample.
3.3.70 Determination of Dissolved Metals (Ca, Fe,
Mg, and Mn) by Inductively Coupled Plasma
Atomic Emission Spectroscopy
This method is applicable to the determination of
dissolved Ca, Fe, Mg, and Mn in natural surface
waters. The method employs a technique for
simultaneous or sequential determination of Ca, Fe,
Mg, and Mn in lake water samples collected for
NSWS. The method is based on the measurement of
atomic emission by optical spectroscopy. Samples
are nebulized to produce an aerosol. The aerosol is
transported by an argon carrier stream to an
inductively coupled argon plasma which is produced
by a radio-frequency (RF) generator. In the plasma,
which is at a temperature of 6,000 to 10,000°K, the
Tsble 3-6. Atomic Absorption Concentration Ranges3
Flame
Furnace*3'0
Metal
Aluminum
Calcium
Iron
Magnesium
Manganese
Potassium
Sodium
Detection Limit
(mg/L)
01
0.01
0.03
0.001
0.01
0.01
0.002
Sensitivity (mg/L)
1
0.08
0.12
0.007
0.05
0.04
0.015
Optimum
Concentration
Range (mg/L)
5 to 50
0.2 to 7
0.3 to 5
0.02 to 0.5
0.1 to 3
0 1 to 2
0.03 to 1
Detection Limit
(mg/L)
0.003
0.001
--
0.0002
-
--
Optimum
Concentration
Range (mg/L)
0.020 to 0.200
-
0.005 too 100
0.001 to 0 030
--
a The concentrations shown are obtainable with any satisfactory atomic absorption spectrophotometer.
b For furnace sensitivity values, consult msrument operating manual.
c The listed furnace values are those expected when a 20-nL injection and normal gas flow are used.
22
-------
analytes in the aerosol are atomized, ionized, and
excited. The excited ions and atoms emit light at their
characteristic wavelengths. The spectra from all
analytes are dispersed by a grating spectrometer,
and the intensities of the lines are monitored by
photomultiplier tubes. The photocurrents from the
photomultiplier tubes are processed by a computer
system. The signal is proportional to the analyte
concentration and is calibrated by analyzing a series
of standards (Fassel, 1982; U.S. EPA, 1983).
A background correction technique must be used to
compensate for the variable contribution that the
background intensity makes to the determination of
trace elements. Background intensity must be
measured adjacent to analyte lines during sample
analysis. The position selected for the background
intensity measurement, on either or both sides of the
analytical line, will be determined by the complexity
of the spectrum adjacent to the analyte line. The
position used must be free of spectral interference. It
also must reflect the same change in background
intensity that occurs at the analyte wavelength
measured. Generally, each instrument has different
background handling capabilities. The instrument
operating manual should be consulted for guidance.
3.4 References
American Public Health Association, American Water
Works Association, and Water Pollution Control
Federation, 1985. Standard Methods for the
Examination of Water and Wastewater, 16th Ed.
APHA, Washington, D.C.
American Society for Testing and Materials, 1984.
Annual Book of ASTM Standards, Vol. 11.01,
Standard Test Method for Anions in Water by Ion
Chromatography, D4327-84. ASTM, Phila-
delphia, Pennsylvania.
Barnard, W. R., and D. K. Nordstrom, 1982. Fluoride
in Precipitation I. Methodology with the Fluoride-
Selective Electrode. Atmos. Environ., v. 16, pp.
99-103.
Barnes, R. B., 1975. The Determination of Specific
Forms of Aluminum in Natural Water. Chem.
Geol., v. 15, pp. 177-191.
Butler, J. N., 1982. Carbon Dioxide Equilibria and
their Applications. Addison-Wesley Publications,
Reading, Massachusetts.
Driscoll, C. T., 1984. A Procedure for the
Fractionation of Aqueous Aluminum in Dilute
Acidic Waters. Int. J. Environ. Anal. Chem., v.16,
pp. 267-283.
Drouse, S. K., D. C. Hillman, L. W. Creelman, and S.
J. Simon, 1986. National Surface Water Survey,
Eastern Lake Survey (Phase l-Synoptic
Chemistry) Quality Assurance Plan. EPA 600/4-
86-008. U.S. Environmental Protection Agency,
Las Vegas, Nevada.
Fassel, V. A., 1982. Analytical Spectroscopy with
Inductively Coupled Plasmas - Present Status
and Future Prospects. In: Recent Advances in
Analytical Spectroscopy. Pergamon Press,
Oxford and New York.
Gales, M. E., Jr., E. C. Julian, and R. C. Kroner,
1966. Method for Quantitative Determination of
Total Phosphorus in Water. J. Am. Water Works
Assoc., v. 58, pp. 1363-1368.
Govett, G.J.S., 1961. Critical Factors in the
Colorimetric Determination of Silica. Anal. Chim.
Acta, v. 25, pp. 69-80.
Gran, G., 1952. Determination of the Equivalence
Point in Potentiometric Titrations. Part II. Analyst,
v. 77, pp. 661-671.
Hillman, D. C., J. F. Potter, and S. J. Simon, 1986.
National Surface Water Survey, Eastern Lake
Survey (Phase l-Synoptic Chemistry) Analytical
Methods Manual. EPA 600/4-86/009. U.S.
Environmental Protection Agency, Las Vegas,
Nevada.
Kissa, E., 1983. Determination of Fluoride at Low
Concentrations with the Ion-Selective Electrode.
Anal. Chem., v. 55, pp. 1445-1448.
Kramer, J. R., 1982. Alkalinity and Acidity. In: R. A.
Minear and L. H. Keith (eds.), Water Analysis,
Vol. 1. Inorganic Species. Part 1. Academic
Press, Orlando, Florida.
LaZerte, B. D., 1984. Forms of Aqueous Aluminum in
Acidified Catchments of Central Ontario: A
Methodological Analysis. Can. J. Fish Aquat. Sci.,
v. 41, n. 5, pp. 766-776.
May, H. M., P. A. Helmke, and M. L. Jackson, 1979.
Determination of Mononuclear Dissolved
Aluminum in Near-Neutral Waters. Chem. Geo!.,
v.24, pp. 259-269.
McQuaker, N. R., P. D. Kluckner, and D. K.
Sandberg, 1983. Chemical Analysis of Acid
23
-------
Precipitation: pH and Acidity Determinations.
Environ. Sci. Techno!., v. 17, n. 7, pp. 431-435.
Mullen, J. B., and J. P. Riley, 1955. The Colorimetric
Determination of Silica with Special Reference to
Sea and Natural Waters. Anal. Chim. Acta, v. 12,
pp. 162-176.
Murphy, J., and J. P. Riley, 1962. A Modified
Single-Solution Method for the Determination of
Phosphate in Natural Waters. Anal. Chim. Acta,
v.27, pp. 31-36.
National Bureau of Standards, 1982. Simulated
Precipitation Reference Materials, IV. NBSIR
82-2581. U.S. Department of Commerce -
NBS, Washington, D.C.
O'Dell, J. W., J. D. Raff, M. E. Gales, and G. D.
McKee, 1984. Technical Addition to Methods for
the Chemical Analysis of Water and Wastes,
Method 300.0, The Determination of Inorganic
Anions in Water by Ion Chromatography. EPA-
600/4-85/017. U.S. Environmental Protection
Agency, Cincinnati, Ohio.
Peden, M. E., 1981. Sampling, Analytical, and Quality
Assurance Protocols for the National Atmospheric
Deposition Program. Paper presented at October
1981 ASTM D-22 Symposium and Workshop on
Sampling and Analysis of Ram. ASTM,
Philadelphia, Pennsylvania.
Silverstein, M. E., S. K. Drouse, J. L. Engels, M. L
Faber, and T. E. Mitchell-Hall, 1987. National
Surface Water Survey, Western Lake Survey
(Phase l-Synoptic Chemistry) Quality Assurance
Plan. U.S. Environmental Protection Agency, Las
Vegas, Nevada.
Skougstad, M. W., M. J. Fishman, L. C. Friedman, D.
E. Erdman, and S. S. Duncan (eds.), 1979.
Method I-2700-78, Automated Molybdate Blue
Colorimetric Method for Dissolved Silica. In:
Methods for Determination of Inorganic
Substances in Water and Fluvial Sediments:
Techniques of Water-Resources Investigations
of the United States Geological Survey, Book 5,
Chapter A1. U.S. Government Printing Office,
Washington, D.C.
Strickland, J.D.H., 1962. The Preparation and
Properties of Silicomolybdic Acid; I. The
Properties of Alpha Silicomolybdic Acid. J. Am.
Chem. Soc., v. 74, pp. 852-857.
Topol, L. E., and S. Ozdemir, 1982. Quality
Assurance Handbook for Air Pollution
easurement Systems: Vol. V. Manual for
Precipitation Measurement Systems, Part II.
Operations and Maintenance Manual. EPA-
600/4-82/042b. U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina.
U.S. Environmental Protection Agency, 1983
(Revised). Methods for Chemical Analysis of
Water and Wastes. EPA-600/4-79/020. U.S.
Environmental Protection Agency, Cincinnati,
Ohio.
Warner, T. B., and D. J. Bressan, 1973. Direct
Measurement of Less Than 1 Part-Per-Billion
Fluoride in Rain, Fog, and Aerosols with an Ion
Selective Electrode. Anal. Chim. Acta, v. 63, pp.
165-173.
Weast, R. C. (ed.), 1972. CRC Handbook of
Chemistry and Physics, 53rd Ed. CRC Press,
Cleveland, Ohio.
Xertex-Dohrman Corporation, 1984. DC-80
Automated Laboratory Total Organic Carbon
Analyzer Systems Manual, 6th Ed. Xertex-
Dohrman, Santa Clara, California.
24
-------
Appendix
Calculation Procedures Required for ANC and BNC Determinations
1.0 HCI Standardization
1.1 Procedure
Step 1--Weigh about 1 g anhydrous Na2C03 to the
nearest 0.1 mg, dissolve it in water, then dilute the
solution to 1.000 L. Calculate the concentration as
follows:
Nx
wt.Na2C03g
X
106.00 g 1 mole 1L
mole 2 eq
NOTE: Fresh solution is prepared just before use.
Step 2--Calibrate the pH meter and electrode as
recommended by the manufacturer.
Step 3-Pipet 1.00 ml standard Na2CO3 plus
40.00 mL C02-free water into a clean, dry titration
vessel. Add a Teflon-coated stir bar, and stir the
solution at a medium speed (no visible vortex).
Step 4-lmmerse the pH electrode. Record the pH
when a stable reading is obtained.
Step 5-Add a known volume of the HCI titrant.
Record the pH when a stable reading is obtained.
Use the following information as a guide to the
volume of titrant that should be added in different pH
ranges.
Maximum Volume
Increment of HCI Titrant
PH
>7.5
4 to 7.5
<4
(mL)
0.2
0.1
0.2
F =(V +V)
lb s
vc
s
K
where
Fib = Gran function
Vs = initial sample volume = 41.00ml_
V = volume of HCI added (in mL)
C = NNa2co3/(2 X dilution factor)
[H + ] = 10-pH
KI = 4.4463 ^ 10'7
K2 = 4.6881 x 10-n
Kw = 1.0123 x 10-1"
Step 7-Plot F^ versus V. Using the points on the
linear portion of the plot, perform a linear regression
of Fib °n V to obtain the coefficients of the line F^
= a + bV.
The correlation coefficient should exceed 0.999. If it
does not, reexamine the plot to make sure that only
points on the linear portion are used in the linear
regression.
Step 8-Calculate the equivalence volume, V,, by
V, = -a/b.
Then calculate the HCI normality by
Continue the titration until the pH is less than 4.
Obtain at least seven data points in the range pH 4 to
7.
Step 6-Calculate F^ for each data pair (volume
acid added, pH) that has a pH between 4 and 7:
N
C°3
HCI
Step 9-Perform the titration and calculation two
more times. Calculate an average NHCI and standard
25
-------
deviation. The %RSD must be less than 2 percent. If
it is not, the entire standardization must be repeated
until the %RSD is less than 2 percent.
Step 10-Cross-check the concentration of every
new batch of HCI titrant. Use the procedure
described in Section 2.2 of this appendix.
Step 11--Store the titrant in a clean polyethylene
bottle. Although the HCI titrant is stable, it must be
restandardized monthly.
1.2 Example Calculations
A 1.00-mL portion of 0.01038 N Na^COg and 40.00
ml_ C02-free deionized water are titrated with HCI
titrant. The titration data are given below.
The resulting (V, Fib) values are given below.
ml HCI
Added
0.00
0.100
0.200
0.300
0.400
0.500
0.600
0.700
0.800
0.900
1.000
1.100
pH
10.23
9.83
9.70
9.54
9.28
8.65
7.20
6.71
6.37
6.03
5.59
4.91
ml_ HCI
Added
1.200
1.300
1.400
1.500
1.700
1.900
2.100
2.300
2.500
pH
4.48
4.26
4.11
4.00
3.84
3.72
3.63
3.56
3.49
V
Flb(x 10-3)
V
Ftb(x 10-3)
0.700
0.800
0.900
1.000
3.57
2.59
1.60
0.64
1.100
1.200
1.300
1.400
1.500
-0.34
-1.33
-2.28
-3.26
-4.23
The plot of FH, versus V is shown in Figure A-1.
The data lie on a straight line and are analyzed by
linear regression to obtain the coefficients of the line
Fib = a + bV.
-1. Plot of F1b versus V for HCI standardization.
Fib IS calculated for each data pair (V, pH) that has a
pH between 4 and 7:
F,. = (V + V)
lb s
V C
s
K
[H + ]
where
Vs =
V =
C =
[H + ] =
Ki =
K2 :
Kw :
initial sample volume = 41.00 ml
volume of HCI added (in ml)
1.266 X 10'4 = (N Na2CO3) / (2 X
10-pn
4.4463 X 10-7
4.6881 X 10-11
1.0123 X 10-14
The results of the regression are
r = 1.0000
a = 0.01038 ± 0.00001
b = -0.009747 ± 0.000012
Then, V^ = -a/b = 1.065 ml, and
41)
NHC1 =
a CO
a2 3
X V,
V.
= 0.009743 eq/L
(0.01038) (1.00)
1.065
26
-------
2.0 NaOH Standardization
Every batch of NaOH titrant is initially standardized
against KHP, and the standardization is cross-
checked against standardized HCI titrant (Section 2.2
of this appendix). Thereafter, the NaOH titrant is
restandardized daily against the HCI titrant (Section
2.3 of this appendix).
2.1 Initial NaOH Standardization
2.1.1 Procedure
Step 1--Weigh about 0.2 g KHP to the nearest
0.1 mg, dissolve it in water, then dilute the solution to
1.000 L. Calculate the normality of the solution as
follows:
N
KHP
wt. KHP g
204.22g
eq
X
1 L
Step 2-Calibrate the pH electrode and meter as
recommended by the manufacturer.
Purge the titration vessel with CO2-free nitrogen,
then pipet 5.00 ml_ standard KHP solution and 20.00
ml C02-free water into the vessel. Maintain a
CO2-free atmosphere above the sample throughout
the titration. Add a Teflon stir bar, and stir the sample
at a medium speed (no visible vortex).
Step 3-lmmerse the pH electrode and record the
reading when it stabilizes.
Step 4-Titrate with the 0.01 N NaOH. Use the
increments specified below.
Maximum Volume
Increment of NaOH
PH
<5
5 to 9
>9
Titrant (ml)
0.10
0.05
0.2
Between additions, record the volume and the pH
(when stable). Continue the titration until the pH is
greater than 10. Obtain at least four data points in the
pH range 5 to 7 and four data points in the pH range
7 to 10.
Step 5-Calculate F^ for each data pair (volume
added, pH) that has a pH between 5 and 10.
F3b =
V C
s
[H + ] -
K
where
Vs
V
[H
K2
Kw
= G ran function
= initial sample volume = 25.00 mL
= volume NaOH added (mL)
= NKHP corrected for initial dilution
= NKHp/5
= 10-pn
= 1.3 X 10-3
= 3.9 X 10-6
= 1.01 X 10-14
Step 6-Plot F3b versus V. Using the points on the
linear portion of the plot, perform a linear regression
of Fab on V to obtain the coefficients of the line Fab
= a + bV.
The correlation coefficient should exceed 0.999. If it
does not, examine the plot to ensure that only points
on the linear portion are used in the linear regression.
Step /--Calculate the equivalence volume, V3 by
V3 = -a/b.
Then calculate the NaOH normality by
N
KHP
V
KHP
NaOH
Vn
27
-------
Step 8--Perform the titration and calculation two
more times. Calculate an average NNaoH and
standard deviation. The %RSD must be less than 2
percent. If it is not, the entire standardization must be
repeated until the %RSD is less than 2 percent.
2.1.2 Example Calculations
A 5.00-mL portion of 9.793 x 1CH N KHP and 20.0
ml C02-free deionized water are titrated with
approximately 0.01 N NaOH. The titration data and
appropriate Gran function values are given below.
Volume
NaOH (ml)
0.000
0.050
0.100
0.150
0.200
0.250
0.300
0.350
0.400
0.450
0.500
0.550
0.600
0.700
0.900
1.100
1.300
PH
4.59
4.78
4.97
5.14
5.31
5.48
5.66
5.87
6.14
6.66
8.99
9.26
9.48
9.95
10.23
10.39
10.51
F3b(x 10-3)
3.90
3.39
2.86
2.34
1.82
1.29
0.79
0.26
-0.25
-0.77
-1.28
-2.29
-4.40
versus V is plotted in Figure A-2. The data lie
on a straight line with the equation F^ = a + bV.
Linear regression is used to calculate the
coefficients.
Figure A-2. Plot of F3b versus V for initial NaOH
standardization with KHP.
4-1
3-
2-
1-
o
"x
The Gran function F3b is calculated for each data pair
that has a pH between 5 and 10. F3h is calculated by
V C
(V + V) !
s (V + V)
-2-
-3-
-4-
-5 -1
01
03
07
09
where
2[H + ]
K
EH'
Vs = initial sample volume = 25.00 mL
V = volume NaOH added
C = NKHP/S = 1.9586 X 10-"
[H + ] = 10-pH
KI = 1.3 X 10-3
K2 = 3.9 X 10-e
Kw = 1.01 X 10-14
The results of the regression are
r = 1 .QOOO
a = 0.004931 ± 0.000008
b = -0.01036 ± 0.00002
From this, calculate V3 by V3 = -a/b = 0.4761 mL.
Then calculate the NaOH normality by
N
KHP X KHP
NaOH
= 0.01028 eq/L
28
-------
2.2 NaOH-HCI Standardization Cross-
check
2.2.7 Procedure
Step 1--Purge a titration vessel with C02-free
nitrogen, then pipet 0.500 ml_ 0.01 N NaOH and
25.00 mL CO2-free water into the vessel. Maintain a
CO2-free atmosphere above the sample. Add a
Teflon stir bar, and stir sample at a medium speed
(no visible vortex).
Step 2--lmmerse the pH electrode, and record the
reading when it stabilizes.
Step 3-Titrate with the standardized 0.01 N HCI.
Use the increments specified below. Between
additions, record the volume and pH (when stable).
Continue the titration until the pH is less than 3.5.
Obtain at least seven data points in the pH range 4 to
10.
PH
Maximum Volume
Increment of HCI Titrant
(mL)
10 to 4
<4
0.2
0.05
0.2
Step 4-Calculate F! for each data pair (V, pH)
that has a pH between between 4 and 10.
is calculated by
K
F, = (V + V)
where
Vs
V
K
[H
w
[H + ]
= Gran function
= initial sample volume = 25.5 mL
= volume of HCI added (mL)
= 1.01 X 10-'4
= 10-PH
Step 5--Plot FI versus V. Using the points on the
linear portion of the plot, perform a linear regression
of FI on V to obtain the coefficients of the line FI =
a + bV.
The correlation coefficient should exceed 0.999. If it
does not, reexamine the plot to ensure that only
points on the linear portion are used in the linear
regression.
Step 6-Calculate the equivalence volume (V,) by
V, = -a/b.
Then calculate the HCI normality (N'HCI) by
N'.
HCL
NaOH X NaOH
V.
where
VNaOH = 0.500
Step /--Calculate the absolute relative percent
difference (RPD) between N'HCI ar)d NHCI (normality
determined in Section 1.1 of this appendix) by
RPD =
N' N
HCL HCI
X 100
The absolute RPD must be less than 5 percent. If it
is not, then a problem exists in the acid
standardization, in the base standardization, or in
both standardizations (bad reagents, out-of-
calibration burets, etc.). The problem must be
identified, and both procedures must be repeated
until the RPD calculated above is less than 5
percent.
2.2.2 Example Calculations
A 0.500-mL portion of 0.00921 N NaOH and 25.0
mL CO2-free deionized water are titrated with
0.0101 N HCI (standardized with Na2CO3). The
titration data and appropriate Gran function values
are given below.
Volume HCL
(mL)
0.000
0.100
0.200
0.250
0.300
0.350
0.400
0.450
0.500
0.550
0.600
0.650
0.700
0.800
PH
10.29
10.15
10.03
9.91
9.78
9.60
9.34
8.39
4.76
4.44
4.26
4.12
4.04
3.88
F, (X 10-3)
2.75
2.09
1.55
1.03
0.57
0.064
-0.45
-0.94
-1.43
-1.98
-2.39
29
-------
The Gran function F! is determined for data in the
pH range 4 to 10. Fj is calculated by
K
Plot of Fi versus V for standardization cross-
check, titration of NaOH with HCL.
where
Vs
V
Kw
[H + ]
= initial sample volume = 25.5 ml
= volume of HCI added
= 1.01 X 10-14
= 10-pH
FI versus V is plotted in Figure A-3. The data are
on a straight line with the equation FI = a + bV. The
coefficients, determined by linear regression, are
r = 0.9994
a = 0.00465 ± 0.00005
b =-0.01016 ± 0.00011
From these values, Vi is calculated by
Vj = -a/b = 0.4577
and N'HCI is calculated by
NaOH X NaOH
NHC1= - - =0.01006
Comparing this value for N'HCI with the previously
determined value of NHCI. the absolute RPD is
RPD in Nu_, values =
rid
X 100 = 0.4%
0.01006 - 0.0101
0.5(0.0101 + 0.01006)
This RPD is acceptable because it is less than 5
percent.
2.3 Daily NaOH Standardization
2.3.7 Procedure
Step 1-Calibrate the pH meter and electrode as
recommended by the manufacturer.
Step 2-Purge the titration vessel with C02-free
nitrogen, then pipet 1.000 ml_ NaOH titrant plus
25.00 mL CO2-free water into the vessel. Maintain a
C02-free nitrogen atmosphere above the sample.
(Smaller volumes of NaOH may be used. A known
volume of CO2-free water should be added to bring
-3J
the solution to a convenient volume.) Add a Teflon
stir bar and stir at a medium speed (no visible
vortex).
Step 3-lmmerse the pH electrode, and record the
reading when it stabilizes.
Step 4--Titrate with the standardized HCI titrant.
Use the increments specified below.
PH
Maximum Volume
Increment of HCI Titrant
(mL)
4 to 10
<4
0.2
0.05
0.2
Between additions, record the volume and the pH
(when stable). Continue the titration until the pH is
less than 4. Obtain at least seven data points in the
pH range 4 to 10.
Step 5-Calculate FI for each data pair (volume
acid added, pH) that has a pH between 4 and 10 by
30
-------
F=(V
K
[H+]
where
Vs
V
Kw
[H + ]
Gran function
initial sample volume = 26.00 mL
volume of HCI added
1.01 X 10-14
10-pH
Step 6--Plot FI versus V. Using the points on the
linear portion of the plot, perform a linear regression
of FI on V to obtain the coefficients of the line FI =
a + bV.
The correlation coefficient should exceed 0.999. If it
does not, reexamine the plot to make sure that only
points on the linear portion are used in the linear
regression.
Step 7-Calculate the equivalence volume, V\, by
Vi = -a/b.
Then calculate the NaOH normality by
N
NHCLXV1
NaOH
V
NaOH
Step 8-Perform the tilration and calculation two
more times. Calculate an average NNa0H and
standard deviation. The %RSD must be less than 2
percent. If it is not, the entire standardization must be
repeated until the %RSD is less than 2 percent.
Because the NaOH titrant can deteriorate readily
through exposure to the air, every effort must be
made to prevent its exposure to the air at all times.
Furthermore, it must be standardized daily or before
every major work shift. Store in a linear polyethylene
or a Teflon container in a C02-free atmosphere
(e.g., under CO2-free air, nitrogen, or argon).
2.3.2 Example Calculations
A 1.000-mL portion of an approximately 0.01 N
NaOH solution and 25.00 mL of C02-free deionized
water are titrated with 0.009830 N HCI. The titration
data are given below.
mLHCI
Added
0.00
0.200
0.400
0.600
0.650
0.700
0.750
PH
10.44
10.30
10.13
9.71
9.51
9.19
5.35
mLHCI
Added
0.800
0.850
0.900
1.000
1.100
1.200
1.400
PH
4.65
4.37
4.22
4.02
3.88
3.78
3.62
FI is calculated for each data pair (V, pH) that has a
pH between 4 and 10 by
F =(V
K
where
V, =
[H
V =
Kw =
+ 1 =
initial sample volume = 25.00 +
26.00 mL
volume of HCI added
1.01 X 10-1"
10-pH
1.00
The new data pairs (V, F,) are tabulated below.
V
0.400
0.600
0.650
0.700
0.750
F,
(X 10-3)
3.56
1.36
0.86
0.41
-0.12
V
0.800
0.850
0.900
1.000
1.100
F,
(X 10-3)
-0.60
-1.14
-1.62
-2.58
-3.57
A plot of F, versus V is shown in Figure A-4, The
data sets that correspond to volumes from V =
0.400 to V = 1.100 lie on a straight line with the
equation F! = a + bV.
The coefficients, determined by linear regression, are
r = 0.9996
a = 0.007488 ± 0.00008
b = -0.0101 ± 0.0001
From these results, V! is calculated by V, = -a/b =
0.741.
31
-------
is calculated by
N
NHC1 X Vl
NaOH
NaOH
= 0.00728
(0.009830) (0.741)
1.000
Figure A-4. Plot of F, versus V for routine NaOH
standardization.
4 T
3 -
2-
1-
-1 -
-2-
-3-
-4
04
10 1.2
32
-------
3.0 Electrode Calibration
Separate electrodes must be used for the acid
titration and for the base titration. Each new electrode
pair must be evaluated for Nernstian response,
according to the procedure described in Section
3.1.1 of this appendix, before any samples are
analyzed. After the initial electrode evaluation, the
electrodes are calibrated daily, according to the
procedure given in Section 3.1.2 of this appendix.
3.1 Calibration Procedures
3. 1. 1 Rigorous Calibration Procedure
This procedure calibrates an electrode in terms of
hydrogen ion concentration. It also evaluates the
Nernstian response of the electrode and familiarizes
the analyst with the characteristic response time of
the electrode.
Step 1 --Following the manufacturer's instructions,
use pH 7 and pH 4 buffer solutions to calibrate the
electrode and meter that are used for acid titrations.
Use pH 7 and 10 buffer solutions to calibrate the
electrode that is used for base titrations.
time required for stabilization. Obtain at least seven
data points that have a pH less than 4.
Step 4-Prepare a fresh aliquot of water and 0.10
M KCI as in step 2.
Step 5-Under a CO2-free atmosphere, titrate the
blank with standardized 0.01 N NaOH. Use the
increments specified below.
pH
Maximum Volume
Increment of HCI Titrant
(mL)
0.10
0.20
Continue the titration until the pH is between 10.5
and 11. After each addition, record the pH. Obtain at
least 10 data points between pH 9 and 10.5.
Step 6--For each titration, calculate the pH for
each data point by
Step 2-Prepare a blank solution by pipetting 50.00
ml CO2-free water and 0.50 ml 0.10 M KCI into a
titration vessel. Add a Teflon stir bar, and use a
magnetic stirrer to stir the solution at a medium
speed (no visible vortex).
Step 3-Titrate the blank with standardized 0.01 N
HCI. Use the increments specified below.
PH
Maximum Volume
Increment of HCI Titrant
(mL)
>4
<4
0.050
0.3
Continue the titration until the pH is between 3.3 and
3.5. After each addition, record the pH, and note the
PH = -log[H + ]
For the acid titration, [H + ] is calculated by
VACA
where
VA = acid volume (in mL)
CA = HCI concentration in eq/L
Vs = sample volume = 50.5 mL
For the base titration, [H + ] is calculated by
K
VBCB
in eq/L
33
-------
where
Kw = 1.01 x 10~14
VB = base volume (in mL)
CB = NaOH concentration in eq/L
Vs = sample volume = 50.5 mL
Step 7--For each titration, plot the measured pH
versus the calculated pH (designated as pH*).
Perform a linear regression on each plot to obtain the
coefficients of the line pH = a + b(pH*).
The plots must be linear: b must be equal to 1.00 ±
0.05 and r must be greater than 0.999. Typically,
some nonlinearity exists in the pH range 6 to 8. The
nonlmearity most likely results from small errors in
titrant standardization, impure salt solutions, or
atmospheric COa contamination. The nonlinear points
should not be used in the linear regression. If the
plots are not linear and do not meet the
specifications given above, the electrode should be
considered suspect. Repeat the electrode
characterization. If the results still are unacceptable,
the electrode must be replaced.
Step 8-The plots for the two titrations should be
coincident. Combine the data from both titrations,
and perform a linear least-squares analysis on the
combined data to obtain new estimates for the
coefficients of pH = a + b(pH*).
The electrodes now are calibrated. Do not move any
controls on the meter.
If the two plots are not coincident (i.e., if the
coefficients a and b do not overlap), repeat the
characterization. If the plots still are not coincident,
the electrode must be replaced.
3.1.2 Daily Calibration Procedure
Generally the calibration curve prepared in Section
3.1.1 of this appendix is stable from day to day. The
daily calibration procedure is designed to verify the
rigorous calibration.
Step 1--Copiously rinse the electrode with water.
Immerse the electrode in 20 mL pH 7 buffer, and stir
the solution for 1 to 2 minutes. Discard the buffer,
and replace it with an additional 40 mL pH 7 buffer.
While the solution is stirred gently, measure the pH.
Adjust the pH meter calibration knob until the pH is
equal to the theoretical pH of the buffer. Record the
theoretical pH and the final, measured pH reading.
The two values should be identical.
Step 2--Copiously rinse the electrode with water.
Immerse it in 20 mL pH 4 QC sample, and stir the
sample for 1 to 2 minutes. Discard the sample, and
replace it with an additional 40 mL pH 4 QC sample.
While the solution is stirred, measure and record the
pH. From the calibration curve of pH versus pH*,
determine the pH* for the observed pH. Compare the
pH* to the theoretical pH of the QC sample. The two
values must agree within 0.05 pH unit. If the two
values do not agree, the rigorous calibration
procedure must be performed before any samples
are analyzed.
Step 3-Repeat step 2 with the pH 10 QC sample.
This sample must be kept under a CO2-free
atmosphere when in use to ensure that acceptable
results can be obtained.
3.2 Example Calculations
This section describes the electrode calibration
procedure. The tables below show the titration data
(V and pH), the calculated pH values (pH*), and the
coefficients for the line pH = a + b( pH*).
Acid Titration Data
V, = 50.50 mL
= 0.00983
r
a
b
Volume HCL
(mL)
0.000
0.025
0.050
0.100
0.150
0.200
0.250
0.300
0.350
0.400
0.450
0.500
0.600
0.800
1.000
1.200
1.500
1.700
2.000
= 1.00
= 0.10 ± 0.01
= 0.971 ± 0.002
pH
5.87
5.25
4.97
4.68
4.51
4.38
4.29
4.22
4.15
4.10
4.05
4.00
3.92
3.80
3.71
3.64
3.55
3.50
3.43
pH*
5.31
5.01
4.71
4.54
4.41
4.31
4.24
4.17
4.11
4.06
4.02
3.94
3.81
3.72
3.64
3.55
3.50
3.43
34
-------
V
r
a
b
Base
s = 40.4 ml_
Volume NaOH
(ml)
0.000
0.050
0.200
0.300
0.400
0.500
0.600
0.820
0.940
1.080
1.200
1.300
1.400
1.500
= 0.99
= 0.08 ± 0.27
= 0.99 ± 0.03
Titration Data
NNaoH = 0.00804
pH
6.66
9.03
9.55
9.66
9.75
9.90
10.00
10.18
10.25
10.31
10.36
10.40
10.43
10.47
pH"
9.00
9.60
9.77
9.90
9.99
10.07
10.20
10.26
10.32
10.37
10.40
10.43
10.46
The data given above are plotted in Figure A-5.
Except for two points in the base titration (at V =
0.300 and V = 0.400), the data lie on a straight line.
(The lines calculated for each titration are essentially
coincident as indicated by their coefficients.)
Excluding these two points, the data are fit to the line
with the equation pH = a + b (pH*). The
coefficients of the line, obtained by linear regression,
are
r = 1.0000
a = -0.014 ± 0.0011
b = 0.999 ± 0.002
Figure A-5. Plot of pH versus pH* for electrode calibration.
10 1 1
35
-------
-------
4.0 Comparison of Titration Values
4.1 Comparison of Initial Titration pH
Values
[ANC]c(neq/D
For the acid and base titrations, the values for
measured pH at V,ltrant = 0 (before KCI spike)
should agree within 0.1 pH unit. If they do not, check
the electrode operation to ensure that cross-
contamination is not occurring.
For a sample that has an ANC less than or equal to
-15 jaeq/L, calculate a value for alkalinity as follows:
[ANC] co = 106 x 10-pH* (PH at V = 0)
(The pH at Vtltrant = 0 is taken from the acid
titration.) If ANC differs from [ANClco by more than
±10 yieq.L, check the electrode operation and
calibration.
4.2 Comparison of Calculated ANC and
Measured ANC
4.2.1 Procedure
A value for ANC can be calculated from the DIG
concentration and the pH for a sample. Two sets of
pH and DIG values are obtained in the laboratory: (1)
the pH* at V = 0 of the base titration and the
associated DIG concentration and (2) the pH of the
air-equilibrated sample and the associated DIG
concentration. Each set can be used to calculate a
value for ANC. The calculated ANC values then can
be compared to the measured ANC value. The
comparison can be used to check the validity of the
carbonate system assumption and to check for
analytical error. ANC is calculated from pH and DIG
as follows:
[ANC] C1 = ANC calculated from the initial pH and
DIG at the time of base titration
[ANC]
C2
= ANC calculated from the air-
equilibrated pH and DIG
PIC
12,011
K
X 10
where
DIC = DIG in mg/L (the factor 12,011 converts
mg/L to moles/L)
[H + ] = 10-pH
K! = 4.4463 x lO-?at25°C
K2 = 4.6881 x 10-11 at 25°C
Kw = 1.0023 x 10-1-* at 25°C
[ANCci and [ANC]c2 are compared as follows:
For [ANClci less than or equal to 100 ueq/L, the
following condition applies:
[ANC]C1 - [ANC]C2
jieq/L
For [ANCJci greater than 100 peq/L, the following
condition applies:
[ANC]C1 - [ANC]C2
[ANC]C2)/2
X 100
If either of these conditions is not satisfied, the pH
and DIC values are suspect and must be
remeasured. It is very important that the pH and DIC
be measured as closely together in time as possible.
If they are not measured closely in time, acceptable
agreement between [ANClci and [ANC]c2 might not
be obtained. When acceptable values for [ANClci
and [ANC]c2 are obtained, their average is
compared to the measured ANC as described below.
37
-------
For [ANCJc-avg 'ess than or equal to 100 peq/L, the
difference (D) and the acceptance window (w) are
D = [ANC]c_av - ANC and w = 15 peq/L
For [ANClc-ave greater than 100 peq/L, D and w are
D =
[ANC],
X 100 and w = 10%
'C - avg
If the absolute value of D is less than or equal to w, it
is valid to assume that the system is a carbonate
system. If D is less than -w, the assumption of a
pure carbonate system is not valid, and the sample
contains noncarbonate protolytes (soluble reacting
species) such as organic species. If D is greater than
w, an analytical problem exists in the pH
determination, the DIG determination, or the acid
titration (such as titrant concentration). In this case,
the problem must be identified, and the sample must
be reanalyzed.
4.2.2 Example Calculations
For the sample analysis described in Section 6.0 of
this appendix, the following data were obtained:
initial pH
DIG
air-equilibrated pH
air-equilibrated DIG
5.09
0.59 mg/L
5.06
0.36
From these data, the calculated ANC values are
computed as follows:
[ANC]c (peq/L)
DIC
12,011
K.
X
The results are
[ANC]Ci = -5.6 peq/L
and [ANC]C2 = -7.2 peq/L
Then
|[ANClci - [ANC]C2l = 1.6 peq/L < 15 peq/L
Because [ANClci and [ANC]c2 are in agreement,
their average value is compared to the measured
value as follows:
[ANC]C-avg = -6.4 peq/L and ANC = -1.8 peq/L
Then
D = |[ANC]C-avg-ANC| = 4.6 peq/L < 15 peq/L
The calculated and measured ANC values agree,
which supports the assumption that the system is a
carbonate system.
4.3 Comparison of Calculated BNC and
Measured BNC
4.3.7 Procedure
Like the determination for ANC, a BNC value can be
calculated by using pH and DIC values. Because the
BNC of a sample changes with changing DIC, only
the initial pH and DIC values measured at the
beginning of the base titration are used to calculate a
BNC value. This calculated BNC then is compared
to the measured BNC value. BNC is calculated by
[BNC]C (peq/L)
DIC
12,011
K
X
[BNClc is compared to BNC as follows:
For [BNClc less than or equal to 100 peq/L
D = [BNC]C - BNC and w = 10-peq/L
For [BNClc greater than 100 peq/L
D =
[BNCL - BNC
O
[BNCL
X 100 and w = 10%
If the absolute value of D is less than or equal to w, it
is valid to assume that the system is a carbonate
system. If D is less than -w, the assumption of a
pure carbonate system is not valid, and the sample
38
-------
contains noncarbonate protolytes such as organic
species.
If D is greater than w, then an analytical problem
exists in the pH determination, the DIG
determination, or the base titration (such as titrant
concentration). In this case, the problem must be
identified, and the sample must be reanalyzed.
4.3.2 Example Calculations
For the sample analysis presented in Section 6.0 of
this appendix, the following data were obtained.
initial pH = 5.09
DIG = 0.59 mg/L
BNC = 69.0 ueq/L
From these data, the BNC is computed as follows:
[BNC]C (ueq/L)
DIG
12,01
K
[H + ]
X 10
The result is [BNC]C = 54.7 ueq/L.
This value is compared to the measured value as
follows:
D = [BNC]C - BNC = -14.7 ueq/L < -10ueq/L
Although this value for D is borderline, it indicates
that other protolytes in the system are contributing to
the measured BNC. This situation might be expected
because the sample also contains 3.2 mg/L DOC.
4.4 Comparison of Calculated Total
Carbonate and Measured Total
Carbonate
4.4.1 Procedure
If the assumption of a carbonate system is valid, the
sum of ANC plus BNC is equal to the total
carbonate. This assumption can be checked by
calculating the total carbonate from the DIG, then by
comparing the calculated total carbonate to the
measured estimate of total carbonate (the sum of
ANC plus BNC). The total carbonate is calculated by
Cc (umole/L) = DIG (mg/L) X 83.26 (umole/mg)
For GC less than or equal to 100 umole/L, Cc is
compared to (ANC + BNC) as follows:
D = Cc - (ANC + BNC) and w = 10 umole/L
For GC greater than 100 umole/L, Cc is compared to
(ANC + BNC) as follows:
D =
Cc - (ANC + BNC)
X 1 00 and w = 1 0%
If the absolute value of D is less than or equal to w,
the assumption of a carbonate system is valid. If D is
less than -w, the assumption is not valid, and the
sample contains noncarbonate protolytes. If D is
greater than w, an analytical problem exists. It must
be identified, and the sample must be reanalyzed.
4.4.2 Example Calculations
For the sample analyzed in Section 6.0 of this
appendix, the following data were obtained.
initial pH = 5.09
DIG = 0.59 mg/L
BNC = 69.5 ueq/L = 69.5 umole/L
ANC = -1.8 ueq/L = -1.8 umole/L
From the DIG value, the total carbonate is calculated
as follows:
Cc = 83.26 X DIC = 49.1 umole/L
This calculated value then is compared to the
measured value as follows:
D = Cc - (ANC + BNC)
= -18.6umoIe/L < -10 umole/L
Although this value for D is borderline, it indicates
that other protolytes are present m the system. This
situation might be expected because the sample also
contains 3.2 mg/L DOC. Note that the same
conclusion was reached for the BNC comparison.
In general, noncarbonate protolytes are significant
(i.e., contribute significantly to the total protolyte
concentration) when they are indicated by one (or
both) of the individual comparisons (the ANC and
BNC comparisons) and by the total carbonate
comparison.
39
-------
-------
5.0 Blank Analysis - ANC Determination
Determine ANC in one blank per batch. The ab-
solute value of the ANC must be less than or equal
to 10 ueq/L If it is not, contamination is indicated.
Determine and eliminate the contamination source
(often the source will be the water or the KC1) prior
to continuation of sample analysis.
5.1 Initial Calculations
5.1.1 Procedure
Step 1From the calibration curve of measured
pH versus calculated pH (pH*), determine pH* for
each pH value obtained during the acid titration and
during the base titration. Next, convert all pH* values
to hydrogen ion concentrations by [H + ] = 10~PH*.
Step 2-Using the acid titration data, calculate the
Gran function Fia for each data pair (Va, pH*) for
which pH* is less than 4:
where
Vs = total initial sample volume (40.00 + 0.400)
mL
Va = cumulative volume of acid titrant added
Step 3-Plot Fia versus Va. The data should be
on a straight line with the equation Fia = a + bV.
Step 4-Perform a linear regression of Fia on Va
to determine the correlation coefficient (r) and the
coefficients a and b. The coefficient r should exceed
0.999. If it does not, examine the data to ensure that
only data on the linear portion of the plot were used
in the regression. If any outliers are detected, repeat
the regression analysis. Calculate an initial estimate
of the equivalence volume (Vi) by Vi = -a/b.
Further calculations are based on this initial estimate
of Vi and on the initial sample pH*. Table A-1 lists
the appropriate calculation procedure for the different
combinations of Vi and initial sample pH*. Several
equations and constants are used frequently
throughout the calculations. These are listed in Table
A-2.
Table A-1. Calculation Procedures for Combinations of
Initial V-) and pH*
Sample Description
Initial Vi
<0
>0
>0
Initial pH*
--
<76
>7 6
Calculation
Procedure
A
B
C
Appendix
Section No.
5.2
5.3
5.4
NOTE: For blank analyses, calculate ANC by ANC = V, Ca/Vsa
Further calculations are not necessary.
5.7.2 Example Calculations
The blank is prepared by adding 0.40 ml_ 0.10 M
NaCI to 40.00 mL deionized water. It is titrated with
0.00983 N HCI. The titration data, the measured pH,
and the calculated pH* values are given below.
Volume HCI
(mL)
0.000
0.080
0.120
0.200
0300
0.400
0.500
0.600
0700
1.000
1.200
1.500
pH
5.84
4.69
4.52
4.31
4.14
4.01
3.91
3.84
3.77
3.62
3.55
3.45
pH"
5.85
4.70
4.53
4.32
4.14
402
3.91
384
3.77
3.62
3.55
345
FI
0.00194
"0.00295
0.00390
0.00503
0.00593
0.00698
0 00993
0.0117
0.0149
The Gran function Fu (Fia = (V8 + V) [H + ]) is
calculated for pH* values less than 4.5 and for the
values given above.
FI versus V is plotted in Figure A-6. The data are
linear and fit the line Fj = a + bV.
41
-------
Table A-2. Frequently Used Equations and Constants
Equation
1
F. = (V + V)
lc s +2 +
C([H 1K-+2KK
1K*> Kv
[H
[H
Equation
2
F = (V + V)
2c s I + 2 , + ,
C([H
[H ]" + [H
+ [rT] -
[HT ]
Vs = total initial sample volume
Constants0 V = cumulative volume of titrant added
and C = total carbonate, expressed in moles/per liter
variable [H + J _ hydrogen ion concentration
descriptions Kt = 4.4463 X 107at25°C
K2 = 4.6881 X 10 u at 25°C
Kw = 1.0023 X 10"u at 25°C
aConstants are taken from Butler, J N, 1982. Carbon Dioxide Equilibria and Their Applications. Addison-Wesley
Publications, Reading, Massachusetts.
Figure A-6. Plot of F1a versus V for alkalinity determination
of blank.
14-
12-
10-
o 8-
x
uT
6-
4-
2-
i i 1 1 1 1 1
02 04 06 0.8 10 12 14
The resulting coefficients are
r = 0.9998
a = (-0.70 ± 5.6) x 10~5
b = 0.00989 ± 0.00007
From this,
= -a/b = 7.05xlO~4mL
[ANC] =
V1CHC,
= 1.7 x 10~7 = 0.17 neq/L
Lj
This value for [ANC] is acceptable.
5.2 Calculation Procedure A (Initial V-|
Less Than 0)
Step 1--From the base titration data, determine
which data set (V, pH*) has the pH* nearest (but not
exceeding) a pH of 8.2. As an initial estimate, set the
equivalence volume \/2 equal to the volume for this
data set. Next, calculate initial estimates of ANC,
BNC, and C by
ANC =
V,C
1 a
Ca = concentration of acid titrant
Vsa = original sample volume (acid titration)
42
-------
BNC =
V2Cb
V
sb
Cb = concentration of base titrant
VSb = original sample volume (base titration)
C = total carbonate = ANC + BNC
Step 2-Use equation 1, Table A-2 to calculate
the Gran function FIC for the first seven to eight
points of the base titration. Plot FIC versus Vb.
Perform a linear regression with the points on the
linear portion of the plot. Determine the coefficients
of the line FIC = a + bV. The coefficient r should
exceed 0.999. If it does not, examine the plot to
ensure that only points on the linear portion are used.
From the coefficients, calculate a new estimate of Vi
by Vi = -a/b.
Step 3-Calculate the Gran function F2C (equation
2, Table A-2) for data from the base titration across
the current estimate of V?. (Use the first four to six
sets with a volume less than ₯2 and the first six to
eight sets with a volume greater than V?.) Plot F2C
versus Vb. The data should lie on a straight line with
the equation F2C = a + bV. Perform a linear re-
gression of F2c on Vb, and determine the coef-
ficients of the line. If r is less than 0.999, reexamine
the data to ensure that only points on the linear
portion were used in the regression. Calculate a new
estimate of V2 by V2 = -a/b.
Step 4-Use the new estimates of Vi and V2 (an
asterisk indicates a new value) to calculate new
estimates of ANC, BNC, and C.
ANC* =
BNC* =
sb
sb
C* = ANC + BNC
Step 5-Compare the latest two values for total
carbonate. If
C-C*
C + C*
>0.001
calculate a new estimate for C by C(new) = (C +
C*)/2.
Step 6--Using the new value for C, repeat the
calculations for the Gran functions and for new
values of C. Continue repeating the calculations until
the relative difference between C and C* is less than
0.001.
Step /--When the expression is less than 0.001,
convert the final values for ANC, BNC, and C to
peq/L by
ANC (peq/L) = ANC (eq/L) x 106
BNC (neq/L) = BNC (eq/L) x 106
C(neq/L) = C(eq/L)xl06
5.3 Calculation Procedure B (Initial Vi
Greater Than 0, Initial pH* Greater Than
or Equal to 7.6)
Step 1-From the base titration data, determine
which data set (V, pH*) has the pH* nearest, but not
exceeding 8.2. As an initial estimate, set the
equivalence volume V2 equal to the volume of this
data set. Next, calculate initial estimates of ANC,
BNC, and C by
ANC =
V,C
1 a
V
BNC =
V
sb
C = ANC + BNC
Step 2-Calculate the Gran function Fjc (equation
1) for data from the acid titration across the current
estimate of Vi. (Use the first four to six sets with
volumes less than Vi and the first six to eight sets
with volumes greater than Vi.) Plot FIC versus Va.
The data should lie on a line with the equation FIC =
a + bV. Perform a linear regression of FIC on Va,
and determine the coefficients of the line.
If r does not exceed 0.999, reexamine the data to
ensure that no outliers were used in the regression.
Calculate a new estimate for Vi by Vi = -a/b.
Step 3-Calculate the Gran function F2C (equation
2) for data from the base titration across the current
estimate of V2- (Use the first four to six sets with
volumes less than V2 and the first six to eight sets
with volumes greater than V2-) Plot F2C versus Vb-
The data should lie on a line with the equation F2C =
a + bV. Perform a linear regression of F2C on Vb and
determine the coefficients of the line. If r does not
43
-------
exceed 0.999, reexamine the data to ensure that only
data on the linear portion were included in the
regression. Calculate a new estimate for V2 by V2 =
-a/b.
Step 4-Use the latest estimates of Vi and V2 to
calculate new estimates of ANC, BNC, and C.
ANC* =
BNC* =
V,C
1 a
V
V2Cb
sb
C* = ANC + BNC
Step 5-Compare the latest two values for total
carbonate. If
equivalence volume \/2 equal to the volume of this
data set. Go to step 4.
Step 3--Using data sets from the acid titration with
pH* values across a pH equal to 7 (use four to six
sets with a pH* less than or equal to 7 and four to six
sets with a pH* greater than or equal to 7), calculate
the Gran function F2a by F2a = (Vt - Va) [H + ].
Step 4-Plot F2a versus Va. The data should lie on
a straight line with the equation F2a = a + bV.
Perform a linear regression of F2a on Va. The
coefficient r should exceed 0.999. If it does not,
reexamine the plot to ensure that only data on the
linear portion were used in the calculation. Calculate
a new estimate for V2 by V2 = -a/b.
ANC =
V.C
1 a
V
C -C*
C + C*
>0.001
calculate a new estimate of C by C(new) = (C +
C*)/2.
Step 6--Using the new value of C, repeat the
calculations for the Gran functions and for new
values of C (steps 2 through 5). Continue repeating
the calculations until the expression given above is
less than 0.001.
When the expression is less than 0.001, convert the
final values for ANC, BNC, and C to peq/L by
ANC (peq/L) = ANC (eq/L) X 10b
BNC (peq/L) = BNC (eq/L) X 106
C(neq/L) = C (eq/L) X 106
5.4 Calculation Procedure C (Initial Vi
Greater Than 0, Initial pH* Greater Than
7.6)
Step 1--Obtain an initial estimate of the
equivalence volume ₯2- Follow the procedure given
in step 2 if the initial sample pH* is greater than or
equal to 8.2. If the initial sample pH* is less than 8.2,
follow the procedure given in step 3.
Step 2-From the acid titration data, determine
which data set (V, pH*) has the pH* nearest to but
not exceeding 8.2. As an initial estimate, set the
BNC =
C = ANC + BNC
Step 5-Calculate estimates of ANC, BNC, and C
by
V.C
1 a
V
ANC =
BNC =
C = ANC + BNC
Step 6-Calculate the Gran function FIC (equation
1) for data from the acid titration across the current
estimate of Vi. (Use the first four to six sets with
volumes less than Vj and the first six to eight sets
with volumes greater than Vi.) Plot FIC versus Va.
The data should lie on a straight line with the
equation FIC = a + bV. Perform a linear regression
of FIC on Va and determine the coefficients of the
line. The coefficient r should exceed 0.999. If it does
not, reexamine the plot to ensure that only data on
the linear portion were included in the regression.
Calculate a new estimate for Vi by Vi = -a/b.
Step 7-Calculate the Gran function F2c (equation
2) for data sets from the acid titration with volumes
across the current estimate of V2. (Use the first four
44
-------
to six sets with volumes less than ₯2 and the first six
to eight sets with volumes greater than V2-) Plot F2C
versus Va. The data should lie on a straight line with
the equation F2C = a + bV. Perform a linear re-
gression of F2c on Va and determine the coefficients
of the line. The coefficient r should exceed 0.999. If it
does not, reexamine the plot to ensure that only data
on the linear portion were included in the regression.
Calculate a new estimate of ₯2 by ₯2 = -sub.
Step 9-Compare the latest two values for total
carbonate. If
C-C*
c + c*
>0.001
Step 8-Calculate new estimates of ANC,
and C using the latest estimates of V1 and V2.
BNC,
ANC* =
V.C
1 a
V
BNC* =
C * = ANC + BNC
calculate a new estimate of C by C(new) = (C +
C*)/2.
Step 10--Using this new value of C, repeat the
calculations in steps 6 through 9. Continue repeating
the calculations until the above expression is less
than 0.001.
Step 11--When the expression is less than 0.001,
convert the final values for ANC, BNC, and C to
ueq/L by
ANC (peq/L) = ANC (eq/L) x 106
BNC(peq/L) = BNC (eq/L) x 106
= C(eq/L)xl06
45
-------
6.0 Sample Analysis
An acid titration and a base titration are necessary to
determine the BNC and ANC of a sample. As part of
each titration, the sample pH is determined. The air-
equilibrated pH is determined in a separate sample
portion.
6.1 Procedure
6.1.1 Acid Titration
Step 1-Allow a sealed water sample (aliquot 5) to
reach ambient temperature.
Step 2--Copiously rinse the electrode with
deionized water, then immerse it in 10 to 20 mL
sample. Stir for 30 to 60 seconds.
Step 3--Pipet 40.00 ml_ sample into a clean, dry
titration flask. Add a clean Teflon stir bar and place
flask on a magnetic stirrer. Stir at a medium speed
(no visible vortex).
Step 4--lmmerse the pH electrode and read the
pH. Record the pH on Form 11 and on Form 13
when the reading stabilizes (1 to 2 minutes). This is
the initial measured pH at Vtltrant = 0.
Step 5--Add 0.40 ml 0.1 M KCI. Read and record
the pH on Form 13. This is the initial measured pH at
Vtitrant = 0 after addition of the KCI spike.
Step 6--Add increments of 0.01 N HCI as
specified below.
pH
>9
7.0 to 9.0
5.5 to 7.0
4.5 to 5.5
3.75 to 4.50
<3.75
Maximum Volume
Increment of HCI
Titrant (mL)
0.1
0.025
0.1
0.05
0.1
0.3
Record the volume of HCI added. Record the pH
when a stable reading is obtained. Adjust the volume
increment of titrant so that readings can be taken at
pH values of 4.5 and 4.2. Continue the titration until
the pH is between 3.3 and 3.5. Obtain at least six
data points that have a pH less than 4.
6.1.2 Base Titration
Step 1-Take a portion of aliquot 5 at this time for
DIG determination. If the DIG is not determined
immediately, the sample must be kept sealed from
the atmosphere. A simple way to do this is to
withdraw the sample for DIG by using a syringe
equipped with a syringe valve. By closing the valve,
the sample is sealed from the atmosphere. (Syringe
valves that fit standard Luer-Lok syringes are
available from most chromatography supply
companies.)
Step 2-Purge the titration vessel with C02-free
air, N2, or Ar.
Step 3--Copiously rinse the electrode with
deionized water, then immerse the electrode in 10 to
20 mL sample for 30 to 60 seconds.
Step 4-Pipet 40.00 mL sample into the CO2-free
titration vessel. Maintain a CO2-free atmosphere
above the sample. Do not bubble the N2 (or other
CO2-free gas) through the sample. Add a clean
Teflon stir bar and place the vessel on a magnetic
stirrer. Do not turn stirrer on at this point.
Step 5-lmmerse the pH electrode, read pH, and
record pH'on Form 11 and on Form 13 when pH
stabilizes. This is the initial measured pH at Vtltrant =
Step 6-Add 0.40 mL 0.10 M KCI. Stir for 10 to
15 seconds. Read pH, and record pH on Form 13.
Step 7-Add 0.025 mL of 0.01 N NaOH and begin
stirring (no visible vortex). Record the NaOH volume,
and record the pH when it stabilizes. Continue the
titration by adding increments of NaOH as specified
below until the pH is greater than 11. After each
addition, record the volume of NaOH and the
resulting pH. Obtain at least 10 data points in the pH
47
-------
range 9 to 10.5. If the initial sample pH is less than 7,
obtain at least five data points below pH 8.
6.2 Example Calculations
6.2.7 Titration Oaf a
A natural lake sample was titrated as described
above. The titration data are given below. Also
6.2.2 Initial Estimate of Vj
The Gran function F\a is calculated for each data
pair from the acid titration that has a pH* less than 4.
The values are given below.
F1a(X10-3)
F1a(X10-3)
included
are values
for the
calculated
pH (pH*).
Acid Titration
Vsa = 40.I
va
0.000
0.040
0.080
0.120
0.140
0.160
0.260
0.280
0.380
DO mL
3983 eq/L
PH
5.10
4.89
4.71
4.56
4.50
4.44
4.24
4.21
4.08
Vsalt =
pH*
5.11
4.90
4.72
4.57
4.51
4.44
4.24
4.21
4.08
0.40 mL
va
0.460
0.550
0.650
0.750
0.900
1.100
1.400
1.700
pH
399
3.91
384
3.77
3.69
3.61
3.50
3.42
pH*
3.99
3.91
384
3.77
3.69
3.61
3.50
342
0.460
0.550
0.650
0.750
where
Fla = (V
a+Vs)
4.18
5.04
5.93
6.99
[H + ]
0.900
1.100
1 400
1 700
8.43
10.2
13.2
160
Fla versus Va is plotted in Figure A-7. A regression
of Fla on Va is performed to fit the data to the line
Fja = a + bV.Tne resulting coefficients are:
r =
a =
0.9999
-0.000217 + 0.000050
b = 0.009548 ± 0.000048
Base Titration
Vsb = 40.00 mL
Cb = 0.00702
vb
0.00
0.015
0.030
0.050
0.080
0.120
0.160
0.200
0.240
0.280
0.320
0.340
0360
0.380
0.400
eq/L
PH
5.08
5.13
5.26
5.35
5.57
5.78
6.06
6.30
6.65
6.98
7.29
7.46
7.62
7.83
8.03
Vsalt =
pH"
5.09
5.14
5.27
5.36
5.58
5.79
6.07
6.31
6.66
7.00
7.31
7.48
7.64
7.85
8.05
0.40 mL
vb
0.425
0.470
0.500
0.540
0.560
0.600
0.660
0.700
0.780
0.900
1.000
1.100
1.405
1.700
2.200
2.500
pH
8.30
8.66
8.85
9.01
9.10
9.21
9.35
9.44
9.57
9.72
9.83
992
10.12
10.26
10.43
10.51
PH*
8.32
8.68
8.87
9.03
9.12
9.23
9.37
9.47
960
9.75
986
995
10.15
1029
1043
10.54
Figure A-7.
16-,
Plot of F1a versus Va for initial determination
of V,.
02 04 06 0.8 10 12 14 16 18
48
-------
From this, the initial estimate of Vi is calculated by
Vi = -a/b = 0.0227 mL
Because Vi is greater than 0 and the initial sample
pH* is less than or equal to 7.6, calculation procedure
B is used to determine the ANC and BNC of the
sample.
6.2.3 Initial Estimates of V2, ANC, BNC, and C
From the base titration data, V% is estimated to be
0.40 mL (the first point that has a pH* less than or
equal to 8.2). Now that initial estimates of Vi and V2
have been obtained, estimates of ANC, BNC, and C
can be calculated.
VC
ANC = = 5.6 X 10~6eq/L
V C
BNC = = 7.02 X 10~5eq/L
Vsb
C = ANC + BNC = 7.58 X 10~5eq/L
6.2.4 Refined Estimates of Vi and V2
The Gran function FIC is calculated for acid titration
data across the current estimate of Vi. The values
are given below.
F1c(x 10'4)
F1c(x 10'4)
0.000
0.040
0.080
0.120
0.140
Flc versus Va
of Fit, °n Va
-1.68
-4.10
-705
-10.4
-12.1
0.160
0260
0.280
0380
-14.4
-23.2
-24.9
-33.8
is plotted in Figure A-8. A regression
is performed. The regression results
are
r = 0.999
a = -0.00006 ± 0.00003
b = -0.00864 ± 0.00016
A new estimate of Vi is
Vi = -a/b = 0.007 mL
Next, the Gran function F2c is calculated from data
sets from the base titration across the current
estimate of V2. The values are given below.
vb
0.340
0.360
0.380
0.400
0.425
F2c(x 1Q-")
1 99
1.28
0555
-0031
-0868
vb
0470
0500
0.540
0560
0600
F1a(x 10'4)
-260
-4.14
-6.03
-7.43
-9.55
F2C versus V^, is plotted in Figure A-9. A regression
of F2c on Vb is performed. (Data for which Vb is
greater than 0.5 are not used in the regression.) The
regression results are
r = 0.999
a = -0.00135 ± 0.000024
b = -0.003400 ± 0.000060
A new estimate of V2 is
V2 = -a/b = 0.398 mL
Figure A-8. Plot of Fic versus Va for V, determination.
-004 0 0.04 012 0.20 028 0.36 0.44
-5-
-10-
-15
o
-
-25-
-30-
-35-
6.2.5 New Estimates of ANC, BNC, and C
From the new estimates of V} and V2, new estimates
of ANC, BNC, and C are calculated.
49
-------
Figure A-
3-
2-
1-
0-
1-
li ~3'
U-
4 _
-5-
-6-
-7-
-8-
Q _
-10-
9. Plot of F2C versus Vb for V2 determination. la c
030 \040
/ / 1 \l
7- / «.
Iteration V,(mL)
1 0.0227
2 -0.0060
3 -0.0067
4 -0.0071
5 -0 0072
6 -0.0074
7 -0.0074
8 -0.0074
9 -0.0074
vb
050
i
\
\
V2(mL)
0.400
0.398
0.397
0.397
0.396
0.396
0.396
0.396
0.396
rti>\_/ i.o j\ j.u eu/Li
v M
sa
v2cb 69gxio_5
060 070 V 6q
! 1 gh
C* = ANC + BNC = 6.83 x 10~5eq/L
6.2.6 Comparison of Latest Two Estimates of
Total Carbonate
C -C*
Of\ A 1 "*-v Cl f\ C\ 1
.U41 ->U.UU1
c + c*
Because C and C* do not agree, a new C is
calculated from their average as follows:
C(new) = (C + C*)/2 = 7.09 X 10~5eq/L
The calculations in Sections 6.2.4 through 6.2.6 are
repeated until successive iterations yield total
carbonate values that meet the criteria given above.
The results from each iteration (including those
already given) are presented below.
ANC BNC C C-C" NewC
(lieq/L) (neq/L) (iieq/L) C + C* (vieq/L)
5.6 70.2 75.8
-1.5 69.9 68.4 0.051 72.1
-1.6 69.7 681 0029 70.1
-1.7 69.7 679 0.016 69.0
-1.8 69.6 67.8 0.009 68.4
-1.8 69.5 67.7 0.005 68.1
-1.8 69.5 67.7 0.003 679
-1.8 69.5 67.7 0.001 678
-1.8 69.5 67.7 0.008
50
U-U.S. Government Printing Office : 1988 - Sib-002/80245
-------
o r\j si i_,
&1 w CD ,
H- O 0>5 CM
C H- .
ft? « O
OC - 3 M
o a
- t) 01 <
.CD - H>
O
01
O
O ri-
CD Ol h[j
CD hj H
c+ f O
« i r4"
h-1 CD
ft) 01 O
o et-
O tr"
B c
a
O
Northern
Rockies (4C)
Central
Rockies (40)
Pacific
Northwest (4B)
California (4A)
Southern
Rockies (4E)
Subregions of the Western Lake Survey - Phase I
------- |