EPA-600/4-79-019
   HANDBOOK FOR ANALYTICAL QUALITY CONTROL
     IN WATER AND WASTEWATER LABORATORIES
                    March 1979
ENVIRONMENTAL MONITORING AND SUPPORT LABORATORY
       U.S. ENVIRONMENTAL PROTECTION AGENCY
       OFFICE OF RESEARCH AND DEVELOPMENT
               CINCINNATI, OHIO 45268
                                    OCR(6/2001) by RYA-R10

-------
                                DISCLAIMER

The mention of trade names or commercial products in this handbook is for illustration
purposes and does not constitute endorsement or recommendation for use by the U.S.
Environmental Protection Agency.
                                                     OCR(6/2001) by RYA-R10

-------
                                  ABSTRACT

One of the  fundamental responsibilities of water and wastewater management is the
establishment of continuing programs to insure the reliability and validity of analytical
laboratory and field data gathered in water treatment and wastewater pollution control
activities.

This handbook is addressed to laboratory directors, leaders of field investigations, and other
personnel who bear responsibility for water and wastewater data.  Subject matter of the
handbook is concerned primarily with quality control (QC) for chemical and biological tests
and measurements. Chapters are also included on QC aspects of sampling, microbiology,
biology,  radiochemistry,  and safety as they relate to  water and wastewater pollution
control.  Sufficient information is offered to allow the reader to inaugurate or reinforce
programs of analytical QC that emphasize early recognition, prevention, and correction of
factors  leading to breakdowns in the validity of water and wastewater pollution control
data.
                                       iii               OCR(6/2001)byRYA-R10

-------
                           ACKNOWLEDGMENTS

This handbook was prepared by the Environmental Monitoring and Support Laboratory
(EMSL) of the United States Environmental Protection Agency. The contributions of the
following individuals in preparing the handbook are gratefully acknowledged:

       J. B. Anderson
       D. G. Ballinger
       E. L. Berg
       R. L. Booth
       R. H. Bordner
       P. W. Britton
       J. F. Kopp
       H. L. Krieger
       J. J. Lichtenberg
       L. B. Lobring
       J. E. Longbottom
       C. I. Weber
       J. A. Winter

Technical editing and preparation of the final manuscript were performed under contract to
John F. Holman & Co. Inc., 1346 Connecticut Avenue, N.W., Washington, D.C. 20036.

Inquiries regarding material contained in the handbook should be made to Environmental
Protection Agency, Environmental Research Center, Environmental Monitoring and Support
Laboratory, Cincinnati, Ohio 45268.
                                      iv              OCR(6/2001)byRYA-R10

-------
                       TABLE OF CONTENTS

                                                                   Page

          PREFACE 	iii
          ACKNOWLEDGMENTS	 v
Chapter

   1      IMPORTANCE OF QUALITY CONTROL	1-1
          1.1   General  	1-1
          1.2   Quality Assurance Programs  	1-1
          1.3   Analytical Methods	1-2
          1.4   Reference  	1-4

   2      LABORATORY SERVICES  	2-1
          2.1   General  	2-1
          2.2   Distilled Water	2-1
          2.3   Compressed Air  	2-5
          2.4   Vacuum	2-5
          2.5   Hood System	2-5
          2.6   Electrical Services  	2-6
          2.7   References	2-6

   3      INSTRUMENT SELECTION	3-1
          3.1   Introduction	3-1
          3.2   Analytical Balances  	3-1
          3.3   pH/Selective-lon Meters	3-3
          3.4   Conductivity Meters	3-6
          3.5   Turbidimeters (Nephelometers) 	3-7
          3.6   Spectrometers	3-8
          3.7   Organic Carbon Analyzers	3-13
          3.8   Gas Chromatographs	3-14
          3.9   References	3-14

   4      GLASSWARE	4-1
          4.1   General  	4-1
          4.2   Types of Glassware	4-2
          4.3   Volumetric Analyses  	4-3
          4.4   Federal Specifications for Volumetric Glassware  	4-4
          4.5   Cleaning of Glass and Porcelain	4-5
          4.6   Special Cleaning Requirements 	4-6
          4.7   Disposable Glassware	4-7
          4.8   Specialized Glassware  	4-7
          4.9   Fritted Ware 	4-8
          4.10  References	4-9

   5      REAGENTS, SOLVENTS, AND GASES  	5-1
          5.1   Introduction	5-1
          5.2   Reagent Quality  	5-1
                                                OCR(6/2001) by RYA-R10

-------
       5.3   Elimination of Determinate Errors	5-4
       5.4   References	5-4

6      QUALITY CONTROL FOR ANALYTICAL PERFORMANCE  ... 6-1
       6.1   Introduction	6-1
       6.2   The Industrial Approach to QC  	6-1
       6.3   Applying Control Charts in Environmental Laboratories  	6-2
       6.4   Recommended Laboratory Quality Assurance Program  	6-9
       6.5   Outline of a Comprehensive Quality Assurance Program  .... 6-10
       6.6   Related Topics  	6-13
       6.7   References	6-13

7      DATA HANDLING AND REPORTING	7-1
       7.1  Introduction  	7-1
       7.2   The Analytical Value 	7-1
       7.3   Glossary of Statistical Terms  	7-3
       7.4   Report Forms  	7-5
       7.5   References	7-11

8      SPECIAL REQUIREMENTS FOR TRACE ORGANIC ANALYSIS 8-1
       8.1   Introduction	8-1
       8.2   Sampling and Sample Handling	8-1
       8.3   Extract Handling	8-4
       8.4   Supplies and Reagents  	8-5
       8.5   Quality Assurance	8-7
       8.6   References	8-10

9      SKILLS AND TRAINING 	9-1
       9.1   General 	9-1
       9.2   Skills	9-2
       9.3   Training	9-4

10     WATER AND WASTEWATER SAMPLING	10-1
       10.1  Introduction	10-1
       10.2  Areas of Sampling  	10-2
       10.3  References	10-6

11     RADIOCHEMISTRY 	11-1
       11.1  Introduction	11-1
       11.2  Sample Collection 	11-1
       11.3  Laboratory Practices 	11-2
       11.4  Quality Control 	11-4
       11.5  References  	11-5

12     MICROBIOLOGY	12-1
       12.1  Background  	12-1
       12.2  Specific Needs in Microbiology  	12-1
       12.3  Intralaboratory Quality Control	12-2
       12.4  Interlaboratory Quality Control	12-2
       12.5  Development of a Formal Quality Assurance Program	12-3
       12.6  Documentation of a Quality Assurance Program  	12-3
                              vi              OCR(6/2001)byRYA-R10

-------
              12.7 Chain-of-Custody Procedures for Microbiological Samples . .  12-3
              12.8 References	12-10

       13     AQUATIC BIOLOGY	13-1
              13.1 Summary of General Guidelines	13-1
              13.2 Discussion	13-2
              13.3 Reference 	13-4

       14     LABORATORY SAFETY 	14-1
              14.1 Law and Authority for Safety and Health	14-1
              14.2 EPA Policy on Laboratory Safety	14-5
              14.3 Laboratory Safety Practices	14-7
              14.4 Report of Unsafe or Unhealthful Condition 	14-15
              14.5 References	14-15

Appendix A-Suggested Checklist for the Safety Evaluation of EPA Laboratory Areas A-l
                                      vii              OCR(6/2001)byRYA-R10

-------
                           LIST OF FIGURES

Figure No.                                                              Page

    4-1   Example of Markings on Glassware  	4-5

    6-1   Essentials of a Control Chart	6-2
    6-2   Shewhart Control Chart for Percent Recovery Data	6-5
    6-3   Procedure for Evaluating QC Data From a Monitoring Study 	6-12

    7-1   Example of Bench Sheet	7-8
    7-2   Example of Summary Data Sheet  	7-9
    7-3   Example of STORET Report Form	7-10
    7-4   Flow Chart of the Sequence of Events During a Controlled Series of
           Laboratory  Measurements	7-12

    12-1  Example of Chain-of-Custody Sample Tag (a) Front, (b) Back	12-5
    12-2  Example of a Sample Log Sheet	12-6
    12-3  Example of a Chain-of-Custody Record	12-7

    14-1  Health and Safety Inspection Checklist	14-8
    14-2  Report of Unhealthful or Unsafe Condition 	14-16
    14-3  Notice of Unhealthful or  Unsafe Condition 	14-17


                            LIST OF TABLES

Table No.                                                              Page

    2-1   Water Purity	2-1
    2-2   Requirements for Reagent Water  	2-2
    2-3   Comparison of Distillates From Glass and Metal Stills  	2-3

    3-1   Performance Characteristics of Typical pFI/Selective-Ion Meter	3-5
    3-2   Electrical Conductivity of Potassium Chloride Reference Solutions  ... 3-7

    4-1   Tolerances for Volumetric Glassware	4-4
    4-2   Fritted Ware Porosity  	4-8

    6-1   Analyses of Total Phosphate-Phosphorus Standards, in Milligrams Per
           Liter of Total PO4 -P  	6-4
    6-2   Estimates of the Range (R =  \A - B\) and the Industrial Statistic (I = \A -
           B | /(A + B)  of Three Different Parameters for Various Concentration
           Ranges  	 6-7
    6-3   Shewhart Upper Control Limits UCL and Critical Range Rc Values for
           the Differences Between Duplicate Analyses Within Specific Con-
           centration Ranges for Three Parameters 	6-8
    6-4   Critical Range Values for Varying Concentration Levels	6-9

    9-1   Skill-Time Rating of Standard Analytical Operations  	9-3
                                  viii              OCR(6/2001)byRYA-R10

-------
10-1   Guidance for Water/Wastewater Sampling	10-2

11-1   Sample Handling, Preservation, Methodology, and Major Instrumentation
       Required	11-2
                                               OCR(6/2001) by RYA-R10

-------

-------
                                    Chapter 1

                     IMPORTANCE OF QUALITY CONTROL
1.1 General

The analytical laboratory provides qualitative and quantitative data for use in decision-
making.   To be  valuable, the data must accurately describe the characteristics and
concentrations of constituents in the samples submitted to the laboratory. In many cases,
because they lead to faulty interpretations, approximate or incorrect results are worse than
no result at all.

Ambient water quality standards for pH, dissolved oxygen, heavy metals, and pesticides are
set to  establish satisfactory  conditions for drinking water,  fishing,  irrigation, power
generation, or other water uses. The laboratory data define whether conditions are being
met and whether the water can be used for its intended purposes. In wastewater analyses,
the laboratory data identify the characteristics of the treatment plant influent and the  final
load imposed upon receiving water resources, as well as the effectiveness of steps in the
treatment process. Decisions on process changes, plant modifications, or the construction
of new facilities may be based upon the results of water laboratory analyses.  The financial
implications of such decisions suggest that extreme care be taken in analysis.

Effective research in water pollution control also depends upon a valid laboratory data base,
which in turn may contribute to sound evaluations of both the progress of the research itself
and the viability of available water pollution-control alternatives.

The analytical data from water and wastewater laboratories may also be used to determine
the extent of compliance of a polluting industry with discharge or surface water standards.
If the laboratory results indicate a violation of a standard, remedial action is required by the
responsible parties.  Both legal and social pressures can be brought to bear to protect the
environment. The analyst should realize not only that he has considerable responsibility for
providing reliable laboratory descriptions of the samples at issue, but also that his profes-
sional competence, the validity of the procedures used, and the resulting values reported
may be challenged (perhaps in court). For the analyst to meet such challenges, he should
support the laboratory data with an adequate documentation program that provides valid
records of the control measures applied to all factors bearing on the final results of investi-
gations.

1.2 Quality Assurance Programs

Because of the importance of laboratory analyses in determining practical courses of action
that may be followed,  quality assurance programs to insure the reliability of the water and
wastewater data are essential. Although all analysts practice quality control (QC) in amounts
depending upon their  training, professional pride, and the importance  of their particular
projects, under  actual  working conditions sufficiently detailed QC may be neglected. An
established, routine, quality assurance program applied to each analytical test can relieve
analysts of the necessity of originating individual QC  efforts.
                                        1-1               OCR(6/2001)byRYA-R10

-------
Quality assurance programs have two primary functions  in the laboratory.  First, the
programs should continually monitor the reliability (accuracy and precision) of the results
reported; i.e., they should continually provide answers to the  question "How good (accurate
and precise) are the results obtained?" This function is the determination of quality. The
second function is the control of quality (to meet the program requirements for reliability).
As an example of the  distinction between the two functions, the processing of spiked
samples may be a determination of measurement quality, but the use of analytical grade
reagents is a control measure.

Each analytical method has a rigid protocol.  Similarly,  QC associated with a test must
include definite required steps for monitoring the test and insuring that its results are correct.
The steps in QC vary with the type of analysis. For example, in atitration, standardization
of the titrant on  a frequent basis is an element of QC.  In any instrumental method,
calibration and checking out of instrumental response are also QC functions.  All of the
experimental variables  that affect the  final results should be considered, evaluated, and
controlled.

In summary, laboratory data, in quantitative terms, e.g., in milligrams per liter, are reported
by the analyst.   These values are  interpreted by industrial plant  engineers to show
compliance ornoncompliance with permits for discharge, by state pollution control agencies
to define the need for additional sampling and analysis to confirm violations, or by EPA to
demonstrate that prescribed waste treatment was sufficient to protect the surface waters
affected by the discharge.

This handbook discusses the basic factors of water and wastewater measurements that
determine the value of analytical results and provides recommendations for the control of
these factors to insure that analytical results are the  best possible.  Quality assurance
programs initiated from, and based upon, these recommendations should increase confidence
in the reliability of the reported analytical results.

Because ultimately a laboratory director must assume full responsibility for the reliability
of the analytical results  submitted,  the laboratory director  must also  assume full
responsibility in both design and implementation for the corresponding  quality assurance
program.

1.3 Analytical Methods

Many analytical methods for common water pollutants have been in use for many years and
are used in most environmental laboratories.  Examples are  tests for chloride, nitrate, pH,
specific conductance, and dissolved oxygen.  Widespread use of an analytical method in
water and wastewater testing usually indicates that the method is reliable, and therefore
tends to support the validity of the reported test results. Conversely, the use of little-known
analytical techniques forces the water and wastewater data user to rely on the judgment of
the laboratory  analyst, who must then defend his choice of analytical technique as well as
his conclusions. Present Federal regulations, notably section 304(h) of Public Law 92-500
(Federal Water Pollution Control Amendments of 1977)  and the Interim Drinking Water
Regulations specifically require the use of EPA-approved methods of analysis.

Uniformity of methodology within a single  laboratory as well as among a group  of
cooperating laboratories is required to remove methodology as a variable when there are
many data  users.  Uniformity of methodology is particularly important when several
                                        1-2              OCR(6/2001)byRYA-R10

-------
laboratories provide data to a common databank (such as STORET*) or cooperate in joint
field surveys. A lack of uniformity of methodology may raise doubts as to the validity of
the reported  results.   If the  same constituents are measured  by different analytical
procedures within a single laboratory, or by a different procedure in different laboratories,
it  may be  asked which procedure is superior, why the  superior method is not used
throughout, and what effects the various methods and procedures have on the data values
and their interpretations.

Physical  and chemical measurement  methods used in water or wastewater laboratories
should be selected by the following criteria:

  a.    The selected methods should measure desired constituents of water samples in the
       presence of normal interferences with sufficient precision and accuracy to meet the
       water data needs.

  b.    The selected procedures should use equipment and skills ordinarily available in the
       average water pollution control laboratory or water supply laboratory.

 c.    The selected methods should be sufficiently tested to have established their validity.

   d.   The selected methods should be sufficiently rapid to permit repetitive routine use
       in the examination of large numbers of water samples.

The restriction to the use of EPA methods in all laboratories providing data to EPA permits
the combination of data from different EPA programs and supports the validity of decisions
made by EPA.

Regardless of which analytical  methods are used in a laboratory, the methodology should
be carefully documented.  In some reports it is stated that  a standard method from  an
authoritative reference (such as ref. 1) was used throughout an investigation, when close
examination has indicated, however, that this was not strictly true. Standard methods may
be modified or entirely replaced  because of recent advances in the state of the art or
personal  preferences of the  laboratory staff. Documentation of measurement procedures
used in arriving at laboratory data should be clear, honest, and adequately referenced; and
the procedures should be applied exactly as documented.

Reviewers can apply the associated precision and accuracy of each specific method when
interpreting the  laboratory  results.   If the  accuracy  and  precision  of the analytical
methodology are unknown or uncertain, the data user may have to establish the reliability
of the result he or she is interpreting before proceeding with the interpretation.

The necessarily strict  adherence to accepted methods in water and wastewater analyses
should not stifle investigations leading to improvements in analytical procedures. Even with
accepted and documented  procedures, occasions arise when the procedures must  be
modified; e.g., to eliminate unusual interferences, orto yield increased sensitivities. When
* STORET is the acronym used to identify the computer-oriented U.S. Environmental
Protection Agency water quality control information system; STORET stands for
STOrage and RETrieval of data and information.
                                       1-3              OCR(6/2001)byRYA-R10

-------
a modification of a procedure is necessary, it should be carefully formulated. Data should
then be assembled using both the original and the modified procedures to  show the
superiority of the latter.  Such results can be brought to the attention of the organizations
responsible for standardization of procedures.   To increase the benefit, the modified
procedures should be written in a standard format for routine use as applicable. The
standard format usually includes scope and application, principle, equipment, reagents,
procedure, calculation of results, and expected precision and accuracy.

Responsibility for the results obtained from use of a nonstandard procedure (i.e., one that
has not become accepted through wide use) rests with the analyst and his supervisor.

In field operations, because it may be difficult to transport samples to the laboratory, or to
examine large numbers of samples (e.g., for gross characteristics), the use of rapid field
methods yielding approximate answers is sometimes required. Such methods should be used
only with a clear understanding that the results obtained  ' are not as reliable as those
obtained from standard laboratory procedures.  The fact that such methods have been used
should be documented, and the results should not be reported in the same context with more
reliable analytical  information. When only approximate values  are  available, perhaps
obtained for screening purposes in the field only, the data user would then be so informed.

1.4 Reference

1.  Standard Methods for the Examination of Water and  Wastewater,  14th Edition,
    American Public Health Association, New York (1975).
                                       1-4              OCR(6/2001)byRYA-R10

-------
                                    Chapter 2

                           LABORATORY SERVICES
2.1 General
Quality control of water and wastewater laboratory analyses involves consideration and
control of the many variables that affect the production of reliable data. The quality of the
laboratory services available to the analyst must be included among these variables. An
abundant  supply of  distilled water, free from  interferences  and  other  undesirable
contaminants, is an absolute necessity. An adequate source of clean, dry, compressed air is
needed.  Electrical power for routine laboratory use and voltage-regulated sources for
delicate electronic instrumentation must be provided.  This chapter, therefore, will be
devoted to describing methods of maintaining the quality of these services, as used in water
and wastewater laboratory operations.

2.2 Distilled Water

Distilled or demineralized water is used in the laboratory for dilution, preparation of reagent
solutions, and final rinsing of glassware. Ordinary distilled water is usually not pure. It may
be contaminated by dissolved gases and by materials leached from the container in which
it has been stored. Volatile organics distilled over from the feed water may be present, and
nonvolatile impurities may occasionally be carried over by the steam, in the form of a spray.
The concentration of these contaminants is usually quite small, and distilled water is used
for many analyses without further purification. However, it is highly important that the still,
storage tank, and any associated piping be carefully selected, installed, and maintained in
such a way as to insure minimum contamination.

Water purity has been defined in many different ways, but one generally accepted definition
states that high-purity water is water that has been distilled or deionized, or both, so that it
will have a  specific resistance of 500,000 Q or greater (or  a conductivity less than 2.0
|j,mho/cm).  This  definition is  satisfactory as a base to work from, but for more critical
requirements, the breakdown shown in table 2-1 has been suggested to express degrees of
purity (1).


                                    Table 2-1
                                WATER PURITY
Degree of Purity
Pure
Very Pure
Ultrapure
Theoretically Pure
Maximum
Conductivity
(|j,mho/cm)
10
1
0.1
0.055
Approximate
Concentration
of Electrolyte
(mg/1)
2-5
0.2-0.5
0.01-0.02
0.00
                                       2-1
OCR(6/2001) by RYA-R10

-------
The American Society for Testing and Materials (ASTM) specifies four different grades of
water for use in methods  of chemical analysis and physical testing.   The method of
preparation of the various grades of reagent water determines the limits of impurities. The
various types of reagent water and ASTM requirements are listed in table 2-2.

Type I grade water is prepared by the distillation of feed water having a maximum
conductivity  of 20  |j,mho/cm at 25 °C followed by polishing with a  mixed bed of ion-
exchange materials and a 0.2-|j,m membrane filter.

Type 11 grade water is prepared by using a still designed to produce a distillate having a
conductivity  of less than 1.0 |j,mho/cm at 25°C.  This may be accomplished by double
distillation or the use of a still incorporating special baffling and degassing features.

Type III grade water is prepared by distillation, ion exchange, or reverse osmosis, followed
by polishing with the 0.45-|j,m membrane filter.

Type IV grade water is prepared by distillation, ion exchange, reverse osmosis, or
electrodialysis.

Properly designed metal stills from reputable manufacturers offer convenient and reliable
sources of distilled water. These stills are usually constructed of copper, brass, and bronze.
All surfaces that contact the distillate  should be heavily coated with  pure tin to prevent
metallic contamination.  The metal storage  tank should be of sturdy construction with a
tight-fitting cover, and have a filter in the air vent to remove airborne dust, gases, and fumes.

For special purposes, an all-glass distillation unit may be preferable to the metal still.  These
stills are usually smaller, and of more limited  capacity than the metal stills.  An  actual
comparison in which the distillates from an all-glass still and a metal still were analyzed
spectrographically for certain trace metal contaminants is given in table 2-3.  It can be seen
that the all-glass still produced a product that had substantially lower contamination from
zinc, copper,  and lead.

All stills require periodic cleaning to remove solids that have been deposited from the feed
water.  Hard water and high-dissolved-solids content promote  scale formation  in the


                                       Table 2-2
                      REQUIREMENTS  FOR REAGENT WATER
Grade
of Water



Type I
Type II
Type III
Type IV
Maximum
Total
Matter
(mg/1)

0.1
0.1
1.0
2.0
Maximum
Electrical
Conductivity
at 25 °C
(|j,mho/cm)
0.06
1.0
1.0
5.0
Minimum
Electrical
Resistivity*
at 25 °C
(MQ - cm)
16.67
1.0
1.0
0.2
pH at 25 °C Minimum Color
Retention Time
ofKMnO4
(min)

60
60
6.2-7.5 10
5.0-8.0 10
 * See reference  2.
                                       2-2
OCR(6/2001) by RYA-R10

-------
                                    Table 2-3
              COMPARISON OF DISTILLATES FROM GLASS AND
                                 METAL STILLS

             Source               Element and Concentration (jJ.g/1)

                          Zn    B    Fe   Mn    Al    Cu    Ni    Pb

         All-Glass Still    <1    12     1    <1    <5     5     <2    <2

         Metal Still         9     13     2    <1    <5    11    <2    26

evaporator, and cleaning frequency will thus depend on the quality of the feed water.  The
boiler of an all-glass still should be drained daily and refilled with clean water.  Buildup of
scale is easily detected, and the boiler and condenser coils should be cleaned  at frequent
intervals.  Metal  stills usually incorporate  a constant bleeder device that retards scale
formation to some extent. Nevertheless, these units should be dismantled and cleaned at
regular intervals.   Cleaning should always be in accordance with  the manufacturer's
instructions.

Pretreatment of the incoming feed water will often improve still performance and raise the
quality of the distillate. For example, preliminary softening of hard water removes calcium
and magnesium prior to distillation.  This reduces scale formation in the  boiler and
condenser, thereby reducing maintenance service. These softeners employ the ion-exchange
principle using a sodium chloride cycle, and are relatively inexpensive to operate. A carbon
filtration system, installed at the feed-water intake, will remove organic materials that might
subsequently  be carried over in the distillate.  If trace concentrations of ions are a major
concern, the distillate may be passed through a mixed-bed ion exchanger.

At least two commercially manufactured systems are available for production of high-purity
water by ion exchange. The Millipore Super-Q System (Millipore  Corp., Bedford, Mass.)
consists essentially of disposable cartridges for prefiltration, organic absorption, deioniza-
tion, and Millipore filtration. The company claims it can produce 10-MQ water, containing
no particulate matter larger than 0.45 |j,m in size, from tap water, at the rate of 20 gal/h.
Continental Water Conditioning Corp.,  El  Paso, Tex., advertises a system that can be
tailored to the needs of the customer. Performance specifications  include minimum flow
rates of 45 gal/h and total dissolved solids of less  than 0.1 mg/1 when required.

Specific conductance  is a rapid and simple measurement for determining the inorganic
quality of distilled water.  Stills of the types previously discussed are capable of producing
a distillate with a specific conductance of less than  2.0 |j,mho/cm at 25°C. This is equivalent
to  0.5 to 1.0 mg/1 of ionized material. Frequent checks should be made to determine that
optimum performance is being maintained. A purity meter installed between the still and
the storage reservoir will monitor the conductivity  of the distillate,   in terms  of the
equivalent in milligrams per liter of sodium chloride. If the reading on the meter begins to
rise above the present limit of conductivity, effective action should be taken to eliminate the
source of contamination.  Organic quality is more  difficult to monitor, but the total organic
carbon determination is a simple and rapid test of organic contaminants.
                                       2-3              OCR(6/2001)byRYA-R10

-------
A piping system for delivering distilled water to the area of use within the laboratory is a
convenient feature.  In this case, special care should be taken that the quality of the water
is not degraded between the still and the point of use. Piping may be of tin, tin-lined brass,
stainless steel, plastic, or chemically resistant glass, depending on the quality of the water
desired, its intended use, and on available funds. Tin is best, but is also very expensive. As
a compromise, plastic pipe or glass pipe with Teflon* O-rings at all connecting joints is
satisfactory for most purposes. The glass pipe has an obvious advantage when freedom from
trace amounts of organic materials is important.

When there is no piped-in supply, distilled water will probably be transported to the
laboratory and stored in polyethylene or glass bottles of about 5-gal capacity.  If stored in
glass  containers, distilled water will gradually leach the more  soluble materials from the
glass  and cause an increase in  dissolved solids. On the other hand, polyethylene bottles
contain organic plasticizers, and traces of these materials may be leached from the container
walls.  These are of little consequence, except in some organic analyses.  Rubber stoppers
often used in storage containers contain leachable materials, including significant quantities
of zinc.  This is usually no problem, because the water is not in direct contact with the
stopper. However, the analyst should be aware of the potential for contamination, especially
when the supply is not replenished by frequent use.

The delivery tube may consist of a piece of glass tubing that extends almost to the bottom
of the bottle, and that is bent downward above the bottle neck, with a 3- to 4-ft piece of
flexible tubing attached for mobility. Vinyl tubing is preferable to latex rubber, because it
is less leachable; however, a short piece of latex tubing may be required at the outlet for
better control of the pinchcock. The venttube in the stopper should be protected against the
entrance of dust.

Ordinary distilled water is quite adequate for many analyses, including the determination
of major cations and anions. Certain needs may require the use of double- or even triple-
distilled water. Redistillation from an alkaline permanganate solution can be used to obtain
a water with  low organic  background.  When  determining  trace organics by solvent
extraction and gas chromatography, distilled water with sufficiently low background may
be extremely difficult to obtain.  In this case, preextraction of the water with the solvent
used in the respective analysis may be helpful in eliminating undesirable peaks in the blank.
Certain analyses require special treatment or conditioning of the distilled water, and these
will now be discussed.

2.2.1  Ammonia-Free Water

Removal of ammonia can be accomplished by shaking ordinary distilled water with a strong
cation exchanger, or by passing distilled water through a column of such material.  For
limited volumes of ammonia-free water, use of the  Quikpure (Box 254, Chicago, 111.)
500-ml bottle is highly recommended. The ion-free water described in section 2.2.3 is also
suitable for use in the determination of ammonia.

2.2.2 Carbon-Dioxide-FreeWater

Carbon-dioxide-free water may be prepared by boiling distilled water for 15 min and cooling
to room temperature.  As an alternative, distilled water may be vigorously aerated with a
^Trademark of E. I. dupont de Nemours & Co.

                                       2-4               OCR(6/2001)byRYA-R10

-------
stream of inert gas for a period sufficient to achieve saturation and CO2 removal. Nitrogen
is most frequently used. The final pH of the water should lie between 6.2 and 7.2. It is not
advisable to store CO2-free water for extended periods.

2.2.3 lon-FreeWater

A multipurpose high purity water, free from trace amounts of the common ions, may be
conveniently prepared by slowly passing distilled water through an ion-exchange column
containing one part of a strongly acidic cation-exchange resin in the hydroxyl form.  Resins
of a quality suitable for analytical work must be used.  Ion-exchange cartridges of the
research grade, available from scientific supply houses, have been found satisfactory. By
using a fresh column and high-quality distilled water, a water corresponding to the ASTM
designation for type I reagent water (2) (maximum 0.1 mg/1 of total matter and maximum
conductivity of 0.06 mho/cm) can be obtained.  This water is suitable for use in the
determination of ammonia, trace metals, and low concentrations of most cations and anions.
It is not suited to some organic analyses, however, because this treatment adds organic
contaminants to the water by contact with the ion-exchange materials.

2.3 Compressed Air

The quality of compressed air required in the laboratory is usually very high, and special
attention should be given to producing and maintaining clean air until it reaches the outlet.
Oil, water, and dirt are undesirable contaminants in compressed  air, and it is important to
install equipment that generates dry, oil-free air. When pressures of iess than 50 psi are
required, a rotary-type compressor, using a water seal and no oil, eliminates any addition of
oil  that would subsequently have to be removed from the system.   Large, horizontal,
water-cooled compressors will usually be used when higher pressures are required.

Compression heats air, thus increasing its  tendency to retain moisture. An aftercooler is
therefore necessary to remove water. Absorption filters should be used at the compressor
to prevent moisture from entering the piping system. Galvanized steel pipe with threaded,
malleable-iron fittings, or solder-j oint copper tubing should be used for piping the air to the
laboratory.

When the compressed air entering the-laboratory is of low quality, an efficient filter should
be installed between the outlet and the point of use to trap oil, moisture, and other
contaminants.   As an alternative,  high-quality compressed air of  the dry grade is
commercially available in cylinders when no other source exists.

2.4 Vacuum

A source of vacuum in the chemical laboratory, while not an absolute necessity, can be a
most useful item. While used primarily as an aid in filtration, it is also  sometimes used in
pipetting and in speeding up the drying of pipets.

2.5 Hood System

An efficient hood system is a requirement for all laboratories.  In addition to removing the
various toxic and hazardous fumes that may be generated when using organic solvents, or
that may be formed during an  acid digestion step,  a hood system may also be used to
remove toxic gases  that may  be formed during  atomic absorption  analyses or other
                                       2-5              OCR(6/2001)byRYA-R10

-------
reactions. A regular fume hood should have a face velocity of 100 ft/min (linear) with the
sash fully open.

2.6 Electrical Services

An adequate electrical system is indispensable to the modern laboratory.  This involves
having both 115- and 230-V sources in sufficient capacity for the type of work that must be
done. Requirements for satisfactory lighting, proper functioning of sensitive instruments,
and operation of high-current devices must be considered. Any specialized equipment may
present unusual demands  on the electrical supply.

Because of the special type of work, requirements for a laboratory lighting system are quite
different from those in other areas. Accurate readings of glassware graduations, balance
verniers, and other measuring lines must be made. Titration endpoints, sometimes involving
subtle changes in color or shading, must be observed. Levels of illumination, brightness,
glare, and location of light sources should be controlled to facilitate ease in making these
measurements and to provide maximum comfort for the employees.

Such instruments as spectrophotometers, flame photometers, atomic absorption equipment,
emission spectrographs, and gas chromatographs have complicated electronic circuits that
require relatively constant voltage to maintain stable, drift-free instrument operation. If the
voltage of these circuits  varies,  there is  a resulting  change in resistance, temperature,
current, efficiency, light output, and component life. These characteristics are interrelated,
and one cannot be changed without affecting the others. Voltage regulation is therefore
necessary to eliminate these conditions.

Many instruments have built-in voltage regulators that perform this function satisfactorily.
In the absence of these, a small, portable, constant-voltage transformer should be placed in
the circuit between the electrical outlet and the instrument.  Such units are  available from
Sola Basic Industries, Elk Grove Village, 111., and are capable of supplying a constant output
of 118 V from an input that varies between 95  and 130V. When requirements are more
stringent, special transformer-regulated circuits can be used to supply constant voltage.
Only the instrument receiving the regulated voltage should be operated from such a circuit
at any given time.  These lines are in addition to,  and separate from, the ordinary circuits
used for operation  of equipment with less critical  requirements.

Electrical heating devices provide desirable heat sources, and should offer continuously
variable temperature control. Hot plates and muffle furnaces wired for 230-V current will
probably give better service than those that operate on 115 V, especially if the lower voltage
circuit is only marginally adequate.  Water baths and  laboratory ovens with maximum
operating temperatures of about 200°C perform well at 115 V. Care must be taken to ground
all equipment that could constitute a shock hazard. The three-pronged plugs that incorporate
grounds are best for this purpose.

2.7 References

1.  Applebaum, S. B., and Crits, G. J., "Producing High Purity Water," Industrial Water
    Engineering (Sept./Oct. 1964).
2.  "Water," Part 31  of  1977 Book of ASTM Standards, p.  20, American Society for
    Testing and Materials, Philadelphia (1977).
                                       2-6              OCR(6/2001)byRYA-R10

-------
                                   Chapter 3

                          INSTRUMENT SELECTION

3.1 Introduction

The  modern  analytical laboratory depends very heavily upon instrumentation.  This
statement may be completely obvious, but it should be remembered that the exceptional
emphasis on electronic equipment has really begun since the development of the transistor
and  the computer.   To a certain  extent, analytical  instrumentation  is always  in the
development  stage, with manufacturers continually  redesigning  and upgrading their
products, striving for miniaturization, better  durability and  sensitivity, and improved
automation.  For laboratory supervisors and staff members the net result is a bewildering
stream of advertising brochures, announcements, and catalogs of newly available equipment.
Consequently, the selection of analytical equipment is always difficult.

The instruments commonly used in water and wastewater analysis include the  following:

       Analytical balance
       pH/selective-ion meter
       Conductivity meter
       Turbidimeter
       Spectrometers (visible, ultraviolet (UV), infrared (IR), and atomic absorption (AA))
       Total carbon analyzer
       Gas chromatograph (GC)
       Gas chromatograph/mass spectrometer (GC/MS)
       Temperature devices (such as ovens and water baths)
       Recorders

These devices represent basic equipment used in routine work and should be the subject of
careful consideration before purchase. Further, their operation and maintenance ought to
be primary considerations in  sustained production of satisfactory  data.   Obviously,
fundamental understanding of instrument design will assist the analyst in the correct use of
instruments and in some cases will aid in detecting instrumental failures. Calibration of all
laboratory instruments with primary standards is encouraged whenever practical. This
normally involves aNational Bureau of Standards standard reference material or calibration
and certification procedures.  Calibration checks with secondary standards, made in each
laboratory or available  from private sources, are encouraged  on a frequent basis if not
required by the analytical method each time an analysis is made.

In the pages that follow an attempt is made to discuss basic instrument design and to offer
some remarks about desirable instrumental features.

3.2 Analytical Balances

The most important piece of equipment in any analytical laboratory is the analytical balance.
The  degree of accuracy of the balance is reflected in the accuracy of all data related to
weight-prepared standards. Although the balance should therefore be the most protected and
cared-for instrument in the laboratory, proper care of the balance is frequently overlooked.
                                       3-1               OCR(6/2001)byRYA-R10

-------
There are many fine balances on the market designed to meet a variety of needs. Types of
balances include top-loading, two-pan, microanalytical, electroanalytical,  semianalytical,
analytical, and other special-purpose instruments. Each type of balance has its own place
in the scheme of laboratory operation, but analytical single-pan balances are by far the most
important in the production of reliable data.

Single-pan analytical balances range in capacity from the 20-g to the popular 200-g models
with sensitivities from 0.01 to  1 mg.  Features of single-pan  balances may include
mechanical and electronic switching of weights, digital readout, automatic zeroing of the
empty balance, and automatic preweighing and taring capabilities. Even with all the design
improvements,  however,  modem analytical  balances  are  still fragile instruments, the
operation of which is subject to shock, temperature, and humidity changes, to mishandling,
and to various other insults.  Some of the precautions to be observed in maintaining and
prolonging the dependable life  of a balance are as follows:

 a.    Analytical balances should be mounted on a heavy, shockproof table, preferably one
       with an adequately large working surface and with a suitable drawer for storage of
       balance accessories. The balance level should be checked frequently and adjusted
       when necessary.

  b.    Balances should be located away from laboratory traffic and protected from sudden
       drafts and humidity changes.

  c.   Balance temperatures should be  equilibrated with room temperature; this is
       especially important if building heat is  shut off or reduced during nonworking
       hours.

  d.    When the balance is not in use, the beam should be raised from the  knife edges, the
       weights returned to the beam, objects such as the weighing dish removed from the
       pan, and the weighing compartment closed.

  e.    Special precautions should be taken to avoid spillage of corrosive chemicals on the
       pan or inside the balance case; the interior of the balance housing should be kept
       scrupulously clean.

  f.    Balances should be checked and adjusted periodically by a company service man
       or balance  consultant; if service is not  available locally, the  manufacturer's
       instructions should be followed as closely as possible.  Service contracts, including
       an automatic preventive maintenance schedule, are encouraged.

  g.   The balance should be  operated at all times according to the manufacturer's
       instructions.

Standardized weights to be used  in checking balance accuracy, traceable  to the National
Bureau of Standards, may be purchased from various supply houses.  A complete set of
directions for checking the performance of a balance is  contained in part 41 of ASTM
Standards (1).

Because all analytical balances of the 200-g capacity suitable for water and wastewater
laboratories  have about the same design specifications with reference to sensitivity,
precision, convenience, and price, it is safe to assume that there is no clear preference for
a certain model, and selection can be made on the basis of availability of service.
                                       3-2              OCR(6/2001)byRYA-R10

-------
3.3 pH/Selective-lon Meters

The concept of pH as a means of expressing the degree of effective acidity or alkalinity
instead of total acidity or alkalinity was developed in 1909 by Sorenson (2). It was not until
about 1940 that commercial instruments were developed for routine laboratory measurement
ofpH.

A basic meter consists of a voltage source, amplifier, and scale or digital readout device.
Certain additional refinements produce varying performance characteristics between models.
Some models incorporate expanded scales for increased readability, solid state circuitry for
operating stability and extreme accuracy, and temperature and slope  adjustment to correct
for asymmetric potential of glass electrodes. Other features are scales that facilitate use of
selective-ion  electrodes, recorder output,  and interfacing with complex data-handling
systems.

In routine pH measurements the glass electrode is used as the indicator and the calomel
electrode as the reference. Glass electrodes have a very fast response time in highly buffered
solutions. However, accurate readings are obtained slowly in poorly buffered samples, and
particularly so when changing from buffered to unbuffered samples. Electrodes, both glass
and calomel,  should be well  rinsed with distilled water after each reading, and should be
rinsed with, or dipped several times into, the next test sample before the final reading is
taken. Weakly buffered samples should be stirred during measurement. When not in use,
glass  electrodes should not  be allowed to become  dry, but should be immersed  in  an
appropriate solution consistent with the manufacturer's instructions.   The first steps in
calibrating an instrument are to immerse the glass and calomel electrodes into a buffer of
known pH, set the meter to the pH of the buffer, and adjust the proper controls to bring the
circuit into balance. The temperature-compensating  dial should be set at the temperature
of the buffer solution. For best accuracy, the instrument should be calibrated against two
buffers that bracket the expected pH of the samples.

The presence of a faulty electrode is indicated by failure to obtain  a reasonably correct
value for the pH  of the second reference buffer  solution after  the meter has been
standardized with the first reference buffer solution.  A cracked glass electrode will often
yield  pH readings that  are  essentially the same for both standards.  The response  of
electrodes may  also be impaired by  failure to maintain the  KC1  level in the calomel
electrode, by improper electrode  maintenance, or by certain materials  such  as oily
substances and precipitates that may coat the electrode surface. Faulty electrodes can often
be restored to normal by an appropriate cleaning procedure. Complete and detailed cleaning
methods are given in part 31  of ASTM Standards (3), and are also usually supplied by the
electrode manufacturer.

Because of the asymmetric potential of the glass electrode, most pH meters are built with
a slope adjustment that enables the analyst to correct for slight electrode errors observed
during calibration with two different pH buffers.  Exact details of slope adjustment and
slope check may vary with different models of instruments.  The slope adjustment must be
made whenever electrodes are changed, subj ected to vigorous cleaning, or refilled with fresh
electrolyte.   The slope adjustment feature is highly desirable  and  recommended for
consideration when purchasing a new meter.

Most pH meters now available are built with transistorized circuits rather than vacuum
tubes, which  greatly reduces the warmup time and increases the stability of the meters.
                                       3-3              OCR(6/2001)byRYA-R10

-------
Also, many instruments are designed with a switching circuit so that the entire conventional
0 to 14 scale of pH may be used to read a single pH unit with a corresponding increase in
accuracy.

This expanded-scale feature is of definite value when the meter is used for potentiometric
titrations and selective-ion work.  It is of dubious value, however, in routine analyses,
because pH readings more precise than ±0.1 are seldom required. Primary standard buffer
salts are available from the National Bureau of Standards* and should be used in situations
where extreme accuracy is necessary. Preparation of reference solutions from these salts
requires some special precautions and handling (3,4) such as the use of low-conductivity
dilution water, drying ovens, and carbon-dioxide-free purge gas.  These solutions should be
replaced at least once each month.

Secondary standard buffers may be prepared from NBS salts or purchased as solutions from
commercial vendors. Routine use of these commercially available solutions, which have
been validated by comparison to NBS standards, is recommended.

The electrometric measurement of pH varies with temperature because of two effects. The
first effect is the change in electrode output with temperature.  This interference can be
controlled by use of instruments having temperature compensation or by calibrating the
instrument system including the electrode at the temperature of the  samples. The second
effect is the change of pH of the sample with temperature. This  error is sample dependent
and cannot be controlled; it should therefore be noted by reporting both the  pH and
temperature at the time of analysis.

Typical characteristics of a conventional expanded-scale meter are shown in table 3-1.

3.3.1 pH  Electrodes

A wide variety of special- and general-purpose pH electrodes are now available to meet all
applications in the general analytical laboratory. A survey through any laboratory supply
catalog may confuse more than clarify the selection process.  A rugged, full-range, glass-
or plastic-bodied combination electrode is  a good choice for routine use.  An added
convenience is an electrode that contains solid geltype filling materials not requiring the
normal maintenance of an electrode containing liquid filling solutions.

3.3.2 Selective-Ion Electrodes

Electrodes have been developed to measure almost every common inorganic ion normally
measured  in the  water and wastewater laboratory.  Application of these electrodes has
progressed at a much slower pace and cuff ently only three are approved for EPA monitoring
applications.

Reference 5  includes  methods for use of fluoride, ammonia,  and dissolved oxygen
electrodes.  Various techniques for use of these and other electrodes are  reviewed in
references 6 through 9. A major problem in measuring the total  parameter with electrodes
is that of relating the ion activity to ion concentration. Because the electrodes only measure
*NBS, Office of Standard Reference Materials, Institute for Materials Research,
Washington, D.C. 20234.

                                       3-4              OCR(6/2001)byRYA-R10

-------
                                                  Table 3-1
                    PERFORMANCE CHARACTERISTICS OF TYPICAL pH/SELECTIVE-ION METERS
Smallest Power
Range Scale Accuracy Reproducibility Temperature Input Requirements
Scale Division Compensation1 Impedance
(°C) (0)
(pH) (mV) (pH) (MV) (pH) (mV) (pH) (mV)
Normal 0-14 ±1,400 0.1 10 ±0.05 ±5 ±0.02 ±2 0-100 =
Expanded 0-1 ±100 .005 .5 ±0.002 (3) ±0.002 ±0.2 0-100 =
(V) (Hz)
>10 2 115/220 50/60
>10 2 115/220 50/60
1 Manual or automatic.
2 May also be powered by self-contained batteries.
3 ±2 percent of reading.
                                                     3-5  OCR(6/2001)byRYA-R10

-------
activity, the challenge is to put all of the parameters of interest into the same measurable
ionic form and then to modify the activity to be proportional to the concentration.  The
technique of known addition (spiking of samples) is recommended when unproven electrode
methods are being used or when sample matrix problems are suspected or not controlled by
prior distillation or separation techniques.

3.4 Conductivity Meters

Solutions of electrolytes conduct an electric current by the migration of ions under the
influence of an electric field. For a constant applied EMF, the current flowing between
opposing electrodes immersed in the electrolyte will vary inversely with the resistance of
the solution.  The  reciprocal of the resistance is called conductance and is expressed in
reciprocal ohms (mhos). For natural water samples where the resistance is high, the usual
reporting unit is micromhos.

Most conductivity meters on the market today use a cathode-ray tube, commonly known as
the "magic eye," for  indicating  solution conductivity.  A stepping switch for varying
resistances in steps of 10 x facilitates reading conductivities from about 0.1 to about 250,000
|j,mho. The sensing element for a conductivity measurement is the conductivity cell, which
normally consists of two thin plates of platinized metal, rigidly supported with a very precise
parallel spacing. For protection, the plates are mounted inside a glass tube with openings
in the side walls and  submersible end for access of sample.  Variations in designs have
included use of hard rubber and plastics for protection of the cell plates.  Glass may be
preferable, in that the plates may be visually observed for cleanliness and possible damage,
but the more durable encasements have the advantage of greater protection and reduced cell
breakage. Selection of various cell designs is normally based on personal preference with
consideration of sample type and durability requirements.

In routine use, cells should be frequently examined to insure that (a) the platinized coating
of plates is intact; (b) plates  are not coated with suspended matter;  (c) plates are not bent,
distorted, or misalined; and (d) lead wires are properly  spaced.

Temperature  has  a pronounced  effect on the conductance  of solutions,  and must be
corrected for when results are reported. The specified temperature for reporting data used
by most analytical groups (and all EPA laboratories)  is 25°C.  Data correction may be
accomplished  by adjusting sample temperatures to 25°C,  or by use of mathematical or
electronic adjustment.

Instrumental troubles  are seldom encountered with conductivity meters because of the
design simplicity. When troubles occur, they are usually in the cell, and for most accurate
work the following procedures should be used:

  a.   Standardize the cell and establish a cell factor by measuring the conductivity of a
       standard potassium chloride solution (standard conductivity tables may be found in
       various handbooks).

  b.   Rinse the cell  by repeated immersion in distilled water.

  c.   Again, immerse the cell in the  sample  several times before  obtaining a reading.
                                       3-6               OCR(6/2001)byRYA-R10

-------
  d.   If the meter is equipped with a magic eye, determine the maximum width of the
       shadow at least twice, once by approaching the endpoint from a low reading
       upward, and once from a high reading downward.

Because the cell constants are subject to slow change even under ideal conditions, and
sometimes to more rapid change under adverse conditions, it is recommended that the cell
constant be periodically established. Table 3-2 can be used for this operation.

For instruments reading in mhos, the cell constant is calculated as follows:

                                     K, + K,
                                 L =
                                     10" K
where
       L =    cell constant

       Kj =   conductivity, in micromhos per centimeter, of the KC1 solution at the
              temperature of measurement

       K2 =   conductivity, in micromhos per centimeter, of the KC1 solution at the same
              temperature as the distilled water used to prepare the reference solution

       Kx =   measured conductance, in mhos

Many different manufacturers produce conductivity meters that perform well on water and
wastewater samples.  Selection  should be made consistent with sampling requirements,
availability of service and sales,  and individual personal preference.

3.5 Turbidimeters (Nephelometers)

Many different instrument designs have been used for the optical measurement of turbidity
by measurement of either transmission or reflection of light.  An equal or even greater

                                  Table 3-2
    ELECTRICAL CONDUCTIVITY OF POTASSIUM CHLORIDE REFERENCE
                                 SOLUTIONS
Solution
A


B


C
Normality
0.1


0.01


0.001
Method of Preparation
7.4365 g/1 KC1 at 20°C


0.7440 g/1 KC1 at 20°C


Dilute 100 ml of B to 1.0 1 at 20°C
Temperature
(°C)
0
18
25
0
18
25
25
Conductivity
(|j,mho/cm)
7,138
11,167
12,856
773
1,220
1,408
147
                                     3-7
OCR(6/2001) by RYA-R10

-------
number of materials have been used or proposed as calibration standards. As described in
reference  10, EPA has standardized on the instrument design and the standard turbidity
suspension of Formazin for instrument calibration.

Both the analyst and the  user  of turbidity data should keep in mind that a turbidity
measurement is not a substitute for particle weight or  residue analysis.  Turbidity
instruments can be calibrated to give gravimetric data on specific sample types, but the
influence of particle geometry, specific gravity, refractive index, and color make estimates
of total weight impractical on a variety of sample types.

For production of data with maximum accuracy and precision, the following precautions
should be  observed:

  a.   Protect the sample cuvette from scratches and fingerprints.

  b.   Use a constant orientation of the sample cuvette while calibrating the instrument
       and analyzing samples.

  c.   Use a well-mixed sample in the sample cuvette; do not take readings until finely
       dispersed bubbles have disappeared.

  d.   Dilute samples containing excess turbidity to some value below 40 nephelometric
       turbidity units (NTU); take reading; and multiply results by correct dilution factor.

3.6 Spectrometers

Because a large portion of routine quantitative measurements are performed colorimetri-
cally, the  spectrometer or filter photometer is usually the workhorse of any analytical
laboratory. Indeed, the versatility of such instruments and the number of demands imposed
upon them have resulted in a variety of designs and price ranges.  Systematic listing and
detailed discussion of all instrumental types are beyond the scope of this chapter; however,
ultraviolet, visible, infrared, and atomic absorption instruments will be discussed.

A spectrometer is an instrument  for measuring the amount of light or radiant energy
transmitted through a solution or solid material as a function of wavelength. A spectrometer
differs from a filter photometer in that it uses continuously variable, and more nearly
monochromatic, bands of light.   Because  filter  photometers lack the  versatility  of
spectrometers, they are used most profitably where standard methodologies are used for
routine analysis.

The essential parts of a spectrometer include the following:

  a.   A source of radiant energy

  b.   Monochromator or other device for isolating narrow spectral bands of light

  c.   Cells (cuvettes) or sample holders for containing samples under investigation

  d.    A photodetector (a device to detect and measure the radiant energy passing through
       the sample)
                                       3-8              OCR(6/2001)byRYA-R10

-------
Each of the essential features listed, especially the monochromator and the photodetector
system, varies  in  design  principles  from one  instrument  to  another.  Some  of the
characteristics of the commonly used Perkin-Elmer model 124 double-beam grating
spectrometer are the following:

Light source
       Visible region	Tungsten
       Lamp UV region	Deuterium lamp

Wavelength accuracy	±0.5 nm

Spectral bandwidth 	0.5, 1.0, and 2.0 nm

Photometric presentation
       Linear transmittance	0 to 100 percent
       Linear absorbance	0 to 1A or 0 to 2A

Photodetector R- 136	190 to 800 nm

Sample cells	1.0 to  10 cm

3.6.1 Visible Range

Desirable features on a visible-range spectrometer are determined by the anticipated use of
the instrument. Simple, limited programs requiring use of only a few parameters  can
probably be supported by inexpensive but reliable filter photometers. On the other hand,
if the laboratory programs require a wide variety of measurements on diverse samples at low
concentrations, more versatile instruments may be needed. One of the prime considerations
is adaptability to various sample cell sizes from 1.0 to 10.0 cm.

3.6.2 Ultraviolet Range

A UV spectrometer is similar in design to a visible-range instrument except for differences
in the light source and in the optics.  The UV light source is a  hydrogen- or deuterium-
discharge lamp, which emits radiation in the UV portion of the spectrum, generally from
about 200 nm to the low-visible-wavelength region. The optical system and sample cells
must be constructed of UV-transparent material, which is usually quartz. A grating used in
a UV system may be specially cut (blazed) in the UV region for greater sensitivity.

3.6.3 Infrared  Range

A number of instrumental modifications are required in the construction of spectrometers
for measurements in the IR region because materials such as glass and quartz  absorb IR
radiant energy,  and ordinary photocells do not respond to it. Most IR  spectrometers use
front-surfaced  mirrors to eliminate the necessity for the transmission  of radiant energy
through quartz, glass, or other lens materials. These mirrors are usually  parabolic to focus
the diffuse IR energy. Such instruments must be protected from high humidities and water
vapor to avoid deterioration of the optical system and the presence of extraneous absorption
bands in the IR.
                                       3-9              OCR(6/2001)byRYA-R10

-------
The energy or light source for an IR instrument may be a Nernst glower or a Globar. Each
of these sources has certain characteristics that recommend it for use, but the more rugged
Globar is more commonly used because it also has a more stable emission. The receiving
or detection unit may be athermocouple, bolometer, thermistor, or photoconductor cell. The
type of analysis being performed dictates the degree of sophistication required in the IR
instrumentation selected to acquire usable data.

3.6.4 Proper Use of Spectrometers

The spectrometer manufacturer's instructions for proper use should be followed in all cases.
Several safeguards against misuse of the instruments, however, are mandatory.

Instruments should be checked for wavelength alinement. If a particular colored solution
is to be used at a closely  specified wavelength, considerable loss of sensitivity can be
encountered  if the  wavelength  control is  misalined.   In visible-range instruments, an
excellent reference point is the maximum absorption for a diluted solution of potassium
permanganate, which has dual peaks at 526 and 546 nm. On inexpensive instruments with
less resolution the  permanganate peak appears at 525 to 550 nm as a single, flat-topped
peak.

For both UV and IR instruments, standard absorption  curves for many organic materials
have been published so that reference material for standard peaks  is easily available.
Standard films of styrene and other transparent plastics are available for IR wavelength
checks.  A very good discussion of factors that affect the  proper performance of
spectrometers and standard reference materials available to calibrate them can be found in
publications of the National Bureau of Standards.* Use of certified standards is encouraged
whenever practical.

Too  much emphasis cannot be placed on care of absorption cells.  All absorption cells
should be kept scrupulously clean, free of scratches, fingerprints, smudges, and evaporated
film  residues. Matched cells should be checked to see that they are equivalent, and any
differences should be accounted for during use or in the final data. Directions for cleaning
cells are given in chapter 4.

Generally speaking, trained technicians may operate any of the spectrometers successfully;
however, because interpretation of data from both the UV and IR instruments is becoming
increasingly complex, mere compliance with the operations manual may not be sufficient
for completely accomplishing the special techniques  of sample preparation, instrument
operation, and interpretation of absorption curves.

3.6.5 Atomic Absorption

There are a number of differences in the basic design and accessories for atomic absorption
(AA) equipment that require consideration before purchase and during subsequent use.
These choices concern the light sources, nebulizer burners, optical systems, readout devices,
and mode conversions.  Because some  of these choices are not readily obvious, the
purchaser or user must be familiar with the types and numbers of samples to be analyzed and
*NBS, Office of Standard Reference Materials, Institute for Materials Research,
Washington, D.C. 20234.

                                      3-10              OCR(6/2001)byRYA-R10

-------
the specific elements to be measured before making the choice.  For a program analyzing
a wide variety of samples for a number of elements at varying concentrations, an instrument
of maximum versatility would be best. Most of the discussion that follows applies to use
of instrumentation in absorption, emission, or fluorescence modes.

3.6.5.1  Lamps

Hollow-cathode (HC) lamps or electrodeless-discharge lamps (EDL) are available for over
70 elements with single-element or multielement capability. Multielement lamps are consid-
erably cheaper per element than single-element lamps, but the savings may not be realized
if the lamps are not used strategically, because all the elements in the  cathode deteriorate
when the  lamp  is used, regardless of which element is  measured.  The deterioration
phenomena result from the different volatilities of metals used in the cathode. One metal
volatilizes (sputters) more rapidly than the others and redeposits upon  an area of the other
cathode  metals.  Thus, with continual use, a drift in signal will be noted with at least one
metal increasing and the other (or others) decreasing. If one can ignore the dubious cost
savings of multielement lamps, use of single-element lamps could result in more precise and
accurate data.

The line intensities of one element in a multielement HC lamp will usually be less than those
of a lamp containing a pure cathode of the same element because this  element must share
the discharge energy with the other elements present. However, this reduction should not
affect the output by a factor of more than one-sixth to one-half, depending on the particular
combination and the number of elements combined. The output can be even greater in some
multielement lamps because alloying may permit a higher operating current than for the case
of the pure cathode. All HC lamps have  life expectancies that are related to the volatility
of the cathode metal, and for this reason the manufacturer' s recommendations for operating
should be closely followed.

Recent improvements in design and manufacture of HC lamps and EDLs have resulted in
lamps with more constant output and longer life. Under normal conditions an HC lamp may
be expected to operate satisfactorily for several years. HC lamps used to be guaranteed for
a certain minimum ampere-hour period, but this has been changed to a 90-day warranty. It
is good  practice to date  newly purchased lamps and to  inspect them immediately upon
receipt.  The operating current and voltage indicated on the lamp should not be exceeded
during use.  An increase in background noise or a loss of sensitivity are signs of lamp
deterioration.

A basic design feature of AA spectrometers is the convenience of the HC lamp changeover
system.   Some  instruments provide for as many as six lamps in a  rotating turret, all
electronically stabilized and  ready for use by simply rotating the lamp turret.  Other
instruments provide for use of only one  lamp at a time in the lamp housing, and require
manual removal and replacement whenever more  than one element is  to be measured.  A
quick-changeover system enables frequent lamp changes during the period of operation. If
necessary lamp changes are infrequent, however, multilamp mounts do not represent a great
convenience.

After the proper lamp has been selected, the HC current should be adjusted according to the
manufacturer's recommendations and allowed to electronically stabilize (warm up) before
use.   During this  15-min period, the monochromator may be positioned at the correct
wavelength, and the proper slit width may be selected. For those instruments employing a
multilamp turret,  a warmup current  is provided  to those  lamps not in use, thereby
                                      3-11             OCR(6/2001) by RYA-R10

-------
minimizing the warmup period after the turret is rotated. In a single-lamp instrument, the
instability exhibited during warmup is minimized by the  use of a double-beam optical
system.

3.6.5.2 Burner Types

The most difficult and inefficient step in the AA process  is converting the metal in the
sample from an ion or a molecule to the neutral atomic state.  It is the function of the
atomizer/burner to produce the desired neutral atomic condition of the elements.  With
minor modifications burners are the same as those used for flame photometry.

Basically there are two different types of burners. They  are the total-consumption, or
surface-mix burner, and the laminar-flow or premix burner. There are many variations of
these two basic types,  such as the Boling, the high-solids, the turbulent-flow, the triflame,
the nitrous oxide burners,  and many others.    As one might expect, there are many
similarities among the various burners, the different names resulting from the different
manufacturers. The element being determined and the type of sample solution dictate the
type of burner to be used.   Generally,  all types and makes of burner can be adjusted
laterally, rotationally, and vertically for selection of the most sensitive area of the flame for
the specific element sought.

Nonflame techniques have gained wide acceptance in AA analysis because of the extreme
sensitivity and the capability to directly  introduce very diverse sample matrices. These
systems, which replace the conventional flame  burner, come in various designs using
electrical resistance to produce temperatures as high as 3,500°C.

3.6.5.3 Single- and Double- (Split-) Beam Instruments

There is a great deal of existing uncertainty among instrument users about the relative merits
of single- and double-beam instruments.  Neither system is  appropriate for all cases.

With a single-beam instrument the light beam from the source passes directly through the
flame to the detector.  In a double-beam system the light from the source is divided by a
beam splitter into two paths. One path, the reference beam, goes directly to the detector.
The second path, the sample beam, goes  through the flame to the detector. The  chopper
alternately reflects and passes each beam, creating two equal beams falling alternately upon
the detector.   If the  beams are equal, they cancel the alternate impulses reaching the
detector, and no signal is generated.  If the beams are different, the resulting imbalance
causes the detector to generate a signal that is amplified and measured.  The difference
between the reference and sample beams is then determined as a direct function of absorbed
light.  The advantages of the double-beam design are that any variations in the source are
of reduced importance, and smaller dependence is placed upon the stability of the power
supply.  Conversely, stabilization of the power supply can eliminate the apparent need for
the split-beam system.  Furthermore, the beam splitter requires additional mirrors or optical
accessories that cause  some loss of radiant energy.

A single-beam system does not monitor source  variations,  but offers  certain  other
advantages. It allows use of low-intensity lamps, smaller slit settings, and smaller gain. As
a consequence, the single-beam instrument, properly designed, is capable of operating with
lower noise and better signal-to-noise ratio,  and therefore with better precision and
improved sensitivity.  Because the simplified optical system conserves  radiant energy,
                                      3-12              OCR(6/2001)byRYA-R10

-------
especially in the shorter wavelengths, it facilitates operation in the low-wavelength range.
With this advantage, it should be possible to obtain better sensitivity for those elements with
a strong resonance line below 350 nm and particularly those below 300 nm. Background
correction techniques are also available in single-beam systems.

Double-beam instruments, however, offer the opportunity to perform more sophisticated
techniques like background correction, two-channel multielement analysis, use of internal
standards, and element rationing. If one of these techniques is necessary, a double-beam
instrument must be considered.

3.6.5.4 Readout Devices

Readout devices in even the lower cost AA instruments include digital concentration
display. Using high-speed digital electronics, data-handling techniques encompass multiple
calibration standards, regression analysis to characterize the calibration curves, mean
variance and standard deviation  statistical programs for sample calculations, and various
forms of printout reports  in addition to recorder output.  Choice of a readout system is
predicated largely upon laboratory needs and availability of budget. In general, any step
toward automation is desirable, but the degree of automation should be compatible with the
laboratory program.

3.6.5.5 Miscellaneous Accessories

A number of instruments contain a mode selector, making an instrument usable for either
absorption or emission. The conversion to emission  may be a desirable feature because
certain elements are more amenable to analysis  by this method.  Some models offer an
option of atomic fluorescence and can also be used as a UV/visible spectrometer.

Automatic sample changers are offered for almost all instruments on the market, and as has
been previously stated, any automation feature is  desirable. However, unless a laboratory
program performs a large number of repetitious measurements daily, an automatic sample
changer is not required. As a practical measure, other commonly used sample-changing
devices, although not expressly  designed for AA use, can be interfaced with most AA
instruments.

3.7 Organic Carbon Analyzers

A number of instruments  designed to  measure total organic carbon (TOC) in waters and
wastes have appeared on the market  in the  past several years.  The first of these units
involved pyrolysis followed by IR measurement of the carbon dioxide formed. Sample
injection of 20 to 200 |j,l in a carrier gas of air or oxygen was performed with a syringe.
Combustion at 800°C to 900°C followed by IR analysis was performed automatically with
final output on an analog recorder.  Systems using these principles are still produced and
represent a large part of the TOC market.

Other techniques of TOC analysis that modify every phase of the original TOC instruments
have been introduced. Sample presentation in small metallic boats and purging of CO2 from
solution are two new techniques.  Wet chemical oxidation, either external to the instrument
or within the instrument, using various oxidants including ultraviolet irradiation is now in
wide use. Measurement of the CO2 by reduction to methane (CH4) and quantitation with a
flame ionization detector are also available. Various methods of data handling are now used,
                                      3-13             OCR(6/2001)byRYA-R10

-------
ranging from recorder output to direct readings and printouts of concentration. Techniques
are also available for measuring materials like soil and  sludges, and also the volatile
component of the TOC.  Sensitivity on some  systems has been extended down to the
microgram per liter level.  The major problems associated with TOC measurements are
interference from forms of inorganic carbon and the difficulty of obtaining a representative
sample in the presence of particulate matter. Each system has its own procedure for sample
pretreatment or for accounting  for these problems. When choosing a TOC instrument,
consideration should be given to the types of samples  to be analyzed, the  expected
concentration range, and the forms of carbon to be measured.

3.8 Gas Chromatographs

Because GC's are available from a large number of manufacturers, selection of a particular
manufacturer may be based on convenience. No single multipurpose GC instrument permits
analysis of a wide range of compounds. In this case, a GC/MS could be considered (11). If,
however, relatively few types of environmentally significant compounds are being surveyed,
an inexpensive system equipped with a glass-lined injection port, electrolytic conductivity
detector, and analog recorder is a good choice. A review of the organic methods (12) to be
used will  give  the analyst all the necessary  information on the  specific instrument,
apparatus, and materials necessary for each type or class of compounds.  A discussion of
various quality control considerations in GC analysis is given in chapter 8 of this manual.

Data handling requirements vary widely, and the need to automate GC data collection is
determined by the extent of the sample load.   In a routine monitoring laboratory, GC
systems incorporating their own microprocessers and report generating capabilities would
be useful in solving this problem. Because such  systems greatly increase the cost, the
overall economy of this choice must be considered.

3.9    References

  1.  "Single Arm Balances Testing," Part 41 of 1976 Book of ASTM Standards, American
     Society for Testing and Materials, Philadelphia (1976).
  2. Sorenson, S. P. L., Biochem. Z., 21, 201 (1909).
  3. "pH of Water and Wastewater," from Part 31 of 1976 Book of ASTM Standards,
     American Society for Testing and Materials,  Philadelphia (1976).
  4. Catalog of NBS Standard Reference Materials, NBS Special Publication 260,
     National Bureau of Standards (1975-76).
  5. Methods for Chemical Analysis of Water and Wastes, U.S. EPA, Office of Research
     and Development, EMSL, Cincinnati (1978).
  6. Rechnitz, G. A., "Ion-Selective Electrodes," Chem.  Eng. News (June 12, 1967), p.
     146.
  7. Riseman, Jean M., "Measurement of Inorganic Water Pollutants by Specific Ion
     Electrode," American Laboratory (July 1969), p. 32.
  8. Koryta, Jiri, "Theory and Application of Ion-Selective Electrodes," Anal. Chim.
     Acta, 67,329-411 (1972).
  9. Covington, A. L., "Ion-Selective Electrodes," CRC Critical Review in Anal. Chem.
     (Jan. 1974), pp. 355-406.
  10. Black, A. P., and Hannah, S. A., "Measurement of Low Turbidities," J. Am. Water
     Works Assoc., 57, 901 (1965).
 11. EPA GC/MS Procedural Manual, Budde, W. L., and Eichelberger, J. W., Editors, 1st
     Edition, Vol. 1, U.S. EPA, Office of Research and Development, EMSL, Cincinnati
     (in press).
                                      3-14             OCR(6/2001)byRYA-R10

-------
12.  Methods for  Organic  Analysis  of  Water  and  Wastes, U.S. EPA, EMSL,
    Environmental Research Center, Cincinnati (in preparation).
                                   3-15             OCR(6/2001)byRYA-R10

-------

-------
                                    Chapter 4

                                 GLASSWARE

4.1 General

The measurement of trace constituents in water demands methods capable of maximum
sensitivity. This is especially true for metals and trace organics such as pesticides, as well
as for the determination of ammonia and phosphorus.  In addition to sensitive methods,
however, there are other areas that require special consideration. One such area is that of
the cleanliness of laboratory glassware.  Obviously, the very sensitive analytical systems are
more sensitive to errors resulting from the improper use or choice of apparatus, as well as
to contamination effects due to an improper method of cleaning the apparatus.  The purpose
of this chapter is to discuss the kinds of glassware available, the use of volumetric ware, and
various cleaning requirements.

4.2 Types of Glassware

Laboratory vessels serve three functions: storage of reagents, measurement of solution
volumes, and confinement of reactions. For special purposes, vessels made from materials
such as porcelain, nickel, iron,  aluminum, platinum, stainless steel, and plastic may be
employed to advantage.  Glass, however, is the most widely used material of construction.
There are many grades and types of glassware from which to choose, ranging from student
grade to others possessing specific properties such as super strength, low boron content, and
resistance to thermal shock or alkali. Soft glass containers are not recommended for general
use, especially for storage of reagents because of the possibility of dissolving of the glass
(or of some of the  constituents of the glass). The  mainstay of  the modem analytical
laboratory is a highly resistant borosilicate glass, such as that manufactured by Coming
Glass Works under the name "Pyrex" or by Kimble Glass Co. as "Kimax." This glassware
is satisfactory for all analyses included in reference 1.

Depending on the particular manufacturer, various trade names are used for specific brands
possessing special properties such as resistance to heat, shock, and alkalies. Examples of
some of these special brands follow:

  a.   Kimax- or Pyrex-brand glass is a relatively inert all-purpose borosilicate glass.

  b.   Vycor-brand glass is a silica  glass  (96 percent) made to withstand continuous
       temperatures up to 900°C and can be down-shocked in ice water without breakage.

  c.   Coming-brand glass is claimed  to be 50 times more resistant to alkalies than
       conventional ware and practically boron-free  (maximum 0.2 percent).

  d.   Ray-Sorb- or Low-Actinic-brand glass is used when the reagents or materials are
       light sensitive.

  e.    Corex-brand labware is harder than conventional borosilicates and therefore better
       able to resist clouding and scratching.

The use of plastic vessels, containers, and other apparatus made of Teflon, polyethylene,
polystyrene, and polypropylene has increased markedly over recent years.  Some of these
                                       4-1              OCR(6/2001)byRYA-R10

-------
materials, such as Teflon,  are quite expensive; however, Teflon stopcock plugs have
practically replaced glass plugs in burets, separatory runnels, etc., because lubrication to
avoid sticking or "freezing" is not required. Polypropylene, a methylpentene polymer, is
available as laboratory bottles, graduates, beakers, and even volumetric flasks. It is crystal
clear, shatterproof, autoclavable, and chemically resistant.

The following are some points to consider in choosing glassware or plasticware:

  a.    The special types of glass listed above, other than Pyrex or Kimax, generally are not
       required to perform the analyses given in "Methods for Chemical Analysis of Water
       and Wastes" (1).

  b.   Unless instructed otherwise, borosilicate or polyethylene bottles may be used for
       the storage of reagents and standard solutions.

  c.   Dilute metal  solutions are prone to plate out on container walls over long periods
       of storage. Thus, dilute metal standard solutions must be prepared fresh at the time
       of analysis.

  d.    For some operations, disposable glassware is entirely satisfactory. One example is
       the use of disposable test tubes as  sample containers for use with the Technicon
       automatic sampler.

  e.   Plastic bottles  of polyethylene and Teflon have been found satisfactory for the
       shipment of water samples. Strong mineral acids (such as sulfuric acid) and organic
       solvents will readily attack polyethylene and are to be avoided.

  f.    Borosilicate glassware is not completely inert, particularly to alkalies; therefore,
       standard solutions  of silica, boron, and the alkali metals are usually stored in
       polyethylene bottles.

For additional information the reader is referred to the catalogs of the various glass and
plastic manufacturers.  These catalogs contain a wealth of information such as specific
properties, uses, and  sizes.

4.3 Volumetric Analyses

By common usage, accurately calibrated glassware for precise measurements of volume has
become known as volumetric glassware. This group includes volumetric flasks, volumetric
pipets, and accurately  calibrated burets.  Less accurate types of glassware  including
graduated cylinders and serological  and measuring pipets also have specific uses in the
analytical laboratory when exact volumes are unnecessary.

The precision of volumetric work depends  in part upon the accuracy with which volumes
of solutions can  be measured. There are certain sources of error that must be carefully
considered.  The volumetric apparatus must be  read correctly; that is, the bottom of the
meniscus should be  tangent to the calibration mark.  There are other sources of error,
however, such as changes in temperature, which result in changes in the actual capacity of
glass apparatus and in the volume of the solutions. The capacity of an ordinary glass flask
of 1000-ml volume increases 0.025 ml/deg with rise in temperature, but if the flask is made
of borosilicate glass, the increase is much less.  One thousand milliliters of water or of most
0.1N  solutions  increases in  volume by approximately 0.20 ml/deg increase  at room
                                       4-2              OCR(6/2001)byRYA-R10

-------
temperature.  Thus solutions must be measured at the temperature at which the apparatus
was calibrated.  This temperature (usually 20°C) will be indicated on all volumetric ware.
There may also be errors of calibration of the apparatus; that is, the volume marked on the
apparatus may not be the true volume.  Such errors can be eliminated only by recalibrating
the apparatus or by replacing it.

Volumetric apparatus is calibrated to contain or to deliver a definite volume of liquid. This
will be indicated on the apparatus with the letters "TC" (to contain) or "TD" (to deliver).
Volumetric flasks are calibrated to contain a given volume and are available in various
shapes and sizes.

Volumetric pipets are calibrated to  deliver a fixed volume.  The usual capacities are  1
through 100 ml although micropipets are also available. Micropipets are most useful in
furnace work and are available in sizes ranging from 1 to 100  |j,l.

In emptying volumetric pipets, they  should be held in a vertical position and the outflow
should be unrestricted. The tip of the pipet is kept in contact with the wall of the receiving
vessel for a second or two after the free flow has stopped.  The liquid remaining in the tip
is not removed; this is most important.

Measuring and serological pipets should also be held in a vertical position for dispensing
liquids; however, the tip of the pipet is only touched to the wet surface of the receiving
vessel after the outflow has  ceased.  For those pipets  where the small amount of liquid
remaining in the tip is to be blown out and added, indication is made by a frosted band near
the top.

Burets are used to deliver definite volumes.  The more common types are usually of 25- or
50-ml capacity, graduated to tenths of a milliliter, and  are provided with stopcocks.   For
precise analytical methods in  microchemistry, microburets are  also used.  Microburets
generally are of 5- or 10-ml capacity, graduated in divisions of hundredths of a milliliter.
Automatic burets with reservoirs are also available ranging in capacity from 10 to 100 ml.
Reservoir capacity ranges from 100 to 4,000 ml.

General rules in regard to the manipulation of a buret are as follows: Do not attempt to dry
a buret that has been cleaned for use, but rinse it two or three times with a small volume of
the solution with which it is to be filled. Do not allow alkaline solutions to stand in a buret
because the glass will be attacked, and the stopcock, unless made of Teflon, will tend to
freeze. A 50-ml buret should not be emptied faster than 0.7 ml/s, otherwise too much liquid
will adhere to the walls and as  the solution drains down, the meniscus will gradually rise,
giving a high false reading. It should be emphasized that improper use or reading of burets
can result in serious calculation errors.

In the case of all apparatus for delivering liquids, the glass must be absolutely clean so that
the film of liquid never breaks at any point. Careful attention must be paid to this fact or the
required amount of solution will not be delivered. The various cleaning agents and their use
are described later.

4.4 Federal Specifications for Volumetric Glassware

Reference 2 contains a description of Federal specifications for volumetric glassware.  The
National Bureau of Standards  no longer accepts stock quantities of volumetric apparatus
                                       4-3              OCR(6/2001)byRYA-R10

-------
from manufacturers or dealers for certification and return for future sale to consumers. This
certification service is still available, but apparatus will be tested only when submitted by
the ultimate user, and then only after an agreement has been reached with the Bureau
concerning the work to be done.

Consequently, the various glass manufacturers have discontinued the listing of NBS-
certified ware.  In its place catalog listings of volumetric glass apparatus that meet the
Federal specifications are designated as  class A and all such glassware is permanently
marked with a large "A." These NBS specifications are listed in table 4-1. The glassware
in question includes the usual burets, volumetric flasks, and volumetric pipets.

In addition to the "A" marking found on calibrated glassware and the temperature at which
the calibration was made, other markings also appear. These include the type of glass, such
                                    Table 4-1
                       TOLERANCES FOR VOLUMETRIC
                                 GLASSWARE1
Type
of
Glassware
Graduated flask







Transfer pipet






Buret3




Capacity2
(ml)
25
50
100
200
250
300
500
1,000
2,000
2
5
10
25
30
50
100
200
5
10
30
50
100
Limit of Error
(ml)
0.03
0.05
0.08
0.10
0.11
0.12
0.15
0.30
0.50
0.006
0.01
0.02
0.025
0.03
0.05
0.08
0.10
0.01
0.02
0.03
0.05
0.01
                  1 Abridged from reference 3.
                  2 Less than and including.
                  3 Limits of error are of total or partial capacity.
                   Customary practice is to test the capacity at five
                   intervals.
                                       4-4               OCR(6/2001)byRYA-R10

-------
as Pyrex, Corex, or Kimax; the stock number of the particular item; and the capacity of the
vessel. If the vessel contains a ground-glass connection, this will also be included along
with the TD or TC symbol. An example of the markings usually found on volumetric ware
is shown in figure 4-1.  Class A glassware need not be recalibrated before use.  However,
if it should become necessary to calibrate a particular piece of glassware, directions may be
found in texts (4) on quantitative analysis.

4.5 Cleaning of Glass and Porcelain

The method of cleaning should be adapted to both the substances that are to be removed, and
the  determination to be performed. Water-soluble substances are simply washed out with
hot or cold water, and the vessel is finally rinsed with successive small amounts of distilled
water. Other substances more difficult to remove may require the use of a detergent, organic
solvent, dichromate cleaning solution, nitric acid, or aqua regia  (25 percent by volume
concentrated HNO3 in concentrated HC1).  In all cases it is good practice to rinse a vessel
with tap water as soon as possible after use.  Material allowed to dry on glassware is much
more difficult to remove.

Volumetric glassware, especially burets, may be thoroughly cleaned by amixture containing
the  following: 30 g of sodium hydroxide, 4 g of sodium hexametaphosphate (trade name,
Calgon), 8 g of trisodium phosphate, and 1 1 of water.  A gram or two of sodium lauryl
sulfate or other surfactant will improve its action in some cases. This solution should be
used with a buret brush.

Dichromate cleaning solution (chromic acid) is a powerful cleaning agent; however, because
of its destructive nature upon clothing and upon laboratory furniture, extreme care must be
taken  when using this mixture.  If any of the  solution  is spilled, it must be  cleaned up
immediately. Chromic acid solution may be prepared in the laboratory by adding 1 1 of
concentrated sulfuric acid slowly, with stirring, to a 3 5-ml saturated sodium dichromate
solution. This mixture must be allowed to stand for approximately  15 min in the vessel that
is being cleaned and may then be returned to a storage bottle.  Following the chromic acid
wash, the vessels are rinsed thoroughly with tap water, then with small successive portions
of distilled water.  The analyst should be cautioned that when chromium is included in the
PYREX
USA
^ ^^^^^^^^V T
| S ^—
^ ^^^^^^*

500 ml ±0.20 ml -*•
TC 20° C «.. 	 - 	 	 	 , 	 -

NO 5680 "*
GLASS CO.
TYPE
STANDARD j
— TAPER 	 » S 4
SIZE 19

TO
CONTAIN
STOCK
NO.
KIMAX
USA
^
V


„. 	 ^, TP ?n°r

^^ NO 28013
                   Figure 4-1.  Example of markings on glassware.
                                       4-5               OCR(6/2001)byRYA-R10

-------
scheme of analysis, it is imperative that the last traces of dichromate be removed from the
apparatus.  To this end, a substitute for dichromate cleaning solution, called Nochromix,*
is available and may be used to advantage. Fuming nitric acid is another powerful cleaning
agent, but is disagreeable to handle. As with dichromate, when the acid becomes dilute, the
cleaning mixture is no longer effective.  A mixture of concentrated sulfuric and fuming
nitric acids is even more efficient but is also hazardous to use. A persistent greasy layer or
spot may be removed by acetone or by allowing a warm solution of sodium hydroxide, about
1 g per 5 0 ml of water, to stand in the vessel for 10 to 15 min; after rinsing with water, dilute
hydrochloric  acid, and water again,  the  vessel is usually clean.   Alcoholic  potassium
hydroxide is also effective in removing grease. To dry glass apparatus, rinse with acetone
and blow or draw air through it.

4.6 Special Cleaning Requirements

Absorption cells, used in spectrophotometers, should be kept scrupulously clean, free of
scratches, fingerprints, smudges, and evaporated film residues.  The cells may be cleaned
with detergent solutions for removal of organic residues, but should not be soaked for
prolonged periods in caustic solutions because of the possibility of etching. Organic solvents
may be used to rinse cells in wl-iich organic materials have been used. Nitric acid rinses are
permissible, but dichromate solutions are not recommended because of the adsorptive
properties of dichromate on glass. Rinsing and drying of cells with alcohol or acetone
before storage is a preferred practice. Matched cells should be checked to see that they are
equivalent by placing portions of the same solution in both cells and taking several readings
of the transmittance (T, percent) or optical density (OD) values.

For certain determinations., especially trace metals, the glassware should also be rinsed with
a 1:1 nitric acid-water mixture. This operation is followed by thoroughly rinsing with tap
water and successive portions of distilled water.  This may require as many as 12 to  15
rinses, especially if chromium is being determined. The nitric acid rinse is also especially
important if lead is being determined.

Glassware to be used for phosphate determinations should not be washed with detergents
containing  phosphates.  This glassware must be thoroughly rinsed with tap  water and
distilled water.  For ammonia and Kjeldahl nitrogen, the  glassware must be rinsed with
ammonia-free water. (See ch. 2.)

Glassware to be  used in the determination of trace organic constituents in water, such as
chlorinated pesticides, should be as free as possible of organic contaminants.  A chromic
acid wash of at least 15 min is necessary to destroy these organic residues. Rinse thoroughly
with tap water and, finally, with distilled water. Glassware may be dried for immediate use
by rinsing with redistilled acetone. Otherwise glassware may be oven dried or drip dried.
Glassware should be stored immediately after drying to prevent any accumulation of dust
and stored inverted or with mouth of glassware covered with foil.

Bottles to be used for the collection of samples for organic analyses should be rinsed
successively with chromic acid cleaning solution, tap water, distilled water, and, finally,
several times with a  redistilled solvent  such as  acetone, hexane, petroleum ether,  or
chloroform. Caps are washed with detergent, rinsed with tap water, distilled water, and
* Available from Godax Laboratories, 6 Varick Street, New York, N.Y. 10013.

                                       4-6              OCR(6/2001)byRYA-R10

-------
solvent.  Liners are treated in the same way as the bottles and are stored in a sealed
container.

4.7 Disposable Glassware

When the risk of washing a pipet for reuse becomes too great, as in the case of use with
toxic materials, or when the cost of washing glassware becomes prohibitive, disposable
vessels may be the answer, provided they meet the necessary specification. Various types
are available including bacteriological, serological, and microdilution pipets. Disposable
glassware generally is made of soft glass although plastic vessels and pipets are also
available.

4.8 Specialized Glassware

The use of vessels and glassware fitted with standard-taper, ground-glass, and ball-and-
socket joints has increased because of certain advantages such as less leakage and fewer
freezeups. Standard-taper, interchangeable ground joints save time and trouble in assembling
apparatus. They are precision-ground with tested abrasives to  insure an accurate fit and
freedom from leakage. Ball and socket joints increase flexibility of operation and eliminate
the need for exact alinements of apparatus.   Symbols and their meaning as applied to
standard joints, stoppers, and stopcocks are shown below.

4.8.1  Standard Taper (I)

The symbol 1" is  used to designate interchangeable joints, stoppers, and stopcocks that
comply with the requirements of reference 5. All mating parts are finished to a 1:10 taper.

The size of a particular piece appears after the appropriate symbol. Primarily because of
greater variety of apparatus equipped with  1" fittings, a number of different types of
identifications are used:

  a.   For joints-a two-part number as 1" 24/40, with 24 being the approximate diameter
       in millimeters at the large end of the taper and 40 the axial length of taper, also in
       millimeters

  b.    For stopcocks-a single number, as ¥ 2, with 2 mm being the approximate diameter
       of the hole or holes through the plug

  c.    For bottles-a single number, as  ¥ 19, with 19 mm being the appropriate diameter at
       top of neck. However, there are differences in dimensions between the bottle and
       flask stoppers

  d.   For flasks and similar containers-a single number, as ¥ 19, with 19 mm being the
       appropriate diameter of the opening at top of neck

4.8.2 Spherical Joints (§)

The designation §>  is for spherical (semiball) joints complying with reference  5. The
complete designation of a spherical joint also consists of a two-part number, as 12/2, with
12 being the approximate diameter of the ball and 2 the bore of the ball and the socket, also
in millimeters.
                                       4-7              OCR(6/2001)byRYA-R10

-------
4.8.3 Product Standard
The symbol 
-------
4.9.2 Cleaning of Used Filters

In many cases, precipitates can be removed by rinsing with water, passed through from the
underside, with the pressure not exceeding 15 lb/in2.  The suggestions that follow will be
helpful in dealing with material that will not be removed by such a reverse water-wash:

       Material                                    Removal Agent

Albumen	Hot ammonia or hydrochloric acid

Aluminous and siliceous residues . . Hydrofluoric acid (2 percent) followed by  concen-
                                    trated sulfuric acid; rinse immediately with water
                                    until no trace of acid can be detected.

Copper or iron oxides  	Hot hydrochloric acid plus potassium chlorate

Fatty materials	Carbon tetrachloride

Mercuric sulfide	Hot aqua regia

Organic matter	Hot, concentrated cleaning solution, or hot,  concen-
                                    trated sulfuric  acid with a few drops of sodium
                                    nitrite

Silver chloride 	Ammonium or sodium hyposulfite

4.10 References

1.   Methods for Chemical Analysis of Water and Wastes, U.S. EPA, Office of Research and
    Development, EMSL (1978).
2.   Hughes, J. C., Testing of Glass Volumetric Apparatus, NBS Circular 602, National
    Bureau of Standards (1959).
3.   Peffer, E. L., and Mulligan, G.  C., Testing of Glass Volumetric Apparatus, NBS
    Circular 434, National Bureau of Standards (1941).
4.   Willard, H. H., and Furman, N. H.,  Elementary Quantitative Analysis-Theory and
    Practice, D. Van Nostrand Co., Inc., New York (1947).
5.   Interchangeable Ground Glass Joints, Commercial Standard CS-21 -30, National Bureau
    of Standards (Aug. 25,  1930).
6.   Maximum Pore Diameter and Permeability of Rigid Porous Filters for Laboratory Use,
    E-128, American Society for Testing and Materials, Philadelphia (1968).
                                      4-9              OCR(6/2001)byRYA-R10

-------

-------
                                    Chapter 5

                     REAGENTS, SOLVENTS, AND GASES
5.1 Introduction
The objective of this chapter is to provide general information and suggestions that will
serve to keep the analyst conscious of his responsibilities in analytical quality control, as
they relate to reagents, solvents, and gases. While the material presented here will assist the
analyst in producing high-quality data, it is by no means complete. It is incumbent on the
analyst  to  obtain  details  of special precautions required to  insure proper selection,
preparation, and storage of reagents, solvents, and gases from the descriptions of individual
methods.

5.2 Reagent Quality

Chemical reagents, solvents, and gases are available in a wide variety of grades of purity,
ranging from technical grade to various ultrapure grades.  The  purity of these materials
required in analytical chemistry  varies with the type of analysis.  The parameter being
measured and the sensitivity and specificity of the detection system are important factors in
determining the purity of the reagents required.  For many analyses, including most
inorganic analyses, analytical reagent grade  is satisfactory.  Other analyses, such as trace
organic and radiological, frequently require special ultrapure reagents, solvents, and gases.
In methods where the purity of reagents is not specified it is intended that analytical reagent
grade be used.  Reagents of lesser purity than that specified by the method should not be
used. The labels on the container should be checked and the contents examined to verify
that the purity of the  reagents meets the needs  of the particular method involved.  The
quality of reagents, solvents, and  gases required for the various classes  of analyses -
inorganic, metals, radiological, and organic - are discussed  in this section.

Reagents must always be prepared and standardized with the utmost of care and technique
against reliable primary standards. They must be restandardized or prepared fresh as often
as required by their stability.  Stock and working standard solutions must be checked
regularly for signs of deterioration; e.g., discoloration, formation of precipitates, and change
of concentration. Standard solutions should be properly labeled as to compound, concentra-
tion, solvent, date, and preparers

Primary standards must be obtained from a  reliable source, pretreated (e.g., dried, under
specified conditions), accurately prepared in  calibrated volumetric glassware, and stored in
containers that will not alter the reagent. A large number of primary standards are available
from the National Bureau of Standards (NBS). A complete listing of available standards is
given in reference 1. Primary standards may also be obtained from many chemical supply
companies. Suppliers for special quality reagents, solvents, and gases are noted in later
discussions of the various classes of analyses.   Reagents and solvents of all grades are
available from many chemical supply houses.

There is some confusion among chemists as to the definition of the terms "Analytical
Reagent Grade," "Reagent Grade," and "ACS Analytical Reagent Grade." A review of the
literature and chemical supply catalogs indicates that the  three terms are synonymous.
Hereafter, in this document, the term "Analytical Reagent Grade" (AR) will be used. It is
                                       5-1              OCR(6/2001)byRYA-R10

-------
intended that AR-grade chemicals and solvents shall conform to the current specifications
of the Committee on Analytical Reagents of the American Chemical Society (2).

References 3  through 5 devote several chapters  to  problems related  to preparation,
standardization, and storage of reagents. The information provided therein is particularly
appropriate to  inorganic determinations. The type of volumetric glassware to be used, the
effect  of certain  reagents  on  glassware, the  effect  of temperature  on  volumetric
measurements, purity of reagents, absorption of gases and water vapor from the air,
standardization of solutions, instability, and the need for frequent standardization of certain
reagents are among the topics discussed.  It is  recommended that the analyst become
thoroughly familiar with these publications.

5.2.1 General Inorganic Analyses

In general, AR-grade reagents and solvents are satisfactory for inorganic analyses.  Primary
standard reagents must, of course, be  used for  standardizing all volumetric solutions.
Commercially prepared reagents and standard solutions are very convenient and may be
used when it is demonstrated that they meet the method requirements. All prepared reagents
must be checked for accuracy.

The individual methods specify the reagents that require frequent standardization, or other
special treatment, and the analyst must follow through with these essential operations. To
avoid waste, the analyst should prepare a limited volume of such reagents, depending  on the
quantity required over a given period of time.  Examples and brief discussions of the kind
of problems that occur are given in section 5.3.

As far as possible, distilled water used for preparation of reagent solutions must be free of
measurable amounts of the constituent to be determined. Special requirements for distilled
water are given in chapter 3 of this manual and in individual method descriptions.

Compressed gases, such as oxygen and nitrogen, used for total organic carbon determination
may be of commercial grade.

5.2.2 Metals Analyses

All  standards used for atomic  absorption  and emission spectroscopy should  be  of
spectroquality. It is recommended that other reagents and solvents also be of spectroquality,
although AR grade is sometimes satisfactory. Standards may be prepared by the analyst in
the laboratory, or spectrographically standardized materials may be purchased commercially.
Standards required for determination of metals in water are not generally available from the
National Bureau of Standards.

Analytical-reagent-grade nitric and hydrochloric acids  must be specially prepared  by
distillation in borosilicate glass and diluted with deionized distilled water. All other reagents
and standards  are also prepared in deionized water.

In general, fuel and  oxidant gases used for atomic absorption can be of commercial grade.
Air supplied by an ordinary laboratory compressor is quite satisfactory, if adequate pressure
is maintained  and necessary precautions are taken to filter oil, water, and possible trace
metals from the line. For certain determinations such as aluminum, AR-grade nitrous oxide
is required.
                                       5-2               OCR(6/2001)byRYA-R10

-------
5.2.3 Radiological Analyses

The great sensitivity of radioactive counting instruments requires that scintillation-grade
reagents and solvents, or equivalent, be used for all radioactivity determinations. Some of
the reagents, for example,  strontium carbonate and yttrium oxide carriers used for the
determination of strontium-90 and yttrium-90, must be stable, that is, free of radioactivity.
Barium sulfate, used for coprecipitation of radium, must be free from all traces of radium.
These reagents and solvents are commercially available from chemical supply houses.

Calibrated standard sources of specific radioactive materials with known count and date of
counting are available from various suppliers. No single company supplies all standards.

Gases used for radioactive counting must be of high purity and extra dry. Gases such as
helium and air are aged for about 30 days to allow radioactive background to decay.  All
gases are checked for background before use. Some cylinders contain inherent radioactivity
that is imparted to the gas.  When this background is above normal, the gas should not be
used for radioactivity determinations.

5.2.4 Organic Chemical Analyses

The  minimum purity of reagents that  can  be  used for organic analyses is AR  grade.
Reference-grade standards should be used whenever available. Special note should be taken
of the assay of standard  materials.   Owing  to  the great sensitivity (nanogram and
subnanogram quantities)  of gas chromatography (GC), which  is often used to  quantitate
organic results, much greater purity is frequently required (6). The specificity of some GC
detectors requires that reagents and solvents be free of certain classes of compounds. For
example,  analyses by  electron capture require that reagents and  solvents be free of
electronegative materials that would interfere with the determination of specific compounds
in the sample. Similarly, use of the flame photometric detector requires that reagents and
solvents be  free from  sulfur and phosphorus interference.  Pesticide-quality solvents,
available from several sources, are required when doing low-level work. AR-grade solvents
are permissible when analyzing industrial waste samples.

However, the contents  of each container must be checked to assure  its suitability for the
analyses.  Similarly, all analytical reagents and other chemicals must also be checked
routinely.

The  quality  of gases required  for GC determinations varies somewhat with the type of
detector. In general, the  compressed gases are a prepurified dry grade. Grade A helium from
the U.S. Bureau of Mines has always been satisfactory. The Dohrmann nitrogen detection
system requires the use  of ultrapure hydrogen for satisfactory results.  Argon-methane used
for electron-capture work should be oxygen free and should have an oxygen trap in the
supply line.  The use of molecular-sieve, carrier-gas filters and drying tubes is required on
combustion gases and is recommended for use on all other gases.  It  is recommended that
the analyst familiarize himself with an article by Burke (7) on practical aspects  of GC.

All reagents, solvents, and adsorbents used for thin-layer chromatography must be checked
to be certain that there are no impurities present that will react with the  chromogenic reagent
or otherwise  interfere with  subsequent qualitative  or quantitative  determinations.
Glass-backed layers prepared in  the  laboratory or precoated layers  supplied  by a
manufacturer may be used; however, precoated layers are more difficult to scrape.  It is
                                       5-3              OCR(6/2001)byRYA-R10

-------
recommended, therefore, that layers prepared in the laboratory be used when zones are to
be  scraped to recover  isolated compounds.   Plastic-backed  layers  are generally
unsatisfactory for this type of analysis.

Adsorbents most commonly used for column chromatographic cleanup of sample extracts
are Florisil,* silica gel, and alumina.  These must be preactivated according to the method
specifications and checked for interfering constituents.

5.3 Elimination of Determinate Errors

To produce high-quality analytical data, determinate errors must be eliminated or at least
minimized. For purposes of this discussion, we assume that a competent analyst and reliable
equipment in optimum operating condition are available.  Thus, determinate errors that
might result from an inexperienced or careless analyst and poor equipment are eliminated.
The remaining sources of error are the reagents, solvents, and gases that are used throughout
the analyses.  The quality of these materials, even though they are AR grade or better, may
vary from one source to another, from one lot to another, and even within the same lot.
Therefore, the analyst must predetermine that all of these materials are free of interfering
substances under the conditions of the analyses.  To do this he must have a regular check
program.  Materials that do not meet requirements are replaced or purified so that they can
be used.

5.3.1 Reagent Blank

The first step the analyst must take is to determine the background or blank of each of the
reagents and solvents used in a given method  of analysis.  The conditions for determining
the blank must be identical to those used throughout the analysis, including the detection
system. If the reagents and solvents contain substances that interfere  with  a particular
analysis, they should be treated  so that they can be used, or satisfactory reagents and
solvents must be found.

5.3.2 Method Blank

After determining the individual reagent or solvent blanks, the analyst must determine the
method blank to see if the cumulative blank interferes with the analyses. The method blank
is  determined by following the procedure step by step, including all of the reagents and
solvents, in the quantity required by the method.  If the cumulative blank interferes with the
determination, steps must be taken to eliminate or reduce the interference to a level that will
permit this combination of solvents  and  reagents to be used.   If the  blank cannot be
eliminated, the  magnitude of the interference must be considered when  calculating the
concentration of specific constituents  in the samples being analyzed.

A method  blank should be determined whenever an analysis is made. The  number of blanks
to  be run  is determined by the method of analysis  and the number of samples being
analyzed at a given time.  In some methods, such as the AutoAnalyzer procedures, the
method blank is automatically and continuously compensated for because a continuous flow
of the reagents passes through  the  detector.   In other procedures,  such  as  the gas
chromatographic determination of pesticides, a method blank is  run with each series of
samples analyzed. Usually this is one blank for every nine samples.
* Trademark of Floridin Co.

                                       5-4              OCR(6/2001)byRYA-R10

-------
5.3.3 Elimination of Interferences and Other Sources of Error

Procedures for eliminating  or at least minimizing impurities that produce  specific
interferences or high general background vary with the reagent and method involved. These
procedures may include the following: recrystallization, precipitation, distillation, washing
with an appropriate solvent, or a combination of these. Examples of procedures used for
various types of analyses are given below.  For complete information, the analyst should
consult the individual methods.

5.3.3.1 General  Inorganic Analyses

Analytical-reagent-grade chemicals and solvents usually present no interference problems
in inorganic analyses. However, some reagents do not always meet methods requirements.
An example is potassium persulfate used in phosphorus and nitrogen determinations.  This
reagent is frequently contaminated with ammonia.  Therefore, it is routinely purified by
passing air through a heated water solution of the reagent. The purified potassium persulfate
is recovered by recrystallization.

A problem more commonly encountered in inorganic analyses is the rapid deterioration of
the standard reagents and other ingredients. To minimize or eliminate this problem, some
reagents, for example, ferrous ammonium sulfate, must be standardized daily.  Others, such
as sodium thiosulfate used for dissolved oxygen determination, may require a substitute
reagent such as phenylarsine oxide.  Solid phenol, which readily oxidizes and acquires a
reddish color, can be purified by distillation.  Starch indicator used for idiometric titrations
may be prepared for each use or preserved by refrigeration or by addition of zinc chloride
or other suitable compounds.

5.3.3.2 Metals Analyses

In general, spectrograde chemicals, solvents, and gases present no interference problems in
atomic absorption or emission spectrographic determinations.  However, standards that do
not meet  the requirements of the method are sometimes obtained. Ordinarily, no effort is
made to purify them. They are simply replaced by new reagents of sufficient purity. Some
reagents  may form precipitates on standing.  Such reagents will reduce the accuracy of
quantitative analyses and should not be used.

5.3.3.3 Radiological Analyses

In general, reagents that do not meet the purity requirements for radiological determinations
are replaced with reagents that are  satisfactory. However, in some instances (for example,
barium sulfate used for coprecipitation of radium) it may be necessary to perform repeated
recrystallization to remove all forms of radium, and reduce the background count to a usable
level. In some instances, solvents that do not meet requirements may be distilled to produce
adequate purity. In some cases, gases having background counts may be usable after aging
as described earlier.  If not, they should be replaced with gases that are satisfactory.

5.3.3.4 Organic Analyses

Many AR-grade chemicals and solvents, and at times pesticide-quality solvents, do not meet
the specifications required forthe determination of specific organic compounds.  Impurities
                                       5-5              OCR(6/2001)byRYA-R10

-------
that are considered trace, or insignificant, for many analytical uses are often present in
greater  quantities than the organic constituents being  measured.  Coupled with  the
several-hundred-fold concentration of the sample extract that is usually required, such
impurities can cause very significant interferences in trace organic analyses. Reagents and
solvents found to be unsatisfactory, under the conditions of the analyses, must be replaced
or cleaned up so that they are usable. Some useful cleanup procedures are -

  a.    Washing the inorganic reagents with each solvent that the reagent contacts during
        the analysis

  b.    Washing the adsorbents, such as silica gel G and Florisil, with the solvents that are
        used  for  a specified  column  or  thin-layer chromatographic procedure,  or
        reactivating the Florisil by firing to 630°C

  c.    Preextracting distilled water with solvents used for the particular analysis involved

  d.    Preextracting aqueous reagent solutions with the solvents involved

  e.    Redistilling solvents in  all-glass systems using an efficient fractionating column

  f     Recrystallizing reagents and dyes used in calorimetric or thin-layer determinations

If the reagents and solvents thus produced  are  not of sufficient purity, they should be
replaced.

Dirty gases (quality less than specified) are particularly troublesome in gas chromatographic
analyses.  They may reduce the sensitivity of the detector, and produce a high or noisy
baseline. If this occurs, the cylinder should be replaced immediately. Similarly, if cylinders
of compressed gases are completely emptied in use, the end volumes of the gas may produce
a similar and often more severe effect. Oils and water may get into the system and foul the
detector. When this occurs the  system must be dismantled and cleaned.  Overhaul of the
detector may be required.  To reduce  chances  of this, it is recommended that  all  gas
cylinders be replaced when the pressure falls to  100 to 200 psi. Filter driers are of little help
in coping with this type of contamination.

5.3.4 Storing and Maintaining Quality of Reagents and Solvents

Having performed the tasks of selecting, preparing, and verifying the suitability of reagents,
solvents, and gases, the analyst must properly store them to prevent contamination and
deterioration prior to their use.  Borosilicate glass bottles with ground-glass stoppers are
recommended for most  standard  solutions and solvents.  Plastic containers such as
polyethylene are recommended  for alkaline solutions.  Plastic containers must not be used
for reagents or solvents intended for organic analyses. However, plastic containers may be
used for reagents not involved with organic analyses if they maintain a constant volume, and
if it is demonstrated that they do not produce interferences and do not absorb constituents
of interest.  It is important that all containers be properly cleaned and stored prior to use.
(Refer to ch. 4 for details.)

Standard reagents and solvents must always be stored according to the manufacturer's
directions.   Reagents or solvents that are  sensitive to the light should be stored in dark
bottles and  in a cool, dark place.  It is particularly important to store materials used for
                                        5-6              OCR(6/2001)byRYA-R10

-------
radiological determinations in dark bottles, because photoluminescence will produce high
background if light-sensitive detectors are used  for counting.   Some reagents require
refrigeration.

Adsorbents for thin-layer and column chromatography are stored in the containers that they
are supplied in, or according to the requirements of individual methods.  When new stock
solutions are  necessary, dilutions of the old and new standard should be compared to
determine their accuracy.

The  analyst should pay particular attention to  the  stability of the standard reagents.
Standards should not be kept longer than recommended by the  manufacturer or in  the
method.  Some standards are  susceptible to changes in normality because of absorption of
gases or water vapor from the air.  Provisions for minimizing this effect are given in
reference 4.

The concentration of the standards will change as a result of evaporation of solvent. This
is especially true of standards prepared in volatile organic solvents. Therefore, the reagent
bottles should be kept stoppered, except when actually in use. The chemical composition of
certain standards may change on standing.  Certain pesticides, for instance, will degrade if
prepared in acetone that contains small quantities of water. Thus, it is essential that working
standards be frequently checked to determine changes in concentration or composition.
Stock solutions should be checked before preparing new working standards from them.

5.4 References

1.   Catalog of Standard Reference  Materials, NBS Special Publication  260, National
    Bureau of Standards (June 1975).
2.   "Reagent  Chemicals,"   American  Chemical  Society  Specifications,  5th Edition,
    American Chemical Society, Washington, B.C. (1974).
3.   Manual on Industrial Water and Industrial Waste Water, 2nd Edition, ASTM Special
    Publication 148-H,  American Society for Testing and Materials (1965), p. 869.
4.   "Standard Methods  for Preparation, Standardization, and Storage of Standard Solutions
    for Chemical Analysis,"  from Part 31 of 1976 Book of ASTM Standards, American
    Society for Testing  and Materials, Philadelphia (1977).
5.   Standard  Methods for the Examination  of Water and Wastewater,  13th Edition,
    American Public Health Association, New York (1971).
6.   Methods  for Organic Pesticides in Water and Wastewater, U.S. EPA, Environmental
    Research  Center, Cincinnati (1971).
7.   Burke,  J., "Gas Chromatography  for Pesticide  Residue Analysis; Some Practical
    Aspects," J. Assoc.  Off.  Anal.  Chem., 48, 1037 (1965).
                                       5-7              OCR(6/2001)byRYA-R10

-------

-------
                                   Chapter 6

           QUALITY CONTROL FOR ANALYTICAL PERFORMANCE

6.1 Introduction

Previous chapters discussed basic elements of quality control (QC) pertaining to laboratory
services, instrumentation, glassware, reagents,  solvents, and gases; the reader should refer
to the appropriate sections to determine necessary specifications and requirements for QC.
Assuming that these basic variables are under QC, that approved methods are being used,
and that the complete system is initially under QC, valid precision and accuracy data must
initially be developed for each method and analyst. Then, to insure that valid data continue
to be produced, systematic daily checks must show that the test results remain reproducible,
and that the methodology is actually measuring the quantity in each sample. In addition, QC
must begin with sample collection and must not end until the resulting data have been
reported.  QC of analytical performance within the laboratory is thus but one vital link in
the dissemination of valid data to the public. Understanding and conscientious use of QC
among all field sampling personnel, analytical personnel,  and management personnel is
imperative. Technical approaches are  discussed in the following sections.

6.2 The Industrial Approach to QC

In the 1920's, Dr. Walter A. Shewhart of Bell Telephone Laboratories, Inc. developed the
theory of  control charts as a basic method for evaluating the  quality of products from
manufacturing processes. His book (1) on statistical QC grew out of this original work.
Later, acceptance of his concepts and  related statistical techniques within industry led to
refined, quantitative evaluations of product quality in manufacturing. Dr. Shewhart's work
on production processes assumed a uniform product manufactured in large numbers and
inspected on a continuous basis through the periodic analysis of samples of n production
units. The re suiting data, X^X2,... Xm were then used to estimate precision, as the standard
deviation S or range R, and accuracy, as the  arithmetic mean X.  These statistics were
calculated as follows:


                            2X2-  2X-
                                    ^
                                    n -1


                      R the largest of the Xl - the  smallest of the Xl

                            n
                           I X,
These statistics were evaluated by plotting them on control charts developed from similar
statistics taken while the process was under properly controlled operation.  The elements
common to such control charts are represented in figure 6-1. They include an expected
value (the central line) and an acceptable range of occurrence (the region between upper and
lower control limits).
                                    6-1                OCR(6/2001)byRYA-R10

-------
                                   UPPER CONTROL LIMIT
                                     CENTRAL LINE
                                  LOWER CONTROL LIMIT
                   05          10        IS        20
                                    SAMPLE NUMBER

                       Figure 6- 1. Essentials of a control chart.

There are many reference sources available that discuss in great detail the classic Shewhart
control charts and related statistics that have since been developed for specific industrial
applications (2-4). In addition, many authors have discussed applications of a related type
of control chart called a cumulative-summation (cusum) chart (2,4). Rather than evaluating
each sample independently, the cusum chart evaluates the cumulative trend of the statistics
from  a series of samples. Because each successive  point is based upon a cumulative data
trend, cusum charts are often considered more effective than control charts in recognizing
process changes and, therefore, may minimize losses from production of unacceptable units;
however, cusum  charts require the more  difficult  calculations, and optimally designed
Shewhart techniques have been found to be almost as effective (2,4), so there is no universal
agreement on the choice between them.

6.3 Applying Control Charts in Environmental Laboratories

In industrial applications, separate control charts are recommended for each product, each
machine, and each operator. Analogous system variables in an environmental laboratory are
the parameter, the  instrument, and the analyst.  However, environmental laboratories
routinely have to contend with a variable that has no industrial counterpart-the true
concentration level of the investigated parameter,  which may  vary considerably among
samples.  Unfortunately, the statistics that work well for industry are sensitive to the
variability in true concentration that is  common in environmental analysis; e.g., the classic
X and R statistic values increase  substantially as concentration increases. This variability
in true concentration means there are no expected values for randomly selected samples, so
that the accuracy of testing methodology must be evaluated indirectly through the recovery
of standards and spikes. As a result, it has been difficult for environmental laboratories to
satisfactorily apply industrial QC techniques.

There are two possible approaches to the  solution of the problem of variation in the true
concentration level; either use  of a  statistic that  is not  sensitive to this  variation or
application of the industrial techniques within restricted concentration ranges. Obviously,
the former should be preferred because it actually solves the problem and does not require
the development and maintenance of a series of charts for each parameter.

6.3.1 Quality Control Charts for Accuracy

Two replacements for the Shewhart X control chart have been suggested for evaluating the
recovery of a series of different standards or spikes.  One of these, a cusum chart using the
                                     6-2                 OCR(6/2001)byRYA-R10

-------
square of the difference between the observed and true values, is described in an EPA
Region VI QC manual (5). The other alternative uses the classic Shewhart technique to
evaluate the percent recovery instead of X. It is recommended that the percent recovery be
calculated as
                                       observed
                                ^=100	
                                        known

for standards, or
                                 observed - background
                                          spike
for recovery of spikes into natural water backgrounds. An example of the linear relationship
between percent recovery and the known concentration of standards and spikes is demon-
strated in the accuracy plots of a recent EPA method study report on analysis of mercury (6).
Both approaches are being used on a daily basis by various environmental laboratories.

The data in table  6-1 were used  in the EPA Region VI manual (5) to illustrate the
development of a cusum chart. The actual data have been reordered here to appear in
ascending order of the known values.  Note that the mean and the range of the d2 values
increase with increasing concentration level, and this violates a basic premise for acceptable
control chart statistics. Because the percent recovery data do not show any such trend, it is
the recommended control chart statistic for controlling accuracy.

From the data in table 6-1, a Shewhart control chart for percent recovery can be calculated
in the following way:
Average percent recovery
                                       ?'•
                                  p = 1±1—
                                        23
                                       2,310
                                        23
                                     = 100.4
The standard deviation for percent recovery
                                          22
                                    6-3                OCR(6/2001)byRYA-R10

-------
                             /234,074-(2,310)2 723
                                       22
                          = V94.0751

                          =  9.70

Therefore, the upper control limit becomes the following:

                          UCL  = P + 3Sp

                                = 100.4 + 3(9.70)

                                = 129.5


                                Table 6-1
   ANALYSIS1 OF TOTAL PHOSPHATE-PHOSPHORUS STANDARDS, IN mg/1
                              TOTAL PO4 -P
Point
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
v f\^ • j Difference ,2
Known Obtained , a,
0.34
0.34
0.40
0.49
0.49
0.49
0.50
0.50
0.50
0.52
0.66
0.66
0.67
0.68
0.83
0.98
1.3
1.3
1.6
2.3
2.3
3.3
4.9
0.33
0.34
0.40
0.49
0.49
0.63
0.47
0.53
0.56
0.59
0.70
0.60
0.65
0.65
0.80
0.75
1.2
1.3
1.7
2.3
2.4
3.3
4.6
0.01
0.00
0.00
0.00
0.00
-0.14
0.03
-0.03
-0.06
-0.07
-0.04
0.06
0.02
0.03
0.03
0.23
0.10
0.00
-0.10
0.00
-0.10
0.00
0.30
0.0001
0.0000
0.0000
0.0000
0.0000
0.0196
0.0009
0.0009
0.0036
0.0049
0.0016
0.0036
0.0004
0.0009
0.0009
0.0529
0.0100
0.0000
0.0100
0.0000
0.0100
0.0000
0.0900
Percent
Recovery
97
100
100
100
100
129
94
106
112
113
106
91
97
96
96
77
92
100
106
100
104
100
94
. »,'
9,409
10,000
10,000
10,000
10,000
16,641
8,836
11,236
12,544
12,769
11,236
8,281
9,409
9,216
9,216
5,929
8,464
10,000
11,236
10,000
10,816
10,000
8,836
         Totals
   2,310
234,074
 'Using a calorimetric method with persulfate digestion.
                                 6-4
OCR(6/2001) by RYA-R10

-------
and the lower control limit becomes

                      LCL   =  100.4-29.1

                              =  71.3

The completed control chart is shown in figure 6-2.

Following normal procedures, the control chart must indicate the conditions under which
it was developed; i.e., laboratory name, parameter, method of analysis, date of preparation,
and any other information unique to the initializing data, such as range of concentration and
identification  of analyst(s).  A control chart is not generally applicable under other
conditions.

To verify the control chart, the initializing data should be checked to be sure that none of
the values exceeds these new control limits. In addition, if its distribution is proper, about
68 percent of the initializing data should fall within the interval P  ± Sp.  It has been
suggested that the control chart is not valid if less than 50 percent of the initializing data
falls within this interval.

In applying the control chart, either of the following two conditions would indicate  an out-
of-control situation:

   a.    Any point beyond the control limits

   b.    Seven successive points on the same side of the value P  of the central line

When an out-of-control situation occurs, analyses  must be stopped until the problem has
been identified and resolved, after which the frequency should be increased for the next few
percent-recovery QC checks. The problem and its solution must be documented,  and all
analyses since the last in-control point must be repeated or discarded.
                140i-
             2  120
             O
             U
             I  100
             t-
             UJ
             g  80
             ui
             o-
                60
                                                                      UCL
          .. P
                                                                      LCL
                                      10        15
                                     SAMPLE NUMBER
20
                                                                   25
              Figure 6-2.  Shewhart control chart for percent recovery data.
                                       6-5
 OCR(6/2001) by RYA-R10

-------
A final note of caution regarding use of a single percent-recovery P control chart over a
broad concentration range is necessary. As noted earlier for the analysis of mercury, a good
linear relationship of the form

                         X = P (known concentration) + K

where K is a constant,  seems appropriate for many parameters.  However, to justify use of
a single percent-recovery control chart, K must be small enough relative to the P (known
concentration) term that it has little or no practical effect upon the value of X. This will
usually be true for moderate or high concentration levels, but may not be true at very low
concentration levels.  As a result, for some parameters it may be necessary to develop a
separate percent-recovery or Shewhart X chart for each standardized low concentration
level sample.

6.3.2 Quality Control Charts for Precision

Because the characteristics of the range statistic change as concentration changes, two
alternatives to Shewhart's R chart have been used in environmental laboratories to evaluate
the precision of routine sample analyses.

One alternative is a cusum chart using the sum of the squared difference between duplicate
determinations on randomly selected routine samples (5). Because the range R for duplicate
analyses is equal to the difference between them, the cusum statistic equals the sum of
squared ranges £ R2. However, if R changes significantly as  concentration level changes,
then R2 is affected even more and, therefore, is not as good a criterion for judging whether
precision of the system is within acceptable limits.

The other alternative uses a chart similar to the R chart, but the chart statistic is either the
percent relative standard deviation (IOOS/X), the coefficient of variation (CV or SIX or the
industrial statistic /. For the duplicate determinations A and B, I equals the absolute value
of their difference divided by their  sum,  or  \A  - B\I(A + B), and can be  shown to be
equivalent to the other two statistics:

                            100(CV)=   100 =
                                            A.

                                              R /V2~
                                     =  100
                                     =  100
                                            (A  + B)/2

                                             2      R
                                                    + B)

                                         200  \A - B\
                                           i (A  + B}
                                         2007
For the sake of computational ease, / seems to be a logical alternative to R.
                                     6-6                OCR(6/2001)byRYA-R10

-------
The next concern is whether/is independent of changes in concentration level. Based upon
experience with duplicates on routine samples taken during the last 2 years by EPA Region
VII, / appears to decrease substantially as concentration increases.  In recognition of this
possible dependency, control charts for / should only be developed from and applied to
results within a limited concentration range. Note that control charts for R could be applied
under similar limitations.

As an illustration of the concentration dependency of these precision statistics, table 6-2
provides estimates of R and/for different concentration ranges of three parameters. These
parameters were selected because approximately 100 sets of duplicates were available that
were well distributed over a reasonably broad concentration range. The ranges for the sum
of duplicates A + B used in table 6-2 were selected because they were convenient and the
data tended to be well distributed among them.  Data judged to be out of control were
discarded before any calculations were made.

Table 6-2 indicates the concentration dependence of both the range R and the industrial
statistic / for these three parameters. Because / is not independent of concentration and is


                                   Table 6-2
 ESTIMATES OF THE RANGE R = \A-B\) AND THE INDUSTRIAL STATISTIC [/ =
 \A - B\I(A + B)]  OF THREE DIFFERENT PARAMETERS FOR VARIOUS CON-
 CENTRATION RANGES1
Parameter
BOD, 5-day (mg/1)






Chromium (Mg/1)





Copper (Mg/1)





„ ,, No. of
Range of „ ,,
A+B Sets of
Duplicates
2 to <20
20 to <50
50to<100
100to<300
300 to <600
600 to <2,000
2,000 up
10to<20
20 to <50
50to<100
100to<300
300to
-------
                                  Table 6-3
          SHEWHART UPPER CONTROL LIMITS (UCL) AND CRIT-
          ICAL RANGE Rc VALUES FOR THE  DIFFERENCES  BE-
          TWEEN DUPLICATE ANALYSES WITHIN SPECIFIC CON-
          CENTRATION RANGES FOR THREE PARAMETERS1
Parameter
BOD, 5-day (mg/1)






Chromium (Mg/1)





Copper (jj.g/1)





Concentration
Range2
lto<10
10to<25
25 to <50
50to<150
150to<300
300to
-------
Table 6-3 also contains a critical range Rc column. Because the data from EPA Region VII
were almost always whole units with only a very occasional half unit reported, the Rc. value
is the UCL value rounded to the nearest whole unit at higher concentration levels and to the
nearest half unit for the lowest concentration level. However, there is an exception to this
rule among the low-concentration Rc values for copper that demonstrates an advantage
beyond the simplicity of using such tables. The UCL value for copper at 25 to 50 |j,g/l is
inconsistent with the UCL values for adjacent concentration levels,  and the Rc value has
been adjusted to resolve this inconsistency. Without the table, such inconsistencies could
very easily go unnoticed.

The examples in table 64 illustrate how to use the Rc values in table 6-3. This technique,
consisting of the development and use of a table of critical-range Rc values at different
concentration levels, is recommended to control precision. Normal control chart procedures
should be followed as in section 6.3.1 regarding identification and verification of the table.
The table  should  be updated periodically as additional, or more current, data become
available,  or whenever the basic analytical system undergoes a  major change.  If any
difference between duplicate analyses exceeds the critical-range value for the appropriate
concentration level, then analyses must be stopped until the problem is identified and
resolved, and the frequency should be increased for the next few precision checks. After
resolution, the problem and its solution must be documented, and all analyses since the last
in-control  check must be repeated or discarded.

6.4 Recommended Laboratory Quality Assurance Program

A minimum laboratory quality assurance  program should include control procedures for
each parameter as described in the following sections.

6.4.1 Standard Curves

A new standard curve should be established with each new batch of reagents, using at least
seven concentration levels.

6.4.2 Quality Control Checks for Each Analytical Run

With each batch of analyses, the following tests should be run:

  a.    One blank on water and reagents


                                   Table  6-4
        CRITICAL RANGE VALUES FOR VARYING CONCENTRATION
                                    LEVELS
Parameter
BOD (mg/1)
Chromium (jJ.g/1)
Copper (jj.g/1)
Duplicates
20 and 24
60 and 75
46 and 5 1
R
4
15
5
RC
6
12
15
R
-------
  b.   One midpoint standard

  c.   One spike to determine recovery

  d.   One set of duplicate analysis

The results from b through d should be compared with previous in-control data by using the
appropriate technique recommended in section 6.3.

6.4.3 Interlaboratory QC

An interlaboratory QC program would require each laboratory to do the following:

  a.   Analyze reference-type samples to provide independent checks on the analytical
       system.  These  may be available from EPA as QC samples,  from the National
       Bureau of Standards as standard reference materials, or from commercial sources.
       If performance  limits are not provided, the results should fall within the routine
       limits of each laboratory for a standard at a level comparable to the specified true
       value.

  b.   Participate in performance evaluation and method studies as available from EPA,
       from the American Society for Testing and Materials, and from other agencies.

6.5 Outline of a Comprehensive Quality Assurance Program

In the following discussion the symbols used  represent the results of analysis according to
the scheme:

           Al  = first replicate of sample A
           A2  = second replicate of sample A
           B  = sample taken simultaneously with sample A
          BSF  = field spike into sample B
          BSL  = laboratory spike into sample B
          DF  = field spike into distilled water
          DL  = laboratory spike into distilled water
            T = true value for all spikes
The laboratory spikes BSL and DL are the only analyses that may not be necessary.  All other
analyses must be done simultaneously.

6.5.1 Steps for the Field Personnel

A comprehensive quality assurance program would include the following steps for each
parameter in the monitoring study:

  a.   Take independent simultaneous samples A and B at the same sampling point.
       Depending on the parameter, this might involve side-by-side grab samples or
       composite samplers mounted in parallel.
                                    6-10               OCR(6/2001)byRYA-R10

-------
  b.    Split sample A into the equal-volume samples Al and A2

  c.    Split sample B into equal volumes and add a spike T to one of them; the latter
        sample becomes sample BSF.   As with all spikes,  the  addition of T should
        approximately double the anticipated concentration level.

  d.    Add the same spike Tto a. distilled water sample furnished by the laboratory and
        designate this sample as DF.

These QC samples must be treated in the same way as routine samples; i.e., the volume, type
of container, preservation, labeling, and transportation must be the same for all.

6.5.2 Steps for the Laboratory

The laboratory personnel should perform the following steps for quality assurance:

  a.     Analyze the blank and midpoint standard recommended in section 6.4. If results are
        unsatisfactory, resolve problems before continuing.

  b.    Analyze sample  DF.  If the percent recovery of T is unsatisfactory (see section
        6.3.1), create a similarly spiked, distilled-water sample DL and analyze to test for
        a systematic error in the laboratory or fundamental problems with the spike. If the
        percent recovery  of T from DL is satisfactory, any systematic error occurred before
        the samples reached the laboratory.

  c.    Analyze samples  B and BSF. IfB is below the detection limit, or if B is greater than
        lOror less than O.IT, disregard the remainder of this step and proceed to step d. If
        the percent recovery of T from BSF is unsatisfactory (see section 6.3.1), spike an
        aliquot of sample B the same way in the laboratory so that a similar recovery can
        be anticipated. Analyze this sample BSL to test for immediate sample interferences
        or a bad background result B.  If the percent recovery from BSL is satisfactory, then
        the interference  must  require a longer delay before  analysis,  or other special
        conditions not present in the laboratory, in order to have a noticeable effect upon
        recovery of the spike.

  d.    Analyze Al and A2.  If the absolute (unsigned) difference between these results
        exceeds the critical value (see section 6.3.2), then precision is out of control.

  e.    Calculate the absolute difference between Al  and B.  If it is unsatisfactory (see
        section 6.3.2), the field sampling procedure did not provide representative samples.

If initial results at each of the  laboratory steps were satisfactory, then the validity of the
related data has been indisputably established.  If results at any step are unsatisfactory,
resolution depends upon the problem identified. Laboratory problems may just require that
the  analyses be repeated,  but field problems will usually require new samples.  Figure 6-3
is intended to clarify the interdependence of the preceding laboratory steps b through e.

In figure 6-3 it must be noted that there is no way to identify additive sample interferences;
i.e., those that have an equal effect upon the background-plus-spike results (BSF or BSL) and
the  background result B.  Recovery of a spike will not show such interferences.
                                     6-11                OCR(6/2001) by RYA-R10

-------
             1100Df/7 WITHIN QC LIMITS?]
No
                                               Yes
I




|lOODt IT WITHIN QC LIMITS? | fl < DETECTION LIMIT. B <0.1 T. or fl > 1077
Yes
No
* +
SYSTEMATIC
ERROR IN
THE FIELD


ERROR IN THE ,
LABORATORY W, - *2|
OR BAD SPIKES WITHIN OC
LIMITS?

f
DATA Yes tA' " S| fc
""i" • ., ~ 	 UUITIIIM nr n
VAI in WIIHINUL u
-!H LIMITS? B


No

Yes

No
*
XCESSIVE
EVIATION
ETWEEN
REPLICATE
No ANALYSES

EXCESSIVE DEVIATION
BETWEEN SIMULTANEOUS
SAMPLES


No
1
I


100(flSF - fl)/r
WITHIN QC LIMITS?

No
100(SSL - BMT
WITHIN QC LIMITS?

Yes
1 '
IMMEDIATE MULTIPLICATIVE
SAMPLE INTERFERENCES
OR BAD BACKGROUND RESULT

MULTIPLICATIVE SAMPLE INTERFERENCES
THAT REQUIRE A LONG TIME PERIOD
OR SPECIAL CONDITIONS TO HAVE A
NOTICEABLE EFFECT UPON RECOVERY
OF THE SPIKE
              Figure 6-3. Procedure for evaluating QC data from a monitoring study.
                  6-12
OCR(6/2001)byRYA-R10

-------
Problems causing systematic errors that may occur in the field include the following:

  a.   Contaminated preservative, distilled water, or containers

  b.   Contamination by sampling personnel

  c.    Deterioration through excessive holding, time or use of an ineffectual preservation
       technique

  d.   Use of a bad field-spiking procedure

6.6 Related Topics

6.6.1 Advanced Laboratory Automation  and Its Effect on QC

Advanced laboratory automation systems under development analyze samples automatically
and use a control computer to interpret the resulting data and produce an analytical report.
The primary benefits of such a system are not only that the data-recording and calculation
errors common to manual analyses have been inherently eliminated, but also that extensive
QC can be accomplished quite easily and cheaply. The computer can be programed to
automatically recognize different kinds of QC samples and to establish or recall appropriate
control limits. Thus the QC overhead is reduced considerably and QC procedures previously
too costly or complex become practical.

As an example of a QC procedure that is considered impractical for manual use, regression
could be used to determine the relationship between concentration change and the accuracy
and precision statistics discussed earlier.  Using these relationships, very responsive, single
accuracy and precision charts could be developed for each parameter. As computer-assisted
analysis becomes common, automated laboratories will very likely replace the manual
procedures recommended earlier in this  chapter with evaluation criteria based upon
regressions.

6.6.2 Method Comparability (Equivalency)

Requirements for method comparability  are under development for proposed alternatives
to the methodology specified in Public Law 92-500, section 304(g). A final version of these
requirements should be  available from EPA at a later date.

6.7 References

1.  Shewhart, W. A., Economic Control of Quality of Manufactured Product, Van Nostrand
    Reinhold Co., Princeton, N.J. (1931).
2.  Duncan, A. J., Quality Control and Industrial Statistics, 3d Edition, Ch. 18, R. D. Irwin,
    Inc., Homewood, 111. (1965).
3.  Grant, E. L., and Leavenworth, R. S., Statistical Quality Control, 4th Edition, McGraw-
    Hill Book Co., Inc., New York (1972).
4.  Davies, O. L., and Goldsmith, P. L., Statistical Methods in Research and Production, 4th
    Edition, Hafner Publishing Co., New York (1972).
5.  Laboratory Quality Control Manual, 2d Edition, U.S. EPA, Region VI, Surveillance and
    Analysis Division, Ada, Okla. (1972).
                                    6-13               OCR(6/2001)byRYA-R10

-------
6.  Winter, J., Britton, P., Clements, H., and Kroner, R., "Total Mercury in Water," EPA
   Method Study 8, pp. 35-36, U.S. EPA, Office of Research and Development, EMSL,
   Cincinnati (Feb.  1977).
                                   6-14               OCR(6/2001)byRYA-R10

-------
                                   Chapter 7

                      DATA HANDLING AND REPORTING

7.1 Introduction

To obtain meaningful  data on water  quality,  the sample  collector must  obtain  a
representative sample and then deliver it unchanged for analysis. The analyst must perform
the proper analysis in the prescribed fashion, complete calculations, and convert results to
final form for permanent recording of the analytical data in meaningful, exact terms. These
results are transferred to a storage facility for future interpretation and use.

The following sections discuss processing of actual values, recording and reporting of data
in the proper way, some means of quality control of data, and the storage and retrieval of
data.

7.2    The Analytical Value

7.2.1 Significant Figures

The term "significant figure" is used, sometimes rather loosely, to describe a judgment of
the reportable digits in a result. When the j udgment is not soundly based, meaningful digits
are lost or meaningless digits are reported. On the other hand, proper use of significant
figures gives an indication of the reliability of the analytical method used.

The following discussion describes the process of retention of significant figures.

A number is an expression of quantity.  A figure or digit is any of the characters  0, 1, 2, 3,
4, 5, 6, 7,  8, 9, which, alone or in combination, serve to  express a number.  A significant
figure is a digit that denotes the  amount of the quantity in the particular decimal place in
which it stands. Reported analytical values should contain only significant figures. A value
is made up of significant figures when it contains all digits known to be true and one last
digit in doubt. For example, if a value is  reported as 18.8 mg/1, the 18 must be firm while
the 0.8 is somewhat uncertain, but presumably better than one of the values 0.7 or 0.9 would
be.

The number zero may or may not be  a significant figure depending on the situation.

Final zeros after a decimal point are always meant to be significant figures.  For example,
to the nearest milligram, 9.8 g is reported as 9.800 g.

Zeros before a decimal point with nonzero digits preceding them are significant. With no
preceding nonzero digit, a zero before the decimal point is not significant.

If there are no nonzero digits preceding a decimal point, the zeros after the decimal point but
preceding other nonzero digits are not significant. These zeros only indicate the position
of the decimal point.

Final zeros in a  whole number may or may not  be  significant.   In a conductivity
measurement of 1,000 |j,mho/cm, there is no implication by convention that the conductivity
is 1,000 ± 1 |j,mho. Rather, the zeros only indicate the magnitude of the number.
                                       7-1              OCR(6/2001)byRYA-R10

-------
A good measure of the significance of one or more zeros interspersed in a number is to
determine whether the zeros can be dropped by expressing the number in exponential form.
If they can, the zeros may not be significant.  For example, no zeros can be dropped when
expressing  a weight of 100.08 g in exponential form; therefore the zeros are significant.
However, a weight of 0.0008 g can be expressed in exponential form as 8 x 10~4 g, so the
zeros are not significant. Significant figures reflect the limits in accuracy of the particular
method of analysis. It must be decided whether the number of significant digits obtained
for resulting values is  sufficient for interpretation purposes.  If not, there is little that can be
done within the limits of the given laboratory operations to improve these values. If more
significant  figures are needed, a further improvement in method or selection of another
method will be required.

Once the number of significant figures obtainable from atype of analysis is established, data
resulting from such analyses are reduced according to set rules for rounding off.

7.2.2 Rounding Off  Numbers

Rounding off of numbers is a necessary operation in all analytical areas. It is automatically
applied by the limits of measurement of every instrument and all glassware. However, when
it is applied in chemical calculations incorrectly or prematurely, it can adversely affect the
final results. Rounding off should be applied only as described in the following sections.

7.2.2.1 Rounding-Off Rules

If the figure following those to be retained is less than 5, the figure is dropped, and  the
retained figures are kept unchanged. As an example, 11.443 is rounded off to 11.44.

If the figure following those to be retained is greater than 5, the figure is dropped, and the
last retained figure is  raised by 1. As an example,  11.446 is rounded off to 11.45.

If the figure following those to be retained is  5, and if there are no figures other than zeros
beyond the five, the figure 5 is dropped, and the last-place figure retained is increased by
one if it is an odd number or it is kept unchanged if an even number. As an example, 11.435
is rounded off to 11.44, while 11.425 is rounded off to  11.42.

7.2.2.2 Rounding Off Arithmetic Operations

When a series of numbers is added, the sum should be rounded off to the same number of
decimal places as the  addend with the smallest number of places. However, the operation
is completed with all decimal places intact, and rounding  off is done afterward.  As an
example,

                                      11.1
                                      11.12
                                    +11.13
                                      33.35

The sum must be rounded off to  33.4.
                                       7-2              OCR(6/2001)byRYA-R10

-------
When one number is subtracted from another, rounding off should be completed after the
subtraction operation, to avoid possible invalidation of the operation.

When two numbers are to be multiplied, all digits are carried through the operation, then the
product is rounded off to the number of significant digits of the multiplier with the fewer
significant digits.

When two numbers are to be divided, the division is carried out on the two numbers using
all digits. Then the quotient is rounded off to the number of significant digits of the divisor
or dividend, whichever has the fewer.

When a number contains n significant digits, its root can be relied on for n digits, but its
power can rarely be relied on for n digits.

7.2.2.3 Rounding Off the Results of a Series of Arithmetic Operations

The preceding rules for rounding off are reasonable for most calculations; however, when
dealing with two nearly  equal numbers, there is a danger of loss of all significance when
applied to a series of computations that rely on a relatively small difference in two values.
Examples are calculation of variance and standard deviation. The recommended procedure
is to carry several extra  figures through the calculations and then to round off the final
answer to the proper number of significant figures.

7.3 Glossary of Statistical Terms

To clarify the meanings of statistical reports and evaluations of water quality data, the
following statistical terms are introduced. They are derived in part from usage (1,2) of the
American Society for Quality Control.

Accuracy - The difference between an average value and the true value when the latter is
known or assumed.

Arithmetic mean - The arithmetic mean (or average) of a set of n values is the sum of the
values divided by n:
                                        I x,
Bias - A systematic error due to the experimental method that causes the measured values
to deviate from the true value.

Confidence limit, 95 percent - The limits of the range of analytical values within which
a single analysis will be included 95 percent of the time,

                             95 percent CL = X ± 1.96S

where CL is the confidence level and S is the estimate of the standard deviation.
                                       7-3              OCR(6/2001)byRYA-R10

-------
Constant - A nonvarying qualitative or quantitative characteristic of the population.

Geometric mean - A measure of central tendency for data from a positively skewed
distribution (log normal),

                              Xg=i}(XJ(X2)...(XJ
                                            n
                                           Z log*,-
                              X  = antilog
                                               n
Interference - A biological or chemical attribute of a test sample that positively or
negatively offsets the measured result from the true value.  If interference that is not
segregated and identified is present, it enlarges or reduces the method bias.

Median - Middle value of all data ranked in ascending order. If there are two middle
values, the median is the mean of these values.

n - The number of values Xt reported for a sample.

N- The total number of values^ of the entire population or universal set of data.

Population - The total set of units, items, or measurements under consideration.

Precision - Relative  to the data from a single test procedure,  the degree  of mutual
agreement among individual measurements made under prescribed conditions.

Precision data - Factors that relate to the variations among the test results themselves; i.e.,
the scatter or dispersion of a series  of  test results, without assumption of any prior
information.

Range - The difference between the highest and lowest values reported for a sample.

Relative deviation (coefficient of variation) - The ratio of the standard deviation S of a set
of numbers to their mean X expressed as percent. It relates standard deviation (or precision)
of a set of data to the size of the numbers:
                           CV=RD(percent) = 100 =
                                                   J\.
Relative error - The mean error of a series of measured data values as a percentage of the
true value Xt,
           t,
                                              X - X
                           RE(percent) = 100	
                                               *(

Sample - Groups of units or portions of material, taken from a larger collection of units or
quantity of material, that provide information to be used for judging the quality of the total
collection or entire material as a basis for action on them or on their production processes.
                                       7-4              OCR(6/2001)byRYA-R10

-------
Series - A number of test results with common properties that identify them uniquely.

Skewness - A measure of the asymmetry of a frequency distribution,

                                K =
This measure is a pure signed number. If the data are perfectly symmetrical, the skewness
is zero. If K is negative, the long tail of the distribution is to the left. If K is positive, the long
tail extends to the right.

Standard deviation - The square root of the variance of the universe,
Standard deviation estimate - The most widely used measure to describe the dispersion
of a set of data. Normally X ± S will include 68 percent, andX ± 2S will include about 95
percent of the data from a study:
                                 I*HX*
Standard deviation, single analyst - A measure of dispersion for data from a single analyst
that is calculated here using an equation developed by Youden for his nonreplicate study
design (3),
where D = the difference in paired values obtained from a single analyst.

Universe - The total set of items or measurements.

Variable - An experimentally determined estimate denotedXfor a particular quality or trait
of the population.

7.4 Report Forms

The analytical information reported should include the measured parameters; the details of
the analysis such as burette readings, absorbance, wavelength, normalities of reagents,
correction factors, blanks; and the reported data values.
                                      7-5              OCR(6/2001)byRYA-R10

-------
To reduce errors in manipulation of numbers a general rule is to reduce handling and
transposition of data to an absolute minimum. Ideally, a report form includes preliminary
information  about the sample and its analysis, and the same form is used for the final
entering of data into a computer; however, such report forms are not yet common.  Rather,
a variety of methods is used to record data.

7.4.1 Loose Sheets

Reporting of data onto loose or ring-binder forms is a means of recording data that allows
easy addition of new sheets, removal of older data, or collection of specific data segments.
However, the easy facility for addition or removal also permits loss or misplacement of
sheets, mixups in date sequence, and ultimately questionable status of the data for formal
display or presentation as courtroom evidence.

7.4.2 Bound Books

The  use of bound books is an improvement in data recording  that tends to result in a
chronological sequence of data insertion. Modification beyond a simple lined book improves
its effectiveness with little additional effort. Numbering of pages encourages use of data in
sequence and also aids in referencing data through a table of contents ordered according to
time, type of analysis, kind of sample, and identity of analyst.

Validation can be  easily accomplished by requiring the analyst to date and  sign each
analysis on the day completed. This validation can be strengthened further by providing
space for the laboratory supervisor to witness the date and the completion of the analyses.

A further  development of the bound notebook is the commercially available  version
designed for research-type  work. These notebooks are preprinted with book and page
numbers, and spaces for title of project, project number, analyst signature, witness signature,
and dates.  Each report sheet has a detachable duplicate sheet that allows up-to-date review
by management without disruption of the notebook in the laboratory. The cost of research-
type bound notebooks is about four times that of ordinary notebooks.*

Use  of bound notebooks has been limited to research and development work where an
analysis is part of a relatively long-term project, and where the recording in the notebook
is the prime disposition of the data until an intermediate or final report is written.

However, bound notebooks can and should be used in routine analytical laboratories such
as those concerned with water quality. The need for repeated information on sampling and
analyses can be answered by use of preprinted pages in the bound notebooks.

7.4.3 Preprinted Report Forms

Most field laboratories  and installations repetitively analyzing fixed parameters develop
their own system of compiling laboratory data that may include bound notebooks, but a
means of forwarding data is also required. Usually, laboratories design forms to fit a related
* Scientific Notebook Co., 719 Wisconsin Ave., Oak Park, 111. 60304.

                                       7-6              OCR(6/2001)byRYA-R10

-------
group of analyses or to report a single type of analysis for a series of samples. As much
information as possible is preprinted to simplify use of the  form.  With loose-sheet,
midticopy forms (using carbon or NCR paper) information can be forwarded on the desired
schedule while also allowing retention of data in the laboratory.  Still, the most common
means for recording data in rough form are internal bench sheets or  bound books. The
bench sheet or book never leaves the laboratory but serves as the source of information for
transfer of data to appropriate report forms. (See fig. 7-1.)

In most instances the supervisor and analyst wish to look at the data from a sampling point
or station in relation to other sampling points or  stations on a particular river or lake. This
review of data by the supervisor prior to release is a very important part of the QC program
of the laboratory; however, such reviews are not easily accomplished with bench sheets. For
review purposes, a summary sheet can be prepared that displays a related group of analyses
from a number of stations such as shown in figure 7-2. Because the form contains space for
all of the information necessary for reporting data, the completed form can also be used to
complete the data forms forwarded to the computer storage and retrieval system.

The forms used to report data to storage  systems provide spaces  for identification of the
sampling point, the parameter code, the type of analysis used, and the reporting terminology.
Failure to provide the correct information can result in rejection of the  data, or insertion of
the data into incorrect parameter fields. As sample analyses are completed, the data values
are usually reported in floating decimal form along with the code numbers for identifying
the parameter data fields and the sampling point data fields. Figure 7-3 shows an example
of a preprinted report form used for forwarding  data to keypunch.

7.4.4 Plastic-Coated Labels and  Forms

A recent addition to good sample handling and data management is the availability of
plastic-coated (blank or preprinted) labels, report forms, and bound report books.   These
materials are waterproof, do not disintegrate when wet or handled, can  be written on while
wet, and retain pencil or waterproof ink markings though handled when wet.

7.4.5 Digital Readout

Instrumental analyses, including automated, wet-chemistry instruments such as the Tech-
nicon AutoAnalyzer, the atomic absorption spectrophotometer,  the  pH meter, and the
selective electrode meter, provide digital readout of concentrations, which can be recorded
directly onto report sheets without further calculation. Programmed calculators can be used
to construct best-fit curves,  to perform  regression analyses,  and to perform  series  of
calculations leading to final reported values.

7.4.6 Keypunch Cards and Paper Tape

Because much of the analytical data generated  in laboratories  is  first recorded on bench
sheets, then transferred to data report forms, keypunched, and manipulated on small terminal
computers (or manipulated and stored in a larger data storage system),  there is a danger of
transfer error that increases with each data copy.  The analyst can reduce this error by
recording data directly from bench sheets onto punch cards that can be retained or forwarded
immediately to the  data  storage system.  Small hand-operated keypunch machines are
available.
                                       7-7              OCR(6/2001)byRYA-R10

-------
NL-C-88
Spectrographic Analyses Bench Data
Sample* 	Date	
                                                        TDS
              ml rnnr tn
                                       ml. Factor ,
                                                                            Test Count
                                                                                                Sec
                                                                       3.
Count
1 7n
^ CA
"» %
4 B
s P
A FP
7 M«
8 Mn
0 Al
m Bf
M Oi
17 AE
13 Ni
id rv.
i<; PK
16 Cr
17 V
18 Ra
19. Sr
Rerun Count



















PPM (Mg/l) PPB
Av. In Cone. Less Orig.
Count Sample Than Sample
n r
rn r
n r
n r
n
rn
rn
rn
i i
rn
rn
rn
r~i
r-i
r~i
n
n
r— i
m
















































































                                            Figure 7-1. Example of bench sheet.
                                                            7-8  OCR(6/2001)byRYA-R10

-------
                                MINERALS ANALYSES OF ZONE B, OHIO RIVER SAMPLES, CONC, mg/1.
STATION
Ohio at Iron ton
Ohio at Greenup Dam
Ohio at Portsmouth
Scioto at Lucasville
Ohio at Maysville
Ohio at Meldahl Dam
Little Miami at Cincinnati
Ohio at Cincinnati
Licking at 1 2th Street
Ohio at Miami Fort
Ohio at Markland Dam
Kentucky at Dam I
Ohio at Madison
Great Miami at Eldean
Great Miami at Sellars Road
Great Miami at Liberty-
Fairfleld Road
Great Miami at American
Materials Bridge
Whitewater at Suspension
Great Miami at Lawrenceburg
(Lost Bridge)
Storet
Number
200152
200001
200139
381710
200153
383070
380090
380037
200523
383072
200521
200522
174304

383047
383015
383007

383071
Date



















Alkalinity



















Hardness



















Chloride



















Sulfate



















Fluoride



















SOLIDS
Total



















Diss.



















Susp.



















Figure 7-2.     Example of summary data sheet.
                                                          7-9  OCR(6/2001)byRYA-R10

-------
LABORATORY BCNCH DATA
STATION DHIQNATION
Y«. I MO. | DAY
OATC OF COMPOSITE SAMPLE
ITEM
1
1 1
ITEM
u

1 1 1 1 1 1
Peal Coliform
i i i i i i a
1 1 1 1 1 1

1


1

1


1

1 1


1 1


1 1 1
4 If
MF/100
1 1 1

ITEM
1
1 1

ITEM
1
1 1
ITEM
I
1 1
ITEM
1
1 1
ITEM
t
1 1
ITEM
1 1
ITEM
t
LL
NHj-N t Org N
i I i i I i •
1 1 1 1 1 1

NH3-N

i 1 1 1 1 1
NOj-N + NO3-N
1 1 1 1 1 1 0
1 1 1 i 1 1
P. Tot»l
MINI
P, Soluble
t i i i i i t
1 1 1 1 1 1
TOC
1111118
1 1 1 1 I 1
Phenol
1 1 1 I 1 1 0
1 1 1 1 1 1


1



1


1

1


1


1


1


1



1


1

1 1


1 1


1


1 1


1 1



1 1


1 1

1 1


1 1


1 1


1 1
-a|fl-
1 1 1

mg/l
1 1 1
mg/1
1 1 1
mg/1
1 1 1
mg/1
1 1 1
mg/1

1 1 1
^
\ \ \

ITEM
1
LL
Cyanide
1111110
1 1 1 1 1 1


1


1 1


1 1
mg/1

1 i 1

COMPUTER CODED DATA
ITATION COOi HHIAL V*.
1 1 1 .1 1
PArtAMETIR COOf
|3h|6|.|6

|3|l |6|?|9
31-39

|0|0|6|3|S
4347

|0|0|6|,[0
ss-st

|0|0|6|3|0
•7-71

|0|0|6J6|5

|fl| 0|6 |6 |6

|0 | 0 | 6 | 8JO

|3|2|7!3|0
5S5»

|OJO|7|2|0
«7-71
1

II

II


II


II

II

II

II

II

II


Jl

1 1
1
VALUI
1 1

IJ
3*-3»

1 1
4H1

-Ur

i 1
7J-79

1 1

1 1

J-L

1 1
60-C3

1 1
72-79



1

1


1


1

1

1

1

1

1


1

j
V

1

1


1


1

1

1

1

1

1


1

MO. DAY
i,; i ; i
i : i ; i
13-16
fXPONINT MMKS
000
000
000
000
COLUMN H (BLANK)
CMO.
0000
500
DDD
ann
000
COLUMN M (BLANK)
CHO,
0000
                  Figure 7-3. Example of STORET report form.
7.4.7 STORET-Computerized Storage and Retrieval of Water Quality Data

Because of their ability to record, store, retrieve, and manipulate huge amounts of data, the
use of computers is a natural outgrowth of demands for meaningful interpretation of the
great masses of data generated in almost all technical activities.

In August 1961, numerous ideas were brought together concerning the basic design of a
system called STORET for storage and retrieval of water pollution control  data.  A
                                     7-10
OCR(6/2001) by RYA-R10

-------
refinement of this system is now operated by the Technical Data and Information Branch,
Division of Applied Technology, Office of Water Programs, EPA.

This is a State/Federal cooperative activity that provides State water pollution control
agencies with direct, rapid access into a central computer system for the storage, retrieval,
and analysis of water quality-control information.

If properly stored, the data can be retrieved according to such descriptors as the point of
sampling, the date, and the specific parameters stored, or all data at a sample point or series
of points can be extracted as a unit.

Full details on use  of the  STORET system are  given in the recently revised STORET
handbook (4).

7.4.8 Automated Laboratory Systems

The use of digital readout, keypunch cards, and paper tape have been overshadowed by the
development of customized, fully automated online computer systems that make measure-
ments, calculate results, perform quality control, and report analytical data simultaneously
from a full range of laboratory instruments. (See fig. 7-4.) Such systems can contain the
following functions:

  a.   Manual or  automatic  sampling and testing  of a series of samples, standards,
       replicates, and check samples

  b.   Detection of the measurement signals from the series of samples

  c.   Conversion of  signals to  concentrations, generation of a standard curve, and
       calculation of sample values in final units

  d.   Calculation of the deviation and recovery values of the results and indication of
       acceptance or nonacceptance based on limits established by the analyst

  e.   Provision of the output in a form designated by the analyst: dial, paper recording
       chart, digital readout, cathode ray tube, or printed report form

The degree of hands-on operation required in the  system is specified by the analyst.

If an automated system is properly designed and operated, most calculation and transposi-
tion errors are avoided and the proper level of quality control is  automatically exerted.
Laboratory automation systems for water analyses are being developed and coordinated by
EMSL-Cincinnati for use in a number of EPA laboratories (5).

7.5 References

1.  "Guide for Measure  of Precision and Accuracy," Anal. Chem.  33, 480 (1961).
2.  "Glossary  of General Terms Used in Quality Control," Quality  Progress, Standard
    Group of the Standards Committee, American Society for Quality Control, 11(1), 21-2
    (1969).
3.  Youden, W. J., Statistical Techniques for Collaborative Tests, Association of Official
    Analytical Chemists, Washington, D.C. (1967).
                                      7-11              OCR(6/2001) by RYA-R10

-------
           c
  START
               OPERATOR
               INPUTS
                   I
       MEASURE TIMING,
       BASELINE, AND
       CALIBRATION STANDARDS
                   I
           MEASURE SAMPLES.
           CHECK STANDARDS,
           REPLICATES, SPIKES,
           AND BLANKS
                   I
         COMPUTE AND DISPLAY
         INTERIM RESULTS AND
         STATISTICS
  NO
ALL
SAMPLES
DONE?
                                        ALERT
                                        OPERATOR
                                        TO
                                        ACTION
Figure 7-4.     Flow chart of the sequence of events
              during a controlled series of laboratory
              measurements.
                       7-12
                        OCR(6/2001) by RYA-R10

-------
"STORET, EPA's Computerized Water Quality Data Base, the Right Answer," U.S.
EPA, Office of Water and Hazardous Materials, Washington, D.C. (1977).
Budde, W. L., Almich, B. P., and Teuschler, J. M., The Status of the EPA Laboratory
Automation Project, EPA-600/4-77-025, U.S. EPA, Office of Research and Develop-
ment, EMSL, Cincinnati (1977).
                                 7-13             OCR(6/2001)byRYA-R10

-------

-------
                                    Chapter 8

        SPECIAL REQUIREMENTS FOR TRACE ORGANIC ANALYSIS
8.1 Introduction

The high sensitivity of the instrumentation used in trace organic chemical analysis and the
low concentration of organic compounds being investigated dictate that special attention
must be given to this analytical field. Contamination of the sample from any possible source
must be  diligently guarded against, and interferences in the sample must be carefully
controlled. Strict attention to method and highly refined technique are required to produce
valid quantitative and qualitative results.

Two manuals have been developed by the Environmental Monitoring and Support Labora-
tory (EMSL) in Cincinnati covering the two broad areas of quantitative and qualitative trace
organic analysis. Reference 1 contains both general and specific quality assurance programs
designed to insure the production of acceptable measurements when using the methods
contained in that manual. Reference 2 contains specific, detailed quality assurance programs
covering both  the performance of the instrument and the interpretation of the resulting
spectra.

This chapter is organized with three sections devoted to general material applicable to all
organic analysis, and one section abstracting highly specialized materials from each of the
two manuals cited.

8.2 Sampling and Sample Handling

Regardless of the intent, all numbers generated by a water quality laboratory are ultimately
represented as  the concentration levels in the sample matrix at the time of collection. Such
numbers tend,  automatically, to endorse  the sample collection, preservation, and shipment
procedures. Thus, quality assurance programs limited to the care of the sample beginning
with its receipt by the laboratory are inadequate. The laboratory must share responsibility
for the preservation and shipment of all samples that it will accredit with concentration
values.  Two  approaches are available that  will generally protect the laboratory from
generating numbers that may not reflect actual conditions of the sample at the time of
collection. The best, but usually least practical solution, is for the laboratory personnel to
collect all samples. The alternative is forthe laboratory to adoptapolicy of sample rejection
based on minimum standards of sample identification and age.  Guidelines for establishing
these standards are discussed in this section. It is recommended that copies  of this material
be supplied to all sample collectors along with an understanding of the specific policy of the
laboratory toward rejecting samples that do not meet these criteria.

In all of the cases to be discussed, it is the responsibility of the project director to  (a)
coordinate his  sampling, preservation, and shipment with the laboratory, (b) obtain clean
sample containers from the laboratory, (c) provide  adequate sample identification and
compositing instructions, and (d) provide duplicates  and blanks as  required by the
laboratory. Additional prearrangements should be made with the laboratory if sample
splitting is desired, to  create separate supernatant and settleable matter samples, or if
calculations on a wet-weight basis in addition to the standard dry-weight calculations are
desired.
                                       8-1              OCR(6/2001)byRYA-R10

-------
Bottles and caps are to be supplied by the laboratory because rigorous cleaning is required
even for new bottles. New and recycled bottles should be washed as described in reference
1. Samples received in bottles of unknown origin or questionable cleanliness should be
rejected by the laboratory. For water samples that are to be analyzed by solvent extraction
methods, narrow-mouthed, screw-cap bottles such as Boston rounds are preferred because
they have less tendency to leak and are easy to handle in the laboratory. One-liter bottles
are generally more expensive than  quart bottles, thus most laboratories use the clear 32-oz
containers. The bottle should be sealed with Teflon (du Pont) lined caps. For water samples
of intermediate pH, aluminum-foil-lined caps may be used when Teflon (du Pont) is not
available. Oil and grease samples can be collected with Polyseal caps - the conical liner pro-
vides an excellent seal against pressure during shipment and sample extraction. Screw-cap,
widemouthed glass bottles are preferred for sediment samples because they are easier to
handle in the field and in the laboratory. Precleaned, 16- or 32-oz bottles with Teflon (du
Pont) or foil-lined screw-cap closures should be provided by the laboratory.  Masking tape
or other suitable labels  should be applied to the dry bottles.

Sampling for purgeable organics requires special consideration and equipment. The sample
container should consist of a 45-ml, screw-cap vial fitted with a Teflon (du Pont) faced, sili-
cone septum. * The vials, septa, and caps should be washed in hot detergent water and thor-
oughly rinsed with tapwater and organic-free water, then dried at 105°C for 1 h. The vials
should then be cooled to room temperature in a contaminant-free area. When cool, the vials
should be sealed with the septa, Teflon (du Pont) side down, and screw cap and maintained
in this sealed  condition until filled with sample.

The bottles used for collecting water samples for solvent extraction should not be overfilled
or prerinsed with sample before  filling because oil and other material that  can cause
erroneously high results may remain in the sample bottle after rinsing.  Bottles should be
filled with sample to about 90 percent of capacity, and  the level should  be marked to
determine if leakage occurs. Because full sample bottles are difficult to pour from during
extractions, complete filling should be avoided. In the collection of sediment samples,
nonrepresentative debris such as large stones or wood should be discarded.

Multiple samples  are usually required for purgeable organics analysis because of leakage
and because the measurement process is destructive to the sample.  All vials  should be
identified with waterproof labels. The water sample vials are filled to overflowing from a
bubble-free source so that a convex meniscus is formed at the top.  They are sealed by
carefully placing the septum, Teflon (du Pont) side down on the opening of the vial and
screwing the cap firmly in place.

Shipment and receipt of samples must be coordinated with the laboratory to minimize time
in transit because it is the prerogative of the laboratory to reject samples where delays in
shipment have caused them to age beyond acceptable holding times. To avoid the need to
resample, the sampler should determine in advance the most efficient and reliable form of
transportation for the samples.  All samples for organic analysis should arrive at the
laboratory on the same day collected, or should be shipped and maintained at less than 4°C
for arrival by the next day. The samples are usually shipped in insulated ice chests. Water
samples should be prechilled before packing to reduce the ice requirements during shipment.
*Vial and septum are available from Pierce Chemical Co., P.O. Box 117, Rockford, 111.
61105. Vial: No. 13074; septum: No. 12722.

                                       8-2              OCR(6/2001)byRYA-R10

-------
The bottles should be stabilized in the container with styrofoam and covered with ice. The
information needed to identify the samples should be attached to the outside of the ice chest.

A part of the quality assurance program for a laboratory must be the development of a clear
policy for accepting or rejecting samples. Because organic analyses are expensive in terms
of manpower and supplies, it is poor management to commit such resources just to obtain
data of questionable validity. Water samples that clearly have not met the preservation
criteria during shipment (e.g., in the case of a spill) should be accepted only if resampling
is impossible.  Results from such samples must be qualified in the laboratory report.

Upon receipt, the samples should be checked for adequate identification, sample temperature
or presence of ice in the chest,  and leakage.  Samples  for volatile organics  should be
checked for air bubbles, although it is extremely difficult to avoid the development of very
small bubbles regardless of the type of sample bottle employed. The samples are logged in
with the receipt time noted. Unless the condition of the samples fails to meet the criteria for
acceptance by the laboratory, required preservatives are added immediately and the samples
are all refrigerated.

The  laboratory staff should be alerted to the arrival of the samples, so that the required
analysis can begin as soon as possible. When sediment samples are to be reported on a wet-
weight basis, or when a water sample is to  be filtered or divided into supernatant and
settleable subsamples for separate analysis, the sample processing should begin promptly.

When analysis of water samples is to be restricted to the water phase only, filtration is
required.  This may be accomplished with 4.7-cm,  glass-fiber filter disks that have been
preextracted with acetone and allowed to dry.  The filter disks must not contain organic
binder.  The disks are placed on a membrane  filter holder and  up to a liter of water is
filtered. The filtrate is transferred without rinsing to a clean sample container and treated
as a normal, whole-water sample.

When separate reporting is required for the settleable and supernatant phases, the water
sample (or a large portion of it) is allowed to settle overnight in a closed glass container at
4°C. If phase separation occurs, the supernatant phase is carefully decanted or siphoned into
a graduated cylinder without disturbing the surface of the settled material.  After the volume
of supernatant is noted, the supernatant is filtered as just described into a clean glass
container and stored at 4°C.  The volume of settled material is determined either by marking
the slurry level on the side of the sample container for later calibration, or by transferring
the slurry without washing into a  calibrated vessel.  A  portion of well-mixed slurry is
removed for a determination of percent solids.  The remaining slurry is stored in a sealed
glass container at 4°C.

When both dry-weight and wet-weight results are required for sediments and  sludges, a
percent-solids determination should be performed  soon  after receipt of the sample.  A
representative portion (ideally 10 to 25 g)  of well-mixed sample is weighed into a tared
Erlenmeyer flask and dried at 105°C to  a  constant weight.  Then  percent  solids  are
calculated for the sample.

Routine laboratory management involves detailed recordkeeping beginning with the initial
contact with the sample collector. A master flowsheet should be prepared for each sample,
listing parameters to be determined and pretreatment operations to be  performed.  The
                                       8-3              OCR(6/2001)byRYA-R10

-------
master  flowsheet is kept at a fixed  location and is  designed to handle dated entries
indicating when operations are completed.  Methods requiring extraction permit double
entry on the master flowsheet so laboratory throughput times can be calculated in terms of
receipt-to-extraction and receipt-to-completion times. A separate set of forms follows the
sample through  the laboratory.  Each sample form records  information for a class of
parameters and details cleanup operations and other method options. Calculations and cross
references to chromatograph files are entered on the sample form also.

A chain-of-custody program lends significant legal support to the results generated.  The
program begins when the laboratory dispatches the sample collectors.  Each time a sample
is collected, a form is initiated stating where and when the sample was collected, with cross
reference to a number on the sample container.  The sample collectors return  to the
laboratory with the samples in their guarded possession and deliver them personally to the
responsible person in the laboratory. The person (custodian) receiving the samples in turn
signs each ledger to acknowledge receipt and locks the sample in a refrigerator. The analyst
then comes to the custodian and signs for the sample that he is taking for analysis.  In this
way full documentation of the sample handling is maintained from sample collection
through completion of the analysis.

8.3 Extract Handling

Each method in reference  1 is prepared  in  sections identified by titles.  This style of
presentation is aimed at presenting the analyst logical places to interrupt his analysis.  Often,
because of the length of the method, the analyst is unable to complete an analysis in a single
day. When planning a partial analysis, certain factors  must be considered.  Because the
organics are generally more stable in solvent than in water, it is always preferable to extract
a sample and hold the extract rather than to hold the sample. Extracts to be held overnight
or longer should first be dried by treatment with sodium sulfate.

The methods include several transfers of the solvent extract from one piece of glassware to
another.  These quantitative transfers are made using several  small portions of solvent to
wash the walls of the previous container.  A 5-ml, luer-lock glass syringe with a 2-in., 20-
gage needle is convenient for this purpose.  Solvents tend to creep up the outside of a
container, such as an ampoule, while pouring. To minimize contamination during transfer,
solvent extracts are poured as rapidly as possible. Extracts can become contaminated not
only from oils from the skin of the handler but also from other extracts handled at the same
time where deposits on the outside of the container are unintentionally transferred from one
container to another. Instead of using labels, contamination problems from this source can
be reduced by etching permanent numbers on ampoules.  Sample log sheets should be used
to index the extract with a numbered ampoule, eliminating the need fortape or wax pencils.

Of the several ways to dry a solvent  extract with sodium sulfate, passing it through a
chromatographic column packed with 2 to 3 in. of anhydrous crystals is the most convenient
and quantitative.  When a wet extract is being transferred from a separatory funnel to a
Kuderna-Danish (K-D) concentrator, it should be drained directly through the column to
eliminate the need for an intermediate piece of glassware and the resulting transfer step.
Prewashing the sodium sulfate column with extracting solvent is recommended, although
interferences can be controlled by preheating the salt in a shallow tray at 400°C for 30 min.
After the extract has passed through the column, 20 to 30 ml of extracting solvent are used
to wash the residual extract from the column.
                                       8-4              OCR(6/2001)byRYA-R10

-------
The standard K -D concentration apparatus consists of athree-ball Snyder column, a 500-ml
flask, and a calibrated ampoule. It is designed to concentrate 100- to 300-ml volumes of
extract to a final volume of 5 to 10 ml.  To use the apparatus, one 10/20 mesh boiling chip
(previously rinsed with solvent and heated for 1/2 h) is added, and the assembly is supported
above a concentric-ring water bath with the tip of the ampoule below the surface  of the
water. The lower ground-glass j oint must be kept above the water.  The water temperature
should be adjusted for mfld distillation, with no chamber flooding or splashing (about 10°C
to 20°C above the boiling point of the solvent).  When the volume of liquid reaches 1  to 2
ml (checked frequently), the  assembly should be removed and allowed to cool.   The
chambers and flask will drain to a final volume near 10 ml.  The flask and the lower ground-
glass joint are rinsed with a minimum of solvent.  The evaporation can be continued with
the microscale K-D concentrator if further concentration is required.

The microscale K-D concentrator is designed to concentrate extracts from 5 to  10 ml to 1.0
ml. A fresh boiling chip must be added, and a two-ball micro-Snyder column is attached to
the  ampoule.   The  ampoule is  supported above  the water bath,  and the extract is
concentrated to about 0.7 ml. The column and ground-glass joint are rinsed with aminimum
of solvent, and the volume is adjusted to 1.0 ml.

The K-D concentrator can be used to exchange solvents. When the sample is dissolved in
a solvent unsuitable for a  cleanup operation or for gas chromatography (GC), it can be
displaced by a suitable higher boiling solvent.  The actual volumes that should be used to
effect the exchange vary with the  solvent pairs depending upon the difference in boiling
points and azeotrope formations, but a general procedure is to concentrate the extract to 10
ml, add 20 ml of the higher boiling solvent, and reconcentrate to 10 ml.

After extractions and subsequent K-D concentrations, solubilities of some materials may be
exceeded. High sulfur levels are a particular problem encountered in sediment extractions.
Extracts should be decanted from ampoules where sulfur has crystallized. In some samples
the  extractable organic levels are so high the  extract tends to solidify and will not
concentrate further. When this occurs, a small aliquot of the extract should be taken and
diluted as appropriate for final analysis.

A significant source of the variation in GC analysis can be attributed to the injection of a
portion of the extract into the analytical system.  Manual injections of 2 to 3 |j,l with the use
of a 10-|j,l syringe will introduce variance even when the injection volumes are determined
to the nearest 0.05 |j,l. Of a variety of injection techniques in use, the  solvent flush technique
has been found to be acceptable for quantitative work. This technique is described in detail
in reference 1.

8.4 Supplies and Reagents

Reference compounds of materials should be assayed and of 98 percent purity or higher.
If the purity is less than 98 percent, the appropriate correction factor must be included in all
calculations of standard concentration.  The reference materials should be cataloged, dated,
and stored in a refrigerator.

Stock solutions of these reference materials should be prepared in a high-boiling, inert
solvent, if possible, to minimize errors due to evaporation or solvent-induced decomposi-
                                       J-5              OCR(6/2001)byRYA-R10

-------
tion. The laboratory should have an accurate six-place balance for preparing small quantities
of reference standards. The following method of preparing stock solutions is recommended:
Weigh 10 mg (to the nearest 0.01 mg) of reference standard into a small aluminum weighing
pan.* Drop the entire pan into a 10-ml volumetric flask. Dissolve the reference material in
about 5 ml of solvent, then dilute to volume.  Label the solution with compound name,
concentration, solvent used, date prepared, and initials of preparers Store the volumetric in
a refrigerator, except when preparing dilutions. When the standard is a replacement for an
existing stock solution, the two solutions should be compared, and the results as well as the
suspected reason for any variation should be permanently recorded in the laboratory files.

Working standards are prepared from one or more stock solutions after they have warmed
to room temperature. As these working standards usually represent three to six orders of
magnitude of dilution of the stock standards, it is obviously necessary to take great care in
preparing them.  Serial dilutions are recommended with a maximum of 1:100 dilution for
each step.  Although 10-|j,l volumes can be read within 1 percent with a 10-|j,l syringe, the
inherent problems with the dead volume in the syringe make the use of such equipment less
desirable for preparing working standards than volumetric pipets. New working standards
should be  prepared frequently unless long-term stability has been demonstrated.  When
several compounds are combined into a single standard for simultaneous GC, they must be
closely monitored for chemical interactions.

Pesticide-quality solvents are usually required, and each new lot should be checked for
interferences prior to use. The solvent check, representing approximately 10 percent more
solvent than required for any method, should be concentrated and analyzed for method
interferences under all GC conditions applicable to that solvent. If interfering peaks  or a
broad solvent front are observed, the solvent should be redistilled in an all-glass distillation
system, with a distillation column.** If interferences persist, the solvent lot should be
discarded. This preliminary lot check does not eliminate the need for routine solvent blanks
to monitor for  purity changes over a period of time.

Diethyl ether must be shown to be free of peroxides before use.  Peroxide test stripsf can
be used for a quick,  convenient test.   The  alumina  column procedure for removing
peroxides, described in literature supplied with the test strips, has been used successfully
to remove all peroxides from the  solvent. The solvent should always be stabilized with 2
percent volume/volume ethyl alcohol. Chromatographic elution patterns are based on ether
containing this alcohol.

Granular sodium sulfate should be purchased in glass containers.  If purchased in a large
container (5-lb bottles or larger), it  should be transferred to smaller bottles for daily  use.
Before sodium sulfate is used for chromatographic work, it should be heated to 400°C for
30 min and shown not to be contaminated. When the sodium sulfate is used to dry extracts
before concentration, the heating is usually unnecessary because impurities will be  removed
by preelution of the drying column with solvent.
* Available from The Perkin-Elmer Corp., Norwalk, Conn.; No. 219-0041.
* * Available from Lab Glass, Inc., North West Blvd., Vineland, N.J. 08360; Widner No.
LG-5930.
tAvailable from Scientific Products, 1430 Waukegan Rd., McGaw Park, 111. 60085;
Quant peroxide test strip, No. PI 126-8.

                                       8-6              OCR(6/2001)byRYA-R10

-------
Florisil (Floridin Co.) is purchased from the manufacturer preactivated at 630°C and trans-
ferred to glass containers with Teflon (du Pont) or foil-lined lids. Prior to use the Florisil
(Floridin Co.) is heated in a shallow open dish for 5  h in a 130°C oven.   The Florisil
(Floridin Co.) can then be transferred into a sealed, glass bottle and stored indefinitely at
130°C until needed. Cleanup methods using Florisil (Floridin Co.) require that a lauric acid
value be determined for each lot before use. In addition, the determination of a pesticide
elution pattern is recommended. These procedures are  described in detail  in reference 1.

Carrier gases are a very important part of the chromatographic system; therefore, special
care should be taken in their selection and handling.  Only high-purity or equivalent carrier
gases should be purchased, and they should be filtered, online, through a 5-A molecular
sieve. Porous polymer and other column packings degrade at elevated temperatures in the
presence of trace quantities of oxygen.  Oxygen in the carrier gas also adversely affects the
performance of electron-capture detectors. Therefore, some type of online oxygen-removal
system is recommended for these applications. The chemical traps should be changed or
regenerated with each new cylinder of carrier gas.  One purifier* has  been found to last
through many cylinders without  replacement.  A better grade of gas  is required for
temperature programing than for isothermal operation. Combustion gases may be of lower
quality but should be at least equivalent to dry air or purified hydrogen.  Regulators should
have stainless steel internal parts and be of two-stage design. External tubing should be of
good quality, such as refrigeration tubing. Such tubing should be rinsed with solvent and
heated at 200°C under gas flow before use in the analytical system.

The purchase of precoated  GC column packings is  strongly recommended over the
preparation of coated materials in the laboratory. The commercial products are, generally,
of higher quality and consistency than those prepared by the average analyst. All tubing
should be cleaned before packing by passing a  series of solvents (e.g., hexane, chloroform,
and acetone) through it.  Glass columns should also be silylated. (Instructions are included
with the purchase of silylating agents.) Dry the tubing thoroughly before packing it.

A vibrator should never be used to settle the packing material in the column. Such vibration
may fracture the solid support material, expose uncoated active sites, and produce inferior
chromatographic separations.  When vacuum alone is  inadequate and further settling is
required, the column may be tapped with a pencil or similar object while the vacuum is
being applied.  Unless otherwise  stated, stainless-steel tubing should be packed before
coiling it to fit the GC apparatus.

8.5 Quality Assurance

8.5.1 Measurements

Most of the quality assurance programs suggested in chapter 6 of this manual cannot easily
be adapted to the methods for organic compounds. The  reasons for this, and the suggested
approach for a suitable program for the organic analytical laboratory are discussed in detail
in reference 1 and only summarized here.

Reference 1  suggests that quality assurance for organic analysis be  divided into  three
separate categories.  The first  category represents the  determination  of purgeable  com-
* Available from Matheson Gas Products, P.O. Box E, Lyndhurst, N.J. 07071; Hydrox
Purifier model No. 8301.

                                       8-7              OCR(6/2001)byRYA-R10

-------
pounds. This determination is performed in a closed analytical system; the complete analysis
can be performed in 1 h; and the number of theoretically possible interferences is somewhat
limited.  The second category represents liquid/liquid partition methods in a regulatory
situation.  Here a very limited number of compounds are being measured; there is a high
occurrence of positive results;  and it is important to establish that the method works
satisfactorily on the particular sample matrix. The third category represents liquid/liquid
partition methods in a monitoring situation. Here a large number of compounds are often
being measured simultaneously; there is a low occurrence, of positive results; and each
sample  matrix  may be different.   Quality assurance is  aimed at establishing that the
laboratory is using  the method correctly.

The purgeable methods are unique among organic methods because the standards are treated
in exactly the same way as the samples, and there is no inherent method bias. The methods
are amenable to a variety of quality assurance programs. The approach that has been found
applicable to all types of samples and provides the maximum data for the expended effort
consists of the addition of one or more internal standards to the matrix before purging.  Data
generated in this program provide a continuous monitoring of the equipment and establishes
matrix applicability for the test.

For liquid/liquid extraction methods in a regulatory situation, the  emphasis is placed on
duplicates and dosed samples. Both field duplicates and laboratory duplicates are used in
the program to establish sampling and subsampling validity. The dosing of samples to
establish method accuracy for the  matrix is an integral part of this program. Where the
analytical program  will extend over a long period of time the construction of control charts
is recommended.

When the liquid/liquid extraction methods are used for monitoring, the emphasis is placed
on an external control series. A standard laboratory matrix is developed. With each series
of samples the matrix is dosed and analyzed with the samples. Data generated over a period
of time  can be  used to monitor the performance of the equipment and the analyst, with
relatively tight  specifications to define problems that arise.  Control charts can be con-
structed to alert the analyst to problems, but there is no provision for rej ection of results for
samples of this type.

8.5.2 Identifications

The combined gas chromatography/mass spectrometer (GC/MS) has emerged as the most
important single instrument at the disposal of the environmental analytical chemist. It alone
can provide both  the  sensitivity  and the high degree  of certainty  necessary  for an
identification culled from a complex environmental matrix. The instrument has generated
an aura of well-deserved respect, and its results are seldom questioned. For these reasons
it is mandatory that strict quality assurance programs be followed in both the generation and
the interpretation of mass spectra.  The EMSL, with the cooperation of many other EPA
GC/MS users, has  produced a  procedural manual (2) generally for use with a Finnigan
quadrupole instrument. A detailed quality assurance program constitutes an integral part
of the manual. The actual detailed program is beyond the scope of this manual but has been
summarized in the following paragraphs.

To  insure that  a quadrupole mass spectrometer generates quality spectra,  the program
provides for  at least daily performance evaluation  with a reference compound,  and
                                                        OCR(6/2001) by RYA-R10

-------
readjustment of the instrument as necessary. The operator prepares a solution of decafluoro-
triphenylphosphine (DFTPP).* When this compound is injected into the GC/MS using any
of several compatible GC columns, the resulting elaborate spectrum can be evaluated using
criteria developed at EMSL-Cincinnati (3).

Decafluorotriphenylphosphine key ions and the ion-abundance criteria that are used for
determining whether the mass spectrometer is generating high-quality spectra are as follows:

                             Abundance Criteria
                 Mass (amu)         Ion Abundance Criteila

                       51    30 to 60 percent of mass 198

                       68    Less than 2 percent of mass 69

                       70    Less than 2 percent of mass 69

                      127    40 to 60 percent of mass 198

                      197    Less than 1 percent of mass 198

                      198    Base peak, 100 percent relative abundance

                      199    5 to 9 percent of mass 198

                      275    10 to 30 percent of mass 198

                      365    1 percent (or greater) of mass 198

                      441    Less than mass 443

                      442    Greater than 40 percent of mass 198

                      443    17 to 23 percent of mass 442

If these specifications are not met, the guidelines provided for adjusting the instrument (2,3)
must be considered.

Having produced quality spectra, the analyst must always ascertain, by analyzing a method
blank under exactly the same analytical and instrumental conditions, that the spectra are
relevant. Most commercial systems have a software program, extracted ion current profile
(EICP), that permits the sample and the blank to be overlayed on a graphic display device
when each is scanned for a particular mass. When large numbers of spectra result from a
sample and the blank must be checked for a match for each one, this technique simplifies
the screening process.

If the spectra are found to be unique to the sample, it is normally processed through a mass-
spectral search-and-match system. The  computerized version of such a system consists
* Available from PCR Research Chemicals, Inc., P.O. Box 1778, Gainsville, Fla. 32602;
No. 11 898-4.

                                       8-9              OCR(6/2001)byRYA-R10

-------
of an organized collection of many thousands of compound spectra stored in a large central
computer accessed by the analyst through telephone linkups. On the basis of 8 or 10 major
masses the computer can rapidly search the complete system for similar spectra.  These
spectra are ranked for similarity to the unknown using a mathematical algorithm.  The
resulting similarity index (SI), which can be further refined to a  quality index (2), is a
measure of the degree of confidence the analyst may place in the identification. If a match
is not found, the analyst must revert to manual interpretation of the  spectra and deduce the
structure of the compound by its fragmentation patterns.

After a tentative identification is made, several other types of supporting experiments
become  possible.  Retention-time GC/MS data of a pure compound (standard) may be
compared with analogous data from the sample component.  Similarly, the mass spectrum
of the standard, obtained under the same conditions that were used for the sample, may be
compared with the sample component spectrum.  The standard may be dissolved in water
at an appropriate concentration, isolated, and measured.  The recovery of this spike in the
same fraction in which the suspected component appeared and the observation of equivalent
mass spectra  for the spike and the sample component constitute strong  evidence for
confirmation of the identification.

8.6 References

1.   Methods for Organic Analysis of Water and Wastes, U.S. EPA, EMSL, Environmental
    Research Center, Cincinnati (in preparation).
2.   EPA GC/MS Procedural Manual, Budde, W. L., and Eichelberger, J. W., editors, 1st
    Edition, Vol. 1, U.S. EPA, Office of Research and Development, EMSL, Cincinnati (in
    press).
3.   Eichelberger, J. W., Harris, L. E., and Budde, W. L., Anal. Chem., 47, 995 (1975).
                                      8-10              OCR(6/2001)byRYA-R10

-------
                                   Chapter 9

                            SKILLS AND TRAINING

9.1 General

Analytical   operations  in the  laboratory can be graded  according to the degree of
complexity.  Some analyses require no sample treatment, and the measurement can be
performed in minutes on a simple  instrument.  Other determinations require extensive
sample preparation  prior to complex instrumental examination.   Consequently,  work
assignments in the laboratory should be clearly defined. Each analyst should be completely
trained and should fully understand all the assignments of his job before being given new
responsibilities.  In this  regard, all analysts,  subprofessional or professional, should be
thoroughly instructed in basic laboratory operations, according to the degree of professional
maturity. Some of the basic operations that should be reviewed periodically with laboratory
personnel follow.

9.1.1 Sample Logging

Routine procedure for recording of samples entering the laboratory and assigning primary
responsibility should be emphasized. The information that is required and the routing of the
sample to the analyst is then established. The stability, preservation, and storage of samples
prior to analyses are then discussed.

9.1.2 Sample Handling

The analyst should understand thoroughly at which points in his procedures the sample is
to be settled, agitated, pipetted, etc., before he removes it from the original container.

9.1.3 Measuring

The analysts, especially new employees and subprofessionals, should be instructed in the
use of volumetric glassware. The correct use of pipettes and graduates should be emphasized
as discussed in chapter 4 of this manual.

9.1.4 Weighing

Because almost every measuring operation in the analytical laboratory is ultimately related
to a weighing operation, the proper use of the analytical balance should be strongly
emphasized. Maintenance of the balance, including periodic standardization, should be
repeatedly emphasized to all personnel. The  correct use and maintenance of balances is
discussed in chapter 3 of this manual.

9.1.5 Glassware

All glassware should be washed and rinsed according to the requirements of the analysis to
be performed. Not only must the personnel assigned to these tasks be instructed, but also all
lab personnel should know the routine for washing and special requirements for particular
uses of glassware. In addition, the precision tools of the laboratory such as pipets, burets,
                                       9-1               OCR(6/2001)byRYA-R10

-------
graduates, and Nessler tubes should be inspected before use for cleanliness, broken delivery
tips, and clarity of marking. Defective glassware should be discarded or segregated.

9.1.6 Instrumentation

Operation and maintenance of analytical instrumentation is of primary consideration in the
production of valid data. All instruments must be properly calibrated, quality-control checks
documented, and standard curves verified on a routine basis.  Details on instrumental
quality control are presented in chapter 3 of this manual.

9.1.7 Data  Handling and Reporting

As with sample logging, the routine procedure for recording results of analyses and pertinent
observations, including quality control  checks, should be emphasized.  Analytical data
should be permanently recorded in meaningful, exact terms and reported in a form that
permits future interpretation and unlimited use. Details are discussed in chapter 7 of this
manual.

9.1.8 Quality Control

The need to  continuously assess precision and recovery values of methodology is a prime
responsibility of the analyst. Self-evaluation through the analyses of replicates and recovery
of spikes  from samples representative  of the daily workload provides  confidence and
documentation of the quality of the reported data.

9.1.9 Safety

Laboratory safety should be discussed on a continuing basis with all employees, but it
should be  emphasized when an employee is assigned to perform new duties.

9.1.10 Improvement

In summary, quality control begins with basic laboratory techniques. Individual operator
error  and  laboratory error can be minimized if approved  techniques are consistently
practiced.  To insure the continued use of good technique, laboratory supervisors should
periodically review the basic techniques and point out areas of needed improvement with
each analyst.

Continuing improvement of technical competence by all laboratory personnel is, of course,
the final responsibility of the laboratory supervisor. In a well-organized laboratory, however,
a big-brother attitude of higher ranking to lower grade personnel should be encouraged; each
person should be eager to share experience, tricks of the trade, special skills, and special
knowledge with subordinates. Obviously, efficiency and results will improve.

9.2 Skills

The cost of data production in the analytical laboratory is based largely upon two factors:
the  pay scale of the analyst,  and the number of data units produced per unit of time.
However,  because of the large variety  of factors involved,  estimates of the  number of
measurements that can be made per unit of time are difficult. If the analyst is pushed to
                                       9-2              OCR(6/2001)byRYA-R10

-------
                               Table 9-1
     SKILL-TIME RATING OF STANDARD ANALYTICAL OPERATIONS
Measurement
Simple Instrumental:
pH
Conductivity
Turbidity
Color
Dissolved Oxygen (Probe)
Fluoride (Probe)
Simple Volumetric:
Alkalinity (Potentiometric)
Acidity (Potentiometric)
Chloride
Hardness
Dissolved Oxygen (Winkler)
Simple Gravimetric:
Solids, Suspended
Solids, Dissolved
Solids, Total
Solids, Volatile
Simple Colorimetric:
Nitrite N (Manual)
Nitrate N (Manual)
Sulfate (Turbidimetric)
Silica
Arsenic
Complex, Volumetric, or Colorimetric:
BOD
COD
TKN
Ammonia
Phosphorus, Total
Phenol (Distillation Included)
Oil and Grease
Fluoride (Distillation Included)
Cyanide
Special Instrumental:
TOC
Metals (by AA), No Preliminary
Skill Required Number
(Rating No.)1 Per Day






















,2
,2





,2

,2
,2
,2
,2

2
2
2
2
2,3

2,3
2,3
2,3
2,3
2,3
2,3
2,3
2,3
2,3

2,3
Treatment 2,3
Metals (by AA), With Preliminary Treatment 2,3
Organics (by GC), Pesticides, Without Cleanup 3,4
Organics (by GC), Pesticides, With Cleanup 3,4

100-125
100-125
75-100
60-75
100-125
100-125

50-75
50-75
100-125
100-125
75-100

20-25
20-25
25-30
25-30

75-100
40-50
70-80
70-80
20-30

215-20
25-30
25-30
25-30
50-60
20-30
15-20
25-30
8-10

75-100
150
60-80
3-5
2-4
1 Skill-required rating numbers are defined as follows:
   1 - aide who is a semiskilled subprofessional with minimum background or
      training, comparable to GS-3 through GS-5. (Continued)
                                  9-3
OCR(6/2001) by RYA-R10

-------
produce data at a rate beyond his capabilities, unreliable results may be produced. On the
other hand, the analyst should be under some compulsion to produce a minimum number of
measurements per unit of time, lest the cost of data production become prohibitive. In table
9-1, estimates are given for the number of determinations that an analyst should be expected
to perform on a routine basis. The degree of skill required for reliable performance is also
indicated.

The time limits presented  in the table are based on use of approved methodology.  A tacit
assumption has been made that multiple analytical units are available for measurements re-
quiring special equipment, as for cyanides, phenols, ammonia, nitrogen, and COD.  For
some of the simple instrumental or simple volumetric measurements, it is assumed that other
operations such as filtration, dilution, or duplicate readings are required; in such cases the
number of measurements  performed per day may appear to be fewer than one would nor-
mally anticipate.

9.3 Training

For more experienced, higher grade personnel, formal  training in special fields, possibly
leading to specialization, should be almost mandatory.  Such training can be fostered
through local institutions and through the training courses provided by the EPA. Regional
policies on after-hours, Government-supported training should be properly publicized.

Formalized training for lower grade personnel, comparable to GS-3 to GS-5, is relatively
scarce. However, skills can be most efficiently improved at the bench level on a personal,
informal  basis by more experienced analysts working in  the same  area.  Exposure of
personnel to pertinent literature should also be a definite program policy.
(Continued)
           2 -  aide with special training or professional with minimum training with
               background in general  laboratory techniques  and  some knowledge of
               chemistry, comparable to GS-5 through GS-7.
           3 -  experienced analyst capable of following complex procedures with good
               background in analytical techniques, professional,  comparable to  GS-9
               through GS-12.
           4 -  experienced analyst specialized in highly complex procedures, profes-
               sional, comparable to GS-11 through GS-13.
       2Rate depends on type of samples.
                                       9-4              OCR(6/2001)byRYA-R10

-------
                                   Chapter 10

                   WATER AND WASTEWATER SAMPLING
10.1 Introduction
The quality of data resulting from water and wastewater sampling surveys is dependent upon
the following six major activities: (a) formulating the particular objectives of the water
sampling program, (b) collecting representative water samples, (c) maintaining the integrity
of the water samples through proper handling and preservation, (d) adhering to adequate
chain-of-custody and sample identification procedures, (e) practicing quality assurance in
the field, and (f) properly analyzing the pollutants in the water samples.  These areas are
equally important for insuring that environmental data are of the highest validity and quality.

The present section addresses aspects of quality control (QC) concerned with the collection
of environmental samples and data in  the field.   It includes a capsule summary of the
specific areas mentioned previously and  a list of references (table 10-1) that provide specific
guidance in these areas, rather than a collection of guidelines on sampling procedures.

10.2 Areas of Sampling

The specific areas that comprise an overall water sampling program are as follows.

10.2.1 Objectives of the Particular Sampling Program

The objectives of the sampling program affect all the other aspects of the sampling program.
Sampling program objectives are determined by the following activities:  (a) planning
(areawide or basin), (b) permitting, (c) compliance, (d) enforcement, (e) design, (f) process
control, and (g) research and development. The types of water sampling programs to be
employed, depending on suitability to program objectives, include reconnaissance surveys;
point-source characterization; intensive surveys; fixed-station-networkmonitoring; ground-
water monitoring; and special surveys involving chemical, biological, microbiological, and
radiological monitoring.

Factors that must be considered in meeting the objectives of the sampling program are the
extent of the manpower resources, the complexity of the parameters of interest, the duration
of the survey, the number of samples, the frequency of sampling, the type of samples (grab
or composite), and the method of sample collection (manual or automatic).

10.2.2 Collection of Representative  Samples

The objective of all water and wastewater sampling is to obtain a representative portion of
the total environment under investigation. The techniques for obtaining representative water
samples may vary with the length, width, and depth of a body of water, its physical and
chemical parameters, and its type to be sampled (such as municipal or industrial effluents,
surface waters and bottom  sediments,  agricultural runoff, and  sludges).  In collecting
representative samples, the following factors should be considered.
                                     10-1               OCR(6/2001)byRYA-R10

-------
                   Table 10-1
 GUIDANCE FOR WATER/WASTEWATER SAMPLING
Sampling Subject Area

Objectives of a Sampling Program
Collection of Representative Samples:
Site Selection
Sample Types1
Automatic Samplers
Flow Measurement
Statistical Approach to Sampling:
Frequency of Sampling
Number of Samples
Probability of Exceeding Standard
Special Sampling Procedures:
Municipal Wastewaters
Industrial Wastewaters
Agricultural Wastewaters
Surface Waters
Bottom Sediments
Sludges
Biology
Microbiology
Radiation
Sample Preservation and Handling:
Volume of Sample
Container Type and Cleaning
Sample Identification
Chain of Custody
Quality Assurance in the Field

1
X
X
X
X
X
X
X
X
X
X

X
X
X
X
X
X
X
X
X
X
X
X
X
X


2

X
X
X
X
X













X
X
X
X
X
X
X
Reference Numbers
3456789
X X

X X
X X
X X X X
XX X

X
X
X







X X
X X
X X
XXX


X
X


10 11 12




X
X X





X








X X
X




:Grab, composite, manual, etc.
10-2
OCR(6/2001) by RYA-R10

-------
10.2.2.1 Site Selection

The location of the sampling site is critical in obtaining representative data. Preferably,
water sampling sites for point sources of pollution from municipal and industrial effluents
are located at points of highly turbulent flow to insure good mixing; however,  inaccessi-
bility, lack of site security, or power unavailability may preclude use of the best sites, but
these impediments should not be used as reasons for collecting samples at unacceptable
locations. Locations of sampling sites for streams, lakes, impoundments, estuaries, and
coastal areas vary, but in general occur in the following bodies: (a) in water bodies for
sensitive uses (swimming and  drinking water supply), (b) in major impoundments or
reservoirs near the mouths of major tributaries and in the rivers entering and leaving the
impoundments, (c) in water bodies polluted by man's activities, (d) in rivers upstream and
downstream from tributaries, and (e) where hydrological conditions  change significantly.

10.2.2.2 Types

The basic types of water and wastewater sampling methods are grab sampling and composite
sampling. Composite  sampling may be conducted manually  or automatically. The six
methods for forming composite samples, all of which depend on either a continuous or
periodic sampling mode, are the following: (a) constant sample pumping rates, (b) sample
pumping rates proportional to stream flow rates, (c) constant sample volumes and constant
time intervals between samples, (d) constant sample volumes and time intervals between
samples proportional to stream flow rates, (e) constant time intervals between samples and
sample volumes proportional to total stream flow volumes since last sample, and (f) constant
time intervals between samples, and sample volumes proportional to total stream flow rates
at time  of sampling.  The choice of using the grab sampling method or one  of the six
compositing sampling methods is determined by program objectives  and the parameters to
be sampled.

10.2.2.3 Automatic Samplers

The use of automatic samplers eliminates errors caused by the human element  in manual
sampling, reduces personnel cost, provides more frequent sampling than practical for
manual sampling, and eliminates the performance of routine tasks by personnel. Criteria for
brand selection of automatic samplers include evaluations of the intake  device,  intake
pumping rates, sample transport lines, sample gathering systems (including pumps and
scoops), power supplies and power controls, sample storage  systems,  and  additional
desirable features to fit particular  sampling conditions.  There are many commercially
available automatic samplers; however, because no single automatic sampler is ideally
suited for all situations, the user carefully selects the automatic sampler most suited for the
particular water or wastewater to be characterized.  Precautions must be taken in regard to
using certain types of samples in potentially explosive atmospheres.

10.2.2.4 Flow Measurement

An essential part of any water or wastewater sampling survey as well  as a  necessary
requirement of the  National Pollution Discharge Elimination System (NPDES) permit
program is accurate flow measurement, which can be divided into four categories:

   a.    Flow measurement in  completely filled pipes under pressure-common devices
        employed are orifices,  Venturi tubes, flow nozzles, Pitot tubes, magnetic flow
        meters, ultrasonic flow meters, and elbow meters.
                                    10-3               OCR(6/2001)byRYA-R10

-------
  b.   Trajectory methods, either full or partially full, measured at the end of the pipe -
       common flow measurement methods are the California and Purdue pipe methods.
       These methods are normally  considered as  estimates rather than accurate
       measurements.

  c.    Flow measurement in open channels and sewers-common methods are the velocity-
       area measurement, time-of-passage measurement, and level measurement methods
       using weirs and flumes.

  d.   Miscellaneous flow measurement methods-common methods include use of Man-
       ning formula, tracer and salt dilution techniques, water meters, pump rates, and
       measurements of level changes in tanks and calibrated vessels.

Flow measurement data may be instantaneous  or continuous.   For continuous measure-
ments, atypical system consists of primary devices such as weirs and flumes and secondary
devices  such as  flow sensors, transmitting  equipment,  recorders, and  totalizers.  The
improper installation or design of a primary  device  or malfunction of any  part of a
secondary device results in erroneous flow data. The accuracy of flow measurement data
also varies widely, depending  principally on the accuracy of the primary device and the
particular flow measurement method used. In any case, an experienced investigator should
be able to measure flow rates within ±10 percent of the true values.

10.2.2.5 Statistical Approach to Sampling

Four factors must be established for every sampling program: (a) number of samples, (b)
frequency of sampling, (c) parameters to be measured, and (d) sampling locations. These
factors are usually determined in varying degrees by details of the pertinent discharge
permits or are more arbitrarily set by the program resource limitations. Nevertheless, the
nature of the statistical methods selected and scientific judgment should be used to establish
the best procedures.

10.2.2.6 Special Sampling Procedures

Special sampling procedures should be employed for municipal, industrial, and agricultural
waters, surface waters  as well as bottom sediments and sludges, and  for biological,
microbiological, and radiological studies.

10.2.3 Sample Preservation and Handling

During and after collection, if immediate analysis is not possible, the sample must be
preserved to maintain its integrity. The only legally binding reference EPA has for sample
preservation methods is the NPDES permit program specified in reference 13.  However,
these sample preservation procedures serve as a guide for other program objectives.

Proper handling of the samples helps insure valid data; consideration must also be given to
care of the  field container material and cap material, cleaning, structure of containers,
container preparation for determination of specific parameters, container identification, and
volumes of samples.
                                    10-4               OCR(6/2001)byRYA-R10

-------
10.2.4 Chain-of-Custody Procedures

All programs involved in water and wastewater surveys should document and implement a
chain of possession and custody of any sample collected, whether or not the resulting data
are to be used in enforcement cases. Such procedures insure that the samples are collected,
transferred, stored, analyzed, and destroyed only by authorized personnel. See section 12.7
for detailed procedures that can be used on all sample types.

10.2.5 Quality Assurance in the Field

Quality assurance programs for sampling equipment and for field measurement procedures
(of such parameters as temperature, dissolved oxygen, pH, and conductance) are necessary
to insure data of the highest quality. A field quality assurance program administered by a
quality assurance coordinator should contain the following documented elements:

  a.     The analytical methodology; the special sample handling procedures; and the preci-
        sion, accuracy, and detection limits of all analytical methods used.

   b.    The basis for selection of analytical and sampling methodology.  For example, all
        analytical methodology for NPDES permits shall be that specified in reference 13,
        or shall consist of approved alternative test procedures. Where methodology does
        not exist, the quality assurance plan should  state how the new method will be
        documented, justified, and approved for use.

   c.    The amount of analyses for quality control (QC), expressed as a percentage of
        overall analyses, to assess the validity of data.  Generally, the complete quality
        assurance program should approximate 15 percent of the overall program with  10
        and 5 percent assigned to laboratory QC and field QC, respectively. The plan should
        include a shifting of these allocations or  a decrease in the allocations depending
        upon the degree of confidence established for collected data.

   d.    Procedures for the recording, processing,  and reporting of data; procedures for
        review of data and invalidation of data based upon QC results.

   e.    Procedures for calibration and maintenance of field instruments and automatic
        samplers.

   f.     A performance  evaluation system, administered through the quality assurance
        coordinator, allowing field sampling personnel to cover the following areas:

        (1)  Qualifications of field personnel for a particular sampling situation.

        (2)  Determination of the best representative sampling site.

        (3)  Sampling technique including location of the points of sampling  within the
            body of water, the choice of grab or composite sampling, the type of automatic
            sampler,  special handling procedures,  sample preservation, and sample
            identification.

        (4)  Flow measurement, where applicable.
                                     10-5                OCR(6/2001)byRYA-R10

-------
       (5)  Completeness of data, data recording, processing, and reporting.

       (6)  Calibration and maintenance of field instruments and equipment.

       (7)  The use of QC samples such as duplicate, split, or spiked samples to assess the
            validity of data.

  g.   Training of all personnel involved in any function affecting the data quality.

Quality assurance in sample collection should be implemented to minimize such common
errors as improper sampling methodology, poor sample preservation,  and lack of adequate
mixing during compositing and testing. The checks listed in the following sections will help
the quality assurance coordinator to determine when the sample collection system is out of
control.

10.2.5.1  Duplicate Samples

At selected stations on a random time frame duplicate samples are collected from two sets
of field equipment installed at the site, or duplicate grab samples  are collected.  This
provides a check of sampling equipment and technique for precision.

10.2.5.2 Split Samples

A representative subsample from the collected sample is removed and both are analyzed for
the pollutants  of interest.  The samples may be reanalyzed by the same laboratory or
analyzed by two different laboratories for a check of the analytical procedures.

10.2.5.3 Spiked Samples

Known amounts of a particular constituent are added to an actual sample  or to blanks of
deionized water at concentrations at which the accuracy of the test method is satisfactory.
The amount added should be coordinated with the laboratory.  This method provides a
proficiency check for accuracy of the analytical procedures.

10.2.5.4 Sample Preservative Blanks

Acids and chemical preservatives can become contaminated after a period of use in the field.
The sampler should add the same quantity of preservative to some  distilled water as
normally would be added to a wastewater sample.  This preservative blank is  sent to the
laboratory for analysis of the same parameters that are measured in the sample  and values
for the blank are then subtracted from the sample values.  Liquid chemical preservatives
should be changed every 2 weeks-or sooner, if contamination increases above predetermined
levels.

10.2.5.5 Precision, Accuracy, and Control Charts

A minimum of seven sets each of comparative data for duplicates, spikes, split samples, and
blanks should be collected to define acceptable estimates of precision and accuracy criteria
for data validation.
                                    10-6               OCR(6/2001)byRYA-R10

-------
10.2.5.6 Calibration of Field Equipment

Plans should be developed and implemented for calibrating all field analysis test equipment
and calibration standards to include the following: (a) calibration and maintenance intervals,
(b) listing of required calibration standards, (c) environmental conditions requiring calibra-
tion, and (d)  a  documented record system.  Written calibration procedures should be
documented and should include mention of the following:

   a.    To what tests the procedure is applicable.

   b.    A brief description of the calibration procedure. (A copy of the manufacturer's
       instructions is usually adequate.)

   c.    A listing of the calibration standard, the reagents, and any accessory equipment
       required.

   d.    Provisions for  indicating  that the field equipment is labeled and contains the
       calibration expiration date.

10.3 References

 1.  Huibregtse, K. R., and Moser, J. H., Handbook for Sampling and Sample Preservation
     of Water and Wastewater, EPA-600/4-76-049 (Sept.  1976).
 2.  Olson, D. M., Berg, E. L., Christensen, R., Otto, H., Ciancia J., Bryant, G., Lair, M. D.,
     Birch, M.,  Keffer, W., Dahl, T., and Wehner, T., Compliance Sampling Manual,
     Enforcement Division, Office of Water Enforcement, Compliance Branch (June 1977).
 3.  Weber, C. I., Biological Field and Laboratory Methods for Measuring the Quality of
     Surface Waters and Effluents, EPA-670/4-73-001, U.S. EPA (July 1973).
 4.  Lauch, R. P., A Survey of Commercially Available Automatic Wastewater Samplers,
     EPA-600/4-76-05 1, U.S. EPA (Sept. 1976).
 5.  Shelly, P. E., and Kirkpatrick, G. A., An Assessment  of Automatic Sewer Flow
     Samplers, EPA-600/2-75-065, U.S. EPA (Dec. 1975).
 6.  Handbook  for Monitoring Industrial Wastewater,  U.S. EPA,  Technology Transfer
     (Aug. 1973).
 7.  Areawide Assessment Procedures Manual, Vol. II, App. D, EPA-600/9-76-014,  U.S.
     EPA (July  1976).
 8.  Winter, J. A., Bordner, R., and Scarpino, P., Microbiological Methods for Monitoring
     the Environment, Part I - "Water and Wastes" U.S. EPA, EMSL, Cincinnati (1977).
 9.  Water Measurement Manual, 2d Edition (Revised), U.S. Department of the Interior,
     Bureau of Reclamation (1974).
10.  Kulin, G., and Compton, P., A Guide to Methods and Standards for Measurement of
     Water Flow, Special Publication 421, National Bureau of Standards (May 1975).
11.  Methods for Chemical Analysis of Water and Wastewater, U.S. EPA Technology
     Transfer (1974).
12.  Harris, D.  J., and Keffer, W. J., Wastewater Sampling  Methodologies and Flow
     Measurement Techniques, EPA-907/9-74-005 (Sept. 1974).
13.  Guidelines Establishing Test Procedures for the Analysis of Pollutants, Federal
     Register, 41, No. 232 (Dec.  1976).
                                    10-7               OCR(6/2001)byRYA-R10

-------

-------
                                   Chapter 11

                               RADIOCHEMISTRY

11.1 Introduction

The objective of this chapter is to provide general information and suggestions that enable
the analyst to execute his responsibilities in analytical quality control (QC) as they relate to
radiochemical analyses. Because chemical and radiochemical responsibilities should be
considered together, the following requirements could be included with the items specified
in the preceding chapters as tasks for the radiochemist: verifying the validity of laboratory
data, recommending methodology, interpreting the results, and examining the  need for
standard procedures.

Environmental radiation measurements are made daily by Federal, State, local, and private
agencies.  The data obtained from these measurements are used by the U.S. EPA and other
agencies for such purposes as estimating doses, describing health effects, establishing
standards  and guides, and conducting regulatory activities. It is imperative to insure the
precision  and  accuracy of the data, so that policy decisions concerning environmental
quality are based on valid and comparable data.

A radiation QC program should be designed to encourage the development and implementa-
tion of QC procedures at all levels of sample collection, analysis, data processing, and
reporting. It should enable the analyst to verify his analyses and document the validity of
the data. In addition, such a program allows the determination of the precision and accuracy
of environmental radiochemical analyses.

11.2 Sample  Collection

Analytical results can be no more meaningful than the integrity of the samples that are
analyzed.  Representative  samples must be collected so the data for any aliquot can be
related to a well-defined pollution source. For most analyses (table 11-1) the sample should
be preserved at the sampling site to maintain its integrity and to minimize activity losses
from absorption on container walls.  If at all possible, analyses should be performed soon
after receipt of the samples at the laboratory.

Sample container types and descriptions have already been discussed. Both plastic and glass
containers have been recommended, and each has its particular merit.  Cost most likely
determines which of the many types of plastic  containers will be used; in general, those
more expensive should be more resistant to adsorption losses.  Because sample analyses are
recommended soon after collection, less expensive plastic ware can be tolerated.  In any
event, containers should be discarded after use to prevent contamination of subsequent
samples.

Glass bottles are popular items, readily  available in all  sizes.  Although the possibility of
breakage  either in handling or shipping is  obviously great, the very small sizes for
radioactive tracer standards and calibration sources can withstand breakage. Use of the
larger size bottles for these purposes should therefore be avoided. It is poor economics to
ship  radioactive samples in fragile  glass containers when unbreakable types  serve the
purpose much  better.
                                     11-1               OCR(6/2001) by RYA-R10

-------
                                                       Table 11-1
          SAMPLE HANDLING, PRESERVATION, METHODOLOGY, AND MAJOR INSTRUMENTATION REQUIRED1
          Parameter
             Sample
          Preservation2
Container3
Instrumentation
Gross Alpha Activity
Gross Beta Activity
Strontium-89
Strontium-90
Radium-226
Radium-228
Cesium-134
Iodine-131
Tritium
Uranium
Photon Emitters
Concentrated HNO3 to pH < 2
Concentrated HNO3 to pH < 2
Concentrated HNO3 to pH < 2
Concentrated HNO3 to pH < 2
Concentrated HNO3 to pH < 2
Concentrated HNO3 to pH < 2
Concentrated HC1 to pH < 2
None
None
Concentrated HNO3 to pH < 2
Concentrated HNO3 to pH < 2
   P or G        Low-background proportional counter
   P or G        Low-background proportional counter
   P or G        Low-background proportional counter
   P or G        Low-background proportional counter
   P or G        Scintillation cell system
   P or G        Low-background proportional counter
   P or G        Low-background proportional counter
   P or G        Low-background proportional counter
   G            Liquid scintillation counter
   P or G        Fluorometer
   P or G        Gamma spectrometer
:See reference 1 for method.
Preservative to be added at the time of collection unless the sample is to be characterized for suspended solids activity.
3P = hard or soft plastic; G = hard or soft glass.
                                  11-2
                     OCR(6/2001) by RYA-R10

-------
11.3    Laboratory Practices

11.3.1 Laboratory Safety

The general principles of laboratory safety are covered in chapter 14. The hazards to be
avoided are listed, and the importance of good housekeeping practices is stressed. These
practices reduce the potential hazard of the many chemical operations being performed. In
the radiochemical laboratory they  eliminate  the  probability of radiological cross-
contamination from sample to sample and from sample to glassware. Many of the materials
used in the laboratory are potentially hazardous because of their chemical properties or their
radioactivity and therefore should be handled with the utmost care and respect.

In the radiochemical laboratory, the prevention of contamination by radioactive materials
requires attention to radiation-protection practices, an ongoing personnel-monitoring pro-
gram, and the designation of a segregated storage area for radioactive sources following use
in radiochemical analyses. Adequate labeling of work areas, of samples for analysis, of
aliquots,  and of separated  fractions helps to  control radiation hazards and to insure
personnel safety.

The handling of radioactive materials involves safety  hazards of a type not usually
associated with  a chemical laboratory. Special precautions and instruments should be used
to insure the greatest personal safety. It is imperative to wear monitoring devices (personal
film badges) at  all times.

A number of health and contamination hazards  are to  be considered. Many  radioactive
aterials are dangerous even in extremely minute quantities if inhaled or ingested.  All
radioactive materials are capable of contaminating laboratories, instruments, and clothing.
All radioactive materials in large concentrations are dangerous because of the effects of their
radiation external to their containers.  To control these hazards, the following rules should
be in effect at all times.

   a.     For the case of radioactive materials that are capable of being volatilized or
        airborne, perform all  work  in a closed area or hood.   Perform distillations,
        evaporations, and other such processes in a well-ventilated hood.

   b.     Do not  bring food or liquid refreshments into a laboratory engaged in work with
        radioactive materials. The  same applies to the  counting rooms.

   c.     Do not  smoke when handling radioactive materials.

   d.     After working with radioactive materials, wash hands thoroughly before eating or
        handling uncontaminated materials.

11.3.2 Laboratory Analyses

Standard radiochemical procedures or their equivalent are needed to comply with sensitivity
detection limits for each nuclide as designated by  the  quality assurance program  (2).
Radiochemical procedures have been compiled for a multitude of nuclides in a multitude of
media (3-9) and  descriptions of specific separations  are to be found in the scientific
literature.  As laboratory techniques become more  sophisticated and as more sensitive
instrumentation is developed, these procedures will be improved.
                                     11-3               OCR(6/2001) by RYA-R10

-------
Depending on the media, the radioactivity levels, and the nuclide composition, there are
several approaches that can be made when a sample is received for QC determination. The
requirements for an acceptable QC program are described in section 11.4.

Before starting such a program, the laboratory should be already set up for radiochemical
analyses and the analysts should have the prescribed education and experience.

11.3.3 Laboratory Radiation Instruments

The types of radiation counting systems (described in ref 2) needed to comply with the
requirements are set forth in the following paragraphs. Only those instruments needed for
analyzing specific radionuclides are required.  Such instruments should meet the specifica-
tions discussed in the next sections.

11.3.3.1 Liquid Scintillation System

A liquid scintillation system must measure tritium with the  sensitivity required by the
National Interim Primary Drinking Water Regulations. Efficiency of the system should be
greater than 57 percent for tritium.  The tritium  figure of merit (E^)2/B should be greater
than 100.

11.3.3.2 Gas-Flow Proportional Counting System or Alternative

A gas-flow proportional counter, or the alternative  described  later, is required for
measurement of gross alpha- and gross beta-particle activities, radium-228, strontium-89,
cesium-134, and iodine-131. The detector may be either windowless (internal proportional
counter) or of the thin-window type. A minimum  shielding equivalent of 5 cm of lead must
surround the detector.  A cosmic (guard) detector should be operated in anticoincidence
with the main detector. The main detector should have an efficiency greater than 20 percent
for polonium-210 and carbon-14 and greater than 40 percent for strontium-90. The detector
background should be less than 1.3 counts per minute.

The detector plateau should be less than 1.5 percent per 100 V and should be at least 100
V wide for carbon-14 and less than 2 percent per 100 V for strontium-90.

A scintillation system designed for alpha- and beta-particle counting may be substituted for
the gas-flow proportional counter described. In such a system a Mylar disk coated with a
phosphor (silver-activated zinc sulfide) is either  placed directly on the sample (for alpha
measurements)  or on the face  of a photomultiplier tube, enclosed within a light-tight
container along with the appropriate electronics (high-voltage supply, amplifier(s), timer,
and sealer). Radiation shielding, although desirable, is not required for this system.

11.3.3.3 Gamma Spectrometer System

A sodium  iodide (Nal)  detector connected to a multichannel analyzer is required for
determination of manmade photon emitters. A 7.5- by 7.5-cm Nal crystal  is satisfactory;
however, a 10- by 10-cm crystal is recommended. The crystal detector must be shielded
with a minimum of 10 cm of iron or equivalent.

It is recommended but not required that the distance from the center of the crystal detector
to any part on the shield should not be less than 30 cm.  The multichannel analyzer, in
                                     11-4               OCR(6/2001) by RYA-R10

-------
addition to the appropriate  electronics (high-voltage supply, preamplifier,  and linear
amplifier), must contain a memory of not less than 200 channels and at least one readout
device.

11.3.3.4 Scintillation Cell System

A scintillation system must  be designed to accept scintillation flasks (Lucas cells) for
measurement of radium-226 by the radon-emanation method. The system consists of a light-
tight enclosure for the scintillation flasks, a detector (phototube), and the appropriate
electronics (high-voltage supply, amplifier, timers, and sealers). The scintillation flasks
required for this measurement may either be purchased from commercial suppliers or
constructed according to published specifications.

11.4 Quality Control

The following requirements are recommended for all laboratories:

   a.    All QC data should be available for inspection to determine validity of laboratory
       results.

   b.    Each laboratory should participate at least  twice each year in EPA laboratory
       intercomparison studies (10).

   c.    Each laboratory should  participate once  each year in an appropriate EPA-
       administered performance study on unknowns. Results must be within the control
       limits established by  EPA for each analysis.

   d.    Counting-instrument operating manuals and calibration protocols should be
       available to analysts and technicians.

  e.    Calibration data and maintenance records on all radiation instruments and analytical
       balances must be maintained in a permanent record.

   f    Minimum daily QC

       (1)   To verify precision of methods, a minimum of 10 percent of the samples shall
            be duplicates.  Checks must be within ±2 standard deviations of the mean
            range.

       (2)   If less than 20 samples per day are analyzed, a performance standard and a
            background sample must be measured. If 20 or more samples are analyzed per
            day, a performance standard and a background sample must be measured with
            each 20 samples. Checks must be within ±2 standard deviations of the mean
            range.

       (3)   Quality control performance  charts  or  performance records must be
            maintained.

11.5 References

  1. Federal Register 41, No. 133 (July 9,  1976).
 2. Jarvis, A. N., et al., The Status and Quality of Radiation Measurements of Water, EPA-
    600/4-76-017 (Apr. 1976).

                                     11-5                OCR(6/2001) by RYA-R10

-------
 3.  Krieger, H. L., Interim Radiochemical Methodology for Drinking Water, EPA-600/4-
    75-008 (Revised), U.S. EPA, Office of Research and Development, Cincinnati (Mar.
    1976).
 4.  Doublas, G. S., Editor, Radioassay Procedures for Environmental Samples, Environ-
    mental Health Series, Publication 999-RH-27,  U.S. DREW, Public Health Service,
    National Center for Radiological Health (Jan. 1967).
 5.  Harley, J. H., HASL Procedures Manual, HASL-300, U.S. ERDA, Health and Safety
    Laboratory, New York (1972).
 6.  Kleinberg, J., Editor, Collected Radiochemical  Procedures, LA-1721 (Revised), Los
    Alamos Scientific Laboratory, Los Alamos (Nov. 1955).
 7.  Krieger, H. L., et al., Radionuclide Analysis of Environmental Samples, Technical
    Report R59-6 (Revised), U.S. DHEW, Public Health Service (Feb. 1966).
 8.  Nuclear Science Series, Reports NAS-NS-3001  to NAS-NS-3115, National Academy
    of Sciences, National  Research Council Technical Information Center, Office of
    Information Services (1960).
 9.  "Water," Part 31 of 1977 Annual Book of ASTM Standards, American Society for
    Testing and Materials,  Philadelphia (1977).
10.  Jarvis, A. N., et al., Environmental Radioactivity Laboratory Intercomparison Studies
    Program 1975, EPA-680/4-75-002b, U.S. EPA,  Office of Research and Development,
    Las Vegas (1975).
11.  Ziegler, L. H., and Hunt, H. M., Quality Control for Environmental Measurements
    Using Gamma-Ray Spectrometry, U.S. EPA, EPA-600/7-77-144 (Dec. 1977).
12.  Kanipe, L. G., Handbook for Analytical Quality Control in Radioanalytical Labora-
    tories, U.S. EPA, EPA-600/7-77-088 (Aug. 1977).
                                   11-6               OCR(6/2001) by RYA-R10

-------
                                   Chapter 12

                                MICROBIOLOGY
12.1 Background

The quality assurance program described in this chapter is a synopsis of a detailed program
prescribed in parts IV and V of the EPA microbiological methods manual (1). Quality
assurance is a program of integration of all intralaboratory and interlaboratory quality
control  (QC), methods standardization, and QC management practices into  a formal,
coordinated, continuing effort.

12.2 Specific Needs in Microbiology

A quality assurance program for microbiological analyses must emphasize the control of
laboratory operations and analytical procedures because the tests measure living organisms
that continually change in response to their environment.  Further, because true values
cannot be provided for the microbial parameters, microbiologists do not yet have the
advantages of analytical  standards, QC charts, and spiked samples available to other
disciplines for measurement of accuracy. Because known values cannot be applied, it is im-
portant that careful and continuous control be exerted over sampling, personnel, analytical
methodology, materials, supplies, and equipment.

12.3 Intralaboratory Quality Control

Intralaboratory QC is the orderly application within  a single laboratory of laboratory
practices necessary to eliminate or reduce systematic  error and to control random error.
This within-laboratory program must be practical, integrated, and time-efficient or it will
be bypassed. When properly administered, such a program helps to insure high-quality data
without interfering with the primary analytical functions of the laboratory.  This within-
laboratory program should be supplemented by participation of the  laboratory in an
interlaboratory quality assurance program such as that conducted by EPA.

Intralaboratory QC for microbiology should  cover the following areas:

        a.     Laboratory operations

              (1)  Sample collection and handling
              (2)  Laboratory facilities
              (3)  Laboratory personnel
              (4)  Laboratory equipment and instrumentation
              (5)  Laboratory supplies
              (6)  Culture media
              (7)  Analytical methodology

        b.     Analytical QC

              (1)  Sterility checks
              (2)  Positive and negative controls
                                       12-1              OCR(6/2001)byRYA-R10

-------
               (3) Duplicate analyses
               (4) Single-analyst precision
               (5) Comparison of results between analysts
               (6) Verification of membrane filter analyses
               (7) Completion of most probable number analyses
               (8) Data handling

12.4 Interlaboratory Quality Control

An interlaboratory quality assurance program  is an agreed-upon system of minimal
requirements necessary to maintain a quality standard among a group of laboratories. Such
a program may be voluntary or compulsory for participants. It may include the following
activities:

  a.    Selection and approval of uniform sampling methodology and analytical methodol-
       ogy

  b.    Collaborative studies to establish the precision and accuracy of selected methodol-
       ogy

  c.    Preparation of guidelines to set minimal group standards for personnel, equipment,
       instrumentation, facilities, and intralaboratory QC programs

  d.   Onsite inspection of laboratory capabilities

  e.   Periodic evaluation of laboratory performance on unknown samples

  f.    Followup on problems identified in onsite inspections and performance evaluations

As a part of its interlaboratory quality assurance program, EPA has selected microbiological
methodology and standards for laboratory operations (1). EMSL-Cincinnati is currently
conducting research on the  development of QC samples for use in performance testing and
method-validation studies.

12.5 Development of a Formal Quality Assurance Program

Unless records are kept of the QC checks and procedures, there is no proof of performance,
no value  in future reference, and for practical purposes,  no quality assurance program in
operation. To insure a viable quality assurance program, management must first recognize
the need and require the development of a formal program, and then commit 15 percent of
the laboratory man-years to QC activities.  The laboratory manager holds meetings with
supervisors  and staff workers to  establish  levels of  responsibility and functions  of
management, supervisor, and analyst in the quality assurance program.  Laboratory person-
nel participate in planning and structuring the program.

Once the quality assurance program is functioning, supervisors review laboratory operations
and QC with analysts on a frequent (weekly) basis. Supervisors use the results of the regular
meetings with laboratory personnel to inform management of the status of the program on
a regular (monthly) basis. These meetings identify problems through participation of labora-
tory personnel and provide the backing of management for actions  required to correct
problems.
                                       12-2              OCR(6/2001)byRYA-R10

-------
12.6 Documentation of a Quality Assurance Program

A laboratory operating manual should be prepared that describes operation, maintenance,
and QC of laboratory operations and analyses as practiced. The review mechanisms and the
frequency of review in the  quality-assurance program are included.

12.6.1 Sampling

A sample log is used to record information on samples received in the laboratory including
details of sample identification and origin, the necessary chain-of-custody information,
analyses performed, and final results.

12.6.2 Laboratory Operations

A QC record  is maintained on media preparation, instrument calibration, purchase of
supplies, QC checks on materials, supplies, equipment, instrumentation, facilities, and
analyses.

12.6.3 Analytical Quality Control

A record of analytical QC checks is maintained on positive and negative controls, sterility
checks, single-analyst precision, precision between analysts, and  use-test  results from
comparison of lots of media, membrane filters, and other supplies.

12.7   Chain-of-Custody Procedures for Microbiological Samples

12.7.1 General

A regulatory agency must demonstrate the reliability of its evidence by proving the chain
of possession and custody of any samples that are offered for evidence or that form the basis
of analytical test results introduced into evidence in any water  pollution case.  It is
imperative that the office and the laboratory prepare procedures to be followed whenever
evidence samples are collected, transferred, stored, analyzed, or destroyed.

The primary objective of these procedures is to create an accurate written record that can
be used to trace the possession of the sample from the moment of its collection through its
introduction into evidence.  A sample is in custody if it is in any one of the following states:

  a.   In actual physical possession

  b.   In view,  after being in physical possession

  c.   In physical possession and locked up so that no one can tamper with it

  d.   In a secured area, restricted to authorized personnel

Personnel should receive copies of study plans prior to the study of a water pollution case.
Prestudy briefings should  then be held to apprise participants  of the objectives, sample
locations, and chain-of-custody procedures  to be followed.  After the  chain-of-custody
samples are collected, a debriefing is held in the field to verify the adherence to the chain-of-
custody procedures and to  determine whether additional  samples are required.
                                       12-3              OCR(6/2001)byRYA-R10

-------
12.7.2 Rules for Sample Collection

An agency or laboratory engaged in sample study activities should follow these rules:

  a.    Involve a minimum number of trained persons in sample collection and handling.

  b.    Establish guidelines for particular procedures to be used for each type of sample
       collection, preservation, and handling.

  c.    Handle samples as little as possible.

  d.    Obtain stream and effluent samples using the appropriate sampling techniques.

  e.    Attach sample tag or label securely (see fig. 12-1) to the sample  container at the
       time the sample is collected. The tag  should contain the  following items as a
       minimum:  the serial number of the tag, the station number and location, the date
       and time taken, the type of sample, the sequence number (e.g., first sample of the
       day-sequence No. 1), the preservative used, the analyses required,  and the name of
       the sample collector. Tags should be completed legibly in waterproof ink.

  f.    Use  bound field notebooks to  record field measurements and other pertinent
       information necessary to reconstruct the sample collection processes in the event
       of a later enforcement proceeding.  Maintain a separate set of field notebooks for
       each study and store them in a safe place where they can be protected and accounted
       for at all times. Establish a sample log sheet with a standard format to minimize
       field entries and include the serial number of the sheet, the date, time, survey, type
       of samples  taken, volume  of each sample, type of analyses, (unique) sample
       numbers, sampling location, field measurements (such as temperature, conductivity,
       dissolved oxygen (DO),  and pH),  and any  other pertinent  information or
       observation. (See fig. 12-2.) The entries should be signed by the sample collector.
       The responsibility for preparing and retaining field notebooks during and after a
       study should be assigned to a study coordinator or his designated  representative.

  g.    The sample collector is responsible for the care and custody of the samples until the
       samples are properly dispatched to the receiving laboratory or given to an assigned
       custodian.  The sample collector must insure that each container is in his physical
       possession or in his view at all times, or stored in a locked place where no one can
       tamper with it.

  h.    Take color slides or photographs of the sample locations and any visible water
       pollution. Sign and indicate time, date, and site location on the back of the photo.
       To prevent alteration, handle such photographs according to the established chain-
       of-custody procedures.

12.7.3 Transfer of Custody and Shipment

Intransfer-of-custody procedures, each custodian or sampler must sign, record, and date the
transfer.  Most environmental regulatory agencies develop chain-of-custody procedures
tailored to their needs. These procedures may vary in format and language but contain the
same essential elements. Historically, sample transfer under chain of custody has been on
                                      12-4              OCR(6/2001)byRYA-R10

-------
r
U»S. ENVIRONMENTAL PROTECTION AGENCY
Station No. Date Time Sequence No.
Station Location grab
	 BOD 	 1 Metals
_____ Solids . Oil and Grease
	 COD - D.O.
' 	 Nutrients 	 Bact.
	 Other
Samplers:
Remarks/Preservative:
(a)
                 ENVIRONMENTAL PROTECTION AGENCY
                              (Local Address)
 (b)
       Figure 12-1. Example of chain-of-custody sample tag. (a) Front. (£) Back.
a sample-by-sample basis, which is awkward and time consuming.  However, the EPA
National Enforcement Investigation Center (NEIC) at Denver has set precedent with its bulk
transfer of samples. Bulk transfer is speedier and reduces paperwork and the number of
sample custodians.  The following description of bulk transfer of custody is essentially that
of the Office of Enforcement (2).
                                    12-5
OCR(6/2001) by RYA-R10

-------














LU
<
a






flfi SHFFT NO
LU
_l
O.
<
»
SIIRVFY
cc
o
u_




Q



3




co
Ul
ui
>
_i
<
2





S5
CC
UJ
Ou
2
OT
LU
-1
a.
<
UJ
LL
O
Ul
a.
t-


3QINVAO
lONBHd
SOINVDUO 30VUJ.
9«3H
S3QIOIlS3d
I10V8
S1W13W
3SV3H90NV1IO
Aiiaisdni
wuojnooivoaj
waOjnoDnvioi
.3UniVU3dW3J.
.AllAUDnONOO
.Hi
oa
Aimnvxnv
sanos asaNadsns
sanos nvioi
001
GOD
aos
SlN3ldinN
PRESERVATIVE
U3NIV1NOO 3dAl
3WmOA1Vi01
STATION DESCRIPTION
z a
O u
il


































































































































































































































































































































i
i
=
z
Si
§



















2 ]
! i






























                                       CQ
                                       X
                                      W
                                       8
                                      I
12-6
OCR(6/2001) by RYA-R10

-------
a.     Samples must be accompanied by a chain-of-custody record that includes the name
      of the study, collectors' signatures, station number, station location, date, time, type
      of sample, sequence number, number of containers, and analyses required. (See fig.
      12-3.) When turning over possession of samples, the transferor and transferee sign,
      date, and time the record sheet. This record sheet allows transfer of custody of a
                        CHAIN OF CUSTODY RECORD

STATION
NUMBER












STATION LOCATION













DATE












Relinquished bv:«.»n«u,,/
Relinquished by:  Date/Time
Received by Mobile Laboratory for field Date/Time
analysis: rj,,™,«.,/
Date/Time
1
Received for Laboratory by: Date/Time
Method of Shipment:
D stribution Ortg. -Accompany Shipment
1 Copy-Survey Coordinator Fed Files
                Figure 12-3.  Example of chain-of-custody record.
                                    12-7
OCR(6/2001) by RYA-R10

-------
       group of samples in the field to the mobile laboratory or to the central laboratory.
       When a custodian transfers a portion of the samples identified on the sheet to the
       mobile laboratory, the individual samples must be noted in the column with the
       signature of the person relinquishing the samples. The laboratory person receiving
       the samples acknowledges receipt by signing in the appropriate column.

  b.    If the custodian has not been assigned, the field custodian or field sampler has the
       responsibility of packaging and dispatching samples to the laboratory for analysis.
       The dispatch portion of the chain-of-custody record must be filled out, dated, and
       signed.

  c.    To avoid breakage, samples must be carefully packed in shipment containers such
       as ice chests. The shipping containers are padlocked for shipment to the receiving
       laboratory.

  d.    Packages must be accompanied by the chain-of-custody record showing identifica-
       tion of the contents. The original must accompany the shipment. A copy is retained
       by the survey coordinator.

  e.    If sent by mail, register the  package with return receipt requested. If sent by
       common carrier, a Government bill of lading should be obtained. Receipts from
       post offices and bills of lading will be retained as part of the permanent chain-of-
       custody documentation.

  f.    If delivered to the laboratory when  appropriate personnel are not there to receive
       them, the samples must be locked in a designated area within the laboratory, so that
       no one can tamper with them or must be placed in a secure area. The recipient must
       return to the laboratory, unlock the samples, and deliver custody to the appropriate
       custodian.

12.7.4 Laboratory Custody  Procedures

Suitable laboratory procedures during custody of samples include the following:

  a.    The laboratory shall designate a sample custodian and an alternate custodian to act
       in his absence. In addition, the laboratory shall set aside a sample  storage security
       area. This should be a clean, dry, isolated room with sufficient refrigerator space
       that can be securely locked from the outside.

  b.    Samples should be handled by the minimum possible number of persons.

  c.    Incoming samples shall be received only by the  custodian who will indicate receipt
       by  signing the chain-of-custody record sheet accompanying the  samples and
       retaining the sheet as a permanent record. Couriers picking up samples at the airport
       or post office shall sign jointly with the laboratory custodian.

  d.    Immediately upon receipt, the custodian places the samples in the sample room,
       which will be locked at all times except when samples are removed or replaced by
       the custodian.  To the maximum extent possible, only the  custodian shall be
       permitted  in the sample room.
                                      12-8              OCR(6/2001)byRYA-R10

-------
  e.   The custodian shall insure that microbiological samples are properly stored and
       maintained at 4°C.

  f.   Only the custodian will distribute samples to personnel who are to perform tests.

  g.   The analyst records in his laboratory notebook or analytical worksheet, identifying
       information describing the sample, the procedures performed, and the results of the
       testing. The notes shall be dated, shall indicate who performed the tests, and should
       include any abnormalities that occurred during the testing procedure. The notes
       shall be retained as a permanent record in the laboratory. In the event that the
       person who performed the tests is not available as a witness at the time of a trial, the
       Government  may be able to introduce the notes in  evidence under the Federal
       Business Records Act.

  h.   Approved methods of laboratory analyses shall be used as required by Public Laws
       92-500, 93-523,  92-532, and amendments.

  i.   Laboratory personnel are responsible for the care and custody of a sample once it
       is handed to them and should be prepared to testify  that the sample was  in their
       possession and view or secured in the laboratory at all times from the moment it
       was received from the custodian until the tests were run.

  j.   The laboratory area shall be maintained as a secured area and shall be restricted to
       authorized personnel.

  k.   Once the sample analyses are completed, the unused portion of the sample,  together
       with identifying labels and other documentation, must be returned to the custodian.
       The returned, tagged sample should  be  retained  in  the  custody room until
       permission to destroy the sample is received by the custodian.

  1.   Samples should be destroyed only upon the order of the laboratory director,  in
       consultation with previously designated enforcement officials, or when it is certain
       that the information is no longer required, or that the samples have deteriorated.
       The same destruction procedure is true for tags and laboratory records.

12.7.5 Evidentiary Considerations

Reducing chain-of-custody procedures and promulgated analytical procedures to writing will
facilitate the admission of evidence under Rule 803 (6) of the Federal Rules of Evidence
(Public Law 93-575).  Under this statute, written records of regularly conducted business
activities may be introduced into evidence as an exception to the hearsay rule without the
testimony of the person(s) who made the record. Although it would be preferable, it is not
always possible for the individuals who collected, kept, and analyzed samples to testify in
court.  In addition, if the opposing party does not intend to contest the integrity of the
sample or testing evidence, admission under Rule 803(6) can save a great deal of trial time.
For these reasons, it is important that the procedures followed in the collection and analyses
of evidentiary samples be standardized and described in an instruction manual, which, if
need be, can be offered as evidence of the regularly conducted business activity followed
by the laboratory or office in generating any given record.
                                       12-9              OCR(6/2001)byRYA-R10

-------
In criminal cases, however, records and reports of matter observed by police officers and
other law enforcement personnel are not included under the business record exceptions to
the hearsay rule previously cited. (See Rule 803(8), Public Law 93-595.) It is arguable that
those portions of the compliance inspection report dealing with matters other than sampling
and analysis results come within this exception. For this reason, in criminal actions, records
and reports of matter observed by field investigators may not be admissible, and the
evidence may still have to be presented in the form of oral testimony by the person(s) who
made the record or report, even though the materials come within the definition of business
records. In a criminal proceeding, the  opposing counsel may be able to obtain copies  of
reports prepared by witnesses (even if the witness does not refer to the records while
testifying), which may be used for cross-examination purposes.

Admission of records is not automatic under either of these sections. The business records
section authorizes admission "unless the source of information or the method or circum-
stances of preparation indicate lack of trustworthiness," and the caveat under the public
records exception reads "unless the sources of information or other circumstances indicate
lack of trustworthiness."

Thus, whether or not the inspector anticipates that a report will be introduced as evidence,
the inspector should make certain that the report is as accurate and objective as possible.

12.8 References

1.  Winter, J. A., Bordner, R., and Scarpino, P., Microbiological Methods for Monitoring
    the Environment, Part I - "Water and Wastes," U.S. EPA, EMSL, Cincinnati (1978).
2.  NPDES Compliance Sampling Manual, U.S. EPA, Office of Water Enforcement (June
    1977).
                                      12-10             OCR(6/2001)byRYA-R10

-------
                                  Chapter 13

                              AQUATIC BIOLOGY
Quality assurance guidelines for aquatic biology programs (fully described in ref. 1) are
summarized in the following section.

13.1 Summary of General Guidelines

Successful  quality assurance programs in aquatic biology are based, on  the following
essential elements:

  a.   An understanding and acceptance of the importance of quality control (QC) and a
       commitment on the part of the biology staff to fully integrate QC practices into field
       and laboratory operations

  b.   A staff with adequate formal training and experience and proper specialization to
       meet program needs

  c.   Adequate field  equipment, storage and laboratory space, instrumentation, and
       taxonomic references

  d.   Careful advance  preparation and design of field and laboratory studies

  e.   Strict adherence to approved methodology, where available, and careful consider-
       ation of the technical defensibility of the methods and their application

  f.    Use of replication in sample collection and analysis where feasible, and determina-
       tion of the accuracy and precision of the data

  g.   Frequent calibration of field and laboratory instruments

  h.   Proper sample identification and handling to prevent misidentification or inter-
       mixing of samples

  i.    Use of blind, split, or other control samples to evaluate performance

  j.    Development and regular use of in-house reference specimen collections, and use
       of outside taxonomic experts to confirm or provide identifications for problem
       specimens

  k.   Meticulous, dual-level review of the results of manual arithmetical data manipula-
       tions  and transcriptions  before the data are used in reports or placed in BIO-
       STORET (2)

  1.    Participation in EPA formal interlaboratory aquatic biology methods studies, and
       use of EPA biological reference materials

  m.   Documentation of methodology and QC practices employed in the program
                                      13-1             OCR(6/2001)byRYA-R10

-------
13.2 Discussion

A brief description of each of the areas mentioned in section  13.1 is provided here to
indicate the scope of the quality assurance program.

13.2.1 Staff Commitment

To establish and maintain an effective quality assurance program in aquatic biology, the
supervisor must actively support and frequently monitor the use of QC practices in all
aquatic biology activities. This will require the commitment of 10 to 15 percent of the total
manpower resources. The supervisor must review field and laboratory operations with his
staff frequently (weekly) to insure that QC practices are followed and properly documented.

13.2.2 Staff

The quality and reliability of the data rest heavily on the competence of the staff. The range
of aquatic organisms studied by biological programs is very broad, and each community
requires unique skills in sample collection and analysis, and in data interpretation. Several
disciplines,  therefore,  must be  represented on the staff to deal  effectively with the
taxonomy and  ecology  of the major groups  of aquatic organisms, which  include the
phytoplankton, zooplankton, periphyton, macrophyton, macroinvertebrates, and fish.

13.2.3 Facilities

The quality of the data also depends upon the  availability and performance of laboratory
equipment. Such items as sampling gear, current meters, spectrophotometers, and micro-
scopes must be available and must meet performance standards related to the biological
parameters measured. Laboratory instrumentation must provide the sensitivity and accuracy
required by the state of the art in sample analysis.  Adequate laboratory and storage space
must also be provided.

13.2.4 Advance Planning

Thorough advance planning of field and laboratory projects is necessary to maintain the
required control over the technical aspects of the project and to insure the collection of
meaningful data. Factors taken into consideration include the objectives of the study, the
parameters to  be measured,  station selection, the sampling frequency and  replication,
seasonal cycles in the properties of communities of aquatic organisms, and QC measures to
be incorporated into the various phases of the project.

13.2.5 Use  of Approved Methodology

Methods in the EPA manual "Biological Field and Laboratory Methods for Measuring the
Quality  of Surface  Waters and Effluents" (1) should be employed where applicable. The
manual  contains concensus methods selected  by a committee of senior Agency aquatic
biologists as the preferred methods for use within EPA.  If program activities require
methods for parameters not covered by reference 1, methods may be selected from other-
sources  if their application is technically defensible.
                                       13-2              OCR(6/2001)byRYA-R10

-------
13.2.6 Replication

Comparisons of biological parameters measured in control and affected regions of water
bodies, and in laboratory experiments, may be meaningless if the precision of the results is
not known. Preliminary measurements must be made in each study to determine the scatter
in the data and to establish the number of replicate samples required to achieve the level of
precision required to detect differences in the responses measured.

13.2.7 Instrument Calibration and Maintenance

Sampling equipment and laboratory instrumentation with mechanical metering devices and
electronic components are calibrated on a regularly scheduled basis to insure the accuracy
of the data. Standards are obtained, such as NBS-certified thermometers for temperature-
measuring devices, class-S weights for balances, and absorption filters for spectrophotom-
eters. Records of calibrations, regular performance checks, and service for each device are
maintained in bound log  books in such a manner that the history of performance of the
instruments may be easily reviewed.  Analytical reagents  are labeled and dated when
received, and are protected from deterioration if labile.  The expected shelf life of each
reagent is recorded on the label, and the material is not used after the expiration date.

13.2.8 Sample Labeling

Samples are securely labeled in the field and recorded  in a bound log. Information on the
label should include the station, date, time of day, depth,  and other relevant information.
A unique lot number is assigned to the sample and recorded on the label. Waterproof paper
and ink must be used for the labels, and are recommended for the field logs. Depending on
requirements, labels are placed inside  samples such as  macroinvertebrates.

13.2.9 Quality Control Samples

The  accuracy of the data from routine  analyses such as  counts  and identification of
organisms and chlorophyll and biomass measurments is determined by introducing blind,
split,  or reference  samples in the sample processing  stream.   These samples are either
prepared by the supervisory aquatic biologist, laboratory quality control officer, or analytical
QC coordinator, or are obtained from EMSL-Cincinnati.  The results are discussed at
regularly scheduled staff meetings and any problems identified are discussed and corrected.

13.2.10 Organism Identification and Reference Specimens

Accurate identification of aquatic organisms to the species  level is essential to the
interpretation of biological data.  A set of reference specimens is established within  each
laboratory, to be used as taxonomic (identification) standards in processing samples and in
training new personnel. The set is representative of the aquatic organisms collected by the
program, and each specimen embodies the morphological characteristics essential to the
identification of that taxon.   The identity of these  specimens is verified  by outside
taxonomic experts, who also examine organisms that pose unusually difficult identification
problems in routine sample analysis.
                                       13-3              OCR(6/2001)byRYA-R10

-------
13.2.11 Data Records, Editing (Proofing), Review

Data collected manually are entered in a bound log or on specialized bench sheets that fully
describe the origin and nature of the sample and that are maintained in a binder or file.
Source data such as organism abundance, metabolic rates, and chlorophyll, which are
manually or electronically manipulated  or transcribed from one record to another, are
doublechecked by a second person.  All manual calculations and all electronic calculations,
where data are manually keyboarded, are  performed twice, except where the source (input)
data are included in the output and can be proofed.  Keyboarded data are carefully proofed
before they are submitted for computer manipulation.

13.2.12 Interlaboratory Methods Studies

The aquatic biology programs participate in formal interlaboratory biological methods
studies performed by EMSL-Cincinnati.  Studies  on chlorophyll and macroinvertebrate
identification methods have been completed, and additional studies on phytoplankton and
periphyton identification methods are planned in the future.

13.2.13 Documentation for the Quality Assurance Program

A laboratory operations manual is available (1) that describes the scope of the program,
organizational structure,  qualifications  of the staff,  available space  and equipment,
methodology employed for sample collection and analysis, and QC procedures.

13.3 References

1. Weber, C. I., Editor, Biological Field and Laboratory Methods for Measuring the
    Quality of Surface Waters and Effluents, 2d Edition, EPA-670/4-73-001, U.S. EPA
    (July 1973).
2. Nacht,  L., and Weber,  C. I., BIO-STORET Final Design  Specification, U.S. EPA
    (1976).
                                      13-4             OCR(6/2001)byRYA-R10

-------
                                   Chapter 14

                            LABORATORY SAFETY

14.1 Law and Authority for Safety and Health

Public Law 91-596 is the Occupational Safety and Health Act of 1970 (1). The purpose of
this act is the following:

       To assure safe and healthful working conditions for working men and
       women; by authorizing enforcement of the standards developed under the
       act; by assisting and encouraging the States in their efforts to assure safe
       and healthful working conditions; by providing for research, information,
       education, and training in the field of occupational safety and health; and
       for other purposes.

The intent of the act (1) is "to assure so far as possible every working man and woman in
the Nation safe and healthful working conditions and to preserve our human resources."

The responsibility for promulgating and enforcing occupational safety and health standards
rests with the Department of Labor.  The Department of Health, Education, and Welfare is
responsible for conducting research on which new standards can be  based, and  for
implementing education  and training programs for  producing an adequate supply of
manpower to implement the purposes of the act. These responsibilities are performed by
the National Institute for Occupational Safety and Health (NIOSH).

Section 19(a) of Public Law 91-596 states the following (1):

       It shall be the responsibility of the head of each Federal agency to establish
       and maintain an effective and comprehensive occu ational safety and health
       program that is consistent with the standards promulgated under section 6.
       The head of each agency shall (after consultation with representatives of
       the employees thereof)-

          (1)   provide safe and healthful places and conditions of employment,
               consistent with the standards set under section 6;

          (2)   acquire,  maintain,  and require the use of safety equipment,
               personal protective equipment, and devices reasonably necessary
               to protect employees;

          (3)   keep adequate records of all occupational accidents and illnesses
               for proper evaluation and necessary corrective action;

          (4)   consult with the  Secretary with regard to the adequacy as to form
               and content of records kept pursuant to subsection (a)(3) of this
               section; and

          (5)   make an annual  report to the Secretary of Labor with respect to
               occupational accidents  and injuries  and the  agency's  program
               under this section. Such report shall include any report submitted
               under section 7902(e)(2) of title 5, United States Code.
                                      14-1              OCR(6/2001)byRYA-R10

-------
Executive Order 11807 was issued in the fall of 1974 to provide general direction to the
heads of Federal agencies and to the Secretary of Labor in establishing occupational safety
and health programs in the Federal Government. The executive order details the following
duties:

  a.   Appointment of a Safety and Health Official.

  b.   Establishment of a management information system.

  c.   Establishment of an occupational safety and health program.

       (1)     Adoption of safety and health standards as effective as the  Secretary of
               Labor's standards.

       (2)     Institution  of procedures forprocessing  reports  from  employees on
               hazardous conditions.

       (3)     Institution of periodic inspection of facilities.

       (4)     Provision for abatement of hazards in facilities.

  d.   Provision for training of agency personnel.

       (1)     Training  of supervisors at all levels.

       (2)     Training  of those responsible for conducting inspections of facilities.

       (3)     Training  of other employees.  Attention is called to the list ofOSHA
               training requirements.

  e.   Assistance of the Secretary of Labor.

       (1)     Compliance with the recordkeeping and reporting requirements.

       (2)     Observation of the guidelines issued in 28 CFR 1960.

       (3)     Cooperation with the Secretary of Labor in the performance of his responsi-
               bilities.

  f.    Issuance of guidelines for a safety and health program.

  g.   Prescription of recordkeeping and reporting requirements.

  h.   Provision of  consulting services to Federal agencies in adoption of standards,
       training agency personnel, and in other matters.

  i.    Upon  request and  subject  to reimbursement, performance of such services as
       evaluation of safety and health conditions in the agency, recommendation on the
       adoption of standards, inspection of facilities for safety and health hazards, and
       training of agency personnel in safety and health matters.
                                       14-2             OCR(6/2001)byRYA-R10

-------
  j.    Evaluation of agency programs and report of the condition to the President. This
       item in the executive order gives to the Secretary of Labor a degree of enforcement
       of safety and health rules and regulations within Federal agencies.

  k.   Section 4 continues the Federal Advisory Council on Occupational Safety and
       Health. Note that at least one-third of the members are to be labor representatives.

14.1.1 Occupational Safety and Health Administration Regulations

The Occupational Safety and Health Act of 1970 covers laboratory workplaces as well as
industrial and manufacturing workplaces (1). Many large laboratories have already come
under the scrutiny of the Occupational Safety and Health Administration (OSHA); small and
medium-sized laboratories can expect direct involvement in the future.

The 1970 act set up NIOSH within the U.S. Department of Health, Education, and Welfare.
NIOSH provides OSHA with the scientific information it needs to effectively perform its
regulatory function.

The initial OSHA regulations were, and in some measure still are, a collection of many well-
established standards taken from industry and from standards-making groups. The merican
National Standards Institute, the National Safety Council, the National Fire Protection
Association, and the American Society for Testing and Materials were some of the prime
sources for these OSHA standards. Certain existing State health and safety regulations that
predated OSHA were used to develop the Federal OSHA regulations.

Efforts are continually underway to refine and quantify current OHSA regulations through
court decisions resulting from appeals of compliance citations, through continuing reviews
of standards by OSHA safety compliance officers, through the work of NIOSH, and through
internal reviews.

Because of the  continual  publication of revisions of the  regulations  consisting  of
refinements and clarifications,  as well as dissemination of newly issued  regulations,
laboratory operators  must expend great efforts to keep strictly up to date.  What was
permitted yesterday may not be permitted today.

The first place to start in keeping up to date on OSHA regulations is to obtain a copy of the
Federal or State regulations that apply.  In States not operating their own OSHA function,
copies of applicable regulations are usually available without charge from the local office
of the U.S.  Department of Labor. In other States, copies of their regulations are available
from similar State offices and boards either free or at a nominal charge.  In either case,
appropriate steps must be taken to keep up to date on revisions and reissues. Usually this
means getting on a mailing list for  automatic receipt of this material as it is published.
Because entire sections  of the  regulations may be reissued to incorporate only a few
changes, it is usually necessary to completely study these reissues to determine the actual
change that has been made.

As OSHA moves more into the complicated areas of chemical and biological laboratory
workplaces, the level of the compliance  inspector's  knowledge  must be increased
accordingly.
                                      14-3              OCR(6/2001)byRYA-R10

-------
14.1.2 Federal and State OSHA

There is no completely identical OSHA standard in force throughout the United States.

A provision of the act of 1970 permits Federal OSHA to certify individual State OSHA
operations  when they are satisfied  that applicable  State  regulations and compliance
enforcement methods fulfill the provisions of the Federal act (1).

While many State OSHA applications are in process, the following States and territories are
certified or at least provisionally certified to permit their own OSHA operation:

       Alaska
       Arizona
       California
       Colorado
       Connecticut
       Hawaii
       Indiana
       Iowa
       Kentucky
       Maryland
       Michigan
       Minnesota
       Nevada
       North Carolina
       Oregon
       South Carolina
       Tennessee
       Utah
       Vermont
       Washington
       Wyoming
       Virgin Islands

All other states and territories are under Federal OSHA jurisdiction.

Even though  the Federal Government supports 50 percent of the cost of State OSHA
programs, these States apparently would prefer that the Federal Government take on the
entire cost burden for them.

An important difference between Federal and  State OSHA operations  lies  in  their
jurisdictions.  While State OSHAs are responsible for State-operated workplaces, Federal
OSHA can only inspect Federal workplaces.   It cannot enforce compliance.  There is,
however, a current effort by the General Accounting Office to urge Congress to remove this
restriction on Federal OSHA's effectiveness in dealing with Federal workplaces.

While Federal OSHA is not permitted to offer a consulting service, most State OSHAs can
and do. Some who were unaware of this difference have called local Federal OSHA offices
for advice.  All they succeeded in doing was to invite an inspection. Normally, however,
                                      14-4              OCR(6/2001)byRYA-R10

-------
inspections  of facilities, be  they manufacturing or laboratories, are triggered by an
employee's complaint or an accident report, either as required by law or through police, fire,
or hospital channels.

Reactions to OSHA operations vary widely.  Many laboratories have been cited for
violations and have been heavily fined. Yet Dr. David Pettit, Technical Director of Kelco's
operations in San Diego, says, "We feel that OSHA standards are not unrealistic, just good
laboratory practice."

14.2 EPA Policy on Laboratory Safety

EPA has issued an Occupational Safety and Health Manual (3) in which policy, responsi-
bilities, accident reporting, inspections, standards, and training are described. The Agency
has also  issued  a Safety Management Manual (4)  that contains  information on safety
protection plans at EPA facilities and air, water, and road safety.

The EPA Occupational Safety and Health Manual (3) establishes policy, responsibilities,
and procedures for the conduct of the EPA safety and health program.

The policy of EPA is to administer its programs in a manner that assures adequate protection
of its own employees and property, and that for which it has a responsibility. Every manager,
supervisor, and employee is responsible for identifying risks, hazards, and unsafe situations
or practices and for taking steps to insure  adequate safety  in the  activities under his
supervision.

A facility safety officer designated for each unit must be responsible for assisting the
officer-in-charge in developing, organizing, directing, and evaluating the safety and health
program  and coordinating illness and inquiry reporting and recordkeeping  requirements;
analyzing accidents and injuries for prevention and control; and providing technical advice
in the implementation of program standards and policy.

Safety in any laboratory requires continuing attention. Use of new or different techniques,
chemicals, and equipment requires careful reading, instruction, and supervision, and may
require consultation with other people with special knowledge or experience.

Prevention of laboratory accidents requires positive attitudes toward safety and training, and
suitable  information  for  understanding laboratory and  chemical hazards  and  their
consequences.

Responsibility for safety within the laboratories for an organization may be considered to
exist at three different levels-individual, supervisory (or instructional), and organizational
(or institutional).

The division of responsibility needs to be clearly assigned and accepted, steps need to be
taken to see that the responsibilities are exercised, and the assignments need to be reassessed
if unexpected problems develop.

Each individual who works in a laboratory is responsible for learning the health and safety
hazards of the chemicals he will be using or producing, and the hazards that may occur from
                                       14-5              OCR(6/2001)byRYA-R10

-------
the equipment  and techniques  he will employ,  so that he can design his setup and
procedures to limit the effects  of any accident. The individual should investigate any
accident that occurs, and record and report the apparent causes and the preventive measures
that may be needed to prevent similar accidents.

The supervisor has the responsibility for giving all the necessary directions, including the
safety measures to be used, and the responsibility of seeing that employees carry out their
individual  responsibilities.  Whoever directs the activities of others  has a concurrent
responsibility to prevent accidental injuries from occurring as a result of the activities.

The organization  of which the laboratories are part has a fundamental responsibility to
provide the facilities, equipment, and maintenance for a safe working environment and to
provide an organized  program to make the improvements necessary for a safe working
environment.  Unless the organization fulfills its responsibilities, it cannot expect its
supervisors, employees, or students to fulfill their responsibilities for laboratory safety.

The Environmental Protection Agency has a designated safety and health official who is
responsible for assuring that formal safety and health inspections are conducted at all EPA
workplaces.  He will  notify the Administrator regarding uncorrected  safety and health
deficiencies.

14.2.1 Formal Safety Inspections

Formal safety and health  inspections at workplaces where there is an increased risk  of
accident, injury, or illness because of the nature of the work performed, as in the case  of
chemical operations and material-handling or material-loading operations, must be made by
a safety and  health specialist.  A "Safety  and Health Specialist" is defined in 29 CFR
1960.2(h) as  a person who meets the Civil Service standards for the position of Safety
Manager/Specialist GS-018, Safety Engineer GS-803, Fire Protection Engineer GS-804,
Industrial  Hygienist GS-690, Fire  Protection  Specialist/Marshall GS-081, or Health
Physicist GS-1306, or who  is an employee of an equally qualified military agency  or
nongovernment organization.

Formal safety and health inspections need not be made by a safety and health specialist at
workplaces where there is little risk involved, but should be conducted by a person having
sufficient training and experience in the safety and health needs of the workplaces involved
to adequately perform the duties of an inspector as set forth in Executive Order 11807.
Inspectors should be accompanied on formal safety and health inspections by representatives
of the  officer in charge of the reporting unit being inspected and representatives of the
employees of such establishments. Management and employee representatives should be
familiar with and maintain OSHA standards.

To insure safe and healthful working conditions for EPA employees,  safety and health
inspectors  are authorized to enter without delay and at reasonable times, any building,
installation, facility, construction site, or other area, workplace, or environment where work
is performed by employees of the Agency, to inspect and investigate during regular working
hours and  at other reasonable times, and  within reasonable limits and in a reasonable
manner, any such place of employment and all pertinent conditions, structures, machines,
                                       14-6             OCR(6/2001)byRYA-R10

-------
apparatus, devices, equipment, and  materials therein; and to question privately any
employee, or any supervisory employee, or any officer in charge of a reporting unit.

14.2.2 Informal Safety Inspections

An informal safety and health inspection is performed on either a scheduled or unscheduled
basis by the facility safety officer, regional safety officer, supervisory management, or
members of a safety and health committee. EPA Form 1440-2, Health and Safety Inspection
Checklist (fig. 14-1), shall be used by the inspector or spokesman of a safety and health
committee conducting the inspection to note safety and health deficiencies identified during
the inspection process.

A more detailed checklist for safety evaluation of the laboratory is given in appendix A.
This inspection list contains many specific recommendations and guidelines for laboratory
safety.

14.3 Laboratory Safety Practices*

14.3.1 Introduction

14.3.1.1 Safe  Use,  Handling, and Storage of Chemicals

Chemicals in any form can be safely stored, handled, and used if their hazardous physical
and chemical properties are fully understood and the necessary precautions, including the
use of proper safeguards and personal protective equipment are observed.

The management of every unit within a manufacturing establishment must give whole-
hearted support to a well-integrated safety policy.

14.3.1.2 General Rules for Laboratory Safety

Supervisory personnel should think "safety." Their attitude toward fire and safety standard
practices is reflected in the behavior of their entire staff.

A safety program is only as strong  as the worker's will to do the correct things at the right
time.

The fundamental weakness of most safety programs lies in too much lip service to safety
rules and not enough  action in putting them into practice.

Safety practices should be practical and enforceable.

Accident prevention is based on certain common standards of education and training of
personnel, and provision of safeguards against accidents.
*This description was prepared by Paul F. Hallbach, Chemist, National Training and
Operational Technology Center, U.S. EPA, Cincinnati, Ohio 45268.

                                       14-7              OCR(6/2001)byRYA-R10

-------
HEALTH AND SAFETY INSPECTION CHECKLIST


TITLE

DATE


ITEM
4. MATERIALS HANDLING EQUIPMENT
6. LADDERS AND STAIRS
7. FLOORS. PLATFORMS AND RAILINGS
8. EGRESS
9. LIGHTING
10, VENTILATION

13. FIRE DETECTION DEVICES

IS. FIRE SUF-P1ESBION line ludinf extinx uitht nl


SAT










UNSAT










ITEM
19. WATER (artn-iypftotf and croit-connecnoni)
2O. ELECTRICAL fjutft. /(rounding, etc.)
25. PORTABLE TOOLS

27. HAZARDOUS WARNING SIGNS

30 M.OICAL SERV.CES, INCLUDING FIRST AID


3t PROVISIONS FOB HANDICAPPED
32. OTHER


SAT










UNSAT










PART It. PROCEDURES AND INSTRUCTIONS (Check each applicable item)
33. MATERIALS HANDLING
34. MOTOR VEHICLE OPERATION

EQUIPMENT
fftrboH monoxide, radittion, etc. I


4V BOATING OPERATIONS
42 DIVING OPERATIONS
























49 FIELD OPERATIONS. OTHER





AND HEALTH
MOTIVATIONAL ACTIVITIES

























 EPA Foim 1*40-3 IR«v 5-771
                         PREVIOUS EOITIC
                   Figure 14-1. Health and safety inspection checklist.
14.3.2 Laboratory Design and Equipment




14.3.2.1  Type of Construction




The construction of the laboratory should generally be fire resistant or noncombustible.
                                      14-8
OCR(6/2001) by RYA-R10

-------
Multiple story buildings should have adequate means of exit.

Stairways should be enclosed with brick or concrete walls.

Laboratories should have adequate exit doors to permit quick, safe escape in an emergency
and to protect the occupants from fires or accidents in adjoining rooms. Each room should
be checked to make sure there is no chance of a person being trapped by fire, explosions,
or release of dangerous gases.

Laboratory rooms in which most of the work is performed with flammable liquids or gases
should be provided with explosion-venting windows.

14.3.2.2 Arrangement of Furniture  and Equipment

Furniture should be arranged  for maximum use of available space and should provide
working conditions that are efficient and safe.

Aisles between benches should be at least 4 ft wide to provide adequate room for passage
of personnel and equipment.

Desks should be isolated from benches or adequately protected.

Every laboratory should have an eyewash station and a safety shower.

14.3.2.3 Hoods and Ventilation

Adequate hood facilities  should  be installed where highly toxic or highly flammable
materials are used.

Hoods  should be ventilated separately and the exhaust  should be terminated at a safe
distance from the building.

Makeup air should be supplied to rooms or to hoods to replace the quantity of air exhausted
through the hoods.

Hood ventilation systems are best designed to have an airflow of not less than 60 ft/min
(linear) across the face of the  hood with all doors open, and 150 ft/min (linear) if toxic
materials are involved.

Exhaust fans should be sparkproof if exhausting flammable vapors and corrosive resistant
if handling  corrosive fumes.

Controls for all services should be located at the front of the hood and should be operable
when the hood door is  closed.

All laboratory rooms should have the air changed continuously at a rate depending on the
materials being handled.

Recent California OSHA regulations require the presence of a means of visual indication
of the existence  of the airflow in the hood and specify the height and type of hood exhaust
permitted.
                                      14-9             OCR(6/2001)byRYA-R10

-------
14.3.2.4 Electrical Services

Electrical outlets should be placed outside of hoods to afford easy access and thus protect
them from spills and corrosion by gases.

Noninterchangeable plugs should be provided for multiple electrical services.

Adequate outlets should be provided and should be of the three-pole-type to provide for
adequate grounding.

Rubber or nonconductive  composition shoe soles should be  required (except when
flammable vapors are present). Shoe soles should not be of atype that readily absorbs water
or other liquids.

14.3.2.5 Storage

Laboratories should provide for adequate  storage space for mechanical equipment and
glassware that will be used regularly.

Flammable solvents should not be stored in glass bottles over 11 in  size.  Large quantities
should be stored in metal safety cans.  Quantities requiring containers larger than 1 gal
should be stored outside the laboratory.

Explosion proof refrigerators  should be  used  for the  storage of highly volatile and
flammable solvents.

Cylinders of compressed or liquefied gases should not be stored in the laboratory.

Alphabetized storage of chemicals should be avoided to prevent the unintentional mixing
of two incompatible chemicals in an accident situation.

An appropriate antidote must be readily available for every stored chemical compound for
which an antidote is specified.

14.3.2.6 Housekeeping

Housekeeping plays an important role in reducing the frequency of laboratory accidents.
Rooms should be kept in a neat and orderly condition. Floors, shelves, and tables should
be kept free from dirt and from all apparatus and chemicals not in use.

A cluttered laboratory is a dangerous place to work.  Maintenance of a clean and orderly
work space is indicative of interest, personal pride, and safetymindedness.

Passageways should be kept clear to all building exits and stairways.

Metal containers should be provided for the disposal of broken glassware and should be
properly labeled.

Separate approved waste disposal cans should  be provided for the disposal of waste
chemicals.
                                      14-10             OCR(6/2001)byRYA-R10

-------
Flammable liquids not miscible with water and corrosive materials or compounds that are
likely to give off toxic vapors should never be poured into the sink.

Laboratory operators must be sure that unguarded rotating equipment such as belt-driven
vacuum pumps is provided with guards all the way around and that the guards are always
in place.

Whenever  heavy  laboratory equipment  must be moved frequently, rollers should be
provided.  In other cases, proper lifting equipment should be available.

14.3.2.7 Fire Protection

Laboratory personnel should be adequately  trained regarding  pertinent fire hazards
associated with their work.

Personnel should know rules of fire  prevention and methods of combating fires.

Fire extinguishers (CO2 type) should be provided at convenient locations and personnel
should be instructed in their use.

Automatic sprinkler systems are effective for the control of fires in chemical laboratories.

14.3.2.8 Alarms

An approved fire-alarm system should be provided.

Wherever a hazard of accidental release of toxic gases exists, a gas alarm system to warn
occupants to evacuate the building should be provided.

Gas masks of oxygen or compressed-air-type should be located near exits and selected
personnel trained to  use them.

14.3.3 Handling Glassware

14.3.3.1 Receiving, Inspection, and Storage

Packages containing glassware should be opened and inspected for cracked or nicked pieces,
pieces with flaws that may become cracked in use, and badly shaped pieces.

Glassware should be stored on well-lighted stockroom shelves designed and having a coping
of sufficient height around the edges to prevent the pieces from falling off.

14.3.3.2 Laboratory Practice

Select glassware that is designed for the type of work planned.

To  cut glass tubing  or a rod, make  a straight, clean cut with a cutter or file at the point
where the piece is to be severed. Place  a towel over the piece to protect the hands and
fingers, then break away from the body.
                                      14-11             OCR(6/2001) by RYA-R10

-------
Large size tubing is cut by means of a heated nichrome wire looped around the piece at the
point of severance.

When it is necessary to insert a piece of glass tubing or a rod through a perforated rubber
or cork stopper, select the correct bore so that the insertion can be made without excessive
strain.

Use electric mantels for heating distillation apparatus, etc.

To remove glass splinters, use a whisk broom and a dustpan.  Very small pieces can be
picked up with a large piece of wet cotton.

14.3.4 Gases and Flammable Solvents

14.3.4.1 Gas Cylinders

Large cylinders must be securely fastened so that they cannot be dislodged or tipped in any
direction.

Connections, gages, regulators, or fittings used with other cylinders must not be inter-
changed with oxygen cylinder fittings because of the possibility of fire or explosion from
a reaction between oxygen and residual oil in the fitting.

Return empty cylinders promptly with protective caps replaced.

14.3.4.2 Flammable Solvents

Store in well-ventilated designated areas.

Flashpoint: the temperature at which a liquid gives off vapor sufficient to form an ignitable
mixture with the air near the surface of the liquid or within the vessel used.

Ignition temperature: the minimum temperature required to initiate or cause self-sustained
combustion independently of the heating or heated element.

Explosive or flammable limits:  for most flammable liquids, gases, and solids there is a
minimum concentration of vapor in air or oxygen below which propagation of flame does
not occur on contact with a source of ignition. There is also a maximum proportion of vapor
or gas in air above which propagation of flame does not occur. These limit mixtures of
vapor or gas with air, which if ignited will just propagate flame, are known as the "lower
and higher explosive or flammable limits."

Explosive range: the difference between the lower and higher explosive or flammable limits,
expressed in terms of percentage of vapor or gas in air by volume.

Vapor density: the relative density of the vapor as compared with air.
                                      14-12             OCR(6/2001)byRYA-R10

-------
Underwater's Laboratories classification: a standard classification for grading the relative
hazard of the various flammable liquids. This classification is based on the following scale:

       Ether class              100
       Gasoline class         90 to 100
       Alcohol (ethyl) class   60 to 70
       Kerosene class        30 to 40
       Paraffin oil class       10 to 20

Extinguishing agents that are appropriate for each of the four classes of fires are required.

14.3.5 Chemical Hazards

14.3.5.1 Acids and Alkalies

Some of the most hazardous chemicals are the strong or mineral acids such as hydrochloric,
hydrofluoric, sulfuric, and nitric.

Organic acids are less hazardous because of their comparatively low ionization potentials;
however, such  acids as phenol (carbolic acid), hydrocyanic, and  oxalic are extremely
hazardous because of their toxic properties.

Classification of acids is according to mineral or organic composition.  Acids should be
stored together, except that perchloric acid should not be placed next to glacial acetic acid.
Picric acid should be stored separately.

14.3.5.2 Oxidizing Materials

Oxidizing agents, in contact with organic matter, can cause explosions and fire.  They are
exothermic and decompose rapidly, liberating oxygen, which reacts with organic  com-
pounds.

Typical hazardous oxidizing agents are-

       Chlorine dioxide
       Sodium chlorate; chlorates
       Potassium chromate
       Chromium trioxide
       Perchloric acid; perchlorates

14.3.5.3 Explosive Power

Many chemicals are explosive or form compounds that are explosive and should be treated
accordingly.

A few of the more common examples of this class of hazardous materials are-

       Acetylides
       Silver fulminate
                                      14-13             OCR(6/2001)byRYA-R10

-------
       Peroxides
       Peracetic acid
       Nitroglycerine
       Picric acid
       Chlorine and ethylene
       Sodium metal
       Calcium carbide

14.3.5.4 Toxicity

Laboratory chemicals improperly stored or handled can cause injury to personnel by virtue
of their toxicity.

There are four types of exposure to chemicals:

  a.   Contact with the skin and eyes

  b.   Inhalation

  c.   Swallowing

  d.   Injection

Special classes of toxic agents:

  a.   Carcinogens-laboratory operators should recheckthe OSHA 1974 regulations on
       carcinogens.

  b.    Mercury-complete cleaning of spills is essential for compliance with OSHA limits.

14.3.6 Precautionary Measures

14.3.6.1  Clothing and Personal Protective Equipment

Chemical laboratories should have  special protective clothing and equipment readily
available for emergency use and for secondary protection of personnel working with
hazardous materials.

Equipment should be provided for adequate-

  a.   Eye protection

  b.   Body protection

  c.   Respiratory protection

  d.   Foot protection

  e.   Hand protection
                                      14-14             OCR(6/2001)byRYA-R10

-------
14.3.6.2 Bodily Injury

Burns, eye injuries, and poisoning are the injuries with which laboratory people must be
most concerned.

First emphasis in the laboratory should be on preventing accidents. This means observing
all recognized safe practices using necessary personal protective equipment and exercising
proper control over poisonous substances at the source of exposure.

So that a physician can be summoned promptly, every laboratory should post the names,
telephone numbers, and addresses of doctors to be called in an emergency requiring medical
care.

A consulting physician should specify the type and extent of first aid materials required for
the laboratory.

14.4 Report of Unsafe or Unhealthful Condition

In EPA a procedure has been established for reporting an unsafe or unhealthful condition
by the employee or supervisor. The procedure also provides for communication between
the employees;  supervisors;  safety official; head of the  unit; and, in  the  case of an
unresolved report, the Department of Labor.

A sample of an unsafe or unhealthful condition reporting form is shown in figure 14-2, and
sample notice of an unhealthful or unsafe condition is shown in figure 14-3.

14.5 References

1.  Public Law  91-596, Occupational Safety and Health Act of 1970 (Dec. 29, 1970).
2.  Occupational Safety and Health Manual, U.S. EPA (Jan. 8, 1976).
3.  Safety Management Manual, U.S. EPA, TN 1440.1 (Dec. 4, 1972).
                                     14-15             OCR(6/2001)byRYA-R10

-------
                                                                                      RNN 1440.01 1
                           REPORT OF UNHEALTHFUL OR UNSAFE CONDITION
 BRIEF DESCRIPTION OP UNHEALTHFUL OR UNSAFE CONDITION
 OCCUPATIONAL SAFETY AND HEALTH STANDARD VIOLATED (If  LOCATION {Inctudt organization, Facility and Building;
 kno\vnj
 ACTION TAKEN BY SUPERVISOR
 SIGNATURE
                                     I EMPLOYING ORGANIZATION
EPA Form 1440-6 (Fl*v. 5-77) PREVIOUS EDITION MAY BE USED UNTIL
                                                                        A COPY OF THIS FORM MUST BE
                                                                        RETAINED BY THE ORGINATOR
               Figure 14-2.  Report of unhealthful or unsafe condition.
                                            14-16
OCR(6/2001) by RYA-R10

-------
NOTICE OF UNHEALTHFUL OR UNSAFE WORKING CONDITION
TO iNamr ofOffictr-in-Chttrge of Reporting Unit)
REPORTING UNIT
An inspection conducted by me on
following violation!*) of EPA Occ
with the Occupational Safety and

NO.


OR SECTION VIOLATED


FROM (Name of Inspector!:
Occupational Health and Safety Office
Washington. n.C. 20460
at reveal ad the
upational Health and Safety Standards. These Standards have been adopted in compliance
ha Health Act of 1970, PL 91-956, Section 19.
DESCRIPTION OF
VIOLATION

LOCATION Of
VIOLATION

NO. OF WORKING DAYS
BV WHICH VIOLATION
MUST BE CORRECTED
AND DATE

Subpart D of 29CFR 1960.33, Safety and Health Provisions for Federal Employees, requires that a copy of this Notice shall
be posted immediately in a prominent place at or near each place that the violationlsl referred to in the Notice occurred. The
Notice must remained posted until all violations cited therein are corrected, or for three (3) working* days, whichever period is
longer. A copy of this Notice shall be sent to the Health and Safety Committee of the establishment or Reporting Unit and to
any person (s) who made a report of the unhealthful or unsafe condition which precipitated this inspection pursuant to the
provisions of 29CFR 1960.31.
Subpart D of 29CFR 1960.34 requires the Officer-in-Charge of the Reporting Unit to immediately submit an abatement plan
to the Designated Agency Safety and Health Official, if, in his judgment, the correction of the violation will not be possible
within thirty (30) working days*. Such plan shall contain an explanation of the circumstances of the de ay in abatement; a
proposed timetable for the abatement, and a summary of steps being taken in the interim to protect employees. A copy of
the plan shall be sent to the Health and Safety Committee of the establishment or Reporting Unit and to any person(s) who
made a report of the unhealthful or unsafe condition which precipitated this inspection pursuant to the provisions of 29CFR
1960.31.
'Under the Occupational Safety and Health Act, the term "Working Day" means Monday through Friday but does not
include Saturday, Sunday, or Federal Holidays.
SIGNATURE OF HEALTH AND SAFETY OFF ICER
DATE
EPA Hq Form 1440-8 I8'77f
           Figure 14-3. Notice of unhealthful or unsafe condition.
                                 14-17
OCR(6/2001) by RYA-R10

-------

-------
                             APPENDIX A




SUGGESTED CHECKLIST FOR THE SAFETY EVALUATION OF EPA LABORATORY AREAS
                                A-l          OCR(6/2001)byRYA-R10

-------
Organization:
Location:	
Room(s):	
	 Date:
 Building:
 By:	!
Item Inspected
Yes
No
N.A.
Comments
                                           Fire Prevention

  1.   Is fire alarm facility available? [.36(b)(7)]

  2.   Are all exits maintained to provide free and unob-
      structed egress from all parts of buflding? [.36(b)(4)]

  3.   Are all exits free of locks or fastening devices that
      could prevent free escape? [.36(b)(4)]

  4.   a. Is the fire detection system in working order?
      b. Is the sprinkler system in working order?
      c. Are fire doors in working order? [.36(d)(2)]
  5.   Are corridors and hallways at least 44 in wide?
  6.   Do all exits discharge directly to a street, yard, court,
      or other open space? [.37(h)(l)]

  7.   Are all exits marked by proper sign and illuminated?
      Are letters in sign not less than 6 in high, 3/4 in
      wide? [.37(q)(8)]

  8.   Is access to exits marked in all cases where the exit
      or the way to reach it is not immediately visible?
      [.37(q)(5)]

  9.   Is care taken to insure that no exit signs are obscured
      by decorations, furniture, or equipment? [.37(q)(3)]

  10.  Is exit access arranged so that it is not necessary to
      travel toward any high hazard area to escape?
      [.37(f)(5)]
Note:  Adapted from a safety inspection work sheet developed by the Center for Disease Control (CDC).
The pertinent section of the Code of Federal Regulations, title 29, part 1910 is given within brackets at the
end of each item.
                                                 A-2
          OCR(6/2001) by RYA-R10

-------
Listed below are explanations given by OSHA to questions that are not immediately clear on the safety checklist
for laboratory areas. Each explanation is numbered according to the corresponding question on the facing page.
    1.    Self-explanatory.

    2.    Self-explanatory.
    3.    At no  time should an exit door be locked or fastened in a manner that prevents it from being
         immediately opened from the inside of the building in the event of emergency.  Safety inspectors
         should check all doors marked "EXIT" to insure that they can be readily opened.

  4a,b.   The building manager should conduct periodic tests (as recommended by the manufacturer or as
         required by local code) to insure proper working order of the fire detection system and the automatic
         sprinkler system.

   4c.    Fire doors are designed to be closed in the event of fire.  Automatic fire doors normally remain open;
         however, the heat produced by a fire will cause them to close. Regular fire doors are designed to stay
         closed  at all times, except for the passage of personnel. No fire door should ever be blocked open as
         this will interfere with its function. Fire doors are typically used to enclose stairways and to separate
         buildings and corridors.

    5.    Self-explanatory.
    6.    An exit should never discharge into a location that could potentially trap personnel.  For example, an
         exit should not discharge into a closed courtyard.

    7.    Self-explanatory.
         A sign reading "EXIT" or similar designation, with an arrow indicating the direction, shall be placed
         in every location where the direction of travel to reach the nearest exit is not immediately apparent.
    9.    Self-explanatory.
   10.    In designing and maintaining exit routes from a building, care should be taken to avoid routing people
         through or near a high hazard area. An example of a high hazard area is a corridor or room where
         flammable liquids are stored.
                                               A-3                 OCR(6/2001)byRYA-R10

-------
Item Inspected
1 1 . Are aisles maintained clear and unobstructed for
movement of personnel and fire-fighting equipment?
12. Is all fire protection equipment and apparatus identi-
fied with the color red? [. 144(a)(l )]
13. Are portable fire extinguishers maintained fully
charged and operable and kept in designated places
at all times? [.157(a)(l)]
14. Are fire extinguishers conspicuously located, readily
accessible, and available along normal paths of
travel? [.157(a)(2)]
15. Are extinguishers and locations conspicuously
marked to indicate intended usage? [.157(a)(4)]
16. Are extinguishers mounted so that the top is not
more than 5 ft above floor; not more than 3'/2 ft if
weight equals more than 40 Ib? [.157(a)(6)]
17. Are all extinguishers mounted in cabinets placed so
that the instructions face outward? [.157(a)(7)]
18. Are extinguishers available suited to the class of fire
anticipated in each area? [. 1 57(b)(l )]
19. Are extinguishers placed according to distances for
proper coverage?
Within 75 ft - class A.
Within 50 in - class B. [.157(c)]
20. Are extinguishers inspected, maintained, and re-
placed by spares when they are discharged or miss-
ing? [.157(d)]
2 1 . Are laboratory rooms with potential fire hazards
equipped with proper extinguishers for emergency
situations? [.157(b)]
22. If flammable liquids are used in a laboratory, is the
mechanical ventilation sufficient to remove vapors
before they reach a hazardous concentration?
[.106(e)(2)(iii)]
Yes














No














N.A.














Comments














A-4
OCR(6/2001) by RYA-R10

-------
 11.    Self-explanatory.
 12.    Self-explanatory.
 13.    Self-explanatory.  During a safety inspection all fire extinguishers should be checked aginst the
       requirements of this question.
 14.   Fire extinguishers should never be located in places where they are concealed from general view.
15.    Fire extinguishers must be clearly labeled to indicate the type of fire they are capable of fighting. The
      following code is used to classify types of fires:

           Class A - fires in ordinary combustible materials such as wood, cloth, paper and rubber
           Class B - fires in flammable liquids, gases, and greases
           Class C - fires that involve energized electrical equipment where the electrical nonconductivity
                of the extinguishing medium is of importance; when electrical equipment is deenergized,
                extinguishers for class A or B fires may be safely used
           Class D - fires in combustible metals such as magnesium, titanium, zirconium, sodium, and
                potassium

 16.   Inspector should make certain that no fire extinguishers (other than wheeled-type extinguishers) are
      placed on the floor.
 17.    Self-explanatory.
18.     See explanation to question No. 15 for classification of fire hazards. Fire extinguishers (of more than
       one type, if necessary) should be selected and located according to anticipated fire hazards.

19.    Self-explanatory.
20.    Self-explanatory.
21.    In certain instances, laboratories can be fire traps. When this is the case, fire extinguishers should be
      provided inside the laboratory so that occupants can fight their way out of a fire emergency.  Safety
      inspectors should check for escape routes to the nearest corridor when considering this question.

22.   Self-explanatory.
                                             A-5                 OCR(6/2001)byRYA-R10

-------
Item Inspected
23. Are hose outlets within easy reach of persons stand-
ing on floor and not over 6 ft from floor?
[1 ^Q/'KVI ViM
• \ A A /J
24. Do all connections on dry standpipes have a sign
reading "DRY STANDPIPE-FOR FIRE DEPART-
MENT USE ONLY"? [.158(b)(7)]
25. Are automatic sprinkler systems provided with at
least one fire department connection with at least 4-
inpipesize?[.159(b)(l)]
26. Are fire department connections designated "AUTO
SPKR" or "OPEN SPKR"? [.159(b)(4)(v)]
27. Is there a water flow detection device on the sprin-
kler system that will activate the fire alarm?
[.159(c)(l)]
28. Are fire alarm boxes readily accessible and within
normal path distance of 200 ft? [.163(b)(3)]
29. Are all fire alarm systems inspected and tested at
weekly intervals? [.163(c)]
30. Are "NO SMOKING" signs posted in prohibited
areas? [.106(d)(7)(iii)]
Yes








No








N.A.








Comments








                                   Flammable Liquid Storage

31.  Are drums which contain flammable liquids con-
    structed of noncombustible materials? [.106(b)(l)(i)]

32.  Are storage drums vented? [.106(b)(2)(iv)]

33.  Are flammable liquids stored in proper containers?
34.  Are safety cans and portable containers of flammable
    liquids painted red with yellow band or name of con-
    tents? [.144(a)(l)(ii)]

35.  Are storage cabinets being used for storing
    flammable liquids? [.106(d)(3)]

36.  Are storage cabinets labeled "FLAMMABLE-KEEP
    FIRE AWAY"? [.106(d)(3)(ii)]
                                             A-6
OCR(6/2001) by RYA-R10

-------
23.    This question applies to buildings equipped with water and hose standpipe systems. Hose outlets are
      normally located in stairways or in the corridor immediately outside the stairway.
24.   Self-explanatory.  The building manager should be able to answer this question.
25.   Fire department connections to automatic sprinkler systems are normally located on the outside of
      the building being protected. The size of the connection is usually embossed on a metal plate or on
      the connection itself.

26.   The designations "ALTTO-SPKR" or "OPEN-SPKR" are found embossed on a plate at the fire
      department connection. Failure to display one of the designations is a violation of OSHA.

27.   This question is self-explanatory. The building manager should know if the facility is equipped
      with a sprinkler waterflow detection device.
28.   Self-explanatory.


29.   This procedure should be part of a regular maintenance program.
30.   The laboratory supervisor shall designate areas where smoking is prohibited and "NO SMOKING"
      signs shall be placed in these areas.
31.   Flammable liquid storage drums should be constructed of steel or some other noncombustible
      material.
32.   Self-explanatory.

33.   Flammable liquids may be safely stored in glass containers if the total capacity of the container is 1
      gal or less. Larger quantities of flammable liquids should be stored in safety cans approved by a
      recognized testing laboratory.

34.   Self-explanatory.
35.   Self-explanatory.


36.   Self-explanatory.
                                            A-7                 OCR(6/2001)byRYA-R10

-------
Item Inspected
37. Is the storage area provided with either a gravity or
mechanical exhaust ventilation system? [.106(d)(4)
(iv)]
38. Are extinguishers available where flammable or
combustible liquids are stored? [.106(d)(7)]
39. Is at least one portable fire extinguisher of rating not
less than 12-B units located outside of, but not more
than 10 ft from, door opening into room used for
storage? [.106(d)(7)(i)(a)]
40. Is at least one portable fire extinguisher with rating
not less than 12-B units located not less than 10 ft
nor more than 25 ft from flammable liquid storage
located outside of a storage room but inside of a
building? [.106(d)(7)(i)(b)]
41 . Are "NO SMOKING" signs posted in the flammable
or combustible liquid storage areas? [.106(d)(7)(iii)]
Yes





No





N.A.





Comments





                                       Electrical Hazards
42. Are all new electrical installations and all replace-
    ments, modifications, or repairs made and being
    maintained in accordance with the National Electri-
    cal Code? [.309]

43. Does the interior wiring system have a grounded
    conductor? i.e., three-wire system? [.309-NEC(200-
    2)]

44. Do all electrical appliances have Underwriter's Labo-
    ratories Inc. approval, or that of some other na-
    tionally recognized testing laboratory? [.309-
    NEC(90-8)]

45. Are the cords of all electrical equipment in good
    condition, not frayed or spliced, etc.? [.309-
    NEC(400-5)]

46. Are cords used properly (not run under rugs)? [.309-
    NEC(4004)]

47. Is there only one plug-in per socket outlet; i.e., no
    multiple plug-ins to one socket? [.309-NEC(240-2)]

48. Are the lighting levels such that good illumination is
    provided in all walking, working, and service areas
    to insure safety? [.309-NEC(110-16(e))]
                                              A-8
OCR(6/2001) by RYA-R10

-------
37.   The building manager or engineering department should know the answer to this question.
38.   Self-explanatory. Fire extinguishers approved for fighting class B fires should be available in
      flammable liquids storage areas.

39.   This question is self-explanatory but should be reread several times for full comprehension.  Fire
      extinguisher ratings are often stamped on the title plate along with other specifications.  Note: This
      question is to be used only when inspecting flammable-liquid storage rooms.

40.   This question is self-explanatory, but should be reread several times for full comprehension.  Note:
      This question is to be used only when inspecting  flammable liquid storage outside of a special
      designated room, but inside of a building.
41.   Self-explanatory.
42.   Prior to adding any new electrical facilities to a laboratory, it is the responsibility of the laboratory
      supervisor or building manager to check with the electrical contractor to insure that the National
      Electrical Code is observed. This rule is retroactive to Mar. 15, 1972.
43.   All convenience outlets shall be of the three-pronged type.  These outlets should be checked with a
      grounding tester to insure proper wiring.

44.   Approved appliances will bear the label of a recognized testing laboratory.
45.   Self-explanatory.
46.   Self-explanatory.
47.   Multiple outlet (cube) adaptors or extension cords with multiple outlet ends should not be used to
      plug more than one appliance into a single socket.

48.   Self-explanatory.
                                             A-9                  OCR(6/2001)byRYA-R10

-------
Item Inspected
49. Are circuit breaker panels and cutoff switches
located so as to be readily accessible? [.309-
NEC(240-25(b)) and (1926.400(e))]
50. Are all circuit breaker switches marked or labeled?
[.309-NEC(240-25(e))]
Yes

No

N.A.

Comments

                                Aisles, Exits, Floors, and Stairways

51.  Are all floors kept clean and dry? [.22(a)(2)]

52.  Are all floors, areas, and passageways free from pro-
    truding nails, splinters, holes, and loose boards?
    [.22(a)(3)]

53.  Are aisles and passageways clear of all obstructions
    and in good repair? [.22(b)(l)]

54.  Are aisles and passageways wide enough to operate
    equipment safely? [.22(b)(l)]

55.  Is floor loading within posted maximum limits?
    [.22(d)]

56.  Does every floor opening having standard guarded
    railings? [.23(a)(l)]
57. Do fixed stairs make an angle to the horizontal
    between 30° to 50°? [.24(e)]
58. Are all opensided floors, platforms, ramps, 4 ft
    above adjacent floor or ground, guarded with
    railing? [.23(c)(l)]
                                             A-10
OCR(6/2001) by RYA-R10

-------
   49.    Circuit breaker cabinets should never be locked.



   50.    Each circuit breaker switch shall be labeled as to the lights, outlets, or appliances it controls.




   51.    Self-explanatory.

   52.    Self-explanatory.



   53.    Self-explanatory.


   54.    Self-explanatory.


   55.    Safe floor loadings must be posted and observed in all buildings and on all floors.
   56.    Specifications for standard railing: A standard railing shall consist of a top rail, intermediate rail,
         and posts. It shall have a vertical height of 42 in nominal from upper surface top rail to floor,
         platform, runway, or ramp  level.  The top rail shall be smooth surfaced throughout its length. The
         intermediate rail shall be approximately halfway between the top rail and the floor, platform,
         runway, or ramp.

         Specifications for standard railing to be used on stairways: A stair railing shall be of construction
         similar to a standard railing (for a floor opening), but the vertical height shall not be more than 34
         in nor less than 30 in from upper  surface of top rail to surface of tread in line with face of visor at
         forward edge of tread.

   57.    All fixed stairways must rise at an angle within the limits shown:
                                                   ACCEPTABLE SAFE ZONE
                                     30°        HORIZONTAL
58.      Self-explanatory.
                                               A-l 1                 OCR(6/2001) by RYA-R10

-------
Item Inspected
59. Are flights of stairs having four or more risers
equipped with railings or handrails? [.23(d)(l)J
60. Are stair treads reasonably uniform and slip resist-
ant? [.24(fj]
61 . Are all places kept clean and orderly and in a
sanitary condition? [.141(a)(l)(i)]
62 . Are trash receptacles of approved use? [ . 1 4 1 (a)(3 )]
63. Is every enclosed work place so constructed and
maintained to prevent entrance and harborage of
rodents, insects, and vermin? [.141(a)(4)]
64. Is drinking water potable and available within 200 ft
of location at which employees work? [.141(b)(l)(i)]
65. Are toilet facilities adequate for both sexes and in
accordance with regulations listed? [.141(c)(l)]
66. Are toilet rooms constructed so that each water
closet occupies a separate compartment equipped
with door, latch, and clothes hanger? [.141(c)(2)(i)]
67. Does door to toilet room have a self-closing device,
and is entrance screened so interior is not visible
from outside? [.141(e)(2)(iii)]
68. Does every water closet have a hinged open-front
seat made of nonabsorbent material? [. 141(c)(3)(ii)]
69. Are suitable washing facilities available and main-
tained in a sanitary condition? [.141(d)(l)]
70. If employees are allowed to lunch on the premises, is
an adequate space provided for that purpose?
71 . Are adequate waste disposal containers provided?
72. Are change rooms provided for each sex where it is
necessary to change clothes? [.141(e)(10]
73. Are change rooms provided with separate storage
facilities for street clothes and protective clothing?
[.141(e)]
74. Are noise levels acceptable? [.95]
Yes
















No
















N.A.
















Comment
















A-12
OCR(6/2001) by RYA-R10

-------
59.   Self-explanatory.


60.   Stair treads should be free of cracks and other uneven areas.


61.   Self-explanatory.

62.   Trash receptacles used for garbage disposal should be equipped with tight-fitting lids.

63.   Self-explanatory.



64.   Self-explanatory.



65.   Self-explanatory.



66.   Self-explanatory.



67.   Self-explanatory.



68.   Self-explanatory.


69.   Suitable washing facilities consist of a lavatory with hot and cold water, a suitable cleansing agent,
      and individual hand towels.

70.   Self-explanatory. Employees are not permitted to eat in any laboratory that uses agents that are
      dangerous to health if ingested. It is the responsibility of the laboratory super-visor to enforce this
      rule.

71.   Self-explanatory.


72.   Self-explanatory.


73.   Self-explanatory.
74.   If any employees are exposed to noise levels greater than 90 dB at any time during the day, a safety
      expert should be consulted to determine if corrective measures need to be taken.
                                            A-13                 OCR(6/2001)byRYA-R10

-------
Item Inspected
75. Is necessary protective equipment provided, used,
and maintained in a sanitary, safe, and reliable
condition? [.132(a)]
76. Are eye protectors provided where machines or
operations present the hazard of flying objects, glare,
liquids, radiation? [.133(a)J
77. Are sufficient washing facilities (including eye
washes and deluge showers) available for all persons
required to handle liquids that may burn, irritate,
etc.? [.151(c)]
78. Are employees in the area exposed to air contami-
nants only in accordance with proper limits?
[.134(a)(l)J
79. Is a respiratory protection program used where
needed? [.134(a)(2)]
80. Are written standard operating procedures governing
the selection and use of respirators established?
81 . Has the user of the respirator been instructed and
trained in the proper use and limitations of the respi-
rator? [.134(b)(3)]
82. Are respirators regularly cleaned, disinfected, in-
spected, and stored in a convenient, clean, and sani-
tary location? [.134(b)(5)]
83 . Has the person assigned the task requiring a respira-
tor been determined physically able to perform the
work and use the equipment by the local physician?
84. Are breathing gas containers marked in accordance
with the American National Standards identifying
contents? [.134(d)(4)]
85 . Can gas mask canisters be identified by properly
worded labels and color code or atmospheric con-
taminant? [.134(g)(l)]
86. Is the compressor for supplying air to respirators
equipped with necessary safety and standby equip-
ment? [.134(d)(2)(ii)]
Yes












No












N.A.












Comments












A-14
OCR(6/2001) by RYA-R10

-------
 75.   Protective equipment, including personal protective equipment, shall be provided, used, and
       maintained in a sanitary and reliable condition wherever work-associated hazards may cause injury
       or impairment in function of any part of the body through absorption, inhalation, or physical
       contact. Laboratory supervisors should recognize such hazards and take the action necessary to
       insure that employees use adequate personal protection to avoid injury.

 76.   Safety glasses or goggles may be obtained through the CDC Safety Office.
 77.    All laboratories using liquids that may bum or irritate must be equipped with some type of emergency
       eye wash equipment.
78.     If laboratory supervisors have any questions concerning safety concentrations of air contaminants, they
       should contact the CDC Office of Biosafety.
 79.   Respirators that are applicable and suitable for the purpose intended shall be provided when such
       equipment is necessary to protect the health of employees. The CDC Office of Biosafety will provide
       the proper respirators for a given hazard.

 80.   Self-explanatory.
 81.   Self-explanatory.
 82.   Self-explanatory. Respirators should be cleaned and disinfected after each use.
 83.   Self-explanatory.
 84.   Self-explanatory.
 85.   Self-explanatory.
 86.   Compressors for breathing air should be equipped with receivers of sufficient capacity to enable the
       respirator wearer to escape from a contaminated atmosphere in the event of compressor failure; alarms
       to indicate compressor failure and overheating shall be installed in the system.
                                             A-15                 OCR(6/2001)byRYA-R10

-------
Item Inspected
87. Is each employee who works in an area where radio-
active material is used furnished and wearing film
badge? [.96(d)(2)]
88. Is radiation exposure of individuals to the body
limited to iwrems per calendar quarter? [.96(b)]
89. Is each radiation area posted with the proper radia-
tion caution sign? [.96(e)(2)]
90. Are all radiation area employees instructed in the
safety problems, precautions, and devices to mini-
mize exposure? [.96(i)]
91 . Are records maintained of the radiation exposure of
all employees who are monitored? [.96(n)]
92. Are radioactive materials stored and disposed prop-
erly? Are storage containers labeled? [.96(j,k)]
Yes






No






N.A.






Comment






                                      Compressed Gases
93.  Is each portable gas container for gases such as hy-
    drogen legibly marked with the name of contents?
94.  Are compressed gas cylinders determined in safe
    condition by visual and other inspection required in
    regulations? [.101(a)]

95.  Does the compressed gas cylinder or tank have an
    installed pressure relief device? [.101(c)]

96.  Are all compressed gas cylinders stored and secured
    so they cannot fall? [.252(a)(2)(ii)(b)]
97.  Are protection caps in place on compressed gas
    cylin-ders except when in use? [.252(a)(2)(ii)(d)]
98.  Is each mobile hydrogen supply unit secured to pre-
    vent movement? [.103(b)(l)(iv)(e)]

99.  Is the hydrogen storage area permanently marked
    "HYDROGEN-FLAMMABLE GAS-NO
    SMOKING OR OPEN FLAMES"? [.103(b)(l)(v)]
                                            A-16
OCR(6/2001) by RYA-R10

-------
   87.   Self-explanatory.



   88.   Self-explanatory.


   89.   Self-explanatory.


   90.   Self-explanatory.



   91.   Self-explanatory.


   92.   Self-explanatory.




   93.   Self-explanatory.



   94.   Deep dents or heavy corrosion of a gas cylinder might indicate a dangerous situation.



   95.   Self-explanatory.


   96.   When either storing or moving a compressed gas cylinder, care must be taken to prevent the cylinder
         from falling. OSHA requires that compressed gas cylinders either be chained or strapped in an upright
         position at all times.

   97.   Self-explanatory.

                                 Preface  to Questions 98 through 107

A gaseous hydrogen system is one in which the hydrogen is delivered, stored, and discharged in the gaseous form
to the consumer's piping. The system includes stationary or movable containers, pressure regulators, safety relief
devices, manifolds, interconnecting piping, and controls.  The system terminates at the point where hydrogen at
service pressure first enters the consumer's distribution piping. Systems having a total hydrogen content of less
than 400 ft3 are not covered by the following questions.

   98.   Mobile supply units should be strapped or chained to prohibit movement.


   99.   Self-explanatory.
                                               A-17                 OCR(6/2001)byRYA-R10

-------
Item Inspected
100. Is the hydrogen system in an adequately ventilated
area? [.103(b)(2)(n)(d)(l)]
101. Is the hydrogen system 20 ft from stored flammable
materials or oxidizing gases? [.103(b)(2)(ii)(d)(2)]
102. Is it 25 ft from open flames, electrical equipment,
or other sources of ignition? [.103(b)(2)(ii)(d)(3)]
103. Is it 25 ft from concentrations of people?
104. Is it 50 ft from intakes of ventilation or air-
conditioning equipment and air compressors?
105. Is it 50 ft from other flammable gas storage?
106. Is it protected from damage due to falling objects or
working activity in the area? [.103(b)(2)(ii)(d)(7)]
107. Are safety relief devices arranged to discharge up-
ward and unobstructed to the open air and prevent
impingement of gas upon container, adjacent struc-
ture, or personnel? [.109(b)(l)(ii)(b)]
Yes








No








N.A.








Comments








                                            Storage

108. Is storage of material stable and secure against slid-
     ing, collapse, falls, or spills? [.176(b)]

109. Are storage areas kept free from accumulation of
     materials that constitute hazards from tripping, fire,
     explosion, or pest harborage? [.176(e)]

110. Are dangerous parts of machines or energized
     equipment, which may injure, colored orange
     where exposed? [.144(a)(2)]

111. Is yellow the color used for designating physical
     hazards and caution in the particular environment?
112. Is the color green used to designate "SAFETY" and
     location of first aid equipment? [.144(a)(4)]

113. Is purple the color used to designate radiation
     hazards? [.144(a)(6)]

114. Are the colors black and white used for traffic and
     housekeeping markings? [.144(a)(7)]
                                             A-18
OCR(6/2001) by RYA-R10

-------
100.   Self-explanatory.






101.   Self-explanatory.






102.   Self-explanatory.






103.   Self-explanatory.






104.   Self-explanatory.









105.   Self-explanatory.






106.   Self-explanatory.






107.   Self-explanatory.
108.   Self-explanatory but very important,






109.   Self-explanatory.









110.   Self-explanatory.









111.   Self-explanatory.









112.   Self-explanatory.






113.   Self-explanatory.






114.   Self-explanatory.
                                           A-19                OCR(6/2001)byRYA-R10

-------
Item Inspected
115. Are all accident prevention signs used to minimize
workplace hazards? [.145(e)(e)]
116. Are all exposed steam and hot water pipes within 7
ft of floor or working platform covered with an
insulating material or guarded? [.264(e)(4)(iii)]
117. At all loading docks employing powered industrial
trucks, are wheel chocks used to prevent trailers or
railroad cars from robing? [.178(k)(l)]
118. Is medical help readily available through personnel
and first aid supplies approved by consulting physi-
cian? [.15 l(b)]
Yes

No

NA.

Comments

A-20
OCR(6/2001) by RYA-R10

-------
115.   Self-explanatory.


116.   Self-explanatory.



117.   Self-explanatory.
118.   In the absence of an infirmary, clinic, or hospital in near proximity to the workplace that is used for
      treating all injured employees, a person or persons shall be adequately trained to render first aid. First
      aid supplies approved by the consulting physician shall be readily available.
                                           A-21                OCR(6/2001)byRYA-R10

-------
j
                                     TECHNICAL REPORT DATA
                              (Please read Instructions on the reverse before completing)
   1. REPORT NO.
     EPA-600A-79-019
                              2.
  4. TITLE ANDSUBT1TLE
     Handbook for Analytical  Quality Control  in Water
     and Wastewater Laboratories
  7. AUTHOR(S)
     Editor:  Robert L.  Booth, EMSL-Cincinnati
              3. RECIPIENT'S ACCESSION»NO
                                                            5. REPORT DATE
                                                             March 1979 issuing^date
                                                            6. PERFORMING ORGANIZATION CODE
                                                           8. PERFORMING ORGANIZATION REPORT NO.
  9. PERFORMING ORGANIZATION NAME AND ADDRESS
                                                            10, PROGRAM ELEMENT NO
     SAME AS BELOW
                                                              1BD801
                                                            11. CONTRACT/GRANT NO,
                                                                 In house
  12. SPONSORING AGENCY NAME AND ADDRESS
     Environmental Monitoring and Support Lab.  -  Cinn, OH
     Office of Research  and  Development
     U.S.  Environmental  Protection Agency
     Cincinnati, Ohio  45268
                                                            13. TYPE OF REPORT AND PERIOD COVERED
                                                              Final                     	
                                                            14. SPONSORING AGENCY CODE
                                                              EPA/600/06
  15. SUPPLEMENTARY NOTES
     Project Officer:  Robert L.  Booth, EMSL-Cincinnati
  16. ABSTRACT

     This handbook is addressed to laboratory  directors, leaders of  field investigations,
     and other personnel who bear responsibility for water and wastewater data.  Subject
     matter of the handbook is concerned primarily with quality control  (QC) for chemical
     and biological tests  and measurements.  Chapters are also included  on QC aspects  of
     sampling, microbiology, biology, radiochemistry, and safety as  they relate to water
     and wastewater pollution control.  Sufficient information is  offered to allow the
     reader to inaugurate  or reinforce programs  of analytical QC that emphasize early
     recognition, prevention, and correction of  factors leading to breakdowns in the
     validity of water and wastewater pollution  control data.


  17.
                                KEY WORDS AND DOCUMENT ANALYSIS
  a.
                  DESCRIPTORS

    Quality Assurance
    Quality Control
    Water Analysis
    Laboratory Performance
    Valid Data
    Analytical Quality Control
b.lDENTIFlERS/OPEN ENDED TERMS
                                              Quality Assurance Techniq
                                              Analytical  Measurements
                                              Laboratory Services
                                              Instrument Selection
                                              Data  Handling
                                              Skills  and Training
c.  COS AT I Field/Group
                           jes

                                  07/D

                                                                                             1
13. DISTRIBUTION STATEMENT
  I  Release to Public
  EPA Form 2220-1 (9-73)
19. SECURITY CLASS (This Report}
Unclassified

                                              20. SECURITY CLASS (This page)

                                              Unclassified
21. NO. OF PAGES
                                                                               164
                                                                           22. PRICE
                                                        GOVERNMENT PRINTING OFFICE: 1979-J658-837/38 Region No. 5-11

-------