Guidance To Operation of Water Quality Laboratories
Guidance To Operation of Water Quality Laboratories
Guidance To Operation of Water Quality Laboratories
TECHNICAL REPORT:
GUIDANCE TO OPERATION
OF WATER QUALITY
LABORATORIES
September 2002
PREFACE
The “Guidance to Operation of Water Quality Laboratories” was drawn up as part of the
activity of the UN/ECE Task Force on Laboratory Quality Management & Accreditation
(UN/ECE TF-LQM&A) under the Convention on the Protection and Use of
Transboundary Watercourses and International Lakes (Helsinki, 1992) by taking into
account the requirements of the Guidelines on Water-Quality Monitoring and
Assessment of Transboundary Rivers (prepared by RIZA, the Netherlands as lead
country of the UN-ECE Task Force on Monitoring and Assessment).
Before drawing-up the guidance the UN/ECE TF-LQM&A completed a survey on the
present practices in the laboratories implementing water quality monitoring in the
countries signatory to the Convention. The survey included: (a) information on the
sampling and analytical methods used, (b) inventory on the laboratory facilities including
instrumentation, and (c) the quality control measures and status of accreditation. 27 out
of 46 countries responded to a Questionnaire, which had been distributed in English and
Russian languages. In addition to the information from this survey, requirements of
laboratory accreditation, guidelines for good laboratory practices, laboratory practices in
developed countries and related documents (e.g., Handbook for Analytical Quality
Control in Water and Wastewater Laboratories, US-EPA) have been taken into account
during drawing-up the guidelines.
The guidance was revised by the members of the Task Force Core Group in December
1999, during the first half of 2000 and in the year 2001. It was discussed during the first
and second meeting of Monitoring and Assessment Working Group too.
PART I.
1 INTRODUCTION........................................................................................................1
1.1 Objectives and Character of the Guidance........................................................... 1
1.2 Monitoring Cycle and Activities Dealt with in this Guidance................................. 1
1.3 Laboratory Quality Management .......................................................................... 2
1.3.1 Methodologies................................................................................................ 4
1.3.2 Laboratory Facilities and Personnel............................................................... 8
2 QUALITY REQUIREMENTS OF WATER QUALITY MONITORING .........................9
2.1 Quality Assurance, Quality Control in Water Laboratories................................... 9
2.2 Requirements for Accreditation .......................................................................... 10
3 COLLECTION OF SAMPLES FROM THE AQUATIC ENVIRONMENT...................13
3.1 Sample Collection from Sediments..................................................................... 16
3.1.1 Sampling Equipment .................................................................................... 16
3.1.2 Storage, Transport and Preservation of Sediment Samples ........................ 19
3.1.3 Selection of Particle Size ............................................................................. 21
3.1.4 Drying of Sediment....................................................................................... 21
4 PREREQUISITES OF QUALITY WORK IN LABORATORIES ................................22
4.1 Laboratory Services............................................................................................ 22
4.1.1 Distilled-Deionised Water ............................................................................ 22
4.1.2 Compressed Air............................................................................................ 25
4.1.3 Electrical Supplies........................................................................................ 25
4.2 Instrumentation................................................................................................... 26
4.3 Laboratory Glassware ........................................................................................ 27
4.3.1 Types of Glassware...................................................................................... 28
4.3.2 Glassware for Volumetric Analyses.............................................................. 29
4.3.3 Cleaning of Glass and Porcelain.................................................................. 30
4.3.4 Disposable Glassware ................................................................................. 31
4.3.5 Specialized Glassware................................................................................. 32
4.4 Reagents, Solvents and Gases .......................................................................... 32
4.4.1 Reagent Quality ........................................................................................... 32
4.4.2 General lnorganic Analyses ......................................................................... 33
4.4.3 Radiological Analyses .................................................................................. 34
4.4.4 Analyses of Organic Compounds................................................................. 34
4.4.5 Storing and Maintaining Quality of Reagents and Solvents ......................... 35
4.5 Elimination of Determinate Errors....................................................................... 36
4.5.1 Reagent Blank.............................................................................................. 36
4.5.2 Method Blank ............................................................................................... 36
4.5.3 Elimination of Interferences.......................................................................... 36
4.6 Laboratory supplies ............................................................................................ 38
5 SPECIAL REQUIREMENTS FOR TRACE ORGANIC ANALYSIS...........................39
5.1 Collection of the Samples................................................................................... 39
5.1.1 Discrete BottIed Samples............................................................................. 39
5.1.2 Carbon Adsorption Samples ........................................................................ 40
5.2 Glassware........................................................................................................... 40
5.3 Reagents and Chemicals ................................................................................... 40
5.4 Common Analytical Operations .......................................................................... 41
6 CONTROL OF ANALYTICAL PERFORMANCE ......................................................43
6.1 Precision and Accuracy ...................................................................................... 43
6.2 Evaluation of Daily Performance ........................................................................ 46
6.2.1 Quality Control Charts.................................................................................. 47
7 DATA HANDLING AND REPORTING .....................................................................49
7.1 The Analytical Result Value ............................................................................... 49
7.1.1 Significant Figures ....................................................................................... 49
7.1.2 Rounding-Off Numbers ................................................................................ 50
7.1.3 Glossary of Terms ........................................................................................ 51
7.2 Report Forms...................................................................................................... 52
7.2.1 Loose Sheets ............................................................................................... 52
7.2.2 Bound Books ................................................................................................ 52
7.2.3 Pre-Printed Report Forms ............................................................................ 53
7.2.4 Digital Read-out ........................................................................................... 54
8 SKILLS.....................................................................................................................55
8.1 Skills ................................................................................................................... 56
8.2 Training .............................................................................................................. 56
References, literature......................................................................................................57
Annex 1. ..........................................................................................................................58
Annex 2. ..........................................................................................................................61
Annex 3. ..........................................................................................................................83
Page 1
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
PART I.
1 INTRODUCTION
This guidance is intended to assist ECE Governments and joint bodies (e.g. bilateral or
multilateral river commissions) in understanding laboratory quality management and
providing guidance to laboratory managers to design or upgrade water laboratories and
to strive for accreditation.
The character of this guidance is technical rather than strategic as the latter aspects are
being dealth with in the relevant chapters of the Guidelines on monitoring and
assessment of transboundary rivers and transboundary groundwaters (see the
documents for the forthcoming second meeting of the Parties to the Convention issued
under symbol MP.WAT/2000/9 and 10).
Water management
Data analysis
Monitoring programmes
Data handling
Data collection
Successive activities in this monitoring cycle should be specified and designed based
on the required information product as well as the preceding part of the chain. In
drawing up programmes for the monitoring and assessment of river basins, riparian
countries should jointly consider all stages of the monitoring process.
The guidelines on monitoring and assessment are supported with five background
reports, which are as follows:
Although the guidelines and the background reports deals with some aspects of the
laboratory work, particular the major quality assurance approaches, there is a need for
guidance in establishment and upgrading laboratory work in water/environmental
laboratories.
The evaluation of the obtained information may lead to new or redefined information
needs, thus starting a new sequence of activities. In this way, the monitoring process
will be improved.
The present guidance on laboratory quality management exclusively deals with the
activity designated in the above figure by “Data collection”. This activity in the
monitoring cycle is composed of a number of sub-activities such as collecting samples,
ensuring quality work in laboratories and reporting analytical results.
The guidance summarised in the present document will address to laboratory managers
designing or upgrading water laboratories and will help these laboratories in the
countries in transition to be prepared for accreditation.
Water quality targets, objectives and standards are set to evaluate the quality of the
water resources, both surface and subsurface water bodies, to characterise ecological
status (for surface waters) and to establish satisfactory condition for intended uses of
the aquifer(s). The laboratory data define whether that condition is being met, and
whether the water is at acceptable quality to fit for the purpose. If the laboratory results
indicate a violation of the standard, action is required by the pollution control authorities.
The analyst must be aware that his professional competence, the procedures he has
used, and the reported values are reliable and may be used with confidence.
elements of the measurement cycle. It starts with the sampling and closes with reporting
the measurement results.
Quality Assurance
Quality Control
Figure 2. The measurement cycle ensuring the quality of the monitoring results.
Water quality monitoring rests upon a firm base of laboratory data. The role of the
analytical laboratory is to provide the qualitative and quantitative data to be used for
different purposes. The data must accurately describe the characteristics or the sample,
or the type and concentration of constituents in the sample analysed in the laboratory. In
many cases, an approximation or invalidated (incorrect) result is worse than no answer
at all, because it will lead to miss interpretations.
The progress of research, effectiveness of pollution control will depend upon the validity
of the results reported by the laboratory. Therefore, the laboratory data must be backed
up by an adequate programme to document the proper control and application of all of
the factors, which effect the final result.
The design of the monitoring programme should describe the targeted water quality
variables (characteristics, parameters, determinands), the matrices in the aquatic
environment from where samples should be taken, analytical accuracy targets, including
detection limits, application ranges and acceptable tolerances. The specialised
laboratories should apply the appropriate sampling and analytical methods, should have
the necessary rooms, should be properly equipped, having the necessary
instrumentation. Skills of laboratory personnel must ensure the quality of the work during
the sampling and the analyses.
Compilation of the guidance on laboratory quality management has been based on the
expectation that sampling and laboratory work consider all kinds of measurements,
which might be required during implementation of the water quality monitoring, detailed
in the monitoring design. The guidance target considerations, which should taken during
design and operation of the laboratories to generate reliable, comparable monitoring
results.
The laboratory management should plan the work in the laboratory before starting with
the sampling. Sampling and analytical methodologies must be defined as the first step of
the implementation of the monitoring.
1.3.1 Methodologies
♦ The method should measure the desired constituent with precision and accuracy
sufficient to meet the data needs even in the presence of interferences, which
might be encountered in the samples. Box 1 and Table 1 show the approach
used in the Danube river basin transnational monitoring network.
Page 5
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
♦ The procedure should utilise the equipment and skills available in the laboratory.
♦ The selected methods should be in use that the methods are properly validated.
♦ The method should be sufficient to produce the measurement results within the
required time-frame.
The resulting lists of determinands for water and sediments as agreed for the Danube
TNMN are presented in Table 1, together with the levels of interest and analytical
accuracy targets, which are defined as follows:
• The minimum likely level of interest is the lowest concentration considered likely to be
encountered or important in the TNMN.
• The principal level of interest is the concentration at which it is anticipated that most
monitoring will be carried out.
• The required limit of detection is the target limit of detection (LOD), which
laboratories are asked to achieve. This has been set, wherever practicable, at one
third of the minimum level of interest. This is intended to ensure that the best
possible precision is achieved at the principal level of interest and that relatively few
"less than results" will be reported for samples at or near the lowest level of interest.
Where the performance of current analyses is not likely to meet the criterion of a
LOD of one third, if the lowest level of interest, the LOD has been revised to reflect
best practice. In these cases, the targets have been entered in italics.
• The tolerance indicates the largest allowable analytical error which is consistent with
the correct interpretation of the data and with current analytical practice. The target is
expressed as “x concentration units or P%“. The larger of the two values applies for
any given concentration. For example, if the target is 5 mg/l or 20% - at a
concentration of 20 mg/l the maximum tolerable error is 5 mg/l (20% is 4 mg/l); at a
concentration of 100 mg/l, the tolerable error is 20 mg/l (i.e. 20%) because this value
exceeds the fixed target of 5 mg/l.
• Analytical accuracy targets for sediments are defined for <63 µm grain-size fraction.
Table 1/a. Determinand list and accuracy targets for water in the Danube TNMN
Regardless of the analytical method used in the laboratory, the specific methodology
should be carefully documented. In some water pollution reports it is customary to state
that “Standard Methods” have been used throughout. Close examination may indicate,
however, that this is not strictly true. In many laboratories, the “Standard Method” has
been modified because of personal preferences of the laboratory staff, etc. In other
cases the “Standard Method” has been replaced with a better one. Statements
concerning the methods used in arriving at laboratory data should be clearly and
honestly stated, if any deviation from the standard procedure occurs, proper validation
of the analytical method is an absolute need. The methods used should be adequately
referenced and the procedures applied exactly as directed. The recommended methods
to be used in water laboratories are listed in Annex 1.
1
fluoranthene, benzo(b)fluoranthene, benzo(k)fluoranthene, benzo(a)pyrene,
benzo(ghi)perylene and benzo(1,2,3-cd)pyrene
2
PCB-28, PCB-52, PCB-101, PCB-118, PCB-138, PCB-153 and PCB-180
Page 8
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
Knowing the specific method, which has been used, the reviewer can apply the
associated precision and accuracy of the method when interpreting the results. If the
analytical methodology is in doubt, the data user may inquire the reliability of the result.
The advantages of strict adherence to accepted methods should not reduce efforts,
investigations leading to improvements in analytical procedures. In spite of the value of
accepted and documented methods, occasions do arise when a procedure must be
modified to eliminate unusual interference, or to yield increased sensitivity. When
modification is necessary, the revision should be carefully worked out and the revised
method should be validated to accomplish the desired result. It is advisable to assemble
data using both the regular and the modified procedure to show the superiority of the
latter. Responsibility for the use of a non-standard procedure rests with the analyst and
his supervisor, since such use represents a deviation from accepted practice.
In field surveys, the problem of transport of samples to the laboratory, or the need to
analyse a large number of samples to arrive at gross values may require the use of
rapid field (screening) methods. Such methods should be used with caution, and with a
clear understanding that the results obtained do not compare in reliability with those
obtained using standard laboratory methods. The fact that such screening methods
have been used should be noted. The data user is entitled to know that approximate
values have been obtained for screening purposes only, and that the results do not
represent the precision and accuracy, which might be obtained in the laboratory.
The quality of the laboratory work should be ensured by appropriate facilities including
instrumentation, as well as by skilled personnel. Related issues will be discussed in
more details in the following chapters of the guidance.
Depending on the size of the laboratory, e.g., more than 15-20 staff member, a quality
manager should be appointed. The responsibility of this person includes: (a) supervision
of the internal quality control measures in the laboratories, (b) ensure regular provision
of spiked samples which are unknown for the analyst before the evaluation of the
results, and (c) co-ordinate participation of the laboratories in performance testing
programmes.
The background reports, particularly on the Quality Assurance, prepared by the UN/ECE
Task Force on Monitoring and Assessment provide general guidance on the quality
requirements and quality assurance needed for water quality monitoring. The present
guidance gives more practical details on these subjects particularly to help laboratories
in their ways for accreditation.
Because of the importance of monitoring data (analytical results) and the resulting
actions, a programme to assure the reliability of the data is essential. It is recognised
that most analysts practice analytical quality control to varying degrees, depending
partly upon their training, professional ambitions, and partly upon the importance of the
work they are doing. However, under the pressure of daily workload, analytical quality
control may be easily neglected. Therefore, an established, routine internal quality
control programme, applied to each analytical test, is important in assuring the reliability
of the measurement results. When the implementation of the monitoring programme
requires participation of several national, or international laboratories, there is a need
for external quality control measures such as organisation and participation in
performance testing schemes (intercalibration programmes) organised by dedicated
institutions.
The quality control programme in the laboratory has two primary functions. First, the
programme should monitor the reliability of the reported results. It should provide an
answer to "How good (true) are the results submitted". This phase may be termed as
"measurement of quality." The second function is the control of quality in order to meet
the programme requirements. For example, the processing of spiked samples is the
measurement of quality, while the use of analytical grade reagents is a control measure.
As each analytical method has an analytical protocol (e.g., in international/national
standards, or in Standard Operational Procedures compiled in the frame of most large
international river basin management programme), so the quality control associated with
that test must also involve definite steps to monitor and assure the correctness of the
results. The steps in quality control vary with the type of analysis. For example, in the
titrimetric measurements, standardisation of the titrant on a frequent basis is an element
of qualify control. In an instrumental method, the check-out of instrument response and
the calibration of the instrument in concentration units is also a quality control function.
The variables, which can affect the final results should be considered and evaluated.
The guidance considers and provides recommendations for the control of the factors,
which go into generating an analytical result, in order to ensure that the best possible
result is obtained. A programme based upon these recommendations will give the
analyst and the laboratory confidence in the reliability of the measurement results
characterising the collected sample.
Page 10
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
• have managerial and technical personnel with the authority and resources needed
to carry out their duties and to identify the occurrence of departures from the quality
system or from the procedures for performing tests and/or calibrations, and to
initiate actions to prevent or minimize such departures;
• have arrangements to ensure that its management and personnel are free from any
undue internal and external commercial, financial and other pressures and
influences that may adversely affect the quality of their work;
• have policies and procedures to ensure the protection of its clients' confidential
information and proprietary rights, including procedures for protecting the electronic
storage and transmission of results;
• have policies and procedures to avoid involvement in any activities that would
diminish confidence in its competence, impartiality, judgment or operational integrity;
• define the organization and management structure of the laboratory, its place in any
parent organization, and the relationships between quality management, technical
operations and support services;
• specify the responsibility, authority and interrelationships of all personnel who
manage, perform or verify work affecting the quality of the tests and/or calibrations;
• provide adequate supervision of testing and calibration staff, including trainees, by
persons familiar with methods and procedures, purpose of each test and/or
calibration, and with the assessment of the test or calibration results;
• have technical management which has overall responsibility for the technical
operations and the provision of the resources needed to ensure the required quality
of laboratory operations;
• appoint a member of staff as quality manager who, irrespective of other duties and
responsibilities, shall have defined responsibility and authority for ensuring that the
quality system is implemented and followed at all times, the quality manager shall
have direct access to the highest level of management at which decisions are made
on laboratory policy or resources;
• appoint deputies for key managerial personnel
The laboratory shall establish, implement and maintain a quality system appropriate to
the scope of its activities. The laboratory shall document its policies, systems,
programmes, procedures and instructions to the extent necessary to assure the quality
of the test and/or calibration results. The system's documentation shall be
Page 11
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
communicated to, understood by, available to, and implemented by the appropriate
personnel.
The laboratory’s quality system policies and objectives shall be defined in a quality
manual. The overall objectives shall be documented in a quality policy statement. The
quality policy statement shall be issued under the authority of the chief executive. It shall
include at least the following:
The quality manual shall include or make reference to the supporting procedures
including technical procedures. It shall outline the structure of the documentation used
in the quality system. The quality manual should include:
The roles and responsibilities of technical management and the quality manager,
including their responsibility for ensuring compliance with this International Standard,
shall be defined in the quality manual.
The laboratory shall established and maintain procedures to control all documents that
form part of its quality system (internally generated or from external sources), such as
regulations, standards, other normative documents, test and/or calibration methods, as
well as drawings, software, specifications, instructions and manuals.
The laboratory shall establish and maintain procedures for the review of requests,
tenders and contracts. The policies and procedures fro these reviews leading to a
contract for testing and/or calibration shall ensure that:
The laboratory shall afford clients or their representatives cooperation to clarify the
client’s request and to monitor the laboratory’s performance in relation to the work
performed, provided that the laboratory ensures confidentiality to other clients.
Page 12
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
The laboratory shall periodically, and in accordance with a predetermined schedule and
procedure, conduct internal audits of its activities to verify that its operations continue to
comply with the requirements of the quality system. The internal audit programme shall
address all elements of the quality system, including the testing and/or calibration
activities. It is the responsibility of the quality manager to plan and organize audits as
required by the schedule and requested by management. Such audits shall be carried
out by trained and qualified personnel who are, wherever resources permit, independent
of the activity to be audited.
PART II.
The design of the sampling plan is part of the monitoring cycle, whereas the
implementation of the sampling belongs to the measurement cycle. Collection of the
representative samples should be the responsibility of the laboratory staff. Because
collection of samples from different environmental matrices can be considered as the
first step in the measurement cycle and very much dependent on the analytical work it is
desirable to collect the samples by laboratory personnel.
Grab samples are single samples collected at a specific spot at a site over a short
period of time (typically seconds or minutes). Thus, they represent a "snapshot" in both
space and time of a sampling area. Discrete grab samples are taken at a selected
location, depth, and time. Depth-integrated grab samples are collected over a
predetermined part or the entire depth of a water column, at a selected location and time
in a given body of water.
Page 14
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
When samples are collected from a river or stream, observed results may vary with
depth, stream flow, and distance from each shore. Selection of the number and
distribution of sites at which samples should be collected depends on study objectives,
stream characteristics, available equipment, and other factors. If equipment is available,
take an integrated sample from top to bottom in the middle of the main channel of the
stream or from side to side at mid-depth. If only grab or catch samples can be collected,
preferably take them at various points of equal distance across the stream; if only one
sample can be collected, take it in the middle of the main channel of the stream and at
mid-depth.
Rivers, streams, lakes, and reservoirs are subject to considerable variations from normal
causes such as seasonal stratification, diurnal variations, rainfall, runoff, and wind.
Choose location, depth, and frequency of sampling depending on local conditions and
the purpose of the investigation.
Although well pumping protocols depend on the objectives of an investigation and other
factors such as well characteristics and available equipment, a general rule is to collect
samples from wells only after the well has been purged sufficiently to ensure that the
sample represents the groundwater. It will be necessary to pump at a specified rate to
achieve a characteristic drawdown, if this determines the zones from which the well is
supplied; record purging rate and drawdown, if necessary. By using methods with
minimal drawdown, purging volumes can be reduced significantly.
Sample carefully to ensure that analytical results represent the actual sample
composition. Important factors affecting results are the presence of suspended matter or
turbidity, the method chosen for removing a sample from its container, and the physical
and chemical changes brought about by storage or aeration. Detailed procedures are
essential when processing (blending, sieving, filtering) samples to be analyzed for trace
constituents, especially metals and organic compounds. Carefully consider the
technique for collecting a representative sample and define it in the sampling plan.
The type of sample container used is of utmost importance. Test sample containers and
document that they are free of analytes of interest, especially when sampling and
analyzing for very low analyte levels. Containers typically are made of plastic or glass,
but one material may be preferred over the other. For example, silica, sodium, and
Page 15
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
boron may be leached from soft glass but not plastic, and trace levels of some
pesticides and metals may sorb onto the walls of glass containers. Thus, hard glass
containers are preferred. For samples containing organic compounds, do not use plastic
containers except those made of fluorinated polymers such as polytetrafluoroethylene
(PTFE).
Some sample analytes may dissolve (be absorbed) into the walls of plastic containers;
similarly, contaminants from plastic containers may leach into samples. Avoid plastics
wherever possible because of potential contamination from phthalate esters. Container
failure due to breakdown of the plastic is possible. Therefore, use glass containers for
all organics analyses such as volatile organics, semivolatile organics, pesticides, PCBs,
and oil and grease. Some analytes (e.g., bromine-containing compounds and some
pesticides, polynuclear aromatic compounds, etc.) are light-sensitive; collect them in
amber-colored glass containers to minimize photodegradation.
N≥
(ts )
2
where:
N = number of samples,
t = Student-t statistic for a given confidence level,
s = overall standard deviation, and
U = acceptable level of uncertainty.
Methods of preservation are relatively limited and are intended generally to retard
biological action, retard hydrolysis of chemical compounds and complexes, and reduce
volatility of constituents. Preservation methods are limited to pH control, chemical
addition, the use of amber and opaque bottles, refrigeration, filtration, and freezing.
Page 16
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
Three major types of solid particles (particulate matter) can be distinguished in rivers and
lakes, even though this classification is not absolute. One type can be converted into
another by the changes in water velocity. Those types are:
Bedload - that part of the particulate matter which remains in almost constant contact
with the river bed and is moved by rolling, sliding or hopping much more slowly than the
water velocity;
Deposited matter - results from a decrease in water energy. This leads to settling, the
rate of which depends on the particle size.
Guidance for sediment sampling can be obtained from the ISO Guide 5667-12: Guidance on
sampling bottom sediment and ISO Guide 5667-17: Guidance on sampling of suspended
sediments.
The procedures and apparatus applied for sediment sampling depend on the type of
sediment required for analysis. The procedures and the equipment used for sampling
suspended sediments are different from those required for bottom sediment. In general, care
must be taken not to lose fine fractions since the association (adsorption) of analytes is
largest on small particles. During bottom sediment sampling the sampler must be usually
able to maintain the integrity of the layers, but it is very difficult particularly in the case of
sampling fast-flowing river bottom sediment.
Samplers used for suspended sediments must allow the collection of a sample
representative of the water-sediment mixture at the sampling point or sampling zone at the
time of sampling and in amount enough to carry out all of the required analysis. Samplers
are of four general types:
* Integrating samplers
* Grab samplers
* Pumping (centrifuging) samplers
* Sedimentation traps
The integrating samplers are designed in a way that the intake nozzle is oriented into, and
parallel to the flow. As a result, the water-sediment mixture moves from the stream into the
intake with minimum acceleration, thus providing the most accurately the representative
sample.
Page 17
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
In smaller rivers, direct hand sampling can be performed by wading, using a simple grab
sampler, such as a bucket if that is all that is available. If the suspended sediment load is
sufficiently high, grab or depth-integrating samplers may be used followed by pressure, or
vacuum filtration through a 0.45 µm filtration system.
Pumped centrifugation can provide relatively large amount of sediment sample, which might
be required for trace organic analysis, but it needs a special portable centrifuge designed for
such type of sample collection.
In lakes, the first two types are unlikely to provide enough material after filtration for even the
most basic analyses. Most of the published work on the suspended matter in lakes has used
sedimentation traps, although microbial processes continuing within them during the lengthy
residence times required to obtain even small samples, and the complex physics of trap
entry raise doubts about how representative the analyses may be. Sedimentation traps for
use in lakes and reservoirs are cylinders of plastic for metallic determinands, and glass or
metal for organic compounds. The cylinder should have a height-to-width ratio larger than
five, with nothing above the opening such as lattices, lids or collars. It is necessary to
prevent the mooring system from moving by using an appropriate sub-surface float. Traps
should be retrieved after no more than one month to reduce mineralization and microbial
degradation of suspended organic matter.
Because the topmost layer of the sediment is the most recent and is the site of the most
intensive microbiological activity, it is usually important that the sampler does not stir the
sediment surface and/or cause change in the sediment layers. Thus, the sampler must be
lowered so as not to create a pressure wave that might displace the softest surface
sediments. The sampler must penetrate far enough into the sediments to include the desired
layers, but not so far that the sediment spills over the top.
For different sediment types the recommended samplers are summarized in Table 2.
Grab samplers
Grab samplers are more commonly used than core samplers for collecting deposited
sediments as they are often much lighter and in some circumstances much easier to use.
There is a large number of variations of bed grabbers. They consist of one or more hinged
buckets which close whilst being raised. During scissors-like closing of the buckets,
sediment is enclosed by them, providing disturbed samples. Probe depths vary from 5 cm to
50 cm, depending upon the size and mass of the sampler and the structure of the bed
material. Due to the grab construction, there is a large chance of losing part of the finer
fraction and/or the top layer. Grabs are available in a variety of designs.
In general the scissor grab is most suited for sampling sediment beds consisting of silt
and/or sand. It is not suitable for sampling peat or clays. A sample taken with a scissor grab
will always be disturbed. Inaccuracies arise because of washing away of the fine fractions.
Depth of the penetration is unknown and dependent on the nature of the bed for any
particular instrument.
Page 18
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
Sand Both grab and corer systems can be used. A sand bed can be
very hard and thus prove difficult for light weight grabs and
manually operated corer systems
Consolidated silt Both grab and corer can be used. If a grab is used, it is not
possible to determine the sample penetration depth
Unconsolidated silt Grab systems are not suitable as they are prone to sinking
through the soft layer. Corer systems are better but when a
frame is used at great depth care is essential to prevent the
frame from sinking through the soft layer. More support can
usually be given to prevent this by adding large plates to the
feet of the frame.
Core samplers
Sampling using a core sampler is based on a principle of driving a hollow tube into the bed
so that the sediment is pushed into it. A sample is obtained by pulling the tube out of the
bed. The tube can be inserted into a bed by means of its weight, vibration device or
manually.
In manually operated corers (e.g., piston drill, Beeker sampler, sealed core sampler, Vrijwit
drill, etc.) the tube is pushed into the bed by means of rods. Penetration is generally up to
maximum 2 meters, depending on the nature of bed materials. Gravels are unlikely to be
suited to this sampling method. Because extension rods are used, there can be problems in
obtaining samples when working from the bank where a distance of more than 4 m must be
bridged by rods. Due to the movement of the vessel, it is often difficult to obtain good
samples from a boat. However, it is possible to obtain reliable samples in a river depth of
approximately 2 m; beyond this, a diver may need to be employed.
Mechanically operated sampling devices employ usually their weight for penetrating into the
bed. In the gravity corer ("Falling bomb" sampler) the core tube is mounted in a weighted
holder, which is dropped freely from a vessel and penetrates the sediment. The method is
fast and efficient because it is not necessary for the vessel to be anchored. This sampler is
not suitable for use in unconsolidated sediments.
Page 19
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
In the "Jenkins mud sampler", the corer is mounted in a frame and due to its large mass
sinks into bed. Once the suspension cable is slackened sufficiently, a closing mechanism is
activated which shuts off the sample tube by the means of hinged arms. This sampler is
suitable for investigation of very soft beds, it is not suitable for hard sediment beds.
The Craib corer consists of a core tube mounted in a frame. When it is lifted out of the
sediment layer, the core tube is first closed off at the top by a valve. As soon as the bottom
is free of the bed, it is closed off by a ball.
The Easy All corer's mass can be increased to approx. 110kg. After the sample has been
taken, the core tube is shut off at the top and the bottom by means of valves. The filled core
tube can be removed from the holder completely once it is aboard.
Glass containers should also be used when organic constituents are to be determined,
whereas polyethylene containers are preferable for sampling those determinands that are
major constituents of glass (e.g., sodium, potassium, boron and silicon) and for sampling of
trace metallic moieties (e.g., mercury). These containers should only be used if preliminary
tests indicate acceptable levels of contamination. If glass containers are used for storing
sediments with pore waters, which are weakly buffered, then borosilicate rather than soda
glass containers should be chosen.
Suspended sediment
As soon as possible after collection, the sample should be filtered. The filtrate can then be
used for measuring the dissolved constituents. The following sample containers and
preservation procedures are recommended:
• Samples for analysis of carbon, heavy metals, nitrogen and phosphorus can be
stored in polyethylene or polypropylene bag or jar frozen to -20oC for the period of
max. 6 months.
• Samples for analysis of chlorinated hydrocarbons, pesticides and oil pollution can be
stored in glass jar or aluminum canister extracted immediately, or frozen to -20oC and
extracted as soon as possible.
The analysis of a suspended sediment sample is often limited on account of the difficulty in
obtaining sufficient material for the several sub-samples required for the different analyses.
A composite of a large number of representative samples may be necessary.
Page 20
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
Bottom sediment
In practice it has become apparent that every project or investigation sets its own particular
demands in the field of sample treatment. The investigation plan prepared for the field
sampling should include a section on the treatment of the samples. This plan should take
into account the particular aim of the project and the requirements for the sample treatment
given by the receiving analyst.
When transferring samples from the collection equipment to the storage container, care
should be taken to ensure the continuance of anaerobic conditions if appropriate to the
planned analysis. The maintenance of anaerobic conditions will to a large extent depend on
the equipment being used. A practice run may be found useful to refine any techniques
developed. In addition, if trace organics are to be studied, use of some plastic implements
during sub-sampling may contribute to interference. Similarly, the use of a metal spatular
should be avoided if trace metals are of interest. The type and composition of sample
transfer tools should be noted in the field report.
Sediment samples should generally be kept in glass containers and stored and transported
cool. If it is necessary to keep them longer than one month, then this should be done in a
deep freezer, giving due regard to the physicochemical changes that can occur to colloids
on freezing. For example, changes in de-watering characteristics may be observed when
specific laboratory sample preparations are used. In all cases sample containers should be
delivered to the laboratory tightly sealed and protected from light and excessive heat,
because the sample may change rapidly due to gas exchange, chemical reactions and the
metabolism of organisms. The build up of gas pressures in the sample container, due to
anaerobic digestion, should not be overlooked, and it may therefore be necessary
periodically to release pressure from the container. This may become necessary if
temperature regulation cannot be provided in warm climates.
If freezing the sample is chosen as the preferred method of preservation as defined by the
sampling programme and specified analytical method, the notice should be taken of
following:
• It is essential for the sample to be completely thawed before use, as the freezing
process may have the effect of concentrating some components in the pore water of
the inner part of the sample, which substantially freezes last. The freezing of samples
can lead to a loss of analytes from pore-water solution by absorption/adsorption on
the precipitating compounds. When the sample thaws, dissolution may be
incomplete, and thus erroneous results for pore water parameters may be produced.
• Chemical preservation techniques should only be used after careful assessment of
the project needs, the requirements of the analytical method and with the specific
guidance of the receiving laboratory about the techniques required for
homogenisation of the sample with the preservative. For example, mineral acid may
be added in an attempt to arrest or inhibit anaerobic digestion of organic matter if a
study of organic acids is being made. Separate sub-samples, therefore, may be
required prior to freezing.
Page 21
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
All preservation steps should be recorded in a site report and the temperature measured
and recorded on site. If appropriate, other physical and chemical parameters should be
determined on site or as soon as possible after sample collection.
In the literature are still very differing proposals for the choice of the grain size fraction
for analyses. It is a fact that the organic micropollutants are mainly adsorbed on the
organic part of the sediment, which can be found with increasing amount in the small
grain-size fractions but it is also valid for the inorganic micropollutants. The larger
particles, mainly quartz (sand), are considered as diluting material.
The sediment sample has to be fractionated prior to sample processing and analysis.
Selection of the fraction for the analysis depends, similarly to the sampling type, on the
general aim of the sediment analysis, and it should reflect the distribution of the particular
analyte as a function of the sediment particle size. Thus, any sediment analysis program
should be preceded by an agreement on the fraction to be analyzed (e.g., 4 - 200 µm or <
63 µm). Because of the tedious sieving process for grain-size smaller than about 50 µm,
grain- size fractions around this value are chosen in many cases for practical reasons.
The clay-silt fraction, smaller than 63 µm, is widespread in different monitoring
programmes. Sieving of the sediment should be done in the wet state at the sampling
site through appropriate sieves. Sieves used for fractionation have to be made from
material that will not interfere with the analytes.
In any cases the dry weight of the sediment has to be determined with a part of the
sample because the concentration of analytes should be related to dry sediment mass,
expressed as mg/kg of dry sediment.
Page 22
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
This chapter provides guidance for laboratory managers and staff concerning the
laboratory services, instrumentation, etc., needed to support and ensure the quality of
the work in the laboratory.
The quality of the work laboratory analyses involves consideration and control of several
factors, which 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.
The ways of maintaining the quality of these services in the laboratory are provided in
the following.
Water purity has been defined in different ways, but one generally accepted definition
states that high purity water is water that has been distilled and/or deionised so that it
will have a conductivity of 2 µS/cm, or smaller. This definition is satisfactory as a base to
work from, but for more critical requirements, Table 3 shows the degrees of purity.
Page 23
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
For special purposes, all-glass distillation unit may be preferable to the metal still. The
all-glass distillation units are usually smaller, and of more limited capacity than the metal
stills. The all-glass still produced a product, which had substantially lower contamination
from zinc, copper, and lead. Other analyses have indicated the same general
relationship, except that a boron concentration of 100 µg/litre was found in water from
the all-glass still on one occasion. This was probably related to the length of time the
distillate had remained in the glass storage reservoir.
All distillation units require periodic cleaning to remove solids, which have been
deposited from the feed water. Hard water and high dissolved solids content promote
scale formation in the evaporator, and cleaning frequency will thus depend on the
quality of the feed water. The boiler of an all-glass unit should be drained daily and
refilled with clean water. Build-up of scale is easily detected, and the boiler and
condenser coils should be cleaned at frequent intervals. Metal stills should be
dismantled and cleaned at regular intervals. Cleaning should always be in accordance
with the manufacturer’s instructions.
Pre-treatment of the incoming feed water will often improve distillation 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, which might subsequently be carried over in the distillate. If trace
concentrations of ions, organic substances are of major concern, the distillate may be
further treated as described later.
A piping system for delivering distilled/deionized 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/deionizer 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 and on available funds. Tin is best,
but is also very expensive. As a compromise, plastic pipe, or (Teflon) 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.
Page 24
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
When there is no piped-in supply, distilled water will be transported to the laboratory
and stored in polyethylene or glass bottles. If stored in glass containers, distilled water
will gradually leach the more soluble materials from the glass and cause an increase in
dissolved solids. Therefore, only borosilicate-free glass containers should be used.
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, since 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, which extends almost to the
bottom of the bottle, and which is bent downward above the bottle neck, with flexible
tubing attached for mobility. Vinyl tubing is preferable to rubber, because it is less
leachable. The vent tube 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. Re-distillation from an alkaline permanganate solution can be used
to obtain a water with low organic background. Certain analyses require special
treatment or conditioning of the distilled water.
The quality of the distilled water should be checked according to the given
determinands, however, it is always include in the method blank determinations when it
is part of the analytical procedures (e.g., trace organic analysis).
The ammonia content of this water should be below the detection limit, as measured in
the blank solution.
Carbon-dioxide-free water may be prepared by boiling distilled water for 15 minutes and
cooling to room temperature. As an alternative, distilled water may be vigorously
aerated with a 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.
A multi-purpose 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 strongly acidic cation-exchange resin in the hydroxyl form. Resins of a quality
suitable for analytical work must be used. By using a fresh column and high quality
distilled water, the water corresponding to designation for referee reagent water
(maximum 0.1 mg/l total matter and maximum conductivity of 0.1 µS/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 may add organic contaminants to the water
by contact with the ion-exchange materials.
Due to the special type of work, requirements for a laboratory lighting system are quite
different from those in other areas. Accurate readings of glassware graduations and
other measuring lines must be made. Titration endpoints, sometimes involving subtle
changes in colour or shading, must be observed. Levels of illumination, brightness, and
Page 26
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
Electrical heating devices (hot plates, muffle furnaces, water baths and laboratory
ovens) provide desirable heat sources. Care must be taken to ground all equipment,
which could constitute a shock hazard.
Special attention should be paid in laboratories where electric supply shuts down
temporarily time-to-time, for different time periods. Several instruments can be
damaged, or data lost from sudden stop of electricity. Therefore, it is important to
establish power cut-off free units, which provide electricity continuously when central
electric power-supply stops.
4.2 Instrumentation
The modem analytical laboratory depends very heavily upon instrumentation. Analytical
instrumentation, to a certain extent, is always in the development stage, with
manufacturers continually redesigning and upgrading their products, striving for better
durability and sensitivity, and improved automation, particularly due to computerization.
The laboratory supervisors and stuff members should consider the competing markets, a
stream of advertising brochures, announcements, and catalogues of newly available
equipment. Consequently, the selection and purchase of analytical equipment is should
be based on careful considerations. Table 4. lists the instruments most commonly used
in aquatic environmental analysis.
Page 27
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
These represent basic equipment used in routine work and more sophisticated
instruments for special research and monitoring purposes. Further, operation and
maintenance of these instruments ought to be a primary consideration in production of
satisfactory data. Obviously, a fundamental understanding of instrument design will
assist the analyst in the correct use of the instrument and in some cases aid in detecting
instrumental failures.
Depending on the particular manufacturer, various trade names are used for specific
brands possessing special properties such as resistance to heat, shock, alkalies, etc.
Examples of some of these follow:
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 materials, such as Teflon, are quite expensive; however, Teflon stopcock
plugs have practically replaced glass plugs in burets, separatory funnels, 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, shatter-proof, autoclavable and chemically resistant.
Some points to consider in choosing glassware and/or plastic-ware are:
• Plastic bottles of polyethylene and/or Teflon have been found satisfactory for the
shipment of water samples. Strong mineral acids (such as sulphuric acid) and
organic solvents will readily attack polyethylene and are to be avoided.
• 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.
The precision of volumetric work depends in part upon the accuracy with which volumes
of solutions can be measured and there are certain sources of error, which 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 per 1°C rise in
temperature, but if made of borosilicate glass the increase is much less. 1000 ml of
water or of most 0.1 N solutions increases in volume by approximately 0.20 ml per 1°C
increase at room 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.
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. They are available
in various shapes and sizes ranging from 1 to 2000 ml capacity.
Volumetric pipets are calibrated to deliver a fixed volume. The usual capacities are 1 to
100 ml although micro-pipets are also available. 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. The band is usually located far enough down so that it will not
touch the technician's lips when liquid is being drawn up or blown out.
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, micro-burets are also used. Micro-
Page 30
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
• Do not attempt to dry a buret which 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 per second, 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 of and/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.
Several standards, such as DIN, NBS, etc., describe the specifications for volumetric
glassware. In general, volumetric glass apparatus, which meet standard specifications
are designated as Class A and all such glassware is permanently marked with a large
"A". 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 as Pyrex, etc., the stock number of the particular item, and the capacity of
the vessel. According to the DIN standard, the tolerance is also marked, such as 25 ml ±
0.030 ml.
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 12-15
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.
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 minutes 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. Store 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 redistilled solvent (e.g., acetone, hexane, petroleum ether,
chloroform).
Caps are washed with detergent, rinsed with tap water, distilled water and solvent.
Liners are treated in the same way as the bottles and are stored in a sealed container.
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
pipets may be the answer, provided they meet the necessary specification. Various
types are available including bacteriological, serological and micro-dilution pipets.
Disposable glassware generally is made of soft glass.
Page 32
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
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,
etc. Standard-taper, interchangeable ground joints save time and trouble in assembling
apparatus. They are tested precision-ground to insure an accurate fit and freedom from
leakage. Ball and socket joints increase flexibility of operation and eliminate the need for
exact alignments of apparatus.
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. It is required that the date of preparation and expiery
be shown on the storage bottle. Each bottle should be labeled showing its content, date
of preparation and the name of the analyst.
Chemical reagents, solvents, and gases are available in a wide variety of grades of
purity, ranging from technical grade to various "ultra pure" 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, e.g., most
inorganic analyses, analytical reagent grade is satisfactory. Other analyses, e.g., trace
organic and radiological, frequently require special "ultra pure" 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
below.
Reagents must always be prepared and standardised with care and appropriate
technique against reliable primary standards. They must be re-standardised 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 concentration. Standard solutions should be properly labeled as to
compound, concentration, solvent, date and prepared.
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,
certified reference materials (CRM) are available from the different suppliers (NBS, etc.).
Page 33
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
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 suppliers.
There is some confusion among chemists as to the definition of the terms ANALYTICAL
REAGENT GRADE, REAGENT GRADE, and ACS ANALYTICAL REAGENT GRADE. In
this document, the term ANALYTICAL REAGENT GRADE (AR) will be used. It is
intended that AR chemicals and solvents shall conform to the current specifications set
up by competent committees, advisory groups.
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 vapour from the air, standardization of solutions,
instability, and need for frequent standardisation of certain reagents are among the
factors to be considered and described in the literature. It is recommended that the
analyst become thoroughly familiar with the relevant publications.
Analytical reagent grade nitric and hydrochloric acids must be specially prepared by
distillation in borosilicate glass and diluted with deionised distilled water. All other
reagents and standards are also prepared in deionised 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, e.g., aluminum,
reagent-grade nitrous oxide is required; for analysis of mercury suprapure acid is
needed.
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 sulphate, used for co-precipitation 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 which is imparted to the gas. When this background is above normal, the
gas should not be used for radioactivity determinations.
The minimum purity of reagents and solvents 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 sub-nanogram quantities) of gas chromatography (GC), which is often
used to quantitate organic results, much greater purify is frequently required. 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 sulphur and/or
phosphorus interference. Pesticide quality solvents are available from several sources.
These are often satisfactory for many organic GC determinations. 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 pre-purified dry grade. The nitrogen-
detection system requires the use of ultra-pure hydrogen for satisfactory results. The
Page 35
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
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/or stored in a cool, dark place. It is particularly important to store materials
used for radiological determinations in dark bottles, since photoluminescence will
produce high background if light sensitive detectors are used for counting. Some
reagents require refrigeration.
Adsorbents for column chromatography are stored in the containers that they are
supplied in, or according to the requirements of individual methods. Activated carbon,
used for collection of samples for organic analyses, must be stored and processed in
areas protected from atmospheric and other sources of contamination.
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 due to absorption of
gases or water vapour from the air.
In order to produce high quality analytical data, determinate errors must be eliminated or
at least minimised. For this purposes, 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 she/he must have a regular check program. Materials that do not meet requirements
should be replaced or purified so that they can be used.
various types of analyses are given below. For complete information, the analyst should
consult the individual methods.
Many AR-grade chemicals and solvents, and at times pesticide quality solvents, do not
meet the specifications required for the determination of specific organic compounds.
Impurities 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
Page 38
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
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 useable. Some useful clean-up
procedures are:
• Washing the inorganic reagents with each solvent that the reagent contacts
during the analysis,
• 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,
• Pre-extracting distilled water with solvents used for the particular analysis
involved,
• Pre-extracting aqueous reagent solutions with the solvents involved,
• Redistilling solvents in all-glass systems using an efficient fractlonating column,
• Re-crystallizing reagents and dyes used in colorimetric 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 defector. 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
100-200 psi. Filter driers are of little help in coping with this type of contamination.
To provide some orientation for the user of this guidance, laboratory supplies needed for
a general water chemical laboratory, for trace inorganic and trace organic laboratory are
summarised in Annex 3.
Page 39
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
The high sensitivity of the instrumentation used in trace organic chemical analysis, and
the low concentration of compounds being investigated, dictate that special attention be
given this field of analytical endeavor. Contamination of the sample from any possible
source must be diligently guarded against, and interferences in the sample must be
carefully controlled. Finally, strict attention to method and highly refined technique are
required to produce valid quantitative results.
Sample collection should be done with wide-mouth glass bottles, equipped with screw
caps fitted with Teflon liners. The use of a screw cap without a Teflon liner may cause
contamination of the sample by the liner or adhesive used in sealing the liner to the cap.
Plastic bottles (polyethylene) are not used because traces of plasticizer may be leached
from the plastic by the water, and can be a source of analytical interference. Moreover,
organics from the sample may be adsorbed on the plastic. It has been suggested that
high grade Teflon bottles may be satisfactory for this use, however, the cost is
prohibitive at present.
To insure freedom from organic contaminants, bottles are rinsed successively with
chromate cleaning solution, running tap water, distilled water, and finally several times
with redistilled solvent (e.g., acetone, hexane, petroleum ether, chloroform). Caps are
washed with detergent, rinsed with tap water, distilled water, and solvent. Liners are
treated in the same way as the bottles and are stored in a sealed container.
Each method designates a recommended sample size for surface water analysis.
Duplicate samples are recommended. If analysis by more than one method is to be
requested on the same sample, sufficient sample must be simultaneously obtained to
supply the needs of each analysis.
It is also recommended that when requesting a non-specific analysis, any information
that could help direct the analytical approach, or aid in interpretation of results, be
supplied. Such information could include industrial or agricultural activities in the area
from which the sample was obtained, spills, or other accidents that may have occurred
in the area. Also mention of similar upstream activity could provide valuable assistance.
Samples should be stored in a cool, dark place, and analysed as soon as possible. If the
sample cannot be analysed immediately, reporting the holding time can help in
interpreting results where die-off rates are known.
Page 40
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
The affinity of the activated carbon for organic substances requires that supplies of
carbon be protected from extraneous sources of contamination. For example, carbon
can adsorb organic substances from the air. Therefore, the activated carbon is stored
and processed in an area adequately protected from such sources of contamination. As
an additional precaution, the ventilating, heating, and air conditioning systems for the
laboratories in which carbon adsorption samples are processed should be completely
isolated from all other laboratories.
5.2 Glassware
The minimum purity of reagents and chemicals should be certified (analytical reagent)
grade. Analytical standards should be reference grade, when available. The analyst
should take special note of the assay of less pure materials (most often pesticides). All
reagents and chemicals should be stored according to manufacturer's instructions to
prevent degradation. Proper storage is especially important if the chemical is to be used
in preparing an analytical standard. Refrigerated chemicals should be allowed to come
to room temperature before exposing them to the atmosphere.
When preparing stock solutions, it is recommended that at least 10 mg of material be
used for greater accuracy in weighing. Solutions should be carefully stored so as to
preserve their concentration, and to protect them from ultraviolet radiation. Usually
storage in ground-glass stoppered bottles, either amber-glass or out of the line of direct
lighting, is sufficient.
Standard solutions should be prepared using precision syringes to measure the volume
of stock solution to be diluted. The syringe barrel should be pre-wetted with solvent and
air bubbles expelled. Dilution should be done in a Class A volumetric flask to insure
accurate measurement. If these solutions are to be used frequently, they are best stored
in a screw-cap, septum, sealed vial. These vials allow instant access to the solution and
offer good protection against concentration changes of the standard solution.
Page 41
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
Evaporation of the solvent caused by repeated removal of the cap is a serious problem
with other containers. If septum vials are not available it is advisable to prepare
standard solutions in a volumetric container of 100 ml or more and transfer a small
portion to a separate container for daily use, then discard that portion at the end of the
day.
All stock or standard solutions should be carefully watched fox signs of changes in
concentration or deterioration. As an aid to monitoring these solutions, it is wise to label
them as to compound, concentration, solvent used, date and prepared. Also, in the case
of GC solutions, it is necessary to retain some evidence of its chromatographic behavior
as a fresh solution for comparison at a later date.
Distilled water used as dosed, control samples must be free of organic interferences. A
very effective way of removing organic interferences from distilled water is to pre-extract
the water with the solvent that is to be used in the analysis, then boil the water to
remove the residual solvent.
Hexane-ethyl ether and benzene are commonly used in the extraction of water and
wastewater in conjunction with analysis by electron-capture gas-liquid chromatography.
Because electron-capture detection methods are extremely sensitive to interferences
normally found in these solvents, the cleanest possible reagent grade or pesticide
quality solvents must be used. Redistillation in the lab in an all-glass system might be
necessary. Solvents in the same lot may also vary and therefore each container should
be checked.
Adequate steps must be taken to eliminate or minimize interferences from solvents and
other materials. A blank should be run simultaneously under the same analytical
conditions as any block of samples analysed. A block of samples is defined as any
group of one or more samples analysed using a common batch of analytical supplies.
Should any one of the supplies be changed (i.e., solvent, silica gel, Florisil, etc.) a new
blank is required.
Concentration of sample extracts to very small volumes for trace analysis requires great
care to avoid loss of constituents. A Kuderna-Danish evaporator is a very useful
apparatus to accomplish this operation. Instructions in the use of this evaporator must
be strictly followed to avoid loss of desired sample. Final concentration in an ampoule or
calibrated tube is accomplished in a warm water bath with a gentle stream of clean, dry
air, if air oxidation is not a problem; otherwise, nitrogen should be used. During the final
concentration, the inside walls should be rinsed repeatedly with the working solvent to
insure the total sample is contained in the bottom of the tube. Complete evaporation of
solvent must be avoided to prevent loss of sample constituents. The step should be
accomplished within 10 or 12 minutes for best results.
Page 43
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
It is assumed that a valid sample has been properly taken, preserved, and delivered to
the laboratory for analyses; that the laboratory analyses were done according to
recognised, standardised methods; and that the recording and reporting of subsequent
laboratory results were done in a systematic and uniform way. It must be recognised and
practiced, however, that quality control begins with the sample collection and does not
end until the resulting data are reported. The laboratory control of analytical
performance is one vital link in obtaining valid data, however, use of quality control
between field sampling, laboratory analyses and management decisions are necessary
to insure this validity.
There are a number of different methods available for the determination of precision.
One method that can be successfully employed, and can be adapted to several
analytical instrumentation and chemical procedures, is described as follows:
• To allow for changes in instrument conditions, the precision study should cover at
least two hours of normal laboratory operation.
• In order to permit the maximum interferences in sequential operation, it is
suggested that the samples be run in the following order: high, low, intermediate,
intermediate. This series is then repeated seven times to obtain the desired
replication.
• The precision statement should include a range of standard deviations over the
tested range of concentration. Thus, four standard deviations will be obtained
over a range of four concentrations, but the statement should contain only the
extremes of standard deviations and concentrations studied.
Table 15. Precision data on river water samples for PO4-P, in mg P/l.
"In a single laboratory, using surface water samples at concentrations of 0.06, and
0.62 mg P/l, the standard deviation was ±0.004”
Thus, the statement contains the number of laboratories involved, the type of samples,
the concentrations used and the resulting standard deviation (s).
Table 16. Accuracy data on river water samples for PO4-P, in mg P/l.
Again, in order to contain the key elements, the accuracy statement would read as
follows:
"In a single laboratory, using surface water samples at concentrations of 0.11 and 0.74
mg P/l, recoveries were 90% and 95%, respectively".
Once collected and documented, these precision and accuracy data may be used in a
number of ways. Two important examples are:
(2) They present evidence that the analyst in question is indeed capable of
analysing the water samples for that particular parameter. That is, he has the
standard method under control, and is capable of generating valid data, and
(3) The data can be used in the evaluation of daily performance in reference to
replicate samples, spiked standards and samples, and in the preparation of
quality control charts.
The above methods can be adapted to other chemical procedures and analytical
instruments. They can be used on manual titration methods for such parameters as
alkalinity, chloride, and hardness; on general inorganic instruments such as pH,
conductivity; on TOC and AOX instruments, etc. Other instruments, such as atomic
absorption and flame emission spectrophotometers could also be evaluated by these
methods, however, radiological instrumentation and gas chromatography systems
require special techniques.
Page 46
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
Once valid precision and accuracy data are available on the method, systematic daily
checks are necessary to insure that valid data are being generated. First of all,
verification of the originally-constructed standard curve is mandatory. As previously
noted, at least two standards (a high and a low) should be analysed routinely along with
a blank to determine that comparable operating conditions exist. If the data do not
substantiate such control, the analyst must systematically trouble-shoot his system until
the problem is corrected.
In order to document that reproducible results are being obtained (i.e., precision of the
method), it is necessary to run replicate samples. Although frequency of such replicate
analyses is, by nature, dependent on such factors as the original precision of the
method, the reliability of the instrumentation involved, and the experience of the analyst,
good laboratory technique is to run duplicate analyses at least ten percent of the time.
The resulting data should agree favorably with the known precision of the method. If
they do not, the system is not under control, and results are subject to question.
Concurrently, quality control should include assurance that the daily system is actually
measuring what is in the sample (i.e., accuracy of the method). Although it is far
preferable to have obtained values check with known or actual values, it should be
recognized that inaccuracy does not destroy the value of data if the degree and
precision of the error is known and taken into account. In order to account for
background contamination and/or sample interferences, and as a matter of routine
practice, spiked samples should be used in addition to standards. As in the case of
duplicate sample analyses, good laboratory technique dictates that spiked samples be
run at least ten percent of the time.
Thus, daily control of analytical performance in the laboratory requires approximately 5-
20 percent of the analyst's time. Considering the elapsed time and combined efforts of
skilled personnel that are represented in a final laboratory result, this is a comparatively
small price to pay for, not a ”number", but a valid concentration value.
A most convenient way of recording the obtained precision and accuracy data is through
the preparation of quality control charts. Plotting of said data systematically answers the
question as to whether the laboratory analyses are under control, and is useful in
observing developing trends of positive or negative bias. Because of its importance in
documenting the quality control being practiced daily in the laboratory, the construction
and uses of quality control charts are treated as a separate topic in the next section.
A broader and somewhat different form of evaluation of daily performance may be made
through routine participation in interlaboratory round-robin studies. Samples analysed in
such a cooperative program should be treated as part of the routine sample load. In so
doing, the analyst is able to compare his individual performance against other laboratory
personnel, and to have a reliable measure of the particular method's capabilities. In
many respects such samples can be regarded as reputable "blind samples"; a
necessary ingredient in the quality control of laboratory results.
Page 47
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
Quality control charts were originally developed for the control of production processes
where large numbers of items were being manufactured and inspected on an essentially
continuous basis. As shown in Fig. 3., a control chart consists of a graphical chart with
the vertical scale plotted in units of the test result, or recovery, and the horizontal scale
in units of time or sequence of results. The upper and lower warning and control limits
shown on the chart are used as criteria for action, or for judging the significance of
variations between duplicate samples. The central line represents the average or the
standard value of the statistical measure being plotted.
Average
The accuracy chart for QC samples is from the average and standard deviation of a
specified number of measurements of the analyte of interest. The accuracy
chartincludes upper and lower warning (WL) and control levels (CL). Common practice
is to use ± 2s and ± 3s limits for the WL and CL, respectively. These values are derived
from results from analysis of reference materials. The number of measurements to
calculate standard deviation should be relevant to the statistical confidence limits of
95% for the WLs and 99% for the CLs. Percent recovery can be used in the case of
samples if the concentration varies.
Page 48
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
The precision chart is also constructed from the average and standard deviation of a
specified number of measurements of the analyte of interest. If teh standard deviation of
the method is known the factors given in the following can be used to constuct the
central line and the warning and control limits:
Number of Observations Factor for Central Line Factor for Control Lines
n D2 D4
2 1.128 3.267
3 1.693 2.575
4 2.059 2.282
5 2.326 2.115
In the precision chart, only upper warning and control limits are meaningful. The
standard deviation is converted to the range so that the analyst need only subtract the
two results to plot the value on the precision chart. The mean range is computed as:
_
R = D2 s
the control limit as
_ _
CL = R ± 3s ( R) = D4 R
and the warning limit as
_ _ _ _
WL = R ± 2s ( R ) = R ± 2 / 3( D4 R − R )
where
D2 = factor to convert s to the range (1.128 for duplicates, as above),
s(R) = standard deviation of the range,
D4 = factor to convert mean range to 3s(R) (3.267 for duplicates, as above).
A precision chart is very simple when duplicate analyses of a standard are used.
To start with a quality control chart it is a practical way to measure seven times a check
solution with a concentration level between 40 to 60% of the range. Calculate the mean
and the standard deviation, and follow the preparation of the control chart as mentioned
above.
Page 49
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
To obtain meaningful data on water quality, the laboratory must first collect a
representative sample and deliver it unchanged for analysis. The analyst must then
complete the proper analysis in the prescribed fashion. Having accomplished these
steps, one other important step must be completed before the data are of use. This step
includes the permanent recording of the analytical data in meaningful exact terms, and
reporting it in proper form to some storage facility for future interpretation and use.
It is very important that starting from the sampling through the laboratory work until
reporting the results, the name of the responsible person(s) be included in the records,
reporting sheets.
The brief sections that follow discuss the data value itself, recording and reporting the
value in the proper way, means of quality control of data handling.
Proper use of significant figures given an indication of the reliability of the analytical
method used. Reported 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/l, the "18" must be firm values
while the "0.8" is somewhat uncertain and may be "7" or "9".
The number zero may or may not be a significant figure:
• Final zeros after a decimal point are always significant figures. For example, 9.8
grams to the nearest mg is reported as 9.800 grams.
• Zeros before a decimal point with other preceding digits are significant. With no
other preceding digit, a zero before the decimal point is not significant.
• If there are no digits preceding a decimal point, the zeros after the decimal point
but preceding other digits are not significant. These zeros only indicate the
position of the decimal point.
Significant figures reflect the limits of the particular method of analysis. It must be
decided beforehand whether the number of significant digits is sufficient for
interpretation purposes. If not, there is little that can be done within the limits of normal
Page 50
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
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 to
produce an increase in significant figures.
Once the number of significant figures is established for a type of analysis, data
resulting from such analyses are reduced according to set rules for rounding off.
Addition: When adding a series of numbers, the sum should be rounded oft to the same
numbers 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 is rounded-off to 33.4.
Subtraction: When subtracting one number from another, rounding off should be
completed before the subtraction operation, to avoid invalidation of the whole operation.
Multiplication: When two numbers of unequal digits 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 less accurate number.
Division: When two numbers of unequal digits 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 digits of the less accurate of the divisor or dividend.
Page 51
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
Powers and Roots: 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.
To clarify the meanings of reports and evaluations of data, the following terms are
defined.
7.1.3.2 Average
In ordinary usage, the arithmetic mean. The arithmetic mean of a set on n values is the
sum of the values divided by n.
7.1.3.3 Characteristic
A property that can serve to differentiate between items. The differentiation may be
either quantitative (by variables), or qualitative (by attributes).
7.1.3.4 Error
The difference between an observed value and its true value.
7.1.3.5 Mean
The sum of a series of test results divided by the number in the series. Arithmetic mean
is understood (X).
7.1.3.6 Precision
Degree of mutual agreement among individual measurements. Relative to a method of
test, precision is the degree of mutual agreement among individual measurements made
under prescribed, like conditions.
• 95% Confidence Limits. The interval within which one estimates a given
population parameter to lie, 95% of the time.
7.1.3.8 Sample
A group of units, or portion of material, taken from a larger collection of units, or quantity
of material, which serves to provide information that can be used as a basis for judging
the quality of the larger quantity as a basis for action on the larger quantity or on the
production process. Also used in the sense of a "sample of observations."
7.1.3.9 Series
A number of test results which possess common properties that identify them uniquely.
7.1.3.10 Unit
An object on which a measurement or observation may be made.
7.1.3.11 Variable
A term used to designate a method of testing, whereby units are measured to determine,
and to record for each unit, the numerical magnitude of the characteristic under
consideration. This involves reading a scale of some kind.
The analytical information reported should include the parameter, the details of the
analysis such as burette readings, absorbance, wavelength, normalities of reagents,
correction factors, blanks, and finally, the reported value.
Reporting of data onto loose or ring-binder forms is an older, but much used means of
recording data. It does allow 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 easy loss or misplacement of sheets, mix-ups as to date sequence, and
questionable status in formal display, or for presentation as evidence.
referencing data, through a table of contents, according to time, type of analysis, kind of
sample, analyst, etc.
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 sign off as to the date and acceptability of the
analysis.
A further development of the bound notebook is the commercially available version
designed for research-type work. These note books 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 its detachable duplicate sheet which allows
for up-to-date review by management without disruption of the book in the laboratory.
The cost is about four times that of ordinary notebooks.
Use of bound notebooks is essentially limited to research and development work where
an analysis is part of a relatively long project, and where the recording in the notebook
is the prime disposition of the data until a status or final report is written.
In most instances the supervisor and analyst wish to look at the data from a sample
point in relation to other sample points on the river or lake. This review of data by the
supervisor, prior to release, is a very important part of the laboratory's quality control
program; however, it is not easily accomplished with bench sheets. For this purpose, a
summary sheet can be prepared which compares a related group of analyses from a
number of stations. An example is shown in Figure 7-2. Since the form contains all of
the information necessary for reporting data it is used also to complete the data forms
forwarded to the storage and retrieval system.
The forms used to report data to data storage systems require a clear identification of
the sample 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 in an incorrect parameter. As a group of analyses is completed on one or
more samples, the values are reported in floating decimal form, along with true code
numbers, for identifying the parameter and the sampling point (station).
Page 54
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
8 SKILLS
• Sample Handling. The analyst should understand thoroughly when the sample is
to be settled, agitated, poured, pipetted, etc., before removal from the container.
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 with each analyst and point out, when
necessary, areas of needed improvement.
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, improved
efficiency and improved data quality will result.
8.1 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, estimates of the number of measurements that can be made per unit of
time are difficult, because of the variety of factors involved. If the analyse is pushed to
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 the following table, 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 arbitrary rating numbers for the
degree of skill required are footnoted in the tables, but are explained more fully below:
A tacit assumption has been made that multiple analytical units are available for
measurements requiring 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 normally anticipate.
8.2 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.
Page 57
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
References, literature
APHA, AWWA, WEF (American Public Health Association, American Water Works
Association, Water Environment Federation): Standard Methods for the Examination of
Water and Wastewater, 20th Edition 1998. American Public Health Association.
European Standard EN 45001: General criteria for the operation of testing laboratories.
1989.
D.T.E. Hunt and A.L. Wilson: “The Chemical analysis of Water”, General Principles and
Techniques. The Royal Society of chemistry, London, 1986.
E. Mullins: Introduction to control charts in the analytical laboratory. Analyst, Vol. 119,
369-375, 1994.
RIZA: Quality Assurance. RIZA report no.: 95.067. Authors: J.G. Timmerman (RIZA),
M.J. Gardner (WRc), J.E. Ravenscroft (WRc). RIZA, Lelystad 1996.
Page 58
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
____________
Annex 1.
ISO 5667-4:1987 Part 4: Guidance on sampling from lakes, natural and man-made
ISO 5667-13:1997 Part 13: Guidance on sampling of sludges from sewage and water
treatment works
ISO 5667-15:1999 Part 15: Guidance on preservation and handling of sludge and
sediment samples
ISO 8265:1988 Water quality. Design and use of quantitative samplers for benthic
macro-invertebrates on stony substrate in shallow freshwaters
Page 59
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
____________
ISO 8692:1989 Water quality. Freshwater algal growth inhibition test with
Scenedesmus subspicatus and Selmastrum capricornutum
ISO 1484:1997 Water analysis. Guidelines for teh determination of total organic
carbon (TOC) and dissolved organic carbon (DOC).
ISO 9297:1989 Water quality. Determination of chloride. Silver nitrate titration with
chromate indicator (Mohr’s method)
Annex 2.
1. Analytical Balances
There are many fine balances on the market designed to meet a variety of needs such
as sensitivity, speed weighing, batch weighing, etc. Types of balances include general
purpose, micro-, electro-, semi-analytical, analytical and other special purpose
instruments. The analytical balance is one of the most important equipment in the
production of reliable data.
Most analytical balances in use today in well-equipped laboratories are of the "single
pan" variety. Single-pan capacities range from 80 grams to the 200-gram models with
sensitivities from 0.01 to 0.1 mg. Features of single-pan balances include mechanical
lifting and substitution of weights, digital readout of weights, and mechanical zeroing of
the empty balance. The advantage of the single-pan balance over the old "two-pan"
balance is in greatly increased weighing speed and improved weighing accuracy
because of mechanical weight handling. With all the design improvements, however, the
modern analytical balance is still a fragile instrument, subject to shock, temperature and
humidity changes, mishandling and various other insults. Some of the precautions to be
observed are as follows:
A basic pH meter consists of a voltage source, amplifier, and readout device, either
scale or digital. Certain additional refinements produce varying performance
characteristics between models. Some models incorporate expanded scales for
increased readability, operating stability and extreme accuracy. All instruments of recent
design also include temperature adjustment 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 analytical work, 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 when changing from buffered to un-buffered samples.
Electrodes, both glass and calomel, should be well rinsed with distilled water after each
reading, and should be rinsed or dipped several times into the next test sample before
the final reading is taken. Weakly buffered samples should be stirred during
measurement. Glass electrodes should not be allowed to become dry. When not in use
they should be immersed in distilled water.
The first step in standardization of the instrument is done by immersing the glass and
calomel electrodes into a buffer of known pH, setting the meter scale or needle to the pH
of the buffer and adjusting the proper controls. The temperature compensating dial
should be set at the sample temperature. The pH of the standard buffer should be within
about two pH units of the sample. For best accuracy, the instrument should be
calibrated against two buffers that bracket the pH of the samples.
Standard buffer solutions, covering a range of pH, may be purchased from several
chemical suppliers and are satisfactory for routine use. Table 5. gives a list of buffers
(easily made in the laboratory) and the resulting pH at different temperatures. The effect
of temperature on pH may be obtained by observing temperature as pH of various
butters shown in the table. A rough rule is that: temperature compensation is about 0.05
pH units per 5 degree increase in temperature.
The method for the use of the fluoride electrode specifies use of multiple standards in
the range between 0 to 2 mg/liter, because this system has supplied very precise data
when compared to the colorimetric methods using the same set of standards. The
system of incremental addition appears to have considerable merit since the electrode
response is established in the presence of possible interference.
When an ion-selective electrode appears to be malfunctioning, the same check system
may be used as for a faulty glass pH electrode. It is unlikely, however, that the electrode
will be cracked, it will probably be dry, or insufficiently filled with the necessary solution.
The probe assembly and instructions for refilling customarily accompany the item when
shipped by the manufacturer and said instructions should be followed by the user.
3. Conductivity Meters
Solutions of electrolytes conduct an electric current by the migration of ions under the
influence of an electric field. 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 or mhos. For
natural water samples where the resistance is high, the usual reporting unit is in
micromhos (µS/cm).
Practically all conductivity meters on the present market use direct read-out meter for
indicating solution conductivity, and include a stepping switch for varying resistance in
steps of 10X. The instruments are therefore capable of reading conductivity from about
0.1 µS/cm to about 250,000 µS/cm.
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.
In routine use, cells should be frequently examined to insure that (a) platinized coating
of plates is intact, (b) plates are not coated with suspended matter, (c) plates are not
bent, distorted, or misalign.
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:
• Standardize the cell and establish a cell factor by measuring the conductivity of a
standard potassium chloride solution. 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 7. can be used for this operation,
• Again, immerse the cell in the sample several times before obtaining a reading.
K1 + K2
L=
1,000,000 x Kx
where
L = cell constant
K1 = conductivity, in µS/cm, of the KCl solution at the temperature of measurement
K2 = conductivity, in µS/cm, at the same temperature, of the distilled water
used to prepare the reference solution
Kx = measured conductance, in µS/cm
4. Spectrophotometers
sensitive, and lack the versatility of spectrophotometers. They are used most profitably
where a single method can be designed to fit the instrument.
Each of the essential features listed, especially the monochromator and the photo-
detector system, vary in design principles from one instrument to another.
The ultraviolet and visible regions of the spectrum are of most interest in fluorimetry and
absorption in these regions causes the excitation of the outermost electrons of the
molecule. The energy associated with radiation of this frequency is quite high, around
100 kilogram calories per einstein, and is sometimes sufficient to break down the
absorbing molecules, as for instance with the fading of dyes by the action of sunlight.
The absorption of light results in the formation of excited molecules which can in turn
dissipate their energy by decomposition, reaction, or re-emission.
Since the emission of fluorescence always takes place from the lowest vibrational level
of the first excited state, the shape of the emission spectrum is always the same, despite
changing the wavelength of exciting light. A plot of emission against wavelength for any
given excitation wavelength is known as the emission spectrum. If the wavelength of the
exciting light is changed and the emission from the sample plotted against the
wavelength of exciting light, the result is known as the excitation spectrum.
Since the amount of energy abstracted is always constant, Raman bands appear
separated from the incident radiation by the same frequency difference, irrespective of
the wavelength of the exciting light. If the Raman band of the solvent coincides with the
fluorescence emission of the solute, separation can be achieved by changing the
excitation wavelength to a lower value, hence the Raman band will also be lowered.
Since the wavelength of fluorescence emission is independant of exciting wavelength
the fluorescence will be separated from the Raman scatter. Raman scatter is only
important at high sensitivity since the phenomenon is weak but must always be born in
mind. The Raman emission of some solvents is quite complex and an emission
spectrum of the solvent blank should always be run prior to the analysis. The
wavelength of the Raman scatter in different solvents compared to selected excitation
wavelengths is shown in Table 9.
All solvents containing hydrogen atoms linked to either carbon or oxygen show a Raman
band shifted approximately 3000 cm-1 from the exciting radiation. If interference is
observed, carbon tetrachloride or chloroform should be considered, since the main
Raman bands of these solvents are much closer to the excitation wavelength.
Page 71
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
____________
Instruments
All fluorescence instruments contain three basic items, a source of light, a sample
holder and a detector. In addition, to be of analytical use, the wavelength of incident
radiation needs to be selectable and the detector signal capable of precise manipulation
and presentation. In simple filter fluorimeters, the wavelengths of excited and emitted
light are selected by filters, which allow measurements to be made at any pair of fixed
wavelengths. Simple spectrofluorimeters have a means of analysing the spectral
distribution of the light emitted from the sample, the fluorescence emission spectrum,
which may be by means of either a continuously variable interference filter or a
monochromator. In more sophisticated instruments monochromators are provided for
both the selection of exciting light and the analysis of sample emission and such
instruments are also capable of measuring the variation of emission intensity with
exciting wavelength, the fluorescence excitation spectrum. In addition, synchronous
scanning of both excitation and emission spectra might be usuful.
In principle the greatest sensitivity can be achieved by the use of filters, which allow the
total range of wavelengths emitted by the sample to be collected, together with the
highest intensity source possible. In practice, to realise the full potential of the
technique, only a small band of emitted wavelengths are examined and the incident light
intensity is not made excensive, to minimise the possible photodecomposition of the
sample and to increase selectivity.
The majority of fluorescence assays are carried out in solution, the final measurment
being made upon the sample contained in a cuvette. Cuvettes may be either circular,
square or rectangular, and must be constructed of a material, e.g., silica, that will
transmit both the incident and emitted light. Square cuvettes, or cells will be found most
precise since the parameters of pathlength and parallelism are easier to maintain during
manufacture.
The cuvette is placed normal to the incident beam, the resulting fluorescence is given
off equally in all directions, and may be collected from either the front surface of the cell,
at right angles to the incident beam, or in-line with the incident beam. Some instruments
will provide the option of choosing which collecting method should be employed, a
choice based upon the characteristics of the sample. A very dilute solution will produce
fluorescence equally from any point along the path taken by the incident beam through
the sample. Under these conditions the right angled collection method should be used
since it has the benefit of minimising the effect of light scattering by the solution and
cell. This is the usual measuring condition in analytical procedures. Although
fluorescence takes place from every point along the light path, only a small fraction of
this emission is actually collected by the instrument and transmitted to the detector.
The result is that much of the solution does not contribute to the fluorescence emission
and the same intensity will be observed from a much smaller volume of solution
contained in a microcell whose dimensions more closely match the optical
considerations of the instrument.
Page 72
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
____________
Other types of cuvette include flowcells, e.g., used in the monitoring of continuous flow
systems, cells with thermostatted water jackets and micro or semi-micro cells.
Fluorescence measurements can also be carried out on solid samples by means of the
front surface collection technique.
Cuvettes
rather than precision cuvettes without appreciable loss in precision. This benefit is
derived from the geometrical layout of simple fluorimeters where only a small central
area of the cuvette is actually viewed by the detector so that the overall dimensions of
the cell are less important. However, this statement needs careful qualification since the
use of laboratory grade test-tubes will result in deviation arising from other sources.
Variations in cell wall width can give rise to errors by distorting the lens action of the
round cell wall from one cell to another and the native fluorescence of the material used
to manufacture the cell will often produce large blank values. Obviously the cell must be
fabricated from a material which will transmit both the excitation and emission
wavelengths of interest and if these are both in the visible region, savings can be made
by the use of glass or even plastic sample containers. Only those cells recommended by
the manufacturer of the instrument should be employed and every new batch should be
carefully checked to make quite sure that their native fluorescence at the analytical
wavelength chosen is minimal. Even the highest grades of silica and glass have some
fluorescence although it should not contribute appreciably to routine measurements.
Cells and other glassware used for fluorimetric analysis should be carefully cleaned,
preferably by boiling in 50% nitric acid followed by thorough rinsing in distilled water.
Page 73
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
____________
The usable wavelength range of some common solvents is above their light absorption,
and in most cases analytical reagent grade will be sufficiently pure. However, it is
essential to examine the quality of solvents periodically since small traces of
contaminants may be enough to produce high black values, particularly in some part of
the féuorescence spectra. If the solvent blank is appreciable it can usually be reduced to
a reasonable value by distillation or washing with acid, base or water. Water is the most
common solvent and deionised-distilled water should always be employed. Solvents
should not be stored in plastic containers since leaking of organic additives and
plasticizers will produce high blank values. Fluorescence derived from contaminants
should not be confused with the Raman scatter from the solvent itself. All other reagents
used in the assay should be carefully controlled and high quality grades are to be
recommended. Reagent blanks should always be carried through the analytical
procedure and the actual value of the blank determined before the instrument is re-
zeroed. High or changed blank values should immediately cast suspicion upon the
solvents and reagents employed.
Inner-filter Effects
There are a number of differences in the basic design and accessories for atomic
absorption equipment that require consideration before purchase and during
subsequent use. These choices concern the light source, the burners, optical systems,
readout devices, and mode of conversion. Some of these choices are not readily
obvious, and require that the purchaser or user be familiar with the types and numbers
of samples to be analysed and the specific elements to be measured before a choice is
made. For a programme analysing a wide variety of samples for a number of elements
at varying concentrations, an instrument of maximum versatility would be required.
Lamp Mounts
Fox optimal use of the instrument, certain precautions should be observed. After the
proper lamp has been selected the hollow cathode current should be adjusted according
to the manufacturer's recommendations and allowed to electronically stabilize (warm up)
before use. This requires approximately 15 minutes. During this period, the
monochromator may be positioned at the correct wavelength and the proper slit width
selected. For those instruments employing a multi-lamp turret, a warm-up current is
provided to those lamps not in use, thereby minimizing the warn-up period when the
turret is rotated. In a single-lamp instrument, the instability exhibited during warm-up is
minimized by the use of a double-beam optical system.
Multi-element lamps are considerably cheaper per element than single element lamps,
but the savings may not 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 results from the different volatilities of
metals used in the cathode. One metal volatilises more rapidly than the others and re-
deposits upon the cathode causing an increase in surface area of that metal, and
decreasing the exposed 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. The use of single-element lamps result in more precise and accurate data.
Page 75
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
____________
Recent improvements in design and manufacture of hollow cathode lamps have resulted
in lamps with more constant output and a longer life. Under normal conditions a HC
lamp may be expected to operate satisfactorily for several years. Operating current and
voltage will be indicated on the lamp and should not be exceeded during use. An
increase in background noise and/or a loss of sensitivity are signs of lamp deterioration.
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 and the 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 pre-mix burner. There are many variations of
these two basic types, such as the Boling, the high-solids, the turbulent-flow, the tri-
flame, nitrous-oxide burner 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 burners can be adjusted laterally, rotationally and
vertically for selection of the most sensitive absorbing area of the flame for the specific
element sought. The vertical adjustment is probably the most important since the
position of greatest sensitivity varies from element to element.
There is a great deal of existing uncertainty among instrument users about the relative
merits of single-beam and double-beam instruments. Neither system is the final answer.
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.
Page 76
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
____________
The chopper alternately reflects and passes each beam, creating two equal beams
falling alternately upon the detector. lf 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, which is amplified and
measured. Any difference between the reference and sample beam is measured as a
direct function of absorbed light. The advantage of the double-beam design, therefore,
is that any variations in the source are of reduced importance, and smaller dependence
is placed upon the stability of the power supply. However, stabilization of the power
supply can eliminate the apparent need for the split-beam system. Further, a beam
splitter requires use of additional mirrors or optical accessories that cause some loss of
radiant energy. Neither system, however, compensates for variation in flame intensity.
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, better signal-to-noise ratio and therefore better precision and
improved sensitivity. Because the simplified optical system conserves radiant energy,
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 strong resonance lines below 350 nm, even those slightly below 300 nm.
Readout Devices
Early models of the AAS instrument offered only a meter, calibrated in percentage
absorption. In the surge of competitive design, more sophisticated readout devices were
built into or offered as accessories to various models. At the present time any desired
readout method may be obtained with almost any instrument. Most of the instruments
offer any combination of built-in digital scalers, calibrated in concentration, external
digital printout in concentration, etc. Even inexpensive instruments are built with
recorder interfacing.
Choice of a readout system is predicated largely upon laboratory needs and availability
of budget. In general, any step toward complete automation is desirable but the degree
of automation should be compatible with the laboratory programme.
Miscellaneous Accessories
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 programme performs a large number of repetitious measurements daily, an
automatic sample changer would not be required.
The manufacturer's instructions for proper use should be followed in all cases. Several
safeguards against misuse of the instruments, however, are mandatory.
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.
Although most instruments contain built-in transformers for stabilization of the electronic
circuits, an exterior, high capacity, constant-voltage transformer is recommended for
general laboratory control.
Concerning the absorption cells, all cells should be kept clean, free of scratches,
fingerprints, smudges and evaporated film residues. 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 %T or absorbance values. If a cell is mismatched it should
be discarded or reserved for rough work.
Generally speaking, trained technicians may operate any of the spectrophotometers
successfully. However, interpretation of data from the sophisticated instruments
becomes increasingly complex, and requires more training and specialization. IR and
fluorescence interpretation requires special training, and because of the special
techniques of sample preparation, instrument operation, and interpretation of absorption
curves and spectra, mere compliance with the operations manual is not sufficient.
Total organic carbon (TOC) as a measure of the carbon content of dissolved and
undissolved organic matter present in the water is one of the principal parameters of
water quality. In routine water analysis the organic pollution is generally described in
terms of so called "non-specific" group parameters, such as biological oxygen demand
(BOD), chemical oxygen demand (COD), total oxygen demand (TOD), total organic
carbon (TOC) and UV absorption. Of these parameters, TOC is considered most
relevant, since this parameter not only reflects the quantitative aspects of the organic
pollution best, but accurate and rapid instrumental techniques are available for its
measurement. TOC like the other "non-specific" parameters does not give information
on the nature of the organic substance.
In addition to organic carbon the water sample may contain carbon dioxide or ions of
carbonic acid. Prior to the TOC determination, it is essential that this inorganic carbon is
removed by purging the acidified sample with a gas which is free from CO2 and organic
compounds. Alternatively, both total carbon (TC) and total inorganic carbon may be
determined and the organic carbon content (TOC) may be calculated by subtracting the
total inorganic carbon from the TC. This method is particularly suitable for samples in
which the total inorganic carbon is less than the TOC.
Page 78
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
____________
Cyanide, cyanate, and particles of elemental carbon (soot) when present in the sample,
will be determined together with the organic carbon.
Total carbon (TC): The sum of organically bound and inorganically bound carbon
present in water, including elemental carbon.
Total inorganic carbon (TIC): The sum of carbon present in water, consisting of
elemental carbon, total carbon dioxide, carbon monoxide, cyanide, cyanate, and
thiocyanate. TOC instruments mostly register as TIC only the CO2 originating from
hydrocarbonates and carbonates.
Total organic carbon (TOC): The sum of organically bound carbon present in
water in dissolved form and/or associated with the suspended matter. Cyanate,
elemental carbon and thiocyanate will also be measured.
Dissolved organic carbon (DOC): The sum of organically bound carbon present in
water originating from compounds which will pass an 0.45 µm pore size membrane
filter. Cyanate and thiocyanate are also measured.
Volatile organic carbon (VOC), (POC): Under the conditions of TOC measurement
purgeable organic carbon.
Non volatile organic carbon (NVOC): Under the conditions of TOC measurement
non-purgeable organic carbon.
The sample handling and processes considered during the TOC measurement are as
follows:
A number of devices designed to measure the organic content of aqueous samples have
appeared on the market within the last decades. The oldest, or first of these instruments,
is the Beckman Carbonaceous Analyzer. The apparatus measures organic carbon as
carbon, by oxidizing a very small sample at a temperature of 900o, in a stream of
oxygen, converting all organically bound carbon to carbon dioxide, which is then
measured by a Golay-type thermal detector. The instrument is able to detect about 20
mg of carbon. A modification of the original Beckman Carbonaceous Analyzer employs a
dual-combustion-tube system operating at different temperatures to distinguish between
total carbon and inorganic carbon. Organic carbon is found by difference.
Because there are several new instruments - even using different principles for
digestion, combustion of the organic matter -, on the market, the determination of TOC
concentrations of the samples in accordance with the instrument manufacturer's
instructions. Some important considerations are asfollows:
• In the case of direct determination of the TOC, the total inorganic carbon (ensure
that the pH is below 2) should be removed prior to analysis. Loss of volatile
organic substances should be carefully minimised.
• The TOC concentration should be within the working rage of the calibration. This
can be achieved by diluting the sample.
• Prior to each batch of TOC determinations (for example 10 determinations) carry
out control experiments at the intervals recommended by the manufacturer,
but at least daily.
• After acidification, blow a stream of pure inert gas free from CO2 and organic
impurities through the system (for approx. 5 min) in order to remove CO2.
Only a few natural substances contain halogens in their molecules, and chlorinated
organic compounds are usually synthetically produced. Because of their favourable
qualities, such as non-inflammability, great dissolving power and high reactivity, they
have found widespread use in households and industry throughout the world. They are
used in degreasing solvents, cooling liquids, insecticides, hydraulic and transformer
fluids, dry cleaning and bleaching processes.
Unfortunately their environmental risk is also considerable. Most of them are resistant to
biodegradation, they are lipophilic and concentrate in fatty tissue. As a consequence
they remain in the biosphere for long periods, causing damage in many different ways.
Although generalisation is not possible, it can be stated that, in addition to being acutely
toxic to human beings, many organohalides are proven or suspected carcinogens. For
example, discussions about dioxins and fluorochloro-hydrocarbons have raised much
Page 80
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
____________
public concern. Ever since the discovery that drinking water contains haloforms being
formed by the reaction of the disinfectant chlorine with organic material, an intensive
investigation of the sources and possible consequences has been in progress.
The extremely large number of organohalides that are of environmental concern has
created a need for a method for their detection in effluents and drinking water. In order
to avoid complicated and time-consuming analysis of single species within this category
of compounds and for convenience, a general method for organohalogens was sought.
The analysis of organohalogens in water itself is not possible, due to their low
concentration and the considerably higher amount of inorganic chlorides. Thus a
pretreatment for the isolation and concentration of the organic compounds and the
separation of interfering inorganic compounds had to be developed.
In the early 1970s Kuhn and Sontheimer in Germany developed the AOX method for the
analysis of organohalogens, consisting of adsorption on activated carbon, rinsing off the
inorganic chlorides, pyrolysis of the organohalogens and microcoulometry. A variation of
this method by Dressman led to the American EPA method 450.1. In Sweden, where the
pulp industry produces large amounts of bleaching effluents, the TOCl method was
developed, comprising adsorption on XAD resin and Schöniger combustion. The AOX
method was standardised in Germany first (DIN 38409H14) and incorporated in official
Iegislation. Having the highest recovery rates and being the most practicable, this
method of determination has meanwhile come into worldwide use. In Sweden it has
superseded the total organic chlorine (TOCl) method.
A criticism often raised is that the AOX method does not differentiate between
hazardous and non-hazardous substances. Unfortunately there is no quick screening
method available to identify hazardous compounds. Moreover, the multitude of organic
materials in industrial effluents that can react in many different ways would render such
methods ultimately unsatisfactory.
Principles
AOX is the determination of the amount of organic halogen that adsorbs on activated
carbon. After combustion with oxygen at 900 oC, halide is detected by microcoulometric
titration with silver.
Using the AOX method, organic chlorides, bromides and iodides can be quantitatively
determined. The result is expressed in milligrams of chlorine per litre, regardless of the
actual composition, because usually the chlorinated compounds constitute the
overwhelming majority.
Prior to adsorption, the water sample is acidified with nitric acid to enhance its affinity for
charcoal. Then it is shaken with activated carbon; within one hour the organic
compounds are adsorbed. A second method, also in common use, employs small
columns filled with charcoal and connected in series, through which the water sample is
pumped. After filtration with a membrane filter and rinsing with nitrate solution to remove
the inorganic halogens, the filter with the carbon is transferred to a furnace. There it is
pyrolysed in an oxygen flow at 1000 oC and the resulting hydrohalides are passed to a
vessel containing sulphuric acid for drying. They then proceed to a microcoulometric
titration cell where the halides react with silver ion, the concentration of which is
maintained electrochemically, and are precipitated. The ion concentration drops and
additional silver ion is generated. The charge required to generate that amount of silver
ion is measured by an integration circuit and expressed in terms of micrograms of
chlorine, which may be read from a digital meter on the instrument or calculated from a
plotter diagram. The analysis of organic fluorides by this method is not possible because
silver fluoride dissolves easily in the electrolyte.
This method was originally developed for drinking water and the detection limit is 10 µg.
When analysing effluents some additional points have to be taken into account. First of
all some samples may need to be homogenised using a magnetic stirrer, which is often
the case with textile effluents. Also the DOC value (dissolved organic carbon) is very
important. The DOC of an AOX sample must not exceed 10 mg/I, otherwise the charcoal
would be overloaded and the adsorption would be incomplete. If the DOC values of
samples exceed the 10 mg/l, the samples have to be diluted accordingly. The contents
of inorganic chloride also have to be checked because more than 1 mg/l could interfere
with the analysis. These conditions prove problematic on samples with high DOC values
and low AOX values (< 0.5 mg/l).
If the sample is not to be analysed immediately after sampling, the liquor is acidified to
pH 2 to prevent hydrolysis of the organic halides. To samples that may contain residual
chlorine, sodium sulphite is added, to prevent further reaction and because free chlorine
would lead to an incorrect AOX value. However, a reduction in AOX levels is observed
by dechlorination with sulphite under neutral and basic conditions. This may be
explained by hydrolysis or reaction of sulphite with the chlorinated compounds. Addition
Page 82
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________________
____________
This method detects hydrophilic and lipophilic or polar and non-polar compounds. The
recovery varies with different substances, e.g., chlorophenols and other aromatic
compounds range from 90 to 95%. Worst are the highly volatile compounds that adsorb
very poorly and are better determined as purgable organic halogen compounds
(POX). For POX analysis the sample is purged with oxygen in a vessel attached to the
analyser and the purgable compounds are passed directly to the combustion unit.
Afterwards the residual aqueous sample is used for AOX analysis.
Considerations
There is still much discussion about the use of AOX as a meaningful environmental
parameter. Environmental scientists urgently need a fast measurement of the degree of
pollution in water, wastewater, soil, etc. Biological tests have proven that the
compounds measured with AOX are really very harmful for all life on earth.
Annex 3.
1.Sampling
2. Sample preparation
3.Analysis
4. Chemicals
pH paper
Acids
Standards
Distilled water
Reagents and measure solutions for methods
Page 85
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________
____________________ ______________________________
1. Sampling
Quantity
Item Size
(pcs.)
100, 250,
Bottles (PE or PP) 100-100
500 ml
Membrane filter unit with
-cellulose nitrate/acetate membranes 0.45 µm 250
Dropping bottles (PE) 50 ml 10
2. Sample preparation
Quantity
Item Size
(pcs.)
Membrane filter unit with
-cellulose nitrate/acetate membranes 0.45 µm 250
100, 250,
Bottles for prepared samples (PE or PP) 100-100
500 ml
10, 25, 50,
Graduated cylinders PMP (TPX) 100, 250, 500, 5-5
1000 ml
10, 25, 50,
Beakers PMP (TPX) 100, 250, 500, 10-10
1000 ml
10, 25, 25-25
Volumetric flasks PMP (TPX) 50, 100, 250, 50-50
500, 1000 ml 5-5
50, 100,
Erlenmeyer flasks PMP (TPX) and glass 50-50
250 ml
Stirrers PP 10
0.4-2 ml 1
Dispensers 2-10 ml 1
5-30 ml 1
Pipettes with tips 10 µl…10 ml 1-1
Funnels 50…150 mm 5-10
Wash-bottles PE 500 ml 5
Page 86
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________
____________________ ______________________________
Quantity
Instruments and accessories Features
(pcs.)
Analytical balance 0.1 mg 1
Top-loading balance 0.1 g 1
Hot plate & stirrer 1
Refrigerator for sample and reagent storage 2
Digestor
a) Microwave system with -a rotor 1
-medium and high pressure vessels (TFM) 12-12
-water cooler 1
-exhausting unit 1
b) Stainless steel coated TFM bombs, 25
-drying oven 1
c) Multi-unit extraction heater with 6-plate
-glass extraction apparatus 15
3. Analysis
Quantity
Instruments and accessories Features
(pcs.)
UV-VIS spectrophotometer with
-cuvettes (glass and/or quartz) different
Flame photometer
AAS (flame, graphite furnace, hydride and cold vapour
technique) with
-hollow cathode lamps,
-EDL power supply,
-sample changer with vessels,
-gas cylinders and regulators,
-compressor,
-graphite tubes
ICP-OES with
-sample charger with vessels,
-gas cylinders and regulators
4. Chemicals
pH papers
Acids (HCl, HNO3, H2O2, H2SO4) Suprapur
Standards
High purity water
Reagents for AAS hydride and cold vapour technique
Matrix modifiers for AAS graphite furnace technique
Page 87
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________
____________________ ______________________________
1. Facilities
2. Sample preparation
Item Size
Quantity
Dropping bottles 5
SPE cartridges 250 mg C18 500
SPE cartridges 500 mg SiOH 500
ENVI C18 disk 47 mm 10 pk
Pipettes 1.2.5.10 mL 50
Beakers (glass) 25.50.75.150.500. 10.20.5.15.5.5
1000 mL
Volumetric Flasks ( glass) 25.50.100.250. 200
500 mL
Petri dishes 50
Evaporating dishes 200
Funnels 10
Graduated cylinders 5.10.50.100.250. 15
1000 mL
Sampler bottles 150 mL 150
1000 mL 50
5000 mL 100
Page 88
Technical report: Guidance to operation of water quality laboratories
__________________________________________________________________________________________________________________________________________________________________________________________________________________
____________________ ______________________________
3. Analysis
4. Chemicals
pH papers
Acids ( HCl, HNO3, H2O2, H2SO4) Suprapure
Organic solvents ( Suprapure for org. Trace analysis)
Standards
Deuterated standards for GC/MS
High purity water
Derivatization agents