ENVIRONMENT AGENCY

The determination of chemical oxygen demand in waters and effluents (2007)

Methods for the Examination of Waters and Associated Materials

2

The determination of chemical oxygen demand in waters and effluents (2007)

Methods for the Examination of Waters and Associated Materials

This booklet contains guidance on methods for the determination of chemical oxygen demand using potassium dichromate solution, and notes for the disposal and recovery of selected reagents. This document revises the document published in 1986. Five methods are described and these include:
A

A reference large scale (10 ml) flask digestion procedure with mercury suppression.

B
A small scale (2 ml) flask digestion procedure with mercury suppression and spectrophotometric determination.
C
A mercury-free large scale (10 ml) flask digestion procedure using chromium(III) potassium sulphate and silver nitrate solutions.
D
A mercury-free small scale (2 ml) flask digestion procedure using chromium(III) potassium sulphate and silver nitrate solutions.
E
A mercury-free small scale (2.5 ml) flask digestion procedure using chromium(III) potassium sulphate and silver nitrate solutions.
Throughout the booklet, the term chemical oxygen demand (COD) is used to express the amount of oxygen consumed during oxidation of a sample with hot acid dichromate solution under defined conditions; the test provides an estimate of the oxidisable matter present in the sample. The result is usually expressed as milligrams of oxygen consumed per litre of sample (mg l-1). All of the methods described in this booklet are empirical, and are based upon the oxidation of organic matter using acid dichromate solution followed by spectrophotometric determination or titrimetric determination of residual dichromate with iron(II) ammonium sulphate solution.
The following booklet in this series contains details on an inter-laboratory trial on COD determinations using some of the methods described in this booklet.
Whilst this booklet may report details of the materials actually used, this does not constitute an endorsement of these products but serves only as an illustrative example.
Equivalent products are available and it should be understood that the performance characteristics of the method might differ when other materials are used. It is left to users to evaluate methods in their own laboratories.
3

Contents
About this series
Warning to users

6
6

The determination of chemical oxygen demand in waters and effluents
1
Introduction
2
Sample collection, preservation and storage
3
Hazards
4
Principle
5
Analytical quality control solutions

7
7
9
10
11
12

A
A reference large scale (10 ml) flask digestion procedure with mercury suppression 13

A1
Performance characteristics of the method
A2
Reagents
A3
Apparatus
A4
Analytical procedure
A5
Calculation
A6
Chloride interference
A7
Modified procedure using mixed reagents
Tables A1 and A2

13
13
15
15
17
17
18
19

B
A small scale (2 ml) closed-tube digestion procedure with mercury suppression and spectrophotometric determination

20

B1
Performance characteristics of the method
B2
Reagents
B3
Apparatus
B4
Analytical procedure
B5
Calculation
B6
Chloride interference
B7
Modified procedure using mixed reagents
B8
Open-tube procedure
B9
Analytical procedure
B10 Calculation
B11 Chloride interference
B12 Modified procedure using mixed reagents
Tables B1 and B2

20
20
22
22
25
25
25
26
27
28
29
29
30

4

C
A mercury-free large scale (10 ml) flask digestion procedure using chromium(III) potassium sulphate and silver nitrate solutions

32

Cl
Performance characteristics of the method
C2
Reagents
C3
Apparatus
C4
Analytical procedure
C5
Calculation
C6
Chloride interference
C7
Modified procedure using mixed reagents
Tables C1 - C4

32
32
34
35
36
37
38
38

D
A mercury-free small scale (2 ml) flask digestion procedure using chromium(III) potassium sulphate and silver nitrate solutions

42

D1
Performance characteristics of the method
D2
Reagents
D3
Apparatus
D4
Analytical procedure
D5
Calculation
D6
Chloride interference
D7
Modified procedure using mixed reagents
Tables D1 - D6

42
42
44
45
47
47
48
49

E
A mercury-free small scale (2.5 ml) flask digestion procedure using chromium(III) potassium sulphate and silver nitrate solutions

52

E1
Performance characteristics of the method
E2
Reagents
E3
Apparatus
E4
Analytical procedure
E5
Calculation
E6
Chloride interference
E7
Modified procedure using mixed reagents
Tables E1 - E7

52
52
54
54
56
56
57
58

F

References

62

G
G1
G2

Waste disposal and recovery of mercury and silver
Recovery of silver
Recovery of mercury

63
63
63

Address for correspondence
Members assisting with these methods

64
64

5

About this series
Introduction
This booklet is part of a series intended to provide authoritative guidance on recommended methods of sampling and analysis for determining the quality of drinking water, ground water, river water and sea water, waste water and effluents as well as sewage sludges, sediments, soil (including contaminated land) and biota. In addition, short reviews of the most important analytical techniques of interest to the water and sewage industries are included.

Performance of methods
Ideally, all methods should be fully evaluated with results from performance tests. These methods should be capable of establishing, within specified or pre-determined and acceptable limits of deviation and detection, whether or not any sample contains concentrations of parameters above those of interest. For a method to be considered fully evaluated, individual results from at least three laboratories should be reported. The specifications of performance generally relate to maximum tolerable values for total error (random and systematic errors) systematic error (bias) total standard deviation and limit of detection. Often, full evaluation is not possible and only limited performance data may be available. In addition, good laboratory practice and analytical quality control are essential if satisfactory results are to be achieved.

Standing Committee of Analysts
The preparation of booklets within the series
“Methods for the Examination of Waters and
Associated Materials”

and their continuing revision is the responsibility of the
Standing Committee of Analysts. This committee was established in 1972 by the Department of the
Environment and is now managed by the Environment
Agency. At present, there are nine working groups, each responsible for one section or aspect of water quality analysis. They are
1 General principles of sampling and accuracy of results 2 Microbiological methods
3 Empirical and physical methods
4 Metals and metalloids
5 General non-metallic substances
6 Organic impurities
7 Biological methods
8 Biodegradability and inhibition methods
9 Radiochemical methods
The actual methods and reviews are produced by smaller panels of experts in the appropriate field, in cooperation with the working group and main committee.
The names of those members principally associated with these methods are listed at the back of the booklet. Publication of new or revised booklets will be notified to the technical press. If users wish to receive copies or advance notice of forthcoming publications, or obtain details of the index of methods then contact the
Secretary on the Agency’s internet web-site
(www.environment-agency.gov.uk/nls) or by post.
Every effort is made to avoid errors appearing in the published text. If, however, any are found, please notify the Secretary.
Dr D Westwood
Secretary
August 2006

__________________________________________________________________________
Warning to users
The analytical procedures described in this booklet should only be carried out under the proper supervision of competent, trained analysts in properly equipped laboratories.

noted. Numerous publications are available giving practical details on first aid and laboratory safety. These should be consulted and be readily accessible to all analysts. Amongst such publications are; “Safe
Practices in Chemical Laboratories” and “Hazards in the
Chemical Laboratory”, 1992, produced by the Royal
All possible safety precautions should be followed
Society of Chemistry; “Guidelines for Microbiological and appropriate regulatory requirements complied with. This should include compliance with the Health Safety”, 1986, Portland Press, Colchester, produced by
Member Societies of the Microbiological Consultative and Safety at Work etc Act 1974 and regulations made under this Act, and the Control of Substances Committee; and “Safety Precautions, Notes for
Guidance” produced by the Public Health Laboratory
Hazardous to Health Regulations 2002 (SI
2002/2677). Where particular or exceptional hazards Service. Another useful publication is “Good Laboratory exist in carrying out the procedures described in this Practice” produced by the Department of Health. booklet, then specific attention is

6

The determination of chemical oxygen demand in waters and effluents
1

Introduction

Methods for the determination of chemical oxygen demand have been previously published(1, 2). These methods have been revised and other procedures developed and are included in this booklet.
Previously published methods
1977 method
1986 method A
1986 method B (closed-tube)
1986 method C

Equivalent method in this booklet
A
B (closed-tube)
B (open-tube)
C
D
E

Initially, methods C, D and E of this booklet were developed in 1986 as alternative methods to the use of method A, which, over a prolonged period of time, utilises significant quantities of mercury. Methods C and D were used for samples with chemical oxygen demand (COD) values of up to 400 mg l-1. Method E was used where large numbers of samples were analysed and where it was considered appropriate to reduce the manipulative operations within the test, especially the dilution of samples. Consequently, method E was used where many samples with COD values of between 400 - 800 mg l-1 were frequently analysed.
It has now been established that varying, biased results have been obtained when the mercury-free methods, i.e. methods C and D of this booklet, were used to determine COD values of certain sewage effluents possessing COD values of up to about 160 mg l-1.
Consequently, to address this issue but at the same time facilitate the reduction of mercury disposed to the environment, a small scale mercury COD method was developed; this is method B of this booklet and is effectively a scaled down version of method A. Method A is the original (and still remaining) reference method. The results obtained using method A have, in the past(1, 2) been compared with the results obtained using methods C, D and E.
Comparative data are also shown in this and the following booklet. Whilst COD values have been determined using procedures similar to those described in method B, these usually have involved test-kits, with the determination of residual dichromate using spectrophotometry.
It is the laboratory’s responsibility to ensure that the validation of methods, used in comparative exercises, is properly carried out to ensure equivalency of results on the samples analysed is demonstrated for the types of samples being analysed.
As stated, the determination of COD is empirical, and experiences have shown that departure from the concentration of reagents used, temperatures and other details specified in both the digestion, and titrimetric or spectrophotometric stages described in methods A - E can lead to significant differences being reported in the results determined.
In addition, when the spectrophotometric determination is carried out, users should ensure that matrix effects (including turbidity) of individual samples are included in the validation of the test.
Where the expected COD is above 400 mg l-1 (but see individual methods for range of application) the sample should be diluted accordingly, see Table 1.

7

Table 1

Example dilutions for expected COD values
Expected COD
(mg l-1)
0 - 400
400 - 800
800 - 2000
2000 - 4000
2000 - 8000
10000 - 20000
10000 - 40000
20000 - 80000

Dilution none 2-fold
5-fold
10-fold
20-fold
50-fold
100-fold
200-fold

Whilst all types of sewage effluents, trade wastes and other polluted waters can be analysed for COD, the test can also be applied to solutions of single organic compounds and inorganic substances. However, the rate and extent of oxidation of organic compounds and inorganic substances varies considerably. Some compounds are easily oxidised under the conditions of the test, while others are less so. For example, many heterocyclic compounds, for example sulphonic acid derivatives, are only partially oxidised under the test conditions. Furthermore, inorganic reducing compounds (such as nitrites, sulphites and iron(II) salts) usually react faster with dichromate than organic compounds, generally, and so contribute to the oxygen demand. If these inorganic compounds are present in relatively high concentrations, the concentration of dichromate can be reduced rapidly. This effect may affect the oxidisability of any organic compounds present in the same sample. Similarly, while certain organic compounds may be completely oxidised when present in solution, alone and in low concentrations, the oxidation may be hindered or inhibited when the same organic compound is present with other compounds that are oxidised faster. The presence of oxidising agents in samples reduces the COD values determined. Interferences are known to occur with the oxidation, especially when chloride is present.
For instance, chlorine may be formed, which although an oxidising agent, may be lost by volatilisation. Similarly, with many organic compounds and their partially oxidised intermediates. In the presence of large amounts of chloride, the presence of ammonia and certain amines may lead to falsely high COD values being determined. Chlorine reacts with these types of nitrogen compounds to form chloramines. Chloramines are then decomposed to give chloride, which is then re-oxidised to form chlorine. If not suppressed, chloride will cause positive bias, the magnitude of which depends on the concentration of chloride and the COD value of the sample. Using standard solutions of potassium hydrogen phthalate, the interference effects decrease as the true COD values increase.
This is probably due to the reduction in the amount of residual dichromate in the refluxing mixture. In addition, the reaction temperature, which is dependent on many variables, including reaction mixture composition, apparatus design, the degree of immersion of the digest solution in the heater, etc, is important, as this and other factors affect the reaction rates, oxidation potential and volatilisation capability.
In methods A and B, mercury is used to suppress the effect of chloride. In methods C, D and E, silver is used as a catalyst, with chromium(III) and extra silver to suppress the effects of chloride. The effects of chloride may be reduced by treating the sample with silver nitrate solution, which precipitates silver chloride, which may then be removed from the sample mixture before the COD determination is carried out. Method C has been used in the past to determine the COD of samples with low values in the presence of
8

chloride, however, the use of methods A or B in this booklet may now be more appropriate, especially for the determination of COD values of final sewage effluents. It has also been shown that procedures involving closed tubes, as opposed to open-tube procedures, have resulted in the reporting of higher COD values. This is probably because closed tubes prevent the loss of volatile compounds, which may be lost when open-tube systems are used. Similar situations also arise when mixed reagents are employed instead of the addition of individual reagents. Falsely high COD values are obtained when pre-mixed reagents are added to the sample.
Whilst methods A and B can be adapted to a spectrophotometric end point determination, this is not usually the case with methods C, D and E. This is due to the high concentrations of chromium(III) and to the turbidity of resulting mixtures. In addition, when a spectrophotometric end point determination is employed, particular care needs to be taken to ensure a suitable solution is obtained which can be spectrophotometrically measured. Factors, such as turbidity, quality of (spectrophotometric) tubes, the matrix being analysed, etc are important considerations. Furthermore, extra work to investigate matrix and other interference effects due to the sample should be undertaken to ensure correct calibration is produced. There are several commercially available test-kits for determining COD values spectrophotometrically. Some of these test-kits involve mercury suppression, but others do not. Users should validate these test-kits with a range of sample types and standard solutions, taking into account known or suspected interferences. The robustness and reliability of the test, especially for methods B, D and E where only
2 ml or 2.5 ml of sample are analysed, is dependent on the representative nature of the sample and any dilutions made on the sample. The use of a 10-ml aliquot of the sample may be more appropriate where the particulate matter of a sample is high. Where 2-ml or
2.5-ml aliquots are used, it should be ensured that the volume is representative of the bulk of the sample. This is particularly important where samples of an in-homogeneous nature are taken for analysis. Attempts to homogenise such samples, especially those containing volatile compounds, may result in the loss of volatile material, or the formation of films, which may adhere to the sampling equipment.
COD tests are required for compliance purposes for a number of EU directives. For substances that are soluble in water, solutions of known concentrations can be examined by any of the procedures described. Whilst this is generally not possible for insoluble and sparingly soluble substances the procedures can be modified to carry out meaningful COD determinations on such materials.
2

Sample collection, preservation and storage

It is as essential to follow agreed sampling procedures, as it is to follow agreed analytical procedures. A sample for analysis should be as representative as possible of the bulk sample from which it is taken. For in-homogeneous material, the initial sample should be sufficient to enable a representative sample to be taken and that a representative analysis can be made. If sampling is undertaken for investigatory purposes, consideration should be given as to whether or not any anomalous material should be included or excluded, or sampled and analysed separately. In addition, any air space above the sample should be kept to a minimum. More specific information on sampling and sampling techniques can be found elsewhere in this series(3).
Care should be taken to ensure that sampling equipment does not react with or
9

contaminate the sample. Details of the procedures used should be recorded with the result for COD determinations on samples that have been filtered or allowed to settle. For example, the supernatant liquid (taken below any floating solids) or filtrate or centrifuged sample, with or without pH adjustment undertaken on the sample, may be used for the
COD determination.
The failure to store samples correctly, or to have them analysed as soon as possible may lead to erroneous results being determined. Samples should be stored in suitable containers.
Changes can be inhibited by refrigerating the sample at a temperature between 2 - 5 oC, or by reducing the pH to a value between 1 - 2 with sulphuric acid. The time for which preservation is effective should be established for each type of sample, as this may vary from a few hours to several days. Sample preservation procedures should be validated before being used routinely. The analyst should be aware of any preservative added before commencing the analysis.
In order to ensure that aliquots of the sample are representative, samples containing large amounts of suspended solid material should be homogenised before any portion is removed. Care should be taken to prevent the loss of volatile compounds, either by cooling the sample or by careful choice of the design of equipment used. If solid material is present, it should be separated from the sample and, if necessary, analysed separately.
Oily samples, especially emulsions, show a tendency to leave a surface film of grease on sample bottles and sampling equipment. Careful temperature control and the possibility of conditioning the glassware (by pre-rinsing) may sometimes avoid significant losses. If possible, oil should not be included in the sample and should be analysed separately, if required. For samples requiring dissolved COD, the sample should be filtered before preservation of the sample is carried out. In addition, precautions should be taken to avoid contamination from the filter paper or from the atmosphere, and the use of membrane glass-fibre filters may be more appropriate. If volatile compounds are known or suspected of being present consideration should be given as to whether the sample should be filtered and if so, suction should not be applied. Where necessary, the filtration rate should be increased by applying inert gas pressure to the sample. Allowing the sample sufficient time to settle may be useful, but should only be considered if microbial action does not take place. Where a sample does not settle, but which would settle if mixed with an inert material, filtration, dilution or pH-adjustment might be considered appropriate. Agreed procedures should, however, be followed and corrections made for any resultant volume changes.
Where samples are to be analysed by small-scale techniques, it is of paramount importance that a representative sample is taken for analysis. Homogenisation techniques may be used to pre-treat samples provided it is shown that the results produced are not unacceptably biased. Prolonged homogenisation has generally been shown to reduce the value of the COD determined.
3

Hazards

Poisonous gases may be evolved from samples which are acidified with sulphuric acid.
The acidification of samples containing, for example, cyanide or sulphide should always be carried out in a fume cupboard. In addition, dangers of spillage arise if the sample contains carbonate. Acidification should, therefore, be conducted in a fume cupboard. Addition of concentrated sulphuric acid to water should always be carried out with care and with
10

gentle swirling of the contents of the flask. The methods described involve the handling of boiling, concentrated solutions of sulphuric acid and potassium dichromate. Protective equipment is essential. In the event of spillage, immediate rinsing with copious amounts of clean water is generally the most effective remedy. Adsorbents are also available which facilitate removal of the bulk of the spillage. Inhalation and ingestion of dichromate dust should be avoided. Care is required when preparing and handling concentrated solutions containing silver or dichromate as these salts are toxic.
In closed-tube methods, the tubes are under pressure during and immediately after the heating stage. Care should, therefore, be exercised when handling, sealing or unseating tubes, the opening of which should always be directed away from staff. The closure, heating, removal from the heating block, and subsequent opening of closed tubes after cooling should be carried out behind a protective screen in a well ventilated hood. Care should be taken when opening the tubes so that any build-up of pressure is released gently and in a controlled manner.
Sealed tubes that contain only water should not be placed in heating blocks as excessive pressures can be generated. In these circumstances, the contents of such tubes may boil explosively if the tubes are opened, or break. The resulting steam will be super-heated.
The disposal of silver and mercury are notifiable wastes and should be recovered from used solutions and not discarded down the drain. Alternatively, residues should be sent to authorised dealers for proper disposal.
4

Principle

The COD test is an empirical test and is applicable to almost any aqueous sample as an index of pollution. The test is not subject to inhibition by toxic components that would be affected by tests that are dependent upon biochemical oxidation. Samples are oxidised in a defined manner by refluxing with sulphuric acid and potassium dichromate. Silver is added to catalyse the oxidation of certain compounds, such as alcohols and low molecular weight acids. Mercury is added to suppress the effects of chloride and ammonia. In the mercury-free methods, chromium(III) is added, as chromium(III) potassium sulphate, which together with excess silver suppresses chloride interference and the effects due to the presence of ammonia. The resulting mixture is refluxed for two hours and residual dichromate is determined by titration with standardised iron(II) ammonium sulphate solution or spectrophotometrically. The amount of dichromate reduced is expressed as
COD in the form of milligrams of oxygen consumed per litre of sample. Blank and analytical quality control determinations should always be undertaken with every batch of samples. 11

The reactions taking place during the oxidation can be expressed by the following series of equations. Oxidation reaction:

Cr2O72- + 8H+ → 2Cr3+ + 4H2O + 3O

Iron(II)/dichromate titration:

6Fe2+ + Cr2O72- + 14H+ → 6Fe3+ + 2Cr3+ + 7H2O

Chloride interference reactions:

Cr2O72- +6Cl- + 14H+ → 2Cr3+ + 3Cl2 + 7H2O :
Hg2+ + 4Cl- → [HgCl4]2- :
Cr3+ + 6Cl- → [CrCl6]3- :
Ag+ + Cl- → AgCl

5

Analytical quality control solutions

A variety of analytical quality control solutions may be used to challenge the oxidation procedure. Some of these solutions are more difficult to oxidise than others and may not achieve 100 % oxidation in the procedure. It may be that these solutions offer an alternative to the use of potassium hydrogen phthalate solution which may be more appropriate for use as a standard solution rather than as an AQC solution.
Analytical quality control solution
Sodium oxalate
Sodium acetate trihydrate
Succinic acid
Glucose

Concentration (mg l-1) equivalent to 1000 mg l-1 COD
8.376
2.126
1.054
0.938

A variety of AQC solutions have been tested with two commercially available closed-tube systems in the COD ranges 0 - 150 mg l-1 and 0 - 1500 mg l-1. Results are shown below and expressed as a percentage recovery of concentrations (in each range) of 50 mg l-1 and 500 mg l-1 respectively.

Sodium oxalate
Sodium acetate trihydrate
Succinic acid
Glucose
Data provided by Palintest

Hach
0 - 150 mg l-1
100
94
94
100

Hach
0 - 1500 mg l-1
100
98
99
100

12

Palintest
0 - 150 mg l-1
100
90
91
100

Palintest
0 - 1500 mg l-1
100
100
100
100

A

A reference large scale (10 ml) flask digestion procedure with mercury suppression A1
Performance characteristics of the method
_______________________________________________________________________
A1.1 Range of application
Up to a COD value of 400 mg l-1. The range can be extended by dilution of the sample with water.
A1.2 Standard deviation

See Table A1.

A1.3 Limit of detection

Typically, 5 - 15 mg l-1.

A1.4 Sensitivity

1 ml of 0.025 M iron(II) ammonium sulphate is equivalent to 20 mg l-1 COD.

A1.5 Bias

See Table A1.

A1.6 Interferences

See Table A2.

A1.7 Time required for analysis

Typically, the total analytical time for 1 - 10 samples is about 3 hours.
_______________________________________________________________________
A2

Reagents

Except where otherwise stated, analytical reagent grade chemicals should be used.
Reagents should be stored in glass bottles. All reagents, with the exception of iron(II) ammonium sulphate solution, may be stored at room temperature for up to one month.
Commercially available mixtures are obtainable for many of the reagents described.
Unacceptable blank values are usually caused by the oxygen demand of the water or sulphuric acid, or the use of dirty apparatus.
A2.1
Water. Water used for blank determinations and preparation of control standards should show negligible interference. Water with conductivity of less than
2 µS cm-1 and total organic carbon content of less than 1 mg l-1 has been shown to be satisfactory. Glassware used for the preparation and storage of water should be cleaned with chromic acid solution.
A2.2

Concentrated sulphuric acid (SG 1.84).

A2.3
Potassium dichromate solution (0.02083 M, i.e. M/48). To a 1000-ml volumetric flask, add 6.129 ± 0.001 g of potassium dichromate (previously dried for one hour at 140 - 150 oC). To the flask, add approximately 950 ml of water (A2.1) and mix the contents to dissolve. Make to 1000 ml with water, stopper and mix well.
A2.4
Silver sulphate in sulphuric acid (10 g l-1). To a glass bottle, add 10.0 ± 0.1 g of silver sulphate and 1000 ± 10 ml of sulphuric acid (A2.2) and stopper. To obtain a satisfactory solution, swirl the initial mixture and allow it to stand overnight. Swirl the contents again until all the silver sulphate dissolves. This solution may be stored in the dark at room temperature for up to an indefinite period.

13

A2.5
1:10 phenanthroline iron(II) indicator solution (“Ferroin” indicator). Dissolve
3.5 ± 0.1 g of iron(II) sulphate heptahydrate in 500 ± 1 ml of water (A2.1). Add 7.4 ± 0.1 g of 1:10 phenanthroline monohydrate, and mix the contents to dissolve. This reagent may be obtained commercially. During the titration using 0.025 M iron(II) ammonium sulphate solution
(A2.6) this indicator solution transfers a small amount of iron(II) to the titration flask. Thus, no more than 2 drops (i.e. 0.1 ml) of indicator solution should be used. Titrations should be made to the same colour end-point using equal amounts of indicator solution.
A2.6
Iron(II) ammonium sulphate solution (approximately 0.025 M). To a 1000-ml volumetric flask, add approximately 9.8 g of iron(II) ammonium sulphate hexahydrate to approximately 100 ml of water (A2.1). Add 20.0 ± 0.5 ml of sulphuric acid (A2.2) and swirl the mixture to dissolve the solid. Cool the solution and make to 1000 ml with water (A2.1).
Stopper and mix the contents. This solution is not stable and should be prepared freshly on the day of use, and standardised prior to use.
Standardise the iron(II) ammonium sulphate solution against 0.02083 M potassium dichromate solution (A2.3) using the following procedure. To approximately 60 ml of water
(A2.1) add 5.00 ± 0.05 ml of 0.02083 M potassium dichromate solution (A2.3). Carefully, add 15.0 ± 0.5 ml of sulphuric acid (A2.2) mix and cool the solution. Add no more than two drops of indicator solution (A2.5) and titrate to the end-point with the iron(II) ammonium sulphate solution to be standardised. Towards the end-point of the titration (see section
A4.3.1, note f) the addition of the iron(II) ammonium sulphate solution (A2.6) from a narrow-bore burette, in quantities of 0.01 - 0.05 ml, facilitates the detection of the endpoint. The titre should be approximately 25 ml.
The molarity, M, of the iron(II) ammonium sulphate solution (A2.6) is given by
M = 0.625 / V where V is the volume (ml) of iron(II) ammonium sulphate solution.
Alternatively, M = (0.020833 x 30) / V

or

M = 5 / (8 x V).

A2.7
Mixed reagent. To a 2-litre flat bottomed borosilicate flask, add 250 ± 10 ml of water (A2.1). Carefully, add 1.53 ± 0.01 g of potassium dichromate and 7.5 ± 0.5 g of silver sulphate. Swirl the contents of the flask to mix. Cautiously, with frequent swirling of the flask, add 750 ± 25 ml of sulphuric acid (A2.2). The contents of the flask should be well mixed in order to dissolve the silver sulphate completely. Allow the solution to stand overnight in the dark. Swirl the flask before use. This mixed reagent may be stored at room temperature for up to 3 months in a stoppered bottle in the dark, but rapidly deteriorates in daylight. Owing to a volume reduction on pre-mixing, 18.5 ml of this solution is equivalent to the addition of separate volumes of 5 ml of potassium dichromate solution (A2.3) and 15 ml of silver sulphate solution (A2.4).
A2.8
Mercury(II) sulphate solution (20 %). Cautiously, add 50 ± 2 ml of sulphuric acid (A2.2) to 450 ± 5 ml of water (A2.1). To this solution, add 100 ± 1 g mercury(II) sulphate. Stir the mixture until the solid dissolves.
A2.9
Analytical quality control solution. For example, the following AQC solution of potassium hydrogen phthalate gives a theoretical COD value of 400 mg l-1. To a 1000-ml volumetric flask, add 0.340 ± 0.001 g of potassium hydrogen phthalate (previously dried at
105 oC for 2 hours) to approximately 950 m of water (A2.1). Make to 1000 ml with water
14

(A2.1). This solution may be stored, without freezing, in a refrigerator for up to one month.
An alternative solution may however be more appropriate, see section 5.
A2.10

Chromic acid used for cleaning.

A3

Alternatively, 50 % nitric acid may be used.

Apparatus

High blank values may result from the presence of trace amounts of contaminants in the boiling flask, the reflux condenser or on the anti-bumping granules. Apparatus should be cleaned (on repeated occasions) by boiling with fresh dichromate/sulphuric acid/silver sulphate mixture until low and consistent blank values are obtained. Apparatus should be reserved solely for COD determinations. Glassware should have standard ground glass joints where appropriate and grease should not be used. When rinsed with water between use, the digestion apparatus should be drained and dried at 105 oC. The use of wet apparatus may cause loss of precision. To improve precision, volumetric glassware should be grade B or better.
A3.1
Boiling flask (150 ml). A distillation tray capable of holding all the contents of the boiling flask in the event of breakage during the digestion stage may also be useful.
A3.2

Pipettes and burette.

A3.3
Water-cooled condenser. For example, at least 150 mm, capable of being easily rinsed. A protective cap for the condenser helps to keep out dust when not in use.
For example a small beaker may be satisfactory.
A3.4
Anti-bumping granules. All anti-bumping granules should be pre-cleaned by using the same digestion procedure.
A3.5
Uniform heating. It is essential to maintain gentle boiling under reflux. Pointsources of heating are unsatisfactory. No part of the flask should be heated to a temperature in excess of the temperature of the refluxing liquid as decomposition of dichromate commences at temperatures only slightly above this. This will lead to high results being determined.
A4

Analytical procedure

_______________________________________________________________________
Step
Procedure
Notes
_______________________________________________________________________
A4.1

Determination of chloride

A4.1.1 Determine the chloride(4) content in the sample, see note a.
A4.2

(a) If the chloride content of the sample is greater than 500 mg l-1, continue as described in section A6.

Digestion

A4.2.1 See note b. Add several anti- bumping granules (A3.4) to the boiling flask (A3.1) and

15

(b) If the COD of the sample is known or expected to be above

add 10.0 ± 0.1 ml of sample or diluted sample, see note c.

400 mg l-1 the sample should be diluted with water. The dilution should be such that the titration of a
10-ml aliquot requires between approximately 5 - 20 ml of iron(II) ammonium sulphate solution (A2.6).
If the amount of chloride in a 10-ml aliquot of diluted sample is more than 5 mg then go to section A6.2.1.
(c) At least two blank solutions
(containing no sample) and an AQC sample (A2.9) should also be analysed.

A4.2.2 Add 1.0 ± 0.1 ml of mercury(II) sulphate solution (A2.8) and swirl the flask.
A4.2.3 To the flask, add 5.00 ± 0.01 ml of potassium dichromate solution (A2.3) followed by
15.0 ± 0.5 ml of silver sulphate-sulphuric acid solution (A2.4) see note d.

(d) Run the silver sulphate-sulphuric acid solution down the side of the flask whilst gently swirling and cooling the flask under running cold water. This procedure reduces loss of volatile compounds.

A4.2.4 Fit the condenser and gently boil under reflux for 120 ± 10 minutes (note e).

(e) Excessive reflux times will result in high blank values.

A4.2.5 Remove the flask and condenser from the source of heat and allow the contents to cool for approximately 10 minutes. Rinse the condenser with 25 ± 5 ml of water (A2.1) and allow the rinsings to collect in the flask.
Disconnect the flask from the condenser and cool the flask and contents to below 20 oC in running water.
A4.3

Determination of residual dichromate

(f) After the first addition of iron(II) ammonium sulphate solution, the solution is blue-green in colour.
During titration, the flask should be well mixed. The end-point of the titration occurs when the colour changes sharply from deep blue to pink. The blue colour may re-appear a few moments later but this should be ignored. This is especially common with samples high in chloride. See Figure A1.
_______________________________________________________________________
A4.3.1 Add no more than two drops (0.1 ml) of indicator solution (A2.5) to the flask and titrate the residual dichromate with iron(II) ammonium sulphate solution (A2.6) (note f).

16

A5

Calculation

The blank value should be the average of at least two determinations. If any blank value differs by more ± 0.5 ml from the average value it should be rejected. In these circumstances, it may be necessary to determine additional blank values.
An acceptable blank determination should require at least 23.5 ml of 0.025 M iron(II) ammonium sulphate solution (A2.6) in the titration. In addition, the difference between a refluxed blank value and an un-refluxed blank value should not exceed 1.5 ml of 0.025 M iron(II) ammonium sulphate solution (A2.6).
Standard chloride solutions may also need to be analysed to establish potential interference effects especially where samples are analysed and low COD values are obtained. The analysis of AQC samples, for example solutions of potassium hydrogen phthalate
(A2.9) or other appropriate solutions establishes that correct procedures are being followed. The COD of the sample is given by mg l-1

COD = 800 x DF x M (Vb - Vs) where Vb is the average volume (ml) of iron(II) ammonium sulphate solution (A2.6) used in the titration of blank solutions (A4.3.1);
Vs is the volume (ml) of iron(II) ammonium sulphate solution (A2.6) used in the titration of the sample (A4.3.1);
DF is the dilution factor, if appropriate (note b, section A4.2.1);
M is the molarity of standardised iron(II) ammonium sulphate solution (A2.6).
A6

Chloride interference

The following procedure is applicable to samples for which the 10-ml aliquot taken for analysis contains more than 5 mg of chloride and employs additional mercury(II) sulphate such that the ratio of mercury(II) sulphate to chloride is 40 to 1.
_______________________________________________________________________
Step
Procedure
Notes
_______________________________________________________________________
A6.1

Determination of chloride

A6.1.1 Following the determination of chloride in the sample, calculate the amount of chloride in
10 ml of sample, note g.

(g) If the amount is less than 5 mg continue as described in section
A4.2.

A6.2.1 Add several anti-bumping granules
(A3.4) to the boiling flask (A3.1) and add
10.0 ± 0.1 ml of sample or diluted sample, see

(h) Ensure that the blank and AQC solutions (A2.9) contain the same amount of mercury(II) sulphate as

17

used for the 10-ml aliquot of sample note c. To the flask, add solid mercury(II) or diluted sample. sulphate equal to forty times the mass of chloride in the 10-ml aliquot (note h). Swirl the flask vigorously for approximately 2 minutes. Continue as described in section A4.2.3.
_______________________________________________________________________
A7

Modified procedure using mixed reagents

For laboratories where very large numbers of samples, or where samples containing volatile compounds, are analysed for COD, the following procedure, using a mixed reagent, reduces the manipulative work per sample. However, the possibility that higher
COD values may be obtained should be recognised.
_______________________________________________________________________
Step
Procedure
Notes
_______________________________________________________________________
A7.1

Digestion

(i) The effects due to chloride and
A7.1.1 Add several anti-bumping granules any dilution should be taken into
(A3.4) into the boiling flask (A3.1). Add account. 10.0 ± 0.1 ml of sample or diluted sample
(note i). Add 1.0 ± 0.1 ml of mercury(II) sulphate solution (A2.8) and swirl the flask. Carefully, add
18.5 ± 0.2 ml of mixed reagent (A2.7) swirling the flask gently. Continue as described in
A4.2.4.
_______________________________________________________________________
Figure A1

Before digestion

Colour of solutions before and after digestion and titration

Following digestion

Immediately before titration end-point

18

Immediately following titration end-point

Table A1

Performance data
Sample type synthetic solution synthetic solution sewage effluent trade effluent synthetic solution synthetic solution synthetic solution synthetic solution synthetic solution synthetic solution effluent effluent synthetic solution synthetic solution synthetic solution synthetic solution distilled water synthetic solution effluent effluent synthetic solution synthetic solution synthetic solution synthetic solution synthetic solution synthetic solution

COD
(mg l-1)
64
400
64
312
250
250
250
250
80
400
80
400
400
350
64
400
290
398
371
250
250
210
240
252
200

St
(mg l-1)
1.7-6.6
1.4-6.2
2.3-5.2
3.5-9.8
3.3
2.4
3.3
3.7
1.7-6.6
1.4-6.2
2.3-5.2
3.5-9.8
4.2
7.0
5.1

RSD
(%)

Bias
(%)

DOF

1.3
1
1.3
1.5

-0.7
-0.6
-0.6
-0.9

28
23
29
29
5
5
4
4

-1.8
-2.2
-1.5

183
10
730
64
89
51
89
33
32
46
21
49
254

1.1
2.1
8

1.6
4.1
1.6
5.5
6.5
6.9
5.1
6.1
8.3
5.4

1.4
4
1.5
2.6
3.4
2.4
2.6
3.3
2.7

-2.3
0.2
0.5
-0.8
0.1
-1.2
-0.2
-1.4
-2.8

Synthetic solutions comprised standard solutions of potassium hydrogen phthalate in distilled water.
St is total standard deviation: RSD is relative standard deviation: DOF is degrees of freedom.

Data provided by various ex-water authority laboratories.
Table A2

Chloride interference using procedures described in section A4.2.2

COD
(mg l-1)
0*
100
200
300
400

Deviation on COD value (mg l-1)
Amount of chloride in 10 ml aliquot (mg)
5
10
15
20
10
1-4
2-10
4-12
4

23
2-8
1-28
6-10
4

41
2-16
1-30
8-14
4-20

*Measurement made by only one laboratory.

19

59
7-34
5-50
8-28
12-26

B

A small scale (2 ml) closed-tube digestion procedure with mercury suppression and spectrophotometric determination

In order to reduce the amount of mercury disposed to the environment, it may be more appropriate to use methods C, D or E for samples with COD values above 300 mg l-1.
B1
Performance characteristics of the method
________________________________________________________________________
B1.1
Ranges of application
Two ranges of up to 160 mg l-1 COD and
1600 mg l-1 COD. The range can be extended up to 3200 mg l-1 by dilution of the sample with water. B1.2

Standard deviation

See Table B1.

B1.3

Limit of detection

Typically, 10 mg l-1.

B1.4

Sensitivity

0.01 absorbance units is equivalent to 3 mg l-1
COD in the 160 mg l-1 range and 20 mg l-1 COD in the 1600 mg l-1 range.

B1.5

Bias

See Table B1.

B1.6

Interferences

Chloride interferes above 500 mg l-1, see
Table B2.

B1.7

Time required for analysis

B2

Reagents

Typical total analytical time for 1 - 20 samples is about 3 hours.
________________________________________________________________________

Except where otherwise stated, analytical reagent grade chemicals should be used.
Reagents should be stored in glass bottles. All reagents may be stored at room temperature for up to one month, except where specified. Commercially available mixtures are obtainable for many of the reagents described. Unacceptable blank values are usually caused by the COD of the water or sulphuric acid, or the use of dirty apparatus. B2.1
Water. Water used for blank determinations and preparation of control standards should show negligible interference. Water with conductivity of less than
2 µS cm-1 and total organic carbon content of less than 1 mg l-1 has been shown to be satisfactory. Glassware used for the preparation and storage of water should be cleaned with chromic acid solution.
B2.2

Concentrated sulphuric acid (SG 1.84).

B2.3
Potassium dichromate solution (0.00833 M, i.e. M/120) for 160 mg l-1 range. To a 1000-ml volumetric flask, add 2.451 ± 0.001 g of potassium dichromate (previously dried for one hour at 140 - 150 °C). To the flask, add approximately 950 ml of water (B2.1) and mix the contents to dissolve. Make to 1000 ml with water, stopper and mix well.

20

B2.4
Potassium dichromate solution (0.0833 M, i.e. M/12) for 1600 mg l-1 range. To a 1000-ml volumetric flask, add 24.51 ± 0.01 g of potassium dichromate (previously dried for one hour at 140 - 150 °C). To the flask, add approximately 950 ml of water (B2.1) and mix the contents to dissolve. Make to 1000 ml with water, stopper and mix well.
B2.5
Silver sulphate in sulphuric acid (10 g l-1). To a glass bottle, add 10.0 ± 0.1 g of silver sulphate and 1000 ± 10 ml of sulphuric acid (B2.2) and stopper. To obtain a satisfactory solution, swirl the initial mixture and allow it to stand overnight. Swirl the contents again until all the silver sulphate dissolves. This reagent may be stored in the dark at room temperature for up to an indefinite period.
B2.6
Mixed reagent for 160 mg l-1 range. To a 2-litre flat-bottomed borosilicate flask, add 250 ± 10 ml of water (B2.1). Carefully, add 0.612 ± 0.001 g of potassium dichromate and 7.5 ± 0.5 g of silver sulphate. Swirl the contents of the flask to mix. Cautiously, with frequent swirling, add 750 ± 25 ml of sulphuric acid (B2.2). The contents of the flask should be well swirled, in order to dissolve the silver sulphate completely. Allow the solution to stand overnight in the dark. Swirl the solution again before use. This mixed reagent may be stored in the dark in a glass bottle, with a polytetrafluoroethylene-lined or glass stopper at room temperature for up to 3 months, but rapidly deteriorates in daylight.
Owing to a volume reduction on pre-mixing, 3.7 ml of this solution is equivalent to the addition of separate volumes of 1 ml of potassium dichromate solution (B2.3) and 3 ml of silver sulphate solution (B2.5).
B2.7
Mixed reagent for 1600 mg l-1 range. To a 2-litre flat-bottomed borosilicate flask, add 250 ± 10 ml of water (B2.1). Carefully, add 6.12 ± 0.01 g of potassium dichromate and 7.5 ± 0.5 g of silver sulphate. Swirl the contents of the flask to mix.
Cautiously, with frequent swirling of the solution, add 750 ± 25 ml of sulphuric acid (B2.2).
The contents of the flask should be well swirled, in order to dissolve the silver sulphate completely. Allow the solution to stand overnight in the dark. Swirl the solution again before use. This mixed reagent may be stored in the dark in a glass bottle, with a polytetrafluoroethylene-lined or glass stopper at room temperature for up to 3 months, but rapidly deteriorates in daylight. Owing to a volume reduction on pre-mixing, 3.7 ml of this solution is equivalent to the addition of separate volumes of 1 ml of potassium dichromate solution (B2.4) and 3 ml of silver sulphate solution (B2.5).
B2.8
Mercury(II) sulphate solution (20 %). Cautiously, add 50 ± 2 ml of sulphuric acid (B2.2) to 450 ± 5 ml of water (B2.1). To this solution, add 100 ± 1 g mercury(II) sulphate. Stir the solution until the solid dissolves.
B2.9
Standard solution (theoretical equivalent to 4000 mg l-1 COD). To a 1000-ml volumetric flask, add 3.40 ± 0.01 g of potassium hydrogen phthalate (previously dried at
105 °C for 2 hours) to approximately 950 ml of water (B2.1). Make to 1000 ml with water
(B2.1). This solution may be stored, without freezing, in a refrigerator for up to one month.
B2.10 Standard solution (theoretical equivalent to 400 mg l-1 COD). To a
1000-ml volumetric flask, add 0.340 ± 0.001 g of potassium hydrogen phthalate (previously dried at 105 °C for 2 hours) to approximately 950 ml of water (B2.1). Make to 1000 ml with water (B2.1). This solution may be stored, without freezing, in a refrigerator for up to one month. An alternative solution may be more appropriate as an AQC sample, see section 5.
B2.11

Chromic acid used for cleaning.

Alternatively, 50 % nitric acid may be used.

21

B3

Apparatus

High blank values may result from the presence of trace amounts of contaminants in the boiling tube. Apparatus should be cleaned by boiling with fresh dichromate, sulphuric acid and silver sulphate mixture (on repeated occasions) until low and constant blank values are obtained. Apparatus should be reserved solely for COD determinations. Between use the digestion tubes should be rinsed with water, drained and dried at 105 °C. The use of wet apparatus may cause loss of precision. To improve precision volumetric glassware should be grade B or better.
B3.1
Pipettes and burette. Graduated pipettes capable of dispensing 1.00 ± 0.01 ml,
2.00 ± 0.02 ml, 0.20 ± 0.02 ml and up to 4.00 ± 0.05 ml, and burette graduated in 0.10 ml divisions. B3.2
Heating source. Thermostatically controlled heating block capable of accommodating the digestion tubes such that the level of the liquid in the tube is coincident with or below the surface of the block. The block should be controlled to give a digest temperature sufficient to gently reflux the oxidation mixture. A temperature of
150 ± 3 °C is suitable. Care should be taken to ensure that the temperature within the block does not rise above this temperature. An alternative heating source for the closedtube procedure is a suitably controlled air oven.
B3.3
Tubes for closed-tube digestion. Many types of tubes have been shown to be suitable. The main criteria are that the tubes can be heated, cooled, and removed from the heater without risk of bursting or spillage of contents. The test data in the tables were obtained using borosilicate-glass culture vials, 125 x 16 mm with plastic screw cap and polytetrafluoroethylene liner. Tubes with any noticeable defect should be discarded. Many commercial systems are available where the tube is used only once and then discarded or recycled according to documented procedures, or returned to the manufacturer.
B3.4
Spectrophotometer or colourimeter. Capable of operating at wavelengths of
420 - 450 nm, and 600 - 620 nm and with cell holder capable of accepting appropriate tubes or glass vials.
B4

Analytical procedure

The digestion procedure described in this method is based on a 1:5 scaled-down version of the flask procedure described in method A. Residual dichromate is determined spectrophotometrically. ________________________________________________________________________
Step
Procedure
Notes
________________________________________________________________________
B4.1
Determination of chloride
B4.1.1

Determine the chloride(4) content in the sample, see note a.

22

(a) If the chloride content of the sample is greater than 500 mg l-1, dilute accordingly so that 2 ml of diluted sample contains no more than 1 mg of chloride. See section
B6. Dilution of the sample to eliminate chloride interference may need to be balanced against reducing the expected
COD value to too low a value.

B4.2

Standard solutions
Depending on the expected COD value carry out the appropriate procedure described in B4.2.1 or B4.2.2.

B4.2.1

Into separate 100-ml volumetric flasks, add (to ± 0.01 ml) 0.0, 5.0, 10.0, 15.0,
20.0, 25.0, 30.0, 35.0 or 40.0 ml of potassium hydrogen phthalate solution
(B2.10). Make each flask to 100 ml with water (B2.1). These solutions are equivalent to 0, 20, 40, 60, 80, 100, 120,
140 and 160 mg l-1 COD respectively when used as described as below.

B4.2.2

Into separate 100-ml volumetric flasks, add (to ± 0.1 ml) 0.0, 5.0, 10.0, 15.0,
20.0, 25.0, 30.0, 35.0 or 40.0 ml of potassium hydrogen phthalate solution
(B2.9). Make each flask to 100 ml with water (B2.1). These solutions are equivalent to 0, 200, 400, 600, 800,
1000, 1200, 1400 and 1600 mg l-1 COD respectively when used as described as below. B4.2.3

Add 2.00 ± 0.02 ml of each standard solution (B4.2.1 or B4.2.2 as appropriate depending on range) to a boiling tube and carry out the closed-tube digestion procedure (B4.3) and determination of dichromate (B4.4).

B4.2.4

Prepare a calibration curve of absorbance (at the appropriate wavelength, see note g) against equivalent COD concentration.

B4.3

Closed-tube digestion

B4.3.1

See note b. Add 2.00 ± 0.02 ml of sample or diluted sample to the boiling tube (B3.3). See note c

(b) If the COD value of the sample is known or expected to be above the range identified in B4.2.1 or B4.2.2 the sample should be diluted with water
(B2.1).
(c) At least three reagent blanks containing 2.00 ± 0.01 ml water (B2.1)

23

in place of sample should also be analysed and an AQC sample (B2.10).
B4.3.2

Add 0.20 ± 0.02 ml of mercury(II) sulphate solution (B2.8) and swirl the tube to mix the contents.

B4.3.3

To the tube, add 1.00 ± 0.01 ml of potassium dichromate solution (B2.3 or
B2.4 depending on range) followed by
3.00 ± 0.05 ml of silver sulphatesulphuric acid solution (B2.5) see note d.

B4.3.4

Close the tube securely, swirl the tube to (e) Excessive reflux times will result in high blank values. mix the contents. and place in the heating source (B3.4). Reflux for
120 ± 10 minutes (note e). Remove the tubes and allow the contents to cool to below 20 °C for approximately 5 minutes under running water or allow to cool in a test tube rack.

B4.3.5

Wipe the outside of each tube, for example with a damp cloth, to remove all extraneous matter from the optical surface of the tube, see note f. Dry with a lint-free cloth.

B4.4

Determination of residual dichromate

B4.4.1

Using a spectrophotometer or colorimeter (B3.4) select the appropriate wavelength for the measurement (see note g).

(g) For the 160 mg l-1 range greater sensitivity is achieved by measuring reduction in yellow colour at
420 - 450 nm, and determining remaining Cr6+. For the 1600 mg l-1 range, the increase in blue colour at
600 - 620 nm determines the amount of
Cr3+ produced.

B 4.4.2

Zero the instrument with water, and measure the absorbance of the three blanks. Determine the average value.
(note h).

(h) Use any light cap or cover provided to exclude light during the measurement. B 4.4.3

(d) Run the silver sulphate - sulphuric acid solution down the side of the tube whilst gently swirling the contents and cooling the tube under running cold water. This procedure reduces loss of volatile compounds.

(f) This should eliminate interference to ensure a correct absorbance reading can be measured.

Insert the tube containing the sample or diluted sample into the cell holder and measure the absorbance (note h).
________________________________________________________________________

24

B5

Calculation

The blank tubes are most important, as all other tubes are compared with these. Three blanks are prepared and the average value absorbance determined. If any blank value differs by more ± 0.01 absorbance units from the average value it should be rejected. In these circumstances, it may be necessary to determine additional blank values.
Standard chloride solutions may also need to be analysed to establish potential interference effects especially where samples are analysed and low COD values are determined. The analysis of AQC samples, for example standard solutions of potassium hydrogen phthalate (B2.10) or other appropriate solutions establishes that correct procedures are being followed.
B6

Chloride interference

The following procedure is applicable to samples for which the 2-ml aliquot taken for analysis contains more than 1 mg of chloride and employs additional mercury(II) sulphate such that the ratio of mercury(II) sulphate to chloride is 40 to 1.
________________________________________________________________________
Step
Procedure
Notes
________________________________________________________________________
B6.1

Closed-tube digestion

(j) Ensure that the blank and AQC
B6.1.1 Add 2.00 ± 0.02 ml of sample or diluted solutions (B2.10) contain the same sample to the boiling tube (B3.3) see note c. To the tube, add solid mercury(II) amount of mercury(II) sulphate used for sulphate equal to forty times the mass of the 2-ml aliquot of sample or diluted chloride in the 2-ml aliquot (note j). Cap sample. the tube and shake vigorously for approximately 2 minutes. Continue as described in section B4.3.3.
________________________________________________________________________

B7

Modified procedure using mixed reagents

For laboratories where very large numbers of samples, or where samples containing volatile compounds, are analysed for COD, the following procedure, using a mixed reagent, reduces the manipulative work per sample. However, the possibility that higher COD values may be determined should be recognised.

25

________________________________________________________________________
Step
Procedure
Notes
________________________________________________________________________
B7.1

Closed-tube digestion

(k) The effects due to chloride and any
B7.1.2 Add 2.00 ± 0.02 ml of sample or diluted dilution should be taken into account. sample (see note k) into the boiling tube
(B3.3). Add 0.20 ± 0.02 ml of mercury(II) sulphate solution (B2.8) and swirl the tube. Carefully, add 3.70 ± 0.05 ml of mixed reagent (B2.6 or B2.7 as appropriate depending on the range).
Continue as described in B4.3.4.
________________________________________________________________________
B8

Open-tube procedure

Alternatively, an open-tube procedure can be used based on the procedures described in sections B4 - B7, but utilising a colourimetric determination. Reagents and apparatus required in addition to those given in sections B2 and B3 include the following.
B8.1
1:10 phenanthroline iron(II) indicator solution. (“Ferroin” indicator). Dissolve
3.5 ± 0.1 g of iron(II) sulphate heptahydrate in 500 ± 1 ml of water (B2.1). Add 7.4 ± 0.1 g of 1:10 phenanthroline monohydrate, and mix to dissolve. This reagent may be obtained commercially. During the titration using iron(II) ammonium sulphate solution (B8.2) this indicator solution transfers a small amount of iron(II) to the titration flask. Thus, no more than 2 drops (i.e. 0.1 ml) of indicator solution should be used. Titrations should be made to the same colour end point using equal amounts of indicator solution.
B8.2
Iron(II) ammonium sulphate solution (approximately 0.008 M). To a 1000-ml volumetric flask, add approximately 3.2 g of iron(II) ammonium sulphate hexahydrate to approximately 100 ml of water (B2.1). Add 20.0 ± 0.5 ml of sulphuric acid (B2.2) and swirl the mixture to dissolve the solid. Cool and make to 1000 ml with water (B2.1). Stopper and mix well. This solution is not stable and should be prepared freshly on the day of use, and standardised prior to use.
Standardise the iron(II) ammonium sulphate solution against 0.00833 M potassium dichromate solution (B2.3) using the following procedure. To approximately 60 ml of water
(B2.1) add 5.00 ± 0.05 ml of 0.00833 M potassium dichromate solution (B2.3). Carefully, add 15.0 ± 0.5 ml of sulphuric acid (B2.2) mix and cool the solution. Add no more than two drops of indicator solution (B8.1) and titrate to the end point with the iron(II) ammonium sulphate solution to be standardised. Towards the end point of the titration (see section
B9.3.1, note f) the addition of the iron(II) ammonium sulphate solution (B8.2) from a narrow-bore burette, in quantities of 0.01 - 0.05 ml, facilitates the detection of the end point. The expected titre should be approximately 30 ml.
The molarity, M, of the iron(II) ammonium sulphate solution (B8.2) is given by
M = 0.25 / V
26

where V is the volume (ml) of iron(II) ammonium sulphate solution.
Alternatively, M = (0.00833 x 30) / V

or

M = 30 / (120 x V).

1 ml of 0.00833 M iron(II) ammonium sulphate is equivalent to 32 mg l-1 COD.
B8.3
Tubes for open-tube digestion. Borosilicate glass tubes, for example 16 mm diameter with a 14/23 ground glass socket and a total volume of not less than 15 ml.
Tubes with any noticeable defect should be discarded.
B8.4

Air condenser.

For example, 13 mm diameter and not less than 150 mm long.

B8.5
Anti-bumping granules. All anti-bumping granules should be pre-cleaned by using the same digestion procedure.
B9

Analytical procedure

The digestion procedure described is based on a 1:5 scaled-down version of the flask procedure described in method A. Residual dichromate is determined colourimterically.
_______________________________________________________________________
Step
Procedure
Notes
_______________________________________________________________________
B9.1

Determination of chloride

B9.1.1 Determine the chloride(4) content in the sample, see note a.

B9.2

(a) If the chloride content of the sample is greater than 500 mg l-1, continue as described in section
B11.

Open-tube digestion

B9.2.1 See note b. Add several anti- bumping granules (B8.5) to the boiling tube (B8.3) and add 2.00 ± 0.01 ml of sample or diluted sample, see note c.

(b) If the COD of the sample is known or expected to be above
160 mg1-1 the sample should be diluted with water. The dilution should be such that the titration of a
2-ml aliquot requires between approximately 1.2 - 5 ml of iron(II) ammonium sulphate solution (B8.2).
If the amount of chloride in a 2-ml aliquot of sample or diluted sample is more than 1 mg, continue as described in section B11.
(c) At least two blank solutions
(containing no sample) and an AQC sample (B2.10) should also be analysed.

B9.2.2 Add 0.20 ± 0.02 ml of mercury(II) sulphate solution (B2.8) and swirl the tube to mix the contents.
27

B9.2.3 To the tube, add 1.000 ± 0.005 ml of potassium dichromate solution (B2.3) followed by
3.00 ± 0.05 ml of silver sulphate-sulphuric acid solution (B2.5) see note d. Swirl the tube to mix the contents.

(d) Run the silver sulphate-sulphuric acid solution down the side of the tube whilst gently swirling and cooling the tube under running cold water. This procedure reduces loss of volatile compounds.

B9.2.4 Fit the condenser to the tube and gently boil under reflux for 120 ± 10 minutes (note e).

(e) Excessive reflux times will result in high blank values.

B9.2.5 Remove the tube and condenser from the source of heat and allow the flask and contents to cool to below 20 oC in running water.
Remove the condenser from the tube, rinse the condenser with approximately 2 ml of water
(B2.1) and allow the rinsings to collect in a titration flask. Quantitatively transfer the contents of the tube to the titration flask, rinsing the tube as appropriate with a small amount of water.
B9.3

Determination of residual dichromate

(f) After the first addition of iron(II)ammonium sulphate solution, the solution is blue-green in colour.
During titration, the flask should be well mixed. The end point of the titration occurs when the colour changes sharply from deep blue to pink. The blue colour may reappear a few moments later but this should be ignored. This is especially common with samples high in chloride. See Figure A1.
_______________________________________________________________________
B9.3.1 Add no more than two drops (0.1 ml) of indicator solution (B8.1) to the flask and titrate the residual dichromate with iron(II) ammonium sulphate solution (B8.2) (note f).

B10

Calculation

The blank value should be the average of at least two determinations. If any blank value differs by more ± 0.1 ml from the average value it should be rejected. In these circumstances, it may be necessary to determine additional blank values. An acceptable blank determination should require at least 28.5 ml of 0.00833 M iron(II) ammonium sulphate solution (B8.2) in the titration. In addition, the difference between a refluxed blank value and an un-refluxed blank value should not exceed 1.5 ml of 0.00833 M iron(II) ammonium sulphate solution (B8.2).
Standard chloride solutions may also need to be analysed to establish potential interference effects especially where samples are analysed and low COD values are determined. 28

The analysis of AQC samples, for example standard solutions of potassium hydrogen phthalate (B2.10) or other appropriate solutions establishes that correct procedures are being followed. The COD of the sample is given by
COD = 4000 x DF x M (Vb - Vs)

mg l-1

where
Vb is the average volume (ml) of iron(II) ammonium sulphate solution (B8.2) used in the titration of blank solutions (B9.3.1);
Vs is the volume (ml) of iron(II) ammonium sulphate solution (B8.2) used in the titration of the sample (B9.3.1);
DF is the dilution factor, if appropriate (note b, section B9.2.1);
M is the molarity of standardised iron(II) ammonium sulphate solution (B8.2).
B11

Chloride interference

The following procedure is applicable to samples for which the 2-ml aliquot taken for analysis contains more than 1 mg of chloride and employs additional mercury(II) sulphate such that the ratio of mercury(II) sulphate to chloride is 40 to 1.
_______________________________________________________________________
Step
Procedure
Notes
_______________________________________________________________________
B11.1

Determination of chloride

B11.1.1 Following the determination of chloride in the sample, calculate the amount of chloride in
2 ml of sample, note g.
B11.2

(g) If the amount is less than 1 mg continue as described in section
B9.2.

Open-tube digestion

(h) Ensure that the blank and AQC
B11.2.1 Add several anti-bumping granules solutions (B2.10) contain the same
(B8.5) to the boiling tube (B8.3) and add amount of mercury(II) sulphate as
2.00 ± 0.01 ml of sample or diluted sample, see used for the 2-ml aliquot of sample note c. To the tube, add solid mercury(II) or diluted sample. sulphate equal to forty times the mass of chloride in the 2-ml aliquot (note h). Swirl the flask vigorously for approximately 2 minutes. Continue as described in section B9.2.3.
_______________________________________________________________________

B12

Modified procedure using mixed reagents

For laboratories where very large numbers of samples, or where samples containing volatile compounds, are analysed for COD, the following procedure, using a mixed reagent, reduces the manipulative work per sample. However, the possibility that higher COD values may be determined should be recognised.
29

______________________________________________________________________
Step
Procedure
Notes
_______________________________________________________________________
B12.1

Open-tube digestion

B12.1.1 Add several anti-bumping granules
(B8.5) into the boiling tube (B8.3) and add
2.00 ± 0.01 ml of sample or diluted sample (see note i). Add 0.20 ± 0.02 ml of mercury(II) sulphate solution (B2.8) and swirl the tube.
Carefully, add 3.70 ± 0.01 ml of mixed reagent
(B2.6) swirling the tube gently. Continue as described in B9.2.4.

(i) The effects due to chloride and any dilution should be taken into account. _______________________________________________________________________

Table B1

Comparison of COD results

Method
B (closed-tube)
B (closed-tube)
B (closed-tube)
B (closed-tube)
B (closed-tube)
B (open-tube)
D (closed-tube)
D (closed-tube)
A

Laboratory*
1
4
5
7
8
7
2
6
11

COD
(mean mg l-1)
108.1
122.4
134.2
131.3
123.2
122.0
124.4
120.0
125.9

NOR

Sw

10
10
10
6
10
10
2
4
10

5.2
3.9
4.2
1.7
6.7
1.7

Standard solution of potassium hydrogen phthalate (125 mg l-1 COD).
NOR is the number of results: Sw is within-batch standard deviation.

2.1
2.8

* See also the following booklet in this series.
Data provided in DOE funded report produced by Yorkshire LabServices.

30

Table B2

Comparison of COD results in the presence of chloride

See also Table 12B in the following booklet in this series.

Laboratory

COD
(mean mg l-1)

NOR

Sw

B (closed-tube)
B (closed-tube)
B (closed-tube)
B (closed-tube)
B (closed-tube)
B (closed-tube)
B (open-tube)
D (closed-tube)

1
4
5
6
7
8
7
2

109.7
115.6
138.5
126.5
126.7
123.9
122.7
155.1

9
10
10
8
6
9
9
8

1.5
1.6
4.2
4.2
3.5
1.1
15.7

D (closed-tube)
D (closed-tube)
A

5
6
11

126.6
128.7
124.2

2
3
10

3.6

Method

-1

Standard solution of potassium hydrogen phthalate (125 mg l COD) in the presence of 500 mg l-1 chloride
NOR is the number of results: Sw is within-batch standard deviation.

Data provided in DOE funded report produced by Yorkshire LabServices.

31

C

A mercury-free large scale (10 ml) flask digestion using chromium(III) potassium sulphate and silver nitrate solutions

C1
Performance characteristics of the method
_______________________________________________________________________
C1.1 Range of application
Up to 400 mg l-1 COD. The range can be extended by dilution of the sample with water.
C1.2 Standard deviation

See Tables C1 and C2. See also Table C3 for a comparison of results obtained using methods
A and C.

C1.3 Limit of detection

Typically, 5 mg l-1.

C1.4 Sensitivity

1 ml of 0.025 M iron(II) ammonium sulphate corresponds to 20 mg l-1 COD.

C1.5 Bias

See Tables C1 and C2.

C1.6 Interferences

Ammonium ion is not oxidised unless chloride is present in sufficient quantities. See Table C4.

C1.7 Time required for analysis

Typical total analytical time for 1 - 10 samples is about 3 hours.
_______________________________________________________________________
C2

Reagents

Except where otherwise stated, analytical reagent grade chemicals should be used.
Reagents should be stored in glass bottles. All reagents, with the exception of iron(II) ammonium sulphate solution, may be stored at room temperature for up to one month.
Commercially available mixtures are obtainable for many of the reagents described.
Unacceptable blank values are usually caused by the oxygen demand of the water or sulphuric acid, or the use of dirty apparatus.
C2.1
Water. Water used for blank determinations and preparation of control standards should show negligible interference. Water with conductivity of less than
2 µS cm-1 and total organic carbon content of less than 1 mg l-1 has been shown to be satisfactory. Glassware used for the preparation and storage of water should be cleaned with chromic acid solution.
C2.2

Concentrated sulphuric acid (SG 1.84).

C2.3
Potassium dichromate solution (0.02083 M, i.e. M/48). To a 1000-ml volumetric flask, add 6.129 ± 0.001 g of potassium dichromate (previously dried for one hour at 140 - 150 oC). To the flask, add approximately 950 ml of water (C2.1) and mix to dissolve. Make to 1000 ml with water, stopper and mix well.
C2.4
Silver sulphate in sulphuric acid (10 g l-1). To a glass bottle, add 10.0 ± 0.1 g of silver sulphate and 1000 ± 10 ml of sulphuric acid (C2.2) and stopper. To obtain a satisfactory solution, swirl the initial mixture and allow it to stand overnight. Swirl the

32

contents again until all the silver sulphate dissolves. This solution may be stored in the dark at room temperature for up to an indefinite period.
C2.5
1:10 phenanthroline iron(II) indicator solution (“Ferroin” indicator). Dissolve
3.5 ± 0.1 g of iron(II) sulphate heptahydrate in 500 ± 1 ml of water (C2.1). Add 7.4 ± 0.1 g of 1:10 phenanthroline monohydrate, and mix to dissolve. This reagent may be obtained commercially. During the titration using 0.025 M iron(II) ammonium sulphate solution
(C2.6) this indicator solution transfers a small amount of iron(II) to the titration flask. Thus, no more than 2 drops (i.e. 0.1 ml) of indicator solution should be used. Titrations should be made to the same colour end point using equal amounts of indicator solution.
C2.6
Iron(II) ammonium sulphate solution (approximately 0.025 M). To a 1000-ml volumetric flask, add approximately 9.8 g of iron(II) ammonium sulphate hexahydrate to approximately 100 ml of water (C2.1). Add 20.0 ± 0.5 ml of sulphuric acid (C2.2) and swirl the mixture to dissolve the solid. Cool and make to 1000 ml with water (C2.1). Stopper and mix well. This solution is not stable and should be prepared freshly on the day of use, and standardised prior to use.
Standardise the iron(II) ammonium sulphate solution against 0.02083 M potassium dichromate solution (C2.3) using the following procedure. To approximately 60 ml of water
(C2.1) add 5.00 ± 0.05 ml of 0.02083 M potassium dichromate solution (C2.3). Carefully, add 15.0 ± 0.5 ml of sulphuric acid (C2.2) mix and cool the solution. Add no more than two drops of indicator solution (C2.5) and titrate to the end point with the iron(II) ammonium sulphate solution to be standardised. Towards the end point of the titration (see section
C4.3.1, note g) the addition of the iron(II) ammonium sulphate solution (C2.6) from a narrow-bore burette, in quantities of 0.01 - 0.05 ml, facilitates the detection of the end point. The titre should be approximately 25 ml.
The molarity, M, of the iron(II) ammonium sulphate solution (C2.6) is given by
M = 0.625 / V where V is the volume (ml) of iron(II) ammonium sulphate solution.
Alternatively, M = (0.020833 x 30) / V

or

M = 5 / (8 x V).

C2.7
Mixed reagent. To a 2-litre flat bottomed borosilicate flask, add 250 ± 10 ml of water (C2.1). Carefully, add 1.53 ± 0.01 g of potassium dichromate and 7.5 ± 0.5 g of silver sulphate. Swirl the contents of the flask to mix. Cautiously, with frequent swirling of the flask, add 750 ± 25 ml of sulphuric acid (C2.2). The contents of the flask should be well swirled in order to dissolve the silver sulphate completely. Allow the solution to stand overnight in the dark. Swirl the contents again before use. This mixed reagent may be stored at room temperature for up to 3 months in a stoppered bottle in the dark, but rapidly deteriorates in daylight. Owing to a volume reduction on pre-mixing, 18.5 ml of this solution is equivalent to the addition of separate volumes of 5 ml of potassium dichromate solution (C2.3) and 15 ml of silver sulphate solution (C2.4).
C2.8
Chromium(III) potassium sulphate solution (250 g l-1). Dissolve 25.00 ± 0.01 g of chromium(III) potassium sulphate dodecahydrate (KCr(SO4)2.12H2O) in 100 ± 1 ml of hot water (C2.1) i.e. water above 50 oC. This solution is saturated at 30 oC. This solution may be stored at room temperature in a glass bottle with a polytetrafluoroethylene

33

stopper, for up to an indefinite period. When required for use, warm to 50 oC and stir before use, to re-dissolve any solids.
C2.9
Silver nitrate (1000 g l-1). Dissolve 100 ± 1 g of silver nitrate in approximately
80 ml of water (C2.1). Warm the solution if necessary. Cool and make to 100 ± 1 ml with water (C2.1). Do not store below 10 oC as this solution is saturated at 0 oC. Stored in the dark in a stoppered glass bottle, this solution should keep indefinitely.
C2.10 Combined mixed reagent. To 185.0 ± 0.2 ml of mixed reagent (C2.7) add
4.00 ± 0.05 ml of chromium(III) potassium sulphate solution (C2.8) and mix well before use. C2.11 Analytical quality control solution. For example, the following AQC solution of potassium hydrogen phthalate gives a theoretical COD value of 400 mg l-1. To a 1000-ml volumetric flask, add 0.340 ± 0.001 g of potassium hydrogen phthalate (previously dried at
105 oC for 2 hours) to approximately 950 m of water (C2.1). Make to 1000 ml with water
(C2.1). This solution may be stored, without freezing, in a refrigerator for up to one month.
An alternative solution may however be more appropriate, see section 5.
C2.10

Chromic acid used for cleaning.

C3

Alternatively, 50 % nitric acid may be used.

Apparatus

High blank values may result from the presence of trace amounts of contaminants in the boiling flask, the reflux condenser or on the anti-bumping granules. Apparatus should be cleaned (on repeated occasions) by boiling with fresh dichromate/sulphuric acid/silver sulphate mixture until low and consistent blank values are obtained. Apparatus should be reserved solely for COD determinations. Glassware should have standard ground glass joints where appropriate and grease should not be used. When rinsed with water between use, the digestion apparatus should be drained and dried at 105 oC. The use of wet apparatus may cause loss of precision. To improve precision, volumetric glassware should be grade B or better.
C3.1
Boiling flask (150 ml). A distillation tray capable of holding all the contents of the boiling flask in the event of breakage during the digestion stage may also be useful.
C3.2

Pipettes and burette.

C3.3
Water-cooled condenser. For example, at least 150 mm, capable of being easily rinsed. A protective cap for the condenser helps to keep out dust when not in use.
For example a small beaker may be satisfactory.
C3.4
Anti-bumping granules. All anti-bumping granules should be pre-cleaned by using the same digestion procedure.
C3.5
Uniform heating. It is essential to maintain gentle boiling under reflux. Pointsources of heating are unsatisfactory. No part of the flask should be heated to a temperature in excess of the temperature of the refluxing liquid as decomposition of dichromate commences at temperatures only slightly above this. This will lead to high results being determined.

34

C4
Analytical procedure
_______________________________________________________________________
Step
Procedure
Notes
_______________________________________________________________________
C4.1

Determination of chloride

A4.1.1 Determine the chloride(4) content in the sample, see note a.
C4.2

(a) If the chloride content of the sample is greater than 2000 mg l-1, continue as described in section C6.

Digestion

C4.2.1 See note b. Add several anti- bumping granules (C3.4) to the boiling flask (C3.1) and add 0.60 ± 0.02 ml of silver nitrate solution
(C2.9) and 10.0 ± 0.1 ml of sample or diluted sample, see note c. Swirl the flask to mix the contents. (b) If the COD of the sample is known or expected to be above
400 mg l-1 the sample should be diluted with water. The dilution should be such that the titration of a
10-ml aliquot requires between approximately 5 - 20 ml of iron(II) ammonium sulphate solution (C2.6).
If the amount of chloride in a 10-ml aliquot of sample or diluted sample is more than 20 mg, continue as described in section C6.2.
(c) At least two blank solutions
(containing no sample) and an AQC sample (C2.11) should also be analysed.

C4.2.2 Add 0.40 ± 0.05 ml of chromium(III) potassium sulphate solution (C2.8) and swirl the flask to mix the contents. To the flask, add
5.00 ± 0.03 ml of potassium dichromate solution
(C2.3) followed by 15.0 ± 0.5 ml of silver sulphate-sulphuric acid solution (C2.4) see note d. Swirl the flask to mix the contents.

(d) Run the silver sulphate-sulphuric acid solution down the side of the flask whilst gently swirling and cooling the flask under running cold water. This procedure reduces loss of volatile compounds.

C4.2.3 Fit the condenser and gently boil under reflux for 120 ± 10 minutes (note e).

(e) Excessive reflux times will result in high blank values.

C4.2.4 Remove the flask and condenser from the source of heat and allow the contents to cool for approximately 10 minutes. Rinse the condenser with 25 ± 5 ml of water (C2.1) and allow the rinsings to collect in the flask.
Disconnect the flask from the condenser and cool the flask to below 20 oC in running water, note f.

(f) The titrimetric end point fades unless the digest solution is titrated between 15 - 20 oC.

35

C4.3

Determination of residual dichromate

(g) After the first addition of iron(II) ammonium sulphate solution, the solution is blue-green in colour.
During titration of the contents, the flask should be well mixed. The end point of the titration occurs when the colour changes sharply from deep blue to pink. The blue colour may reappear a few moments later but this should be ignored. This is especially common with samples high in chloride.
_______________________________________________________________________
C4.3.1 Add no more than two drops (0.1 ml) of indicator solution (C2.5) to the flask and titrate the residual dichromate with iron(II) ammonium sulphate solution (C2.6) (note g).

C5

Calculation

The blank value should be the average of at least two determinations. If any blank value differs by more ± 0.5 ml from the average value it should be rejected. In these circumstances, it may be necessary to determine additional blank values.
An acceptable blank determination should require at least 23.5 ml of 0.025 M iron(II) ammonium sulphate solution (C2.6) in the titration. In addition, the difference between a refluxed blank value and an un-refluxed blank value should not exceed 1.5 ml of 0.025 M iron(II) ammonium sulphate solution.
Standard chloride solutions may also need to be analysed to establish potential interference effects especially where samples are analysed and low COD values are obtained. The analysis of AQC samples, for example standard solutions of potassium hydrogen phthalate (C2.11) or other appropriate solutions establishes that correct procedures are being followed.
The COD of the sample is given by mg l-1

COD = 800 x DF x M (Vb - Vs) where Vb is the average volume (ml) of iron(II) ammonium sulphate solution (C2.6) used in the titration of blank solutions(C4.3.1);
Vs is the volume (ml) of iron(II) ammonium sulphate solution (C2.6) used in the titration of the sample (C4.3.1);
DF is the dilution factor, if appropriate (note b, section C4.2.1);
M is the molarity of standardised iron(II) ammonium sulphate solution (C2.6).

36

C6

Chloride interference

The following procedure is applicable to samples for which the 10-ml aliquot taken for analysis contains more than 20 mg of chloride.
_______________________________________________________________________
Step
Procedure
Notes
_______________________________________________________________________
C6.1

Determination of chloride

C6.1.1 Following the determination of chloride in the sample, calculate the amount of chloride in
10 ml of sample, note g.
C6.2

(g) If the amount is less than 20 mg continue as described in section
C4.2.

Digestion

C6.2.1 Add several anti-bumping granules
(C3.4) to the boiling flask (C3.1) and add
1.20 ± 0.05 ml of silver nitrate solution (C2.9) and 10.0 ± 0.1 ml of sample or diluted sample
(note h). Swirl the flask vigorously for approximately 2 minutes.

(h) Ensure that the blank and AQC solutions contain the same amount of silver nitrate as used for the 10-ml aliquot of sample or diluted sample.

C6.2.2 Add 0.80 ± 0.05 ml of chromium(III) potassium sulphate solution (C2.8). Swirl the flask again to mix the contents. To the flask, add
5.00 ± 0.03 ml of potassium dichromate solution
(C2.3) followed by 15.0 ± 0.5 ml of silver sulphate-sulphuric acid solution (C2.4) see note d. Swirl the flask to mix the contents.
Continue as described in section C4.2.3, note i.

(i) If the colour change at the end point to the titration is now less easy to observe, use of a potentiometric titrator may be more appropriate.

_______________________________________________________________________

37

C7

Modified procedure using mixed reagents

For laboratories where very large numbers of samples, or where samples containing volatile compounds, are analysed for COD, the following procedure, using a mixed reagent, reduces the manipulative work per sample. However, the possibility that higher COD values may be determined should be recognised.
_______________________________________________________________________
Step
Procedure
Notes
_______________________________________________________________________
C7.1

Digestion

C7.1.1 Add several anti-bumping granules
(C3.4) into the boiling flask (C3.1). Add
0.60 ± 0.02 ml of silver nitrate solution (C2.9) and 10.0 ± 0.1 ml of sample or diluted sample
(note j). Swirl the flask and leave for approximately 2 minutes. Add 0.40 ± 0.05 ml of chromium(III) potassium sulphate solution (C2.8) and swirl the contents to mix. Carefully, add
18.50 ± 0.05 ml of mixed reagent (C2.7) swirling the flask gently. Continue as described in
C4.2.3.

(j) The effects due to chloride and dilution should be taken into account. C7.1.2 Alternatively, add several anti-bumping granules (C3.4) into the boiling flask (C3.1). Add
0.60 ± 0.02 ml of silver nitrate solution (C2.9) and 10.0 ± 0.1 ml of sample or diluted sample
(note j). Swirl the flask and leave for approximately 2 minutes. Carefully, add
18.90 ± 0.05 ml of combined mixed reagent (C2.10) swirling the flask gently. Continue as described in C4.2.3.
_______________________________________________________________________
Table C1

Performance data

COD
(mg l-1)

Sw
(range
(mg l-1)
0.1 to 3.3
0 to 3.4
1.4 to 3.3
1.4 - 5.0
0 to 7.2

0
100
200
300
400

mean)
(mg l-1)
0.9
1.5
2.4
3.0
3.1

Bias
(range
(mg l-1)
-1.6 to 1.6
-13.4 to 4.8
-12.2 to 0.6
-14.6 to 5.0
-23.4 to 0.6

Based on standard solutions of potassium hydrogen phthalate.
The average mean bias is approximately -2 %.
Sw is within-batch standard deviation.

38

mean)
(mg l-1)
0.9
-2.2
-3.9
-9.6
-8.5

Number of Batch laboratories size
5
6
6
6
7

5-10
5-10
5-10
4-10
2-10

Table C2

Relative standard deviations on samples

Sample type

Crude sewage

Industrial effluents

Sewage works effluents

Mean COD Relative
(mg l-1) standard deviation
(%)
47.6
3.6
48
0
206
0.7
269
2.9
346
1.7
388
1.9
480
3.1
611
1.5
617
2.2
656
0.1
679
1
9.8
9.2
71.5
1.4
74.4
3
95.4
9.0
146
1.4
187
1.4
216
1.1
311
0.5
420
2.5
567
2.1
1018
7.1
1041
1.2
2828
2.7
2840
0.5
36.7
8.2
39.0
4.6
41.9
4.8
47.6
3.6
50
1.8
55.3
1.5
121
1.4

39

Degrees of freedom
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4

Table C3

Comparison of COD results determined using methods A and C

Sample type
Industrial effluent

Crude sewage

Sewage works effluent

Method C
(mg l-1)
8
63
63
94
137
174
258
310
403
557
1003
1011
2779
2872
3323
207
256
338
368
467
593
605
623
660
665
864
34
35
36
42
45
45
50
51
121

Method A
(mg l-1)
6
55
60
92
128
162
308
309
387
547
988
982
2730
2903
3251
208
241
330
349
454
576
593
587
664
651
892
28
31
35
35
42
43
45
52
121

Whilst the agreement is good for some matrices others show significant differences, illustrating agreement might be sample dependent.

40

Table C4
Sample type

Chloride interference
COD
(mg l-1)

Phthalate standards 0
100
200
300
400
0
100
200
300
400
0
100
200
300
400
0
100
200
300
400
0
100
200
300
400
Industrial effluent 187*
567*
Ammonia
0

Chloride
(mg l-1)

Mean deviation due to chloride (mg l-1)

Degrees of freedom

500
500
500
500
500
1000
1000
1000
1000
1000
2000
2000
2000
2000
2000
5000
5000
5000
5000
5000
10000
10000
10000
10000
10000
3500
3480
5000

4.7
6.2
- 0.5
- 0.6
0.5
13.5
5.7
- 0.8
5.5
0.6
11.5
6.4
1.6
4.7
1.6
42.5
14.5
8.8
12.3
11.9
94.1
62.6
34.7
23.8
24.7
3.2
10
45.4

4
4
4
4
4
6
4
4
4
6
6
4
4
4
5
6
4
4
4
6
8
5
5
5
6
4
4
2

Ammonia present as 100 mg l-1 NH3.
*These samples initially contained negligible chloride, were analysed, spiked with chloride and then reanalysed.

41

D

A mercury-free small scale (2 ml) flask digestion procedure using chromium(III) potassium sulphate and silver nitrate solutions

D1
Performance characteristics of the methods
_______________________________________________________________________
D1.1 Range of application
Up to 400 mg l-1 COD. The range can be extended by pre-dilution of the sample with water. D1.2 Standard deviation

See Tables D1 - D4.

D1.3 Limit of detection

Open-tube method - typically 10 mg l-1
(17 degrees of freedom).
Closed-tube method - typically 9 mg l-1
(15 degrees of freedom).

D1.4 Sensitivity

1 ml of 0.025 M iron(II) ammonium sulphate solution corresponds to 100 mg l-1 COD.

D1.5 Bias

Open-tube digestion - see Tables D1 and D2.
Closed-tube digestion - see Tables D3 and D4.

D1.6 Interferences

See Tables D5 and D6.

D1.7 Time required for analysis

Typical total analytical time for from 1 - 36 samples is about 3.5 hours.
_______________________________________________________________________
D2

Reagents

Except where otherwise stated, analytical reagent grade chemicals should be used.
Reagents should be stored in glass bottles. All reagents, with the exception of iron(II) ammonium sulphate solution, may be stored at room temperature for up to one month.
Commercially available mixtures are obtainable for many of the reagents described.
Unacceptable blank values are usually caused by the oxygen demand of the water or sulphuric acid, or the use of dirty apparatus.
D2.1
Water. Water used for blank determinations and preparation of control standards should show negligible interference. Water with conductivity of less than
2 µS cm-1 and total organic carbon content of less than 1 mg l-1 has been shown to be satisfactory. Glassware used for the preparation and storage of water should be cleaned with chromic acid solution.
D2.2

Concentrated sulphuric acid (SG 1.84).

D2.3
Silver nitrate solution (1200 g l-1). Dissolve 120 ± 1 g of silver nitrate in approximately 80 ml of water (D2.1). Warm the solution if necessary. Cool and make to
100 ± 1 ml with water (D2.1). Do not cool below about 10 oC as this solution is saturated at 0 oC. Stored in the dark in a stoppered glass bottle, this solution may be stored at room temperature for up to an indefinite period.

42

D2.4
Chromium(III) potassium sulphate (250 g l-1). Dissolve 25.00 ± 0.01 g of chromium(III) potassium sulphate dodecahydrate (KCr(SO4)2.12H2O) in 100 ± 1 ml of hot water (D2.1) i.e. water above 50 oC. This solution is saturated at 30 oC. Stored in a glass bottle with a polytetrafluoroethylene stopper, this solution may be stored for up to an indefinite period. When required for use, warm the contents to 50 oC and stir before use, to re-dissolve any solids.
D2.5
Mixed reagent. Add 250 ± 10 ml of water (D2.1) into a 2 litre flat-bottomed borosilicate flask. Carefully, add 1 53 ± 0.01 g of potassium dichromate and 7.5 ± 0.5 g of silver sulphate. Swirl the contents of the flask to mix. Cautiously, with frequent swirling of the flask, add 750 ± 25 ml of sulphuric acid (D2.2). The contents of the flask should be well swirled in order to dissolve the silver sulphate completely. Allow the solution to stand overnight in the dark. Swirl the flask before use. This mixed reagent may be stored for up to 3 months in a stoppered bottle in the dark, but rapidly deteriorates in daylight. Because of a volume reduction on pre-mixing, 3.7 ml of this solution is equivalent to separate volumes of 1 ml of potassium dichromate solution (D2.8) and 3 ml of silver sulphate solution (D2.9).
D2.6
1:10 phenanthroline iron(II) indicator solution (“Ferroin” indicator). Dissolve
3.5 ± 0.1 g of iron(II) sulphate heptahydrate in 500 ± 1 ml of water (D2.1). Add 7.4 ± 0.1 g of 1:10 phenanthroline monohydrate, and mix to dissolve. This reagent may be obtained commercially. During the titration using iron(II) ammonium sulphate (D2.7) the use of this indicator solution transfers a small amount of iron(II) to the titration flask. Hence, not more than 2 drops (i.e. 0.1 ml) of indicator solution should be used. Titrations should be made to the same colour end point using equal amounts of indicator solution.
D2.7
Iron(II) ammonium sulphate solution (approximately 0.025 M). To a 1000-ml volumetric flask, add approximately 9.8 g of iron(II) ammonium sulphate hexahydrate to approximately 100 ml of water (D2.1). Add 20.0 ± 0.5 ml of sulphuric acid (D2.2) and swirl to dissolve the solid. Cool and make to 1000 ml with water (D2.1). Stopper and mix well.
This solution is not very stable and should be prepared freshly on the day of use, and standardised before use.
Standardise the iron(II) ammonium sulphate solution against 0.02083 M potassium dichromate solution (D2.8) using the following procedure. To approximately 60 ml of water
(D2.1) add 5.00 ± 0.05 ml of 0.02083 M potassium dichromate solution (D2.8). Carefully add 15.0 ± 0.5 ml of sulphuric acid (D2.2) and cool the solution. Add no more than two drops of indicator solution (D2.6) and titrate to the end point with the iron(II) ammonium sulphate solution to be standardised. Towards the end point of the titration, addition of the iron(II) ammonium sulphate solution (D2.7) from a narrow-bore burette, in quantities of
0.01 - 0.05 ml, facilitates the detection of the end point. The titre should be approximately
25 ml.
The molarity, M, of the iron(II) ammonium sulphate solution (C2.7) is given by:M = 0.625 / V where V is the volume (ml) of iron(II) ammonium sulphate solution (D2.7).
Alternatively, M = (0.020833 x 30) / V

or

43

M = 5 / (8 x V)

D2.8
Potassium dichromate solution (0.02083 M, ie M/48). To a 1000-ml volumetric flask, add 6.129 ± 0.001 g of potassium dichromate (previously dried for one hour at
140 - 150 oC) in approximately 800 ml of water (D2.1) in. Make to 1000 ml with water, stopper and mix well.
D2.9
Silver sulphate in sulphuric acid (10 g l-1). To a glass bottle, add 10.0 ± 0.1 g of silver sulphate in a 1000 ± 10 ml of sulphuric acid (D2.2) and stopper. To obtain a satisfactory solution, swirl the initial mixture and allow it to stand overnight, Swirl again until all the silver sulphate dissolves. Store in the dark at room temperature. This solution may be stored for up to an indefinite period.
D2.10 Analytical quality control solution.
For example, the following AQC solution of potassium hydrogen phthalate gives a theoretical COD value of 400 mg l-1. To a 1000-ml volumetric flask, add 0.340 ± 0.001 g of potassium hydrogen phthalate (previously dried at
105 oC for 2 hours) to approximately 950 ml of water (D2.1). Make to 1000 ml with water
(D2.1). This solution may be stored, without freezing, in a refrigerator for up to one month.
An alternative solution may however be more appropriate, see section 5.
D3

Apparatus

High blank values may result from the presence of trace amounts of contaminants in the digestion tube, the reflux condenser or on the anti-bumping granules. Apparatus should be cleaned (on repeated occasions) by boiling with fresh dichromate/sulphuric acid/silver sulphate mixture until low and consistent blank values are obtained. Apparatus should be reserved solely for COD determinations. Glassware should have standard ground glass joints where appropriate and grease should not be used. When rinsed with water between use, the digestion apparatus should be drained and dried at 105 oC. The use of wet apparatus may cause loss of precision. To improve precision, volumetric glassware should be grade B or better.
D3.1
Tubes for open-tube digestion. Borosilicate glass tubes, for example 16 mm diameter with a 14/23 ground glass socket and a total volume of not less than 15 ml.
Tubes with any apparent defects should be discarded.
D3.2

Air condenser.

For example, 13 mm diameter and not less than 150 mm long.

D3.3
Tubes for closed-tube digestion. Many types of tubes have been shown to be suitable. The main criteria are that the tubes can be heated, cooled, removed from the heater and opened without risk of bursting or spillage. The test data in the tables were obtained using borosilicate-glass culture vials, 125 x 16 mm with plastic screw-cap and polytetrafluoroethylene liner. Tubes with any apparent defect should be discarded. Many commercial systems are available where the tube is used once, and discarded or sent for disposal. D3.4
Heating source. Thermostatically controlled heating block capable of accommodating the digestion tubes such that the level of the liquid in the tubes is coincident with the surface of the block. The block should be controlled to give a digest temperature of 150 ± 3 oC. Care should be taken to ensure that the temperature within the block does not rise above 153 oC. An alternative heating source is a suitable air oven, controlled so as to give a digest temperature of 150 ± 3 oC.
D3.5

Pipettes and burette.

Graduated pipettes capable of dispensing
44

0.100 ± 0.005 ml and 0.20 ± 0.01 ml, and burette graduated in 0.02 ml divisions.
D3.6
Anti-bumping granules. All anti-bumping granules should be pre-cleaned by using the same digestion procedure.
D4

Analytical procedure

The digestion procedure described is based on a 1:5 scaled-down version of the flask procedure described in method C.
_______________________________________________________________________
Step
Procedure
Notes
_______________________________________________________________________
D4.1

Determination of chloride

D4.1.1 Determine the chloride(4) content in the sample, see note a.
D4.2

(a) If the chloride content of the sample is greater than 2000 mg l-1, continue as described in section D6.

Open-tube digestion

D4.2.1 See note b. Add several anti- bumping granules (B3.6) to the boiling tube (D3.1) and add 0.100 ± 0.002 ml of silver nitrate solution
(D2.3) add 2.00 ± 0.02 ml of sample or diluted sample, see note c. Swirl the tube and allow the tube and contents to stand for approximately
2 minutes.

(b) If the COD of the sample is known or expected to be above
160 mg l-1 the sample should be diluted with water. The dilution should be such that the titration of a
2-ml aliquot requires between approximately 5 - 20 ml of iron(II) ammonium sulphate solution (D2.7).
If the amount of chloride in a 2-ml aliquot of diluted sample is more than 4 mg, continue as described in section B6.2 or B6.3.
(c) At least two blank solutions
(containing no sample) and an AQC sample (D2.10) should also be analysed. D4.2.2 To the tube, add 0.100 ± 0.005 ml of chromium(III) potassium sulphate solution (D2.4) and swirl the tube to mix the contents.
D4.2.3 Add 1.00 ± 0.01 ml of potassium dichromate solution (D2.8) and 3.00 ± 0.05 ml of silver sulphate-sulphuric acid solution (D2.9) see note d. Swirl the tube to mix the contents.

(d) Run the silver sulphate-sulphuric acid solution down the side of the tube whilst gently swirling and cooling the tube under running cold water. This procedure reduces loss of volatile compounds.

D4.2.4 Fit the condenser to the tube and

(e) Excessive reflux times will result
45

gently boil under reflux for 120 ± 10 minutes
(note e).

in high blank values.

D4.2.5 Remove the tube and condenser from the source of heat and allow the contents to cool to below 20 oC in running water, note f. Remove the condenser from the tube, rinse the condenser with approximately 2 ml of water (D2.1) and allow the rinsings to collect in a titration flask.
Quantitatively transfer the contents of the tube to the titration flask, rinsing the tube as appropriate with a small amount of water. Continue as described in section D4.4.

(f) The titrimetric end point fades unless the digest solution is titrated between 15 - 20 oC.

D4.3

Closed-tube digestion

D4.3.1 See note b. Add several anti-bumping granules (D3.6) to the boiling tube (D3.3) and add 0.100 ± 0.005 ml of silver nitrate solution
(D2.3) and add 2.00 ± 0.01 ml of sample or diluted sample, see note c. Swirl the tube to mix the contents.
D4.3.2 To the tube, add 0.100 ± 0.005 ml of chromium(III) potassium sulphate solution (D2.4) and swirl the tube to mix the contents.
D4.3.3 Add 1.00 ± 0.01 ml of potassium dichromate solution (D2.8) and 3.00 ± 0.05 ml of silver sulphate-sulphuric acid solution (D2.9) see note d.
D4.3.4 Close the tube and ensure it is secure.
Swirl the tube to mix the contents. Place the tubes in the heating source and reflux for 120 ± 10 minutes (note e) Remove the tubes and allow the contents to cool for approximately 5 minutes under running water to below 20 0C, note f.
Cautiously unseal the digestion tube.
D4.3.5 Quantitatively transfer the contents of the tube to a titration flask, rinsing the tube as appropriate with a small amount of water.
D4.4

Determination of residual dichromate

D4.4.1 Add no more than two drops (0.1 ml) of indicator solution (D2.6) to the flask and titrate the residual dichromate with iron(II) ammonium sulphate solution (D2.7) (note g).

46

(g) After the first addition of iron(II)ammonium sulphate solution, the solution is blue-green in colour.
During titration, the flask should be well mixed. The end point of the titration occurs when the colour

changes sharply from deep blue to pink. The blue colour may reappear a few moments later but this should be ignored. This is especially common with samples high in chloride. _______________________________________________________________________
D5

Calculation

The blank value should be the average of at least two determinations. If any blank value differs by more ± 0.5 ml from the average value it should be rejected. In these circumstances, it may be necessary to determine additional blank values.
An acceptable blank determination should require at least 23.5 ml of 0.025 M iron(II) ammonium sulphate solution (D2.7) in the titration. In addition, the difference between a refluxed blank value and an un-refluxed blank value should not exceed 1.5 ml of 0.025 M iron(II) ammonium sulphate solution(D2.7).
Standard chloride solutions may also need to be analysed to establish potential interference effects especially where samples are analysed and low COD values are determined. The analysis of AQC samples, for example standard solutions of potassium hydrogen phthalate (D2.10) or other appropriate solutions establishes that correct procedures are being followed.
The COD of the sample is given by mg l-1

COD = 4000 x DF x M (Vb - Vs) where Vb is the average volume (ml) of iron(II) ammonium sulphate solution (D2.7) used in the titration of blank solutions (D4.4.1);
Vs is the volume (ml) of iron(II) ammonium sulphate solution (D2.7) used in the titration of the sample (D4.4.1);
DF is the dilution factor, if appropriate (note b, section D4.2.1);
M is the molarity of standardised iron(II) ammonium sulphate solution (D2.7).
D6

Chloride interference

The following procedure is applicable to samples for which the 2-ml aliquot taken for analysis contains more than 4 mg of chloride.
_______________________________________________________________________
Step
Procedure
Notes
_______________________________________________________________________
D6.1

Determination of chloride

D6.1.1 Following the determination of chloride
47

(g) If the amount is less than 4 mg

in the sample, calculate the amount of chloride in
2 ml of sample, note g.
D6.2

Open-tube digestion

D6.2.1 Add several anti-bumping granules
(D3.6) to the boiling tube (D3.1) and add
0.20 ± 0.01 ml of silver nitrate solution (D2.3) add 2.00 ± 0.02 ml of sample or diluted sample, see note c. Swirl the tube to mix the contents.
Allow the tube and contents to stand approximately 2 minutes. To the tube, add
0.20 ± 0.01 ml of chromium potassium sulphate
(D2.4) and swirl the tube to mix the contents.
Continue as described in section D4.2.3, note i.
D6.3

continue as described in section
D4.2 or D4.3.

(h) Ensure that the blank and AQC solutions (D2.10) contain the same amount of silver nitrate and chromium potassium sulphate as used for the 2-ml aliquot of sample or diluted sample.
(i) If the colour change at the end point to the titration is less easy to observe, the use of a potentiometric titrator may be more appropriate.

Closed-tube digestion

D6.3.1 Add several anti-bumping granules
(D3.6) to the boiling tube (D3.3) and add
0.20 ± 0.01 ml of silver nitrate solution (D2.3) and 2.00 ± 0.02 ml of sample or diluted sample.
Swirl the tube to mix the contents. Allow the tube and contents to stand approximately 2 minutes.
To the tube add 0.20 ± 0.01 ml of chromium(III) potassium sulphate (D2.4) and swirl the tube to mix the contents. Continue as described in section D4.3.3, note i.
_______________________________________________________________________
D7

Modified procedure using mixed reagents

For laboratories where very large numbers of samples, or where samples containing volatile compounds, are analysed for COD, the following procedure, using a mixed reagent, reduces the manipulative work per sample. However, the possibility that higher
COD values may be obtained should be recognised.
_______________________________________________________________________
Step
Procedure
Notes
_______________________________________________________________________
D7.1

Open-tube digestion

D7.1.1 Add several anti-bumping granules
(D3.6) into the boiling tube (D3.1) and add
0.100 ± 0.001 ml of silver nitrate solution (D2.3) and 2.00 ± 0.02 ml of sample or diluted sample
(see note i). Swirl the tube to mix the contents.
To the tube, add 0.100 ± 0.005 ml of chromium(III) potassium sulphate solution (D2.4) and swirl to mix the contents. Carefully, add
48

(i) The effects due to chloride and any dilution should be taken into account. 3.70 ± 0.02 ml of mixed reagent (D2.5) swirling the tube gently. Continue as described in D4.2.4.
D7.2

Closed-tube digestion

D7.2.1 Add several anti-bumping granules
(D3.6) into the boiling tube (D3.3) add
0.100 ± 0.005 ml of silver nitrate solution (D2.3) and 2.00 ± 0.02 ml of sample or diluted sample
(see note i). Swirl the tube to mix the contents.
To the tube, add 0.100 ± 0.005 ml of chromium(III) potassium sulphate solution (D2.4) and swirl to mix the contents. Carefully, add
3.70 ± 0.01 ml of mixed reagent (D2.5). Continue as described in D4.3.4.
_______________________________________________________________________
Table D1

Standard deviation and bias (open-tube method)

COD
(mg l-1)

Sw
Bias of the mean Number of
-1
(mg l )
(mg l-1)
(laboratories

Batch size 0
100
200
300
400

4.3
3.0
4.5
4.1
4.5

5
5
5
5
5

-2
-2
-2
-3
-4

1
1
1
1
1

Sw is within-batch standard deviation.
Based on standard solutions of potassium hydrogen phthalate.

Table D2

Relative standard deviation (open-tube method)

Sample type

COD
(mg l-1)

Relative standard deviation
(%)

Degrees of freedom

Industrial effluent
Industrial effluent
Sewage works effluent

142
293
44

4.6
5.2
2.6

4
4
4

Table D3

Standard deviation and (closed-tube method)

COD
(mg l-1)

Sw
(mg l-1)
Range
1.9 to 5
1.2 to 3.9
3.8 to 6.5
3.6
2.3 to 4.5

0
100
200
300
400

Number of laboratories Batch size Mean

Bias of the mean
(mg l-1)
Range
Mean

3.2
2.7
5.3
3.3
3.5

-5 to 12.5
3.2
-1.7 to 2
1.1
1 to 7
3.2
-4 to 1
-1.6
-14.6 to -8.0 -10.8

3
4
3
3
4

5
5
5
5
5

Sw is within-batch standard deviation.

49

Table D4

Relative standard deviations on samples (closed-tube method)

Sample type

COD
(mg l-1)

Relative standard deviation
(%)

Degrees of freedom

Crude sewage

38.0
66.5
169
208
209
903
39.3
44
95.4
142
184
188
212
272
279
293
60
62
79

8.0
2.1
2.2
1.5
1.8
1.8
7.1
5.9
9.0
3.2
3.6
5.4
1.8
0.8
1.4
1.8
8.5
6.8
3.7

4
4
4
4
4
4
4
4
4
4
4
4
5
5
5
4
4
4
4

Industrial effluent

Sewage works effluent

Table D5

Chloride interference (open-tube method)

Sample type
Phthalate standards

Industrial effluent
Ammonia

COD
Chloride
(mg l-1) (mg l-1)
0
100
200
300
400
0
100
200
300
400
0
100
200
300
400
164*
0

500
500
500
500
500
1000
1000
1000
1000
1000
2000
2000
2000
2000
2000
3500
5000

Mean deviation due to chloride
(mg l-1 COD)
5
4
4
3
-6
7
9
10
4
-3
9
16
9
3
2
10
97

Degrees of freedom 4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1

Ammonia present as 100 mg l-1 NH3
* This sample initially contained negligible chloride, was analysed, spiked with chloride and then re-analysed.

50

Table D6
Sample type

Chloride interference (closed-tube method)
COD Chloride
(mg l-1) (mg l-1)

Phthalate standards 0
100
200
300
400
0
100
200
300
400
0
100
200
300
400
0
100
200
300
400
0
100
200
300
400
0
100
200
300
400
Industrial effluents 168*
Ammonia
0

500
500
500
500
500
500
500
500
500
500
1000
1000
1000
1000
1000
2000
2000
2,000
2000
2000
5000
5000
5000
5000
5000
10000
10000
10000
10000
10000
3370
5000

Mean deviation due to chloride
(mg l-1 COD)

Degrees of freedom (in each of 2 laboratories) 5.6
6.6
10.5
5.5
4.4
13
14
13
12
-5
2.6
19.6
8
33
-5
8
19
16
12
-2
86
25
7
7
-5
86
92
81
12
19
-16.2
20

4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1

Ammonia present as 100 mg l-1 NH3
* This sample initially contained negligible chloride, was analysed, spiked with chloride and then re-analysed.

51

E

A mercury-free small scale (2.5 ml) flask digestion procedure using chromium(III) potassium sulphate and silver nitrate solutions

E1
Performance characteristics of the method
_______________________________________________________________________
E1.1 Range of application

Up to 800 mg l-1 COD. The range can be extended by dilution of the sample with water.

E1.2 Standard deviation

See Tables E2 and E3.

E1.3 Limit of detection

Typically, 10 mg l-1 (17 degrees of freedom).

E1.4 Sensitivity

1 ml of 0.025 M iron(II) ammonium sulphate solution is equivalent to 80 mg l-1 COD.

E1.5 Bias

See Tables E1 - E4.

E1.6 Interferences

See tables E5 - E7.

E1.7 Time required for analysis

Typical total analytical time for from 1 - 36 samples is about 3.5 hours.
_______________________________________________________________________
E2

Reagents

Except where otherwise stated, analytical reagent grade chemicals should be used.
Reagents should be stored in glass bottles. All reagents, with the exception of iron(II) ammonium sulphate solution, may be stored at room temperature for up to one month.
Commercially available mixtures are obtainable for many of the reagents described.
Unacceptable blank values are usually caused by the oxygen demand of the water or the sulphuric acid, and dirty apparatus.
E2.1
Water. Water used for blank determinations and preparation of control standards should show negligible interference. Water with conductivity of less than
2 µS cm-1 and total organic carbon content of less than 1 mg l-1 has been shown to be satisfactory. Glassware used for the preparation and storage of water should be cleaned with chromic acid solution.
E2.2

Concentrated sulphuric acid (SG 1.84).

E2.3
Silver nitrate solution (500 g l-1). Dissolve 50.0 ± 0.5 g of silver nitrate in approximately 80 ml of water (E2.1). Warm the solution if necessary. Cool and make to
100 ± 1 ml with water (E2.1). Do not cool below about 10 oC as this solution is saturated at 0 oC. Stored in the dark in a stoppered glass bottle, this solution may be stored at room temperature for up to an indefinite period.
E2.4
Chromium(III) potassium sulphate (250 g l-1). Dissolve 25.0 ± 0.1 g of chromium(III) potassium sulphate dodecahydrate (KCr(SO4)2.12H2O) in 100 ± 1 ml of hot water (E2.1) i.e. water above 50 oC. This solution is saturated at 30 oC. Stored in a glass bottle with a polytetrafluoroethylene stopper, this solution should keep indefinitely. When required for use, warm to 50 oC and stir before use, to re-dissolve any solids.
52

E2.5
Potassium dichromate solution (0.03473 M). To a 1000-ml volumetric flask, add 10.216 ± 0.001 g of potassium dichromate (previously dried for one hour at
140 - 150 oC) in approximately 800 ml of water (E2.1). To the flask, add, with external cooling and stirring, 167 ± 1 ml of sulphuric acid (E2.2). Cool and make to 1000 ml with water (E2.1). Mix thoroughly. This reagent may be stored at room temperature for up to 3 months in a stoppered bottle in the dark.
E2.6
Silver sulphate in sulphuric acid (50 g l-1). To a glass bottle, add 50.0 ± 0.1 g of silver sulphate and 1000 ± 10 ml of sulphuric acid (E2.2). To obtain a satisfactory solution, swirl the initial mixture and allow it to stand overnight. Swirl again until all the silver sulphate dissolves. Store in the dark at room temperature. This solution may be stored for up to an indefinite period.
E2.7
1:10 phenanthroline iron(II) indicator solution (“Ferroin” indicator). Dissolve
3.5 ± 0.1 g of iron(II) sulphate heptahydrate in 500 ± 1 ml of water (E2.1). Add 7.4 ± 0.1 g of 1:10 phenanthroline monohydrate, and mix to dissolve. This reagent can be obtained commercially. During the titration using iron(II) ammonium sulphate, the use of this indicator solution transfers a small amount of iron(II) to the titration flask. Hence, no more than 2 drops (i.e. 0.1 ml) of indicator solution should be used. Titrations should be made to the same colour end point using equal amounts of indicator solution.
E2.8
Iron(II) ammonium sulphate solution (approximately 0.025 M). To a 1000-ml volumetric flask, add approximately 9.8 g of iron(II) ammonium sulphate hexahydrate to approximately 100 ml of water (E2.1). Add 20.0 ± 0.5 ml of sulphuric acid (E2.2) and swirl to dissolve the solid. Cool and make to 1000 ml with water (E2.1). Stopper and mix well.
This solution is not stable and should be prepared freshly on the day of use, and standardised prior to use.
To standardise the iron(II) ammonium sulphate solution against 0.02083 M potassium dichromate solution (E2.9) add 5.00 ± 0.05 ml of 0.02083 M potassium dichromate solution
(E2.9) and dilute with water (E2.1) to approximately 60 ml. Carefully, add 15.0 ± 0.5 ml of sulphuric acid (E2.2) and cool. Add no more than two drops of indicator solution (E2.5) and titrate to the end point with the iron(II) ammonium sulphate solution to be standardised. Towards the end point of the titration, addition of the iron(II) ammonium sulphate solution (from a narrow-bore burette) in quantities of 0.01 - 0.05 ml, facilitates the detection of the end point. The titre normally expected is approximately 25 ml.
The molarity, M, of the iron(II) ammonium sulphate solution is given by:M = 0.625 / V where V is the volume (ml) of iron(II) ammonium sulphate solution.
Alternatively, M = (0.020833 x 30) / V

or

M = 5 / (8 x V)

E2.9
Potassium dichromate solution (0.02083 M, i.e. M/48). To a 1000-ml volumetric flask, add 6.129 ± 0.001 g of potassium dichromate (previously dried for one hour at 140 - 150 oC). To the flask add approximately 950 ml of water (E2.1) and mix to dissolve. Make to 1000 ml with water, stopper and mix well.
E2.10
Combined mixed reagent. To a 1000 ml volumetric flask, add 10.216 ± 0.001 g of potassium dichromate, 16.70 ± 0.01 g of chromium(III) potassium sulphate and
53

approximately 600 ml of water (E2.1). Cautiously, add 167 ± 1 ml of sulphuric acid (E2.2).
Cool and make to 1000 with water (E2.1).
E2.11
Analytical quality control solution.
For example, the following AQC solution of potassium hydrogen phthalate gives a theoretical COD value of 800 mg l-1. To a 1000-ml volumetric flask, add 0.680 ± 0.001 g of potassium hydrogen phthalate (previously dried at
105 oC for 2 hours) to approximately 950 m of water (E2.1). Make to 1000 ml with water
(E2.1). This solution may be stored, without freezing, in a refrigerator for up to one month.
An alternative solution may however be more appropriate, see section 5.
E3

Apparatus

High blank values may result from the presence of trace amounts of contaminants in the digestion tube, the reflux condenser or on the anti-bumping granules. Apparatus should be cleaned (on repeated occasions) by boiling with fresh dichromate/sulphuric acid/silver sulphate mixture until low and constant blank values are obtained. Apparatus should be reserved solely for COD determinations. Glassware should have standard ground glass joints where appropriate and grease should not be used. When rinsed with water between use, the digestion apparatus should be drained and dried at 105 oC. The use of wet apparatus causes loss of precision. To improved precision, volumetric glassware should be grade B or better.
E3.1
Tubes for open-tube digestion. Borosilicate glass tubes, for example 16 mm diameter with a 14/23 ground glass socket and a total volume of not less than 15 ml.
Tubes with any apparent defects should be discarded.
E3.2

Air condenser.

For example, 13 mm diameter and not less than 150 mm long.

E3.3
Heating source. Thermostatically controlled heating block capable of accommodating the digestion tubes such that the level of liquid in the tube is coincident with the surface of the block. The block should be controlled to give a digest temperature of 150 ± 3 oC. Care should be taken to ensure that the temperature within the block does not rise above 153 oC. An alternative heating source is a suitable air oven, controlled so as to give a digest temperature of 150 ± 3 oC.
E3.4
Pipettes and burette. Graduated pipettes capable of dispensing
0.100 ± 0.005 ml and 0.20 ± 0.01 ml, and burette graduated in 0.02 ml divisions.
E3.6
Anti-bumping granules. All anti-bumping granules should be pre-cleaned by using the same digestion procedure.
E4

Analytical procedure

The digestion procedure is based on the open-tube procedure in method D using a more concentrated solution of the oxidising reagent, potassium dichromate.
_______________________________________________________________________
Step
Procedure
Notes
_______________________________________________________________________
E4.1

Determination of chloride

E4.1.1 Determine the chloride(4) content in the
54

(a ) If the chloride content of the

sample is greater than 2000 mg l-1, continue as described in section E6.

sample, see note a.
E4.2

Open-tube digestion

E4.2.1 See note b. Add several anti-bumping granules (E3.6) to a digestion tube (E3.1) and add 0.100 ± 0.001 ml of silver nitrate solution
(E2.3) and 2.50 ± 0.02 ml of sample (note c).
Swirl the tube to mix the contents.

(b) If the COD of the sample is known or expected to be above
800 mg l-1 the sample should be diluted with water. The dilution should be such that the titration of a
2.5-ml aliquot requires between approximately 5 - 20 ml of iron(II) ammonium sulphate solution (E2.8).
If the amount of chloride in a 2.5-ml aliquot of diluted sample is more than 5 mg, continue as described in section E6.2.
(c) At least two blank solutions
(containing no sample) and an AQC sample (E2.11) should also be analysed. E4.2.2 To the tube, add 0.100 ± 0.001 ml of chromium(III) potassium sulphate solution
(E2.4). Mix well and stand for at least 2 minutes.
E4.2.3 Add 1.50 ± 0.01 ml of potassium dichromate solution (E2.5).
E4.2.4 To the tube, add 3.50 ± 0.01 ml of silver sulphate-sulphuric acid solution (E2.6) note d.
Swirl the tube to mix the contents. Allow any evolved gas to escape.

(d) Run the silver sulphate-sulphuric acid solution down the side of the tube whilst gently swirling and cooling the tube under running cold water. This procedure reduces loss of volatile compounds.

E4.2.5 Place the condenser on top of the tube.
Place the tube in the heating source and allow the solution to digest at 150 ± 3 oC for 120 ± 10 minutes, note e. Remove the tubes and allow the solution to cool for 5 minutes, then cool under running to below 20 oC (note f). Remove the condenser from the digestion tube. Rinse the condenser with 2.0 ± 0.1 ml of water (E2.1) and allow the rinsings to run into a titration flask.
Quantitatively transfer the contents of the tube into the flask using 10 ± 1 ml of water as appropriate. (e) Excessive reflux times will result in high blank values.

55

(f) The titrimetric end point fades unless tubes are kept at between
15 - 20 oC.

E4.3

Determination of residual dichromate

E4.3.1 Add no more than two drops of indicator solution (E2.7) to the flask and titrate the residual dichromate with standardised iron(II) ammonium sulphate (E2.8) note g).

(g) After the first addition of iron(II) ammonium sulphate solution, the solution is blue-green in colour.
During titration the flask should be well mixed. The end point of the titration occurs when the colour changes sharply from deep blue to pink. The blue colour may reappear a few moments later but this should be ignored. This is especially true for samples high in chloride.

_______________________________________________________________________
E5

Calculation

The blank value should be the average of at least two determinations. If any blank value differs by more ± 0.5 ml from the average value it should be rejected. In these circumstances, it may be necessary to determine additional blank values.
An acceptable blank determination should require at least 23.5 ml of 0.025 M iron(II) ammonium sulphate solution (E2.8) in the titration. In addition, the difference between a refluxed blank value and an un-refluxed blank value should not exceed 1.5 ml of 0.025 M iron(II) ammonium sulphate solution (E2.8).
Standard chloride solutions may also need to be analysed to establish potential interference effects especially where samples are analysed and low COD values are obtained. The analysis of AQC samples, for example standard solutions of potassium hydrogen phthalate (E2.11) establishes that correct procedures are being followed.
The COD of the sample is given by mg l-1

COD = 3200 x DF x M (Vb - Vs) where Vb is the average volume (ml) of iron(II) ammonium sulphate solution (E2.8 used in the titration of blank solutions (E4.3.1);
Vs is the volume (ml) of iron(II) ammonium sulphate solution (E2.8) used in the titration of the sample (E4.3.1);
DF is the dilution factor, if appropriate (note b, section E4.2.1);
M is the molarity of standardised iron(II) ammonium sulphate solution (E2.8).
E6

Chloride interference

The following procedure is applicable to samples for which the 2.5-ml aliquot taken for analysis contains more than 5 mg of chloride.

56

_______________________________________________________________________
Step
Procedure
Notes
_______________________________________________________________________
E6.1

Determination of chloride

E6.1.1 Following the determination of chloride in the sample, calculate the amount of chloride in
2 ml of sample, note h.
E6.2

(h) If the amount is less than 5 mg continue as described in section
E4.2.

Open-tube digestion

(i) Ensure that the blank and AQC
E6.2.1 Add several anti-bumping granules solutions contain the same amount
(E3.6) to the boiling tube (E3.1) and add of silver nitrate and chromium(III)
0.20 ± 0.01 ml of silver nitrate solution (E2.3) potassium sulphate as used for the and 2.50 ± 0.02 ml of sample or diluted sample,
2.5-ml aliquot of sample or diluted see note c. Swirl the tube to mix the contents. sample. Allow the tube and contents to stand approximately 2 minutes. To the tube add
(j) If the colour change at the end
0.20 ± 0.01 ml of chromium(III) potassium point to the titration is less easy to sulphate (E2.4) note i. Swirl the tube to mix the observe, the use of a potentiometric contents. Continue as described in section titrator may be more appropriate.
E4.2.3, note j.
_______________________________________________________________________
E7

Modified procedure using mixed reagents

For laboratories where very large numbers of samples or where samples containing volatile compounds are analysed for COD, the following procedure, using a mixed reagent, reduces the manipulative work per sample. However, the possibility that higher COD values may be obtained should be recognised.
_______________________________________________________________________
Step
Procedure
Notes
_______________________________________________________________________
E7.1

Open-tube digestion

(k) The effects due to chloride and
E7.1.1 Add several anti-bumping granules any dilution should be taken into
(E3.6) into the boiling tube (E3.1) and add account. 0.100 ± 0.001 ml of silver nitrate solution (E2.3) and 2.50 ± 0.02 ml of sample or diluted sample
(see note k). Swirl the tube to mix the contents.
Carefully, add 1.50 ± 0.01 ml of combined mixed reagent (E2.10) swirling the tube gently.
Continue as described in E4.2.4.
_______________________________________________________________________

57

Table E1

Recovery tests

Substance

Concentration
(mg l-1)

Glutamic acid
Sodium benzoate
Pyridine
3-Picoline
Nicotinic acid
Acetic acid
Methanol
Sodium dodecylsulphate
Benzene
Toluene
Oxalic Acid

500
250
200
200
200
508
300
250
176
173
1530

Theoretical
COD
(mg l-1)
490
416
445
481
286
542
450
499
541
542
280

Actual
COD
(mg l-1)

Recovery
(%)

505
416