* Approved by Standard Methods Committee, 1993.


4500-O G. Membrane Electrode Method


1. General Discussion


    Various modifications of the iodometric method have been developed to eliminate or minimize effects of interferences; nevertheless, the method still is inapplicable to a variety of industrial and domestic wastewaters.1 Moreover, the iodometric method is not suited for field testing and cannot be adapted easily for continuous monitoring or for DO determinations in situ.

    Polarographic methods using the dropping mercury electrode or the rotating platinum electrode have not been reliable always for the DO analysis in domestic and industrial wastewaters because impurities in the test solution can cause electrode poisoning or other interferences.2,3 With membrane-covered electrode systems these problems are minimized, because the sensing element is protected by an oxygen-permeable plastic membrane that serves as a diffusion barrier against impurities.4-6 Under steady-state conditions the current is directly proportional to the DO concentration.*

    Membrane electrodes of the polargraphic4 as well as the galvanic5 type have been used for DO measurements in lakes and reservoirs,8 for stream survey and control of industrial effluents,9,10 for continuous monitoring of DO in activated sludge units,11 and for estuarine and oceanographic studies.12 Being completely submersible, membrane electrodes are suited for analysis in situ. Their portability and ease of operation and maintenance make them particularly convenient for field applications. In laboratory investigations, membrane electrodes have been used for continuous DO analysis in bacterial cultures, including the BOD test.5,13

    Membrane electrodes provide an excellent method for DO analysis in polluted waters, highly colored waters, and strong waste effluents. They are recommended for use especially under conditions that are unfavorable for use of the iodometric method, or when that test and its modifications are subject to serious errors caused by interferences.

    a. Principle: Oxygen-sensitive membrane electrodes of the polarographic or galvanic type are composed of two solid metal electrodes in contact with supporting electrolyte separated from the test solution by a selective membrane. The basic difference between the galvanic and the polarographic systems is that in the former the electrode reaction is spontaneous (similar to that in a fuel cell), while in the latter an external source of applied voltage is needed to polarize the indicator electrode. Polyethylene and fluorocarbon membranes are used commonly because they are permeable to molecular oxygen and are relatively rugged.

    Membrane electrodes are commercially available in some variety. In all these instruments the "diffusion current" is linearly proportional to the concentration of molecular oxygen. The current can be converted easily to concentration units (e.g., milligrams per liter) by a number of calibration procedures.

    Membrane electrodes exhibit a relatively high temperature coefficient largely due to changes in the membrane permeability.6 The effect of temperature on the electrode sensitivity, Φ (microamperes per milligram per liter), can be expressed by the following simplified relationship:6


                                            log Φ = 0.43 mt + b



     t = temperature, °C,

    m = constant that depends on the membrane material, and

    b = constant that largely depends on membrane thickness.


If values of Φ and m are determined for one temperature (Φ0 and t0), it is possible to calculate the sensitivity at any desired temperature (Φ and t) as follows:


                                        log Φ = log Φ0 + 0.43 m (t - t0)


Nomographic charts for temperature correction can be constructed easily7 and are available from some manufacturers. An example is shown in Figure 4500-O:2; in which, for simplicity, sensitivity is plotted versus temperature on semilogarithmic coordinates. Check one or two points frequently to confirm original calibration. If calibration changes, the new calibration should be parallel to the original, provided that the same membrane material is used.

    Temperature compensation also can be made automatically by using thermistors in the electrode circuit.4 However, thermistors may not compensate fully over a wide temperature range. For certain applications where high accuracy is required, use calibrated nomographic charts to correct for temperature effect.

    To use the DO membrane electrode in estuarine waters or in wastewaters with varying ionic strength, correct for effect of salting-out on electrode sensitivy.6,7 This effect is particularly significant for large changes in salt content. Electrode sensitivity varies with salt concentration according to the following relationship:


                                            log ΦS = 0.43 mSCS + log Φ0




ΦS, Φ0 = sensitivities in salt solution and distilled water, respectively,

    C  = salt concentration (preferably ionic strength), and

    mS  = constant (salting-out coefficient).


If Φ0 and mS are determined, it is possible to calculate sensitivity for any value of CS. Conductivity measurements can be used to approximate salt concentration (CS). This is particularly applicable to estuarine waters. Figure 4500-O:3 shows calibration curves for sensitivity of varying salt solutions at different temperatures.

    b. Interference: Plastic films used with membrane electrode systems are permeable to a variety of gases bedsides oxygen, although none is depolarized easily at the indicator electrode. Prolonged use of membrane electrodes in waters containing such gases as hydrogen sulfide (H2S) tends to lower cell sensitivity. Eliminate this interference by frequently changing and calibrating the membrane electrode.

    c. Sampling: Because membrane electrodes offer the advantage of analysis in situ they eliminate errors caused by sample handling and storage. If sampling is required, use the same precautions suggested for the iodometric method.






*Fundamentally, the current is directly proportional to the activity of molecular oxygen.7



2. Apparatus


    Oxygen-sensitive membrane electrode, polarographic or galvanic, with appropriate meter.


3. Procedure


    a. Calibration: Follow manufacturer’s calibration procedure exactly to obtain guaranteed precision and accuracy. Generally, calibrate membrane electrodes by reading against air or a sample of known DO concentration (determined by iodometric method) as well as in a sample with zero DO. (Add excess sodium sulfide, Na2SO3, and a trace of cobalt chloride, CoCl2, to bring DO to zero.) Preferably calibrate with samples of water under test. Avoid an iodometric calibration where interfering substances are suspected. The following illustrate the recommended procedures:

    1) Fresh water – For unpolluted samples where interfering substances are absent, calibrate in the test solution or distilled water, whichever is more convenient.

    2) Salt water – Calibrate directly with samples of seawater or waters having a constant salt concentration in excess of 1000 mg/L.

    3) Fresh water containing pollutants or interfering substances – Calibrate with distilled water because erroneous results occur with the sample.

    4) Salt water containing pollutants or interfering substances – Calibrate with sample of clean water containing the same salt content as the sample. Add a concentrated potassium chloride (KCl) solution (see Conductivity, Section 2510 and Table 2510:I) to distilled water to produce the same specific conductance as that in the sample. For polluted ocean waters, calibrate with a sample of unpolluted seawater.

    5) Estuary water containing varying quantities of salt – Calibrate with a sample of uncontaminated seawater or distilled or tap water. Determine sample chloride or salt concentration and revise calibration to account for change of oxygen solubility in the estuary water.7

    b. Sample measurement: Follow all precautions recommended by manufacturer to insure acceptable results. Take care in changing membrane to avoid contamination of sensing element and also trapping of minute air bubbles under the membrane, which can lead to lowered response and high residual current. Provide sufficient sample flow across membrane surface to overcome erratic response (see Figure 4500-O:4 for a typical example of the effect of stirring).

    c. Validation of temperature effect: Check frequently one or two points to verify temperature correction data.





4. Precision and Bias


With most commercially available membrane electrode systems an accuracy of ± 0.1 mg DO/L and a precision of ± 0.05 mg DO/L can be obtained.





5. References

  1. McKeown, J.J., L.C. Brown & G.W. Gove. 1967. Comparative studies of dissolved oxygen analysis methods. J. Water Pollut. Control Fed. 39:1323.

  2. Lynn, W. R. & D. A. Okun. 1955. Experience with solid platinum electrodes in the determination of dissolved oxygen. Sewage Ind. Wastes 27:4.

  3. Mancy, K.H. & D. A. Okun. 1960. Automatic recording of dissolved oxygen in aqueous systems containing surface active agents. Anal. Chem. 32:108.

  4. Carritt, D.E. & J.W. Kanwisher. 1959. An electrode system for measuring dissolved oxygen. Anal. Chem. 31:5.

  5. Mancy, K.H. & W.C. Westgarth. 1962. A galvanic cell oxygen analyzer. J. Water Pollut. Control Fed. 34:1037.

  6. Mancy, K.H., D. A. Okun & C.N. Reilley. 1962. A galvanic cell oxygen analyzer. J. Electroanal. Chem. 4:65.

  7. Mancy, K.H. & T. Jaffe. 1966. Analysis of Dissolved Oxygen in Natural and Waste Waters. Publ. No. 999-WP-37, U.S. Public Health Serv., Washington, D.C.

  8. Weiss, C.M. & R. T. Oglesby. 1963. Instrumentation for monitoring water quality in reservoirs. American Water Works Assoc. 83rd Annual Conf., New York, N.Y.

  9. Cleary, E.J. 1962. Introducing the ORSANCO robot monitor. Proc. Water Quality Meas. Instrum. Publ. N. 108, U.S. Public Health Serv., Washington, D.C.

  10. MacKereth, F.J.H. 1964. An improved galvanic cell for determination of oxygen concentrations in fluids. J. Sci. Instrum. 41:38.

  11. Sulzer, F. & W.M. Westgarth. 1962. Continuous D.O. recording in activated sludge. Water Sewage Works 109:376.

  12. Duxbury, A.C. 1963. Calibration and use of a galvanic type oxygen electrode in field work. Limnol. Oceanogr. 8:483.

  13. Lipner, H.J., L.R. Witherspoon & V.C. Champeaus. 1964. Adaptation of a galvanic cell for microanalysis of oxygen. Anal. Chem. 36:204.


©Standard Methods for the Examination of Water and Wastewater. 20th Ed. American Public Health Association, American Water Works Association, Water Environment Federation.