The rise of chlorine as a disinfectant for water treatment has made the measurement of residual free chlorine a high priority. Until now, such measurements required one sensor for measuring free chlorine and another to correct the reading due to changes in pH. New research findings, however, purport to eliminate the need for a separate pH sensor for compensation, thus reducing maintenance and calibration costs by up to 50 percent and eliminating the need to stock two separate sensors as spares.

Analytical chemistry of free chlorine has been developed in the last century based on colorimetry and electrochemistry.[1] Colorimetric methods rely on the product of reactions between a reagent and the chlorine species, which is only suitable for grab sample bench analysis. The electrochemical method relies on the redox characteristics of the chlorine species themselves. The chlorine species are reduced at the cathode of an electrochemical cell, producing a current that is proportional to the concentration of the species. The technology is called amperometric sensing and has been developed into sensor technology for online applications.[2]

Free chlorine is defined as species of chlorine gas dissolved into water, which can be described in the following reactions:

Cl2 + H2O → HClO (1)
HClO ⇔ H+ + ClO- (2)

The total of hypochlorous (HClO) acid and the hypochlorite ion (ClO-) are considered as free chlorine. The acid dissociation constant pKa for hypochlorous acid at room temperature is 7.54.[3] Based on the pKa value and equation (2), the pH dependence of the ratio of hypochlorous acid and hypochlorite ion
in free chlorine can be plotted as shown in Figure 1.

pH Dependence of the Ratio of HCIO and CIO- in Free Chlorine

All free chlorine is in the form of hypochlorite ion at pH above 10, as shown in Figure 1. At pH below 5, free chlorine is in the form of hypochlorous acid. The form of free chlorine is highly dependent on the pH value.

Electrode Process of Free Chlorine
The electrode process of free chlorine is a reducing process as illustrated in Figure 2. When hypochlorous acid and hypochlorite ion reach the surface of a negatively charged electrode (cathode), they accept electrons from the electrode and get reduced into chloride ion, which generate a sensing current in an amperometric free chlorine sensor. [2]

Since hypochlorous acid is a neutral molecule, and hypochlorite ion is negatively charged, the accessibilities of hypochlorite ion to the negatively charged electrode is much lower than that of the hypochlorous acid. An amperometric sensor is much more sensitive to hypochlorous acid than to hypochlorite ion. Therefore, the amperometric-free chlorine sensor exhibits a strong pH dependence due to the pH dependence of free chlorine species shown in Figure 1.

Figure 2. Electrode Process of Free Chlorine at the Electrode

Figure 3 shows the pH dependence of an amperometric sensor output at a constant free chlorine concentration. The sensing current decreases significantly with the increase of pH as shown in Figure 3. The electrode process described above is responsible for this behavior.

Diffusion Through Porous Membranes
Tests have shown that the electrode process is not the only phenomenon that results in pH dependence of an amperometeric free chlorine sensor. The surface property of the porous sensing membrane also affects the pH dependence of the sensor output. Figure 4 shows the test results comparing sensors with hydrophobic and hydrophilic sensing mebranes, respectively.

A hydrophobic sensing membrane makes the sensor more pH dependent, while a hydrophilic membrane makes the sensor less pH dependent, as shown in Figure 4.

In the design of an amperometric sensor, a sensing membrane is placed in front of the sensing electrode to hold the fill solution (supporting electrolyte of the electrochemical cell), while allowing the analytes in the sample solution to diffuse through. The results shown in Figure 4 indicate that the hydrophobic membrane tends to let the neutral hypochlorous acid diffuse through, but blocks the hypochlorite ions. Hydrophilic membranes tend to allow both species to diffuse through, as illustrated in Figure 5.

Realization of Ph Independence[4]

Figure 3. pH Dependence of an Amperometric Sensor at Constant Free Chlorine Concentration

It is then clear that in order to eliminate pH dependence, a hydrophilic sensing membrane must be used to allow all free chlorine species to diffuse through. This can be easily achieved because porous hydrophilic membranes are readily available due to their wide application in bioengineering.

The second phenomenon that controls the pH dependence is the electrode process. It is difficult to modify the electrode process. However, the pH value inside the sensor can be controlled at a value below pH 5, and all free chlorine will be in the form of hypochlorous acid, according to Figure 1. In this case, the electrode process will not see the difference in the pH value of the sample solution. This concept was quickly proven by using an acetic pH 4 buffer solution as the sensor fill solution. However, since the sensing membrane is hydrophilic, the pH 4 fill solution was lost in the sample solution through diffusion within a week’s operation. This is not acceptable for an online sensor.

The final challenge then falls on the reduction of the diffusion flux of a pH stabilizer through the sensing membrane. The diffusion flux can be described by using Fick’s first law:

(3)

where f is the diffusion flux; D is the diffusion rate of the pH stabilizer through the sensing membrane; A/d is the area/thickness of the sensing membrane; and C is the concentration of the pH stabilizer. Since D, A, and d are factors related to the sensitivity of the sensor, the only factor that can be changed in equation (3) is C, the concentration.

To increase the life of the fill solution with the pH stabilizer, it is desirable to add as much pH stabilizer into the fill solution as possible. However, the more pH stabilizer that is added in the fill solution, the higher its concentration. According to equation (3), the diffusion flux is proportional to C, which means that the more pH stabilizer added into the fill solution, the faster it will diffuse through the sensing membrane and become lost. Increasing the concentration of the pH stabilizer in the fill solution will not increase the operation life.

Figure 4. Comparison of pH Dependence of Sensors with Hydrophilic and Hydrophobic Sensing Membrane

The final question is how can more pH stabilizer be added into the fill solution while maintaining a low concentration. The answer is to use a weak acid with low solubility in water. By using such an acid, it is possible to maintain a low concentration regardless of the quantity of acid added. The low concentration is controlled by the acid’s low solubility in water.

One suitable acid is adipic acid (hexanedioic acid). The solubility of adipic acid in water is only 0.1 M. Its pKa is about 4.4. Thus, a saturated adipic acid solution will have a pH at about 2.8, which will convert all free chlorine into the form of hypochlorous acid (Figure 1). Figure 6 shows the pH dependence of an amperometric free chlorine sensor by using such a pH stabilizer in the fill solution in comparison with conventional amperometric free chlorine sensors. Tests have shown that the sensor will last for at least three months without changing the fill solution.

A

B

Figure 5. Amperometric Sensors with (A) a Hydrophobic Sensing Membrane and (B) a Hydrophilic Sensing Membrane

Analysis of the free chlorine chemistry and the sensing mechanism of an amperometric free chlorine sensor shows that the pH dependence of the sensor is caused by the reduction of hypochlorous acid at the electrode and the diffusion process of free chlorine through the sensing membrane. It is possible to employ a pH independent free chlorine sensor by using a hydrophilic sensing membrane and controlling the fill solution pH with a pH stabilizer of low solubility.

This technology provides reliable, stable readings of free chlorine while reducing the costs associated with traditional methods.

About the Authors

Figure 6. Comparison of pH Dependence of Sensors with Conventional Fill Solution and Fill Solutions with pH Stabilizer

Chang-Dong Feng, Ph.D., is manager of sensor technology for Emerson Process Management, Rosemount Analytical, Liquid Division. Dr. Feng has contributed more than 20 papers to industry publications on various topics and spoken at well-known conferences such as ISA, IFPAC, and AWWA. Dr. Feng has been awarded four U.S. patents and received two ISA-GTF awards. He is a member of the American Chemical Society and the Electrochemical Society. Dr. Feng can be reached at CD.Feng@Emerson Process.com. Joshua Xu, Ph.D., is a research scientist working on sensor development for Rosemount”s Liquid Division. He previously worked as a research associate in protein mass spectrometry at the Indiana University in Bloomington, Ind. Dr. Xu has had more than 10 papers published in chemistry journals and publications and is a member of the American Chemical Society and of the Electrochemical Society. He can be reached at
Joshua.Xu@Emerson Process.com.

For More Information: www.emersonprocess.com


This article is based on a paper that received the third place Gilmer/Thomason/Fowler (GTF) Award at the ISA 50th Analysis Division Symposium. The award is presented annually to authors of juried papers presented at the symposium.


References
1. Jonas, O., “Critical Overview of Power Station Sampling and Analytical of Water and Steam,” Power Plant Instrumentation for Measurement of High-Purity Water Quality, ASTM STP 742, R.W. Lane and Gerard Otten, Eds., American Society for Testing and Materials, 1981, p11-23.
2. J. Xu and C.D. Feng, “Improved Free Chlorine Amperometric Sensor: Proceedings of the 49th ISA Analysis Division Symposium,” April 2004, p67- 73.
3. J.C. Morris, “The Acid Ionization Constant of HOCl From 5 to 35 ºC,” The Journal of Physical Chemistry, 70(12), 1966, p3798 – 3805.
4. C.D. Feng, J. Xu, and J. Garrestson, “Free Chlorine Sensor,” U.S. Patent Pending.