A Study on the Recovery from Iron Poisoning Caused by Cation Exchange Resins

Abstract: Cation exchange resins used in water softeners are susceptible to iron ion contamination. Using a uniform design approach, regression analysis, and optimization techniques, the effects of each component in the resuscitator on the cation resuscitator effect were investigated, resulting in the optimal resuscitator formulation and the development of a novel resuscitator method. This method features low corrosivity and good resuscitator performance, providing a valuable reference for the resuscitator of resins produced on-site.

Keywords: Cation exchange resin; Uniform design; Regression analysis; Optimization

CLC Number: TU425.23 Document Code: A Article Number: 1009-2455(2003)04-0016-04

A Study on Resuscitation of lron- Poisoned Cation Resin

JIA Bo, ZHOU Bai-qing，LI Qin
(college of power and Mechanical Engineering,Wuhan University,Wuhan 430072,china)

Abstract: The cation resin used in softeners is easily contaminated by iron ion.The effects of different components in detoxicants on the result of resuscitation of the cation have been studied using the method of uniform design，regression analysis and optimization technology，based on which an optimum formula for resuscitation has been obtained and a new method for resuscitation has been developed.This new method features lower corrosion trend and good result of resuscitation，providing useful reference for the resuscitation of the resin on the production sites.
Keywords: cation resin; uniform design; regression analysis; optimization

Currently, cation exchange resins used in industrial boilers to reduce feedwater hardness commonly suffer from "iron poisoning." The process of removing iron from the resin is called "recovery" or "detoxification." In actual production, a common recovery method is soaking the resin in high-concentration hydrochloric acid. However, this method has a significant problem: the corrosion of the softener by HCl. This paper presents an experimental study on this issue, employing a reduction recovery method. Based on uniform design and regression analysis, a detailed study was conducted on the influencing factors in the recovery agent formulation. Optimization experiments yielded good results.

1. Experimental Section

1.1 Preparation of Resins with Different Degrees of Poisoning

In actual production, the required recovery conditions vary depending on the degree of iron ion contamination of the resin. To simulate this situation, the paper prepared resins with different degrees of poisoning using the following methods:

① Preparation of "all-iron type (R3Fe)," "all-calcium type (R2Ca)," and "all-magnesium type (R2Mg)" resins: The cation exchange resins were completely transformed using excess FeCl3, CaCl2, and MgSO4 solutions, respectively, to become resins with a single exchange group.

② Preparation of Iron-Poisoning Resin: A degraded iron-poisoning resin was prepared using the three resins mentioned above in a specific ratio. In this resin, the molar fraction X1 of R3Fe varied depending on the degree of poisoning, while the molar fractions of R2Ca and R2Mg were 1 - X1, and the molar ratio of the exchange groups of R2Ca and R2Mg was fixed at 3:1.

③ Preparation of the Degraded Resin Layer: The three resins were mixed thoroughly and directly loaded into the exchange column for later use. For ease of description, X1 is referred to as the poisoning rate.

1.2 Working Exchange Capacity Loss Rate of Resins with Different Poisoning Degrees

Regeneration and operational tests were conducted on exchange columns using resins with different poisoning degrees under the following conditions:

① Regeneration conditions: Co-current regeneration; regenerator: 5% NaCl; regeneration flow rate: 4 m/h; regenerator consumption: 150 g/mol; water temperature: 15℃.

② Operational conditions: Forward wash flow rate: 12 m/h; influent water hardness: 5.29 mmol/L; operating flow rate: 12 m/h; water temperature: 15℃; operational failure endpoint: hardness 40 μmol/L.

The test results obtained under the above conditions are shown in Table 1.

Table 1 Working Exchange Capacity Loss Rate Test

Note: Y₁ represents the cycle water production; Y₂ represents the working exchange capacity.

In Table 1, the working exchange capacity loss rate (η) is defined as the percentage reduction in working exchange capacity of contaminated resin (i.e., X1 > 0%) relative to the working exchange capacity of uncontaminated resin (i.e., X1 = 0%). For ease of description, the working exchange capacity of the resin will be referred to as "working exchange" in the following sections. Considering that artificially prepared resin layers require several transitional operating cycles to reach stable operating conditions, the average water production rate h of the 3rd and 4th operating cycles is used as the basis for evaluating the impact of iron poisoning on resin performance.

As shown in Table 1, with the increase of iron poisoning, i.e., with the increase of X1, the working exchange capacity of the resin continuously decreases. When X1 is 50%, the working exchange loss rate (η) reaches as high as 55.68%. Simultaneously, it was observed in the experiment that the color of the entire resin layer gradually darkens with the increase of X1.

2. Resin Resuscitation Test

2.1 Hydrochloric Acid Resuscitation Method

According to the characteristics of the resuscitation process, the resin was resuscitated with hydrochloric acid of different concentrations at a temperature of 30℃. Individual resuscitation process: The "iron poisoned" resin was first soaked in a resuscitation solution of one resin volume for 1.5h, and then the remaining resuscitation solution was passed through the resin layer at a flow rate of 3m/h. After resuscitation, the resin layer was rinsed with demineralized water until neutral. The subsequent regeneration and operation process was the same as in Section 1.2 of this paper. Four influencing factors were considered in the experiment, which were denoted as X1, X2, X3, and X4, as shown in Table 2. Among them, X2 is the mass fraction of hydrochloric acid solution, %; X3 is the time of soaking the resin in hydrochloric acid solution, h; X4 is the multiple of the volume of hydrochloric acid solution used to the volume of the resin layer. X1-X3 each had 10 levels, and X4 had 5 levels.

The experimental design adopted the uniform design method proposed by Fang Kaitai [1]. This experiment used a mixed level table of U10 (103×5), and the experimental results are shown in Table 2.

Comparing the experimental results in Tables 1 and 2, it can be seen that hydrochloric acid resuscitation is effective for mild iron poisoning resin (such as a poisoning rate of less than 10%), while the resuscitation effect on resin with a higher degree of poisoning is not significant. 2.2 Reduction Resuscitation Method 2.2.1 Principle of Reduction Resuscitation Method The traditional resin resuscitation method uses H+ or Na+ to replace Fe3+ on the resin. However, based on the selectivity coefficient of strongly acidic hydrogen ion exchange resin, the selectivity coefficient of Fe3+ is much greater than that of Na+ and H+. Therefore, such exchange is relatively difficult. The basic principle of the reduction resuscitation method is to reduce Fe3+ existing in the ionic state on the resin to the more soluble Fe2+. The latter has a lower affinity for the resin than the former [2]. This makes it easier to exchange Fe3+ from the resin, thereby reducing the amount of regenerator, lowering the concentration of regenerated solution, and shortening the regeneration time. Through experiments, Na2SO3 was selected as the most ideal reducing agent. The redox reaction process of Na2SO3 with ferric iron is shown below:

2Fe3+ + SO32- + H2O → 2Fe2+ + SO42- + 2H+

This reaction proceeds relatively thoroughly, with some Fe2+ being further replaced by Na+ in Na2SO3, and no ferric hydroxide precipitate is produced during this process. The author combined a certain concentration of hydrochloric acid and NaCl with Na2SO3, utilizing the dissolving effect of hydrochloric acid on Fe3+ and the replacement effect of Na+ ions in NaCl to revitalize the resin, achieving good results. The experimental design adopted a uniform design table U20(46), and the experimental results are shown in Table 3. The revitalization process is the same as the revitalization process using hydrochloric acid as the revitalizing agent in Section 2.1 of this paper. In Table 3: X3 is the mass fraction of Na2SO3 solution, %; X4 is the mass fraction of NaCl solution, %; X5 is the multiple of the total volume of the recovery liquid to the total volume of the resin layer; X6 is the volume ratio of hydrochloric acid, NaCl, and Na2SO3 solutions in the recovery liquid. X1-X6 each have four levels. Y1 and Y2 are the resin flow rates before and after recovery, mol/m3; Y3 is the flow rate recovery rate, i.e., the percentage of flow rate after recovery of contaminated resin to flow rate of uncontaminated resin, %.

Note: Since the volume fractions of hydrochloric acid solution and Na2SO3 solution in the resuscitation solution need to be determined in the regression equation, they are represented by X7 and X8 respectively in Table 3.

2.2.2 Experimental Results and Discussion

The data in Table 3 were subjected to regression analysis using a centered quadratic regression model [2]. The stepwise regression method was used to screen the variables entering the regression equation. The test thresholds were F1=0.10 and F2=0.11, and the following regression equation was obtained, where Y represents the working fluid of the resin after resuscitation:

Y=576.533-2.068(X1-35)-222.289(X3-5.5)(X7-0.47)-15.954(X3- 5.5)2-1.609(X1-35)(X4-5.5)+16.022(X2-4.5)2+10.312(X5-5.5)2-7.021(X3-5.5)-6.328(X4-5.5)(X5-5.5)-5.958(X2-4.5)(X5-5.5)

The multiple correlation coefficient R=0.976, F=22.086>F0.995(7,12)=5.52, indicating a significant regression equation. The regression equation shows that three terms in the resuscitation solution are related to Na2SO3, indicating that Na2SO3 plays a crucial role in the resuscitation solution. The magnitude of the regression coefficients reveals that the concentration of Na2SO3 solution and the interaction between its concentration and the volume fraction of hydrochloric acid solution in the resuscitation solution have a significant impact on the processing of the resuscitated resin. In addition, hydrochloric acid solution and sodium chloride solution also have a certain impact on the industrial cross-linking of the revived resin. Experiment 5 has solved the pollution situation with a poisoning rate of 20% quite well.

3. Optimization Experiment.

3.1 Selection of Optimization Starting Point Different degrees of resin poisoning require separate revival to determine the corresponding optimal revival conditions. The following is an explanation of the case with a poisoning rate of 50%. Within the experimental range, the constr function [3] in the optimization toolbox of MATLAB software was used to calculate the revival solution formula corresponding to the maximum value of the industrial cross-linking after revival. After experimental confirmation, the following optimized experimental starting points were obtained: X2=3%, X3=6%, X4=3%, X5=4, X7=0.25, X8=0.55. The corresponding industrial cross-linking is: 630mol/m3.

3.2 Optimization Experiment In order to achieve the best revival effect, four additional experiments were conducted around the above conditions. The experimental design and results are shown in Table 4. The optimal composition of the resuscitation solution obtained after optimization experiments is: X2=3%, X3=6.5%, X4=3%, X5=4%, X7=0.30%, X8=0.5%, and the working fluid of the resin after resuscitation is 640 mol/m3.

4. Conclusion

① As the degree of "iron poisoning" of the resin deepens, the amount of Na2SO3 required in the resuscitation solution should also increase accordingly. The content of hydrochloric acid in the resuscitation solution after the optimization test is lower and the corrosiveness is weakened.

② After using the Na2SO3 reduction resuscitation method to resuscitate the "iron poisoned" resin, the appearance and color of the resin were restored, and no abnormalities were found in the physical and chemical properties. The industrial traffic was also well restored, and the resuscitation effect was good.

③ Using uniform design and regression analysis, the expected results were obtained with a smaller number of tests. The test results show that this optimization method is reliable and has strong practicability.

References:

[1] Fang Kaitai. Orthogonal and uniform experimental design[M]. Beijing: Science Press, 2001.

[2] C M Tilsley. Clean-up of fouled ion exchange resin beds[J]. Effluent and water txeatment journal, 1975, 11(15): 560-563.

[3] Shi Yang, Li Jun, Wang Huigang, et al., Matlab Language Toolbox—Toolbox Practical Guide [M]. Xi'an: Northwestern Polytechnical University Press, 1998.

About the author: Jia Bo (1973-), male, Taiyuan, Shanxi, 2000-level master student in the Department of Water Quality Science and Control Engineering, Wuhan University, Tel: (027) 87888473, jb_99@163.com.