Chlorine/UV induced photochemical degradation of total ammonia nitrogen (TAN) and process optimization

Abstract

Total ammonia nitrogen (TAN) is a pollutant which is spread throughout the world in surface water systems, and this leads to a reduction in water treatment efficiency and the generation of highly toxic disinfection by-products. Although the chlorine/UV process has been demonstrated to be effective for TAN degradation, no information is available to describe the effects of the reaction parameters on TAN control. In this study, experiments were conducted to investigate the effects of the chlorine : TAN molar ratio, UV dose and pH on TAN removal efficiency. Experimental results indicated that both the chlorine : TAN molar ratio and UV dose represented positive effects, while pH showed a complex effect on TAN removal. The interaction between the chlorine : TAN molar ratio and UV dose was also significant, implying a synergistic effect in the chlorine/UV process. Response surface methodology (RSM) was applied to establish the regression model and optimize the chlorine/UV process. Based on the optimization parameter, the combined process was used to treat TAN-containing raw water. The experimental results matched well with the predicted response, indicating the proposed model was accurate and reliable. The main photodecomposition products were nitrate, accompanied with a little amount of nitrite. The mechanism of TAN decomposition in the chlorine/UV process was proposed as UV induced photolysis of chloramine and intermediate radical oxidation. This study demonstrated the chlorine/UV process as an effective strategy for TAN control in water treatment.

---

1. Introduction

Total ammonia nitrogen (TAN) is considered as one of the most important pollutants because of its highly toxic nature and ubiquity in water resources.¹ TAN contamination not only leads to a reduction in water treatment efficiency, but also promotes the generation of highly toxic disinfection by-products (DBPs).^2,3 The Chinese government has established the limit for total ammonia nitrogen (TAN) in drinking water to be 0.50 mg L⁻¹ as nitrogen.⁴

Conventional methods for drinking water treatment such as coagulation, filtration and activated carbon adsorption have very limited efficiency for TAN degradation.⁵ Although TAN could be oxidized by breakpoint chlorination, this process would consume 8–10 fold more chlorine than the disinfection process.^6,7 Furthermore, the generation of toxic DBPs restricts their application in ammonia control.^8,9 Therefore, there is a great need to develop a stable, inexpensive and non-toxic method for TAN control.

Recently, the advanced oxidation process (AOP) has been demonstrated to be an efficient water treatment method. As a novel AOP technology, UV₂₅₄ induced photolysis of chlorine would generate oxidative radicals such as the hydroxyl radical (OH˙) and the chlorine radical (Cl˙), which could effectively oxidize and/or mineralize numerous pollutants.¹⁰ The chlorine/UV process has been proven to be a more effective AOP than UV/H₂O₂, due to a relatively higher absorptivity and quantum yield.^11,12

The chlorine/UV process has been evaluated for the decomposition of emerging contaminants,¹³ natural organic matter (NOM)^14,15 and disinfection by-products (DBPs).¹⁶ This process is also a potential alternative disinfection process based on its performance in inactivating waterborne pathogenic microorganisms.¹⁷ Our previous study has confirmed that the chlorine/UV process represented more advantages than breakpoint chlorination for TAN decomposition, in terms of saving chemicals and having a short reaction time.¹⁸ However, more systematic studies are needed to understand the effect of the reaction variables on the TAN removal efficiency in this combined process.

The main objectives of this study were to investigate the effects of three independent factors (chlorine dose, UV₂₅₄ dose, and pH) on TAN removal and optimize the reaction parameters for the chlorine/UV process. Response surface methodology (RSM) was applied to evaluate the interaction of the independent factors. A mathematical model was established to predict and optimize the operating conditions for TAN removal in the chlorine/UV process.^19,20 The verification experiments were conducted to assess the feasibility of the regression model. The decomposition products of TAN containing raw water formed in the chlorine/UV process were determined. An apparent mechanism was also proposed.

2. Materials and methods

2.1 Materials

The chemicals and reagents used in the experiments were of the highest purity available and they were purchased from Sigma-Aldrich (St. Louis, MO, U.S.). The solutions were prepared with ultrapure water (Thermo Fisher Scientific Inc., Waltham, MA, U.S.). The chlorine stock solutions (1000 mg L⁻¹ as Cl₂) were diluted from a sodium hypochlorite solution (10%, by weight). The TAN stock solutions (500 mg L⁻¹ as N) were prepared by the dilution of ammonium chloride (CAS no. 12125-02-9, purity more than 99.5%). The stock solutions were stored in the dark at 4 °C. 0.05 mol L⁻¹ sulfuric acid and 0.10 mol L⁻¹ sodium hydroxide were used to adjust the pH. 0.01 mol L⁻¹ phosphate buffer was used to maintain the pH in solution.

2.2 The chlorine/UV process

The experimental solutions were freshly prepared before experiments. The initial concentration of TAN in the experimental sample was 0.07 mmol L⁻¹ (1 mg L⁻¹ as N). A series of chlorine solutions were added into the TAN samples to achieve a desired chlorine : TAN molar ratio from 0.33 to 1.17 for 180 min chlorination in the dark. Then, the samples were pumped into a 1.50 L UV reactor (L = 61.00 cm, d = 8.00 cm) equipped with a low pressure Hg lamp (HNG, Germany) with a characteristic wavelength of 254 nm. The schematic of the photochemical reactor was described in our previous published paper.¹⁸ The UV dose imposed on the water samples ranged from 40.00 mJ cm⁻² to 140.00 mJ cm⁻², and this was adjusted by controlling the flow rate of water samples from 56.00 mL s⁻¹ to 16.00 mL s⁻¹. All experiments were conducted in triplicate and expressed as the average value.

2.3 Detection method

In the experiments, the concentrations of chlorine, TAN, nitrite and nitrate were determined based on the standard methods for the examination of water and wastewater published by the American Public Health Association (APHA).²¹ Specifically, TAN was monitored according to the ammonia-selective electrode method (4500-NH₃ D) by an ammonia-selective electrode (Orion Co., 95-12, USA). Free chlorine was determined by the DPD colorimetric method (4500-Cl G). Nitrate and nitrite were measured by the colorimetric method (4500-NO₂⁻ B) and the ultraviolet spectrophotometric screening method (4500-NO₃⁻ B), respectively. A pH meter (Thermo Orion Co., 720A, USA) was used to measure the pH of the solutions. A UV-Visible spectrometer (Cary 300 BIO, Agilent Technologies, CA, U.S.) was applied in absorbance determinations. The TAN removal was calculated as eqn (1), where TAN₀ and TAN_t were the initial TAN concentration and the final TAN concentration, respectively.

2.4 RSM experimental design

Response surface methodology (RSM) is an effective and widely applied statistical method for optimizing the processes of water treatment.^22–24 In this study, a three-factor and five-level central composite design based on RSM was applied to investigate the effects of three independent factors (the chlorine : TAN molar ratio (X₁), the UV dose (X₂) and the pH (X₃)) on the response value of TAN removal (Y) in the chlorine/UV process. The experimental parameters were selected according to our preliminary experimental results¹⁸ and summarized in Table 1.

Table 1 Experimental design matrix and response

Std	Chlorine : TAN molar ratio		UV dose (mJ cm^−2)		pH		Actual value (%)	Predicted value (%)
	Coded value	Actual value	Coded value	Actual value	Coded value	Actual value
1	−1	0.50	−1	60.00	−1	6.50	20.35	20.97
2	1	1.00	−1	60.00	−1	6.50	53.44	53.17
3	−1	0.50	1	120.00	−1	6.50	24.90	26.12
4	1	1.00	1	120.00	−1	6.50	66.87	66.34
5	−1	0.50	−1	60.00	1	8.50	23.11	23.83
6	1	1.00	−1	60.00	1	8.50	59.23	58.20
7	−1	0.50	1	120.00	1	8.50	30.35	30.80
8	1	1.00	1	120.00	1	8.50	73.63	73.20
9	−1.68	0.33	0	90.00	0	7.50	13.89	12.19
10	1.68	1.17	0	90.00	0	7.50	73.48	74.91
11	0	0.75	−1.68	40.00	0	7.50	41.99	42.05
12	0	0.75	1.68	140.00	0	7.50	59.17	58.84
13	0	0.75	0	90.00	−1.68	5.82	36.23	35.70
14	0	0.75	0	90.00	1.68	9.18	43.61	43.87
15	0	0.75	0	90.00	0	7.50	53.92	52.84
16	0	0.75	0	90.00	0	7.50	52.68	52.84
17	0	0.75	0	90.00	0	7.50	53.18	52.84
18	0	0.75	0	90.00	0	7.50	51.94	52.84
19	0	0.75	0	90.00	0	7.50	54.71	52.84
20	0	0.75	0	90.00	0	7.50	50.57	52.84

---

2.5 Statistical analyses

Design Expert 8.0 software (Stat-Ease Inc.) was applied to analyze the variance, establish the mathematical model, and optimize the parameters of the chlorine/UV process. The experimental data were fitted to a second-order polynomial model as defined in eqn (2), where Y means the predicted response; X_i represents the independent variables, and β₀, β_i, β_ij and β_ii represent the intercept, linear coefficients, interaction coefficients and quadratic coefficients, respectively.²⁵

3. Results and discussion

3.1 Model fitting and statistical analysis

Table 1 shows the RSM design and experimental results of the decomposition of TAN in the chlorine/UV process. The TAN removals ranged from 20.35% to 73.63% with the chlorine : TAN molar ratio ranging from 0.33 to 1.17 and the UV dose ranging from 40.00 to 140 mJ cm⁻² at a pH ranging from 5.82 to 9.18. All of these parameters were selected at typical ranges applied in drinking water treatment. It can be seen in Table 1 that TAN removal highly depended on the experimental parameters of the three factors. These results were used to establish the empirical quadratic polynomial model described in eqn (3), where Y represents the TAN removal, and X₁, X₂ and X₃ represent the chlorine : TAN molar ratio, UV dose, and pH, respectively.

As shown in Fig. 1, the predicted values calculated by eqn (3) matched well with the actual decomposition results of TAN in the chlorine/UV process. These results suggested that this regression model was adequate in describing the variability of TAN removal as a function of the chlorine : TAN molar ratio, UV dose and pH value.

Fig. 1 Comparison of predicted values and actual values for TAN removal in the chlorine/UV process. Experimental conditions: [TAN]₀ = 0.07 mmol L⁻¹; 0.01 mol L⁻¹ phosphate buffer.

In order to check the significance of the quadratic model,²⁶ a significance test (F-test) and an analysis of variance (ANOVA) was performed and the results were summarized in Table 2. It can be seen that the probability value (p) of the regression model was less than 0.001, suggesting the model was significant for the predicted response. The coefficient of determination (R²) of the model was 0.996, meaning 99.6% variations of the response could be explained by this model. The fact that the adjusted R² (0.993) was close to R² also confirmed the adequacy of the second-order model. The p value of the lack-of-fit was insignificant which demonstrated that the pure error had an insignificant effect on the modal fitting.

Table 2 Analysis of variance for the response surface model

Source	Sum of squares	Degree of freedom	Mean square	F-value	p-value
Model	5633.95	9	625.99	307.28	<0.001
Residual	20.37	10	2.04
Lack of fit	9.60	5	1.92	0.89	0.548
Pure error	10.77	5	2.15
Cor total	5654.32	19
R-squared	0.996
Adj R-squared	0.993

---

3.2 Effects of the main and interaction variables

The significances of the main and interaction variables on the response were evaluated by the analysis of the p values, as illustrated in Table 3. The p values of all three independent variables were less than 0.001, indicating the linear effects were highly significant for the predicted responses. Moreover, the p values of the interaction variables including X₁X₂ and X₁X₃ were less than 0.05, whereas the p value of X₂X₃ was higher than 0.05. Therefore, the interactions of X₁X₂ and X₁X₃ were significant on the response, but the interaction of X₂X₃ were insignificant terms.

Table 3 Analysis of variance for the effects of the three variables

Source	Sum of squares	Degree of freedom	Mean square	F-value	p-value
X1 – chlorine/TAN molar ratio	4749.33	1	4749.33	2331.32	<0.001
X2 – UV dose	343.66	1	343.66	168.69	<0.001
X3 – pH value	80.57	1	80.57	39.55	<0.001
X1X2	32.16	1	32.16	15.79	0.003
X1X3	2.35	1	2.35	1.16	0.048
X2X3	1.67	1	1.67	0.82	0.386
X1^2	155.57	1	155.57	76.37	<0.001
X2^2	10.43	1	10.43	5.12	0.047
X3^2	307.23	1	307.23	150.81	<0.001

---

Fig. 2 shows the effects of one independent factor on TAN removal with the other two factors fixed at a central level. It can be observed that the chlorine : TAN molar ratio presents a significant positive effect on TAN removal, which was consistent with our previous study.¹⁸ Chlorination of TAN would form chloramine at the range of molar ratio used in these experiments. It has been reported chloramine was more sensitive to UV irradiation than parent TAN.²⁷ The UV dose also showed a positive effect for TAN removal, due to the photolysis of chloramine which was dependent on the UV dose.²⁷ However, pH represented a complicated influence on TAN degradation. The optimal pH of TAN removal was roughly 7.5. This result may be attributed to the fact that pH affected the generation of chloramine by influencing the protonation of both chloramine and TAN. It has also been indicated that alkaline conditions could promote the consumption of radicals, thereby reducing the degradation rate of TAN.¹⁷

Fig. 2 The effects of the independent variables on the TAN removal in the chlorine/UV process. (a) Chlorine : TAN molar ratio; (b) UV dose; (c) pH value. Experimental conditions: [TAN]₀ = 0.07 mmol L⁻¹; 0.01 mol L⁻¹ phosphate buffer.

Fig. 3 presents 3D response surfaces for TAN removal with the interactions of the three independent variables, which also played important roles on TAN removal.²⁸ Fig. 3(a) demonstrates the interaction of the chlorine : TAN molar ratio and the UV dose, which was obviously significant. The degradation of TAN was more apparent with an increase in both the molar ratio and UV dose, indicating a synergistic effect in the chlorine/UV process. There is also an interaction effect between the chlorine : TAN molar ratio and pH. As shown in Fig. 3(b), higher ammonia removal was obtained at a larger chlorine : TAN molar ratio and with the pH around 7.5. These results confirmed the former analysis that pH would affect the extent of chlorination of TAN. At a chlorine and TAN molar ratio of 0.77, mono-chloramine was dominant in the system at a pH ranging from 6.0 to 9.0. Although the photolysis of mono-chloramine was UV dose dependent, the interaction between the UV dose and pH value was not significant, resulting in the minor effect of pH on mono-chloramine formation, as shown in Fig. 3(c).

Fig. 3 Response surfaces plotted on the effects of two independent variables on TAN removal in the chlorine/UV process. (a) Chlorine : TAN molar ratio and UV dose; (b) chlorine : TAN molar ratio and pH; (c) UV dose and pH. Experimental conditions: [TAN]₀ = 0.07 mmol L⁻¹; 0.01 mol L⁻¹ phosphate buffer.

3.3 Optimization of the chlorine/UV process

The optimization of the chlorine/UV process was conducted in order to obtain feasible, economical and maximum parameters for TAN removal. Design Expert 8.0 software was applied to optimize the reaction process based on the ridge analysis.¹⁹ The constraints of the three variables were defined as the minimum for both chlorine : TAN molar ratio and UV dose, whereas the pH was kept in the range of 6.0 to 9.0. Table 4 shows the optimum parameters for the different target TAN removals. Confirmation experiments were also performed and the experimental results matched well with the predicted response. These results confirmed the model could successfully optimize the chlorine/UV process for TAN removal.

Table 4 Optimum conditions at different target TAN removals

No.	Experimental parameters			TAN removal (%)
	Chlorine : TAN molar ratio	UV dose (mJ cm^−2)	pH value	Target value	Actual value	Error	STDEV
Test 1	0.68	112.76	7.39	50	50.32	0.32	±0.49
Test 2	0.88	81.13	7.72	60	59.58	0.42	±0.56
Test 3	0.95	130.34	8.77	70	70.72	0.72	±0.67

---

TAN-containing raw water was used to monitor the pH variation and determine the nitrogenous products during the chlorine/UV process. Water samples were collected from a drinking water treatment plant with an initial TAN concentration of 1 mg L⁻¹ and a background nitrate concentration of 1.42 mg L⁻¹. Unbuffered water samples were used to investigate the variation of pH with a fixed chlorine :  TAN molar ratio of 0.88. A decrease in pH from 7.3 to 7.0 was observed with an increase of UV doses from 40.00 mJ cm⁻² to 140.00 mJ cm⁻². The slight decrease of pH was probably due to the formation of a proton during the oxidation of TAN. In addition, the evolution of various products may also lead to pH variation. As shown in Fig. 4, the concentration of TAN gradually decreased with an increase in UV exposure and then it kept stable when the UV dose was around 80.00 mJ cm⁻². It can be observed that nitrate was the main product in the photodecomposition of TAN, accompanied with a slight formation of nitrite. The generation of nitrite may result from the photo decay of nitrate at a relative high UV dose.²⁹

Fig. 4 Nitrogenous products for TAN decomposition in the UV/chlorine process with background nitrate 1.42 mg L⁻¹. Experimental condition: [TAN]₀ = 1 mg L⁻¹ as N (0.07 mmol L⁻¹); chlorine : TAN molar ratio = 0.88; UV₂₅₄ irradiance was 8.0 × 10⁻⁵ W cm⁻²; pH = 7.5; 0.01 mol L⁻¹ phosphate buffer.

3.4 Mechanism of TAN decomposition in the chlorine/UV process

The mechanism of degradation of TAN chlorine in the chlorine/UV process could be summarized in two aspects: direct photodecomposition of chloramine and indirect oxidation by intermediate radicals. Firstly, at a chlorine : TAN molar ratio ranging from 0.3 : 1 to 1.2 : 1, chloramine was formed due to the high reaction rate constant of the chlorination of TAN,³⁰ accompanied with the residual TAN in the system. A previous study indicated the absorptivity of TAN at 254 nm was close to 0,¹⁸ while chloramine represented a relative high sensitivity of UV₂₅₄ irradiation.²⁷ It has been reported that the molar absorptivity of mono-chloramine was 388 M⁻¹ s⁻¹ and its quantum yield was 0.62.²⁷ These data indicated that UV₂₅₄ induced photo decay was primarily imposed on the chloramine in solution.

Secondly, the oxidative radicals would promote the degradation of TAN. The photodegradation of chloramine induced the cleavage of the Cl–N bond, yielding aminyl radicals and chlorine radicals, as shown in eqn (4).¹⁶ In addition, the photodecomposition of free chlorine would produce hydroxyl radicals and chlorine radicals expressed in eqn (5).¹¹ As described in eqn (6) and (7), both the hydroxyl radicals and the chlorine radicals were effective for TAN oxidation and formed aminyl radicals.³¹

Based on the oxidation process shown in eqn (8)–(10), the aminyl radicals would be further oxidized to nitrite, nitrate, and N₂O by the presence of dissolved oxygen in the system.^27,32,33 These photo decay products were conformable with the products determined in our research, as shown in Fig. 4.

NH₃ + ˙OH → ˙NH₂ + H₂O (6)

NH₃ + ˙Cl → ˙NH₂ + H⁺ + Cl⁻ (7)

˙NH₂ + O₂ → ˙NH₂O₂ (8)

2NH₂O₂˙ → 2HNOOH → 2HNO + H₂O₂ → N₂O + 2H₂O + O₂ (10)

4. Conclusions

This study investigated the influence factors on the removal efficiency for TAN decomposition in the chlorine/UV induced photochemical process. The experimental results demonstrated the chlorine/UV process was effective for TAN degradation. Three independent factors (chlorine : TAN molar ratio, UV dose and pH) represented significant effects on TAN removal. Specifically, the effects of both the chlorine : TAN molar ratio and the UV dose were positive for TAN removal, while the effects of the initial pH value were complex. The interaction between the chlorine : TAN molar ratio and the UV dose is significant, implying a synergistic effect in the chlorine/UV process. A mathematical fitting model was established according to response surface methodology (RSM). ANOVA analysis and confirmation experiments demonstrated that the proposed regression model matched well with the actual chlorine/UV process. The optimum parameter for a TAN removal of 70% was a chlorine : TAN molar ratio of 0.95, a UV dose of 130 mJ cm⁻² and the pH at 8.77, respectively. The main products in this process were nitrate, accompanied with a little amount of nitrite. The proposed decomposition pathways showed that there was the photo decay of chloramine, as well as the oxidation of intermediate radicals. These conclusions suggest that the UV/chlorine process may represent an effective option for TAN control in water treatment.

Acknowledgements

This project was financially supported by the Heilongjiang province Research Council under project no.: 2013G0217 and Heilongjiang province funds for distinguished young scientists (JC 200708).

References

1. H. Zhang, H. J. Liu, X. Zhao, J. H. Qu and M. H. Fan, J. Hazard. Mater., 2011, 190, 645–651.
2. L. S. Wang, H. Y. Hu and C. Wang, Environ. Sci. Technol., 2007, 41, 160–165.
3. T. Bond, J. Huang, M. R. Templeton and N. Graham, Water Res., 2011, 45, 4341–4354.
4. Ministry of Health, China National Standard, 2006, 5749.
5. V. K. Gupta, I. Ali, T. A. Saleh, A. Nayak and S. Agarwal, RSC Adv., 2012, 2, 6380–6388.
6. C. T. Jafvert and R. L. Valentine, Environ. Sci. Technol., 1992, 26, 577–586.
7. A. D. Shah and W. A. Mitch, Environ. Sci. Technol., 2011, 46, 119–131.
8. C. M. M. Bougeard, E. H. Goslan, B. Jefferson and S. A. Parsons, Water Res., 2010, 44, 729–740.
9. D. L. Sedlak and U. von Gunten, Science, 2011, 331, 42–43.
10. Y. Feng, D. W. Smith and J. R. Bolton, J. Environ. Sci. Technol., 2007, 6, 277–284.
11. M. J. Watts and K. G. Linden, Water Res., 2007, 41, 2871–2878.
12. M. J. Watts, E. J. Rosenfeldt and K. G. Linden, J. Water Supply: Res. Technol.-AQUA, 2007, 56, 469–477.
13. C. Sichel, C. Garcia and K. Andre, Water Res., 2011, 45, 6371–6380.
14. X. Zhang and W. Li, Adv. Mater. Res., 2013, 807, 466–471.
15. W. Liu, Z. Zhang, X. Yang, Y. Xu and Y. Liang, Sci. Total Environ., 2011, 414, 576–584.
16. J. Li and E. R. Blatchley, Proc. Water Environ. Fed. Annu. Conf., 68th, 2007, 2007, 128–133.
17. J. Jin, M. G. El-Din and J. R. Bolton, Water Res., 2011, 45, 1890–1896.
18. X. Zhang, W. Li, E. R. Blatchley, X. Wang and P. Ren, Water Res., 2015, 68, 804–811.
19. E. E. Mitsika, C. Christophoridis and K. Fytianos, Chemosphere, 2013, 93, 1818–1825.
20. C. Sahoo and A. Gupta, J. Hazard. Mater., 2012, 215–216, 302–310.
21. APHA, Standard Methods for the Examination of Water and Wastewater, 2005.
22. J. Wu, H. Zhang, N. Oturan, Y. Wang, L. Chen and M. A. Oturan, Chemosphere, 2012, 87, 614–620.
23. K. Cruz-González, O. Torres-Lopez, A. M. García-León, E. Brillas, A. Hernández-Ramírez and J. M. Peralta-Hernández, Desalination, 2012, 286, 63–68.
24. W. Jiang, J. A. Joens, D. D. Dionysiou and K. E. O’Shea, J. Photochem. Photobiol., A, 2013, 262, 7–13.
25. P. D. Saha, A. Dey and P. Marik, Desalin. Water Treat., 2012, 39, 95–102.
26. S. Chowdhury and P. D. Saha, Desalin. Water Treat., 2012, 37, 331–336.
27. J. Li and E. R. Blatchley, Environ. Sci. Technol., 2009, 43, 60–65.
28. I. Oller, S. Malato and J. Sánchez-Pérez, Sci. Total Environ., 2011, 409, 4141–4166.
29. M. N. Chong, B. Jin, C. W. Chow and C. Saint, Water Res., 2010, 44, 2997–3027.
30. M. Deborde and U. von Gunten, Water Res., 2008, 42, 13–51.
31. S. Xiao, J. Qu, X. Zhao, H. Liu and D. Wan, Water Res., 2009, 43, 1432–1440.
32. J. de Laat, N. Boudiaf and F. Dossier-Berne, Water Res., 2010, 44, 3261–3269.
33. K. Clarke, R. Edge, V. Johnson, E. Land, S. Navaratnam and T. Truscott, J. Phys. Chem. A, 2008, 112, 1234–1237.

---