Development of an H₂S emission model for wastewater treatment plants

ABSTRACT

Modeling and prediction of H₂S emission from wastewater are important since gaseous H₂S will induce significant corrosion and odor problems. Most previous studies focused on H₂S emission of wastewater in pipeline systems, which may not be fit for H₂S emission in wastewater treatment plants (WWTPs). This study provided a two-phase mass transfer model for prediction of H₂S emission concentrations. The model is based on the mass transfer rate equation of the mass transfer impetus, expressed by the concentration difference. The main parameters of the model are the mass transfer coefficient, the carrier gas flow rate and the concentration of H₂S in liquid phase. The results showed that the model can simulate and predict H₂S emission concentrations of various processes in WWTPs. Moreover, the model can analyze and predict the influences of different pH values, mass transfer coefficients and carrier gas flow rates on H₂S emission concentrations and loads. Therefore, the model provides theoretical guidance for design of WWTPs regarding H₂S emissions.

Implications: Modeling and prediction of H₂S emission from wastewater are quite important since gaseous H₂S will induce significant corrosion and odor problems. Most of previous studies are focused on H₂S emission from wastewater in pipeline system, which may not be fit for H₂S emission in wastewater treatment plants (WWTPs). Thus in this study, a model for predicting H₂S emission from typical units of WWTPs is established and verified. Moreover, the influences of pH values, mass transfer coefficients and carrier gas flow rates on H₂S emission are analyzed. The model can be a useful tool to predict the H₂S concentration in odor gas collection system of WWTP and understand the behaviors of H₂S emission under different WWTPs operating conditions.

Introduction

While improving the water environment, urban wastewater treatment plants (WWTPs) are also associated with malodorous pollution, which is one of the most typical environmental problems in WWTPs (Pang et al. 2016). Eight types of compounds, ammonia, hydrogen sulfide, methyl mercaptan, methyl sulfide, dimethyl disulfide, carbon disulfide, styrene and xylene used by WWTPs can produce malodors (Yanji 1992). Hydrogen sulfide (H₂S) is considered to be a dominant contributor to odorous compound emissions in WWTPs (Li, Yang, and Li 2018). In addition, H₂S volatilized from water surfaces is the main component of sulfide odor pollution. H₂S concentrations above 1 mg/L can not only seriously affect the surrounding living environment but can also harm human health (Wu et al. 2006). Therefore, odor contamination in WWTPs has recently become a significant global environmental problem (He et al. 2018).

Over the past few years, capping and other malodorous treatment techniques, such as biodegradation, adsorption and photocatalysis, have been used to control H₂S emissions in WWTPs (Estrada et al. 2011; Xi and Hu 2006). Each method has different characteristics regarding the processes and the scope of its applications. Thus, research into the emission characteristics of odorous substances can act as a basis for the selection and design of deodorization schemes for WWTPs. However, many factors affect the emissions of odorous substances (Jung et al. 2017), and it is difficult to control the site conditions of WWTPs. Therefore, it is important to predict H₂S emissions using structural parameters and sewage conditions, which are combined to simulate H₂S emissions.

Both wastewater quality and turbulence can affect the release of H₂S (Beghi et al. 2012). Thus, it is necessary to create a model of the H₂S emission concentrations of wastewater to evaluate the pollution levels of H₂S under different circumstances (Sharma et al. 2008). Lahav et al. (2004), Lahav, Sagiv, and Frisdler (2006) built a model of H₂S released from wastewater in pipes based on the two-film theory. In this model, the effects of the wastewater temperature, average flow velocity and sectional area on the release of H₂S were analyzed in terms of the mass transfer coefficient. Yongsiri, Vollertsen, and Hvitved-Jacobsen (2005) established a similar model of H₂S released from wastewater in pipes; this model was also based on the two-film theory. However, in the study by Yongsiri et al., the calculation of the mass transfer coefficient was connected to the aeration rate of wastewater. Santos et al. (2012) constructed a model of H₂S release in WWTPs based on the two-film theory and used a sodium sulfide solution instead of wastewater. However, the discussion focused on the effect of the air flow rate above the wastewater on the mass transfer coefficient of H₂S, and the effect was not clear. Generally, H₂S release models are based on the two-film theory, and they simulate the mass transfer coefficient of H₂S in different ways.

At present, the WATER9 model, compiled by the US Environmental Protection Agency, is the dominant model used for gas emissions from sewage (Calvo et al. 2018). This model relies on parameters related to the structure of WWTPs, wastewater properties and theoretical models, such as gas volatilization, biodegradation, chemisorption, photochemical reactions and hydrolysis (Dai Xiaobo et al. 2012). In the model, the volatilization rates of volatile organic compounds (VOCs) in structures are calculated to estimate exhaust emissions in the wastewater treatment process. The WATER9 model provides a database of more than 100,000 organic compounds. The model only calculates the emissions of organic compounds and not those of inorganic substances.

Currently, most studies are focused on wastewater in pipes (Liang et al. 2019), which is not suitable for WWTPs. In addition, the WATER9 model is used to calculate the emissions of VOCs. Few studies have focused on H₂S emissions in WWTPs, and the analysis of the influencing factors is insufficient. Therefore, H₂S emissions in a WWTP in Suzhou were analyzed and studied here. An H₂S emission concentration model was established and combined with a two-phase mass transfer process. The results showed that the model can accurately and quickly predict H₂S emissions from wastewater.

Model development

Model assumptions

To simplify the model analysis and resolution, the following assumptions are made (Wei et al. 2018; Xi et al. 2015):

H₂S volatilization on a water surface conforms to the two-mode theory of mass transfer from a liquid phase to a gas phase (Jin and Kim 2018).

The absorption rate of the liquid-to-gas mass transfer process is mainly controlled by the absorption resistance of the liquid film.

This is a steady-state model.

Conceptual model and equations

Under these assumptions, the H₂S emission concentration model can be simplified to a conceptual model, as shown in Figure 1.

Figure 1. Conceptual model of the H₂S emission process

The difference in H₂S concentration between the gas-liquid two-phase flow is used to indicate the mass transfer driving force (Jeanmairet et al. 2015; Jin and Kim 2018). According to the mass transfer rate equation, the following equation can be defined (Seader, Henley, and Roper 1998):

(1) $K_L=\frac{N}{\left(C_{H2S}-{C^{\ast }}_{H2S}\right)}$(1)

where K_L is the mass transfer coefficient with the liquid phase concentration difference as the mass transfer power (m/min), N is the emission flux of H₂S (mg/(m²·min)), C_H2S is the concentration of H₂S in the liquid phase (mg/m³) and C*_H2S is the H₂S concentration in the liquid phase equilibrated with the gas phase (mg/m³).

The emission flux can be defined using the following equation (Seader, Henley, and Roper 1998):

(2) $N=\frac{CMv}{LA}$(2)

where C is the H₂S concentration in the gas phase (ppm), M is the molar mass of H₂S (34 g/mol), v is the carrier gas flow (m³/min), L is the molar volume of the gas (22.4 L/mol) and A is the water area (m²).

The H₂S concentration in the liquid phase is related to the pH value and sulfide content of the sewage, and the correlation can be derived as follow (Tian et al. 2020):

(3) $C_{H2S}=S_T\times \frac{34}{32\left(1+K_{S1}/{10}^{-pH}+K_{S1}{K}_{S2}/{10}^{-2pH}\right)}$(3)

where K_S1 and K_S2 are the primary and secondary ionization equilibrium constants of H₂S, respectively, C_H2S is the H₂S concentration in the liquid phase and S_T is the concentration of total dissolved sulfide in the wastewater (mg/m³).

The H₂S concentration in the liquid phase when equilibrated with the gas phase is expressed as C*_H2S. According to Henry’s law (Skinner 1975), the following equation can be drawn:

(4) ${C^{\ast }}_{H2S}=\frac{CP\rho M}{E{M}_S}$(4)

where P is the standard atmospheric pressure (101.3 kPa), ρ is the density of water (1.0 × 10³ kg/m³), E is the Henry coefficient (kPa) and M_S is the molar mass of water (18 g/mol).

The H₂S emission concentration model is obtained by substituting Equations (2)(2) $N=\frac{CMv}{LA}$(2) and (4(4) ${C^{\ast }}_{H2S}=\frac{CP\rho M}{E{M}_S}$(4) ) into Equation (1)(1) $K_L=\frac{N}{\left(C_{H2S}-{C^{\ast }}_{H2S}\right)}$(1) :

(5) $C=\frac{22.4AE{K}_L{C}_{H2S}}{34vE+4.3\times {10}^{6}A{K}_L}$(5)

Materials & methods

Overview of the WWTP

The validation experiments were carried out in a WWTP located in Suzhou. The first-stage treatment scale of the WWTP was 8 × 10⁴ m³/d, and the second-stage treatment scale was 10 × 10⁴ m³/d. The sewage was first pumped into the grid and the primary sedimentation tank for primary treatment and subsequently entered the UNITANK process for secondary treatment. The secondary effluent was discharged after flocculation, precipitation and ultraviolet disinfection. The sludge was passed through concentration and mud storage tanks, dewatered by a belt filter press and then transported.

Pilot test equipment

In November 2015, a pilot plant test was conducted to continuously monitor the emission concentrations of H₂S. The pilot plant was located near the water intake area of the WWTP, in an area including the intake pump, sewage regulating tank, sewage tank, pipeline and air pump. The sewage was pumped up to the sewage regulating tank and then flowed into the sewage tank. Air was injected above the liquid level of the sewage tank, and the H₂S concentration in the gas was continuously measured at the outlet of the whole sewage tank after the gas had through the tank. The pilot device was completely closed, and a schematic diagram of the device is shown in Figure 2.

Figure 2. Schematic diagram of the pilot experiment

Monitoring index and method

The dissolved sulfide concentration, pH, water temperature, H₂S concentration, carrier gas flow and other indicators were monitored. The dissolved sulfide was filtered through 0.45-μm filter membranes and the sulfide concentration was determined using the methylene blue spectrophotometric method. The water temperature and pH were measured using a portable water quality analyzer (KRK KP-5Z, Beijing Jing Ke Weiye Experimental Equipment Co., Ltd.). The H₂S concentration was measured with a H₂S sensor (CairClip, Cairpol France). The carrier gas flow was controlled by opening and closing the valve and was measured by an impeller anemometer (TESTO 416, German Instrument International Trading (Shanghai) Co., Ltd.).

Parameter estimation and sensitivity analysis

Mass transfer coefficient

In Equation (5)(5) $C=\frac{22.4AE{K}_L{C}_{H2S}}{34vE+4.3\times {10}^{6}A{K}_L}$(5) , K_L, which represents the mass transfer coefficient with the liquid phase concentration difference as the mass transfer power, was determined using a pilot test. The results and parameters of the pilot test are shown in Table 1.

Table 1. Experimental results and parameter values

C (ppm)	v (m^3/min)	A (m^2)	T (°C)	E (kPa)	C*H2S (mg/m^3)	ST (mg/L)	pH	CH2S (mg/m^3)	KL (m/min)
15	0.08	0.66	19	47480	60.45	0.66	7.33	196.91	0.0202
17	0.02	0.66	20	48900	66.52	0.23	7.05	102.94	0.0215
19	0.06	0.66	19	47480	76.57	0.66	7.33	196.91	0.0218
21	0.04	0.66	20	48900	82.17	0.36	6.98	175.57	0.0207
25	0.04	0.66	19	47480	100.75	0.66	7.33	196.91	0.0239
32	0.02	0.66	19	47480	128.96	0.66	7.33	196.91	0.0217
38	0.01	0.66	19	47480	153.14	0.66	7.33	196.91	0.0200
50	0.01	0.66	20	48900	195.65	0.52	7.00	247.62	0.0221

According to the H₂S emission concentration model, the following equation relating to the mass transfer coefficient (K_L) can be drawn:

(6) $34vEC=K_L\left(22.4AE{C}_{H2S}-4.3\times {10}^{-6}AC\right)$(6)

By setting X as 22.4AEC_H2S-4.3 × 10⁻⁶AC and Y as 34vEC, the linear equation can be expressed as Y = K_L X. The results of the pilot test in Table 1 are substituted into Equation (6)(6) $34vEC=K_L\left(22.4AE{C}_{H2S}-4.3\times {10}^{-6}AC\right)$(6) , and the following graph in Figure 3 can be drawn.

Figure 3. Linear regression equation of KL

The calculation results shown in Figure 3 indicate that the mass transfer coefficient (K_L) value is relatively stable at 0.02 m/min.

Parameter values

Based on the experimental conditions, the parameter values selected for the model are shown in Table 2.

Table 2. Values of model parameters

symbol	Parameter	Value	Unit
C	H2S volatile concentration at the surface	Measured	ppm
v	Carrier gas flow	Measured	m^3/min
A	Water area	0.66	m^2
T	Water temperature	Measured	°C
E	Henry’s coefficient	Cited(Skinner 1975)	kPa
ST	Total soluble sulfide in sewage	Measured	mg/m^3
pH	pH value	Measured	-
Ks1	Primary ionization equilibrium constants of H2S	1.3 × 10^−7	-
Ks2	Secondary ionization equilibrium constants of H2S	7.1 × 10^−15	-
CH2S	Concentration of H2S in the liquid phase	Calculated by Equation (3)	mg/m^3
KL	Mass transfer coefficient	0.02	m/min

Model verification

To verify the accuracy of the model, the concentration of H₂S in the gas phase near the intake area was investigated and compared with the closed WWTP structures. Indexes, including the H₂S emission concentration, pH value, water temperature and carrier gas flow, were continuously monitored from the pump house and the primary sedimentation tank in the WWTP. The predicted concentrations of H₂S were obtained after the relevant parameters were inputted into the model. The correlation between the predicted concentrations and the measured concentrations was analyzed, and the graph in Figure 4 was plotted. The calculation results shown in Figure 4 demonstrated that the predicted concentrations were well-correlated with the measured concentrations, with an R² of 0.8714. This suggested that the above H₂S emission concentration model is reliable.

Figure 4. Correlation analysis between the predicted concentrations and actual concentrations of H₂S

Model application

The H₂S emission concentration model established in this study can predict the emission concentrations of structures in WWTPs. Different factors affecting H₂S emission concentrations can be analyzed and predicted in this model. Assuming that the water area of a WWTP was 100 m², the water temperature was 20°C, and the concentration of total dissolved sulfide in the sewage was 0.5 mg/L, the effects of the pH value, mass transfer coefficient (K_L) and carrier gas flow (v) on H₂S emission concentrations were investigated.

Effects of pH value on H₂S emission concentrations

Based on a mass transfer coefficient of 0.02 m/min and carrier gas flow of 20 m³/min, the results in Figure 5 show the changes in H₂S emission concentrations under different pH levels (0–14).

Figure 5. Changes in H₂S emission concentrations and emission loads under different pH values predicted by the model

The results in Figure 5 show that the variation trend of H₂S emission concentrations with varying pH values is similar to that of the equivalent emission loads. With a pH value less than or equal to 6, the H₂S emission concentrations and emission loads were both very high. When the pH value was greater than 6 but less than or equal to 9, the H₂S emission concentrations and emission loads decreased as the pH value increased. When the sewage was highly alkaline, the H₂S emission concentrations and emission loads were at their lowest and could almost be ignored. Yan, Ye, and Zhang (2018) revealed the effect of pH on H₂S emissions from sewage sludge. The results indicated that H₂S volatile concentrations at the surface increased with decreasing pH values, and the amount of released H₂S varied depending on the pH range. A similar conclusion was obtained by Chen et al. (2017), whereby the H₂S emission concentrations of treatment units in the inlet area were negatively correlated with the pH of the sewage, and the H₂S concentrations dropped below 5 mg/m³ when the pH value of sewage was higher than 7.2.

Effects of the mass transfer coefficient parameter on H₂S emission concentrations

Based on a pH equal to 7 and carrier gas flow equal to 20 m³/min, the results in Figure 6 show the changes in H₂S emission concentrations under different mass transfer coefficients (0.02, 0.04, 0.06, 0.08, 0.1, 0.2, 0.3, 0.4 and 0.5 m/min).

Figure 6. Changes in H₂S emission concentrations and emission loads under different mass transfer coefficients predicted by the model

The results in Figure 6 show that the variation trend of H₂S emission concentrations with varying mass transfer coefficients was similar to that of emission loads. The emission concentrations and loads of H₂S increased with increasing K_L. In particular, this increase was rapid when K_L was less than 0.1 m/min, and the increase was slow and then stable when K_L was greater than 0.1 m/min. It has been reported that a low mass transfer coefficient affects the removal efficiency of H₂S (Liu et al. 2015). This may be related to the lower H₂S volatile concentration at the surface when the mass transfer coefficient is low.

Effects of the carrier gas flow parameter on H₂S emission concentrations

Based on a pH equal to 7 and a mass transfer coefficient of 0.02 m/min, the results in Figure 7 show the changes in H₂S emission concentrations under different carrier gas flows (0, 20, 40, 60, 80, 100, 120, 140, 160, 180 and 200 m³/min).

Figure 7. Changes in H₂S emission concentrations and emission loads under different carrier predicted by the model

The results in Figure 7 suggest that the H₂S emission concentrations decreased and the H₂S emission loads increased with increasing carrier gas flow. When the carrier gas flow increased from 0 to 20 m³/min, the changes in H₂S emission concentrations and emission loads were rapid. When the carrier gas flow was more than 20 m³/min, the changes were first slow and then stable.

Conclusion

A theoretical model of H₂S emission is constructed in this study. and is verified with a WWTP in Suzhou, China. Main conclusions are as follow: H₂S emission concentration and H2s emission load is unaffected in extremities of the pH range, but greatly decreases between 5–8. Gradient of H₂S emission concentration gradually flatten with the increase in mass transfer coefficient. Dramatic increase in H₂S emission load is evident with the increase in carrier gas flow from 0–20 m³/min, while the inverse was seen in H₂S emission concentration. Although additional verification would further corroborate the accuracy of the model, evidence gathered hitherto confirmed the relevance and efficacy of the model presented in this study, and would provide theoretical ground for estimations of H₂S emissions in WWTPs in general.

Acknowledgment

The Major Science and Technology Program for Water Pollution Control and Treatment of China (Grant No. 2011ZX07301-003).

Disclosure statement

The authors declare no conflicts of interest. The funding sponsors had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

Supplementary material

Supplemental data for this paper can be accessed on the publisher’s website.

Additional information

Funding

This work was supported by the The Major Science and Technology Program for Water Pollution Control and Treatment of China [Grant No. 2011ZX07301-003].

Notes on contributors

Lan Tian

Lan Tian is a ph.D. student in School of Environment, Tsinghua University majoring in Environmental Science.

Caiyun Han

Caiyun Han was a research fellow in Research Institute for Environmental Innovation (Suzhou) Tsinghua.

Jingyu Zhang

Jingyu Zhang was a research fellow in School of Environment, Tsinghua University.

Yun Ouyang

Yun Ouyang was a master student in School of Environment, Tsinghua University.

Jinying Xi

Jinying Xi is an associate professor in School of Environment in Tsinghua University.