001.docx CHEMICAL ENGINEERING TRANSACTIONS VOL. 83, 2021 A publication of The Italian Association of Chemical Engineering Online at www.cetjournal.it Guest Editors: Jeng Shiun Lim, Nor Alafiza Yunus, Jiří Jaromír Klemeš Copyright © 2021, AIDIC Servizi S.r.l. ISBN 978-88-95608-81-5; ISSN 2283-9216 A Study on Using Cyperus Alternifolius for Horizontal Subsurface Flow Constructed Wetland in Municipal Wastewater Treatment Toan D. Nguyena,*, Ha D. Tranb, Huong M. T. Vic aInstitute of Natural Resources and Environment Training, Ministry of Natural Resources and Environment, No 83 Nguyen Chi Thanh, Dong Da, Ha Noi, Viet Nam bDepartment of Water Supply and Sanitation, Faculty of Environmental Engineering, National University of Civil Engineering, No 55 Giai Phong, Hai Ba Trung, Ha Noi, Viet Nam cDepartment of Environmental Engineering, Thai Nguyen University, Tan Thinh Ward, Thai Nguyen, Viet Nam toantnmt@gmail.com Constructed wetlands (CWs) are engineered systems that utilize natural materials including wetland vegetation, soils and their associated microbial assemblages. Horizontal Subsurface flow (HSSF) as a kind of subsurface flow CW, used commonly in the world since the 1990s but its application in municipal wastewater treatment in Vietnam is still limited. Cyperus alternifolius is wetland plants in HSSF CW to remove pollutants such as N, P and heavy metals. This study aimed at evaluating pollutants removal efficiency of HSSF CW using Cyperus alternifolius as in wetlands under Vietnamese conditions. Experimental model included HSSF CW and aerobic pond (AP) in municipal wastewater treatment systems operated from December 2014 to May 2016. The results of the research show that Cyperus alternifolius grow quickly, remain young and average humidity of the tree is 80.28 %. The growth of Cyperus alternifolius promotes the conversion of organic matter (BOD5) and nutrients (NH4+ − N, NO3− − N, PO43− − P) with decomposition kinetics coefficients kBOD5 is 0.084- 0.150 m/d; kNH4+−N is 0.022-0.046 m/d; kNO3−−N is 0.029-0.059 m/d; kPO43−−P is 0.011-0.030 m/d. The BOD5 removal efficiency reaches 75.30 %; TN, NH4+ − N and NO3− − N removal efficiency are 53.92 %; 58.32 % and 62.82 %; PO43− − P removal efficiency is 62.85 %. 1. Introduction Centralized wastewater treatment system (WWTP) is appropriate for densely populated urban areas. Technologies applied to these systems including aerobic tank, oxidation trench, sequencing batch reactor, etc. required large capital and operational costs. Decentralized treatment is more suitable solution for suburban areas and rural areas. CW and AP are low-cost municipal wastewater treatment technologies for urban areas and suburban residential areas. Surface water bodies such as ponds and lakes, can be utilized as a biological treatment facility as well as ecological landscape. The combination of these two technologies in a treatment system improve pollutants removal efficiency, overcome certain disadvantages of each facility. CW have been widely used in the world since 1990 such as Germany, North America; UK; Italy; Denmark; Czech Republic; Netherlands, Portugal, Slovenia, France, Estonia, Norway, Switzerland (Vymazal and Kröpfelová, 2008) but the application is still limited in Viet Nam. Cui et al. (2009) used Cyperus alternifolius in simulated VFCWs to remove TN in wastewater. Nascimento et al. (2017) studied on oily wastewater treatment by CWs. Bilgin (2014) indicated that activated sludge-vertical flow subsurface constructed wetland systems (VFSCW) with planted Cyperus alternifolius was an economical option for N and P removal in the effluent of secondary treatment such as activated sludge, trickling filter of oxidation ponds. Cyperus ligularis and Echinocloa colona¸ planted in Constructed Wetlands in two local plants of Colombian Caribbean region, was highly effective in removal of dissolved organic matter (COD) and nutrients (NH4+ − N, NO3− − N, PO43− − P) from domestic wastewater (Lizcano et al., 2017). CWs was utilized as an economically and energetically efficient unit process to treat greywater for reuse purposes (Arden and Ma, 2018). CW provided a sustainable DOI: 10.3303/CET2183088 Paper Received: 15/07/2020; Revised: 18/09/2020; Accepted: 13/10/2020 Please cite this article as: Nguyen T.D., Tran H.D., Vi H.M.T., 2021, A Study on Using Cyperus Alternifolius for Horizontal Subsurface Flow Constructed Wetland in Municipal Wastewater Treatment, Chemical Engineering Transactions, 83, 523-528 DOI:10.3303/CET2183088 523 mailto:toantnmt@gmail.com solution to manage resource needs for food and energy production (Avellán and Gremillion, 2019). Wetland roof (WR), a combination of shallow constructed wetlands (SCWs) and green roof is a promising secondary treatment technology to adapt to climate changes and in accordance with the development strategy of green cities (Bui et al., 2019). Tran et al. (2019) studied on pollutant removal by Canna Generalis in tropical CW for domestic wastewater treatment and proved that the technology was highly effective. The HSSF CW is a promising engineering technique to remove excess nutrients and certain pollutants from wastewater. C. articulatus could be a promising wetland vegetation for domestic wastewater treatment in the Colombian Caribbean region (Osorio et al., 2017). Cyperus alternifolius has been widely used in wetlands due to its cost-effectiveness. It can also be ultilized as forage for livestock and aquaculture(Ebrahimi et al., 2017). The HSSF CWs using T. geniculata and C. articulatus as wetland vegetation achieved significant removal effciciency of ammonium, phosphates and COD. CWs using T. geniculata removed higher proportions of NH4 + − N and PO43− − P, CWs using C. articulates generated more biomass (Lizcano et al., 2019). Cyperus alternifolius is a kind of tropical and subtropical vegetation with high nutrient and cellular metabolism (Mburu et al., 2015). This perennial shrub grows quickly in wet or swampy areas, prefers sunlight and resistant to shade. Aquatic plants can be easily propagated by seeds or a plant part. It is utilized for landscapes, fences, harvested to produced paper, hats, bags and also preventing soil erosion. Aquatic vegetation were studied as wetland plants in CWs for wastewater pollutants removal, especially, nutrients as N, P and heavy metals (Liao et al., 2005). This paper focused on studying the feasibility of applications of combined HSSF CW with Cyperus alternifolius and aerobic pond in the low-cost domestic wastewater treatment plant in Viet Nam. The pilot experiment was conducted in suburban residential areas of Cau river basin, Bac Ninh province Viet Nam. 2. Materials and methods 2.1 Experimental model set-up and operation The experimental set-up was described in Figure 1 and previous publication of co-authors (Tran et al., 2019). To assess the capacity of Cyperus alternifolius in removal of pollutants (BOD5, TN, NH4+ − N, NO3− − N and PO4 3− − P ). The experiments were designed in two stages and divided into 5 phases, from December 2014 to May 2016. Stage 1 includes 1 phase (phase 1), evaluating the efficiency of pollutants removal of CW without Cyperus alternifolius. Water sampling and analysis were conducted from March 8th, 2015 to August 29th, 2015 with a two-week frequency. Stage 2 included 4 phases (phase 2,3,4,5), assess the bearing capacity of the model when changing the inflow of wastewater and the efficiency of pollutants removal of CW with Cyperus alternifolius planted. Water sampling and analysis were conducted in 4 periods November 8th, 2015 – December 13th, 2015; December 27th, 2015 – January 31st, 2016; February 28th, 2016 – April 3rd, 2016 and April 17th, 2016 – May 29th, 2016 with a weekly frequency. Figure 1: Experimental model Experimental operating parameters and average pollutant loading rates of inlets were presented in Table 1 and Table 2. HF1 is HSSF without Cyperus alternifolius. HF1’ is HSSF which planted Cyperus alternifolius. 524 Table 1: Operating parameters of the experimental model Experimental phase Work Q (L/h) HRT (d) HLR (m3/m2/d) Phase 1: December 7th, 2014 - August 29th, 2015 HF 1 2 5.72 0.05 Aerobic pond 4 20 - Phase 2: September 26th, 2015 - December 13th, 2015 HF 1’ 2 4.572 0.05 Aerobic pond 5 16 - Phase 3: December 13th, 2015 - February 21st, 2016 HF 1’ 3 3.05 0.075 Aerobic pond 6 13.33 - Phase 4: February 21st, 2016 - April 3rd, 2016 HF 1’ 3.5 2.61 0.0875 Aerobic pond 7 11.43 - Phase 5: April 3rd, 2017 - May 29th, 2016 HF 1’ 4 2.29 0.1 Aerobic pond 8 10 - Table 2: The average pollutant loading rates of the inlet No Pollutants Average of pollutants concentration (mg/L) 1 BOD5 83.39 ±13.62 2 TN(*) 35.19 ± 8.84 3 NH4+ − N 33,17 ±10.71 4 NO3− − N 2.07 ±0.70 5 PO43− − P 1.51 ±0.98 2.2 Research method The samples were taken from inlet and outlet of the HSSF, then analysed in the laboratory of National University of Civil Engineering using Vietnamese. Standard methods applied including: TCVN 6001-1995: Water quality – Determination of biochemical oxygen demand after 5 d - Dilution and seeding method for BOD5 concentration; TCVN 5988-1995 (ISO 5664-1984) – Water quality - Determination of aminonium - Distillation and titration method for NH4+ − N determination; TCVN 6180-1996 (ISO 7890-3-1988) - Water quality - Determination of nitrate - Spectrometric method using sulfosalixylic acid for TN and NO3− − N determination; TCVN 6202:2008 (ISO 6878:2004) - Water quality - Determination of phosphorus - Spectrometric method using ammonium molybdate for PO43− − P determination Determination of kinetic coefficients: Decomposition coefficients of pollutants (BOD5, NH4+ − N, NO3− − N, PO4 3− − P) in HSSF in the experimental model was determined based on the kinetic equation of Kadlec and Knight (1996). Kadlec and Knight (1996) consider plant constructed wetlands as adhesive bioreactors. Kadlec and Knight created a first-stage reaction flow model for all pollutants. The model is based on constant coefficients on first stage reaction, without depending on the temperature. As a result, the Kadlec and Knight models are less sensitive to different climatic conditions. The coefficient of reaction speed was presented in Eq (1) to (3) calculated the speed of hydraulic loading. 𝑙𝑙𝑙𝑙�𝑋𝑋𝑒𝑒−𝑋𝑋 ∗ 𝑋𝑋𝑖𝑖−𝑋𝑋∗ � = −𝑘𝑘 𝑞𝑞 (1) 𝑙𝑙𝑙𝑙�𝑋𝑋𝑖𝑖−𝑋𝑋 ∗ 𝑋𝑋𝑒𝑒−𝑋𝑋∗ � = 𝑘𝑘 𝑞𝑞 (2) q = Q As (3) Where, As is treatment area of HSSF (m2); Xe is the concentration of pollutants in the outflow (mg/L); Xi is the concentration of pollutant in the inflow (mg/L); X* is average concentration of pollutants (mg/L); k is the coefficients of reaction speed of stage 1 (m/d); q is the speed of hydraulic loading (m3/m2.d or m/d) and Q is the average flow rate through the constructed wetlands (m3/d). Average concentration X* of NH4+ − N, NO3− − N and PO43− − P equalled to 0 mg/L. That of BOD5 in HSSF was determined by Eq(4): X∗ = 3.5 + 0.053. Xo (4) Where, Xo is concentration of BOD5 in the sewage and selected as 2 mg/L with HF (Ebrahimi et al., 2013). 525 3. Results and discussion 3.1 The results examining the development of Cyperus alternifolius The results reflecting the development of Cyperus alternifolius at HF1’ during the research period is shown in Table 3. Cyperus alternifolius began to be planted in HF1’ from October 4th, 2015. In November 2015, the tree began to produce the first young branches. In December 2015, the tree grew rapidly, with dozen of new sprouting. During the study period from October 2015 to May 2016, biomass harvest was conducted only twice times on February 28th, 2016 and March 20th, 2016. The total amount of biomass obtained was 2,297.38 g, with an average height of trees in the two harvest periods was 1.28 m and 1.6 m. Thus, the Cyperus alternifolius have adapted and grown well in the HF1’. Table 3: Results of examining the development of Cyperus alternifolius at HF1’ No Harvesting date Wet weight (g) Dry weight (g) Water content (%) Average height (m) Hmax 1 28 February 2016 4,400 867.68 80.28 1.28 1.45 2 20 March 2016 7,250 1,429.7 1.6 1.8 Total 11,650 2,297.38 3.2 Determination of pollutants decomposition coefficients Results of determination of pollutants decomposition factors kBOD5, kNH4+−N, kNO3−−N, kPO43−−P in municipal wastewater in HSSF in the experimental model is shown in Table 4, Table 5. The coefficient of determination between kBOD5 of the HSSF and HLR in the experimental model was close to 1, meant that kBOD5 and HLR were linearly correlated. The coefficient kBOD5 was from 0.084 to 0.150 m/d, reaching the highest at HLR as 0.10 m3/m2/d. This value is consistent with the published results as 0.101 m/d (37 m/y) (Kadlec, 2009); 0.1123 m/d (45 m/y) (Vymazal and Kröpfelová, 2008); 0.060-0.260 m/d (22-95 m/y) (Ngo et al., 2010). Table 4: Coefficient of organic compound decomposition rate (𝑘𝑘𝐵𝐵𝐵𝐵𝐵𝐵5) HLR(m3/m2/d) Load of pollutants (kg/ha/d) kBOD5(m/y) kBOD5 (m/d) R 2 0.05 42.42 31 0.084 ±0.005 0.87 0.075 62.54 35 0.095 ±0.019 0.94 0.088 72.86 47 0.127 ±0.001 0.95 0.1 81.95 55 0.150 ±0.036 0.98 The coefficient kNH4+-N was negatively correlated with HLR, decreased from 0.046 to 0.022 m/d regarding HLR increased from 0.05 to 0.10 m3/m2/d (Table 5). This value was in accordance with the published results 0.024 m / day (Vymazal and Kröpfelová, 2008) and 0.031 m / day (11.4 m/y) (Kadlec, 2009). However, this result is lower than the published result of Kadlec and Knight (1996) with the value of kNH4+−N is 0.093 m/d (Liao et al., 2005). The same correlation was found between kNO3−−N and HLR as well as kPO43−−P and HLR. Table 5: The coefficient of decomposition rate of 𝑁𝑁𝑁𝑁4 + − 𝑁𝑁, 𝑁𝑁𝑁𝑁3 − − 𝑁𝑁 and 𝑃𝑃𝑁𝑁4 3− − 𝑃𝑃 HLR (m3/m2/d) Load of pollutants (gN/m2/d) kNH4+−N (m/d) R2 Load of pollutants (gN/m2/d) kNO3−−N (m/d) R2 Load of pollutants (gN/m2/d) kPO43−−P (m/d) R2 0.05 15.98 0.046 ±0.005 0.89 6.35 0.055 ±0.008 0.81 5.75 0.030 ±0.003 0.99 0.075 26.85 0.029 ±0.010 0.91 11.03 0.059 ±0.008 0.91 11.63 0.013 ±0.003 0.99 0.088 29.65 0.022 ±0.002 0.96 19.69 0.029 ±0.019 0.84 21.18 0.014 ±0.003 0.99 0.1 43.67 0.019 ±0.005 0.97 24.3 0.033 ±0.02 0.84 28.4 0.011 ±0.006 0.99 The coefficient tended to decrease gradually as the HLR index increases into constructed wetlands, ranging around 0.011-0.030 m/d. This result of kPO43−−P was consistent with previous studies of Kadlec and Knight (1996) and Kadlec (2009). These values were lower than in study of Ngo et al. (2010) from 0.112 to 0.230 526 m/d. These were also lower than the result of synthesizing the k value for TP of HSSF for all types of wastewater in general and for municipal wastewater with the values of 0.065 and 0.035 m/d (Vymazal and Kröpfelová, 2008). 3.3 Outcome water quality results after treatment of HSSF The parameters of effluent quality after treatment of HSSF and pollutants removal efficiency are shown in Figure 2a and 2b, Figure 3. The ability of the HF1 's treatment degradable organic matter was lower than that of the HF1. This was because the HF1 formed a water layer on the surface, there was the growth of algae and an increase in the exchange of oxygen with the air, so an aerobic decomposition zone was formed in this water layer. The average BOD5 treatment performance of HF1‘ decreased gradually from phase 1 to phase 5, with a slow speed (down from 75.30 % to 61.56 %). This proved that the HF1 was capable of removing BOD5 stably in the studied HLR range from 0.05 to 0.10 m3/m2/d. (see Figure 2a) (a) (b) Figure 2: Efficiency of (a) biodegradable compound removal, and (b) Nitrogen compounds removal Treatment efficiency of TN, NH4+ − N and NO3− − N in the HF1 and HF1’ is (90.38 %; 92.82 %; 9.45 %) and (53.92 %; 58.32 %; 62.82 %). The HF1 has much higher treatment efficiency for TN and NH4+ − N and has much lower efficiency of treating NO3− − N than the HF1'. Treatment efficiency of TN, NH4+ − N and NO3− − N of HF1’ is not high, it proved that the crop has a negligible influence on the nitrogen compounds treatment capacity of the HSSF (see Figure 2b). The PO43− − P treatment efficiency of HF1’ is higher than that of HF1. Because the HF1 without plants, there is no mechanism of absorption of plants like HF1’. The reason for HF1’ is more efficient because the Cyperus alternifolius begins to grow in this experiment, biomass increases rapidly and the need to absorb nutrients also grows. It leads to Increase absorption of PO43− − P (see Figure 3). Figure 3: Efficient treatment PO43- - P 527 4. Conclusions The study found that Cyperus alternifolius had fast-growing speed and its growth promoted the conversion of organic matter (BOD5) and nutrients (NH4+ − N, NO3− − N, PO43− − P). The ability to treat degradable organic substances in HF1’ is lower than that in HF1. The ability to treat nitrogen compounds in municipal wastewater. The removal performance in the model with Cyperus alternifolius was higher than that without this plant regarding NO3− − N and — P. The TN and NH4+ − N removal performance was lower in the model with Cyperus alternifolius. References Arden S., Ma X., 2018, Constructed wetland for greywater recycle and reuse: A review, Science of the Total Environment, 630, 587-599. Avellán T., Gremillion P., 2019, Constructed wetlands for resource recovery in developing countries, Renewable and Sustainable Energy Reviews, 99, 42–57. Caselles-Osorio A., Vega H., Lancheros C.J., Casierra-Martínez A.H., Mosquera E.J., 2017, Horizontal subsurface-flow constructed wetland removal efficiency using Cyperus articulatus L, Ecological Engineering, 99, 479-485. Cui L.H., Ouyang Y., Chen Y., Zhu X.Z., Zhu W.L., 2009, Removal of total nirogen by Cyperus alternifolius from wasteater in simulated vertical – flow constructed wetlands, Ecological Engineering, 35(8), 1271- 1274. Ebrahimi A., Taheri E., Ehrampoush M.H., Nasiri S., Jalali F., Soltani R., Fatehizadeh A., 2013, Efficiency of constructed wetland vegetated with Cyperus alternifolius applied for municipal wastewater treatment, Journal of Environmental and Public Health, 2013, 815-962. Kadlec R.H., Knight R.L., 1996, Treatment wetlands: CRC Press, Boca Raton, Florida,USA. Kadlec R.H., 2009, Comparison of free water and horizontal subsurface treatment wetlands, Ecological Engineering, 35(2), 159–174. Lianos-Lizcano A.H., Barraza E., Narvaez A., Casierra-Martínez A.H.L., Charris-Olmos C.J., Caselles-Osorio A., Parody-Muñoz E.A., 2017, Organic Matter and Nutrients Removal in Tropical Constructed Wetlands Using Cyperus ligularis (Cyperaceae) and Echinocloa colona (Poaceae), Water Air Soil Pollution, 228(9), 1 – 10. Liao X., Luo S., Wu Y., Wang Z., 2005, Comparison of nutrient removal ability between Cyperus alternifolius and Vetiveria zizanioides in constructed wetlands, The Journal of Applied Ecology, 16(1), 156–160. (in Chinese). Mburu N., Rousseau D., Bruggen J., Lens P., 2015, Use of macrophyte Cyperus papyrus in wastewater treatment, Chapter In: Jan Vymazal (Ed.), The Role of Natural and Constructed Wetlands in Nutrient Cycling and Retention on the Landscape, Springer, Cham, Switzerland, 293-314. Nascimento L., Rocha E Silva N., Santos K., Sarubbo L., Santos V., Benachour M., 2017, Improved Efficiency of Constructed Wetlands for Oily Water Treatment with Aid of Microbubbles, Chemical Engineering Transactions, 57, 535-540. Ngo T.D.T., Konnerup D., Schierup H.H., Nguyen H.C., Le A.T., Brix H., 2010, Kineritcs of pollutant removal from domestic wastewater in a tropical horizontal subsurface flow constructed wetland system: Effects of hydraulic loading rate, Ecological Engineering, 36(4), 527–535. Tran D.H., Vi T.M.H., Dang T.T.H., Narbaitz M.R., 2019, Pollutant removal by Canna Generalis in tropical constructed wetlands for domestic waste water treatment, Global Journal of Environmental Science and Management, 5(3), 331-344. Vo T.D.H., Bui X.T., Lin C., Nguyen V.T., Hoang T.K.D., Nguyen H.H., Nguyen P.D., Ngo H.H., Guo W., 2019, A mini-review on shallow-bed constructed wetlands: a promising innovative green roof, Current Opinion in Environmental Science & Health,12, 38–47. Vymazal J., Kröpfelová L., 2008, Wastewater treatment in Constructed wetlands with Horizontal Sup-surface flow, Springer, Dordrecht, Netherlands. 528