Development of a Software to Simulate Effluent
Content
82
Int. J. Process Systems Engineering, Vol. 1, No. 1, 2009
Development of a software to simulate effluent characteristics of textile wastewater treatment Adel Al-Kdasi
Department of Chemical and Environmental Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia and Faculty of Agriculture, Sana’a University, Yemen E-mail: [email protected]
Azni Idris
Waste Technology Centre, Department of Chemical and Environmental Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia E-mail: [email protected]
Luqman Chuah Abdullah
Department of Chemical and Environmental Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia E-mail: [email protected]
Mohanad El-Harbawi*
Department of Chemical Engineering, Universiti Teknologi Petronas, Bandar Seri Iskandar, 31750 Tronoh, Perak, Malaysia E-mail: [email protected] *Corresponding author
Mogeeb Alzokry
Department of Software Engineering, Faculty of Computer Science and Information Technology, Universiti Malaya, 50301, Kuala Lumpur, Malaysia E-mail: [email protected]
Copyright © 2009 Inderscience Enterprises Ltd.
Development of a software to simulate effluent characteristics
83
Chun-Yang Yin
Faculty of Chemical Engineering, University Technology Mara, 40430 Shah Alam, Selangor, Malaysia E-mail: [email protected]
Abstract: The composition of textile wastewaters is extremely varied due to the large spectrum of dyes and chemicals used in wet process resulting in greater variability in the success of different treatment processes. A software was developed to determine the required advanced oxidation process after biological treatment and to evaluate the possibility of applying advanced oxidation processes to any textile treatment plant. Regression equations related to the formulation of the process treatment options were successfully developed into a software program called TexTreat with integrated graphic user interface (GUI) component. TexTreat has the ability to determine the best option of advanced oxidation processes (AOPs) and predict characteristics of wastewater discharge in different retention times with an overall simulation accuracy of more than 89%. Validation of the process treatment options and TexTreat display their applicability to utilise with different textile wastewater plants. Keywords: textile wastewater; advanced oxidation process; AOP; software. Reference to this paper should be made as follows: Al-Kdasi, A., Idris, A., Chuah Abdullah, L., El-Harbawi, M., Alzokry, M. and Yin, C-Y. (2009) ‘Development of a software to simulate effluent characteristics of textile wastewater treatment’, Int. J. Process Systems Engineering, Vol. 1, No. 1, pp.82–99. Biographical notes: Adel Al-Kdasi is currently an Assistant Professor in the Faculty of Agriculture-Sana’a University. Prior to this, he obtained his PhD from the Department of Chemical and Environmental Engineering, Universiti Putra Malaysia (UPM). He has published several papers in textile treatment field. Azni Idris received his PhD in Environmental Engineering from the University of Newcastle upon Tyne, UK. He is currently the Head of Department of Chemical and Environmental Engineering at Faculty of Engineering, Universiti Putra Malaysia. He is the Director of Technology Commercialisation, University Business Center, University Putra Malaysia (UPM). He is also the Director of Waste Technology Center, UPM. As an active researcher, he had published more than 100 publications in international journals and proceedings. In addition, he was awarded for research excellence and his contribution in the production of biofilter [holder of patent on organic waste treatment process (BioFil sysytem)]. Luqman Chuah Abdullah is currently an Associate Professor in the Department of Chemical and Environmental Engineering, Universiti Putra Malaysia (UPM). He has also served as Head of Research Laboratory in the Institute of Tropical Forest and Forest Products (INTROP), UPM. As an active researcher, he had published 160 publications on journals and proceedings, of which many of them are internationally cited and reputable in chemical and environmental engineering. In addition, he won several awards in teaching and research, including Young Engineer Award from the Institution of Engineers Malaysia (IEM) in year 2006.
84
A. Al-Kdasi et al.
Mohanad El-Harbawi received his Doctorate in Chemical and Environmental Engineering from the University Putra Malaysia in 2006. He is part of lecturing team at the Department of Chemical Engineering at Universiti Teknologi PETRONAS (UTP). Currently, he is working diligently in the development of a GIS-based air pollution dispersion modelling software. He has developed several softwares for chemical risk assessment and air pollution assessment. Furthermore, he has several publications including ISI-indexed journal articles and conference proceedings. In addition, he is also a reviewer for Journal of Engineering Science and Technology (JESTEC). Mogeeb Alzokry received his BSc in Software Engineering from Technology University – Iraq in 2002. In 2003, he was appointed as a Staff at Taiz University in the Faculty of Engineering. He is currently pursuing his PhD at the University of Malaya. To date, he has published several papers in indexed journals and conference proceedings. Chun-Yang Yin received his PhD in Chemical Engineering from the University of Malaya and was appointed as a Visiting Scholar at the prestigious Ivy League Institution, Columbia University from December 2008 to March 2009. He has over 60 publications to date including ISI-indexed journal articles and conference proceedings. He is the recipient of numerous universities, national and international-level awards including the IEM Tan Sri Raja Zainal Prize in 2005 and 2009 as well as listed in the 25th Silver Anniversary Edition of Marquis Who’s Who in the World. He is currently a Reviewer for several Elsevier journals and is on the Editorial Board of Malaysian Journal of Chemical Engineering.
1
Introduction
Several conventional methods are used to treat textile dye wastewaters such as biological treatment, chemical oxidation, coagulation, adsorption and filtration. However, the efficiencies of these methods depend strongly on the types of dye in wastewater and concentration of contaminants. Colour removal from textile effluents in aerobic biological treatment is not an effective process since the biodegradation products are toxic to the organisms used in the process and these result in various problem such as sludge bulking, rising sludge and pin-point floc formation (Straley, 1984; Paprowicz and Slodczyk, 1988; Pagga and Brown, 1986; Lin and Peng, 1996; Antonio et al., 1997; Wilmott et al., 1998). According to Antonio et al. (1997), biological treatments are reliable, but there are certain substances which are unable to deal with. Thus, there are a lot of combinations of chemical oxidation and biological treatment which can be arranged for organic removal from toxic wastewater. However, bio-chemical treatment of textile wastewater effluents that contain dyes and their hydrolysis products can be a cost-effective alternative when the effluents are pretreated chemically prior to treatment in the biological unit (Lin and Peng, 1996). Physico-chemical methods such as coagulation/flocculation, activated carbon adsorption and reverse osmosis techniques have been developed in order to remove the colour (Dae-Hee et al., 1999; Bes-Piá et al., 2003; Maria et al., 2004). However, these methods can only transfer the contaminants from wastewater to solid waste leaving the problem essentially unsolved (Mariana et al., 2002; Georgiou et al., 2002; Arslan et al.,
Development of a software to simulate effluent characteristics
85
2002). Therefore, special attention should be focused on the techniques that lead to the complete destruction of the dye molecules. Advanced Oxidation Processes (AOPs) is one of these developing processes that has been used to generate hydroxyl free radicals by different techniques to destroy the contaminants in textile wastewater. Ozone (O3), hydrogen peroxide (H2O2) and UV irradiation are recently used to accelerate the generation of hydroxyl radical (HO•). This strong oxidant can destroy compounds that can not be readily oxidised by conventional oxidants such as ozone, oxygen and chlorine. In addition, AOPs do not generate chemical sludge and toxicity of wastewater is generally reduced. Several studies have been reported about the successful application of the AOPs (Nilsun, 1999; Arslan and Isil, 2001; Arslan et al., 2002; Georgiou et al., 2002; Yung-Shuen and Deng-Kae, 2002; Tanja et al., 2003; Azbar et al., 2004; Ulusoy, 2004; Shu et al., 2004; Shu and Chang, 2005). Previous studies were carried out within a certain range of different kind of synthetic dyes and synthetic textile wastewater in the batch mode. Few studies were reported in the literature that deal with the actual wastewater due to their inability to cope with the complex background matrices and high pollution load normally encountered in real wastewater (Beltran et al., 1997; Wenzel et al., 1999). On the other hand, extraordinary high oxidant doses of chemicals make the industrial scale stand-alone and the application of chemical unfeasible from the economic point of view (Masten and Davies, 1993; Beltran et al., 1997). As such, an integrated system of biological treatment and AOPs would mean a cheaper option for total organic degradation and colour removal from the textile wastewater. Having different, easily controlled and successful processes that can deal with the different strength of textile wastewater is the best way to ensure efficient colour and organic pollutants removal from textile wastewater. Previous related studies were focused on development of models to predict performance of wastewater treatment systems without integration of graphic user interface (GUI) to enhance user-friendly feature of the model. Examples of such reported studies include Bernard et al. (2001), Hamed et al. (2004) and Joksimovic et al. (2006) which focused on development of new models for wastewater parameters estimation as applied for anaerobic, urban domestic and reclaimed wastewater respectively. In a recent study on software development, Hidalgo et al. (2007) developed a multi-criteria analysis user friendly software to assist responsible authorities in determining the most efficient solutions in terms of health and safety for the agricultural reuse of the produced effluent. From our understanding, the concept or idea to develop a software with GUI to determine the best advanced oxidation process after biological treatment and predict characteristics of treated textile wastewater has not been considered in previous studies. As such, the objectives of this study were to develop software (designated as TexTreat) to determine the required advanced oxidation process after biological treatment and evaluate the possibility of applying AOPs to any textile treatment plant. Development of TexTreat was divided into three distinct stages. The first stage assigned a reaction order for each category which was determined by using the integration method. Reaction rate coefficient (rate constant) was evaluated by substitution using the test data at different retention time for colour, total organic carbon (TOC) and chemical oxygen demand (COD). The second stage assigned the best fitting regression equations between different values of the parameters and their reaction rate coefficients were determined at different retention times using SPSS software. The final stage was to write the program for the TexTreat software.
86
A. Al-Kdasi et al.
2
Materials and methods
2.1 Textile wastewater collection
To achieve the formulation treatment option, samples of textile wastewater with different concentrations were collected and studied. Each sample was studied separately using integrated system of biological and advanced oxidation process (data not shown). The textile wastewater was collected every five days from a local manufacturing factory effluents of Kim Fashion Knitwear Sdn Bhd located in Senawang, Negeri Sembilan, Malaysia. Wastewater was produced from several finishing and dyeing units, which were used to dye fabrics, hanks and socks of different natural or synthetic fibres and mixture of both. Three different textile wastewater factories were used to validate the TexTreat application and the formulation of process treatment options. These three factories were Kim Fashion Knitwear (M) Sdn Bhd located in Senawang, Negeri Sembilan, Pacific Peninsula Textile Sdn Bhd located in Johor Bahru and Ramatex Textiles Industrial Sdn Bhd located in Batu Pahat, Johor Bahru.
Figure 1 Design configuration of bio-photochemical reactor
Development of a software to simulate effluent characteristics
87
2.2 Experimental model setup
The laboratory setup consisted of activated sludge process (extended aeration) followed by AOP. The reactor was designed to be operated in continuous mode. The laboratory-scale model reactor used in this study is shown in Figure 1.
2.2.1 Activated sludge model
The activated sludge process reactor consisted of a feeding tank, aeration tank and settling tank made of polyvinyl chloride (PVC). Raw wastewater pH in the feeding tank was adjusted to pH 7 using 1 M hydrochloric acid and mixed continuously using a motorised stirrer (Eurostar Digital, Germany) at a rate of 50 rpm. A continuous flow rate of 1.26 l/h using a peristaltic pump (Heidolph PD5006, Germany) was used. Complete mixing and aeration were assured by the continuous flow of air into the aeration tank. The dissolved oxygen concentration in aeration tank was kept between 2 to 3 mg/l. The aeration tank was connected to the settling tank to remove suspended solids. Biotreated effluent was subsequently discharged through a plastic pipe to the advanced oxidation process reactor. All experiments were conducted at room temperature range of 25–28°C. Samples for analysis were collected from feeding tank, discharge point after aeration tank and discharge point after AOP. Microorganisms of the aeration tank were acclimatised to receive high-level of colour and organic matter in the wastewater using actual textile wastewater without any pretreatment for three months. The system was operated at F/M ratio of around 0.05 (g BOD /g VSS.d) and dissolved oxygen concentrations from 2 to 3 mg/l.
2.2.2 Advanced oxidation model
AOP reactor was made of 304 stainless steel with 2.5 litre capacity. The reactor was cylindrical and equipped with a UV lamp (wavelength, λ = 254 nm). Both the UV-C and quartz lamp housings were centred in the reactor tube. Five outlet discharge points were pointed corresponding to different retention times in the system. O3 generator (OWA 350) with an O3 production rate of 350 mg/hr was used for this experiment. The O3 was produced by natural intake of air from surrounding and was bubbled into reactor by diffusers. The applied O3 concentrations for this experiment were 183, 152 and 101 mg/L. Ozone was introduced into the bottom of the reactor through two diffuser points. The exhaust gas was vented from the top of the reactor passing through two bottles of potassium iodide solution (KI) and 2% absorption solution was then vented into a laboratory hood. Different concentrations of hydrogen peroxide (30%) were applied using a peristaltic pump. Samples were collected from the sampling point at regular time intervals (15, 30, 60, 90 and 120 minutes) and kept for further analyses.
2.3 Analytical methods
True colour (PtCo) of the samples was measured using a spectrophotometer (HACH DR 2500, USA) calibrated according to standard platinum-cobalt method. Prior to colour measurement, both the influent and effluent were filtered using membrane filter paper
88
A. Al-Kdasi et al.
with a pore size of 0.45 µm (Whatman, Japan). The mixed liquor suspended solids (MLSS), total suspended solids (TSS), COD and biochemical oxygen demand (BOD) measurements were determined according to standard methods for examination of water and wastewater (APHA et al., 1998). Ozone concentration in feed gas was determined by KI-starch titration method (APHA et al., 1985). The TOC concentration in solution was determined using TOC analyser (Shimadzu 5050, Japan). Wastewater pH was measured by means of a laboratory multi-meter (WTW, InoLab Multi Level 1, Germany). The average values of the results were obtained by three repeated experiments. Statistical software (SPSS, Version 12) and standard Microsoft Visual Basic 6.0 for windows computer program was used for the regression and developing of the software, respectively.
3
Results and discussion
3.1 Formulation treatment options
Samples of textile wastewater with different concentrations were studied to obtain formulation process treatment option for textile wastewater. Biotreated textile wastewater was successfully categorised into four groups; corresponding to these categories, different AOPs were formulated for use. Table 1 shows the categories of biotreated textile wastewater and the best methods of advanced oxidation process for each category. From the formulation treatment options, a software (TexTreat) was developed to assist operation in selecting the most appropriate AOP for the textile wastewater and also to estimate the discharge effluent concentration.
Table 1 Categories of biotreated textile wastewater and best selected methods of advanced oxidation Condition Colour ≤ 400 and TC ≤ 80 mg/l Colour ≤ 400 and TC > 80 mg/l Category 2 Category 3 Category 4 400 < colour ≤ 800 800 < colour ≤ 1200 Colour > 1200 Best method of AOPS 0.25 ml/l H2O2/UV 0.5 ml/l H2O2/UV 0.75 ml/l H2O2/UV/50 mg/l O3 1.5 ml/l H2O2/UV/134 mg/l O3 2 ml/l H2O2/UV/183 mg/l O3
Category Category 1
3.2 Determination of the reaction order and reaction rate coefficient
Integration method was used to determine the order of the reaction for the selected process. Three reaction orders (zero, first and second order) were plotted. As shown in Figures 2–4, the experimental data fitted the second-order (straight line for the parameters removal) the best among the three orders. Only the second order curve was shown for reason of brevity. As shown in Figure 2, the second order give the straight line for the colour removal where R2 (coefficient of determination), were higher than R2 for plots C and –log[C/Co] (zero and first order).
Development of a software to simulate effluent characteristics
Figure 2 Colour graphical analysis for the determination of reaction order for Category 1
0.18 0.16 0.14 0.12 1/Colour 0.1 0.08 0.06 0.04 0.02 0 0 15 30 45 60 75 90 105 120 135
89
Category 1
R2 = 0.9961
Retention Time (min)
Figure 3 TOC graphical analysis for the determination of reaction order for Category 1
0.14 0.12 0.1 0.08 0.06 0.04 0.02 0
Category 1
R2 = 0.6141
1/T O C
0
15
30
45
60
75
90
105
120
135
Retention Time (min)
Figure 4 COD graphical analysis for the determination of reaction order for Category 1
0.05 0.045 0.04 0.035 1/C O D 0.03 0.025 0.02 0.015 0.01 0.005 0 0 15 30 45 60 75 90 105 120 135 Retention Time (min)
Category 1
R2 = 0.9194
90
A. Al-Kdasi et al.
R2 for the second, third and fourth categories are shown in Table 2. Reaction rate coefficients were determined for colour at the different retention times for the different categories. Table 2 shows reaction rate coefficients for colour at the different retention time for the different categories. These reaction rate coefficients were utilised in TexTreat to predict the residual of parameters undergoing AOPs.
Table 2 Reaction rate coefficient for colour removal at different retention times Reaction rate coefficient 15 min 0.0015205 0.0010198 0.0004342 0.0000977 30 min 0.0014929 0.0009644 0.0005370 0.0001482 60 min 0.0012348 0.0009576 0.0006033 0.0002188 R2 0.9961 0.9769 0.9912 0.9614
Categories Category 1 Category 2 Category 3 Category 4
The reaction orders for TOC and COD were determined and the reaction rate coefficients were evaluated as well. As shown in Figures 3–4, TOC and COD removal data fit the second-order quite well in which R2 was higher than R2 for zero and first-order. R2 for the second, third and fourth categories are shown in Table 3.
Table 3 Reaction rate coefficients for TOC and COD removals Reaction rate coefficient 15 min 0.002187 0.0014545 0.0013813 0.0003819 0.0008333 0.0006871 0.0006030 0.0001502 30 min TOC Category 1 Category 2 Category 3 Category 4 Category 1 Category 2 Category 3 Category 4 0.0018628 0.0010726 0.0010085 0.0004395 COD 0.0004605 0.0004825 0.0006792 0.000233 0.0003293 0.0004198 0.000621 0.0003635 0.919 0.984 0.893 0.955 0.0009978 0.0009206 0.0008118 0.0004803 0.614 0.961 0.981 0.998 60 min R2
Categories
The reaction rate coefficients for TOC and COD were evaluated at different retention time for all the categories. Table 3 shows the reaction rate coefficients for TOC and COD at different retention time for all the different categories. The entire reaction rate coefficients were used to get the relationship between the reaction rate coefficients and parameters and then used in the TexTreat to determine the values of the parameters after the different retention time of AOPs. Once the reaction rate and coefficient are known, the second integrated form could be used. The concentration of the sample could be determined at selected retention time treatment of AOPs if the initial concentration of parameter, retention time and reaction rate coefficient are known. The integrated form for-second order can be written as follows:
Development of a software to simulate effluent characteristics
1 1 − = kt C CO 1 1 = + kt C CO C = CO / (1 + k * t *CO )
91 (1) (2) (3)
where C = concentration of constituent CO = concentration of constituent at zero time k = reaction rat coefficient, (mg/L3) n–1/min t = time
3.3 Determination of the best fitting regression equation
Relationship between different values of the parameters and their rate constants were determined at different retention times. For each parameter, three formulas were determined (first for the 15 min and second for 30 min and the last for 60 min). SPSS was used to determine the best fitting curve and the regression equations. The second-order was the best equation that adequately describes the relationship between the concentration of parameters and 1/k. The coefficients of determinations (R2) were more than 0.9 and the standard errors were around zero with the second regression equations. This indicated all the input data fitted the second-order very well and that since all the R2 values for all regression coefficients were all very close to unity, it implied that all these equations are applicable for data input. Figures 5–7 show the regression equations fitted between different parameters at 15 min and 1/k using Microsoft Excel. All regression fitting were similar to SPSS regression. The regression equations at the different retention times were written in the TexTreat code. Table 4 shows the regression equations and R2 at different retention time.
Figure 5 Relationship between 1/k and colour at 15 min
12000 10000 8000 1/K 6000 4000 2000 0 0 1000
y = 5E-05x 2 + 1.8787x - 148.85 R2 = 0.9982
Data Equation plot
2000
3000 Colour
4000
5000
6000
92
A. Al-Kdasi et al.
Figure 6 Relationship between 1/k and TOC at 15 min
3000 2500 2000 1/k 1500 1000 500 0 10 20 30 40 TOC 50 60 70 80 Data Equation plot y = 0.6153x2 - 21.382x + 685.15 R2 = 0.9943
Figure 7 Relationship between 1/k and COD at 15 min
7000 6000 5000 1/K 4000 3000 2000 1000 0 50 70 90
y = 0.2736x2 - 44.725x + 3103.2 R2 = 0.9987
Data Equation plot
110
130
150 COD
170
190
210
230
250
Table 4
Regression equations Time 15 min 30 min 60 min Regression equations Y=(5E-05x2 + 1.8787x – 148.85)–1 Y=(–2E-05x + 1.4466x + 103.5)
2 –1
Parameter Colour
R2 0.998 0.999 0.998 0.994 0.985 1 0.999 0.989 0.999
Y=(–4E-05x2 + 1.0801x + 382.33)–1 Y=(0.6153x2 – 21.382x + 685.15)–1 Y=(0.1339x2 + 17.1x + 197.21)–1 Y=(0.1967x2 + 0.44x + 912.68)–1 Y=(0.2736x2 – 44.725x + 3103.2)–1 Y=(0.3133x2 – 80.454x + 6706.6)–1 Y=(0.3196x2 – 98.492x + 8864.7)–1
TOC
15 min 30 min 60 min
COD
15 min 30 min 60 min
Development of a software to simulate effluent characteristics
93
3.4 Formulation of computer algorithm
All equations that determined the removal of colour, TOC and COD at the different retention time for the parameters at different retention times were written in TexTreat code. The program was written in standard Microsoft Visual Basic 6.0 and distributed in the source code. The computational of the mathematical models for the result estimation was calculated by using VB program code. The main interface has three buttons (enter, information and exit) as shown in Figure 8. The enter button allows the user to go into the decision support interface for textile wastewater treatment. The information button provides the user detailed description about TexTreat. It is designed using front page and HTML (hyper text markup language). The exit button is used to close the TexTreat.
Figure 8 The main and submenus interfaces (see online version for colours)
94
A. Al-Kdasi et al.
The decision support interface is used to insert the characteristics of textile biotreated wastewater. The decision support interface of TexTreat contains three main stages. The different stages and instructions to use the TexTreat are shown in Figure 9. In the first stage, input data stage, the characteristics of the biotreated textile wastewater were inserted to the processes of running the TexTreat. In the input data stage, different buttons and text boxes for different parameters of characteristics of textile wastewater and the run (process) of the program were used. In the second stage, the required treatment method is achieved in text box. This stage also has buttons to estimate the results by using the determination method. In the final stage, the estimated results using the determination method appear at different text boxes. The results estimated appear for different parameters (colour, TOC and COD) in different retention time (15, 30 and 60 min). Three different separated columns were used to show the result for the different retention time for the different parameters.
Figure 9 Flowchart of instructions of using the TexTreat (see online version for colours)
Figure 10 shows the TexTreat algorithm. An algorithm is basically a succession of instructions or a process used for calculation and data processing. At the beginning of the algorithm sequence, data pertaining to characteristics of wastewater are put into the TexTreat. After commencement of a run, TexTreat will basically determine the level of hydrogen peroxide/UV needed based on two main parameters, namely, colour and total
Development of a software to simulate effluent characteristics
95
carbon. Subsequent to initiation of ‘treatment’, the characteristics of the effluent will be predicted. The TC and colour values for the algorithm are obtained from Table 1.
Figure 10 Flow diagram of the TexTreat algorithm
96
A. Al-Kdasi et al.
3.5 Validation of TexTreat and process treatment options 3.5.1 Sample collection and characteristics
Three different textile wastewater factories were applied as database test for TexTreat. The characteristics of raw and biotreated wastewater for the different factories are shown in Table 5. All factories show different characteristics in wastewater which are ideal for the purpose of validation.
Table 5 The characteristics of raw and biotreated wastewater for the validation purpose Kim fashion knitwear Raw Colour TOC COD 486 138 563 Biotreated 326 23.3 134 Pacific peninsula textile Raw 1744 195 676 Biotreated 1200 41.4 154 Ramatex textiles industrial Raw 1119 236 689 Biotreated 680 46.12 136
Parameter
Note: All unites mg/l except colour PtCo.
3.5.2 Software validation
Based on the characteristics obtained from the biotreated process in textile wastewaters of different industries, TexTreat had recommended different methods of advanced oxidation process to be used. Based on the characteristics of biotreated wastewater of Kim Fashion Knitwear, Pacific Peninsula Textile and Ramatex Textile Industrial, 0.25 ml/l of H2O2/UV, 1.5 ml/l of H2O2/134 mg/l of O3/UV and 0.75 ml/l of H2O2/50 mg/l of O3/UV were recommended, respectively. Figure 11 shows the predicted and lab results of colour obtained using the recommended methods. Lab result shows that obtained colour were 0, 41 and 0 PtCo after 60 min while simulated results using TextTreat were 12, 26 and 18 PtCo for Kim Fashion Knitwear, Pacific Peninsula Textile and Ramatex Textile Industrial, respectively. The difference in terms of removal between TexTreat predicted results and lab results was lower than 4%.
Figure 11 TexTreat and lab result of colour value
1400 1200 1000 Colour PtCo 800 600 400 200 0 -200 0 15 30 45 60 75 Kim software result Kim lab result Pacific software result Pacific lab result Ramatex software result Ramatex lab result
Retention Time (min)
Development of a software to simulate effluent characteristics
97
As shown in Figure 12, lab results of the TOC of biotreated textile wastewater after advanced oxidation of Kim Fashion Knitwear, Pacific Peninsula Textile and Ramatex Textile Industrial were 8, 16 and 18 mg/l while the predicted results were 9, 14 15 mg/l, respectively. Figure 13 shows the lab and simulated results using TextTreat of COD. After 60 min, COD of the lab results were 31, 34 and 36 mg/l whereas the simulated results were 20, 19 and 20 mg/l for Kim Fashion Knitwear, Pacific Peninsula Textile and Ramatex Textile Industrial respectively. However, the differences in terms of removal of TOC and COD between TexTreat and lab results were lower than 6% and 11%, respectively.
Figure 12 TexTreat and lab result of TOC value
50 45 40 35 30 25 20 15 10 5 0 0 15 30 45 Retention Time (min) Kim software result Kim lab result Pacific software result Pacific lab result Ramatex software result Ramatex lab result
TO C (m g/l)
60
75
Figure 13 TexTreat and lab result of COD value
180 160 140 COD (mg/l) 120 100 80 60 40 20 0 0 10 20 30 40 50 60 70 Retention Time (min) Kim software result Kim lab result Pacific software result Pacific lab result Ramatex software result Ramatex lab result
4
Conclusions
Regression equations related to the formulation of the process treatment options were successfully developed into a software code (TexTreat). The results have shown that treatment process options were successfully simulated using TexTreat. The difference between the lab results and TexTreat predicted results were less than 11%. It was also shown that the applied GUI and algorithm were simple and therefore will be useful to aid
98
A. Al-Kdasi et al.
in training of wastewater technicians. While it is understandable that the characteristics of textile wastewater are statistically more varied from parameters presented here, it is fair to imply that prediction of COD, TOC and colour effluent parameters using TexTreat will significantly aid in decision support regarding the type and level of oxidative treatment most suitable to treat textile wastewater. TexTreat could also be used to predict the capability of the combined advanced oxidation process to treatment plants of any factory textile plants in Malaysia.
References
Antonio, M., Santiago, E. and Gabriele, S. (1997) ‘How and why combine chemical and biological processes for wastewater treatment’, Wat. Sci. Tech., Vol. 35, pp.321–327. Arslan, I.A. and Isil, A.B. (2001) ‘Advanced oxidation of raw and biotreated textile industry wastewater with O3, H2O2/UV-C and their sequential application’, J. Chemical Technology and Biotechnology, Vol. 76, pp.53–60. Arslan, I.A., Isil, A.B. and Detlef, W.B. (2002) ‘Advanced oxidation of a reactive dyebath effluent: comparison of O3, H2O2/UV-C and TiO2/UV-A processes’, Water Research, Vol. 36, pp.1143–1154. Azbar, N., Yonar, T. and Kestioglu, K. (2004) ‘Comparison of various advanced oxidation processes and chemical treatment methods for COD and colour removal from a polyester and acetate fibre dying effluent’, Chemosphere, Vol. 55, pp.35–43. American Public Health Association (APHA), American Water Works Association and Water Environmental Federation (1998) Standard Methods for the Examination of Water and Wastewater, Washington. American Public Health Association (APHA), American Water Works Association and Water Pollution Control Federation (1985) Standard Methods for the Examination of Water and Wastewater, Washington Beltran, F., Gonzalez, M. and Gonzalez, J. (1997) ‘Industrial wastewater advanced oxidation, Part 1 UV radiation in the presence and absence of hydrogen peroxide’, Water Research, Vol. 31, pp.2405–2414. Bernard, O., Hadj-Sadok, Z., Dochain, D., Genovesi, A. and Steyer, J.P. (2001) ‘Dynamical model development and parameter identification for an anaerobic wastewater treatment process’, Biotechnology and Bioengineering, Vol. 75, pp.424–438. Bes-Piá, A., Mendoza-Roca, J.A., Roig-Alcover, L., Iborra-Clar, A., Iborra-Clar, M.I. and Alcaina-Miranda, M.I. (2003) ‘Comparison between nonfiltration and ozonation of biologically treated textile wastewater for its reuse in the industry’, Desalination, Vol. 157, pp.81–86. Dae-Hee, A., Won-Seok, C. and Tai-Il, Y. (1999) ‘Dyestuff wastewater treatment using chemical oxidation, physical adsorption and fixed bed biofilm process’, Process Biochemistry, Vol. 34, pp.429–439. Georgiou, D., Melidis, P., Aivasidis, A. and Gimouhopoulos, K. (2002) ‘Degradation of azo-reactive dyesby ultraviolet radiation in the presence of hydrogen peroxide’, Dyes and Pigments, Vol. 52, pp.69–78. Hamed, M.M., Khalafallah, M.G. and Hassanien, E.A. (2004) ‘Prediction of wastewater treatment plant performance using artificial neural networks’, Environmental Modelling & Software, Vol. 19, pp.919–928. Hidalgo, D., Irusta, R., Martinez, L., Fatta, D. and Papadopoulos, A. (2007) ‘Development of a multi-function software decision support tool for the promotion of the safe reuse of treated urban wastewater’, Desalination, Vol. 215, pp.90–103.
Development of a software to simulate effluent characteristics
99
Joksimovic, D., Kubik, J., Hlavinek, P., Savic, D. and Walters, G. (2006) ‘Development of an integrated simulation model for treatment and distribution of reclaimed water’, Desalination, Vol. 188, pp.9–20. Lin, S.H. and Peng, C.F. (1996) ‘Continuous treatment of textile wastewaters by combined coagulation, electrochemical oxidation and activated sludge’, Water Res., Vol. 30, pp.587–593. Maria, K., Adamantini, K., Katia, L. and Georgios, P. (2004) ‘Integrated aerobic biological treatment and chemical oxidation with Fenton’s reagent for the processing of green table olive wastewater’, Process Biochemistry, Vol. 39, pp.1653–1660. Mariana, N., Ilie, S., Ayfer, Y. and Antonius, K. (2002) ‘Kinetics of decolorization and mineralization of reactive azo dyes in aqueous solution by UV/H2O2 oxidation’, Dyes and Pigments, Vol. 53, pp.93–99. Masten, S. and Davies, H. (1993) ‘The use of ozone and other strong oxidation for hazardous waste management’, in Nriagen, J.O. and Simmons, M.S. (Eds.): Environmental Oxidants, John Wiley and Sons, Inc. Nilsun, H.I. (1999) ‘Critical effect of hydrogen peroxide in photochemical dye degradation’, Water Research, Vol. 33, pp.1080–1084. Pagga, U. and Brown, D. (1986) ‘The degradation of dyestuffs: part II behaviour of dyestuffs in aerobic biodegradation tests’, Chemosphere, Vol. 15, pp.479–491. Paprowicz, J. and Slodczyk, S. (1988) ‘Application of biologically activated sorptive columns for textile wastewater treatment’, Environ Technol Lett., Vol. 9, pp.271–80. Shu, H-Y. and Chang, M-C. (2005) ‘Pilot scale annular plug flow photoreactor by UV/H2O2 for the decolorization of azo dye wastewater’, Journal of Hazardous Materials B, Vol. 125, pp.244–251. Shu, H-Y., Chang, M-C. and Huan-Jung, F. (2004) ‘Decolorization of azo dye acid black 1 by the UV/H2O2 process and optimization of operating parameters’, Journal of Hazardous Materials B, Vol. 113, pp.201–208. Straley, J. (1984) ‘Selected bacterial cultures for textile wastewaters’, Am Dyestuff Rep., Vol. 73, pp.46–49. Tanja, K., Alenka, M.L.M. and Darinka, B.V. (2003) ‘Comparison of H2O2/UV, H2O2/O3 and H2O2/Fe2+ processes for the decoloristion of vinylsulphone reactive dyes’, Dyes and Pigments, Vol. 58, pp.245–252. Ulusoy, B. (2004) ‘Application of Box-Wilson experimental design method for the photodegradation of textile dyestuff with UV/H2O2 process’, Dyes and Pigments, Vol. 60, pp.187–195. Wenzel, A., Gahr, A. and Niessner, R. (1999) ‘TOC-removal and degradation of pollutants in leachate using a thin-film photoreactor’, Water Research, Vol. 33, pp.937–946. Wilmott, N., Guthrie, J. and Nelson, G. (1998) ‘The biotechnology approach to colour removal from textile effluent’, JSDC, Vol. 114, pp.38–41. Yung-Shuen, S. and Deng-Kae, W. (2002) ‘Development of photoreactor design equation for the treatment of dye wastewater by UV/H2O2 process’, Journal of Hazardous Materials B, Vol. 89, pp.267–277.
Int. J. Process Systems Engineering, Vol. 1, No. 1, 2009
Development of a software to simulate effluent characteristics of textile wastewater treatment Adel Al-Kdasi
Department of Chemical and Environmental Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia and Faculty of Agriculture, Sana’a University, Yemen E-mail: [email protected]
Azni Idris
Waste Technology Centre, Department of Chemical and Environmental Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia E-mail: [email protected]
Luqman Chuah Abdullah
Department of Chemical and Environmental Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia E-mail: [email protected]
Mohanad El-Harbawi*
Department of Chemical Engineering, Universiti Teknologi Petronas, Bandar Seri Iskandar, 31750 Tronoh, Perak, Malaysia E-mail: [email protected] *Corresponding author
Mogeeb Alzokry
Department of Software Engineering, Faculty of Computer Science and Information Technology, Universiti Malaya, 50301, Kuala Lumpur, Malaysia E-mail: [email protected]
Copyright © 2009 Inderscience Enterprises Ltd.
Development of a software to simulate effluent characteristics
83
Chun-Yang Yin
Faculty of Chemical Engineering, University Technology Mara, 40430 Shah Alam, Selangor, Malaysia E-mail: [email protected]
Abstract: The composition of textile wastewaters is extremely varied due to the large spectrum of dyes and chemicals used in wet process resulting in greater variability in the success of different treatment processes. A software was developed to determine the required advanced oxidation process after biological treatment and to evaluate the possibility of applying advanced oxidation processes to any textile treatment plant. Regression equations related to the formulation of the process treatment options were successfully developed into a software program called TexTreat with integrated graphic user interface (GUI) component. TexTreat has the ability to determine the best option of advanced oxidation processes (AOPs) and predict characteristics of wastewater discharge in different retention times with an overall simulation accuracy of more than 89%. Validation of the process treatment options and TexTreat display their applicability to utilise with different textile wastewater plants. Keywords: textile wastewater; advanced oxidation process; AOP; software. Reference to this paper should be made as follows: Al-Kdasi, A., Idris, A., Chuah Abdullah, L., El-Harbawi, M., Alzokry, M. and Yin, C-Y. (2009) ‘Development of a software to simulate effluent characteristics of textile wastewater treatment’, Int. J. Process Systems Engineering, Vol. 1, No. 1, pp.82–99. Biographical notes: Adel Al-Kdasi is currently an Assistant Professor in the Faculty of Agriculture-Sana’a University. Prior to this, he obtained his PhD from the Department of Chemical and Environmental Engineering, Universiti Putra Malaysia (UPM). He has published several papers in textile treatment field. Azni Idris received his PhD in Environmental Engineering from the University of Newcastle upon Tyne, UK. He is currently the Head of Department of Chemical and Environmental Engineering at Faculty of Engineering, Universiti Putra Malaysia. He is the Director of Technology Commercialisation, University Business Center, University Putra Malaysia (UPM). He is also the Director of Waste Technology Center, UPM. As an active researcher, he had published more than 100 publications in international journals and proceedings. In addition, he was awarded for research excellence and his contribution in the production of biofilter [holder of patent on organic waste treatment process (BioFil sysytem)]. Luqman Chuah Abdullah is currently an Associate Professor in the Department of Chemical and Environmental Engineering, Universiti Putra Malaysia (UPM). He has also served as Head of Research Laboratory in the Institute of Tropical Forest and Forest Products (INTROP), UPM. As an active researcher, he had published 160 publications on journals and proceedings, of which many of them are internationally cited and reputable in chemical and environmental engineering. In addition, he won several awards in teaching and research, including Young Engineer Award from the Institution of Engineers Malaysia (IEM) in year 2006.
84
A. Al-Kdasi et al.
Mohanad El-Harbawi received his Doctorate in Chemical and Environmental Engineering from the University Putra Malaysia in 2006. He is part of lecturing team at the Department of Chemical Engineering at Universiti Teknologi PETRONAS (UTP). Currently, he is working diligently in the development of a GIS-based air pollution dispersion modelling software. He has developed several softwares for chemical risk assessment and air pollution assessment. Furthermore, he has several publications including ISI-indexed journal articles and conference proceedings. In addition, he is also a reviewer for Journal of Engineering Science and Technology (JESTEC). Mogeeb Alzokry received his BSc in Software Engineering from Technology University – Iraq in 2002. In 2003, he was appointed as a Staff at Taiz University in the Faculty of Engineering. He is currently pursuing his PhD at the University of Malaya. To date, he has published several papers in indexed journals and conference proceedings. Chun-Yang Yin received his PhD in Chemical Engineering from the University of Malaya and was appointed as a Visiting Scholar at the prestigious Ivy League Institution, Columbia University from December 2008 to March 2009. He has over 60 publications to date including ISI-indexed journal articles and conference proceedings. He is the recipient of numerous universities, national and international-level awards including the IEM Tan Sri Raja Zainal Prize in 2005 and 2009 as well as listed in the 25th Silver Anniversary Edition of Marquis Who’s Who in the World. He is currently a Reviewer for several Elsevier journals and is on the Editorial Board of Malaysian Journal of Chemical Engineering.
1
Introduction
Several conventional methods are used to treat textile dye wastewaters such as biological treatment, chemical oxidation, coagulation, adsorption and filtration. However, the efficiencies of these methods depend strongly on the types of dye in wastewater and concentration of contaminants. Colour removal from textile effluents in aerobic biological treatment is not an effective process since the biodegradation products are toxic to the organisms used in the process and these result in various problem such as sludge bulking, rising sludge and pin-point floc formation (Straley, 1984; Paprowicz and Slodczyk, 1988; Pagga and Brown, 1986; Lin and Peng, 1996; Antonio et al., 1997; Wilmott et al., 1998). According to Antonio et al. (1997), biological treatments are reliable, but there are certain substances which are unable to deal with. Thus, there are a lot of combinations of chemical oxidation and biological treatment which can be arranged for organic removal from toxic wastewater. However, bio-chemical treatment of textile wastewater effluents that contain dyes and their hydrolysis products can be a cost-effective alternative when the effluents are pretreated chemically prior to treatment in the biological unit (Lin and Peng, 1996). Physico-chemical methods such as coagulation/flocculation, activated carbon adsorption and reverse osmosis techniques have been developed in order to remove the colour (Dae-Hee et al., 1999; Bes-Piá et al., 2003; Maria et al., 2004). However, these methods can only transfer the contaminants from wastewater to solid waste leaving the problem essentially unsolved (Mariana et al., 2002; Georgiou et al., 2002; Arslan et al.,
Development of a software to simulate effluent characteristics
85
2002). Therefore, special attention should be focused on the techniques that lead to the complete destruction of the dye molecules. Advanced Oxidation Processes (AOPs) is one of these developing processes that has been used to generate hydroxyl free radicals by different techniques to destroy the contaminants in textile wastewater. Ozone (O3), hydrogen peroxide (H2O2) and UV irradiation are recently used to accelerate the generation of hydroxyl radical (HO•). This strong oxidant can destroy compounds that can not be readily oxidised by conventional oxidants such as ozone, oxygen and chlorine. In addition, AOPs do not generate chemical sludge and toxicity of wastewater is generally reduced. Several studies have been reported about the successful application of the AOPs (Nilsun, 1999; Arslan and Isil, 2001; Arslan et al., 2002; Georgiou et al., 2002; Yung-Shuen and Deng-Kae, 2002; Tanja et al., 2003; Azbar et al., 2004; Ulusoy, 2004; Shu et al., 2004; Shu and Chang, 2005). Previous studies were carried out within a certain range of different kind of synthetic dyes and synthetic textile wastewater in the batch mode. Few studies were reported in the literature that deal with the actual wastewater due to their inability to cope with the complex background matrices and high pollution load normally encountered in real wastewater (Beltran et al., 1997; Wenzel et al., 1999). On the other hand, extraordinary high oxidant doses of chemicals make the industrial scale stand-alone and the application of chemical unfeasible from the economic point of view (Masten and Davies, 1993; Beltran et al., 1997). As such, an integrated system of biological treatment and AOPs would mean a cheaper option for total organic degradation and colour removal from the textile wastewater. Having different, easily controlled and successful processes that can deal with the different strength of textile wastewater is the best way to ensure efficient colour and organic pollutants removal from textile wastewater. Previous related studies were focused on development of models to predict performance of wastewater treatment systems without integration of graphic user interface (GUI) to enhance user-friendly feature of the model. Examples of such reported studies include Bernard et al. (2001), Hamed et al. (2004) and Joksimovic et al. (2006) which focused on development of new models for wastewater parameters estimation as applied for anaerobic, urban domestic and reclaimed wastewater respectively. In a recent study on software development, Hidalgo et al. (2007) developed a multi-criteria analysis user friendly software to assist responsible authorities in determining the most efficient solutions in terms of health and safety for the agricultural reuse of the produced effluent. From our understanding, the concept or idea to develop a software with GUI to determine the best advanced oxidation process after biological treatment and predict characteristics of treated textile wastewater has not been considered in previous studies. As such, the objectives of this study were to develop software (designated as TexTreat) to determine the required advanced oxidation process after biological treatment and evaluate the possibility of applying AOPs to any textile treatment plant. Development of TexTreat was divided into three distinct stages. The first stage assigned a reaction order for each category which was determined by using the integration method. Reaction rate coefficient (rate constant) was evaluated by substitution using the test data at different retention time for colour, total organic carbon (TOC) and chemical oxygen demand (COD). The second stage assigned the best fitting regression equations between different values of the parameters and their reaction rate coefficients were determined at different retention times using SPSS software. The final stage was to write the program for the TexTreat software.
86
A. Al-Kdasi et al.
2
Materials and methods
2.1 Textile wastewater collection
To achieve the formulation treatment option, samples of textile wastewater with different concentrations were collected and studied. Each sample was studied separately using integrated system of biological and advanced oxidation process (data not shown). The textile wastewater was collected every five days from a local manufacturing factory effluents of Kim Fashion Knitwear Sdn Bhd located in Senawang, Negeri Sembilan, Malaysia. Wastewater was produced from several finishing and dyeing units, which were used to dye fabrics, hanks and socks of different natural or synthetic fibres and mixture of both. Three different textile wastewater factories were used to validate the TexTreat application and the formulation of process treatment options. These three factories were Kim Fashion Knitwear (M) Sdn Bhd located in Senawang, Negeri Sembilan, Pacific Peninsula Textile Sdn Bhd located in Johor Bahru and Ramatex Textiles Industrial Sdn Bhd located in Batu Pahat, Johor Bahru.
Figure 1 Design configuration of bio-photochemical reactor
Development of a software to simulate effluent characteristics
87
2.2 Experimental model setup
The laboratory setup consisted of activated sludge process (extended aeration) followed by AOP. The reactor was designed to be operated in continuous mode. The laboratory-scale model reactor used in this study is shown in Figure 1.
2.2.1 Activated sludge model
The activated sludge process reactor consisted of a feeding tank, aeration tank and settling tank made of polyvinyl chloride (PVC). Raw wastewater pH in the feeding tank was adjusted to pH 7 using 1 M hydrochloric acid and mixed continuously using a motorised stirrer (Eurostar Digital, Germany) at a rate of 50 rpm. A continuous flow rate of 1.26 l/h using a peristaltic pump (Heidolph PD5006, Germany) was used. Complete mixing and aeration were assured by the continuous flow of air into the aeration tank. The dissolved oxygen concentration in aeration tank was kept between 2 to 3 mg/l. The aeration tank was connected to the settling tank to remove suspended solids. Biotreated effluent was subsequently discharged through a plastic pipe to the advanced oxidation process reactor. All experiments were conducted at room temperature range of 25–28°C. Samples for analysis were collected from feeding tank, discharge point after aeration tank and discharge point after AOP. Microorganisms of the aeration tank were acclimatised to receive high-level of colour and organic matter in the wastewater using actual textile wastewater without any pretreatment for three months. The system was operated at F/M ratio of around 0.05 (g BOD /g VSS.d) and dissolved oxygen concentrations from 2 to 3 mg/l.
2.2.2 Advanced oxidation model
AOP reactor was made of 304 stainless steel with 2.5 litre capacity. The reactor was cylindrical and equipped with a UV lamp (wavelength, λ = 254 nm). Both the UV-C and quartz lamp housings were centred in the reactor tube. Five outlet discharge points were pointed corresponding to different retention times in the system. O3 generator (OWA 350) with an O3 production rate of 350 mg/hr was used for this experiment. The O3 was produced by natural intake of air from surrounding and was bubbled into reactor by diffusers. The applied O3 concentrations for this experiment were 183, 152 and 101 mg/L. Ozone was introduced into the bottom of the reactor through two diffuser points. The exhaust gas was vented from the top of the reactor passing through two bottles of potassium iodide solution (KI) and 2% absorption solution was then vented into a laboratory hood. Different concentrations of hydrogen peroxide (30%) were applied using a peristaltic pump. Samples were collected from the sampling point at regular time intervals (15, 30, 60, 90 and 120 minutes) and kept for further analyses.
2.3 Analytical methods
True colour (PtCo) of the samples was measured using a spectrophotometer (HACH DR 2500, USA) calibrated according to standard platinum-cobalt method. Prior to colour measurement, both the influent and effluent were filtered using membrane filter paper
88
A. Al-Kdasi et al.
with a pore size of 0.45 µm (Whatman, Japan). The mixed liquor suspended solids (MLSS), total suspended solids (TSS), COD and biochemical oxygen demand (BOD) measurements were determined according to standard methods for examination of water and wastewater (APHA et al., 1998). Ozone concentration in feed gas was determined by KI-starch titration method (APHA et al., 1985). The TOC concentration in solution was determined using TOC analyser (Shimadzu 5050, Japan). Wastewater pH was measured by means of a laboratory multi-meter (WTW, InoLab Multi Level 1, Germany). The average values of the results were obtained by three repeated experiments. Statistical software (SPSS, Version 12) and standard Microsoft Visual Basic 6.0 for windows computer program was used for the regression and developing of the software, respectively.
3
Results and discussion
3.1 Formulation treatment options
Samples of textile wastewater with different concentrations were studied to obtain formulation process treatment option for textile wastewater. Biotreated textile wastewater was successfully categorised into four groups; corresponding to these categories, different AOPs were formulated for use. Table 1 shows the categories of biotreated textile wastewater and the best methods of advanced oxidation process for each category. From the formulation treatment options, a software (TexTreat) was developed to assist operation in selecting the most appropriate AOP for the textile wastewater and also to estimate the discharge effluent concentration.
Table 1 Categories of biotreated textile wastewater and best selected methods of advanced oxidation Condition Colour ≤ 400 and TC ≤ 80 mg/l Colour ≤ 400 and TC > 80 mg/l Category 2 Category 3 Category 4 400 < colour ≤ 800 800 < colour ≤ 1200 Colour > 1200 Best method of AOPS 0.25 ml/l H2O2/UV 0.5 ml/l H2O2/UV 0.75 ml/l H2O2/UV/50 mg/l O3 1.5 ml/l H2O2/UV/134 mg/l O3 2 ml/l H2O2/UV/183 mg/l O3
Category Category 1
3.2 Determination of the reaction order and reaction rate coefficient
Integration method was used to determine the order of the reaction for the selected process. Three reaction orders (zero, first and second order) were plotted. As shown in Figures 2–4, the experimental data fitted the second-order (straight line for the parameters removal) the best among the three orders. Only the second order curve was shown for reason of brevity. As shown in Figure 2, the second order give the straight line for the colour removal where R2 (coefficient of determination), were higher than R2 for plots C and –log[C/Co] (zero and first order).
Development of a software to simulate effluent characteristics
Figure 2 Colour graphical analysis for the determination of reaction order for Category 1
0.18 0.16 0.14 0.12 1/Colour 0.1 0.08 0.06 0.04 0.02 0 0 15 30 45 60 75 90 105 120 135
89
Category 1
R2 = 0.9961
Retention Time (min)
Figure 3 TOC graphical analysis for the determination of reaction order for Category 1
0.14 0.12 0.1 0.08 0.06 0.04 0.02 0
Category 1
R2 = 0.6141
1/T O C
0
15
30
45
60
75
90
105
120
135
Retention Time (min)
Figure 4 COD graphical analysis for the determination of reaction order for Category 1
0.05 0.045 0.04 0.035 1/C O D 0.03 0.025 0.02 0.015 0.01 0.005 0 0 15 30 45 60 75 90 105 120 135 Retention Time (min)
Category 1
R2 = 0.9194
90
A. Al-Kdasi et al.
R2 for the second, third and fourth categories are shown in Table 2. Reaction rate coefficients were determined for colour at the different retention times for the different categories. Table 2 shows reaction rate coefficients for colour at the different retention time for the different categories. These reaction rate coefficients were utilised in TexTreat to predict the residual of parameters undergoing AOPs.
Table 2 Reaction rate coefficient for colour removal at different retention times Reaction rate coefficient 15 min 0.0015205 0.0010198 0.0004342 0.0000977 30 min 0.0014929 0.0009644 0.0005370 0.0001482 60 min 0.0012348 0.0009576 0.0006033 0.0002188 R2 0.9961 0.9769 0.9912 0.9614
Categories Category 1 Category 2 Category 3 Category 4
The reaction orders for TOC and COD were determined and the reaction rate coefficients were evaluated as well. As shown in Figures 3–4, TOC and COD removal data fit the second-order quite well in which R2 was higher than R2 for zero and first-order. R2 for the second, third and fourth categories are shown in Table 3.
Table 3 Reaction rate coefficients for TOC and COD removals Reaction rate coefficient 15 min 0.002187 0.0014545 0.0013813 0.0003819 0.0008333 0.0006871 0.0006030 0.0001502 30 min TOC Category 1 Category 2 Category 3 Category 4 Category 1 Category 2 Category 3 Category 4 0.0018628 0.0010726 0.0010085 0.0004395 COD 0.0004605 0.0004825 0.0006792 0.000233 0.0003293 0.0004198 0.000621 0.0003635 0.919 0.984 0.893 0.955 0.0009978 0.0009206 0.0008118 0.0004803 0.614 0.961 0.981 0.998 60 min R2
Categories
The reaction rate coefficients for TOC and COD were evaluated at different retention time for all the categories. Table 3 shows the reaction rate coefficients for TOC and COD at different retention time for all the different categories. The entire reaction rate coefficients were used to get the relationship between the reaction rate coefficients and parameters and then used in the TexTreat to determine the values of the parameters after the different retention time of AOPs. Once the reaction rate and coefficient are known, the second integrated form could be used. The concentration of the sample could be determined at selected retention time treatment of AOPs if the initial concentration of parameter, retention time and reaction rate coefficient are known. The integrated form for-second order can be written as follows:
Development of a software to simulate effluent characteristics
1 1 − = kt C CO 1 1 = + kt C CO C = CO / (1 + k * t *CO )
91 (1) (2) (3)
where C = concentration of constituent CO = concentration of constituent at zero time k = reaction rat coefficient, (mg/L3) n–1/min t = time
3.3 Determination of the best fitting regression equation
Relationship between different values of the parameters and their rate constants were determined at different retention times. For each parameter, three formulas were determined (first for the 15 min and second for 30 min and the last for 60 min). SPSS was used to determine the best fitting curve and the regression equations. The second-order was the best equation that adequately describes the relationship between the concentration of parameters and 1/k. The coefficients of determinations (R2) were more than 0.9 and the standard errors were around zero with the second regression equations. This indicated all the input data fitted the second-order very well and that since all the R2 values for all regression coefficients were all very close to unity, it implied that all these equations are applicable for data input. Figures 5–7 show the regression equations fitted between different parameters at 15 min and 1/k using Microsoft Excel. All regression fitting were similar to SPSS regression. The regression equations at the different retention times were written in the TexTreat code. Table 4 shows the regression equations and R2 at different retention time.
Figure 5 Relationship between 1/k and colour at 15 min
12000 10000 8000 1/K 6000 4000 2000 0 0 1000
y = 5E-05x 2 + 1.8787x - 148.85 R2 = 0.9982
Data Equation plot
2000
3000 Colour
4000
5000
6000
92
A. Al-Kdasi et al.
Figure 6 Relationship between 1/k and TOC at 15 min
3000 2500 2000 1/k 1500 1000 500 0 10 20 30 40 TOC 50 60 70 80 Data Equation plot y = 0.6153x2 - 21.382x + 685.15 R2 = 0.9943
Figure 7 Relationship between 1/k and COD at 15 min
7000 6000 5000 1/K 4000 3000 2000 1000 0 50 70 90
y = 0.2736x2 - 44.725x + 3103.2 R2 = 0.9987
Data Equation plot
110
130
150 COD
170
190
210
230
250
Table 4
Regression equations Time 15 min 30 min 60 min Regression equations Y=(5E-05x2 + 1.8787x – 148.85)–1 Y=(–2E-05x + 1.4466x + 103.5)
2 –1
Parameter Colour
R2 0.998 0.999 0.998 0.994 0.985 1 0.999 0.989 0.999
Y=(–4E-05x2 + 1.0801x + 382.33)–1 Y=(0.6153x2 – 21.382x + 685.15)–1 Y=(0.1339x2 + 17.1x + 197.21)–1 Y=(0.1967x2 + 0.44x + 912.68)–1 Y=(0.2736x2 – 44.725x + 3103.2)–1 Y=(0.3133x2 – 80.454x + 6706.6)–1 Y=(0.3196x2 – 98.492x + 8864.7)–1
TOC
15 min 30 min 60 min
COD
15 min 30 min 60 min
Development of a software to simulate effluent characteristics
93
3.4 Formulation of computer algorithm
All equations that determined the removal of colour, TOC and COD at the different retention time for the parameters at different retention times were written in TexTreat code. The program was written in standard Microsoft Visual Basic 6.0 and distributed in the source code. The computational of the mathematical models for the result estimation was calculated by using VB program code. The main interface has three buttons (enter, information and exit) as shown in Figure 8. The enter button allows the user to go into the decision support interface for textile wastewater treatment. The information button provides the user detailed description about TexTreat. It is designed using front page and HTML (hyper text markup language). The exit button is used to close the TexTreat.
Figure 8 The main and submenus interfaces (see online version for colours)
94
A. Al-Kdasi et al.
The decision support interface is used to insert the characteristics of textile biotreated wastewater. The decision support interface of TexTreat contains three main stages. The different stages and instructions to use the TexTreat are shown in Figure 9. In the first stage, input data stage, the characteristics of the biotreated textile wastewater were inserted to the processes of running the TexTreat. In the input data stage, different buttons and text boxes for different parameters of characteristics of textile wastewater and the run (process) of the program were used. In the second stage, the required treatment method is achieved in text box. This stage also has buttons to estimate the results by using the determination method. In the final stage, the estimated results using the determination method appear at different text boxes. The results estimated appear for different parameters (colour, TOC and COD) in different retention time (15, 30 and 60 min). Three different separated columns were used to show the result for the different retention time for the different parameters.
Figure 9 Flowchart of instructions of using the TexTreat (see online version for colours)
Figure 10 shows the TexTreat algorithm. An algorithm is basically a succession of instructions or a process used for calculation and data processing. At the beginning of the algorithm sequence, data pertaining to characteristics of wastewater are put into the TexTreat. After commencement of a run, TexTreat will basically determine the level of hydrogen peroxide/UV needed based on two main parameters, namely, colour and total
Development of a software to simulate effluent characteristics
95
carbon. Subsequent to initiation of ‘treatment’, the characteristics of the effluent will be predicted. The TC and colour values for the algorithm are obtained from Table 1.
Figure 10 Flow diagram of the TexTreat algorithm
96
A. Al-Kdasi et al.
3.5 Validation of TexTreat and process treatment options 3.5.1 Sample collection and characteristics
Three different textile wastewater factories were applied as database test for TexTreat. The characteristics of raw and biotreated wastewater for the different factories are shown in Table 5. All factories show different characteristics in wastewater which are ideal for the purpose of validation.
Table 5 The characteristics of raw and biotreated wastewater for the validation purpose Kim fashion knitwear Raw Colour TOC COD 486 138 563 Biotreated 326 23.3 134 Pacific peninsula textile Raw 1744 195 676 Biotreated 1200 41.4 154 Ramatex textiles industrial Raw 1119 236 689 Biotreated 680 46.12 136
Parameter
Note: All unites mg/l except colour PtCo.
3.5.2 Software validation
Based on the characteristics obtained from the biotreated process in textile wastewaters of different industries, TexTreat had recommended different methods of advanced oxidation process to be used. Based on the characteristics of biotreated wastewater of Kim Fashion Knitwear, Pacific Peninsula Textile and Ramatex Textile Industrial, 0.25 ml/l of H2O2/UV, 1.5 ml/l of H2O2/134 mg/l of O3/UV and 0.75 ml/l of H2O2/50 mg/l of O3/UV were recommended, respectively. Figure 11 shows the predicted and lab results of colour obtained using the recommended methods. Lab result shows that obtained colour were 0, 41 and 0 PtCo after 60 min while simulated results using TextTreat were 12, 26 and 18 PtCo for Kim Fashion Knitwear, Pacific Peninsula Textile and Ramatex Textile Industrial, respectively. The difference in terms of removal between TexTreat predicted results and lab results was lower than 4%.
Figure 11 TexTreat and lab result of colour value
1400 1200 1000 Colour PtCo 800 600 400 200 0 -200 0 15 30 45 60 75 Kim software result Kim lab result Pacific software result Pacific lab result Ramatex software result Ramatex lab result
Retention Time (min)
Development of a software to simulate effluent characteristics
97
As shown in Figure 12, lab results of the TOC of biotreated textile wastewater after advanced oxidation of Kim Fashion Knitwear, Pacific Peninsula Textile and Ramatex Textile Industrial were 8, 16 and 18 mg/l while the predicted results were 9, 14 15 mg/l, respectively. Figure 13 shows the lab and simulated results using TextTreat of COD. After 60 min, COD of the lab results were 31, 34 and 36 mg/l whereas the simulated results were 20, 19 and 20 mg/l for Kim Fashion Knitwear, Pacific Peninsula Textile and Ramatex Textile Industrial respectively. However, the differences in terms of removal of TOC and COD between TexTreat and lab results were lower than 6% and 11%, respectively.
Figure 12 TexTreat and lab result of TOC value
50 45 40 35 30 25 20 15 10 5 0 0 15 30 45 Retention Time (min) Kim software result Kim lab result Pacific software result Pacific lab result Ramatex software result Ramatex lab result
TO C (m g/l)
60
75
Figure 13 TexTreat and lab result of COD value
180 160 140 COD (mg/l) 120 100 80 60 40 20 0 0 10 20 30 40 50 60 70 Retention Time (min) Kim software result Kim lab result Pacific software result Pacific lab result Ramatex software result Ramatex lab result
4
Conclusions
Regression equations related to the formulation of the process treatment options were successfully developed into a software code (TexTreat). The results have shown that treatment process options were successfully simulated using TexTreat. The difference between the lab results and TexTreat predicted results were less than 11%. It was also shown that the applied GUI and algorithm were simple and therefore will be useful to aid
98
A. Al-Kdasi et al.
in training of wastewater technicians. While it is understandable that the characteristics of textile wastewater are statistically more varied from parameters presented here, it is fair to imply that prediction of COD, TOC and colour effluent parameters using TexTreat will significantly aid in decision support regarding the type and level of oxidative treatment most suitable to treat textile wastewater. TexTreat could also be used to predict the capability of the combined advanced oxidation process to treatment plants of any factory textile plants in Malaysia.
References
Antonio, M., Santiago, E. and Gabriele, S. (1997) ‘How and why combine chemical and biological processes for wastewater treatment’, Wat. Sci. Tech., Vol. 35, pp.321–327. Arslan, I.A. and Isil, A.B. (2001) ‘Advanced oxidation of raw and biotreated textile industry wastewater with O3, H2O2/UV-C and their sequential application’, J. Chemical Technology and Biotechnology, Vol. 76, pp.53–60. Arslan, I.A., Isil, A.B. and Detlef, W.B. (2002) ‘Advanced oxidation of a reactive dyebath effluent: comparison of O3, H2O2/UV-C and TiO2/UV-A processes’, Water Research, Vol. 36, pp.1143–1154. Azbar, N., Yonar, T. and Kestioglu, K. (2004) ‘Comparison of various advanced oxidation processes and chemical treatment methods for COD and colour removal from a polyester and acetate fibre dying effluent’, Chemosphere, Vol. 55, pp.35–43. American Public Health Association (APHA), American Water Works Association and Water Environmental Federation (1998) Standard Methods for the Examination of Water and Wastewater, Washington. American Public Health Association (APHA), American Water Works Association and Water Pollution Control Federation (1985) Standard Methods for the Examination of Water and Wastewater, Washington Beltran, F., Gonzalez, M. and Gonzalez, J. (1997) ‘Industrial wastewater advanced oxidation, Part 1 UV radiation in the presence and absence of hydrogen peroxide’, Water Research, Vol. 31, pp.2405–2414. Bernard, O., Hadj-Sadok, Z., Dochain, D., Genovesi, A. and Steyer, J.P. (2001) ‘Dynamical model development and parameter identification for an anaerobic wastewater treatment process’, Biotechnology and Bioengineering, Vol. 75, pp.424–438. Bes-Piá, A., Mendoza-Roca, J.A., Roig-Alcover, L., Iborra-Clar, A., Iborra-Clar, M.I. and Alcaina-Miranda, M.I. (2003) ‘Comparison between nonfiltration and ozonation of biologically treated textile wastewater for its reuse in the industry’, Desalination, Vol. 157, pp.81–86. Dae-Hee, A., Won-Seok, C. and Tai-Il, Y. (1999) ‘Dyestuff wastewater treatment using chemical oxidation, physical adsorption and fixed bed biofilm process’, Process Biochemistry, Vol. 34, pp.429–439. Georgiou, D., Melidis, P., Aivasidis, A. and Gimouhopoulos, K. (2002) ‘Degradation of azo-reactive dyesby ultraviolet radiation in the presence of hydrogen peroxide’, Dyes and Pigments, Vol. 52, pp.69–78. Hamed, M.M., Khalafallah, M.G. and Hassanien, E.A. (2004) ‘Prediction of wastewater treatment plant performance using artificial neural networks’, Environmental Modelling & Software, Vol. 19, pp.919–928. Hidalgo, D., Irusta, R., Martinez, L., Fatta, D. and Papadopoulos, A. (2007) ‘Development of a multi-function software decision support tool for the promotion of the safe reuse of treated urban wastewater’, Desalination, Vol. 215, pp.90–103.
Development of a software to simulate effluent characteristics
99
Joksimovic, D., Kubik, J., Hlavinek, P., Savic, D. and Walters, G. (2006) ‘Development of an integrated simulation model for treatment and distribution of reclaimed water’, Desalination, Vol. 188, pp.9–20. Lin, S.H. and Peng, C.F. (1996) ‘Continuous treatment of textile wastewaters by combined coagulation, electrochemical oxidation and activated sludge’, Water Res., Vol. 30, pp.587–593. Maria, K., Adamantini, K., Katia, L. and Georgios, P. (2004) ‘Integrated aerobic biological treatment and chemical oxidation with Fenton’s reagent for the processing of green table olive wastewater’, Process Biochemistry, Vol. 39, pp.1653–1660. Mariana, N., Ilie, S., Ayfer, Y. and Antonius, K. (2002) ‘Kinetics of decolorization and mineralization of reactive azo dyes in aqueous solution by UV/H2O2 oxidation’, Dyes and Pigments, Vol. 53, pp.93–99. Masten, S. and Davies, H. (1993) ‘The use of ozone and other strong oxidation for hazardous waste management’, in Nriagen, J.O. and Simmons, M.S. (Eds.): Environmental Oxidants, John Wiley and Sons, Inc. Nilsun, H.I. (1999) ‘Critical effect of hydrogen peroxide in photochemical dye degradation’, Water Research, Vol. 33, pp.1080–1084. Pagga, U. and Brown, D. (1986) ‘The degradation of dyestuffs: part II behaviour of dyestuffs in aerobic biodegradation tests’, Chemosphere, Vol. 15, pp.479–491. Paprowicz, J. and Slodczyk, S. (1988) ‘Application of biologically activated sorptive columns for textile wastewater treatment’, Environ Technol Lett., Vol. 9, pp.271–80. Shu, H-Y. and Chang, M-C. (2005) ‘Pilot scale annular plug flow photoreactor by UV/H2O2 for the decolorization of azo dye wastewater’, Journal of Hazardous Materials B, Vol. 125, pp.244–251. Shu, H-Y., Chang, M-C. and Huan-Jung, F. (2004) ‘Decolorization of azo dye acid black 1 by the UV/H2O2 process and optimization of operating parameters’, Journal of Hazardous Materials B, Vol. 113, pp.201–208. Straley, J. (1984) ‘Selected bacterial cultures for textile wastewaters’, Am Dyestuff Rep., Vol. 73, pp.46–49. Tanja, K., Alenka, M.L.M. and Darinka, B.V. (2003) ‘Comparison of H2O2/UV, H2O2/O3 and H2O2/Fe2+ processes for the decoloristion of vinylsulphone reactive dyes’, Dyes and Pigments, Vol. 58, pp.245–252. Ulusoy, B. (2004) ‘Application of Box-Wilson experimental design method for the photodegradation of textile dyestuff with UV/H2O2 process’, Dyes and Pigments, Vol. 60, pp.187–195. Wenzel, A., Gahr, A. and Niessner, R. (1999) ‘TOC-removal and degradation of pollutants in leachate using a thin-film photoreactor’, Water Research, Vol. 33, pp.937–946. Wilmott, N., Guthrie, J. and Nelson, G. (1998) ‘The biotechnology approach to colour removal from textile effluent’, JSDC, Vol. 114, pp.38–41. Yung-Shuen, S. and Deng-Kae, W. (2002) ‘Development of photoreactor design equation for the treatment of dye wastewater by UV/H2O2 process’, Journal of Hazardous Materials B, Vol. 89, pp.267–277.
