Ind. Eng. Chem. Res. 2007, 46, 9107-9113
9107
PROCESS DESIGN AND CONTROL Targeting for Total Water Network. 1. Waste Stream Identification Denny Kok Sum Ng and Dominic Chwan Yee Foo* School of Chemical and EnVironmental Engineering, UniVersity of Nottingham Malaysia, Broga Road, 43500 Semenyih, Selangor, Malaysia
Raymond R. Tan Chemical Engineering Department, De La Salle UniVersity-Manila, 2401 Taft AVenue, 1004 Manila, Philippines
Over the past decades, numerous research works have been dedicated to in-plant water reuse/recycle. After the opportunities for maximum water recovery are exhausted through water reuse/recycle, water flow rates may be further reduced with regeneration processes. Before wastewater is discharged to the environment, wastewater will be treated to meet the requirements given in the emission legislation. In this series of papers, an overall framework called the “total water network” is analyzed. A total water network consists of water reuse/recycle and water regeneration, as well as wastewater treatment for final discharge. Part 1 of this series of papers presents a new targeting procedure utilizing the recent developed graphical and algebraic approaches to identify individual wastewater streams that are emitted from a water network. As will be shown in Part 2 of the series, identification of the individual waste streams is necessary to investigate the interactions among different elements of the total water network. Two literature examples are solved to illustrate the proposed approaches. Introduction Because of stringent emission legislation and the increase in waste treatment costs, waste minimization has been a primary concern in the process and manufacturing industries. One active area for cost reduction activities has been that of resource conservation via in-plant material reuse/recycle, where quantity of raw material, waste and their associated treatment are reduced significantly. Over the past decade, numerous research works have been performed to address in-plant water reuse/ recycle systematically; these works have addressed topics from graphical pinch analysis techniques to mathematical optimization approaches.1-23 After the maximum water recovery potential has been exhausted via reuse/recycle, the fresh water demand of a water network may be further reduced via partial treatment of water sources, or water regeneration. In the seminal work that considers water reuse/recycle and regeneration for fixed load problems, Wang and Smith1 proposed to utilize the limiting composite curve in guiding regeneration placement. In their work, fresh water is used to satisfy the water-using processes until the water impurity concentration reaches the pinch concentration; it then is sent for regeneration. This approach locates the minimum regeneration flow rate for a water network and further reduces the network fresh water and wastewater flow rates. However, as noted by Kuo and Smith,24 in some cases, the pinch concentration migrates to a new position after regeneration has occurred. This leads to a sub-optimum target for the water network. In a later work, Kuo and Smith24 proposed a new graphical procedure for regeneration flow rate targeting with stream * To whom correspondence should be addressed. Tel.: +60-38924-8130. Fax: +60-3-8924-8017. E-mail addresses: Dominic.Foo@ nottingham.edu.my (C.Y.F.),
[email protected] (D.K.S.N.),
[email protected] (R.R.T.).
migration, to overcome the problem in the earlier work. Although the proposed procedure is useful in simultaneously targeting fresh water, wastewater, and regeneration flow rates, the stream migration operation is iterative in nature and, thus, is time-consuming. Moreover, both works of Wang and Smith1 and Kuo and Smith24 are only applicable for fixed-load problems where the water-using processes are modeled as mass-transferbased operations. More recently, Agrawal and Shenoy20 extended the limiting composite curves method1 for fixed-flowrate problems. Similarly, regeneration is conducted on water streams that have reached the pinch concentration, but they did not clearly indicate the procedure for the selection of the regeneration streams. On the other hand, water regeneration placement for fixedflow-rate problems were first reported by Hallale.13 The author proposed to place a regeneration unit across the pinch concentration, in which water is purified from the region with concentration higher than the pinch (higher-concentration region) where water is in excess, to the region with concentration lower than the pinch (lower-concentration region) where water is in deficit. Utilizing the same principle, El-Halwagi25 used the material recovery pinch diagram (MRPD) to explore water saving with a regeneration unit. On the other hand, Foo et al.21 proposed to regenerate water sources in the descending order of source concentration to achieve zero discharge. More recently, Ng et al.26 proposed a new procedure to determine the ultimate water targets for a given quality (regeneration outlet concentration) of regenerated water. The ultimate flow rate targets include the lowest possible fresh water, wastewater, and regeneration flow rates after water reuse/recycle has been maximized among all water-using processes.26 Although this work locates the essential regeneration flow rate target prior to detailed network design, no guideline has been given for the selection of appropriate regeneration streams. It
10.1021/ie071095h CCC: $37.00 © 2007 American Chemical Society Published on Web 11/28/2007
9108
Ind. Eng. Chem. Res., Vol. 46, No. 26, 2007
Figure 1. (a) Interactions between the elements of the water system. (b) General water flow scheme.30
is also interesting to note that the proposed procedure for stream selection for regeneration in the fixed-flow-rate problems13,21,25 is completely different than the fixed-load problems,1,20,24 where the stream(s) at the pinch concentration is(are) selected. No comparison has been made thus far on the merits of these different approaches, and will remain as a scope of study in this series of papers (to be discussed in Part 227). On the other hand, a considerable amount of work has been dedicated for the synthesis of a distributed effluent treatment network. Minimum flow rate targeting for a distributed effluent treatment network was initiated by Wang and Smith,28 who utilized the wastewater treatment composite curves. The main objective of this work is to obtain the minimum wastewater flow rate to be treated in a distributed effluent treatment network, which will eventually lead to the minimum fixed cost of the treatment system. Although the proposed methodology presented by Wang and Smith28 provides valuable insights into the design of distributed effluent treatment systems, the proposed method fails to predict the lowest possible treatment flow rate target in some cases.29 Viewing this limitation, Kuo and Smith,29 in turn, proposed the use of a multiple treatment system that leads to a minimum total treatment cost. However, the proposed method requires repetitive, iterative steps to achieve the optimal wastewater network. In a later work, Kuo and Smith30 noted that there are interactions between water reuse/recycle, regeneration system, and effluent treatment system (see Figure 1a). Therefore, all these systems should be addressed and synthesized as a whole, such as that shown in Figure 1b. Kuo and Smith30 also highlighted the similarities between water regeneration and final effluent treatment. When water of better quality is needed for the purpose of reuse/recycle, partial or full treatment (which is called “regeneration”30) is required. As reported in previous works,1,13,15,20,21,24-26 regeneration allows further reduction of the water flow rates after the opportunities for water saving via direct reuse/recycle have been exhausted. After the water has become too contaminated for reuse/recycle, and/or the regenerated water can no longer be reused/recycled any further (when the water network has reached its maximum extent of using regenerated water of a given quality), it must be treated before it can be discharged to the environment, to fulfill the requirements of the environmental legislation.30 Hence, the methodology that integrates the individual elements of water-using processes (via reuse/recycle) and water-treating systems (for regeneration and/or discharge), within a single network, is called
“total water network” synthesis. Although the work of Kuo and Smith30 provides the seminal contribution for total water network synthesis, it is limited to fixed-load problems. In a more recent work, Bandyopadhyay et al.31 proposed a new source composite curve to identify waste streams generated from a water reuse/recycle network for the fixed-flow-rate problem. The identification of individual waste streams is important, because smaller waste stream flow rates lead to lower cost in the distributed waste treatment system.28,29 Besides, as will be shown in Part 2 of this series of papers,27 the identification of waste streams serves as a good guideline in identifying streams for water regeneration (for reuse/recycle), as well as for final treatment (for discharge). However, Bandyopadhyay et al.31 did not explore the opportunity for further reuse/recycle, even though the water quality after treatment is clean enough to be sent to the water sinks. Hence, further work is needed to explore the interactions among individual elements in a total water network framework (water reuse/recycle, regeneration, and treatment for discharge) for fixed-flow-rate problems, as has been reported by Kuo and Smith30 for fixed-load problems. This is the subject of this series of papers. In this series of papers, a new targeting procedure for the total water network is presented. In Part 1 of the series, a waste stream identification technique is presented to identify individual wastewater streams that are discharged from a water network, utilizing the graphical technique of MRPD14,16,25 and the algebraic tool of the water cascade analysis (WCA).15,21 As will be shown in Part 2 of the series,27 waste identification is an essential step in selecting appropriate wastewater streams for regeneration and waste treatment. Two literature examples are used to illustrate the new waste identification technique, each for the graphical and algebraic approaches. Waste Stream Identification Techniques In this work, a novel targeting procedure is developed to identify the individual wastewater streams and their respective flow rates that are discharged from a water reuse/recycle network. Graphical and algebraic approaches that were originally developed for flow rate targeting in a reuse/recycle network are adapted here. These tools supplement each other well: the graphical technique provides conceptual insights to problem analysis, while the algebraic technique yields the flow rate targets rapidly and accurately.
Ind. Eng. Chem. Res., Vol. 46, No. 26, 2007 9109
Figure 2. Material recovery pinch diagram (MRPD) for (a) pure freshwater and (b) an impure freshwater feed.14,19,25
1. Graphical Approach: Material Recovery Pinch Diagram (MRPD). El-Halwagi et al.,14 as well as Prakash and Shenoy,16 individually presented the MRPD to locate the water flow rate targets for a reuse/recycle network. The MRPD is plotted on a diagram of cumulative load versus cumulative flow rate, such that the slopes of the segments correspond to the impurity concentrations of the sinks and sources that are individually plotted in an ascending order to form the sink and source composite curves (see Figure 2a). Next, the source composite is moved horizontally until it touches the sink composite, with the source composite being below and to the right of the sink composite curve. Minimum fresh water and wastewater targets are obtained from the overhang of the sink and source composite curves, respectively. When the water network is supplied by impure fresh water feed, an impure fresh locus is needed to slide the source composite curve until it touches the sink composite curve (see Figure 2b).14,19,25 The composite curves in the MRPD divide the water network into two separate regions at the pinch concentration. Fresh water is used in the lower-concentration region after the available water sources for reuse/recycle to the sinks have been exhausted. On the other hand, in the higher-concentration region, the available water sources exceed what is required by the sinks; hence, the unused sources are discharged as wastewater. From this observation, it is noted that all wastewater streams are generated from sources in the higher-concentration region. The subsequent step in the targeting procedure calls for the segregation of water sinks and sources at the pinch, followed by a new MRPD that should be plotted for the higherconcentration region. During streams segregation, all water sinks and sources are located in their respective regions, either in the higher- or lower-concentration regions. For the source that lies at the pinch concentration, its water allocation targets is to be identified (e.g., using various targeting techniques13-16,21), to determine the distribution between the higher- and lowerconcentration regions. In the higher-concentration region, the cleanest available water source is supplied at the pinch concentration, i.e., by the pinch-causing source (because no fresh water is used in this region). The quality of the excess water in this region can be maximized by utilizing streams that will just meet the concentration limits of the sinks. The minimum flow rate of the pinchcausing source that will satisfy all flow rate and load requirements in the higher-concentration region can be targeted with
Table 1. Limiting Data for Example 1a
a
sink, SKj
flow rate, Fj (t/h)
concentration, Cj (ppm)
1 2 3 4
20 100 40 10
0 50 50 400
Σj Fj
170
source, SRi
flow rate, Fi (t/h)
concentration, Ci (ppm)
1 2 3 4
20 100 40 10
100 100 800 800
Σi Fi
170
Data taken from ref 1.
the MRPD. When targeting the minimum pinch flow rate, the allocated flow rate of the pinch-causing source to the higherconcentration region is to be excluded from the MRPD. Due to the impurity content of the pinch-causing source, targeting is to be performed using the MRPD in Figure 2b.14,19,25 This flow rate target is called the “minimum pinch flow rate”. Water sources in the higher-concentration region that are emitted as wastewater streams can then be identified. One of these wastewater sources will always be the pinch-causing source, because of the excessive flow rate that is supplied to this region, whereas the others will be the water sources with higher concentration. The wastewater flow rate from the pinchcausing source can be determined by deducting the minimum pinch flow rate from the allocated flow rate of the pinch-causing source to the higher-concentration region. Other than that, the water source(s) that is/are not reused/recycled to the sink(s) in this higher-concentration region will be discharged as wastewater. This often includes the water source(s) at the highest concentration level. Example 1 is used to illustrate the proposed approach. Example 1. Table 1 shows the limiting data of a fixed-load example that consists of four sources and four sinks.1 The minimum water targets for reuse/recycle case are reported as 90 t/h for both fresh water and wastewater, and the pinch concentration is identified to be 100 ppm.1,15,16 Figure 3 shows the MRPD for the reuse/recycle case.
9110
Ind. Eng. Chem. Res., Vol. 46, No. 26, 2007
Figure 3. MRPD for Example 1 (reuse/recycle network).
Figure 4. MRPD for the higher-concentration region (Example 1).
After the pinch concentration has been identified, the sinks and sources data are categorized into lower- and higherconcentration regions. The water allocation targets15 of the pinch-causing source (either SR1 or SR2, which are located at 100 ppm), correspond to a flow rate of 70 t/h that is sent to the lower-concentration region (FL,Pinch) and a flow rate of 50 t/h that goes to the higher-concentration region (FH,Pinch); these have also been included in their respective regions. The segregation of the water sinks and sources leads to SK4, SR3, and SR4 (the pinch-causing source is omitted) being allocated to the higher-concentration region. The new MRPD plotted for this region is shown in Figure 4, which provides a magnified view of the region above and to the right of the pinch point in Figure 3. As shown, the minimum pinch flow rate (FPW) is targeted to be 5.71 t/h. The individual wastewater streams from the higher-concentration region can now be determined. The wastewater flow rate from the pinch concentration corresponds to the difference between the minimum pinch flow rate (5.71 t/h) and the allocated flow rate of the pinch-causing source to the higher-concentration region (50 t/h), i.e., 44.29 t/h. Figure 4 also shows that 45.71 t/h of wastewater is emitted from the highest-concentration sources at 800 ppm (i.e., SR3 and SR4). One can easily verify the total wastewater flow rate by summing the individual wastewater streams (44.29 t/h + 45.71 t/h ) 90 t/h), which matches the overall wastewater target identified in Figure 3 exactly.
2. Algebraic Approach: Water Cascade Analysis (WCA). In this section, the water cascade analysis (WCA) technique that was developed to target the minimum flow rates for a reuse/ recycle network15,21 is revised to identify the flow rates of the individual wastewater streams that are discharged from the water reuse/recycle network. The water cascade table (WCT) in Table 2 summarizes how the WCA is conducted for flow rate targeting in a water network. As shown, the concentration levels (Ck) are arranged in an ascending order (k ) 1, 2, ..., n) in the first two columns of Table 2. The flow rates of the water sink (Fj) and source (Fi) are summed at their respective concentration level k in columns 3 and 4 of Table 2. Column 5 of Table 2 represents the net flow rate (Σi Fi - Σj Fj) between the water sources and sinks at each concentration level k; a positive value indicates a surplus, and a negative value indicates a deficit. Next, the net water flow rate surplus/deficit is cascaded down the concentration levels to yield the cumulative surplus/deficit flow rate (FC,k) in column 6 of Table 2, with an assumed zero freshwater flow rate (FFW ) 0). This assumed flow rate is to facilitate the search for the minimum water flow rate and will be revised after the rigorous freshwater target is located. The next step in the targeting involves setting up the cumulative impurity load cascade (denoted as Cum. ∆m) to fulfill the load constraints. The impurity load (∆mk), given in column 7 of Table 2, is obtained as the product of the cumulative flow rate (FC,k) and the concentration difference across two
Ind. Eng. Chem. Res., Vol. 46, No. 26, 2007 9111 Table 2. Water Cascade Table (WCT) for Water Flow Rate Targetinga k
Ck
Σj Fj
Σi Fi
Σi Fi - Σj Fj
FC,k
k
Ck
(Σj Fj)1
(Σi Fi)1
(Σi Fi - Σj Fj)1
FFW
k+1
Ck+1
(Σj Fj)k+1
(Σi Fi)k+1
(Σi Fi - Σj Fj)k+1
l l l n-2
Cn-2
l l l (Σj Fj)n-2
l l l (Σi Fi)n-2
l l l (Σi Fi - Σj Fj)n-2
n-1
Cn-1
(Σj Fj)n-1
(Σi Fi)n-1
(Σi Fi - Σj Fj)n-1
n
Cn
a
l l l
∆mk
FC, k
∆mk
FC, k+1 l l
∆mk+1 l l l
FC, n-2
∆mn-2
FC, n-1 ) FWW
∆mn-1
Cum. ∆mk
FFW,k
Cum. ∆mk+1
FFW, k+1
l l l
l l l
Cum. ∆mn-1
FFW, n-1
Cum. ∆mn
FFW, n
Data taken from refs 15 and 21.
concentration levels (Ck+1 - Ck). Cascading the impurity load down the concentration levels yields the cumulative load (Cum. ∆mk) in column 8 of Table 2. A feasible water network is characterized by all positive Cum. ∆m values in column 8. A negative Cum. ∆mk value means that the impurity load is transferred from the lower-concentration level to the higherconcentration level, which is infeasible. In such a case, an interval freshwater flow rate (FFW,k, column 9 in Table 2) is calculated by dividing Cum. ∆mk by the concentration difference between level k (Ck) and the freshwater concentration (CFW), i.e.,
FFW,k )
Cum. ∆mk Ck - CFW
(1)
The absolute value of the largest negative FFW,k value will then replace the earlier-assumed zero freshwater flow rate in the flow rate targeting (column 6 of Table 2) to obtain a new feasible flow rate cascade set, and a feasible load cascade (positive values for all Cum. ∆m and Cum. ∆m ) 0 at the pinch).15,21 This new freshwater flow rate represents the minimum freshwater flow rate (FFW) of the network, whereas the final row in column 6 in Table 2 represents the wastewater flow rate (FWW) generated from the network. A more-detailed procedure for performing the WCA is presented elsewhere.15,21 The overall concept to identify the individual wastewater flow rate using the WCA is similar to that of the graphical technique, with the first step being to locate the minimum flow rates and pinch concentration of the water reuse/recycle network. Next, the water sources and sinks are segregated into the lower- and higher-concentration regions. This includes the allocated flow rates of the pinch-causing source that are identified from the WCT, i.e., concentration intervals just above and below the pinch in the FC column.15,21 Next, a new WCA is conducted for the higher-concentration region to locate the minimum pinch flow rate (FPW,k), via the modified eq 1, as follows:
FPW,k )
Cum. ∆mk Ck - CPinch
Table 3. Limiting Data for Example 2a
(2)
where Cpinch is the pinch concentration. Similar to the case of graphical targeting, the allocated flow rate of the pinch-causing source is omitted in the pinch flow rate targeting in this higherconcentration region. The pinch flow rate target is the minimum flow rate requirement (supplied at the pinch concentration) needed to satisfy all water sinks in the higher-concentration region.
a
sink, SKj
flow rate, Fj (t/h)
concentration, Cj (ppm)
1 2 3 4 5 6
120 80 80 140 80 195
0 50 50 140 170 240
ΣFj
695
source, SRi
flow rate, Fi (t/h)
concentration, Ci (ppm)
1 2 3 4 5 6
120 80
100 140
140 80 195
180 230 250
Σ Fi
615
Data taken from ref 5.
As in the case of the graphical targeting, the difference between the minimum pinch flow rate and the allocated flow rate of the pinch-causing source to the higher-concentration region gives the wastewater generated at the pinch concentration. In addition, the unused source in this region will be discharged as wastewater. Example 2 will be used to demonstrate the algebraic targeting approach. Example 2. Example 2 is a fixed-flow-rate example from Sorin and Be´dard;5 the limiting data are shown in Table 3. The minimum water targets for reuse/recycle are first determined, using WCA (Table 4), to be 200 t/h freshwater (FFW) and 120 t/h wastewater (FWW), whereas the pinch concentrations are identified to be 100 and 180 ppm.13-15 Table 4 indicates that there are two pinch concentrations (100 and 180 ppm) in the network, and these pinch concentrations separate the water sinks and sources into three different regions: a region with an excess of water (higher concentration than the upper pinch), a region that is self-sustained (between the lower and upper pinches), and a region with a water deficit (lower concentration than the lower pinch). Hence, the wastewater is expected to originate from the region with excess water. Note that Table 4 also shows the allocation flow rates of the upper pinch-causing source, where 100 t/h is sent to the region between the pinch points, which is self-sufficient, in terms of the water flow rate, whereas 40 t/h of water is sent to the region with excess water (intervals just above and below 180 ppm in column FC).15
9112
Ind. Eng. Chem. Res., Vol. 46, No. 26, 2007
Table 4. Water Cascade Analysis (WCA) Targeting for the Water Reuse/Recycle Case (Example 2) k
Σ Fj
C
Σ Fi
Σ Fi - Σ F j
1
0
120
-120
2
50
160
-160
3
100
120
4
140
140
5
170
80
-60 -80
6
180
140
140
7
230
80
80
8
240
9
250
10
-195
195 195
Cum. ∆mk
FFW, k
195
1000000
Cum. ∆mk
FFW ) 200
-120
-6
-280
-14
-160
-6.4
-220
-6.6
-300
-3
-160
-8
-80
-0.8
-275
-2.75
-80
∆mk
FC
0
120
80
∆mk
FC
-79980
-6
-120.00
-20
-200.00
-26.4
-188.57
-33
-194.12
80
4
-80
-4
4.00 0.00 (lower pinch)
40
1.6
-20
-0.6
-100
-1 2
1.60 1.00
-36
-200.00
-44
-191.30
40
-44.8
-186.67
120
1.2
-47.55
-190.20
-75
-0.75
-80027.55
FWW ) 120
-80.03
0.00 (upper pinch) 2.00 3.20 2.45
119970
119972.45
Table 5. WCA Targeting for Waste Stream Identification for Regions with a Higher Concentration than the Upper Pinch (Example 2) k
C
1
180
2
230
3 4 5
240 250 1000000
Σ Fj
Σ Fi
Σ Fi - Σ F j 0
80
80 -195
195 195
195
∆mk
FC
Cum. ∆mk
FFW, k
80 -115 80
Cum. ∆mk
FPW ) 5
0 0
∆mk
FC
0 0.8 -1.15 79980
Next, the water sinks and sources are segregated into different regions. Sinks SK1, SK2, and SK3 and a portion of source SR1 (80 t/h) are allocated to the region with the water deficit. Sinks SK4 and SK5, as well as sources SR2 and a portion of SR1 (40 t/h) and SR4 (100 t/h), are allocated to the self-sustained region between the pinch points. In addition, sink SK6 and sources SR5 and SR6 (the pinch-causing source SR4 is omitted), which have concentrations higher than the upper pinch concentration (180 ppm), are allocated to the region with an excess of water. A new WCA with an impure fresh feed at 180 ppm (upper pinch concentration) is conducted for the region with this excess water to determine the individual wastewater streams that are emitted. As shown in Table 5, the minimum pinch flow rate (FPW) is determined to be 5 t/h, whereas 85 t/h of wastewater is emitted from the final concentration level of 250 ppm. As stated previously, 40 t/h of the upper pinch-causing source (180 ppm) is supplied to this region. Hence, the wastewater that will be emitted from the upper pinch (180 ppm) for treatment is determined to be 35 t/h (40 t/h - 5 t/h ) 35 t/h). Summing the individual wastewater streams from the upper pinch, as well as the highest-concentration source, yields a total wastewater flow rate of 120 t/h (35 t/h + 85 t/h ) 120 t/h), which matches the overall wastewater target identified previously in the reuse/ recycle network (see Table 4). Conclusion A new targeting procedure for waste stream identification is presented in Part 1 of this series of papers. The procedure extended the recent developed graphical and algebraic flow-rate targeting approaches to identify individual wastewater streams that are emitted from a water network. Note that the summation of the individual waste stream flow rates matches the total wastewater flow rate targeted in the reuse/recycle network.
0 0.8 -0.35 79979.65
0.00 13.33 -5.00 79.99
5
0.25
85
0.85
-110 85
-1.1 84978.75
0.25 1.10 0.00 (pinch) 84978.75
Nomenclature Ci ) impurity concentration of source i Cj ) impurity concentration of sink j Ck ) concentration levels CPinch ) pinch concentration Cum. ∆m ) cumulative impurity load FC,k ) cumulative surplus/deficit flow rate of interval k FFW ) freshwater flow rate FFW,k ) interval freshwater flow rate FH,Pinch ) flow rate of the pinch-causing source that is sent to the higher-concentration region Fi ) flow rate of source i Fj ) flow rate of sink j FL,Pinch ) flow rate of the pinch-causing source that is sent to the lower-concentration region FPW ) minimum pinch flow rate FPW,k ) interval pinch flow rate FWW ) wastewater flow rate FW ) freshwater i ) index of sources j ) index of sinks ∆mk ) interval impurity load SKj ) sink j SRi ) source i Literature Cited (1) Wang, Y. P.; Smith, R. Wastewater Minimisation. Chem. Eng. Sci. 1994, 49, 981-1006. (2) Wang, Y. P.; Smith, R. Wastewater Minimization with Flowrate Constraints. Trans. Inst. Chem. Eng., Part A 1995, 73, 889-904. (3) Dhole, V. R.; Ramchandani, N.; Tainsh, R. A.; Wasilewski, M. Make Your Process Water Pay for Itself. Chem. Eng. 1996, 103 (1), 100-103.
Ind. Eng. Chem. Res., Vol. 46, No. 26, 2007 9113 (4) Olesen, S. G.; Polley, G. T. A Simple Methodology for the Design of Water Networks Handling Single Contaminants. Trans. Inst. Chem. Eng., Part A 1997, 75, 420-426. (5) Sorin, M.; Be´dard, S. The Global Pinch Point in Water Reuse Networks. Trans. Inst. Chem. Eng., Part B 1999, 77, 305-308. (6) Polley, G. T.; Polley, H. L. Design Better Water Networks. Chem. Eng. Prog. 2000, 96 (2), 47-52. (7) Savelski, M.; Bagajewicz, M. On The Optimality of Water Utilization Systems in Process Plants with Single Contaminant. Chem. Eng. Sci. 2000, 55, 5035-5048. (8) Savelski, M.; Bagajewicz, M. Design of Water Utilization Systems in Process Plants with a Single Contaminant. Waste Manage. 2000, 20, 659-664. (9) Bagajewicz, M.; Savelski, M. On The Use of Linear Models for The Design of Water Utilization Systems in Process Plants with a Single Contaminant. Trans. Inst. Chem. Eng., Part A 2001, 79, 600-610. (10) Savelski, M. J.; Bagajewicz, M. J. Algorithmic Procedure to Design Water Utilization Systems Featuring a Single Contaminant in Process Plants. Chem. Eng. Sci. 2001, 56, 1897-1911. (11) Go´mez, J.; Savelski, M; Bagajewicz, M. On a Systematic Design Procedure for Single Contaminant Water Utilization Systems in Process Plant. Chem. Eng. Commun. 2001, 186, 183-203. (12) Feng, X.; Seider, W. D. New Structure and Design Method for Water Networks. Ind. Eng. Chem. Res. 2001, 40, 6140-6146. (13) Hallale, N. A New Graphical Targeting Method for Water Minimisation. AdV. EnV. Res. 2002, 6 (3), 377-390. (14) El-Halwagi, M. M.; Gabriel, F.; Harell, D. Rigorous Graphical Targeting for Resource Conservation via Material Recycle/Reuse Networks. Ind. Eng. Chem. Res. 2003, 42, 4319-4328. (15) Manan, Z. A.; Tan, Y. L.; Foo, D. C. Y. Targeting the Minimum Water Flowrate Using Water Cascade Analysis Technique. AIChE J. 2004, 50 (12), 3169-3183. (16) Prakash, R.; Shenoy, U. V. Targeting and Design of Water Networks for Fixed Flowrate and Fixed Contaminant Load Operations. Chem. Eng. Sci. 2005, 60 (1), 255-268. (17) Prakash, R.; Shenoy, U. V. Design and Evolution of Water Networks by Source Shifts. Chem. Eng. Sci. 2005, 60 (7), 2089-2093. (18) Aly, S.; Abeer, S.; Awad, M. A New Systematic Approach for Water Network Design. Clean Technol. EnViron. Policy 2005, 7 (3), 154161. (19) Almutlaq, A. M.; Kazantzi, V.; El-Halwagi, M. M. An Algebraic Approach to Targeting Waste Discharge and Impure Fresh Usage Via Material Recycle/Reuse Networks. Clean Technol. EnViron. Policy 2005, 7, 294-305.
(20) Agrawal, V.; Shenoy, U. V. Unified Conceptual Approach to Targeting and Design of Water and Hydrogen Networks. AIChE J. 2006, 52 (3), 1071-1082. (21) Foo, D. C. Y.; Manan, Z. A.; Tan, Y. L. Use Cascade Analysis to Optimize Water Networks. Chem. Eng. Prog. 2006, 102 (7), 45-52. (22) Ng, D. K. S.; Foo, D. C. Y. Evolution of Water Network with Improved Source Shift Algorithm and Water Path Analysis, Ind. Eng. Chem. Res. 2006, 45, 8095-8104. (23) Gomes, J. F. S., Queiroz, E. M.; Pessoa, F. L. P. Design Procedure for Water/Wastewater Minimization: Single Contaminant. J. Cleaner Prod. 2007, 15, 474-485. (24) Kuo, W. C. J.; Smith, R. Design of Water-Using Systems Involving Regeneration. Trans. Inst. Chem. Eng., Part B 1998, 76, 94-114. (25) El-Halwagi, M. M. Process Integration; Elsevier: Amsterdam, 2006. (26) Ng, D. K. S.; Foo, D. C. Y.; Tan, Y. L.; Tan, R. R. Ultimate Flowrate Targeting with Regeneration Placement. Chem. Eng. Res. Des. 2007, 85 (A9), 1253-1267. (27) Ng, D. K. S.; Foo, D. C. Y.; Tan, R. R. Targeting for Total Water Network. 2. Waste Treatment Targeting and Interactions with Water System Elements. Ind. Eng. Chem. Res. 2007, 46, 9114-9125. (28) Wang. Y. P.; Smith, R. Design of Distributed Effluent Treatment Systems. Chem. Eng. Sci. 1994, 49 (18), 3127-3145. (29) Kuo, W. -C. J.; Smith, R. Effluent Treatment System Design. Chem. Eng. Sci. 1997, 52 (23), 4273-4290. (30) Kuo, W. C. J.; Smith, R. Designing for the Interactions between Water-Use and Effluent Treatment. Trans. Inst. Chem. Eng., Part A 1998, 76, 287-301. (31) Bandyopadhyay, S.; Ghanekar, M. D.; Pillai, H. K. Process Water Management. Ind. Eng. Chem. Res. 2006, 45, 5287-5297. (32) Bagajewicz, M. A Review of Recent Design Procedures for Water Networks in Refineries and Process Plants. Comput. Chem. Eng. 2000, 24, 2093-2113.
ReceiVed for reView August 10, 2007 Accepted September 20, 2007 IE071095H
