Format And Type Fonts CCHHEEMMIICCAALL EENNGGIINNEEEERRIINNGG TTRRAANNSSAACCTTIIOONNSS VOL. 45, 2015 A publication of The Italian Association of Chemical Engineering www.aidic.it/cet Guest Editors: Petar Sabev Varbanov, Jiří Jaromír Klemeš, Sharifah Rafidah Wan Alwi, Jun Yow Yong, Xia Liu Copyright © 2015, AIDIC Servizi S.r.l., ISBN 978-88-95608-36-5; ISSN 2283-9216 DOI:10.3303/CET1545266 Please cite this article as: Sharan P., Bandyopadhyay S., 2015, Integration of multiple effect evaporators with background process, Chemical Engineering Transactions, 45, 1591-1596 DOI:10.3303/CET1545266 1591 Integration of Multiple Effect Evaporators with Background Process Prashant Sharan, Santanu Bandyopadhyay* Department of Energy Science and Engineering, Indian Institute of Technology Bombay, Powai, Mumbai 400 076, India. santanub@iitb.ac.in In multiple effect evaporators (MEE), evaporation takes place in stages; vapour coming from a stage loses its latent heat of vaporisation and acts as a heat source for the next stage. This process goes on till the last stage of evaporator where the desired concentration is achieved. MEE can be represented as a set process streams at different pressure/ temperature levels and its grand composite curve (GCC) can be generated. These process streams can be integrated with the background process to minimise the total energy consumption of the entire plant, rather than minimising the energy consumption of MEE and background process individually. An analytical methodology is developed to integrate (MEE) with the background process, such that the energy consumption for the entire pant can be minimised. A mathematical theorem is proposed to determine variation in overall energy requirement during integration of MEE with the background process. Proposed methodology is illustrated with the help of an example and up to 15.5 % of energy can be saved if the systems are integrated properly. 1. Introduction Evaporation in multiple effect evaporators (MEE) takes place in stages. The vapour coming out from first stage acts as a heat source for second stage, similarly vapour coming from second stage acts as a heat source for the third one and this process continues till the desired concentration is achieved. External heat is required to supply thermal energy in the first stage. MEE are applied in various food processing, desalination industries etc. MEE is an energy intensive process, for pulp and paper industry MEE consumes 25-30 % of total energy consumed by the plant (Khanam and Mohanty, 2010). Therefore it becomes essential to optimise energy consumption of MEE, to reduce its energy costs. Mathematical modelling serves as an excellent tool for modelling of MEE. Kaya and Sarac (2007) developed a linear model of MEE, by initially assuming effect temperature. Often the focus is on minimizing the energy consumption of MEE as standalone system instead of minimizing the energy consumption of entire plant. Hillenbrand and Westerberg (1988) first developed a method for integrating MEE with background process. They used problem table algorithm to calculate the sensible heat transfer for each effect. Urbaniec et al. (2000) used process integration technique to integrate process streams and evaporators. Westphalen and Maciel (2000), for the given evaporator temperatures, calculated the possible energy integration with the background process. Accordingly they calculated the maximum vapour bleeding for each effect. Higa et al. (2009) showed how number of effect and vapour extraction for process demand effects total steam consumption using process integration techniques. Mesfun and Toffolo (2013) developed a method for integrating MEE, background process and heat and power cycle. They showed that bark liquor consumption can be reduced to half if total site integration is done for same thermal demand and power requirement. Ensinas et.al. (2013) carried out process integration and used multi- objective optimization for designing sugarcane ethanol plant. The objective of the current work is to develop an analytical methodology for integrating MEE with background process for given effect temperatures to minimize the energy consumption of the entire plant. mailto:santanub@iitb.ac.in 1592 2. Grand composite curve of MEE In MEE there are number of streams entering and leaving the evaporator and undergo energy and mass transfer. These streams can be represented as process streams for integration with the background. Figure 1 shows n th effect of a MEE with different streams. Energy is required in every effect for vaporisation of water at constant effect temperature (Y) and may be represented as a cold stream. Its heat capacity (mCpenergy) is given as: 1 1energy n n n n n n mCp V H l h l h      (1) where, V is mass flow rate of vapour, H is enthalpy of vapour, l is liquor flow rate, and h is liquor enthalpy. The vapour generated from an effect acts as a heat source for the other effect. The vapour loses its heat in three steps. First it gets de-superheated, then it condenses at constant pressure, and at last it is cooled to atmospheric temperature. Vapour coming out from any stage gets superheated because of rise in boiling point (BPR) due presence of solute in liquor. Heat capacity of the vapour, as it gets de-superheated from effect temperature to saturation temperature (T), given as:   11 1, 1, -sup 1 1 1 - n nn n Y n T de erheating n n V H H mCp Y T        (2) Heat capacity for condensation is given as: 1 1condensation n n mCp V     (3) where, λ is latent heat of vaporisation. Heat capacity of the condensate up to atmospheric temperature is given as: 1cooling n water mCp V Cp   (4) Liquor enters the MEE at a temperature different than the effect temperature. Similarly the product temperature might be different from last effect temperature. These streams are considered for heat integration. For standalone MEE, there are four streams for every intermediate effect and five streams for the first and the last effects. Once the stream data are known, the utility requirement can be calculated by using modified problem table algorithm (MPTA), proposed by Bandyopadhyay and Sahu (2010), and the grand composite curve (GCC) can be generated. A typical GCC for MEE with two effects is shown in Figure 2. 3. Mathematical results Using mass and energy balance equations vapour evaporated in every stage (V) can be calculated for three effect MEE for forward feed flow by using Eq (5). 1 2 1 1 21 2 1 2 2 2 3 2 3 2 3 3 2 3 2 2 3 3 ( - ) -( - ) -( - ) 0 -( - ) ( - ) -( - ) ( - ) - 1 1 1 V F h h E Eh h H h h h h h H h V F h h E E V W                                (5) where, V is mass of vapour, W is total amount of vapour to be evaporated, F is the feed flow rate, and E is the distance by which individual effect are shifted away from Y axis in GCC. The mathematical results about MEE are established in this section. Lemma: When ∆En amount of energy shift is taken place between n th and (n+1) th effect in GCC of an MEE, the change in utility requirement for the first effect (∆Sn) is approximately given as:       n n K - n ΔS = ΔE K (6) where, n is the effect under consideration and K is total number of effects. ∆En is defined as:   n n+ 1 n ΔE ΔE ΔE (7) Due to brevity, only outline of the proof is given here. It can be seen from Figure 2 that Effect 1 and 2 are separated from each other by E1. Assuming that both effects are pinched initially and its hot utility requirement is H. The hot utility requirement for first effect is increased by (E1/2). So the evaporation taking place in effect one increases by (E1/2) as a result its cooling duty also increases by same amount. The 1593 heat duty for second effect approximately decreases by (E1/2), as this much amount of evaporation has already taken place in first effect. So net energy shift between Effect 1 and 2 is given as:    1 1 1/ 2 – / 2 .E E E  (8) Figure 1: n th effect evaporator Figure 2: GCC for two effect MEE The same concept can be used for generalising it for K effects. Change in utility requirement for effect 1, i.e., ∆Sn can be calculated from Eq(6). Net change in utility requirement for effect 1 (S) is given as: 2 43 1 ....- ...1 K K E E E S E E KK           (9) Because of shifting of effect one away from pinch, external utility might be required by other effects (second effect requires external utility in Figure 2). Net change in utility requirement for MEE is given as: ln-1,Yn-1,hn-1 l n ,Y n ,h n V n ,Y n ,H n l n-1 ,Yin n-1 ,hin -1 V n-1 ,Y n-1 ,H n-1 Vapour from n-1 th effect Vapour De-Superheating Condensate of n-1 th effect V n-1 ,T n-1 ,h cond-1 Concentrated Liquor Liquor from n-1 th effect Energy Requirement Condensation n th Effect 0 50 100 150 0 300 600 900 1200 T e m p e ra tu re ( o C ) Heat (kW) E1 Feed Pre-heating Energy required for evaporation Condensation Condensate Cooling BPR +∆Tmin Effect 1 Effect 2 Condensate Cooling 1594 1 2 43 ....... K E E E E Z E K    (10) This proves the following theorem. Theorem: If the effects are shifted away from the temperature axis by En, the net change in utility requirement for MEE is given as  K n n KE 1 / . Form this theorem, it can be said that whenever there is shift between two effects, the utility requirement for MEE increases. Therefore, En = 0 for all effects, i.e., all effects should be pinched. Corollary: In a standalone MEE, the minimum energy requirement corresponds to En = 0 for all stages. Based on these results, integration of MEE with background is presented as a flowchart (Figure 3). 4. Illustrative example. An example of concentrating sugar juice is considered for the purpose of illustration. In this example, a triple-effect forward feed MEE is considered. Feed of 1 kg/s with 15 % concentration is supplied at 80 °C. Product concentration is 60 % and must leave the battery limit at 80 °C. Effect temperatures are assumed to be 115, 80, and 60 °C with ∆Tmin of 10 °C. Background streams are listed in Table 1. GCC for the background process is shown in Figure 4 (one with the thin line). GCC drawn is reflected about temperature axis and its hot utility requirement comes out to be 190 kW. Table 1: Stream data for the background process Stream Tin ( o C) Tout ( o C) MCp (kW/ o C) H1 100 99 250 H2 100 60 2 H3 85 60 2.5 C1 60 120 5 C2 70 100 4 Assuming equal vaporisation, C1 initial C2 initial and C3 initial values are calculated to be 0.2, 0.3, and 0.6. Mass of vapours for three effects are 0.222 kg/s, 0.259 kg/s, and 0.269 kg/s. Concentrations after second iteration, are calculated to be 0.193, 0.289, and 0.6 respectively. Corresponding mass of vapour are 0.222, 0.258, and 0.27 kg/s. Process streams data are extracted after second iteration and GCC is generated using MPTA (thick line in Figure 4). Minimum hot utility requirement is targeted to be 605.4 kW. Values of X1, X2, and X3 are calculated to be 184.85 kW, 126.3 kW, and 130.47 kW, respectively. Therefore, Xo is 184.85 kW. Values of E1, E2, and E3 are 0, 58.55 kW, and 54.38 kW, respectively. Using the proposed theorem net change in utility requirement for MEE, i.e. Z is determined to be 37.64 kW. GCC of the integrated systems is shown in Figure 5. Total hot utility requirement for the overall system is 648.19 kW. Related design data for the integrated MEE are given in Table 2. If the system is designed independently with same MEE temperature then total utility requirement for the system is 767.4 kW. So by proper integration 15.5 % energy can be saved. Table 2: Final design parameter for integrated MEE Concentration of liquor from stage one (%) 19.70 Liquor flow from stage 1 (kg/s) 0.761 Concentration of liquor from stage one (%) 29.30 Liquor flow from stage 2 (kg/s) 0.512 Vapour flow from stage 1 (kg/s) 0.239 Liquor flow from stage 3 (kg/s) 0.25 Vapour flow from stage 2 (kg/s) 0.249 Vapour flow from stage 3 (kg/s) 0.262 5. Conclusion MEE is an energy intensive equipment for many process industries. Often the focus is on minimizing the energy consumption of MEE as a standalone system rather than minimizing the energy consumption of entire plant. In this paper an analytical methodology for integrating background process with MEE is proposed, proved mathematically, and demonstrated through an illustrative example. GCCs for MEE and the background process are matched appropriately to target the minimum overall hot utility requirement. Background GCC 1595 Procedure for optimally integrating the background process with MEE for given effect temperature is also discussed. It is proved that to minimize the energy consumption of a standalone MEE as well as of a MEE Figure 3: Flowchart describing the procedure for integrating MEE with background process Input Parameters: Number of effects, effect temperature, type of feed, feed temperature, liquor concentration, mass flow rate of feed, minimum temperature driving force and background process stream Start Assume equal evaporation between the stages, calculate initial liquor concentration (Cn initial ) Calculate liquor and vapour enthalpy Use Eq(6) to calculate vapour (V) evaporating from every stage, Calculate new liquor concentration (Cn new ) If (Cn initial )-(Cn new ) < desired accuracy Cn new =Cn initial Define MEE process stream, draw its GCC using MPTA and calculate its utility requirement (H) Draw GCC for background process on negative X axis and calculate its utility requirement (HB/G) Calculate maximum passible integration for each sage (Xn) max( ) o n X X n o n E X X  Calculate Z using Eq(10) New utility requirement for MEE (Hnew) newH H Z  Net utility requirement of the system=Hnew+HB/G-Xo Stop Calculate final design parameter using Eq(6) YES NO 1596 integrated with the background process, all effects must be pinched. Example shows if the systems are integrated properly then 15.5 % energy can be saved. 70 100 130 T e m p e ra tu re ( o C ) X2=126.3 X1=184.8 X3=130.5 MEE GCC H= 605.4 HB/G =190 Figure 4: GCCs of MEE and background process 60 80 100 120 140 T e m p e ra tu re ( o C ) 648.19 KW Reflected background GCC Figure 5: Integrated GCCs for MEE and background process References Bandyopadhyay S., Sahu G. C., 2010, Modified problem table algorithm for energy targeting, Ind. Eng. Chem. Res., 49(22), 11557-11563. Ensinas A.V., Codina V., Marechal F., Albrallei J., Silva M.A., 2013, Thermo-Economic optimisation of Integrated First and Second Generation Sugarcane Ethanol Plant, Chemical Engineering Transactions, 35, 523-528. Higa M., Freitas A.J., Bannwart A.C., Zemp R.J., 2009, Thermal integration of multiple effect evaporator in sugar plant, Applied Thermal Engineering, 29, 515-522. Hillenbrand J.B., Westerberg A.W., 1988, Synthesis of multiple effect evaporator system using minimum utility insights-I. A cascaded heat representation, Computers & Chemical Engineering, 12(7), 611-624. Kaya D., Sarac H.I., 2007, Mathematical modelling of multiple-effect evaporators and energy economy, Energy, 32(8), 1536-1542. Khanam S., Mohanty B., 2010, Energy reduction scheme for multiple effect evaporators, Applied Energy, 87(4), 1102-1111. Mesfun S., Toffolo A., 2013, Optimization of process integration in a Kraft pulp and paper mill – Evaporation train and CHP system, Applied Energy, 107, 98-110. Urbaniec K., Zalewski P., Zhu X.X., 2000, A decomposition approach for retrofit design of energy, Applied Thermal Engineering, 20, 1431-1442. Westphalen D.L., Maciel M.R.W., 2000. Pinch analysis of evaporation system, Brazilian Journal of Chemical Engineering, 17, 4-7.