PROCEEDINGS PAPER LENGTH

0 downloads 0 Views 2MB Size Report
review the impact of treating waste streams in a common large scale unit ... on Engineering for Waste and Biomass Valorisation - August 25-28, 2014 - Rio de ..... The objective function should be an indicative index of the performance of the ...
5th International Conference on Engineering for Waste and Biomass Valorisation - August 25-28, 2014 - Rio de Janeiro, Brazil

A systems approach for the integrated management of waste in multiproduct bio-refineries: analysis and optimization of a real-life industrial plant A. MOUNTRAKI1, M. TSAKALOVA1, A. PANTELI1, A. I. PAPOUTSI1and A. KOKOSSIS1,* 1 School of Chemical Engineering, National Technical University, Athens, Greece. *Corresponding author: [email protected], +30-2107724275 Keywords: Biorefinery, waste handling, superstructure, synthesis

Abstract This paper presents an integrated assessment of waste treatment technologies (including reuse and regeneration) for biorefinery effluents using systems engineering approach. The modelbased approach integrates different scenarios in design and applies optimization techniques to select and compare options. Methodology is based on a superstructure development that is formulated into a MINL optimization problem and is applied on a real life lignocellulosic biorefinery, whose waste effluents are up to forty nine streams with a choice of twenty-two treatment technologies (six for the liquids, four for the solids, seven for the air pollutants, two for reuse of relatively clean water streams and two for catalyst regeneration). In this work waste management features options for decentralized or centralized management, aiming to review the impact of treating waste streams in a common large scale unit compared to the use of smaller units with integrated waste streams. For this purpose the perspective of installation cost savings for a treatment unit within the plant of a biorefinery is examined. Results favor centralized scenarios and illustrate that the integrated use of waste treatment technology could offer reductions more than 50% to the installation cost.

5th International Conference on Engineering for Waste and Biomass Valorisation - August 25-28, 2014 - Rio de Janeiro, Brazil

1- INTRODUCTION Strong bio-based economies are expected to create, both directly and indirectly, significant revenues and jobs. They are also expected to increase farmer income and to improve economic activity in developing rural regions. Βiorefinery is the most powerful concept towards a Bio-based Economy [1] and is founded by the innovative and cost-efficient use of biomass for the production of food, feed, energy and chemicals. Lignocellulosic biorefineries process the most abundant type of biomass through intermediates that include cellulose, hemicellulose and lignin [2, 3]. Biorefineries account for the most efficient valorization of biomass but, in order to remain competitive to the chemical industry, they should build efficiency with process integration. The latter relates to the efficient use of energy [4, 5, 6] water [7, 8, 9, 10, 11], and process flowsheeting [12, 13]. Integrated design of waste treatment has been largely underestimated even though the biorefinery waste streams can be numerous and different in nature. The integrated technology could not only save costs but often produce profits. Industrial waste is distinguished to gas, liquid and solids with each class regulated by separate and different terms. Waste treatment ensures discharge under environmental regulation standards but there is a multitude of treatment technologies to use [14, 15]. Solid waste treatment can be subdivided to biological and thermal treatment. Thermal processing can be combined with energy recovery. The majority of the thermal treatment refers to incineration with some limited use of pyrolysis and/or gasification [16]. Air pollution control methods largely depend on the pollutants: choices differ between suspended particles, volatile organic compounds (VOCs), acid gas (SO2) or CO2 emissions [17, 18]. Wastewater technologies are classified as chemical, physical and biological. The main benefit of biological treatment is the recovery of stabilized organic matter and nutrients. Physical pretreatment requires less energy but is less effective in pollutant reduction; biological treatment demands higher energy but yields much higher reductions; chemical processes stand as intermediate choices. Wastewater treatment is laid out across primary, secondary, and tertiary systems that are benefitted by combining different technologies. Treatment of waste-water includes the removal of specific contaminants as well as the removal and control of nutrients [14, 19]. The abundance of technologies complicates the analysis whereas their complementary merits highlight the potential to compound benefits by integrating each other. Instead, common practice applies intuition and is pre-occupied on popular choices. The paper introduces an approach that copes with the challenge and enables a high-throughput evaluation of options in large-scale applications. Systems technology has already proved its potential to address waste treatment. The Pinch Technology offers the most popular branch of methods in Oil & Gas [4, 8, 20, 21]. Modelbased methods have been applied to pharmaceutical industries using computer-aided synthesis to identify tasks by examining waste properties against sets of environmental targets [22]. Databases offer access to options; nonlinear models predicted residues and costs whereas superstructure models have been set up to evaluate the better options. An extension of the work addressed uncertainties [23], and retrofitting problems [24]. A separate class of technologies includes systems based on ontology models developed to account for treatment options and embedding rules to support decisions [25, 26]. In the case of biorefineries, it is important to holistically address both the waste streams as well as the waste treatment technologies. The proposed methodology highlights a systematic and generic approach. The potential of the work is illustrated with a real-life lignocellulosic biorefinery that consists of 11 routes. Waste streams have been first mapped to treatment options. Next, the proposed mapping is structured as a synthesis problem that is formulated and optimized in the form of

5th International Conference on Engineering for Waste and Biomass Valorisation - August 25-28, 2014 - Rio de Janeiro, Brazil

an MINLP. The optimization selects technologies; optimal solution determines the technology to use. 2- INDUSTRIAL CASE This work is based on a real-life biorefinery [27] which valorizes the biomass feedstock through the organosolv technology developed by CIMV. Organosolv allows the separation without degradation of all the constituents of the vegetable matter with exceptional valorization of cellulose, lignins and xylose under mild reaction conditions by a cost-effective process in which solvents are recovered and recycled at the end of the process [28]. These three main products (cellulose, hemi-cellulose and lignin) can be converted into fuels and chemicals through a number of valorization pathways, as shown in Figure 1.

Figure 1: Perimeters of environmental assessment. Other than the core pretreatment process (organosolv) and the hydrolysis of sugars (cellulose to glucose and hemi-cellulose to xylose), the biorefinery includes eleven valorization paths. Hydrolyzed hemicelluloses (xylose) can be converted to xylitol either by biological or catalytic process. Hydrolyzed cellulose (glucose) can be converted to itaconic acid. Lignin can be converted either to Poly-Urethanes or Phenol Formaldehyde resins. Ethanol can be produced from celluloses, hemicelluloses or from a combination of those two. Ethanol can be further converted to ethylene and PVC. These processes produce 49 waste streams, which need to be classified and properly treated. Waste streams are first classified and mapped onto lists of eligible technologies. Depending on their water content, waste streams are classified as liquids (higher than 50% water content) or solids (less than 50% H2O). Streams with water content close to 50% can be treated as liquids as well as solids (binaries). Liquid streams with water content higher than 99,5% are considered recyclable; if water content is between 97-99,5% they can be reused after treatment. If the effluent stream contains a catalyst it is treated for its recovery and regeneration. A more detailed classification depends on the C/N content, their Lower Heating Value (LHV) and their BOD concentration. Gaseous waste streams are classified as rich in carbon dioxide. Those lower than 4% carbon dioxide content may be released into the

5th International Conference on Engineering for Waste and Biomass Valorisation - August 25-28, 2014 - Rio de Janeiro, Brazil

environment. The effluent gas streams are sufficiently rich to carbon dioxide, which can be considered "green", as product of fermentation processes. In this paper, the scenarios of CO2 capture examine the commercial applications of this product and therefore the possibility of profit from its sale. There is always the possibility of tightening the laws. The proposed classification produced 10 gaseous effluents, 4 solid streams, 1 binary, 14 liquids, 2 streams driven for regeneration, 1 reusable and 17 recyclables streams. Section 3,4 and 5 illustrate the methodology as it was applied in the industrial problem. 3- SYSTEMS APPROACH The screening for the appropriate paths can be carried out by assessing each path separately. To achieve an integrated and holistic overview of streams and technologies such approach of the problem is inadequate. Thus superstructure modelling and optimization is proposed. The objective of the systems approach is to solve a problem whose inputs are the nature and the volumes of waste streams, the sets of co-products, the energy demand, the costing and the conversion for each set of applicable technology treatments. The optimal solution determines the network of treatment technologies and the suitable integration of streams. Product revenues and production scales are also optimized. The paper advocates a model-based superstructure approach. Synthesis blocks include waste streams, treatment technologies and their products (energy, sludge, etc). Wastes streams are assigned to different technologies according to their nature and treatment required. Streams are connected to technologies according to the allocation algorithms presented in section 4. Treatment technologies are regressed separately deploying a list of models that are applied under different restrictions and constraints. Mathematical optimization accounts for mass and energy balances as well as costing (operating and capital) and takes the form of a MINLP model. Binaries relate to the selected technologies as well as the selection of the regressed model that is deemed appropriate for each case Treatment technologies and the models is introduced in section4 and 5 followed by the mathematical formulation in section 6. 4- WASTE TREATMENT TECHNOLOGIES Technologies selected for liquid waste treatment include anaerobic digestion (AD), activated sludge (AS), trickling filter (TF), Rotating Biological Contactors (RBC), aerated lagoon (AL) and stabilization pond (SP); they all account for biological technologies. Catalysts can be regenerated either by chemical precipitation or by ionic exchange, while streams with need of treatment in order to be reused can be driven to reverse osmosis, nano-, macro- or ultrafiltration. Treatment technologies produce sludge and a relatively clean effluent of water; anaerobic digestion yields also biogas. The treatment capacity of each technology differs with respect to organic loading (BOD), while some are not capable of treating streams rich in nitrogen. An allocation mapping (Figure 2) has been accordingly developed to assign options.

5th International Conference on Engineering for Waste and Biomass Valorisation - August 25-28, 2014 - Rio de Janeiro, Brazil

Liquid s Trickling Filter

Aerated Lagoon

OR

BOD: 120 - 310

BOD: 120 - 230 YES

Waste stream 50-97% H2O

H2O

Biogas

Sludge Rotating Biological Contactors

Anaerobic Digestion

N2

BOD: 25 - 1800

BOD: 100 - 380

Activated Sludge

Stabilization Pond

BOD: 120 - 380

BOD: 80 - 290

Figure 2: Logical diagram allocating liquid waste streams. Reuse technologies are restricted by the amount of suspended solids (TSS) they process. Regeneration technologies depend on the substance they regenerate; the allocation map proposed is presented in Figure 3.

H2O

Catalyst

Chemical Precipitation

OR

YES

Waste stream Reuse/ Regeneration

Catalyst

Ion Exchange

H2O YES

H2O>97%

Chemical Precipitation OR

H2O

Ion Exchange

Figure 3: Logical diagram allocating liquid waste streams. Options for solid waste management include incineration, combustion, gasification, pyrolysis or torrefaction. Streams containing less than 40% H2O are considered burnable; otherwise desiccation pretreatment is necessary. When nitrogen, cells, sugars or yeasts are present, the stream can be used for animal feed or fertilizer. The logical diagram of the algorithm developed for the allocation of solid streams is shown on Figure 4.

5th International Conference on Engineering for Waste and Biomass Valorisation - August 25-28, 2014 - Rio de Janeiro, Brazil

Feed

Solids

YES

Zymo Sugars Cell

Liquid Gas

Ash

Fertilize r

YES

Combustion

Waste stream 50 < H2O

Burnable

N2

Incineration

YES OR

Pyrolysis

Gasification Drying

Figure 4: Logical diagram allocating solid waste streams. The gaseous emissions need treatment when COx, NOx and SOx, are at inacceptable levels. In the case of processing gaseous emissions, only membrane reactor technology requires the existence of carbon monoxide. All carbon dioxide capture technologies produce a gas stream rich in air that may be discharged into the environment and a stream rich in carbon dioxide, which is compressed and stored. The carbon dioxide is produced from fermentation processes, thus is considered green. It is intended though to study the profitability of such investment, because CO2 has numerous industrial applications [17, 18]. The logical diagram of the algorithm developed for the allocation of gaseous emissions is shown on Figure 5. Gas

CO2

Gas

Membrane Reactor

Cryogenic Methods

Ca -Loop

YES

OR

Air Pollutant

CO

Discharge

COx, SOx, NOx

YES

Chemical Absorption MEA Chemical Absorption MDEA

OR

Physical Absorption Rectisol Physical Absorption Selexol

Figure 5: Logical diagram allocating gaseous emissions. According to the algorithms presented, the waste streams are not able to be associated to all technologies. For this reason, discrete variables are used to indicate the existence or not of the waste stream for each technology.

5th International Conference on Engineering for Waste and Biomass Valorisation - August 25-28, 2014 - Rio de Janeiro, Brazil

5. SYNTHESIS MODELS The superstructure models consist of the synthesis blocks, which include the biorefinery processes (BP), the waste streams (WS), mixers (M), the treatment processes (TP) and co products (CP). These building blocks are connected to develop a superstructure. To illustrate the generation of the network superstructure, an example of three processes TP1,TP2,TP3, is shown in Figure 6 .

Figure 6: Synthesis Blocks of the superstructure superstructure (left-had) and a supestructure of three treatment processes (right hand) A simple path is for example, when a valorisation process produces a liquid waste stream followed by a treatment process. The connection of the waste stream to the particular process respects the allocation algorithms presented in section 4. Prior to each treatment process stands a mixer that mixes all the waste streams lead to the particular treatment adjusting new features to the total stream (BOD, TSS etc.). Finally, a co-product is produced which in many cases (especially when the waste streams are solids) is energy generated. Figure 7 presents a generic superstructure and a specific example from liquid manufacture streams.

Figure 7: A real life treatment configuration Individual modeling units are based on a conceptual design while the modeling information of the process units is based on input-output models each featuring process variables in relation to the performance of each block and its material and energy streams. Costing features are based on literature. The multiproduct biorefinery that has been used as reference [27] has been formulated into a supestructure network including 54 streams, 14 production processes and 12 products (cellulosic pulp, glucose, lignin powder, C5 sugar syrup, xylose, itaconic acid, xylitol, PF resin, poly-urethanes, ethylene, PVC, ethanol,). Each process produces a variety of waste

5th International Conference on Engineering for Waste and Biomass Valorisation - August 25-28, 2014 - Rio de Janeiro, Brazil

streams presented in Table 1. Treatment technologies that can be used for each waste type of the biorefinery, allocated with respect to the allocation algorithms presented, are summarized in Table 2. The co-products are biogas and electric or thermal energy, sludge, cleaned water, compressed CO2 and the regenerated catalyst. Production Process Gas Liquid for Reuse Solid Catalyst regeneration CIMV 1 1 C5 to xylose 1 to xylitol bio 1 1+1* 1* to xylitol cat 1 1 1 to ethanol 1 1 C6 to glucose 1 to Itaconic Acid 3 1 to ethanol 1 1 Lignin to PF to PU 1 SSH ethanol 1 1 SHF ethanol 1 1 Ethanol to ethylene 1 Ethylene to PVC 3 2 Table 1: Waste streams from each production processes

Treatment Technologies

Chemical absorption

Physical absorption Ca-loop

Gas MEA

Liquid

MDEA

Rectisol

Anaerobic Activated Trickling Digestion Sludge Filters

Reuse

Reverse Osmosis

Solid

Combustion

Selexol

Rotating Biological Contactors

Membrane Cryogenic reactors methods

Aerated Stabilizatio Lagoon n Pond

Micro filtration Incinerati Gasificati on on

Torrefaction

Slow Fast Pyrolysis Pyrolysis

Chemical Ion Catalyst Regeneration precipitation exchange

Table 2: Treatment technologies implied

5th International Conference on Engineering for Waste and Biomass Valorisation - August 25-28, 2014 - Rio de Janeiro, Brazil

6- OPTIMIZATION MODELS Synthesis structure is formulated as an optimization model. The model is composed of basic mass and energy balances, economic flows and logical constraints. Following the allocation according to the characteristics of each stream the model needs to be formulated as a discretization problem. The formulation requires the following definitions. The liquid treatment technologies are denoted by the index set and the waste streams by the index set . The set of the discretization will be denoted by , the set of waste products by and finally, the set of BOD by . Based upon the above index sets, the continuous variables are: is the waste stream i in technology j is the total inlet to technology j, is the total waste from technology j going to discretization k, is the total BOD to technology j going to discretization k, is the total TSS to technology j, is the total flow of TSS from technology j going to discretization k is the product p from technology j is the product p from all the waste management processes is the energy needs of technology j is the total energy needs of the waste management processes, is the installation cost of technology process j, is the installation cost of biorefinery, is the energy cost of technology process j, is the total energy cost of biorefinery, is the revenue from technology j, the total revenue from waste management of biorefinery, is the total cost from waste management treatment. The integer variables are defined as follows: is the binary variable associated with the technology j and the discretization k and is the binary variable associated with the waste stream i and the technology j The parameters of the problem include the conversion of waste stream to products, the BOD of waste liquid stream as well as the maximum and minimum BOD and TSS going to discretization. The objective function should be an indicative index of the performance of the technologies networks. The objective of the problem is the maximization of the total cost, which is calculated in million $ per year. The expression of the objective function is followed: (1) The fixed cost, the energy cost and the revenue are calculated in million $ per year. Logical Constraints include logical constraints of mass flows (2) (3) ,

(4) (5)

The parameter is a large number which provides a reasonable upper bound for the variable if the unit is activated (i.e. or take the value 1). If however the unit is not selected (i.e. or take the value 0) the above inequality forces the positive continuous variable to zero values.

5th International Conference on Engineering for Waste and Biomass Valorisation - August 25-28, 2014 - Rio de Janeiro, Brazil

Since the technologies network should always be active, the logical constraints of binary variables: (6) Additional logical constraints are assigned to enforce the selection of a technology in case that the preceding or reference unit is not selected. 7. RESULTS AND DISCUSSION Analysis approaches the design in two levels: At first, it is necessary to understand the impact of the waste treatment allocation by splitting the problem to decentralized and centralized waste stream manufacture. The scope is to understand the cost driven options preferred for each individual stream- decentralized option. The supestructure is now simplified so that each individual stream that is connected to the process complies with the allocation algorithm. Figure 8 illustrates a simplified superstructure.

Figure 8: The simplified superstructure of decentralized liquid manufacture processes

For the illustration of the methodology it is essential to introduce the superstructure of BIOCORE biorefinery which is used as a base case of our study. The valorization paths include the biotechnological production of xylitol, the production of cellulosic ethanol and Poly-Urethanes. The decentralized solution of the problem is summarized at Table 3. Treatment Technology Organosolv Biomass fractionation Rotating Biological Contactors (RBC) C5 Sugar Hydrolysis Stabilization Pond (SP) Xylitol Production Rotating Biological Contactors (RBC) Ethanol Production Stabilization Pond (SP) Poly-Urethane Anaerobic Digestion (AD) Table 3:Decentralized Approach Optimal Solution. At a second level, the approach considers the problem holistically. The supestructure is used with a full development of the streams. The idea here is to examine the feasibility of the system while increasing the degrees of freedom. Individual streams have the opportunity to be mixed and lead towards one or more technologies. The degrees of freedom are related to the number N of technologies. The scope is to gradually increase the number of technologies allowed to co-exist with the view to assess the possibilities of integration. As N increases, the complexity of the problem will increase as well as the objective function does.

5th International Conference on Engineering for Waste and Biomass Valorisation - August 25-28, 2014 - Rio de Janeiro, Brazil

The analysis is presented with two scenarios where the number of technologies increases gradually. The results indicate significant changes between the two cases, revealing benefits with the integrated and more complex approach. The same biorefinery case study is used in scenario A, where the number of technologies N allowed for all the liquid streams is 1. According to allocation methodology the only common treatment technology for the five liquid streams is Anaerobic Digestion. In scenario B the complexity of problem has been increased as N increases to more than 2 technologies. Scenario A (all liquid streams allocated to anaerobic digestion) is compared to a relaxed option on the selection of treatment network, Scenario B (more than two different technologies may exist). Figure 9 shows a representation of the problem which relates to five liquid waste streams possible to allocate among six different treatment technologies. The superstructure of liquid manufacture processes is presented in Figure 10.

Figure 9: Treatment technologies option for the specific biorefinery design

Figure 10: The holistic superstructure of centralized liquid manufacture processes

5th International Conference on Engineering for Waste and Biomass Valorisation - August 25-28, 2014 - Rio de Janeiro, Brazil

Figure 11: Case A: All liquids to Anaerobic Digestion Scenario A: Centralized management in Anaerobic Digestion The installation cost of a unit that leads to a large scale unit whose cost is calculated to be approximately 15 million $/year. Figure 11 shows a representation of the problem where all liquid streams are allocated to anaerobic digestion. Table 4 includes further cost elements related to energy cost and revenues resulting from the production of biogas. Liquid (1+2+3+8+11) Technology AD Fixed Cost 16,404 Energy Cost Revenue 1,474 Total Cost 14,930 Table 4: Anaerobic Digestion costing features.

Scenario B: Multiple technologies options In Scenario B the complexity increases as the number of technologies that is allowed to be used for treatment is increased gradually. P In the first case number of technologies N can be at least 2. The results are illustrated at Table 4. The synthesis model allocates three of the waste streams to rotating biological contactors (RBC) and two others to anaerobic digestion (AD) unit. It is highlighted that the liquid stream coming from the production of PU elastomers can be allocated only by Anaerobic Digestion. Thus in the particular example within the plant of the biorefinery it is unavoidable to use a small unit of AD. Furthermore it seems that installing a smaller scale of AD unit and using an extra treatment process leads to savings of 33% of the total cost.

5th International Conference on Engineering for Waste and Biomass Valorisation - August 25-28, 2014 - Rio de Janeiro, Brazil

Technology Fixed Cost Energy Cost Revenue Total Cost million $/year

Liquid 1 RBC

Liquid 2 RBC 0.185

Liquid 3 RBC

Liquid 8 AD

Liquid 11 AD 4.722

4.907 0.172 0.172 0.845 0.845 4.234 Table 5: Two unit selection costing features.

Figure 12: Scenario B optimal Solution Preceding the concept of complexity the number of freedoms increases as one extra unit it is allowed to be use (Figure 13).

Figure 13: Schematic Representation of the optimal solution

5th International Conference on Engineering for Waste and Biomass Valorisation - August 25-28, 2014 - Rio de Janeiro, Brazil

The final option (Figure 14) includes the adding of the unit of stabilization pond. It is highlighted that to sustain the scale of the units installed the waste streams can be integrated and split to more than one treatment process (Table 5 liquid 8).

Figure 14: Superstructure representation of the optimal solution The difference in cost seems to be extremely interesting since the coexisting of more units in smaller scales can lead to a significant decrease in installation of waste management unit within the plant of biorefinery.

Technology Fixed Cost

Liquid 1 RBC 0.139

Total Cost million $/year

Liquid 3 SP

Liquid 8 SP AD

0.029 0.304 0.129

Energy Cost Revenue

Liquid 2 SP

Liquid 11 AD 0.136 -

0.129 -

-

0.013

0.013 0.420 Table 5: Three unit selection costing features.

8. CONCLUSIONS The paper outlines a structured methodology towards the systematic evaluation of waste management in a multi-product biorefinery. The integration of technologies and streams is systematic leading to the holistic selections of less costly processes. The methodology is illustrated via a real life biorefinery producing ethanol, xylitol and PU elastomers. The scenarios that were examined were related to the use of a large scale waste treatment unit compared to smaller units that can integrate and combine waste streams. The result of such integration is that including a waste treatment unit within a biorefinery plant can affect cost depending on the scale of the units. According to the methodology it is preferable to split the

5th International Conference on Engineering for Waste and Biomass Valorisation - August 25-28, 2014 - Rio de Janeiro, Brazil

waste streams allocating them to two or three smaller units. The savings from one big scale treatment unit to three smaller can save more than 50% of installation cost. 9. ACKNOWLEDGMENTS Financial support from the European Research Program BIOCORE (FP7-241566) is gratefully acknowledged. Consortium of FP7 Marie Curie project "RENESENG" is also acknowledged for s further support of the work. REFERENCES [1] International Energy Agency (IEA), www.ieabioenergy.com (2010). Accessed 29 May 2014 [2] Ding SY, Himmel ME.: The maize primary cell wall microfibril: A new model derived from direct visualization. J Agric Food Chem 54(3), 597–606 (2006) [3] Zhang YH P., Ding SY, Mielenz J. R., Cui JB, Elander R. T., Laser M., Himmel M. E., McMillan James R., Lynd L. R.: Fractionating recalcitrant lignocellulose at modest reaction conditions. Biotechnology and Bioengineering, 97(2), 214–223 (2007) [4] Smith R.: Chemical Process Design and Integration. John Willey and son, Chichester, UK, (2005) [5] Smith R., Jobson M., Chen L.: Recent development in the retrofit of heat exchanger networks, Applied Thermal Engineering, 30, 2281-2289 (2010) [6] Gadalla M., Jobson M. and Smith R.: Optimization of existing heat-integrated refinery distillation systems, Trans IChemE, 81, Part A, 147-152 (2003) [7] Gunaratnam M, A. Alva-Argez, AC Kokossis, J.-K. Kim, and R. Smith, Automated Design of Total Water Systems, Industrial and Engineering Chemistry Research, 44, pp 588-599, (2005) [8] Nikolakopoulos A., Karagiannakis P., Galanis A., Kokossis A.: A water saving methodology for the efficient development of biorefineries, 22st European Symposium on Computer Aided Process Engineering, 30, pp 6-10, Elsevier, (2011) [9] Wang, Y.P. and Smith, R. “Wastewater Minimisation”, Chem Eng Sci, Vol 49, pp 9811006 (1994). [10] Alva-Argaez A, Kokossis AC, and Smith R.: Wastewater minimization of industrial systems using an integrated approach, Computers and Chemical Engineering, 22, ppS741-S744 (1998). [11] Alva-Argaez, Vallianatos A., Kokossis A.: A multi-contaminant transshipment model for mass exchange networks and wastewater minimization problems. Computers and Chemical Engineering 23 1439–1453 (1999) [12] Stefanakis M. E., Pyrgakis K., Mountraki A.D., Kokossis A. C.: The Total Site Approach as a synthesis tool for the selection of valorization paths in lignocellulosic biorefineries, ESCAPE 23 Finland, June 9-12, (2013) [13] El-Halwagi, M. M.: Process integration, Academic Press, San Diego (2006) [14] Woodard, F, Industrial Waste Treatment Handbook, Butterworth–Heinemann (2001) [15] Chakraborty A., Linninger A.A, Plant-Wide Waste Management. 1. Synthesis and Multiobjective Design, Ind. Eng. Chem. Res. (41), 4591-4604 (2002). [16] Arnold M. E. and Merta E. S.: Towards energy self-sufficiency in wastewater treatment by optimized sludge treatment, Water Practice & Technology Vol 6 No 4, IWA Publishing (2011) [17] Koppmann R., von Czapiewski K. and Reid J. S., A review of biomass burning emissions, part I: gaseous emissions of carbon monoxide, methane, volatile organic

5th International Conference on Engineering for Waste and Biomass Valorisation - August 25-28, 2014 - Rio de Janeiro, Brazil

compounds, and nitrogen containing compounds, Atmos. Chem. Phys. Discuss., 5, 10455–10516 (2005) [18] Brinckerhoff P.: Accelerating the uptake of CCS: Industrial use of captured carbon dioxide, Global CCS Institute (2011) [19] Economic and Social Commission for Western Asia (ESCWA): Waste-Water Treatment Technologies: A General Review, United Nations (2003). [20] Strauss K.J.: Application of pinch technology in water resource management to reduce water use and wastewater generation for an area, WRC Report No. 1241/1/06 (2006) [21] El-Halwagi, M. M.: Pollution prevention through process integration, Academic Press, San Diego (2003) [22] Linninger, A. A.; Chakraborty, A.: Synthesis and optimization of waste treatment flowsheets. Comput. Chem. Eng. 23, 1415-1425 (1999) [23] Linninger, A. A.; Chakraborty, A.; Colberg R. D. Planning of waste reduction strategies under uncertainty. Comput. Chem. Eng. 24, 1043-1048 (2000) [24] Cavin L., Dimmer P., Fischer U. and Hungerbu1hler K.: A Model for Waste Treatment Selection and Costing under Uncertainty, Ind. Eng. Chem. Res, 40, 2252-2259 (2001) [25] Muñoz E., Capón-García E., Hungerbühler K., Espuña A., Puigjaner L.: Decision Making Support Based on a Process Engineering Ontology for Waste Treatment Plant Optimization, AIDIC Conference Series, 11, 261-270 (2013) [26] Muñoz E., Capón-García E., Hungerbühler K., Espuña A., Puigjaner L.: Decision Making Support based on a Process Engineering Ontology for Waste Treatment Plant Optimization, Chemical Engineering Transactions, 32, 277-282 (2013) [27] Mountraki A.D., Nikolakopoulosα A., Mlayah B.B., Kokossis A.C.: BIOCORE– A systems integration paradigm in the real-life development of a lignocellulosic biorefinery, 21st European Symposium on Computer Aided Process Engineering – ESCAPE 21 (2011) [28] Delmas M., Vegetal Refining and Agrichemistry, Chemical Engineering Technology, 31, No. 5, 792-797 (2008)