Integrated facility for Power plant waste processing - Industrial

Oct 24, 2018 - In this work a process has been developed for the transformation of gypsum from coal based thermal plants into sodium sulphate. The cur...
1 downloads 0 Views 2MB Size
Subscriber access provided by University of Sunderland

Process Systems Engineering

Integrated facility for Power plant waste processing Paula Manteca, and Mariano Martín Ind. Eng. Chem. Res., Just Accepted Manuscript • DOI: 10.1021/acs.iecr.8b04029 • Publication Date (Web): 24 Oct 2018 Downloaded from http://pubs.acs.org on October 25, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Integrated facility for Power plant waste processing Paula Manteca, Mariano Martín1 Department of Chemical Engineering. University of Salamanca. Plz. Caídos 1-5. 37008. Salamanca (Spain) Abstract In this work a process has been developed for the transformation of gypsum from coal based thermal plants into sodium sulphate. The current lack of market of gypsum and the traditional use of mineral gypsum makes this process an interesting alternative within the circular economy strategy to reuse a waste from the power industry. The process consists of the synthesis, the crystallization of sodium sulphate and the possibility of producing the hydrated and/or the anhydrous crystal. This section is integrated within the desulfurization unit by regenerating a fraction of the raw material required, CaCO3. The optimal operating conditions for the evaporation and crystals recovery have been computed using a mathematical programming approach. The economics of the process is attractive for the production of the anhydrous crystal, 0.21 €/kg with an investment of 36 M€, as well as for the production of the decahydrated one, 0.06 €/kg and 26 M€ of investment, processing the gypsum of a 350 MW power plant. Keywords: Desulfurizer, Gypsum, Solvay, mathematical optimization, Na2SO4

1

Corresponding author. Tel.: +34 923294479 Email address: [email protected]

1 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 30

1.-Introduction Power plants using coal are responsible for a fair share of the total power production in any country. In their operation flue gas and solid residues are produced.1-3 Most of the work in the literature focuses on the flue gas treatment to avoid emitting pollutants such as SO2, NOx and even CO2. Industrial experience, governmental offices and research studies have evaluated the performance of the various technologies available to control the emissions of these chemicals.3-9 Even though CO2 capture has received attention lately, emission control is devoted to NOx and SO2 removal from the flue gas.1 In the case of NOx, catalytic or non-catalytic denitrifiers are used to reduce the NOx into nitrogen and water. The removal of SO2 can be performed using various technologies including wet scrubbers, the most used one representing about 85% of all the desulfurizers used, spray driers, installed in around 12% of the facilities, while dry injection systems are used in 3% of the plants. Spray dryer systems are typically used in smaller facilities since their removal efficiency reaches up to 90% in the last designs, while wet scrubbers overpass that efficiency achieving easily 95% removal.2,4,10 Wet scrubbers add lime and air so that the SO2 is oxidized and transformed into CaSO4 that precipitates. Desulfurizer gypsum and ash are the main solid residues generated by coal power plants. The amount of gypsum produced in the operation of a regular coal-based power plant represents 30% of the amount of solid residues.1 While studies on flue gas treatment are widely established,2-9 the treatment of the solid residues is mostly focused on evaluating the removal of heavy metals.11 Gypsum and ash are typically used within the construction industry. Compared to the natural counterpart, synthetic gypsum presents a higher degree of purity.12,13 On the other hand, it also has a higher wet content (8- 10%) than natural gypsum (1-3%), leading to a higher energy consumption for drying.14 The uses of synthetic gypsum are spread in a number of industries. Gypsum has an interesting value within the construction industry for the production of plaster and plaster-board.15 Natural gypsum is particularly interesting for the manufacture of building plasters due to its content in clays. The synthetic one is widely used in the manufacture of plasterboard. Among other uses of gypsum, it is possible to highlight its use in confectionary, the brewing industry, pharmaceuticals, paper, glass, sugar-beet refining, oil absorbent, soil, as textile and paints filler, detergents, as a fertilizer, for dental modes and toothpaste.12,16,17 However, currently in many cases the desulfurizer gypsum is being sent to dump resulting in an economic burden, due to the insurance required and the monitoring of the dump and 2 ACS Paragon Plus Environment

Page 3 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

transportation of the gypsum, and an environmental issue, since large amounts of gypsum must be transported and stored over long periods of time. Alternative uses for gypsum must be found. Circular Economy is a current trend that aims at reducing the pressure on the use of natural resources by means of reusing waste from current production system and society.18 Sodium sulphate is a product of interest in a variety of industries. From the pulp and paper industry within the Kraft process, to its use as a filler in powdered detergents, or as a fining agent to remove air bubbles in the glass production. It plays a role in the textile industry and as a raw material for the production of other chemicals such as potassium sulphate, sodium silicate, etc. The production of sodium sulphate is well known in industry over the last centuries. It can be obtained directly from nature in the form of Glauber Salt or through the Mannheim process that uses sodium chloride and sulphuric acid. From the natural brine the Glauber process is used to produce the sodium sulphate crystals and several modifications have been suggested to reduce energy consumption including the use of a spray dryer of the slurry of molten Glauber’s salt, the Na2SO4·10H2O.19 Sodium sulphate is also a by-product of other processes such as the production of the viscose-fibre, the production of sodium dichromate or ascorbic acid. Alternatively, gypsum can be used as raw material to obtain sodium sulphate together with sodium carbonate that is the main product from the Solvay’s process.20 In this case, one of the residues of the operation of power plants can be further used to obtain sodium sulphate, and the byproduct calcium carbonate reused to capture SO2, see Figure 1.

Figure 1.- Intensification within the power plant This study presents an alternative for the use of gypsum. In this work an optimized process is developed for the use of desulfurizer gypsum in the production of sodium sulphate. Mathematical optimization techniques are used to 3 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 30

design the process and determine the operating conditions of the evaporator system.21 Finally, a technoeconomic analysis is carried out to estimate the investment and production costs. The production of sodium sulphate does not only represent diversification of the industry but also provides a way to drive the coal industry into the circular economy reducing the needs for natural CaCO3 intensifying the operation of the power plant. The rest of the work is organized as follows. Section 2 describes the flowsheet. Section 3 presents the modelling assumptions and the correlations developed to estimate the thermodynamics of the system. Section 4 describes the design procedure. Section 5 shows the main results of the mass and energy balances and the economic evaluation. Finally, section 6 presents some conclusions.

2. Process model. The production of sodium sulfate starts with mixing the gypsum and the sodium carbonate. While the gypsum is already in the form of fine power, the sodium carbonate must be milled for the reaction to take place. Sodium carbonate is the product of the Solvay process. The salt particle size is to be reduced to 15 m.20 The reaction takes place at 25ºC as follows, obtaining 98% of conversion when using an excess of sodium carbonate.

CaSO 4  Na 2 CO 3  CaCO3   Na 2SO 4 Calcium carbonate precipitates and can be separated using a centrifuge. We can use this precipitate and reuse it to capture SO2, intensifying the current flue-gas desulfurization process. Next, the solution is sent to a multieffect evaporator system to concentrate the sulfate and crystallize. The presence of residues of sodium carbonate prevent from full recovery, if a pure crystal is to be obtained. Alternatively, Ca(OH)2 can be used to precipitate the sodium carbonate as calcium carbonate. This alternative is to be evaluated within the design procedure. The high energy intensity of the process requires integration in the form of a multi-effect evaporation system. Finally, the crystals can be dehydrated in a furnace. Figure 2 shows the flowsheet for the process.

4 ACS Paragon Plus Environment

Page 5 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Figure 2.- Sodium sulfate production process This process can be integrated with the present power plant process consisting of the boiler, the flue gas treatment and the power island to process the gypsum and recycle the calcium carbonate as raw material for the desulfurizer, see Figure 1.

3.-Modelling Each of the units involved in the process is modeled using different approaches. First principle based short-cut models are used for heat exchangers and mixers, solid-liquid equilibrium and the stoichiometric of the reactions for the reactor and mechanical separation.22 However, more detail models are required for the evaporators and the furnace. Experimental data and the thermodynamics of the solutions are used to model the evaporation and precipitation phenomena within the multieffect evaporator system and the dehydration of the product.23 The set of chemicals is given in the following list J= {Wa, CaSO4, Na2CO3, CaCO3, Na2SO4, Ca(OH)2, CaSO4·2H2O, Na2SO4·10H2O, Na2CO3·10H2O, NaOH} Milling The energy consumption depends on the material humidity. Gypsum is provided as fine powder, see Figure 3, Therefore, no milling energy is assumed. However, sodium carbonate must be broken down for the reaction to take pace. 88.92 kJ /kg are consumed based on rules of thumb from the literature.24

5 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 30

Figure 3. Sample of gypsum Mixing The reaction taking place is shown below:

8H 2 O  CaSO 4 ·2H 2 O  Na 2 CO 3  CaCO3   Na 2SO 4 ·10H 2 O It operates at 25ºC under stirring and with 10% excess of Na2CO3 to achieve 98% conversion in 4h.20 The exit is a saturated stream. Thus, the reaction time together with the solubility allows computing the reactor size. This step is modelled as given by the stoichiometry with a conversion on 98% of the limiting reagent. The solubility of the sulfate is correlated from the experimental data in Lietzke and Marschall.25 Sol (0-40 ºC) [g/L]= -1.2083·10-03·T 4 + 8.0167·10-02·T 3 - 1.2392·T 2 + 9.5833·T + 50

(1)

Sol (40-100 ºC) [g/L]= 1.4184·10-02·T2 - 3.0143·T + 584.21 VolReactor= fc(Na2SO4hy) ·Rtime/Sol

(2)

The energy consumption for stirring is computed from rules of thumb 10 HP/1000gal.26 The sodium carbonate and the sulphates leave the reactor hydrated as well as the unreacted gypsum. Next, the product stream is processed in a hydrocyclone to remove the CaCO3 that has precipitated. We assume 100% recovery of it. Calcium carbonate is recycled to the desulphurization unit reducing the cost of raw material as well as the milling energy required in reducing the particle size from fresh CaCO3. This recycle represents an intensification of the current desulphurization process. Subsequently, the liquid stream is heated up to be fed to the evaporators system. The energy balance of the heat exchanger that feeds the evaporator system is computed as follows: QHX 1  F·( 

Tout

Tref

CpdT )

(3) 6 ACS Paragon Plus Environment

Page 7 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Where the heat capacity of the solution is a function of the concentration of sulfate and the temperature as given by Zaytsev and Aseyev.27 Following a two-stage procedure for the two variables involved, eq (4) is developed: 2 c p ( solution)  (7432.2·CSulf  4769.2·CSulf  4113.1)  (2.0993CSulf  3.3327)·T 

(3.4465·102 CSulf  4.7466·102 )·T 2  (1.86384 CSulf  2.6591·104 )·T 3

(4)

The feed temperature to the evaporator system is a variable that depends on its operation. Evaporators The first evaporator removes mostly water. For this effect the temperature at the evaporating chamber is the highest and therefore a cocurrent system is used for the operation of the evaporators. Each effect is modelled using mass and energy balances where the thermodynamics of saturation, solubility and crystallization are considered, see Figure 4 for a scheme. The evaporation chamber is fed from a solution of sulphate from the previous effect or the heat exchanger. The exit streams of the chamber can be a crystal phase if any, a solution (Saturated or not), and steam. The first effect uses utility steam as heating agent, but the following ones are fed by the overheated steam coming from the previous unit. The increment in the boiling point is a function of the composition of the liquid phase exiting each effect. The model for each one of the effects is given in eqs. (5)- (18) Steam

Vapor(E) To Ev n+1

Solution (L) Condensate

To Ev n+1

Crystals ( C) Feed (F)

Figure 4.-Scheme of one effect evaporator The mass balance to the solute is as follows,

7 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 30

j  effects ,i  Salts

Fj ·x j , f ,i  C j ,i  L j x j ,l ,i

0 i  Na2 SO4 ·10 H 2O  C j ,i   C j ,i  Solubiliy(T) i  Na2 SO4 ·10 H 2O x  xs for crystalization

(5)

xs  eq (1) The mass balance to the water is formulated as eq. (6)

fWa  VapI  lWa

(6)

Next the energy balance to the effects is performed. For the first effect commercial steam is used, thus the balance becomes:

HS  H F  Hs  H E  H L  HC

(7)

However, for effect number 2 onwards the evaporated water in the previous unit is used to provide the energy required in the one downstream. The energy balance becomes:

HE  H F  He  H L  HC

(8)

In both cases the enthalpy of the solution is computed considering the formation of each one from the chemical species, the dissolution, crystallization and heating, see supporting information. The enthalpy of a liquid stream, either the feed to an effect or the liquid product can be computed as eq (9) and (10) respectively: T

H F   fi ·hi , f  F i

c

p

( X , T )dT (9)

Tref

F   fi i

T

H L   li ·hi ,l  L  c p ( X , T )dT i

(10)

Tref

L   li i

Where the complexity of the solution suggests the development of a correlation for the heat capacity of the solution as a function of the temperature and the concentration in the main salt, the sulphate, eq. (11)

8 ACS Paragon Plus Environment

Page 9 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

2 3 4 6 c p ( saturated ) (kJ / kg )  4.18·(0.803·T 1.14·10  T   3.2·10 · T  )  cp    c p ( solution ) , eq (4)  

(11)

Where cp(saturated) was developed from experimental data in the literature.28 The enthalpy of formation involves the formation of the chemicals in the solution and the heat of solution, eq. (12)

h i ,(l or f )  H f  H sol

i  Salts

(12)

The enthalpy of the steam generated in each effect is computed as that of an overheated steam as per eq. (13):

H E =E·(H f ,Wa (liq )  c p ,liq (Teb  Tref )   (Teb )  c p ,vap (T  Teb ))

(13)

The enthalpy of the stream of crystals generated is estimated as that of the formation of the crystal as follows,

H crystals =C· H f ,Crystal

(14)

Finally, for the first effect alone, the energy is provided by condensing commercial steam where the enthalpy of the steam, HS, and the saturated liquid, Hs, are given by eq. (15):

H S =W·(H f ,Wa (liq )  c p ,liq (Teb  Tref )   (Teb )) H s  W·(H f ,Wa (liq )  c p ,liq (Teb  Tref ))

(15)

The saturation pressure as a function of the temperature is computed using Antoine correlation.29

Pj  exp(18.3036 

3816.44 ) Teb  227

(16)

Note that the presence of the salts results in an ebulloscopic increment computed as per eq. (17) that provides the actual operating temperature at the evaporation chamber.

TC  TE  TL  T j ,Evchamber

(17)

TL  Teb  3·0.512·m

Where m is the molality of the solution. Finally, for the heat transfer to be possible feasibility constraints on the temperature and pressure profile along the multieffect system are added to the problem

Pj 1  Pj T j , Steamchamber  T j ,Evchamber (C , E , L )  10

(18)

9 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 30

Dehydration In the last stage the crystals are dehydrated to obtain pure sulphate following the reaction below:30 Na2SO4·10H2O  Na2SO4 + 10H2O The unit is modelled following the stoichiometry of the reaction and an energy balance to the furnace. Alternatively, if the consumer does not require dehydrated sulfate, we can remove this unit and still sell the decahydrated crystals.

4.-Design procedure The process design involves decisions on the recovery of the excess of sodium carbonate added to secure a high conversion in the reactor and on the number of effects of the multieffect evaporator system. Thus, models for the different processes are developed using the models described for each of the units as presented in section 3. The models for all the units described above written in GAMS as a set of mass and energy balances including salt thermodynamics (i.e. heat of dilution, crystallization, ebulloscopic increase) and solubility. The main variables are the stream flows, the temperatures and pressures. The problem is formulated as an NLP, whose number of equations depends on the number of evaporators involved. Each problem is solved using a multistart optimization approach with CONOPT 3.0 as preferred solver. The solution time is below 1 min for all the problems when using an Intel Core i7 processor with 16Gb of Ram. Note that we do not claim global optimum due to the non-convex terms that appear in the formulation of the models, in particular the energy balances and the enthalpy of the salt solutions. The first consideration to comment about is related to the difficulty of crystallizing Na2SO4·10H2O in presence of Na2CO3·10H2O. They both can precipitate together since their solubility is similar. However, note that the presence of the carbonate in the solution is due to the excess provided to reach the 98% conversion reported in the literature. Therefore, the concentration is low. Two alternatives are considered. First, the precipitation of the carbonate in the form of CaCO3 using Ca(OH)2. A preliminary study was carried out comparing the additional profit due to the total recovery of the sulfate. The second alternative consists of recovering the sulfate until both are in the same concentration since the solubility of both, Na2SO4·10H2O and Na2CO3·10H2O is similar. Two NLP problems are developed, one per alternative, 10 ACS Paragon Plus Environment

Page 11 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

considering 1 evaporator alone for the comparison and modelling the units as presented in section 3. The objective function is a simple profit given as eq. (19). Each problem consists of around 500 variables and 315 eqs. Z=PNa2SO4·fc(Na2SO4) -(PGN)·(Q('Furnace1')+Q('HX1')+Q('Evap1')) -PCaOH2·fc(CaOH2)

(19)

Due to the additional evaporation energy involved as a result of the larger amount of product, an increase of around 10% is achieved, a parametric study is carried out comparing the production cost of the sodium sulphate as a function of the price of calcium hydroxide. Figure 5 shows the results. A price of 0.44 €/kg is required so that the additional yield obtained mitigates the additional costs for crystalizing it even if the Ca(OH)2 is for free. The value reaches 0.5 €/kg for the current price of Ca(OH)2. No additional income is considered out of the CaCO3. In short, the high cost of Ca(OH)2, the fact that only an additional 10% recovery of crystals and the fact that gypsum is currently a waste does not recommend the use of Ca(OH)2.

Figure 5.-Effect of the price of Ca(OH)2 on the price of Na2SO4 for its use to be promising The second major decision is related to the number of evaporators used. It is common knowledge that the larger the number of effects, the higher the steam economy. Different NLP optimization problems are set-up modelling processes that use 1, 2, 3 and 5 evaporator effects. Note that a disjunctive optimization approach could have been used to formulate a single optimization problem, but numerical issues and the small number of problems to be analyzed suggested a sensitivity analysis instead. The size of the problem ranges from 464 variables and 311 eqs. for the one

11 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 30

effect process and reaches up to around 900 variables and 600 equations when 5 effects are used. The objective function to compare on the alternatives is given by eq. (19). Figure 6 shows the decrease in the thermal energy required for the production of Na2SO4·10H2O as the number of units increase. An asymptote is starting to be reached for 5 units. Furthermore, the annualized cost of the units is shown in Figure 7. It is possible to see that 5 effects are slightly cheaper than 3. The reason is that for 3 effect two of the evaporators required more than 2000 m2 due to the low temperature gradient and large heat transfer involved, requiring the use two units in parallel.26 As a result the actual number of units to be bought is the same when 3 and 5 effects are used. To estimate the cost the film resistances are taken from the rules of thumb26 for each of the effects, assuming condensation and evaporation in each of the units. The energy balance allows computing the area, A, as per typical design of heat exchangers. The temperature gradient, T, is considered from the condensation temperature of the heating agent and the temperature of the evaporation chamber, eq. (20) Q=U·A·T

(20)

Finally, the cost is estimated from the area using the correlations developed in a previous work.31 Based on the two curves we see that the optimal corresponds to 5 effects. Therefore, 5 units are used for the rest of the analysis.

Figure 6.-Heat load consumed with the number of effects

12 ACS Paragon Plus Environment

Page 13 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Figure 7.- Annualized multi-effect evaporation system costs 5.- Results In this section the major results of the operation of the facility are presented for a process that consists of 5 evaporators and does not use calcium hydroxide to eliminate the excess of sodium carbonate. It is divided into two subsections. The first shows the operating conditions of the process, including the raw materials and product flows, as well as the working conditions at the multi-effect evaporator system. The second subsection elaborates a detailed economic evaluation. 5.1.- Optimal mass and energy balances In this section the main operating variables are presented. Table 1 shows the thermal and electrical demand of energy of the various units involved. Note that since gypsum is provided in small particle size, see Figure 2, no power is consumed for milling. The total power production adds up to 1.2 MW corresponding to the mixing in the reactor and the milling of the sodium carbonate. Additional savings of 0.3 MW are obtained if the sodium comes in the proper particle size. With regards to the thermal energy, the evaporator system, that includes the heat exchanger that heats up the fed to the first evaporator, represents more than 50% of the demand of heat, 22.3 MW. The needs to dehydrate the product results in a total consumption of thermal energy of 40.9 MW, which represents over 10% of the energy produced in the group that generates the gypsum.1 The consumption of electrical energy is negligible compared to the thermal energy required. Table 2 summarizes the consumption of raw materials and the flows of main products. It is expected that the CaCO3 produced and recovered is reused within the power plant to capture the SO2 within the coal. Therefore, it provides an additional asset for the process and reduces the need for natural resources exploitation.

13 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 30

Table 1.-Thermal and electrical energy consumption Thermal energy (MW) Electrical energy (MW) Mill2 0.3 Reactor -3.8 0.9 HX1 7.9 Evaporator 14.4 Furnace 18.6

CaSO4·2H2O Na2CO3 Steam Na2SO4 CaCO3

Table 2.-Major flows of material Raw Materials (kg/s) Products (kg/s) 4.86 3.30 6.82 3.51 2.77

The most complex system within the process is the multi-effect evaporator system. Table 3 summarizes the results. Note that the first unit only evaporates water and no crystals are recovered. The second one recovers 7% of the total sulfate and the last three affects almost recover 30% each. The recovery increase as we progress on the number of effects. Ev1 P (mmHg) 729 T (ºC) 108 Fraction Recovery 0 Na2SO4·10H2O

Table 3.-Effects operating variables Ev2 Ev3 460 277 88.9 76.5 7.2% 29.8%

Ev4 157 64.0 30.7%

Ev5 76 51.1 32.4%

5.2.-Economic evaluation. We divide this section into investment and production cost estimations. Investment cost is computed using the factorial method29 that relies on estimating the equipment cost. The equipment cost is estimated by performing a shortcut sizing of the units involved in the process such as mills, reactor, heat exchanger and multi-effect evaporator system and furnace. The characteristic variables that allow the estimation of the cost are the energy involved for the furnace, the volume of the reactor for a jacketed stirred tank, the area for the heat exchangers and the evaporator and the diameter of the mill. The equipment is sized and its cost estimated as presented in previous works31,32 where the source of the cost is the Matche web page.33 Table 4 shows the values of the characteristic variable for each one of the 14 ACS Paragon Plus Environment

Page 15 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

units and the cost. The total equipment cost adds up to 8.1 M€. As a result, for a facility that handles solids and liquids the factors used are 3.15 to compute the fixed cost29 and 1.4 for the total investment29 resulting in 36 M€. This is a small amount for a facility that require denitrifiers and desulfurizers whose cost goes over 100M€ each to avoid emitting NOx or SO2.10

Table 4.- Unit cost Unit Furnace Reactor HX1 Multieffect evaporator system Mill

Var Power (kW) Volume (m3) Area (m2) Areas (m2) Diameter (m)

Value 18000 434 132

Cost (€) 2216816 1969905 192681

250; 1016; 1149;1198;1221 3

2464512 1300400

Figure 8 shows the breakdown of the contribution of the units to the cost. The 5-effect evaporator system shows the larger share, 30%, but the furnace and the reactor show similar contribution, around 25% each.

Figure 8.- Equipment breakdown cost Production cost is estimated using. Eq. (7):34 Cproduction=M1 + 1.5·M2 + M3 +0.3 ·I

(7)

15 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 30

Where M1 represents the raw material costs. So far the company pays for gypsum disposal and monitoring of the dump. We assume that instead of paying for gypsum disposal we are paid that amount for treatment, 8€/t. It adds up to 1.1 M€ of savings a year, that corresponds to 5% of the production costs of the process. The cost of Na2CO3 is assumed to be 0.2€/kg.35 The asset expected from fine powder of CaCO3 is taken to be 0.25€/kg.36 M2 is direct labor costs, estimated as the positions created per million invested, 0.57,37 assuming 21.891 €/yr per position.38 M3 corresponds to the utilities, Steam, at 0.0077€/kg since it is low pressure steam,32 and natural gas, at 0.03€/kWh,39 to heat the furnace. If hydrate product is to be produced this cost will be removed. Finally, I is the investment as computed above. The raw materials cost accounts for 2.9 M€/yr, utilities add up to 9.7 M€ /yr and labor represents 0.4M€ /yr. As a result, the production cost results in 0.21 €/kg. Typical costs of sodium sulphate are below 0.2 €/kg.40 Therefore, renewable sodium sulphate is competitive with regular sodium sulphate but the benefit is small for the facility to be considered profitable beyond a waste treatment technology. Another consideration can be made due to the cost of dehydration of the sodium sulphate. If hydrated sodium sulphate can be sold as such, it is possible to save around 50% of the thermal energy required, 18.6 MW, but also the investment cost of the furnace, 2,2 M€ and its installation. As a result, the production cost decreases around 4.5 M€/yr for the energy usage in the furnace, and the fact that the product mass contains water. As a result, the production cost decreases down to 0.06€/kg. The facility investment cost without the furnace adds up to 26M€, representing almost a 33% reduction with respect to the original plant.

6.-Conclusions. In this work we have evaluated the production of sodium sulphate in the form of crystals from gypsum from desulphurization units in coal thermal plants and sodium carbonate. A process has been developed and optimized using a mathematical optimization approach for the intensification of the desulfurization of flue gas. Finally, a technoeconomic evaluation is carried out to evaluate investment and production cost. We process the gypsum generated from a 350MW group of a power plant. The recovery of sodium sulphate crystals reaches just below 90%. The cost of using Ca(OH)2 to precipitate the excess of sodium sulphate in the form of 16 ACS Paragon Plus Environment

Page 17 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

CaCO3 is economically unattractive. As a result, a fraction of the produced sodium sulphate will remain with the excess sodium sulphate in the solution since both have similar solubilities. A 5-effect evaporation system is suggested using a concurrent feed system for a competitive production cost of 0.21€/kg and a total investment of 36M€ for the anhydrous crystals and 0.06€/kg and 26M€ for the decahydrated sodium sulphate. While the production cost is comparable with the market price of the product, the consumption of energy represents 5-10% of the energy produced by the facility. Supporting Information Data on the species enthalpies can be found in the supporting information. This material is available free of charge via the Internet at http://pubs.acs.org. Acknowledgments The authors thank the University of Salamanca for software licenses, TCUE program for the award to PM, PSEM3 and the JCyL grant SA026G18 7.-Nomenclature A. Heat exchanger area (m2) C: Flow of crystals produced (kg/s) Cp: Heat capacity (kJ/kg) E: Evaporated water (kg/s) F: Total feed flow to evaporators (kg/s) fi: Flow of species i in evaporator feed (kg/s) fc(i): Flow of species i (kg/s) Hq: Enthalpy of stream q (kW) L: Total solution flow exiting evaporator (kg/s) li: Flow of species i in solution flow (kg/s) M: Production cost item (€/yr) P: Pressure at evaporation chamber (mmHg) Q: Thermal energy Flow (kW) T: Temperature (ºC) Sol: Solubility (g/L) U: Global heat transfer coefficient (kJ/m2 ºC) W: Flow of steam (kg/s) : Heat of evaporation Hf: Formation enthalpy (kJ/kg) Hsol: Solution formation (kJ/kg) T: Temperature gradient (ºC) Subindexes i: Species j: Effects F: Feed C: Crystals L: Solution 17 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 30

E: Vapor e: Condensated vapor S: Steam s: Condensated steam eb: Ebullition Liq: Liquid Vap: Vapor 8.-References. [1] Gas natural Unión Fenosa (2015) Declaración Medioambiental EMAS 2015. Central térmica La Robla. [2] EPA. Air pollution control Technology Fact Sheet. EPA-452/F-03-034, 2003. https://www3.epa.gov/ttn/catc/dir1/ffdg.pdf Last accessed March 2018 [3] EPA Economic and Cost Analysis for Air Pollution Regulations https://www.epa.gov/economic-and-cost-analysis-air-pollution-regulations [4] NLWA (2014) NORTH LONDON HEAT AND POWER PROJECT – FLUE GAS TREATMENT PLANT OPTIONS http://northlondonheatandpower.london/documents/141117_NLWA_Flue_Gas_Treatment_Technology_Options_Consu ltation_V1.pdf Last accessed March 2018 [5] Kikkawa H, Ishizaka H, Kai K, Nakamoto, T. DeNOx, DeSOx, and CO2 Removal Technology for Power Plant. Hitachi Review. 2008; 57: 174-178 http://www.hitachi.com/rev/pdf/2008/r2008_05_104.pdf. Last accessed March 2018 [6] Sorrels JL, Randall DD, Schaffner KS, Richardson Fry C (2016) Chapter 1: Selective Noncatalytic Reduction. https://www3.epa.gov/ttn/ecas/docs/SNCRCostManualchapter7thEdition2016.pdf [7] Sorrels JL, Randall DD, Schaffner KS, Richardson Fry C (2015) Chapter 2 Selective Catalytic Reduction https://www3.epa.gov/ttnecas1/models/SCRCostManualchapter_Draftforpubliccomment6-5-2015.pdf [8] Luo, X., Zhang, B., Chen, Y., & Mo, S.. Operational planning optimization of multiple interconnected steam power plants considering environmental costs. Energy, 2012, 37(1), 549-561 [9] Luo, X., Hu, J., Zhao, J., Zhang, B., Chen, Y., & Mo, S. Multi-objective optimization for the design and synthesis of utility systems with emission abatement technology concerns. Applied Energy, 2014, 136, 1110-1131. [10] Guerras, L.S.; Martín M. Optimal gas treatment and coal blending for reduced emissions in power plants: A case study in Northwest Spain. Revision submitted. Energy 2018 [11] Meawad, A., Yakimova-Bojinova, D., Pelovski, Y., An overview of metals recovery from thermal power plant solid wastes. Waste Manag. 2010, 30(12), 2548-2559 [12] British Geological Survey. Gypsum. London. Office of the Deputy Prime Minister 2006. [13] Eurogypsum (2017) http://www.eurogypsum.org/wp-content/uploads/2015/04/whatisgypsum.pdf Last accessed September 2018 [14] Ecofys. Methodology for the free allocation of emission allowances in the EU ETS post 2012 Sector report for the gypsum industry. 07.0307/2008/515770/ETU/C2, 2009. 18 ACS Paragon Plus Environment

Page 19 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

[15] Dunster, A.M. Flue gas desulphrization (FGD) gypsum in plasterboard manufacture. WRT 177 / WR0115. 2017. [16] Integrated Laboratory systems, Inc. (2006) Chemical Information Review Document for Synthetic and Naturally Mined Gypsum https://ntp.niehs.nih.gov/ntp/htdocs/chem_background/pubnomsupport/gypsum1_508.pdf. Last accessed September 2018. [17] Gilani, N. (2017) Uses of gypsum powder. SCIENCING https://sciencing.com/uses-gypsum-powder-6395376.html Last accessed September 2018 [18] Circle Economy, 2017, www.circle-economy.com [19] Von Plessen, H. Sodium Sulfates. In Ullmann’s Encyclopedia of Industrial Chemistry. 2012. 10.1002/14356007.a24_355 [20] Trendafelov, D., Christov, C., Balarew, C., Karapetkoca, A. Study of the conversion of CaSO4 to CaCO3 within the CaSO4 + Na2CO3 = CaCO3 + Na2SO4 four-component water–salt system Collect. Czech. Chem. Commun.1995, 60, 2107-2111 [21] Ahmetovic, E., Suljkanovic, M., Kravanaja, Z., Marechal, F., Ibric, N., Kermani, M., Bogataj, M., Cucek, L. Simultaneous optimization for multiple effect evaporation systems and heat exchanger network. Chem. Eng. Trans. 2017, 61, 1399-1404 [22] Martín, M. Industrial chemical process. Analysis and design. Elsevier Oxford. 2016. [23] Hougen, O.A.; Watson, KM.; Ragatz, RA. Chemical Process principles vol1 Material and Energy Balances. Wiley New York, NY 1947 [24] Genç, O. Chapter 6. Energy efficient technologies in cement grinding.In High Performance Concrete Technology and Applications. Yilmaz, S., Ozmen, H.B. Eds. 2016 http://dx.doi.org/10.5772/64427 [25] Lietzke, M.H.; Marschall, W.L. Sodium Sulfate Solubilities in High Temperature Aqueous Sodium Chloride and Sulfuric Acid Solutions - Predictions of Solubility, Vapor Pressure, Acidity, and Speciation. J. Solution. Chem., 1986, 15, 11, 903-917 [26] Walas, S. Chemical Process Equipment. Selection and Design. Butterworth-Heineman Series in Chemical Engineering. MA. 1990 [27] Zaytsev, I.D.; Aseyev, G.G. Properties of Aqueous Solutions of Electrolytes. CRC Press. Boca Raton. USA. 1992. [28] Kobe, K.A.; Anderson, C.H. The heat capacity of Saturated sodium sulfate solution. J. Ohys, Chem. 1985, 40, 4 , 429- 433 [29] Sinnot, R.; Coulson, K.; Richardson. Chemical Engineering. 3ªEd. Butterworth Heinemann, Singapore, 1999 [30] Vladyko, P.I.; Stactsov, A.K.; Nazarov, A.A Continuous crystallization of anhydrous sodium sulfate from the precipitation liqour in rayon spinning. Institute of General Inorganic Chemistry of the Academy of Sciences of the USSR; Kiev Man-Made Fibre Combine. Translated from Khimicheskie Volokna, 1975, 3, 53 – 54, May - June, 1975. Original article submitted April 30, 1974. 19 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 30

[31] Almena, A.; Martín, M. Techno-economic analysis of the production of epiclorhidrin from glycerol. Ind. Eng. Chem. Res. 2015, 55 (12), 3226-3238 [32] Martín, M.; Grossmann, I.E. Energy optimization of bioethanol production via gasification of Switchgrass. AIChE J. 2011; 57 (12), 3408-3428 [33] Matche (2014) http://matche.com/equipcost/SizeReduction.html. Last accessed July 2018 [34] Vian Ortuño, A. El pronóstico económico en química industrial. Eudema Universidad S.A., Madrid. 1991. [35] https://www.made-in-china.com/products-search/hot-china-products/Na2co3_Price.html. Last accessed June 2018 [36] https://www.icis.com/resources/news/2003/12/05/540629/prices-up-slightly-for-ground-calcium-carbonate/. Last accessed June 2018 [37] Martín, M. RePSIM metric for design of sustainable renewable based fuel and power production processes. Energy, 2016, 114: 833-845 [38] Feique (2017) Acta para el incremento salarial de 2017 del XVIII Convenio General de la Industria Química. http://www.feique.org/firmada-el-acta-para-el-incremento-salarial-de-2017-del-xviii-convenio-general-de-la-industriaquimica/ Last accessed May 2018 [39] http://ec.europa.eu/eurostat/statistics-explained/index.php/Natural_gas_price_statistics. Last accessed June 2018 [40] Valmet. High Power Generation from Recovery Boilers - What Are the Limits? Technical paper series. 2015.https://www.valmet.com/globalassets/media/downloads/white-papers/power-andrecovery/high_power_generation_recovery_boilers_whitepaper.pdf

20 ACS Paragon Plus Environment

Page 21 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Table of contents

21 ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1.- Intensification within the power plant 277x99mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 22 of 30

Page 23 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Figure 2.- Sodium sulfate production process 272x108mm (300 x 300 DPI)

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3. Sample of gypsum 90x67mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 24 of 30

Page 25 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Figure 4.-Scheme of one effect evaporator 90x58mm (300 x 300 DPI)

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5.-Effect of the price of Ca(OH)2 on the price of Na2SO4 for its use to be promising 90x53mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 26 of 30

Page 27 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Figure 6.-Heat load consumed with the number of effects 90x49mm (300 x 300 DPI)

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 7.- Annualized multi-effect evaporation system costs 90x47mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 28 of 30

Page 29 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Figure 8.- Equipment breakdown cost 90x54mm (300 x 300 DPI)

ACS Paragon Plus Environment

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Table of contents 90x57mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 30 of 30