Conversion of CO2-Rich Natural Gas ... - ACS Publications

Subscriber access provided by University of Groningen

Process Systems Engineering 2

Conversion of CO-rich Natural Gas to Liquid Transportation Fuels via Tri-reforming and Fischer-Tropsch Synthesis: Model-based Assessment José Eduardo Graciano, B. Chachuat, and Rita Maria Brito Alves Ind. Eng. Chem. Res., Just Accepted Manuscript • DOI: 10.1021/acs.iecr.8b00135 • Publication Date (Web): 11 Apr 2018 Downloaded from http://pubs.acs.org on April 12, 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 57 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

Conversion of CO2-rich Natural Gas to Liquid Transportation Fuels via Tri-reforming and FischerTropsch Synthesis: Model-based Assessment José E. A. Graciano †, ‡, Benoît Chachuat ‡, Rita M. B. Alves*,† †

Department of Chemical Engineering, Escola Politécnica, Universidade de São Paulo, São Paulo, Brazil

‡

Department of Chemical Engineering, Centre for Process Systems Engineering, Imperial College London, London, UK *

e-mail: [email protected]; Tel: +55 11 3091-2265

Keywords: tri-reforming, Fischer-Tropsch, liquid fuels, natural gas, economic analysis, mathematical modeling, process simulation Abstract This paper presents a model-based analysis of a process coupling tri-reforming and FischerTropsch technologies for the production of liquid fuels from CO2-rich natural gas. The process also includes an upgrading section based on hydrocracking, a separation section, a water gas shift unit, and a Rankine cycle unit for recovering the excess thermal energy produced by the FischerTropsch reactor. Simulations are carried out in the process simulator Aspen Plus using standard

ACS Paragon Plus Environment

1

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 57

unit operation models where applicable, while modeling the non-conventional units, such as the Fischer-Tropsch and hydrocracking reactors, using Aspen Custom Modeler. The proposed process could achieve a carbon conversion efficiency upwards of 50% in the analyzed scenario, despite a natural gas feedstock with 30 mol% CO2. The analysis also reveals that the plant-wide electricity consumption could be covered nearly entirely by the Rankine cycle unit, enabling significant cost savings alongside a reduction of the overall global warming potential by about 10% in this specific case study. Finally, the results of a detailed economic assessment indicate that cheap natural gas is a prerequisite to the economic viability of the process, which would remain attractive in the current US scenario, yet presents a major impediment for its deployment in Brazil.

1. Introduction

The discovery of large petroleum fields off the Brazilian coast has increased the internal availability of natural gas (NG), from 21.6 Mm3/day in 2006 to 97.4 Mm3/day in August 2017.1,2 This NG is a strategic resource for the country development since, in addition to its primary importance as a fuel, it is also a source of hydrocarbons for petrochemical feedstocks. Nonetheless, exploration of these new offshore fields is hampered by high concentrations of carbon dioxide (CO2) in the NG. In some instances, CO2 can reach close to 80 mol% of the NG composition.3 This high CO2 content decreases the NG heating value and increases its sourness, requiring sweetening prior to transportation and distribution. Separation processes for CO2 removal from NG are energy intensive and often increase the operating cost significantly. Of course, there is also the issue of what to do with the recovered CO2. Possible alternatives include direct CO2 release to the atmosphere, which raises


2

Page 3 of 57 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


environmental concerns; CO2 reinjection into the wells for enhanced oil recovery (EOR) and CO2 storage, which may be energy intensive; and CO2 utilization as a raw material to produce high added-value chemicals.4 A large body of research has also focused on the direct utilization of CO2 rich-natural gas for syngas production ‒ a mixture of carbon monoxide (CO) and hydrogen (H2) ‒ used as an intermediate feedstock by the chemical and petrochemical industry to produce important chemicals such as methanol, acetic acid, olefins, gasoline, oxo-alcohols, synthetic liquid fuels, etc. A relatively new process for producing syngas with a desired H2:CO ratio from CO2-CH4 mixtures is tri-reforming (TR), a synergetic combination of the three main reforming routes (steam reforming, dry reforming and partial oxidation) into a single reactor.5 TR was originally proposed by Song6 in order to make reforming a more energy efficient process via the combination of endothermic and exothermic reactions. This process has the potential to mitigate coke deposition and to reduce the energy consumption by incorporating water and oxygen into the reaction system. Another important feature is its ability to consume CO2 as a feedstock and hence mitigate direct CO2 emissions. For these reasons, TR has attracted significant attention in recent years, with a particular focus on developing new TR reactor configurations and integrating TR with other unit operations.7–9 Díez-Ramírez et al.10 studied the energy and exergy efficiency of TR for the production of syngas with a desired H2:CO ratio for use in methanol or FischerTropsch synthesis. Zhang et al.11 analyzed the technical and economic viability of integrating TR and methanol production, and concluded that it might be possible to produce methanol with a net profit. More recently, Zhang et al.12 integrated TR with a dimethyl ether (DME) production process. Minutillo and Perna13 investigated the co-generation of electricity and methanol, by


3


Page 4 of 57

combining TR with a power plant, and showed that this process is able to produce high-purity methanol from flue gas. Another process alternative entails the conversion of syngas from TR to liquid transportation fuels (LTF) using Fischer-Tropsch (FT) synthesis. Over the last decade or so, the FT process has drawn significant attention due in part to its economic competitiveness with natural oil, which reached the historical peak of $140/bbl in June 2008, alongside other environmental benefits. This process promotes CO reduction by H2 to produce a mixture of paraffins, olefins and oxygenates, similar to those found in natural oils. It is noteworthy that the LTF produced from such synthetic oils (synoil) lead to lower emissions than their natural oil counterpart, due to physical and chemical properties such as reduced density, ultra-low sulphur levels, low aromatic content, and high cetane number.14 It is estimated that around 20 FT plants are in operation worldwide, half of which for commercial purpose. The conversion of NG to LTF, also known as GTL process, has been studied extensively due to its importance, see for instance the recent literature review by Floudas et al.15 However, a majority of the modeling studies concerned with GTL processes have relied on simplified models to describe the unit operations, or consider classical reforming options such as auto-thermal reforming (ATR).16–18 The main contribution in this paper is an assessment of a complete process converting CO2rich natural gas to FT liquids via the TR route. A first-principles modeling approach is adopted to describe the main unit operations and conduct the analysis. The performance of the combined TR+FT process is assessed, and its viability is analyzed in terms of cash flow over 30 years of operation. The rest of the paper is organized as follows: section 2 presents the conceptual design for the proposed plant and details the modeling of each operating unit; section 3 presents the


4

Page 5 of 57 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


method used to assess the economic viability of the process; the results are discussed in section 4; and final conclusions are drawn in section 5. 2. Steady-State modeling of integrated TR+FT process

The proposed TR+FT process comprises six major sections, as depicted in Figure 1. In the TR unit, the reactor converts a mixture of CO2-rich NG, steam and oxygen into syngas. The syngas is then mixed with a surplus hydrogen stream and compressed to feed the slurry FT reactor. The liquid product from the FT reactor goes to the upgrading section, where a hydrocracking reactor converts both the heavy synoil and recycled wax into lighter fractions. Meanwhile the vapor stream from the FT reactor is fractionated to recover the light synoil from the unconverted reagents. The light synoil fraction is sent to an atmospheric distillation column (ADC) for separation into gasoline, kerosene and diesel products. The water gas shift (WGS) unit produces hydrogen for use in the upgrading section and to adjust the H2:CO ratio in the feed of the FT reactor. Finally, the heat produced by the FT reactor is converted to electrical power by the Rankine cycle (RC) unit.

Figure 1 – Conceptual design of the integrated TR+FT process. The proposed flowsheet is implemented in the software Aspen Plus® (v8.8), using standard first-principles models for the common unit operations. Nonstandard equipment, such as the FT


5


Page 6 of 57

and hydrocracking reactors, are modeled using Aspen Custom Modeler® and exported to the plant-wide simulator in Aspen Plus. The analyzed case study is that of a NG stream at 27.22 kg/s, composed of about 48 mol% CH4, 30 mol% CO2, 10 mol% C2H6, 6 mol% C3H8, and a mixture of C4-C8 alkanes and N2 for the remaining 6 mol%. A total of 61 components, including H2, O2, N2, CO, CO2, H2O, alkanes, alkenes and alcohols are considered, with their properties modeled by the Predictive Soave-Redlich-Kwong (PSRK) equation of state. Further details about the model and performance of each unit are given in the following subsections. 2.1. Tri-reforming section

The TR reactor combines standard endothermic and slightly exothermic reforming reactions – steam reforming (Equation 1.a), dry reforming (Equation 1.b) and partial oxidation (Equation 1.c) – within a single bed reactor. The water-gas-shift reaction (Equation 1.d) also takes place in the reactor, leading to a reduction in the CO2 conversion rate subject to an increase in the feed water.19 CH4 +

H2O ↔ CO + 3 H2

(∆H◦ = 206.3 kJ/mol)

(1.a)

CH4 +

CO2 ↔ 2 CO + 2 H2

(∆H◦ = 247.3 kJ/mol)

(1.b)

↔ CO + 2 H2

(∆H◦ = − 35.6 kJ/mol)

(1.c)

H2O ↔ CO2 + H2

(∆H◦ = − 41.1 kJ/mol)

(1.d)

CH4 + 1/2 O2 CO +

As discussed by Cho et al.20, any linear combination of three independent reactions is sufficient to describe the possible reactions among the 6 species (CH4, H2, CO, CO2, O2 and H2O) in TR. The equivalent set of reactions in Equations 2 is adopted herein to describe the methane TR process. This choice is advantageous due to the large volume of information available in the literature, including the species consumption rates in terms of the catalyst mass and various effectiveness factors accounting for intra-particle transport limitations.21 The corresponding parameters are reported in the Supporting Information material (Table S1 and S2).


6

Page 7 of 57 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


CH4 + H2O ↔ CO + 3 H2

(∆H◦ = 206.3 kJ/mol)

(2.a)

CH4 + 2 H2O ↔ CO2 + 4 H2

(∆H◦ = 164.9kJ/mol)

(2.b)

CH4 + 2 O2

(∆H◦ = −802.7kJ/mol)

(2.c)

(∆H◦ = − 41.1 kJ/mol)

(2.d)

↔ CO2 + 2 H2O

CO + H2O ↔ CO2 +

H2

The former sets of reactions describe methane TR, but the actual NG stream consists of a mixture of different hydrocarbons, which follow different reaction mechanisms and rates. To overcome this issue, a strategy to simulate the complete TR reactor as a sequence of two different reactor models has been adopted (Figure 2). The Gibbs reactor (R1) accounts for the reforming reactions involving alkanes with two or more carbon atoms. The model for R1 assumes adiabatic operation under a pressure drop of 0.1 bar, and considers methane, nitrogen and carbon dioxide are not reacting. The subsequent plug-flow reactor (R2) implements the set of reactions (Equations 2) for the methane TR. The model for R2 assumes a reactor with 0.7m diameter and 1 meter length, operating adiabatically with no pressure drop, catalyst bed void fraction of 0.36 and a particle density of 1.5625 g/cm3.22

Figure 2 – Process flow diagram of the tri-reforming unit. In the TR flow diagram (Figure 2), NG is mixed with medium pressure steam, before being heated by the TR reactor effluent stream to approximately 950 °C. After that a 98% pure oxygen


7


Page 8 of 57

stream is added to the TR reactor feed. Then, the reactions take place over the sequence of adiabatic reactors R1 and R2. A parametric analysis is conducted by varying the mass fractions of oxygen and steam streams in the TR feed, and simulating the changes in H2:CO ratio, H2 flow rate, and CO2 conversion at the TR outlet. The results of this analysis are shown in Figure 3. Both the H2:CO ratio and H2 flow rate are increasing as more steam is added, whereas they present a minimum and a maximum, respectively, at an O2:NG feed ratio of about 50 wt% (Figure 3.a-b). Conversely, the CO2 conversion is decreasing with the addition of steam, although a maximal conversion of ca. 55% is once again obtained for a O2:NG feed ratio close to 50 wt% (Figure 3.c). Notice also that a positive CO2 conversion is only predicted in a rather small operating region around this maximum.

(b) (a)


8

Page 9 of 57 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


(c) Figure 3 - Parametric analysis of the H2O:NG and O2:NG feed ratios in terms of: (a) H2:CO ratio, (b) H2 flow rate, and (c) CO2 conversion (Note: both H2O:NG and O2:NG axes are inverted in Figure 3.c to better visualize the profile). The desired operating point (red dots in Figure 3) is determined by maximizing the H2 flow rate in the syngas stream, subject to a minimum CO2 conversion of 25%. The corresponding H2O:NG and O2:NG feed ratios are 0.39 and 0.52, respectively. Under this operation, the reactor converts 98.7% of the NG into a syngas stream with a H2:CO ratio equal to 1.57.

2.2. Fischer-Tropsch unit

The FT reactor converts syngas (CO and H2) into compounds with long carbon chains producing a synthetic oil of composition similar to that of natural oil. Its chemistry is much more complex than the mere polymerization of -CH2- groups and formation of the main products: nParaffins, α-Olefins and Oxygenates (Reactions 3.a-c). n CO + (2n+1) H2 → CnH2n+2

+

n H2 O

(3.a)

n CO +

+

n H2 O

(3.b)

2n H2 → CnH2n


9


n CO +

2n H2 → CnH2n+2O + (n-1) H2O

Page 10 of 57

(3.c)

Because of this high complexity, the range of products in the FT reactor is often predicted in terms of distribution functions, such as the Anderson-Schultz-Flory (ASF) chain length distribution.23–25 Simpler, macroscopic reaction rates are also widely used to simulate the FT synthesis, e.g. for the design and scale-up of these reactors. Another key feature of the FT process is its high exothermicity, with reaction enthalpies as large as -170 kJ.mol-1. For this reason, early FT industrial reactors were built as multi-tubular fixed-bed reactor (FBR), with the catalyst packed inside the tubes and surrounded by coolant.26 This type of reactors is easy to scale up, but it also proves costly to operate due to high pressure drops, low catalyst utilization and poor heat transfer. Another type of reactors, largely used in commercial FT plants, is the slurry bubble column reactor (SBCR), in which fine solid catalyst particles are dispersed into the liquid phase, in a countercurrent arrangement with the gaseous mixture flowing in the upward direction. This configuration enables a better temperature control and results in larger syngas conversion rates in comparison with an FBR.27 The range of FT products depends mostly on process conditions such as temperature and pressure. For instance, the products’ chain length is inversely proportional to the reactor temperature. High temperature FT synthesis (HTFT: 300-350°C, ~20 bar) tends to produce shorter molecules, in the range of gasoline, while low temperature FT (LTFT: 175-250°C, 10-45 bar) generates long carbon chains, in the range of diesel, heavy oil and wax.28 Conversely, the products’ chain length increases with the reactor pressure, which should thus be set at an economic optimum by considering the energy required to compress the syngas feed stream. In order to promote a higher diesel selectivity, as driven by economic and strategic incentives, the present study considers a SBCR. A validated microkinetic model is used to predict the FT


10

Page 11 of 57 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


product distribution, based on a carbide mechanism using a Langmuir-Hinshelwood-HougenWatson (LHHW) expression for relating the rate of hydrocarbon formation to the partial pressure of the reacting gas mixture.29 The kinetic parameters have been estimated from experimental data, obtained in a slurry FT reactor with Co-Re/Al2O3 catalyst for a wide range of operating conditions. This kinetic model accounts for different formation rates of n-paraffin and α-olefins, as given in Equations 4 and 5, respectively. Moreover, methane and ethylene have their own specific rates of formation (Equations 4.a and 5.a). According to Visconti al. (2007), it is possible to obtain hydrocarbon chains as many as 50 carbons.30 Due to the lack of thermodynamic information for n-paraffins with more than 30 carbons and α-olefins with more than 20 carbons, these higher compounds are represented by two pseudo-components with formula C30H62 and C20H40 (Equations 4.c and 5.c) herein. = . . [] =

. .

(4.a)

[] , = 2 … 29

(4.b)

"#$ = % & . . []'

(4.c)

= (),* e, - []

(5.a)

= (,* e - [] , = 3 … 19

(5.b)

,

## = % &(,* e - []'

,

(5.c)


11


Page 12 of 57

with R, the rate of species formation; P, the partial pressure of species; k, the kinetic constants for the rate-determining steps; K, the equilibrium constants for the non-rate-determining steps; α, the chain growth probabilities (Equation 6); and [S], the fraction of vacant sites (Equation 7). It is worth noting that expressing [S] in terms of the reagents partial pressures can provide more accurate predictions.28 = = =

0 0 +

(6.a)

0 0 + + () e,

(6.b)

0 , = 3 … 50 0 + + (,* e,

(6.c)

1 1 1 0 [] = 41 + - + - 51 + + + 7 % 8 9 6 6 6

:

8

(7)

The FT process also produce oxygenates due to the water activity in the catalyst medium. These oxygenates typically correspond to about 1% of the products,31 and they are considered herein as a set of alcohols with 1-6 carbon chains (Equation 8). The rates of formation for H2O, CO and H2 are calculated stoichiometrically (Equations 9-11). Other species, such as O2, CO2 and N2 are considered as inerts. A full set of values for the kinetic parameters and the detailed description of the elementary kinetic equations can be found in Todic et al. 29. ;0

0.01 = &% + % ' , = 1 … 6 6

(

0 = % + % + %= − 1?;0

(

−0 = % + % + % ;0

(8)

(9)

(10)


12

Page 13 of 57 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


(

− = %=2 + 1? + % 2 + % 2;0

(11)

The SBCR is implemented in Aspen Custom Modeler as a reactive vapor-liquid equilibrium stage with fixed temperature, pressure and mass of catalyst. Both the liquid and vapor phases are considered to be perfect mixtures. The vapor-liquid equilibrium relies on fugacity coefficients calculated using the PSRK equation of state, which can predict the behavior of non-polar and polar mixtures in combination with light gases. The complete model comprises >1,000 states, 35 parameters, and 3 input variables (temperature, pressure and mass of catalyst). For its integration within the complete process in Figure 1, the reactor is scaled-up for conversion of approximately 80% of CO, at 240°C and 30 bar 32,33, requiring 80 t of catalyst. The reactor volume is calculated by considering a catalyst load of 28% in volume for SBCR and a particle density of 646 kg/m3, resulting in 442 m3 reactor for production of about 30t/h of FT liquids 28. These values are consistent with those reported by Maretto and Krishana 33. Figure 4 depicts the process flow diagram for the FT unit. Syngas from the TR unit exchanges energy with the Rankine cycle (HXR1A), before cooling down to about 40°C to remove the excess water. Then, the dehydrated syngas is mixed with the hydrogen stream from the WGS unit to obtain a mixture with H2:CO ratio of 2. This mixture is compressed to 30bar, in a 2-stage compressor before feeding into the FT reactor. The SBCR liquid outlet stream (FTLOUT), composed of long hydrocarbon chains is sent to the hydrocracking unit, while the SBCR vapor outlet stream (FTVOUT), composed of light hydrocarbons and unconverted syngas, passes through a series of coolers and flash drums to separate the light synoil stream (SYNOIL) from the OFFGAS1 and the water phase (WW2). Lastly, the SYNOIL stream is sent directly to the separation unit, while the OFFGAS1 stream goes to the WGS unit, where hydrogen is recovered and recycled to the FT reactor or sent to the upgrading unit.


13


Page 14 of 57

Figure 4. Process flow diagram of the Fischer-Tropsch synthesis unit. 2.3. Upgrading section

Production strategies aiming to maximize FT diesel fuel can be traced back to the early 1980s.26 Basically, they integrate LTFT synthesis (using SBCR) with upgrading units, where hydrocracking reactors (HCR) convert the produced heavy oil and wax into lighter fractions. This conversion requires C-C scission, which is promoted at high temperatures (360-440°C) and in the presence of a catalyst. The reactions are carried out in an adiabatic reactor fed with hydrogen in a ratio of 800-1800 Nm3 per Nm3 of liquid feed.34 The HCR model implemented herein is based on Bhutani et al.35 and Mohanty et al.36, but with the pseudo-components replaced by real components, n-alkanes from 1-30 carbons. For this reason, it is assumed that all α-olefins and oxygenates are hydrogenated to n-alkanes with the same number of carbons. Other components such as H2O, N2, CO, CO2 are considered as inerts. This reactor is modelled as a plug flow reactor, using kinetic parameters which are functions of the normal boiling point of each component (n-alkanes). Equation 12 describes the mass balances for the liquid phase (n-alkanes) in a differential catalyst element: H

@AB EFG O = P + 2 QRO P S 7 = − D + % 8 8 D8 51 + 7 with N O = 8 QRO P V 7 @C EF8

(12)

8I


14

Page 15 of 57 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


The first term on the right-hand side denotes the consumption rate of component P and the second term denotes its formation rate. It is assumed that light species – with a normal boiling

point below 400 K (i=7, heptane) – do not undergo cracking. Moreover, the rate of formation of P

is given by cracking of higher components starting from P +2 carbons. AB is the mass flow rate

of component P in [kg/h], D is the mass fraction, is the rate of cracking of component P [kg-

reactant/kg-catalyst], 8 is the probability of formation of P from the cracking of j. The

correction term W1 +

XY XZ

[ enforces the mass balance via the addition of one atom of hydrogen

in each cleaved molecule, with: EFG , the molar mass of hydrogen in [kg/kmol]; and EF8 , the molar mass of component \ in [kg/kmol]. The overall hydrogen consumption is determined from

a mass balance on the flow rate of the n-alkanes.

The cracking rate constant ( ) for component P is given as a function of its normal boiling

point, heavy oil and catalyst specific mass, as reported by Bhutani et al.;35 whereas the

probability of cracking ( 8 ) is a function of the normal boiling point of components i and j, as proposed by Mohanty et al.36 The energy balance is computed by assuming an adiabatic process, so the variation of the total enthalpy is equal to zero. This HCR model is implemented in Aspen Custom Modeler, where the set of differential equations is discretized over the reactor mass and solved as a system of nonlinear equations with 3,300 variables, 8 parameters and 2 input variables (mass of catalyst and pressure). The reactor is scaled-up by considering a liquid flow rate of approximately 70 m3/h, hydrogento-liquid ratio of 800 Nm3/m3, WHSV (weight hourly space velocity) of 1 h-1,34 catalyst bed with 0.55 void fraction, and catalyst density of 4800 kg/m3.37 This resulting reactor has a diameter of 3.56 m and a length of 15 m.


15


Page 16 of 57

Figure 5 depicts the PFD used to simulate the upgrading unit, which is fed by a hydrogen makeup stream from the WGS unit (H2TOHDC), heavy synoil from the SBCR (FTLOUT), and wax recycle from the separation unit (WAXREC). These streams are heated up to 410ºC before entering the HCR (HDCRACKR), where the heavy synoil is cracked into lighter fractions. The reactor effluent stream is first cooled down to 350°C to recover the major part of the LTF fuels (FD4); then, the temperature is further reduced in the second flash drum (FD5) to achieve a H2 fraction of 90% in the hydrogen recycle (SS20).35 A small fraction of the stream SS20 is purged (OFFGAS2) to maintain the hydrogen-to-liquid ratio at 800 Nm3/m3, and the remaining stream SS21 is heated and

compressed before recycling. Lastly, the collected cracked synoil (CRACKOIL) is sent to the separation unit.

Figure 5 – Process flow diagram of the upgrading unit. 2.4. Water-gas shift unit

The WGS unit is responsible for recovering the unconverted hydrogen from the FT reactor. In addition, this unit takes advantage of the WGS reaction (Equation 1.d) to produce extra hydrogen from the unconverted carbon monoxide in the stream OFFGAS1 (see Figure 5). The WGS reaction is slightly exothermic, presenting a favorable thermodynamic behavior at low temperatures and a kinetically favorable performance at high temperatures. For this reason, a


16

Page 17 of 57 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


WGS unit often comprises two adiabatic reactors, a high temperature reactor (400-500ºC) followed by a low temperature one (about 200ºC), in order to achieve high conversions more efficiently.38 Instead, Pasel et al.

39

discuss the use of isothermal operation for WGS units, and

they point out certain advantages over the usual adiabatic process, in terms of reducing the process complexity and reaching the thermodynamic equilibrium at higher temperatures. The present work considers such an isothermal WGS operation by using a multi tubular fixed bed reactor at low temperature (200°C), cooled with boiling water (at 150°C). This reactor is simulated by a standard RPlug model in Aspen Plus with 3400 tubes of diameter 1 inch and length 1 meter and constant heat flux of 6781.15 [cal/s.m2] (this value is based on a typical overall heat-transfer coefficient for water and hydrogen mixtures40). The catalyst bed has a void fraction of 0.3, particle density of 6877.2 [kg/m3], and particle size of 0.62 cm.38,41 The pressure drop is estimated by Ergun’s equation and the kinetics are given by Moe’s model42 apud Seo et al. 43, which is based on industrial data for low temperature WGS catalysts. Figure 6 shows the process flow diagram of the WGS unit. The unit receives unconverted syngas from FT unit (OFFGAS1), which is mixed with steam (ST2WGS) to shift the equilibrium to the hydrogen production side. The output mixture passes through a heat exchanger (HX5) to set its temperature to 200°C, and then enters the WGS reactor to convert CO and H2O into CO2 and H2. The product stream (SS26) has less than 0.5% of carbon monoxide. This stream is processed in a membrane separation unit, modeled as a mathematical separation module (MS1), and retaining 90% of the hydrogen at 95% of purity

44

. The pressurized off-gas stream

(OFFGAS3) is expanded in a turbine to produce electricity. Lastly, the recovered hydrogen stream (H2RECY) is recycled back to the FT and hydrocracking units.


17


Page 18 of 57

Figure 6 – Process flow diagram of the water-gas shift unit. 2.5. Separation unit

An atmospheric distillation column (ADC1 in Figure 7) is used to separate the synoil into offgas (C1-C4), naphta (C5-C10), kerosene (C10-C14), diesel (C15-C22) and wax cuts (C22+) 34

. This column is simulated by the rigorous PetroFrac model in Aspen Plus. The main tower

comprises 66 theoretical stages, two side strippers with 8 stages each (with liquid draw at stages 8 and 20 of the main tower) and a partial condenser operating at 40 °C and 1.31bar. Steam is injected into the last stage of each tower at a ratio of 2.267 kg of steam per barrel (bbl) of product. The feed stream – composed by the light synoil from SBCR (SYNOIL) and the hydrocracker product (CRACKOIL) – is preheated to 99% of the vapor fraction at 1.45 bar in a furnace (FUR1) and enters the ADC1 flash zone at stage 65. Naphtha is obtained in the top distillate, while Kerosene and diesel are obtained in the bottom of each stripping tower, and wax is obtained in the bottom of the main column. Around 73 % of the wax stream is recycled to the upgrading unit for cracking into lighter fractions.


18

Page 19 of 57 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 – Process flow diagram for the separation unity. 2.6. Rankine cycle

The Rankine cycle (RC) unit is responsible for recovering the excess thermal energy produced by the exothermic FT reactions in the SBCR (Figure 4). The SBCR produces saturated vapor at 230°C 33 (FTRHX, Figure 8), which is used in the RC unit to produce electricity. Figure 8 shows the proposed process flow diagram of the RC unit. The saturated vapor produced in the SBCR (represented as the heater FTRHX) is superheated by the combustion of the streams OFFGAS2 and OFFGAS4, and it is then expanded in a turbine to produce electricity. Next, the saturated vapor (stream SS33) condenses in the heat exchanger HX9 and is compressed to 28 [bar]. Lastly, the compressed liquid exchanges energy with the syngas stream (HXR1B), before feeding back into the FT reactor (FTRHX).


19


Page 20 of 57

Figure 8 – Process flow diagram of the Rankine cycle unit 3. Economic assessment method

A detailed economic analysis is carried out according to the method presented by Baliban et al. 16

. The capital cost is adjusted based on cost indices for the first quarter of 2017 and considers

that the plant will be built in Brazil by adopting a factor of 1.4. The values of the product prices are estimated for the Brazilian scenario for the same period. The cash flow after taxes (CFAT, Equation 13) is calculated over 30 years of operation, in which the total permanent investment (TPI, Table 1) is divided among the first three years using distribution factors @Q , @Q , @Q of 0.25, 0.5 and 0.25, respectively. After Year 3, CFAT is calculated based on the revenues (R), the total operating cost (TOC, showed in Table 2) and the capital depreciation (DEP), where t is the Brazilian tax rates 45 (see Table 4). DEP is calculated by the straight-line method, considering a period of 10 years and salvage value of 20% of TPI, after 30 years of operation. After Year 13, CFAT is a function of R and TOC only, until the last year of operation when the salvage value is recovered.

DA]^ = @Q ^ _ =1 + 0.05? P = 1, 2, 3

(13.a)


20

Page 21 of 57 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


DA]^ = = − ^`D? =1 − a? + a bc P = 4 … 13

DA]^ = = − ^`D? =1 − a? P = 14 … 29 DA]^ = = − ^`D? =1 − a? + ^ _ 0.20 P = 30

(13.b) (13.c) (13.d)

Table 1 – Procedure to compute the total permanent investment TEC - Total equipment cost TEC1 LF - Location factor, Brazil 1.4 DPI - Direct permanent investment cost Brazil TEC × BF G&A - Capital overhead DPI × 0.03 CF - Contract fees DPI × 0.03 TDC - Total depreciable cost DPI + G&A + CF CC - Capital contingences TDC × 0.18 TPI - Total permanent investment TDC + CC 1 TEC - total equipment cost, detailed in Supporting Information material (Table S5) Table 2 - Procedure to compute the total operating cost AMC - Annual maintenance cost TPI × 0.04 45 * NO - Number of operators 4.5 × (6.29 + 0.23 × NEQ )0.5 LC - Labor cost NO × labor price OL&M – Operating labor and maintenance AMC + LC RM - Raw materials RM UT - Utilities UT SOC - Subtotal operating cost OL&M + RM + UT G&A - Operating expenses SOC × 0.08 PO - Plant Overhead OL&M × 0.5 TOC - Total operating cost SOC + G&A + PO *

NEQ – number of equipment

The operation starts in Year 4, with a yearly operating capacity (YOP) of 8,000 hours. The annual value for revenue (R), utilities (UT) and raw materials (RM) are computed using the YOP and setting their flow rates and prices according to the Brazilian scenario during the first quarter of 2017 (Supporting Information material, Table S3). These values are adjusted in the respective operating year using escalator factors listed in Table 3.


21


Page 22 of 57

Table 3 – Additional economic parameters Products scalation Raw material scalation Utilities escalation Labor escalation t - Tax rate 46 RR - Rate of return

1 1 1 1 32 15

[%/year] [%/year] [%/year] [%/year] [%/year] [%/year]

The net present value (NPV) accounts for the cash flow after taxes (CFAT) and a desired rate of return (RR, Table 3), as follows:

e f = %

DA]^ =1 + ?

(14)

4. Results and Discussion 4.1. Base case

The proposed plant processes 27.22 kg/s of natural gas with 30 mol% fraction of CO2, which corresponds to approximately 45 wt% of the feed stream, and it produces 9.58 kg/s (6,750 bbl/day) of FTL. This production breaks down into: 2.60 kg/s of gasoline, 2.66 kg/s of kerosene, 3.31 kg/s of diesel, and 1.01 kg/s of waxes. Table 4 gives the mass flow rate for the main streams referenced in Figure 1. Notice also that the TR reactor consumes 25% of the feed CO2, producing a syngas stream with a H2:CO ratio of 1.57. This syngas is mixed with hydrogen from the WGS unit to bump the H2:CO ratio to 2. About 37 wt% of the products from the FT unit is light synoil that can be sent directly to the separation unit, while the other 63 wt% is mixed with the wax recycle before upgrading. The hydrocracking reactor is responsible for the production of 43 wt%, 57 wt%, 46 wt% and 17 wt% of the total C5-C9, C10-C14, C15-C22 and C23-C29 FTL produced.


22

Page 23 of 57 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 4 – Component mass flow rate [kg/s] in the main stream of the integrated TR+FT process.

Component

CO2 FT-light FT-heavy Syngas rich NG synoil synoil

Stream HC-light wax gasoline kerosene diesel wax synoil recycle

H2

—

3.28

—

—

—

—

—

—

—

—

CO

—

29.06

0.01

—

—

—

—

—

—

—

CO2

12.10

9.08

0.11

0.02

0.05

—

—

—

—

—

H 2O

—

9.26

0.02

0.01

0.01

—

0.02

—

—

—

C1-C4

14.02

0.07

0.19

0.01

0.25

—

0.20

—

—

—

C5-C9

0.91

—

1.49

0.04

1.44

—

2.37

0.32

—

—

C10-C14

—

—

0.99

0.15

1.68

0.01

0.01

2.34

0.31

—

C15-C22

—

—

0.81

0.96

2.48

0.29

—

—

2.89

0.10

C23-C29

—

—

0.15

1.46

1.78

1.33

—

—

0.11

0.49

C30+

—

—

0.03

3.74

1.54

1.15

—

—

—

0.42

Total

27.03

50.75

3.80

6.39

9.23

2.78

2.60

2.66

3.31

1.01

One of the dominant operating costs associated with the proposed plant is related to the electricity required to compress the syngas feed to the SBCR. Therefore, the cogeneration of electricity is crucial to make this process economically viable. Overall, the production of 1 kg of FTL product requires 3.22 MJ of electricity, 4.47 MJ of heating energy, and 27.95 MJ of cooling energy. •

97.7% of this electricity is consumed by the compressor (C1) in the FT unit (Figure 4). However, 99.8% of the plant electricity requirement is covered by the RC unit.

•

53.7% of the heating energy is consumed by the furnaces located in the separation and upgrading units, 44.8% is used as high-pressure steam in the heat exchanger HX5 (Figure 6), and the remaining 1.5% is used as medium-pressure steam in other units. About 7% of the total heat requirement is covered by medium-pressure steam produced in the WGS reactor.


23


•

Page 24 of 57

99.1% of the cooling energy is provided by cooling water at 20°C.

The large amount of CO2 in the natural gas feedstock results in a reduction of the H:C ratio, and therefore has a negative impact on the overall carbon conversion. The presence of CO2 in the FT reactor feed also affects the reaction rates by decreasing the partial pressure of H2 and CO, and hence its overall performance. The overall carbon conversion of NG to liquid products obtained in the proposed plant is 54%, which is vastly superior to typical biomass or hybrid biomass/coal to liquids processes (e.g. 33%)16, but remains inferior to conventional gas to liquids processes using high quality natural gas (e.g. 68%)47. The global warming potential (GWP) obtained in the proposed process is 2.69, 2.08, 3.12 and 3.80 kg(CO2-eq) per kg of gasoline, kerosene, diesel and wax produced, respectively. 89.7% of the total GWP is caused by direct emissions from the waste streams WW1, WW2, WW3, OFFGAS5 and OFFGAS6; the remaining fraction is attributed to indirect emissions from the utilities, of which 55% comes from the NG burned in furnaces within the separation and upgrading units, 39% is produced by the generation of high pressure steam used in the WGS unit (HX5, Figure 5), and the last 6% is distributed among refrigeration and medium-pressure steam production. The RC unit plays a key role in the reduction of the GWP, reducing the emissions by about 10%. At this point, it is important to summarize the advantages of the proposed process coupling TR and FT synthesis over a conventional GTL plant using auto-thermal reforming.21 Firstly, the proposed TR+FT process does not require sweetening of the NG, which reduces both the capital and operational costs. It also mitigates the GWP by reducing the direct CO2 emissions by 8.7%, in comparison with the integration of a conventional separation process (with 95% methane recovery and 2% of CO2 in the product stream48) and a GTL plant (with 68% of carbon


24

Page 25 of 57 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


conversion efficiency). Lastly, direct processing of the CO2-rich NG reduces the raw material cost, since this NG is estimated to be 42% cheaper than conventional NG due to its lesser Lower Heating Value.

4.2. Analysis in the Brazilian scenario

Figure 9 depicts the monetary distribution of the sales, raw materials and utilities. Diesel is the main product, making for almost 40% of the sales, followed by gasoline, kerosene and wax. The cost breakdown in terms of raw materials confirms that natural gas is the dominant cost (86%). This large contribution is due to the high price of Brazilian NG – 5.28$/MMBtu in February 2017 – which is nearly twice as expensive as United States (US) NG – 2.85$/MMBtu over the same period

49

. The cost breakdown in terms of utilities identifies cooling water as the main

contributor with 40%, while the electricity accounts for 0.5% only. It is noteworthy that the energy recovery through the RC unit reduces the plant electricity requirement by 99.8%, from 30.9 MW down to a mere 57 kW. Without the RC unit, the electricity cost would instead represent 73.6% of the utilities, leading to a negative cash flow over the entire process lifetime.

Figure 9 – Monetary breakdown: (a) Sales; (b) raw material cost; (c) utilities cost. Figure 10 shows the investment cost breakdown per process unit. The FT unit is the most expensive one, representing 60% of the total investment cost. Only two pieces of equipment in


25


Page 26 of 57

this FT unit correspond to more than 98% of the cost – the FT reactor (59.9%) and the compressors (38.2%). These figures show the importance that the gas compression can play in such plants, since the investment cost associated to these compressors represents nearly 25% of the total investment and the required electricity would lead to a negative cash flow rate without the inclusion of the RC unit. For this reason, the development of new FT catalysts that are effective at lower pressures is critical from an economic standpoint. The hydrocracking unit is the second most expensive, accounting for 29% of the total investment cost, with the HCR as the most expensive equipment (79.2%). Overall, four out of the six units (TR, WGS, separation, RC) represent less than 10% of the total equipment cost, while the three major pieces of equipment (SBCR, compressor, HCR) account for over 80% of the TEC. It is important to note that the investment cost associated with the syngas production is small due to the use of the tri-reforming process. The reactor design in the TR unit holds similarities with an auto-thermal reforming reactor, which consists basically of a combustion chamber followed by a relatively small adiabatic catalyst bed (1 meter of diameter by 2 meters of length). Another factor contributing to the cost reduction in the TR unit is the absence of a syngas cleanup area, due to the low concentrations of acid gases in the syngas, less than 7 vol% CO2 and complete removal of H2S from the NG before its processing (the H2S removal cost is already factored in the NG price).


26

Page 27 of 57 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 10 – Investment cost breakdown by process unit. The cash flow over the 30 years of operation is shown in Figure 11.a. The cumulative cash flow becomes positive after about 16 years of operation. However, the positive cash flow – over the operating years – is not sufficient to balance the investment cost at the desired rate of return (15%), resulting in a negative NPV of -142 MMUS$ after 30 years of operation (see Figure 11.b). Under these conditions, it is possible to estimate an internal rate of return of 5.58% for the proposed plant in the Brazilian scenario.

(b) (a) Figure 11 – Parameters of the economic analysis in the Brazilian scenario: (a) Cash flow after taxes over the operating period; (b) Net present value considering rates of return of 15% and 5.58% a year. The total permanent investment of the proposed plant is about 42,000 US$/bbl/day. This investment is comparable to the installed cost reported by Sasol back in 2001,50,51 namely 47,000 US$/bbl/day after correction to the Brazilian scenario of 2017. By contrast, the raw material cost of 53.1 US$/bbl is quite high in comparison to previous studies reporting 3.7 times cheaper values.52 Consequently, the predicted annual cash flow is unlikely to be sufficient for overcoming the investment cost under the current Brazilian scenario. This trend is expected to get worse after recent contracts for exploiting new NG reserves, which could increase the NG price by up to 40%.2 In this context, the GTL technology is unlikely to be economically viable in


27


Page 28 of 57

the Brazilian scenario, despite the fact that similar conversion efficiencies as conventional GTL processes could be achieved with the TR+FT process for CO2-rich NG. The NG price that would be needed to obtain an NPV break-even in the Brazilian scenario is 3.28 US$/MMBtu, which remains 15% more expensive than US NG over the same period. This analysis suggests that the implementation of the proposed plant in other locations could make the process economically viable. In locations where lean NG is available, the required CO2 in the feed composition could even be provided by recycling part of the produced CO2, thereby increasing the overall carbon conversion efficiency and reducing the GWP simultaneously.

4.3. Analysis in the US scenario

In this section, we evaluate the viability of the proposed plant in the US, which provides a more favorable scenario due to lower NG and electricity prices, and lower investment cost. The previous economic assessment is modified by reducing the location factor (LC) from 1.4 to 1 and increasing the tax rate (t) from 32% to 40%. The raw material and product prices are also changed to the US scenario during the same period (see Table S4, Supporting Information material). Figure 12.a compares the cash flow after taxes between the Brazilian and US scenarios. Clearly, the US scenario presents a higher annual cash flow, with a dramatic reduction of the period over which the cumulative cash flow is negative, from 16 to just 5 years. Figure 12.b compares the NPV for the Brazilian and US scenarios. In the US, the proposed plant presents a positive NPV of 187 MMUS$ after 30 years, with a payback period of 7.8 years and an internal rate of return of 27.6%. The total investment cost (TIC) is approximately 30,000 US$/bbl/day, while the maximum TIC estimated still to obtain an economically viable plant would be about 59,000 US$/bbl/day. These results therefore suggest that the proposed plant


28

Page 29 of 57 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


could be economically viable in a favorable scenario, whereby the NG price is low in comparison to that of the LTF products.

(b) (a) Figure 12 - Parameters of the economic analysis comparison between Brazil and the US: (a) Cash flow after taxes over the operating period for Brazil and the US; (b) Net present value for the Brazilian and US scenarios.

5. Conclusions

This paper has presented an alternative technology to monetize CO2-rich NG by coupling the tri-reforming and Fischer-Tropsch technologies. The conceptual plant is designed such that the excess thermal energy produced in the FT reactor can be converted to electricity in a RC unit, covering almost 100% of the plant electricity needs. Furthermore, the results show that it is technically possible to convert NG with about 30 mol% of CO2 in a plant with 54% carbon conversion efficiency. The detailed economic analysis has shown that the viability of the proposed process is highly dependent on the NG price. It suggests that the proposed plant may not be economically feasible in the Brazilian scenario, considering economic data from the first quarter of 2017. Conversely, the proposed plant presents a positive NPV in the more favorable US scenario where NG currently is cheaper. The developed plantwide model provides a useful tool to assess the integrated TR+FT process under different operating conditions. As part of future work, an analysis of the effect of model, process and external uncertainties will be carried out. Moreover, a plantwide optimization with


29


Page 30 of 57

respect to environmental and/or economic objectives will be conducted, which may bring better results than with standard operating conditions from the literature or based on the optimization of each unit separately. Acknowledgements

The authors would like to thank the sponsorship of Shell and FAPESP through the “Research Centre for Gas Innovation - RCGI” (FAPESP Proc. 2014/50279-4), hosted by the University of Sao Paulo, and the strategic importance of the support given by ANP (Brazil’s National Oil, Natural Gas and Biofuels Agency) through the R&D levy regulation, as well the fellowship from CNPq Brasil. Financial support from the Sustainable Gas Institute (SGI) at Imperial College London is also gratefully acknowledged. Supporting information

Section S1 provides the kinetic parameters used in the simulation of the tri-reforming reactor. Section S2 lists the parameters used to conduct the economic assessment, including raw material cost and product prices in both the Brazil and US scenarios as of February 2017, and a complete list of equipment for each unit. This information is available free of charge from http://pubs.acs.org/

Acronyms

ADC AMC ASF ATR CC CF CFAT DEP

atmospheric distillation column annual maintenance cost Anderson-Schultz-Flory auto-thermal reforming capital contingences contract fees cash flow after taxes capital depreciation


30

Page 31 of 57 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


DME DPI EOR FBR FT G&A GTL GWP HCR HTFT LC LF LHHW LTF LTFT NEQ NG NO NPV OL&M PO PSRK RC RM RR SBCR SOC TDC TEC TOC TPI TR UT WGS WHSV YOP

dimethyl ether direct permanent investment cost enhanced oil recovery fixed-bed reactor Fischer-Tropsch capital overhead gas to liquids global warming potential hydrocracking reactors high temperature Fischer-Tropsch synthesis labor cost location factor Langmuir-Hinshelwood-Hougen-Watson liquid transportation fuels low temperature Fischer-Tropsch synthesis number of equipment natural gas number of operators net present value operating labor and maintenance Plant Overhead predictive Soave-Redlich-Kwong Rankine cycle raw materials rate of return slurry bubble column reactor subtotal operating cost total depreciable cost total equipment cost total operating cost total permanent investment tri-reforming utilities water gas shift weight hourly space velocity yearly operating capacity


31


Page 32 of 57

References (1)

Boletim mensal de acompanhamento da indústria de gás natural - Abril 2007 http://www.mme.gov.br/web/guest/secretarias/petroleo-gas-natural-e-combustiveisrenovaveis/publicacoes/boletim-mensal-de-acompanhamento-da-industria-de-gas-natural (accessed Jan 8, 2018).

(2)

Boletim Mensal de Acompanhamento da Indústria de Gás Natural - Fevereiro 2017 http://www.mme.gov.br/web/guest/secretarias/petroleo-gas-natural-e-combustiveisrenovaveis/publicacoes/boletim-mensal-de-acompanhamento-da-industria-de-gas-natural (accessed Jan 8, 2018).

(3)

Gaffney, C. & A. Exame e Avaliação de Dez Descobertas e Prospectos Selecionadas no Play

do

Pré-sal

em

Águas

Profundas

na

Bacia

de

Santos,

Brasil

www.anp.gov.br/SITE/acao/download/?id=33422 (accessed Jan 9, 2018). (4)

de Queiroz, F. A.; de Medeiros, J. L.; Alves, R. M. B. CO2 Utilization: A Process Systems Engineering Vision. In CO2 Sequestration and Valorization; InTech, 2014.

(5)

Lee, S.-H.; Cho, W.; Ju, W.-S.; Cho, B.-H.; Lee, Y.-C.; Baek, Y.-S. Tri-Reforming of CH4 Using CO2 for Production of Synthesis Gas to Dimethyl Ether. Catal. Today 2003, 87 (1), 133–137.

(6)

Song, C. Tri-Reforming: A New Process for Reducing CO2 Emissions. Chem. Innov. 2001, 31 (1), 21–26.

(7)

Farniaei, M.; Abbasi, M.; Rahnama, H.; Rahimpour, M. R.; Shariati, A. Syngas Production in a Novel Methane Dry Reformer by Utilizing of Tri-Reforming Process for


32

Page 33 of 57 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


Energy Supplying: Modeling and Simulation. J. Nat. Gas Sci. Eng. 2014, 20, 132–146. (8)

Rahnama, H.; Farniaei, M.; Abbasi, M.; Rahimpour, M. R. Modeling of Synthesis Gas and Hydrogen Production in a Thermally Coupling of Steam and Tri-Reforming of Methane with Membranes. J. Ind. Eng. Chem. 2014, 20 (4), 1779–1792.

(9)

Rahimpour, M. R.; Arab Aboosadi, Z.; Jahanmiri, A. H. Synthesis Gas Production in a Novel Hydrogen and Oxygen Perm-Selective Membranes Tri-Reformer for Methanol Production. J. Nat. Gas Sci. Eng. 2012, 9, 149–159.

(10)

Díez-Ramírez, J.; Dorado, F.; Martínez-Valiente, A.; García-Vargas, J. M.; Sánchez, P. Kinetic, Energetic and Exergetic Approach to the Methane Tri-Reforming Process. Int. J. Hydrogen Energy 2016, 41 (42), 19339–19348.

(11)

Zhang, Y.; Cruz, J.; Zhang, S.; Lou, H. H.; Benson, T. J. Process Simulation and Optimization of Methanol Production Coupled to Tri-Reforming Process. Int. J. Hydrogen Energy 2013, 38 (31), 13617–13630.

(12)

Zhang, Y.; Zhang, S.; Benson, T. A Conceptual Design by Integrating Dimethyl Ether (DME) Production with Tri-Reforming Process for CO2 Emission Reduction. Fuel Process. Technol. 2015, 131, 7–13.

(13)

Minutillo, M.; Perna, A. A Novel Approach for Treatment of CO2 from Fossil Fired Power Plants. Part B: The Energy Suitability of Integrated Tri-Reforming Power Plants (ITRPPs) for Methanol Production. Int. J. Hydrogen Energy 2010, 35 (13), 7012–7020.

(14)

Gill, S. S.; Tsolakis, A.; Dearn, K. D.; Rodríguez-Fernández, J. Combustion Characteristics and Emissions of Fischer–Tropsch Diesel Fuels in IC Engines. Prog.


33


Page 34 of 57

Energy Combust. Sci. 2011, 37 (4), 503–523. (15)

Floudas, C. A.; Elia, J. A.; Baliban, R. C. Hybrid and Single Feedstock Energy Processes for Liquid Transportation Fuels: A Critical Review. Comput. Chem. Eng. 2012, 41, 24–51.

(16)

Baliban, R. C.; Elia, J. A.; Floudas, C. A. Toward Novel Hybrid Biomass, Coal, and Natural Gas Processes for Satisfying Current Transportation Fuel Demands, 1: Process Alternatives, Gasification Modeling, Process Simulation, and Economic Analysis. Ind. Eng. Chem. Res. 2010, 49 (16), 7343–7370.

(17)

Baliban, R. C.; Elia, J. A.; Floudas, C. A. Optimization Framework for the Simultaneous Process Synthesis, Heat and Power Integration of a Thermochemical Hybrid Biomass, Coal, and Natural Gas Facility. Comput. Chem. Eng. 2011, 35 (9), 1647–1690.

(18)

Baliban, R. C.; Elia, J. A.; Misener, R.; Floudas, C. A. Global Optimization of a MINLP Process Synthesis Model for Thermochemical Based Conversion of Hybrid Coal, Biomass, and Natural Gas to Liquid Fuels. Comput. Chem. Eng. 2012, 42, 64–86.

(19)

Song, C.; Pan, W. Tri-Reforming of Methane: A Novel Concept for Catalytic Production of Industrially Useful Synthesis Gas with Desired H2/CO Ratios. Catal. Today 2004, 98 (4), 463–484.

(20)

Cho, W.; Song, T.; Mitsos, A.; McKinnon, J. T.; Ko, G. H.; Tolsma, J. E.; Denholm, D.; Park, T. Optimal Design and Operation of a Natural Gas Tri-Reforming Reactor for DME Synthesis. Catal. Today 2009, 139 (4), 261–267.

(21)

Arab Aboosadi, Z.; Jahanmiri, A. H.; Rahimpour, M. R. Optimization of Tri-Reformer Reactor to Produce Synthesis Gas for Methanol Production Using Differential Evolution


34

Page 35 of 57 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


(DE) Method. Appl. Energy 2011, 88 (8), 2691–2701. (22)

De Groote, A. M.; Froment, G. F.; Kobylinski, T. Synthesis Gas Production from Natural Gas in a Fixed Bed Reactor with Reversed Flow. Can. J. Chem. Eng. 1996, 74 (5), 735– 742.

(23)

Kim, Y. H.; Jun, K.-W.; Joo, H.; Han, C.; Song, I. K. A Simulation Study on Gas-toLiquid (Natural Gas to Fischer–Tropsch Synthetic Fuel) Process Optimization. Chem. Eng. J. 2009, 155 (1–2), 427–432.

(24)

Pondini, M.; Ebert, M. Process Synthesis and Design of Low Temperature FischerTropsch Crude Production from Biomass Derived Syngas, Master's thesis, Chalmers University of Technology, Göteborg, Sweden, 2013.

(25)

van Vliet, O. P. R.; Faaij, A. P. C.; Turkenburg, W. C. Fischer–Tropsch Diesel Production in a Well-to-Wheel Perspective: A Carbon, Energy Flow and Cost Analysis. Energy Convers. Manag. 2009, 50 (4), 855–876.

(26)

Leckel, D. Diesel Production from Fischer−Tropsch: The Past, the Present, and New Concepts. Energy & Fuels 2009, 23 (5), 2342–2358.

(27)

Guettel, R.; Turek, T. Comparison of Different Reactor Types for Low Temperature Fischer–Tropsch Synthesis: A Simulation Study. Chem. Eng. Sci. 2009, 64 (5), 955–964.

(28)

De Deugd, R. M. Fischer-Tropsch Synthesis Revisited; Efficiency and Selectivity Benefits from Imposing Temporal And/or Spatial Structure in the Reactor; TU Delft, Delft University of Technology, 2004.

(29)

Todic, B.; Bhatelia, T.; Froment, G. F.; Ma, W.; Jacobs, G.; Davis, B. H.; Bukur, D. B.


35


Page 36 of 57

Kinetic Model of Fischer–Tropsch Synthesis in a Slurry Reactor on Co–Re/Al 2 O 3 Catalyst. Ind. Eng. Chem. Res. 2013, 52 (2), 669–679. (30)

Visconti, C. G.; Tronconi, E.; Lietti, L.; Zennaro, R.; Forzatti, P. Development of a Complete Kinetic Model for the Fischer–Tropsch Synthesis over Co/Al2O3 Catalysts. Chem. Eng. Sci. 2007, 62 (18), 5338–5343.

(31)

Sudiro, M.; Bertucco, A. Synthetic Fuels by a Limited CO 2 Emission Process Which Uses Both Fossil and Solar Energy. Energy & Fuels 2007, 21 (6), 3668–3675.

(32)

Elia, J. A.; Baliban, R. C.; Floudas, C. A. Toward Novel Hybrid Biomass, Coal, and Natural Gas Processes for Satisfying Current Transportation Fuel Demands, 2: Simultaneous Heat and Power Integration. Ind. Eng. Chem. Res. 2010, 49 (16), 7371– 7388.

(33)

Maretto, C. Modelling of a Bubble Column Slurry Reactor for Fischer–Tropsch Synthesis. Catal. Today 1999, 52 (2–3), 279–289.

(34)

De Klerk, A. Fischer-Tropsch Refining, University of Pretoria Pretoria South Africa, 2008.

(35)

Bhutani, N.; Ray, A. K.; Rangaiah, G. P. Modeling, Simulation, and Multi-Objective Optimization of an Industrial Hydrocracking Unit. Ind. Eng. Chem. Res. 2006, 45 (4), 1354–1372.

(36)

Mohanty, S.; Saraf, D. N.; Kunzru, D. Modeling of a Hydrocracking Reactor. Fuel Process. Technol. 1991, 29 (1–2), 1–17.

(37)

Leckel, D.; Liwanga-Ehumbu, M. Diesel-Selective Hydrocracking of an Iron-Based


36

Page 37 of 57 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


Fischer−Tropsch Wax Fraction (C 15 −C 45 ) Using a MoO 3 -Modified Noble Metal Catalyst. Energy & Fuels 2006, 20 (6), 2330–2336. (38)

Smith, R. J.; Loganathan, M.; Shantha, M. S.; others. A Review of the Water Gas Shift Reaction Kinetics. Int. J. Chem. React. Eng. 2010, 8 (1).

(39)

Pasel, J.; Samsun, R. C.; Schmitt, D.; Peters, R.; Stolten, D. Test of a Water–gas-Shift Reactor on a 3kWe-Scale—design Points for High- and Low-Temperature Shift Reaction. J. Power Sources 2005, 152, 189–195.

(40)

Green, D. W.; Perry, R. H. Perry’s Chemical Engineers’ Handbook/edición Don W. Green Y Robert H. Perry.; 1973.

(41)

CHEN, W.; LIN, M.; JIANG, T.; CHEN, M. Modeling and Simulation of Hydrogen Generation from High-Temperature and Low-Temperature Water Gas Shift Reactions. Int. J. Hydrogen Energy 2008, 33 (22), 6644–6656.

(42)

Moe, J. M. Design of Water-Gas Shift Reactors. Chem. Eng. Prog. 1962, 58 (3).

(43)

Seo, Y.-S.; Seo, D.-J.; Seo, Y.-T.; Yoon, W.-L. Investigation of the Characteristics of a Compact Steam Reformer Integrated with a Water-Gas Shift Reactor. J. Power Sources 2006, 161 (2), 1208–1216.

(44)

Liu, F.; Zhang, N. Strategy of Purifier Selection and Integration in Hydrogen Networks. Chem. Eng. Res. Des. 2004, 82 (10), 1315–1330.

(45)

Turton, R.; Bailie, R. C.; Whiting, W. B.; Shaeiwitz, J. A. Analysis, Synthesis and Design of Chemical Processes, 3o.; Prentice Hall PTR: Boston, 2012.


37


(46)

Page 38 of 57

Contribuição Social sobre o Lucro Líquido CSLL — Secretaria da Receita Federal do Brasil http://idg.receita.fazenda.gov.br/acesso-rapido/tributos/CSLL (accessed Jan 9, 2018).

(47)

O’Rear, D. J. Reduction of Carbon Dioxide Emissions from Fischer-Tropsch GTL Facility by Aromatics Production. US 6,693,138 B2, 2004.

(48)

Qi, R.; Henson, M. A. Membrane System Design for Multicomponent Gas Mixtures via Mixed-Integer Nonlinear Programming. Comput. Chem. Eng. 2000, 24 (12), 2719–2737.

(49)

Henry

Hub

Natural

Gas

Spot

Price

(Dollars

per

Million

Btu)

https://www.eia.gov/dnav/ng/hist/rngwhhdM.htm (accessed Jan 4, 2018). (50)

Fleisch, T.; Sills, R.; Briscoe, M. A Review of Global GTL Developments. J. Nat. Gas Chem. 2002, 11, 1–14.

(51)

Vosloo, A. C. Fischer-Tropsch: A Futuristic View. Fuel Process. Technol. 2001, 71 (1–3), 149–155.

(52)

Wood, D. A.; Nwaoha, C.; Towler, B. F. Gas-to-Liquids (GTL): A Review of an Industry Offering Several Routes for Monetizing Natural Gas. J. Nat. Gas Sci. Eng. 2012, 9, 196– 208.


38

Page 39 of 57 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


39

Industrial & Engineering Chemistry Research 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17


Page 40 of 57

Page 41 of 57 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 – Conceptual design of the integrated TR+FT process 500x179mm (96 x 96 DPI)



Figure 2 – Process flow diagram of the tri-reforming unit 277x78mm (96 x 96 DPI)


Page 42 of 57

Page 43 of 57 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 - Parametric analysis of the H2O:NG and O2:NG feed ratios on: (a) H2:CO ratio, (b) H2 flow rate, and (c) conversion of CO2 (Note: the axes H2O:NG and O2:NG are inverted in Figure 3.c to better visualize the profile). 107x83mm (300 x 300 DPI)



Figure 3 - Parametric analysis of the H2O:NG and O2:NG feed ratios on: (a) H2:CO ratio, (b) H2 flow rate, and (c) conversion of CO2 (Note: the axes H2O:NG and O2:NG are inverted in Figure 3.c to better visualize the profile). 105x77mm (300 x 300 DPI)


Page 44 of 57

Page 45 of 57 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 - Parametric analysis of the H2O:NG and O2:NG feed ratios on: (a) H2:CO ratio, (b) H2 flow rate, and (c) conversion of CO2 (Note: the axes H2O:NG and O2:NG are inverted in Figure 3.c to better visualize the profile).. 104x75mm (300 x 300 DPI)



Figure 4. Process flow diagram of the Fischer-Tropsch synthesis unit 313x98mm (96 x 96 DPI)


Page 46 of 57

Page 47 of 57 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 – Process flow diagram of the upgrading unit 289x106mm (96 x 96 DPI)



Figure 6 – Process flow diagram of the water-gas shift unit 339x93mm (96 x 96 DPI)


Page 48 of 57

Page 49 of 57 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 – Process flow diagram for the separation unity 273x152mm (96 x 96 DPI)



Figure 8 – Process flow diagram of the Rankine cycle unit 283x139mm (96 x 96 DPI)


Page 50 of 57

Page 51 of 57 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 9 – Monetary breakdown: (a) Sales; (b) raw material cost; (c) utilities cost. 237x100mm (96 x 96 DPI)



Figure 10 – Investment cost breakdown by process unit. 89x101mm (96 x 96 DPI)


Page 52 of 57

Page 53 of 57 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 11 – Parameters of the economic analysis in the Brazilian scenario: (a) Cash flow after taxes over the operating period; (b) Net present value considering rates of return of 15% and 5.58% a year. 174x80mm (96 x 96 DPI)



Figure 11 – Parameters of the economic analysis in the Brazilian scenario: (a) Cash flow after taxes over the operating period; (b) Net present value considering rates of return of 15% and 5.58% a year. 212x92mm (96 x 96 DPI)


Page 54 of 57

Page 55 of 57 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 12 - Parameters of the economic analysis comparison between Brazil and the US: (a) Cash flow after taxes over the operating period for Brazil and the US; (b) Net present value for the Brazilian and US scenarios. 158x72mm (96 x 96 DPI)



Figure 12 - Parameters of the economic analysis comparison between Brazil and the US: (a) Cash flow after taxes over the operating period for Brazil and the US; (b) Net present value for the Brazilian and US scenarios. 190x82mm (96 x 96 DPI)


Page 56 of 57

Page 57 of 57 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


For Table of Contents Only 44x23mm (300 x 300 DPI)


Conversion of CO2-Rich Natural Gas ... - ACS Publications

Recommend Documents