Plant-wide Simulation of an Integrated Zero-Emission Process to ‎Convert Flare Gas to Gasoline

1. Introduction

Humans need energy to power our technology and to broaden and improve quality of life (Fawole, Cai, & MacKenzie, 2016). Natural gas will continue over coming decades to provide an essential component of the global primary energy mix providing economic growth in many developing and developed countries (Tabak, Chitnis, McGihon, & Zhao, 2009). Until now, there are primarily two commercially viable methods for converting natural gas into liquid fuels (GTL) through intermediate gasification processes (Fig. 1). These are: 1) Fischer-Tropsch Synthesis (FTS), discovered in Germany in the 1920’s using coal as feedstock. FTS can produce a range of valuable hydrocarbon liquids and waxes from various hydrocarbon feedstock (e.g., coal, natural gas or biomass) using different catalysts and pressure-temperature condition. 2) Hydrocarbon conversion to gasoline (HTG) processes with methanol production as an intermediate step are established, but less extensively commercialized than the FTS methods.

Minimizing or eliminating emissions of carbon dioxide (CO₂), a major greenhouse gas (GHG) arising from anthropogenic activities, is also now a priority for most hydrocarbon processing technologies, in order to inhibit its potential climate change impacts. Gas flaring is the process of burning-off associated gas from producing oil wells, hydrocarbon processing plants or refineries, either as a means of disposal or as a safety measure to relieve pressure (Emam, 2015). Flaring makes a significant contribution to CO₂ emissions.

Gas flaring involves the rapid oxidation and combustion of the component gases contained in natural gas leading to the release of CO₂ into the atmosphere. The composition and quantity of flare gas determines the amount of CO₂ and other combustion products emitted as pollutants into the atmosphere. In addition, the prevailing meteorological conditions, flare size and design and other combustion variables also determine the exact quantity of CO₂ emitted during the flaring process (Giwa, Nwaokocha, Kuye, & Adama, 2013). Hydrocarbons, including natural gas, burnt in thermal power plants provide a major source of CO₂ emissions entering the atmosphere, and reducing those emissions has become globally one of our major challenges in the twenty-first  century (Er-rbib, Bouallou, & Werkoff, 2012). Methane (CH₄), the major constituent of natural gas is the second most abundant GHGs, after CO₂, but with far more potent greenhouse impacts on the upper atmosphere, making the venting and fugitive emissions of natural gas even more important to avoid. Identifying methods that can reduce the concentration of methane and CO₂ released to the atmosphere is therefore a priority for many researchers. Converting methane to alternative forms more-easily handled energy, such as methanol, is one potential method for achieving this [6].

Several methods have been proposed for conversion of flare gas to valuable products based on two categories a) separation b) conversion:

1- Power generation (conversion mode, (Heidari, Ataei, & Rahdar, 2016; Ojijiagwo, Oduoza, & Emekwuru, 2016))

2- LNG production (separation mode, (Soltanieh, Zohrabian, Gholipour, & Kalnay, 2016))

3- CNG production (separation mode, (Soltanieh et al., 2016))

4- LPG production (separation mode, (Hajizadeh, Mohamadi-Baghmolaei, Azin, Osfouri, & Heydari, 2018))

5- GTL process (conversion mode, (Wood, Nwaoha, & Towler, 2012))

6- GTG process (conversion mode)

However, as a result of the increasing global demand for transport fuels and the adverse effects of fluctuating energy prices, searching for alternatives to crude oil has also become a priority (Fu, Chang, Shao, & Li, 2017). Furthermore, Gasoline is an important liquid hydrocarbon-based fuel derived primarily from fractional distillation of the petroleum fractions in crude oil conducted in refineries (Galadima & Muraza, 2015). Therefore, natural gas to gasoline (GTG) process is one of the HTL and GTL technologies of particular interest to the energy sector. Fischer-Tropsch Synthesis (FTS) and Gas to Gasoline (GTG) are the two main GTL technologies being evaluated commercially at this time [9].

A flow diagram distinguishing the component processes of the two primary GTL technologies is illustrated in Fig. 2. According to this figure, a synthesis gas production unit is the primary stage for both technologies, but their subsequent process stages are distinct. The main difference between FTS and GTG are their hydrocarbon products. Naphtha, diesel and long-chain hydrocarbon waxes are the main products of FTS [10], whereas gasoline is the primary hydrocarbon product of GTG. Table 1 list of final products and composition of two GTL processes compared with the GTG process in a historic New Zealand plants are (Maiden, 1988). The naphtha /gasoline selectivity of the GTG process in Table 1 is about 82.3%, while it is only about 36% or 19% for the high-temperature and low-temperature FTS processes, respectively.

Published studies have identified key characteristics of the GTL process such as catalyst type and synthesis (Nakamura, Wood, Hou, & Wise, 1981), reactor types (Schanke et al., 2001), energy and exergy analysis (Iandoli & Kjelstrup, 2007; Van Vliet, Faaij, & Turkenburg, 2009). Several modifications have been developed over the years to improve the FTS process, including those introduced for the Qatar Pearl Plant, the largest FTS plant in the world, by Qatar Petroleum and Shell commissioned in 2011 [15]. However, most of these improvements have focused on catalysts and reactor configurations in order to increase the liquid production, increase energy efficiency and decrease the CO₂ formation during this process.

Figure 1. Routes for coal and natural gas to liquid

Table 1. Comparison between MTG & FT units of New Zealand [11]

Figure 2. GTL and GTG process block diagram

The environmental benefits of using transport fuels produced by FTS based GTL technologies are well known [9]. These include low emissions of sulfur compounds and NO_x. In addition, the lower aromatic content of GTL fuels reduces their toxicity and the particulate matter generated when they are combusted, in comparison to typical oil refinery produced fuels. The higher Cetane number (70-80) of FTS diesel enables superior performance in a range of diesel engines (Bao, El-Halwagi, & Elbashir, 2010).The ability of FTS processes  to produce highly valued non-fuel products such as lubricants and waxes further adds to the value of their product slates. Behroozsarand and Zamaniyan (Behroozsarand & Zamaniyan, 2017a), have recently simulated an optimized integrated GTL process for converting of flare gas to valuable hydrocarbons such as gasoline and diesel. This study builds upon that work.

It is clear from Fig.2 that there are similarities between the early steps in the FTS and GTG processes; both require synthesis gas production. Moreover, the methanol production process from synthesis gas is a well-established commercialtechnologywithmultipletechnologysuppliers. Therefore, the key distinctive production technology stage in the GTG process is the methanol to gasoline (MTG) stage. Three companies are the main patent holders of the MTG process:

1)      ExxonMobil, USA

2)      Holder Topsoe, Denmark

3)      Primus Green Energy (syngas to gasoline STG+) for small-scale applications

In this study, an integrated FGTG process for converting flare gas to gasoline, LPG and fuel gas is simulated using the Aspen HYSYS v. 8.8 simulator. The steam methane reforming (SMR) unit, the syngas to methanol unit, and the MTG unit are configured for simulation as an integrated FGTG process. In order to reduce carbon dioxide gas emissions to the atmosphere, a novel closed arrangement for the FGTG process (recycling configuration) is described and simulated. Several researchers have been reported in that Aspen HYSYS is employed for simulation of different processes. Ghorbani and et al. (Hamedi, Shirmohammadi, Ghorbani, & Sheikhi, 2015) used Aspen HYSYS software for Advanced Exergy Evaluation of an Integrated Separation Process with Optimized Refrigeration System. Also, Ghorbani and et al. in other work (Ghorbani, Hamedi, & Amidpour, 2016) utilized this software for exergoeconomic evaluation of an integrated nitrogen rejection unit with LNG and NGL co-production processes based on the MFC and absorption refrigeration systems. Shariati Niasar and et al. (Shariati Niasar et al., 2017) have been used Aspen HYSYS and plus for simulation of ammonia water cycle in proposing of superstructure of cogeneration of power, heating, cooling and liquid fuels using gasification of feedstock with primary material of coal for employing in LNG process.

2. Process Description of GTG

Simulated plant wide GTG process in this study has three subsections (Fig. 3):

1) Syngas production unit

2) Methanol production unit

3) MTG unit (production and gasoline separation)

2.1. Synthesis Gas Production Unit

The initial stages in the GTG process are similar to FTS in producing synthesis gas (H2+CO). The ratio of hydrogen to carbon monoxide (H2:CO) in producing synthesis gas product is the key difference between the synthesis gas production units of the two processes. In GTG plants the second process is the production of methanol for which 3.8 is the suitable H2:CO ratio, whereas a 1.0-2.5 H2:CO ratio favors the FTS process. The synthesis gas product in the GTG process passes directly to a methanol production plant, whereas it passes into a FT unit in the FTS process. Two sequential reactors (conversion and equilibrium types) are considered [16] for simulating the steam reforming process to convert methane into synthesis gas using the Aspen HYSYS V.8.8 simulation environment (Fig. 4).

The first reactor is operated as a pre-reformer for reforming the heavier hydrocarbon components (i.e., natural gas liquids). The second reactor reforms the methane. All conversion and equilibrium reactions producing H2 and CO involved in these reactors are listed in Table 2 and 3, respectively (Behroozsarand & Zamaniyan, 2017a).

Figure 3. Block Flow Diagram (BFD) of simulated FGTG process

Table 2. List of conversion type reactions of primary steam reformer used in simulation model
(Behroozsarand & Zamaniyan, 2017a)

Table 3. List of equilibrium type reactions of secondary steam reformer used in simulation model for steam methane reformer (SMR) process unit for FGTG (Behroozsarand & Zamaniyan, 2017a)

* All equilibrium data come from Aspen HYSYS software reaction library.

Figure 4. Schematic of SMR Process as Configured in the Aspen HYSYS software environment

Figure 5. Schematic of MeOH process as configured in the Aspen HYSYS software environment

Figure 6. Schematic of MTG Process as Configured in the Aspen HYSYS Software Environment

2.2. Methanol Production Unit

Several companies have patented methanol processes, including “Linde Engineering”, “Johnson Mattey Process Technology”, “TOYO”, “Lurgi”, and others. In most commercial methanol production plants, the yield of produced methanol is more 98%. The main reactions involved in methanol reactors are exothermic for an equilibrium model involving three reactions (Eq. 1 to 3).

This process may be summarized as follows:

As the conversion process is limited by the equilibrium achieved for each pass of the reactants through the reactor, a simple loop arrangement recirculates the unreacted syngas back through the methanol converter. Fig.5 provides schematic diagram of the methanol conversion process in the Aspen HYSYS software environment.

2.3. Methanol to Gasoline (MTG) Unit

Main component process of a GTG plant is the MTG process unit. Mobil operated the first MTG plant in New Zealand from 1987 to 1995 [11, 19], producing a Sulphur-free gasoline of approximately 92-RON quality, based on a process developed in the 1970’s [20].  Mobil's novel synthetic gasoline process, based on the conversion of methanol to hydrocarbons over zeolite catalysts [19, 20], was a key departure  from the Fischer-Tropsch process (Behroozsarand & Zamaniyan, 2017a).

MTG reaction paths are summarized by Eq. 4 to 6:

In the patented commercial-scale MTG process, methanol is vaporized and fed into the fixed bed dimethyl ether (DME) reactor [19], where it is catalytically equilibrated to a mixture of DME, methanol and water. The catalyst used for the MTG process is of customized alumina at a reaction temperature and pressure of 310-320°C and about 26 bar, respectively. A schematic of the MTG process configured in the Aspen HYSYS software environment is illustrated in Fig. 6. This is applied to an integrated plant wide FGTG process for converting of flare gas to hydrocarbon fuel products including gasoline and LPG.

3. GTG Process Simulation and Results

The FGTG process is simulated using four model packages of the Aspen HYSYS 8.8 process simulator:

1) Syngas Basis (PRSV model, (Behroozsarand & Zamaniyan, 2017b))

2) DME Basis (UNIQUAC or NRTL models, (Kim, Kim, Cho, & Yoon, 2010))

3) Methanol Basis (PRSV, (Van-Dal & Bouallou, 2013))

4) MTG Basis (Peng-Robinson, (Kim et al., 2010))

Models 1, 2, 3, and 4 include of 104, 58, 9, and 58 components, respectively. Figs. 3, 4, 5, and 6 provide process flow diagrams of the simulated FGTG process in Aspen HYSYS v. 8.8 software.

Fig. 4, illustrates how the primary flare gas is separated into two flow streams:

a) Gas for steam methane reforming (SMR) feed

b) Gas for fuel gas to the furnace burners.

The ratio a: b is about 64:36. This ratio confirms that more than one-third of the flare input gas is consumed as fuel for the furnace burners in the SMR process. The H₂/CO ratio of 3.8 is the main specification of synthesis gas (syngas) product. However, there are three outlet streams from the SMR unit:

1) Reformed gas (syngas)

2) High pressure steam (HPS)

3) Separated sour water

3.1. Validation of the FGTG Process Simulation

Before simulation of FGTG process it is necessary to validate the overall simulation procedure and results. For this reason, a published MTG process case (Jones & Zhu) is simulated using the proposed Aspen HYSYS 8.8 process configuration, and the simulation results compared to the data published for a biomass to gasoline simulation case study employing the MTG process [21]. According to Table 4, the comparison of results show that our proposed simulation configuration involves reasonable deviations from the data published for the published case study.  Deviation error values in Table 4 for the gasoline and LPG streams are the most relevant as these are the main products of a MTG plant. The maximum deviations (our simulation configuration versus published case study [21]) for the LPG and gasoline streams are: 283.47 lb/hr for the propane product (representing 26% of the total error for the LPG flow stream) and 165.22 lb/hr for the cis2-Butene product (representing 15% of the total error for the gasoline flow stream), respectively.

Table 4. Comparison of simulated MTG unit and published MTG case study [17]

Although the deviations for these specific components are significant, the overall mass flow deviations for the two streams (for LPG, 7417 versus 7621 lb/hr; for gasoline, 30705 versus 30937 lb/hr; Table 4) are negligible. The deviation observed for the propane component in the mass flow of the LPG stream represents 3.7% of the total LPG flow stream. The deviation observed for the cis2-Butene component in the mass flow of the gasoline stream represents just 0.53% of the total gasoline flow stream.

Based on the results of this comparison, we assume that our proposed simulation configuration is an acceptable starting point for evaluating variations in feed parameters such as the process arrangement, feed gas composition, flow rate, and other variables.

3.2. Iran’s Asaluyeh Flare Gas Case Study Simulation Results for base Configuration

The Asaluyeh gas processing plant, forming part of the Pars Special Energy Economic Zone, is a large-scale gas processing plant consisting of multiple trains located on the shore of the Persian Gulf some 270 km southeast of the provincial capital of Bushehr. The Asaluyeh facility has significant flare gas available for processing and could provide feed gas for a FGTG plant. For the purposes of this study the input feed (flare) wet gas from the Asaluyeh facility to the steam reformer stage of the FGTG plant is assumed to have a mass flow of 100,000 lb/hr (Table 5) with low sulphurous and carbon dioxide components.

Table 5. Specification of feed and products of syngas production, methanol, and MTG units

As shown in Fig. 3, six streams distinct streams require consideration for the overall simulation of a GTG plant:

1) Feed gas stream (flare gas to synthesis gas)

2) Reformed gas from the SMR process as the syngas unit product

3) Methanol as methanol unit product

4) LPG as MTG unit product

5) Gasoline as MTG unit product

6) Fuel gas as MTG unit product

The main product of the syngas unit is reformed gas and with mass flow of 206,298 lb. /hr. (i.e. 63.82% of the feed gas mass flow). The produced syngas from the SMR unit has significant quantities of CO₂ (about 4 mole %) and water (about 21 mole %). Table 5 shows mass balance for the feed gas and products for the SMR syngas unit of the FGTG process. Table 5 also indicates that the H₂/CO in reformed syngas is about 3.8, which is suitable for use in the methanol plant, which, based on equations 1 to 3, can produce methanol with purity of 97%. The conversion of the carbon monoxide component of the syngas in the methanol unit is given by Eq. (9):

As the methanol mole fraction in the outlet products of the MTG unit in the simulation is zero, it can be concluded that the entire input methanol to the MTG unit is converted into final products (gasoline, LPG, and fuel gas in the proportion 74%; 17%; 8%; Table 5). For detailed evaluation of methanol conversion to component-based products, mole fraction of mixed- product output (Gasoline+LPG+Fuel gas) is provided in Table 5. Results indicate that propane and i-pentane have maximum mole fractions among all the components of mixed-product output. The mass base conversion of methanol to these components is about 7.21% as indicated by applying Eq. (10) to the data from Table 5.

3.3. Novel Recycling Configuration of FGTG Plant with Recycling of all Emission Gases

In all stages of the FGTG process, two streams are the main candidates for recycling and reuse:

1- H₂ rich off gas from methanol unit

2- CO₂ and H₂ rich off gas from MTG unit

Specification of these streams before and after the membrane section are listed in Table 6.

The off-gas specifications reveal that, CH₄, H₂ and CO₂ have highest molar composition in these two streams. On the other hand, the main objective of the FGTG process is to maximize gasoline productivity. So, producing fuel gas for recycling to the syngas unit and synthesis gas for recycling to the methanol unit are both potentially attractive options for maximizing methanol productivity and thereby obtaining maximum gasoline productivity. Synthesis gas is a mixture of H₂ and CO in varying proportions. In order to efficiently produce methanol, the syngas H₂:CO ratio needs to be close to 3.8 when it enters the methanol unit. Unfortunately, the CO content of off gas streams is very low. Therefore, the water-gas-shift (WGS) reaction (Eq. (11)) is required to promote the conversion of CO₂ to CO.

Block flow diagrams (BFD) of the membrane section and a new configuration of the FGTG process are shown in Figs. 7 and 8.

Firstly, a mixed-off-gas stream from the methanol and MTG unit enters the membrane subunit, where H₂ purification is performed through a palladium membrane package. Pure hydrogen is then fed from this unit to a WGS reactor, where it reacts with makeup CO₂ to produce synthesis gas with the required hydrogen to carbon monoxide ratio of 3.83. Secondly, the methane-rich fuel gas stream derived (Table 6) is recycled to the synthesis gas production subunit.

Various dense metal membranes, such Cu, Al, Pd, Ni, Mo, Pt, Au, Nb, Fe, and Ta, have potential to purify hydrogen in a dissociated form, and consequently they can theoretically provide unbounded selectivity. Among these dense metal membranes, palladium (Pd) membranes have recently received considerable attention for the following reasons: excellent permeability, high tolerance to hydrocarbon flows and self-catalyzing behavior of the H₂ dissociation reactions (Rahimpour, Samimi, Babapoor, Tohidian, & Mohebi, 2017). Another important feature of a Pd membrane is its excellent resistance to hydrogen embrittlement and catalytic ability for hydrogen recombination (Gade, Thoen, & Way, 2008).

The material balance and membrane specifications for hydrogen purification in the membrane subunit is provided in Table 7.

Table 6. Specification of off gas before and after recycling from membrane unit

Figure 7. Block Flow Diagram (BFD) of membrane separation process

Figure 8. Block Flow Diagram (BFD) of simulated new FGTG process

Table 7. Results of material balance and membrane specification for hydrogen purification in membrane subunit

3.4. Results of Sensitivity Analysis for Iran’s Asaluyeh Flare Gas Case Study in Base and Recycling Configurations

A key objective for the FGTG process is selectivity of the gasoline product. Three main products are derived from a FGTG plant (Table 5):

1- Gasoline

2- LPG

3- Fuel gas

Among them, gasoline is the product with the highest value and demand, so maximization of gasoline productivity and selectivity need to be the primary focus of FGTG process suppliers. For the purpose of monitoring selectivity and productivity of the gasoline product, flare gas mass flow as feed to the FGTG  plant is selected as the adjustable variable. One reason for this selection is that flare gas mass flow changes dynamically in Iran’s Asaluyeh facility. Consequently, the capacity of an FGTG plant needs to vary between 70,000 to 130,000 lb./hr. according to maximum and minimum mass flow rates observed over an annual period. Table 8 presents the results of productivity and selectivity of all products of the FGTG plant for a range feed gas flow rates.

Table 8. Results of productivity and selectivity of all Products

* Total productivity = Gasoline productivity + LPG productivity

** Productivity enhancement =100× (Productivity in Recycling FGTG Plant Configuration - Productivity in Base FGTG Plant Configuration) / Productivity in Base FGTG Plant Configuration

*** Gasoline selectivity enhancement =100× (Selectivity in Recycling FGTG Plant Configuration - Selectivity in Base FGTG Plant Configuration) / Selectivity in Base FGTG Plant Configuration

**** Energy Index enhancement =100× (Energy Index in Recycling FGTG Plant Configuration - Energy Index in Base FGTG Plant Configuration)/ Energy Index in Base FGTG Plant Configuration

The gasoline selectivity remains approximately constant (75%) in all capacities of flare gas fed through the FGTG process in the base configuration (Table 8). On the other hand, productivity does not remain constant, but increases, as the feed gas mass flow rate is increased. Table 8 compares total productivity for the base configuration and the novel recycling configuration for the FGTG plant. It reveals that total productivity and gasoline selectivity are significantly increased when the recycling configuration is used instead of the base configuration. The results (Table 8) show average 53%  and 10% increases, respectively, for total productivity and gasoline selectivity, as the main outcomes of the process optimization achieved by the recycling configuration.

In the bottom section, calculations have been made on the economic benefits of increasing the production of products. The base price of the products is based on Platts price referance. As the calculations show, the increase in product production will result in a 53 percent increase in profits due to the reform of the process arrangement from base FGTG to recycling FGTG.

Total productivity = Gasoline productivity + LPG productivity

Base FGTG: 43453 (lb./hr.) = 35429 (lb./hr.) + 8024 (lb./hr.)

Recycling FGTG: 66399 (lb./hr.) = 54474 (lb./hr.) + 11925 (lb./hr.)

LPG Density = 5.185 (lb./ft³)

Gasoline Density = 43.96 (lb./ft³)

Base cost of Gasoline = 93 $/bbl = 0.585 $/L = 585 $/m³ = 831 $/Ton = 0.377 $/lb.

Base cost of LPG = 576$/Ton = 0.262 $/lb.

Economiy calculation:

Base FGTG: Sale Income = 35429 (lb./hr.) × 0.377 $/lb.+ 8024 (lb./hr.) × 0.262 $/lb. = 15459 $/hr.

Recycling FGTG: Sale Income = 54474 (lb./hr.) × 0.377 $/lb.+ 11925 (lb./hr.) × 0.262 $/lb. = 23661 $/hr.

Sale Income Enhancement (%) = (23661 $/hr. - 15459 $/hr.) / 15459 $/hr. = + 53%

Other crucial parameters for evaluating the performance of process units, and in particular the FGTG process, is energy and utilities consumption. Table 9 lists the total  utilities and energy consumption for a range of feed gas flow rate cases for the base configuration and the novel recycling configuration for the FGTG plant. The results indicate that increasing the flare gas mass flow as feed streams has, as to be expected, a direct effect on increasing the energy consumption of the FGTG unit. Minimum and maximum utilities consumption for the base FGTG configuration vary following a consistent trend withthe minimum and maximum of flare gas flow rate. This consistent trend is not seen in  the proposed recycling configuration for the FGTG plant (Table 9).

A suitable index for comparing of the various FGTG capacities on the basis of energy consumption for the two configurations (i.e., base and recycling) is the ratio total utilities consumption to total productivity (i.e., termed here the Energy Index, Table 9, expressed in Btu/lb). The highest feed-gas capacity case (130,000 lb/hr.) demonstrates the lowest Energy Index values than the lower- capacity natural gas mass flow cases for both FGTG plant configurations considered (4.21 for old arrangement and 1.83 for new arrangement model).

Significantly, the recycled FGTG configuration yields a significantly more favorable Energy Index than the base configuration (Fig.9). For example, for a feed-gas capacity of 100,000 lb./hr. (the expected average operating conditions) the flare gas capacity Energy Index is 3.46 for the recycled FGTG configuration versus 4.31 for the base configuration. Perhaps, the most important advantage of the recycled FGTG plant configuration, in addition to producing more of the key products (gasoline and LPG) and consuming less energy per unit of production, is its lack of emissions of greenhouse gases (except from the fuel gas consumed in the process burners) into the atmosphere.

Fig. 10 Summarizes the Plant Fuel Losses and Productivity from the Simulation Results for: a) base FGTG Plant Configuration and b) Recycling FGTG Plant Configuration.

Table 9. Results of utility and energy consumption in all scenarios

*Energy Index = Total Productivity/ Total Utilities/10⁴

Figur 9. Comparing of FGTG various cases versus defined energy index

Figure 10. Comparison of fuel losses and productivity of base FGTG and recycling FGTG plants

a) FGTG base; b) FGTG recycling.