Influence factor research on deacidification process for high carbon content gas field by numerical simulation—a case study of the Oudeh gas field

Introduction High concentrations of CO2 in natural gas affect its calorific value and corrode the equipment and pipelines related to its transportation and usage. Therefore, strict control over the H2S and CO2 contents in natural gas is essential. CO2 is an important industrial gas that can bring a great deal of economic profit when it is fully utilized. Case description The natural gas produced at the Oudeh gas field in Syria contains high carbon content natural gas, in which the CO2 content is in the range of 17.5–18.8 %, while the H2S content is in the range of 2.8–3.2 %. However, there have been few studies conducted on treatment solutions for natural gas with high carbon contents. In this paper, several commonly used methods for deacidification of natural gas were introduced. Among these methods, the most suitable one was chosen for desulfurization and decarbonization of the natural gas produced at the Oudeh gas field based on its gas quality. Conclusions Optimization and analysis of the primary operating parameters for the desulfurization and decarbonization processes were conducted to obtain the optimized values for the input temperature of the lean solution (42 °C), reflux ratio (0.8), number of trays in the absorber unit (17) , and circulation rate of the lean solution (330 m3/h), etc. Additionally, the influence of the operating pressure of the regenerator unit on the regeneration system was also investigated. The energy consumption of the apparatus and the corrosion level of sour gas to the apparatus were reduced after optimization. Based on the investigation of the natural gas treatment for this gas field, it can serve as a reference for the purification of high carbon contents natural gas.

and RSH (organic sulfur). H 2 S is highly toxic and burns into SO 2 , causing air pollution and cracking of pipes after severe corrosion. High levels of CO 2 directly affect the calorific value of natural gas and corrode the system and pipelines installed for its transportation and usage. Therefore, the levels of H 2 S and CO 2 in natural gas have to be strictly controlled. The quality of purified natural gas must meet the second standard of China's quality specification for natural gas, in which the content of H 2 S must be less than 7 mg/ m 3 and the content of CO 2 must be less than 3 %.
In 1880s, the American company Union Carbide is the pioneer company formulated the MDEA solution, Ucarsol HS, and further developed it into a series of products. Thereafter, formulated CR solutions, primarily used for CO 2 removal, were also developed (Yu 1985). The Dow chemical company based in the United States also developed similar formulated solutions such as the Gas/Spec SS and CS series solutions. In the 1950s, German companies Lurgi GmbH and Linde AG jointly developed low-temperature methanol absorption technology. Owing to its excellent performance, it was widely used in urban gas as well as the oil and gas industries. Currently there are over 80 installations operating low-temperature methanol absorption technologies worldwide (Niu et al. 2003;Hoochgesand 1970). In 1965, American company Allied developed the Selexol solution (polyethylene glycol dimethyl ether) as a physical absorption method that was used for the removal of CO 2 from natural gas starting in the early 1980s. Recently, the Selexol solution has been widely employed for the removal of H 2 S, CO 2 , COS, sulfides, mercaptans, and other hazardous components from natural gas, fuel gas, and syngas (Qin et al. 2008;Cheng et al. 2002). The ARI Company successfully developed the LO-CAT process used for catalytic liquid-phase oxidation-reduction desulfurization using iron-based catalysts. It is suitable for the purification of natural gas with 0.2 ~ 10 T/d sulfur content. The difference between the LO-CAT and oxyamine desulfurization processes is that the former is an oxidation-reduction process for the removal of H 2 S but does not remove CO 2 . Since the successful development of the LO-CAT desulfurization process in the 1980s, as many as 232 installations have been set up worldwide including those that have been completed and those under construction as of 1998 (Xu et al. 2004). In the early 1980s, the Monsanto Company developed membrane separation technology for the removal of CO 2 from natural gas. However, the application of this technology was hindered by the high production cost of the membrane (Wang et al. 1999;Long and Long 1993).
In China, the oxyamine process is the most commonly used method for the desulfurization and decarbonization of natural gas. In the 1980s, Nanjing Chemical Industry Company successfully developed an absorption solution called NHD that resembled the physical and chemical properties of Selexol, which was later developed into a NHD purification process. The NHD process is currently widely used for desulfurization and decarbonization purposes in small fertilizer plants, ammonia plants, and methanol plants (Hoochgesand 1970). Since 1993, the research institute of PetroChina Southwest Oil and Gasfield has been cooperating with Zhejiang Zhenhai Petrochemical Engineering General Plant in an effort to substitute MEA for a MDEA method in order to reduce the energy consumption of vapor by 40 %. Until now, pure MDEA or formulated solvents with a mixture of MDEA and other physical solvents and chemical additives have been used in most purification installations for natural gas and gas refineries in China (Chen 2003;He et al. 2009).

Mathematical model
HYSYS is a simulation software developed by Hyprotech Company based in Canada that plays a dominant role in simulation techniques in the petrochemical industry around the world. It is also used in wide range of fields, including petroleum extraction, storage and transportation, petrochemical industries, natural gas processing, fine chemical industries, refining industries and pharmaceutical industries.

Vapor-liquid equilibrium
The calculation of liquid phase equilibrium is essential to every operating unit. Therefore, any variation in temperature, pressure, or composition may cause changes in the initial equilibrium system until it reaches a new equilibrium state. The (ф-γ) method is used for the calculation of vapor-liquid equilibrium.
where K is the phase equilibrium constant; γ i is the fugacity coefficients of i component in liquid phase; f o i is the fugacity of i component in liquid phase under standard conditions; φ V i is the fugacity coefficients of i component in vapor phase.

Balance equation of acid gas in the amine solution
The balance equation in a H 2 S-CO 2 -Am-aqueous system involves seven chemical equations, two Henry's Law, three mass-balance equations, and one electroneutrality equation.

Reaction equations
where K 1 is the equilibrium constant, i is the reaction number, [CO 3 2− ] and etc. are the concentrations of ions or compounds in the solution.

Henry's law
where P s , P c is the partial pressure of H 2 S and CO 2 , respectively, kPa; H s , H s are the Henry's constant of H 2 S and CO 2 , respectively, kPa L/mol.

Mass-balance equation
where m is the concentration of amine, mol/L; α c and α s are the equilibrium solubility of CO 2 and H 2 S in amine, respectively, mol/mol.

Project profile
The natural gas investigated in this paper was high carbon content natural gas extracted from the Oudeh gas field in Syria (Belongs to Oudeh oilfield, the owner of the land gave permission to conduct the study on this site. (1) I state that no specific permissions were required for these locations/activities. (2) I confirm that the field studies did not involve endangered or protected species). The composition of its feed gas was illustrated in Table 1.
The processing capacity of the installment, flow rate, pressure (absolute pressure), and temperature of the feed gas were set as 250 × 10 4 Nm 3 /d, 230 × 104, 7.5 MPa, and 25 °C, respectively.
The quality of the natural gas has to be taken into consideration when choosing the most suitable method for desulfurization and decarbonization processes. For example, the oxidation-reduction method is mainly used for the removal of H 2 S and is unsuitable for systems with CO 2 content. The CO 2 content of the purified gas does not reach the specification of being less than 3 % after undergoing the membrane separation process. In addition, it is highly unstable, and generates high production costs (Ren and An 2005). The physiochemical absorption method has only found success in small installations but it is still not stable enough to be widely used (Gupta 1986). The simulation result shows that 39.1 % of methane will be lost using the Selexol method, showing a lack of economic benefit. Compared to those methods, the chemical absorption method seems to be the most suitable desulfurization and decarbonization process to purify natural gas with high CO 2 content.
During the deacidification process, monoethanolamine (MEA) in the solution degrades CO 2 to produce some non-renewable substances, which results in partial loss of its desulfurization capability. Hence, this method is not being selected. Methyldiethanolamine (MDEA) is highly selective towards H 2 S, is less corrosive, requires low energy consumption for regeneration, and has wide applications. Although this method is applicable for H 2 S removal, this study primarily dealt with the removal of CO 2 , and hence, this method was also unsuitable (Xue 1995). In the F1exsorb PS process, the activated MDEA and sterically-hindered amines are suitable for use with natural gas with high carbon content. However, its high cost has hindered its wide applications. Compared to MEA, diethanolamine (DEA) has relatively high solubility, lower degradability, is less corrosive, and is not selective towards either H 2 S or CO 2 . It is suitable to be used in natural gas, manufactured gas, and feed gas with high contents of organic sulfides. Thus, it was selected for this system.
It is known that feed gas contains high CO 2 content. Although the MDEA solution possesses a number of advantages, it has a slower reaction rate in the removal of CO 2 . To overcome this problem, a certain amount of DEA (MEA) was mixed with the MDEA solution to improve the absorption rate of CO 2 . As a result, the MDEA-DEA mixedamines solution exhibited a faster absorption rate, higher CO 2 removal efficiency, and lower circulation rate of lean solution (Li and Zhao 2009;Wang and Wang 2003). After comparing the simulations of solutions with different ratios and the DEA method, a mixed-amines solution with a ratio of 43 %(w) MDEA + 12 %(w) DEA was chosen as the absorber for the desulfurization and decarbonization processes after purification efficiency, economic benefits, and recycling issues were taken into account.

Numerical simulation method validation
Based on the field test data provided by the literature (Ma et al. 2012), a set of decarbonization device were put into production on 2009.12 at Changling gas field, with a operation capacity of 120 × 10 4 m 3 /d. The mixed-amines solution is 45 %(w) MDEA + 5 %(w) Activator + 50 %(w) Water, the CO2 content in processed natural gas is 24 % on average in the test.
A numerical simulation is conducted on the base of this test, as presented in Table 2.

Numerical simulation
In numerical simulation, the process was designed as follows.
The feed gas (F1) was firstly introduced into the horizontal separation to remove impurities and water droplets (F2) that had been carried along with the feed gas (F1). The feed gas (F1) entered the bottom of the absorber unit after the filtration and separation processes. The up-flowing feed gas eventually had countercurrent contact with the down-flowing 55 % (w) mixed-amines lean solution that entered from the top of the tower. Most of the H 2 S and CO 2 were absorbed by the lean solution here. The wet purified gas (F2) left the tower and subsequently entered the vertical separator for removal of the remaining alcohol amine droplets (F4) before leaving the installation. The wet purified gas (F5) was then transferred into the molecular sieve and dehydration unit. The H 2 S and CO 2 contents of the wet purified gas (F5) were the contents after being purified.
The DEA-rich solution (F6) that exited from the bottom of the absorber unit was introduced into the flash drum. Hydrocarbon gasses that were dissolved in the solution were released after undergoing the flash evaporation. In the scrubbing section, the upflowing flash gas (F7) had countercurrent contact with the down-flowing mixed-amines lean solution to remove a portion of the CO 2 and H 2 S content from the flash gas. After its pressure was regulated, the flash gas (F7) eventually entered the fuel gas system of the purification plant.
The mixed-amines-rich solution (F8) coming from the bottom of the flash drum passed through the filtration system to purge the degradation products and mechanical impurities that had been dissolved in the rich solution. After that, the mixed-aminesrich solution (F9) entered the heat exchanger and was warmed by the lean solution (F10) coming out from the bottom of the regenerator unit. Then, the rich solution (F11) re-entered the regenerator unit from the top and had countercurrent contact with the up-flowing vapor, leading to further desorption of the H 2 S and CO 2 gases to achieve regeneration of the solution. A reboiler at the bottom of the tower provided the heat for regeneration. After passing through the heat exchanger, the warm mixed-amines lean solution (F12) was removed from the bottom of the regenerator unit and cooled by the rich solution. It (F12) then entered the air cooler and cooler to be further chilled. Subsequently, the lean solution (F13) was transferred back to the MDEA-DEA absorber unit by means of a circulating pump to complete the cycle. The acidic components and water vapor (F14) leaving from the top of the regenerator unit entered the condenser, and a substantial amount of water vapor was condensed. The acidic condensate (F15) being separated out was directed back to the top of the regenerator unit to recycle the alcohol amines vapor that was carried along by the acidic airflow. The acid gas (F16) was transferred to the sulfur recovery unit or incinerator.
Parameters of the simulated process are shown in Tables 3, 4. The flow diagram of the process is shown in Fig. 1.

Results
The mass flow parameters of the process using 43 %(w) MDEA + 12 %(w) DEA calculated by the simulation software are shown in Table 5.
As can be seen in Table 5, the CO 2 and H 2 S contents of the purified gas after undergoing the deacidification process using mixed-amines solution were 2.44 and 0.0004 %, respectively, meeting the quality specifications for purified gas. The heat load of the apparatus used during the process is illustrated in Table 6.

Discussions
During the simulated calculations for the desulfurization and decarbonisation process of natural gas, most of the parameters were derived from empirical or assumed values. Therefore, this paper investigated the input temperature of the lean solution, the reflux ratio, the number of trays in the absorber unit, the circulation rate of the lean solution, and the operating pressure of the regenerator unit and their influences on the efficiency of the purification process.

Input temperature of the lean solution
The purpose of the simulation calculation was to obtain an optimum input temperature for the lean solution while the other operating conditions remained unchanged. The simulated results using the input temperature of the lean solution as a variable are shown in Fig. 2.   The results shown in Fig. 2 revealed that an increase in the input temperature of the lean solution caused a slight increase in the H 2 S content and a continuous decrease in the CO 2 content in the wet purified gas, as well as a decrease in the flow rate of the wet purified gas leaving the tower. Based on this trend, a decrease in the input temperature of the lean solution reduced the CO 2 content significantly while the H 2 S content remained almost constant. This result demonstrated that the input temperature of the lean solution primarily affected the absorption rate of CO 2 , while the MDEA-DEA mixed-amines solution did not significantly affect the absorption rate of H 2 S when the operating temperature was lower than 47 °C.

Reflux ratio
The molar ratio between the water dissolved in the acid gas and the desorbed acid gas leaving the regenerator unit is known as the reflux ratio (mol water/mol sour gas). In the simulation calculations, the reflux ratio of the condenser was a variable and the input temperature of the lean solution was set at 42 °C, while other operating conditions remained unchanged. The simulated results are shown in Fig. 3.   Fig. 2 The relationship between the input temperature of the lean solution and the efficiency of the purification process Fig. 3 The relationship between the reflux ratio of the condenser and the efficiency of the purification process Figure 3 shows that the content of the acid gas in the wet purified gas began to level off after the reflux ratio of the condenser reached 0.8 (acid gas load of the lean solution was 0.0029). This revelation illustrated that the influence of the reboiler load on the acid gas load in the rich solution became smaller when the desorption of the acid gas in the rich solution reached a certain level, indicating the effect of the absorption of acid gas after this point became less significant even though the reflux ratio continue to increase.

Number of trays in the absorber unit
Before conducting the simulation calculations, the input temperature of the lean solution and reflux ratio of the condenser were set at 42 °C and 0.88, respectively. The number of trays in the absorber unit was set as the variable to investigate its influences on the quality of the purified gas. The relationship between the number of trays in the absorber unit and the H 2 S and CO 2 contents in the wet purified gas is shown in Fig. 4. Figure 4 shows that the content of CO 2 and H 2 S decreased with the increasing number of trays in the absorber unit. If most of the CO 2 gas and trace amounts of H 2 S gas were absorbed by the MDEA-DEA mixed-amines solution when the number of trays was increased to 17 trays. The selectivity of the amine solution decreased with the increasing number of trays and caused a substantial amount of CO 2 to be absorbed. As a result, the flow rate of the wet purified gas dropped accordingly, causing a drop in the revenue of the apparatus and the acid gas quality in the regenerator unit. Therefore, the optimum number of trays in the absorber unit was 17 trays theoretically.

Circulation rate of the lean solution
Based on previous optimized parameters, the number of trays in the absorber unit, the reflux ratio of the condenser at the top of the regenerator unit, and the input temperature of the lean solution were set at 17 trays, 0.88, and 42 °C, respectively, while the other operating parameters remained unchanged. The step size of the circulation rate was calculated as 10 m 3 /h using the circulation rate of the lean or semi-lean solution as a variable in the simulation calculations. The simulation software was also used to Fig. 4 The relationship between the number of trays in the absorber unit and the efficiency of the purification process estimate the influence of the circulation rate of the lean solution on the desulfurization and decarbonisation processes. The influence of the circulation rate of the lean solution on the desulfurization and decarbonisation processes is shown in Fig. 5.
The contents of CO 2 and H 2 S in the wet purified gas decreased with the increasing circulation rate of the lean solution. However, the CO 2 absorption rate eventually dropped when the circulation rate of the lean solution was increased beyond 320 m 3 /h. This was because the reaction rate between CO 2 and MDEA decreased. To remove a large amount of CO 2 , the retention time of the lean solution on the tray had to be maximized. Thus, increasing the circulation rate inevitably shortened the retention time of the lean solution on the tray, which slowed down the CO 2 absorption rate. In addition, the reboiler load increased with the increasing circulation rate. Based on the simulation results, the circulation rate of the lean solution was estimated to be 330 m 3 /h when the design margin of the apparatus was taken into account.

Operating pressure of the regenerator unit
Based on previous optimized parameters, the number of trays in the regenerator unit, the number of trays in the absorber unit, the circulation rate of the lean solution, and the reflux ratio were set at 25 trays, 17 trays, 330 m 3 /h, and 0.88, respectively. The simulation results using the operating pressure of the regenerator unit as a variable are shown in Fig. 6.
As shown in Fig. 6, the CO 2 content increased and the H 2 S content did not change much when the operating pressure of the regenerator unit increased. As a result, the quality of the acid gas degraded. Due to the operating pressure of the regenerator unit increasing almost proportionally with the reboiler load (Wen et al. 2007), the operating pressure should not be too high. Combining the simulation and experimental results, it was more appropriate to set the operating pressure of the regenerator unit at 200 kPa.

Conclusions
Based on previous analyses of the optimization of inlet temperature of the lean solution, reflux ratio of the condenser at the top of the regenerator unit, number of trays Fig. 5 The relationship between the circulation rate of the lean solution and the efficiency of the purification process of the absorber unit, circulation rate of the lean solution, and operating pressure of the regenerator unit, a comparison was made on the data before and after the optimization of the apparatus used for the desulfurization and decarbonisation process employing the MDEA-DEA mixed-amines absorption method, as shown in Table 7.
The flow rates of the wet purified gas remained almost the same before and after optimization. The revenue of the sales gas changed slightly before and after optimization. However, energy consumption of the apparatus, vapor flow rate, flow rate of the recycling cooled water, and the flow rate of the deoxygenated water decreased to different extents. The total costs of energy consumption were reduced. The reboiler load was reduced to a certain extent. Additionally, the acid gas load of the rich solution also dropped slightly, decreasing the probability of causing excessive corrosion of the apparatus and pipe materials. Based on the simulation of the deacidification of the natural gas produced by the Oudeh gas field, the conclusions of the optimization were as follows: 1. When the inlet temperature of the lean solution was less than 47 °C, the CO 2 absorption rate was mainly governed by the temperature of the lean solution, while the MDEA-DEA mixed-amines solution had less effect on the H 2 S absorption rate. Fig. 6 The relationship between the operating pressure of the regenerator unit and H 2 S and CO 2 contents in the sour gas