The effect of increasing CO 2 concentration and flow rate on amine still performance in meeting gas sale specifications

. The main purpose of this study is to examine the effect of increasing CO 2 removal and flow rate on performance of an amine still. The amine still is located in Field X in South East Sumatra at a new gas well producing gases with a rich CO 2 content. The still uses activated MDEA as the amine and has an IMTP 40-type packing column. Two film and desorption equilibrium curve theories were employed to analyse the amine still design conditions. Design equations were utilized to find the slope of the equilibrium curve. A slope of the equilibrium curve of 45° in the amine still is obtained in this study. The maximum liquid CO 2 composition of the amine still feedstock (x o ) which can be separated to produce lean amine according to the specification design flow rate is 0.0307. The total flow rate of CO 2 -rich amine at x o = 0.029 is 761,157.6 kg/hour; the total flow rate of CO 2 -rich amine at x o = 0.0295 is 628,861.1 kg/hour; the total flow rate of CO 2 rich amine at x o = 0.03 is 513,962.6 kg/hour; and the total flow rate of CO 2 -rich amine at x o = 0.0305 is 409,575.3 kg/hour.


Introduction
A significant drop in the flow rate of gas wells has led to the creation of new gas sources in the study area. The recently created gas well in Field X in South East Sumatra has a rich CO2 composition. The sudden surge of CO2 from the well will have an immediate effect on the amine unit used for CO2 separation. Design barriers result in this increased CO2 composition directly leading to a decrease in gas production.
The amine unit referred to in this study is a processing unit used to remove CO2 from gases. This separation is required because of the absence of heating value of CO2 and its corrosive effect when it reacts with water. The CO2 composition of gas for commercial sale must therefore be below specified limits to conform to the specifications of the gas buyer. Increased CO2 from gas wells leads to increased CO2 concentrations in rich amine in the still.
Increasing the CO2 content of rich amine will require optimized amine still performance, so that lean amine returns to the amine contactor to further separate CO2 within the still as needed. Desorption performance in the optimized amine still requires low pressure and high temperature. Low pressure creates the low partial pressure required in the gas phase and high temperature creates the high vapour pressure required in the liquid phase. The pressure deviation between the partial gas phase and the larger liquid phase are required for better mass transfer between phases.
This research is valuable because of the importance of identifying and understanding the conditions for amine unit design, specifically amine stills used for the desorption process. Height transfer unit (HTU) and number transfer unit (NTU) design equations are used to determine the amine still design conditions. The amine still is the location in which the desorption process between MDEA and CO2 occurs. The amine still design used in Field X uses packing with IMTP type 40. The amine still design data and design equations enable us to obtain the equilibrium slope curves (myx) which are used to analyse the limits of CO2 increase and flow rate that can be processed.

HTU and NTU equations designs for the amine still
The HTU and NTU equations are required to calculate the slope of the equilibrium curve (myx). The data required for this design equation are drawn from material design data, dimensions and the amine still process. High packing (H) is the design data used to determine the myx value.
In order to achieve a high packing value (H), the myx guess value must be entered at the beginning of the process. The value of myx is compared to the convergence criteria. In the condition where the value of myx is smaller or equal to the convergence criterion, the myx value is assumed to be achieved. If the myx value is greater than the convergence value, re-guessing of the the myxvalue is required.
Regardless of its current phase (gas or liquid), the HTU is a function of the mass transfer coefficient of both the liquid phase (kL) and the gas phase (kG). The NTU is a function of liquid load transfer and gas load transfer in both the bottom and overhead sections of the still towers.
The slope of the equilibrium curve will be used to determine the amine still design conditions with variations of concentrations and flow rates.

Analysis of the impact of CO2 increase on amine still design capability
The analysis of the impact of CO2 increase on amine still design capability is executed by varying the value of xo (CO2 composition in the liquid phase of the amine feed) at a fixed flow rate. Subsequently, the value of xu (CO2 composition in liquid phase of the amine still) is solved by employing trial and error. As a result, the high value of the amine still remains in accordance with the design. The numerical analysis is performed using the slope value at equilibrium (mxy) which was obtained from this calculation.
Variations are performed with the following assumptions 1. An increase in yo with the assumption that all the increase in CO2 in the amine still (xo) will flow to the top of the amine still. 2. xu is 0.01 mol/mol MDEA (assuming CO2 loading of lean amine) is 0.0013 mole fraction. This value is the maximum value of CO2 allowed in lean amine. 3. The increase in yu from the design value is proportional to the increase in the value of xu.

Analysing the impact of increase in flow rate on amine still design capability
This step of the analysis focuses on the impact of the amine still design conditions on the variation of the flow rate for several CO2 compositions (xo and yo). The equilibrium curve slope value (mxy) was obtained from the previous calculation. The xo value uses numbers inbetween the design value and the maximum value obtained from the previous xo variation. The value of xu is solved by employing trial and error to ensure that high amine value in the still is in accordance with the design specification.
Variations are performed with the following assumptions: 1. The liquid phase flow rate (L) and the gas phase flow rate (V) rise by the same ratio. 2. xu is 0.01 mol/mol MDEA and is 0.0013 mole fraction. This value is the maximum value of CO2 allowed in lean amine. 3. The increase in yu is equivalent to the increase in xu value. 4. yo increases with the assumption that all the addition of CO2 in the amine still feed (xo) will flow to the top of the amine still.

Coefficient of gas CO2 diffusion and coefficient of liquid CO2 diffusion
The determination of the gas CO2 diffusion coefficient can be obtained using equation 1.
The data obtained from the mechanical data sheet of the amine still for the above calculation is as follows: Dimensionless quantity (Ω) is recognized as the function of k B T/ε 12 . T is the operation temperature obtained from the mechanical data sheet of the amine still. Thus, k B T/ε 12 = 362.594/397.4 = 0.912 By looking at the Ω value for k B T/ε 12 = 0,912 , Ω = 1.52 The CO2 diffusion coefficient obtained from equation 1 is as follows: The diffusion value of the gas coefficient ranges from 10 -5 -10 -6 m 2 /s (10 -1 -10 -2 cm 2 /s). This can be seen in the example of the diffusion coefficient value of the gas in other experimental results. The diffusion coefficient of the gas can be increased by raising the temperature or lowering the pressure. The underlying cause of this phenomenon is the faster movement of molecules at high temperatures or at lower pressure.
The determination of the liquid CO2 diffusion can be acquired from equation 4.
The viscosity values of each component at the temperature and pressure of column operations obtained from Simulation are as follows: From equation 6, the CO2 diffusion coefficient value is as follows: .792 x 10 -9 x 0.375 = 2.5457 x 10 -9 m 2 /s. From the above calculation of the gas diffusion coefficient it is found that the obstacles faced by the liquid diffusion coefficient have resulted in its lower value in comparison to the gas diffusion coefficient.
The value of the diffusion coefficient in the liquid can be increased by raising the temperature, as a result of the faster movement of liquid molecules at higher temperatures than at low temperatures. This can be seen by referring to the sample diffusion coefficient value of the other experimental results. The diffusion value of the CO2 coefficient in water at temperature 293 °K (20 °C) is 2.5 x 10 -5 cm 2 /s (2.5 x 10 -9 m 2 /s). In the amine still with a temperature of 89.4 °C, the diffusion value of the CO2 coefficient in water is 6.8 x 10 -9 m 2 /s, which is higher than the diffusion coefficient at 20 °C.

Effective wetted packing area (aw)
Equation 7 can be used to determine the effective wetted packing area (aw).
The data obtained from the mechanical data sheet for the amine still is as follows: In comparison to the surface area per packing volume, the effective wetted packing area value is 60.76%. This number signifies the effectiveness of fluid system contact in the packing. As shown in the calculation in equation 7, the value of the wetted area is inversely proportional to the surface tension of the packing material (σc). Hence, the smaller the value of σc, the greater the wetted area and the more effective the contacts in the packing. The effective wetted packing area can be increased by changing the packing material to one that has a lower surface tension. Polyethylene plastic has the lowest surface tension value compared to other materials according to the table of surface tension material values.

Liquid mass transfer coefficient (kL) and gas mass transfer coefficient (kG)
Equation 11 can be used to determine the liquid mass diffusion coefficient: The Schmidt number (ScL) of the liquid is determined using equation 12. The Schmidt number (ScL) of the liquid is 448.372.
The value of ̅ can be calculated using the mechanical data sheet for the amine still as follows: V = gas flow rate = 58,333. 7 From the mechanical data sheet for the amine still and the calculation from the previous section, the following number is obtained: μG = gas viscosity = 0.016 cp = 1.6 x 10 -5 kg/m.s ρG = gas density = 0.1 lb/ft 3 = 1.705 kg/m 3 DG = gas diffusion coefficient = 4.12 x 10 -6 m 2 /s Therefore, Sc = 1,6 × 10 −5 1,705 × 4,12x10 −6 = 2,277 The value of gas mass transfer coefficient (kG) from equation 13 is as follows: The diffusion coefficient of the gas affects the value of the mass transfer coefficient. The greater the gas diffusion coefficient, the greater the value of the gas mass transfer coefficient. Similar to liquid HTU, gas HTU is inversely proportional to the mass transfer coefficient of the gas. Thus, the higher (the more effective) the coefficient of mass transfer of gas the smaller the HTU of gas needed.

Calculation of equilibrium slope (myx)
The equilibrium curve for the linear desorption process for low-load component transfer in gas Y and in liquid X. In other words, the slope of myx in relation to Y = f (X) can be considered fixed. The equilibrium curve can be found using equation 17.

=
(17) The goal-seek value for high packing (H) is obtained in accordance with the design data for the HTU-NTU model to determine the height of bed packing. Overall height transfer unit (HTUo) for the liquid phase (HTUOL) relates to the respective phases of HTUV and HTUL using stripping factor λ. High packing (H) is determined from the unit transfer in the liquid phase (NTUOL). NTUOL is a function of the load factor for liquid load components at both overhead (Xo) and bottom (Xu) levels and the load factor components of both the overhead (Yo) and bottom (Yu) levels.
From the flow diagram process the fraction of liquid mole data obtained are as follows: xo = 0.0287 (fraction at amine still operating pressure using HYSYS because the PFD data is still at high pressure).  A connection is apparent between the packing height values NTUOL and HTUOL. The height from the mechanical data sheet for the amine still is calculated as 12 m and the mYX value can be determined by performing a goal-seek analysis. The analysis generates an mYX value of 45. This value indicates that the gas phase transfer load component (Y) is higher than that of the liquid phase transfer load component (X) in both the overhead and bottom sides of the still. The transfer load component is a function of the mole fraction, which explains its greater magnitude in comparison to the liquid mole fraction. The amine still desorption process is the mass transfer of the CO2 component from the liquid phase to the gas phase. Mass transfer occurs to achieve equilibrium both in the gas phase and liquid phase.
The slope value is a fixed variable used in high packing calculation design. Since the packing height serves as the design data, this slope will be used as a fixed variable of variation for the increase in concentration and the flow rate.
The following numbers are the value of NTUOL and HTUOL after determining the mYX value:

Amine still design conditions for CO2 concentration in the amine still (xo)
We utilized the above design equations and results of equilibrium curve calculations to create differentiation for the liquid phase of CO2 composition in the amine still (xo) feed (rich amine). To ensure that the amine still fulfils the design conditions (12 m), a goal-seek analysis is performed for the liquid phase of CO2 composition out of the amine still (xu) (lean amine). The slope of the equilibrium curve in the desorption process can be considered constant at all concentrations from top to bottom of the amine still, hence the myx values can be used. The results of the xo to xu variations are as follows: The xo towards xu graph obtained is presented in Figure 1. As depicted, an increase in xo will lead to an increase in xu. Mass transfer in the liquid phase uses concentration, which is not in an equilibrium state, as the driving force. Hence, a voltage gradient exists in the contact surface. This situation is also known as the Marangoni effect. Consequently, the voltage gradient increases the mass transfer rate.  Figure 1, an increase in xo will cause an increase in xu, where the xu result is still within the design boundary. The xu result will exceed the design boundary (0.0013) at xo value of above 0.0307. To sum up, with the flow rate of L and V equal to the design flow rate, the maximum of CO2 composition in rich amine which can be treated by the amine still to produce the CO2 composition of lean amine which is still in conformity with the design is 0.0307.

Amine still design condition towards flow rate
The variation of the amine still design to the flow rate serves as a way to obtain the xu value of 0.0013 mole fraction, which happens to be the maximum value of CO2 allowed in lean amine. The value of the CO2 composition in the liquid phase at the bottom of the packing (xu) with a maximum limit value of 0.0013 in each variation. Using the above equations and myx results, variations in flow rates (L and V) are performed for the amine still with different xo values. The slope of the equilibrium curve in the desorption process can be considered constant at all concentrations from top to bottom of the amine still so that the myx values of the calculations can be used.
The xo value used is between the design value (0.0287) and the maximum value (0.0307). Table 3 shows the variation at xo of 0.029. Table 4 shows the variation at xo of 0.0295. Table 5 presents the variation at xo of 0.03. Table 6 demonstrates the variation at xo of 0.0305. A goalseek analysis was then performed on the liquid phase CO2 composition out of the amine still (xu) (lean amine) so that the high amine still remained in accordance with the design (12 m).
The total flow rates of xu obtained are as presented in Figure 2. As demonstrated, the increase in flow rate will cause an increase in xu. This occurs because the increase in flow rate causes fluid turbulence inside the column to increase. In turbulent flow, the molecules in the fluid will move in any direction, increasing the rate of collision between them and these collisions form gaps where the gas will be trapped. This causes increased mass transfer from one phase to another.  Table 7 presents the maximum total flow rate that can be separated by the amine still with CO2 feed composition in liquid phase (xo). The value of CO2 composition in the liquid phase at the bottom part of the packing (xu) exceeds the maximum limit value of 0.0013 in each variation. The higher the value of xo, the lower the maximum flow rate. The assorted CO2 feed (xo) composition contributes to the maximum total flow rate that can be separated, so as to produce lean amine as per specifications which will be lower for large xo values. At the value of CO2 of rich amine of xo = 0.029, the total flow rate value is 761.157.6 kg/hour. At the value of CO2 rich amine of xo = 0.0295, the total flow rate value is 628.861.1 kg/hour. At the value of CO2 rich amine of xo = 0.03, the total flow rate value is 513.962.6 kg/hour. At the value of CO2 rich amine of xo = 0.0305, the total flow rate value is 409.575.3 kg/hour.

Conclusion
This study attempts to analyze the correlation between amine still design in Field X and increase in CO2. Utilizing HTU and NTU designs, the findings suggest that an equilibrium curve slope (myx) in the amine still desorption process of 45 is acquired from the data and design equation. The maximum liquid CO2 composition of the amine still (xo) feedstock which can be separated to produce lean amine according to the specification (maximum xu 0.01 mol/mol MDEA of 0.0013 mole fraction) for the design flow rate is 0.0307. The total flow rate of CO2. rich amine at xo = 0.029 is 761,157.6 kg/hour. The total flow rate of CO2 rich amine (xo = 0.0295) is 628,861.1 kg/hour. The total flow rate of CO2 rich amine (xo = 0.03) is 513,962.6 kg/hour. The total flow rate of CO2 rich amine (xo = 0.0305) is 409,575.3 kg/hour.