ProTreat Link in VMGSim
Angela Solano - VMG Calgary
VMGSim 9.5 introduces the ProTreat Link unit operation. This highly-anticipated feature was developed in an exclusive partnership with Optimized Gas Treating to enable a live, bi-directional link between VMGSim and the ProTreat® simulation software. Users can now integrate their models seamlessly and take advantage of the rigorous calculations and specialized features that each software package has to offer in a unified environment.
ProTreat is the industry standard for rate-based simulation of gas treating processes. This simulator models separation in a column as a rate-based mass transfer process, which eliminates the need for empirical adjustments to match new applications. Columns are modeled rigorously with the number of real trays, the actual physical depth of packing, and even such details as the number of tray passes. In addition, the package supports a wide variety of solvents including proprietary ones such as INEOS GAS/SPEC, SELEXOL™, and Genosorb®.
In this article we model methanol injection in a gas gathering network and study its effect downstream on an amine unit and an acid gas injection compression train. The gathering network and compression train are both modeled in VMGSim and the amine unit is modeled in ProTreat. The entire analysis can be carried out within VMGSim because the models are linked through the ProTreat Link unit operation.
Setting up a Link
VMG is excited to offer a bridge to the powerful ProTreat engine from within the comprehensive VMGSim 9.5 simulation environment. This is possible through the ProTreat Link unit operation, which allows for seamless integration of a ProTreat model inside a VMGSim case.
A link can be set up in 3 simple steps:
(1) Drag and drop a ProTreat stencil onto the PFD,
(2) Select the ProTreat model to link to, and
(3) Define the material streams that should be linked between the models.
The ProTreat user interface can be launched directly from VMGSim after the model is selected. This allows users to view or modify the ProTreat model.
The ProTreat model is stored inside the VMGSim case file when the simulation is saved, so the combined model can be shared easily.
Material information of linked streams is automatically transferred live between programs as model conditions change.
Case Study: Methanol in Gas Processing
In this case study we combined the unique strengths of ProTreat and VMGSim to analyze the effect that methanol injection can have downstream of the gas gathering network, on both the amine unit and the acid gas injection compression train.
We started by modeling a gas gathering network in VMGSim. This model was derived from Younger’s “Gas Gathering and Processing Principles and Technology” Lecture Notes.
Hydrate unit operations were used to predict the hydrate formation approach temperature at key points in the network. The lowest approach was found in the main gathering line, with an 11.5°F approach to hydrate formation.
Two strategies for hydrate inhibition were added at this point: a methanol injection stream and a heater.
A case study was set up with each inhibition strategy operated to meet hydrate approach temperatures of 15°F and 30°F. The results are summarized in Table 1.
Table 1: Comparison of Hydrate Inhibition Operation Requirements
Methanol in the Amine UnitMethanol injection may seem more attractive at this point in terms of operating costs and scalability, but it is important to take into account the effect that trace methanol quantities will have on downstream units.
Inlet gas from the gathering system is sent directly to an amine unit for removal of CO2 and H2S. Methanol injection could disturb this amine unit. Aqueous amines absorb methanol so that it builds up in the system and affects vapor-liquid equilibrium. A rigorous simulation is necessary to predict whether this would have an effect on overall performance of the unit.
We modeled our amine system using ProTreat, the only simulation tool that can simulate methanol in an amine system on a mass transfer basis. A ProTreat Link unit operation was added in the VMGSim model to link the inlet from the gas gathering network to the amine unit.
A case study was set up in VMGSim to analyze performance of the amine unit with and without methanol injection in the gathering network. Results showed that the presence of methanol had no effect on the absorption of H2S and CO2 in the contactor (see “Treated Gas” in Table 2).
Table 2: Effect of Methanol Injection on Amine Unit Performance
This was not surprising when we examined the internal path of methanol in the contactor. The graphs below were generated in ProTreat and show that the amine solvent completely absorbs methanol within the first bottom stages of the tower. Such brief exposure is not significant enough to have an effect on the vapor-liquid equilibrium in the contactor.
Most of the methanol transferred directly from the sour gas to the rich solvent. Trace methanol in the solvent had no effect on the stripping of H2S and CO2 in the regenerator (see “Acid Gas” in Table 2), but stripping methanol proved to be difficult. Tables 3 and 4 show that almost 60% of methanol in the rich solvent is recycled and accumulates in the lean solvent.
Table 3: Concentration of Methanol in the Regenerator
Table 4: Methanol Recovery in the Regenerator
The figures below show how a trace 7 ppm methanol in the sour gas leads to a bulge of nearly 150 ppm in the vapor phase of the regenerator. Increased traffic in the tower could translate into higher reboiler duties, although within the trace amounts studied here duty remained constant.
Lastly, an important point to consider in the amine unit is emissions from the flash drum. The flash emissions tool in VMGSim was used to determine that a methanol injection of 750 ppm at the gathering system leads to an added 0.003 ton/year of methanol emissions at the flash drum. This amount is insignificant due to high H2S and CO2 content in the stream, but in any case further treatment for the flash gas would also have to manage methanol emissions.
Methanol in the AGI Compression Train
Acid gas from the amine unit is sent to a compression train to prepare it for acid gas injection. The compression train was modeled in VMGSim to study the effect of the remaining methanol on hydrate formation. Acid gas from the amine unit model was linked to the compression train through the ProTreat Link unit operation.
The inlet gas is 6 MMSCFD of 74% H2S, 18% CO2, 8% water, and trace methane and nitrogen. The compressor train was set up to compress this gas in 5 stages from 22 psia to a reservoir pressure of 2450 psia. A variety of VMGSim tools were used to generate the phase envelope and hydrate formation curve of the inlet gas. The process was designed to keep the water in the vapor phase throughout the injection circuit to avoid the need for dehydration.
Figure 1 : Process Path Relative to Dew Point and Hydrate Curves (No Methanol Injection)
Note that the gas is in a supercritical state at the end of the process, thus ensuring that there is no danger of freezing or hydrate formation during injection.
It is expected that methanol carried over from the injection upstream may act as a hydrate inhibitor in the compression process. Unfortunately, the methanol that has reached the acid gas stream is not sufficient to really affect the hydrate formation envelope in the cases that we studied (see Table 5).
The phase envelope and hydrate curve remain the same, but the slight change in composition is enough to now enter the two-phase region in the final compression stage, as shown below.
Figure 2: Process Path Relative to Dew Point and Hydrate Curves (With Methanol Injection)
Process conditions must be adjusted to once again keep the water in the vapor phase throughout the circuit. Cooling the gas an extra 10°F after the 4th compression stage is enough to accomplish this. The change comes at an energy cost of less than 1000 BTU/h, but it might not have been expected if only considering the positive effect of methanol as a hydrate inhibitor without studying its stronger, and in this case negative, effect on phase behaviour.
Changes in any area of a process can have significant consequences much farther downstream, and the wide scope of such consequences is most efficiently studied in an integrated simulation environment. The integration of ProTreat models inside VMGSim offers users the ability to run even more extensive rigorous case studies that include gas treating applications.