Menu

APR for Natural Gas 2

R. Cota, C. Landra, Y.-K. Li, J. Montelongo, L. Staudt and M. Satyro

The Advanced Peng-Robinson for Natural Gas 2 (APRNG2) is a new property package system developed by Virtual Materials Group as the new standard for gas and hydrocarbon system calculations. It supersedes the APR and APRNG property packages.

APRNG2 was specifically designed to provide consistent modeling for hydrocarbon rich and water rich phases using a flexible and consistent mixing rule, capable of modeling hydrocarbon or aqueous systems with ease using a small number of adjustable parameters, in particular systems containing methanol, water and hydrocarbons. APRNG2 is fully integrated with VMG’s multiphase envelope calculator and can provide a rich picture of highly complex mixtures of interest for the hydrocarbon processing industry.

The mixing rule formulation in APRNG2 is completely generic and can handle non-polar or polar mixtures; the necessary interaction parameters are easily accessible by the user through the Basis or Model Regression environments ensuring quick response from our Technical Support team in case client specific interaction parameters are missing as well an efficient way to continually add new interactions parameters to the model as new quality equilibrium data is published.

In addition to its flexible mixing rule for vapour-liquid-liquid equilibrium computations APRNG2 accurately models vapour pressures of non-polar and polar compounds and it includes the same special algorithms for the calculation of accurate enthalpies and liquid heat capacities used in APRNG for mixtures of water and glycols. The APRNG2 property package will be released with the upcoming versions of VMGSim 9.0 and VMGThermo 9.0.

Water / Hydrocarbon Systems

Invariably hydrocarbon production involves water. It is therefore important to be able to effectively and accurately model the presence of water in hydrocarbon streams and compute water contents in gas, solubility of water in hydrocarbons and solubility of hydrocarbons in water.

In addition to water content in gas, acid gases like carbon dioxide and hydrogen sulfide liquefy at temperatures and pressures of interest for gas processing. In particular carbon dioxide has a critical temperature close to ambient and a significant change in the water content happens when the system temperature changes from sub to supercritical as compared to the carbon dioxide critical temperature.

These effects are illustrated in figures 1 and 2.

2.png

Figure 1: Water content in carbon dioxide as a function of temperature and pressure. The average absolute error in water content in 8.7% for APRNG2 and 8.6% for APRNG. Note the dispersion in the experimental data [2] and the change from liquid to gas equilibrium near the carbon dioxide critical temperature (87.9 F). Experimental data from RR-210 [3].

1111-0001.png

Figure 2: Water content in hydrogen sulfide as a function of temperature and pressure. The average absolute error for APRNG2 is 31% while for APRNG it is 33%. Notice the dispersion in the experimental data and the change between liquid and gas equilibrium near the hydrogen sulfide critical temperature (212.7 F). Experimental data from RR-210 [3].

The performance of the APRNG2 model can be seen when calculating the water content over a series compositions is shown in figure 3.

1.png

Figure 3: APRNG2 model predictions for several binary, ternary and quaternary mixtures of methane, H2S, CO2 and propane. Experimental data from RR-210 [3].

Methanol and Water

In order to develop the APRNG2 model a series of key binary interaction parameters had to be defined based on quality experimental vapour-liquid or liquid-liquid equilibrium data. We start with the water/methanol binary, required for the correct calculation of methanol distribution between hydrocarbon and aqueous phases. The results for the methanol / water binary are summarized in figure 4.

2.jpg

Figure 4: Experimental and calculated bubble and dew temperatures for mixtures of methanol and water. Key to experimental data: 26.66, 46.66 and 66.66 kPa [4]; 101.3 kPa [4]; 303.98 and 506.63 kPa [6]; 427.43 kPa [7].

Methanol distribution over aqueous and gas phases

The proper representation of light hydrocarbons and methanol is also key for the construction of reliable models to represent the behaviour of methanol, water and hydrocarbon mixtures. It is useful to study carefully the behaviour of the main natural gas components in methanol at representative temperatures and pressures as shown for methane, ethane and propane in figure 5. 

3.jpg

4.jpg

Figure 5: Water content and methanol content of gas over aqueous solutions. Blue points represent 24 wt% methanol solutions while red points represent 49.4 wt% methanol solutions. The estimated uncertainty in the methanol or water content data is 15%. The average error in water content is 21% and the average error in methanol content is 51%. Data from [8] table 7.

Hydrates

A comprehensive evaluation of currently available hydrate data and evaluation is available on-line and discussed in the Virtual Materials Group On-Line Hydrate Validation Library section.

GPSA Figure 20-26 – Hydrate Conditions for Gases Containing C1, C3 and H2S

The results from the empirical methods recommended by the GPSA handbook and the values calculated by APRNG and APRNG2 are shown in table 1 and figures 6 and 7.

Table 1 Hydrate formation temperatures and pressures

table_1-0001.PNG

 

5.jpg

Figure 6: Sour gas Hydrate formation temperature dispersion plot. Data from [9].

6.jpg

Figure 7: Sour gas hydrate formation pressure dispersion plot. Data from [9].

Hydrate Suppression vs. Inhibitor Concentration in mole%

It is constructive to examine the performance of the APRNG2 method against other hydrate formation temperature lowering methods. Data from GPA Research Reports 92 and 106 were collected at diverse inhibitor concentrations and the formation temperature was computed using APRNG and APRNG2. The results are summarized in figure 8 as a hydrate formation temperature difference from the value obtained without inhibitor.

7.jpg

Figure 8: Hydrate suppression using methanol and ethylene glycol. Based on figure 20-53 from [9].

Figure 8 is quite interesting and some of its key features are not be evident. The points as labeled as in figure 20-53 from the GPSA handbook are hard to interpret in an unambiguous manner because they are points calculated using different mixtures and apparently they are average depression temperatures for these mixtures.

The mixtures range from methane, ethane, propane, carbon dioxide, hydrogen sulfide, gas condensate and lean and rich gases as defined in GPA RR-92 and GPA RR-106 and the inhibitors are methanol and ethylene glycol at diverse concentrations.

If instead of plotting calculated averages we were to plot the actual calculated hydrate depression temperatures the plot would like figure 9. Only the APRNG2 results are shown to avoid overcrowding the figure.

Note that there’s a significant spread in the calculated hydrate depression temperatures which is caused by the differences in fluid composition and inhibitor concentrations, and the depression temperatures suggested by figure 20-53 can be subject to significant error. The actual errors are impossible to quantify based on the available data but they increase with inhibitor concentration and a tentative guideline is proposed in table 3.

It must be stressed that table 1 is not a table of errors in the usual sense but rather it is a tentative guess of uncertainties in the estimated hydrate depression temperatures based on experimental errors, model errors and fluid compositions. The “uncertainty” is defined in this context as 2 times the standard deviation of the hydrate depression temperature for a fixed fluid composition and inhibitor concentration at the reported formation pressure as calculated by the APRNG2 model.

Table 2 Tentative uncertainties in hydrate formation temperatures for use with GPSA figure 20-53

 table_2-0001.PNG 

Correlations for the tentative uncertainty of depression temperatures for inhibitors are shown in equations 1 and 2.

image020-0001.png(1)

image022-0001.png(2)

We note that there’s a definitive trend for increased uncertainty as the inhibitor concentration increases. The small number calculated for the 16% inhibitor mole percent value comes from the value being based on a single methanol depression point for condensate.

Notwithstanding the usefulness of the data presented in the GPSA handbook some of the data is complex and a simple presentation hides significant information and critical thinking is required before interpreting its results.

last.jpg

Figure 9: Hydrate suppression calculations now showing not the average depression temperature but the depression temperatures calculated using the actual fluid composition and reported formation pressures. Note the spread in depression temperature (green circles) versus reported values in figure 20-53 (red squares).

 

Bibliography

  1. Huron, M.J. and Vidal, J.; "New Mixing Rules in simple equations of state for representing vapor-liquid equilibria of strongly non-ideal mixtures"; Fluid Phase Equilibria 3, 1979
  2. ThermoData Engine (TDE) version  8.0 (pure compounds, binary mixtures, ternary mixtures and chemical reactions). NIST standard reference database. Developed by Micheal Frenkel Robert D. Chirico, Vladimir Diky, Kenneth Kroelein, Chris D, Muzny, Andrei F, Kazakovv, Joseph W. Magee. Llmutdin M, Abdulahatov and Eric W Lemmon, Thermodynamic research centre (TRC), Thermophysical properties Division, NIST, Boulder CO 80305-3337. 
  3. Research Report RR-210 – Acid Gas Water Content, An Update of Engineering Data Book I; K.Y. Song, T. Vo; M. Yarrison and W.G. Chapman; Gas Processors Association, June 20, 2012
  4.  Othmer, D. F.; Benenati, R. F.; Ind. Eng. Chem., 1945, 37, 299.
  5. Kurihara, K.; Nakamichi, M.; Kojima, K.; J. Chem. Eng. Data, 1993, 38, 446.
  6. Hirata, M.; Ohe, S.; Nagahama, K.; Computer Aided Data Book of Vapor-Liquid Equilibria, Kodansha Ltd. and Elsevier.
  7. Swami, D. R.; Rao, V. N. K.; Rao, M. N.; Trans., Indian Chem. Eng., 1956, 9, 32
  8. Hong, J. H.; Kobayashi, R.; Fluid Phase Equilib., 1988, 41, 269
  9. GPSA Engineering Data Book 13th Edition (electronic) FPS Volumes I and II
  10. www.naesb.org/pdf2/wgq_bps100605w2.pdf last accessed November 20th 2014
  11.  Research Report RR-92; The Effect of Ethylene Glycol or Methanol on Hydrate Formation in Systems Containing Ethane, Propane, Carbon Dioxide, Hydrogen Sulfide or Typical Gas Condensate; Gas Processors Association, September 1985
  12. Research Report RR-106; The Influence of High Concentrations of Methanol on Hydrate Formation and the Distribution of Glycol in Liquid-Liquid Mixtures; Gas Processors Association, April 1987
  13. Hong, J. H.; Malone, P. V.; Jett, M. D.; Kobayashi, R.; Fluid Phase Equilib., 1987, 38, 83. Ohgaki, K.; Sano, F.; Katayama, T.; J. Chem. Eng. Data, 1976, 21, 55.
To Top