Single Well Tracer Test

(SWTT) Modeling with

STARS

STARS v2019.10
Builder - Results v2019.10
Tutorial
Table of Contents

TABLE OF CONTENTS ..........................................................................................................................1

INTRODUCTION .................................................................................................................................2

EXERCISE 1 – BUILDING BASE MODEL FOR SWTT .................................................................................3

Overview of Exercise 1 3
Creating Base model 3

PLOTTING THE RESULTS.................................................................................................................... 21

SENSITIVITY TO THE RESIDUAL OIL SATURATION ............................................................................... 26

SENSITIVITY TO THE DISPERSION COEFFICIENT .................................................................................. 30

SENSITIVITY TO THE HYDROLYSIS REACTION RATE ............................................................................ 33

THERMAL MODELING OF SWTT (NON-ISOTHERMAL) ........................................................................ 38

1
SWTT Modeling with STARS
Introduction
Single Well Tracer Test (SWTT) is a technique used to determine residual oil saturation. This
method has been applied in several fields around the world and is widely accepted to make an
in-situ measurement, given that the radius of investigation can be several feet around the well
(15-30 ft).

Numerical simulation can be used to model the planning or interpretation of a SWTT. In this
document the procedure of how to build a model for Single Well Chemical Tracer Test (SWCTT)
using CMG’s STARS simulator is presented. For this tutorial, the basic information to create the
model was taken from field data provided in SPE paper 164341*. However, not all the data was
available and, so, data was either taken from other sources or assumed.

This tutorial has the following exercises and structure:

 Building a base model for SWTT

 Plotting the results
 Sensitivity to the main parameters that must be adjusted to match the tracer’s profiles:
residual oil saturation, dispersion and FREQFAC.
 Study the effect of injecting cold water (with respect to the temperature of the in-situ
reservoir water) in the profile

This workflow is designed to be used with any test although specific data entries are shown in
these examples which can be modified to suit your particular case or conditions.

* Farida Abdulla, Hashem Sayed Hashem, Badria Abdulraheem, Meqdad Al-Naqi, Abrar Al-Qattan and
Huw John, P. R. P. Cunningham, P. J. Briggs and Roxy Thawer, “First EOR Trial using Low Salinity Water
Injection in the Greater Burgan Field, Kuwait”, SPE 164341, SPE Middle East Oil and Gas Show and
Conference, 10-13 March, Manama, Bahrain, 2013.

2
SWTT Modeling with STARS
Exercise 1 – Building Base model for SWTT
Overview of Exercise 1
A step by step procedure it is presented in this section for modelling Single Well Tracer Test
with STARS.

Creating Base model

The procedure for building the simulation model is described below:
1. Open CMG Launcher. Double-click the Builder icon.
2. In the top menu of Builder select File|New.
3. Select STARS as the simulator, Field working units, Single porosity and a start date of
2000-01-01. Click OK twice to confirm the settings.

4. Create a radial grid by clicking Reservoir|Create Grid|Radial (Cylindrical). Use the

following parameters:
 Number of division along r = 490
(*Outer radius/Block Widths I direction = 35 ft / 0.0714 ft = 490)
 Theta = 1
 Number of division along K = 1
 Inner radius of innermost block = 0.25 ft
 Outer radius of outermost block = 35 ft*

3
SWTT Modeling with STARS
 *The radius of the model is calculated based on the volume of water injected and the
reservoir properties (check the excel file Calculation.xlsx). This value is very important
given that model must be big enough to avoid the production of the tracer at the
boundary wells.

 Sweep = 3600
 Grid block Widths in I direction = 490*0.0714
 J-direction = 360

Click OK to create the grids.

5. Click Probe mode , then click the Specify Property button. Select the property from the
drop-down list in Go to Property and enter the following values for these properties:
 Grid Top (Layer 1) = 6640
 Grid thickness (Whole Grid) = 28

4
SWTT Modeling with STARS
  Porosity (Whole Grid) = 0.18
 Permeability I (Whole Grid) = 350
 Permeability J (Whole Grid) = 350
 Permeability K (Whole Grid) = 35

6. Click OK twice to populate the properties in the blocks.

7. In the upper menu select the option File|Save as; and replace the name CMGBuilder00.dat
with swctt-tutorial-base.dat.

During the SWTT an ester is injected into the formation which will react with the
water to produce an alcohol. For this test, Ethyl formate (ester) is injected and
ethanol is the alcohol produced during the hydrolysis. The reaction is detailed below:

In general two additional tracers are injected in SWTT: the mass balance tracer and
the cover tracer to check the quality or any deviation during the test. Both are very
important but won’t be included in this example given that the data was not available.
Another ester that is frequently used is ethyl acetate. Simulation reactions must be
defined based on the type of ester.

8. The following components will be defined for modelling the SWCTT: Water, Esther-Ethyl
Formate (Etf), Ethanol, and Dead_Oil. Formic Acid (Acid) will be also included to balance the
reaction but its profile won’t be used for further analysis. To create these components, click
Components|Add/Edit Components. The Components and Phase Properties dialog box
opens.

9. Click Add/Edit Component , click and select Add a component.

Enter the name of the component as Water. Click OK.
10. Enter the Molecular weight as 18.02 lb/lbmole and the reference phase as ‘Aqueous’. As
shown below, also enter ‘0’ in the ‘Critical Pressure’ and ‘Critical Temperature’ boxes. Click
OK.
11. Repeat steps 9 and 10 to create the other required components EtF, Ethanol, Acid and
Dead_Oil. The molecular weights for EtF, Ethanol, and Acid are as follows: 74.08, 46.07, and
46.03 lb/lbmole, respectively. The molecular weight that will be used for Dead_Oil is 128
lb/lbmole and the reference phase is Oleic (oil-like). Refer to the following dialog boxes and
when completed, click OK.

5
SWTT Modeling with STARS
 6
SWTT Modeling with STARS
12. To define the Liquid-Liquid k-value for the component EtF, click Components and then K-
Values, and enable the Enter K value data check box.
13. Click , select Liquid-Liquid under K value tables for and then select
Component EtF. Click , click OK in the pop-up screen, and then enter the
parameters shown in the following dialog box. Finally click OK.

14. In the K Value Table dialog box, shown below, enter the K-values as 15.059 and then
click OK.

7
SWTT Modeling with STARS
 In this example, the lab measured Liquid-Liquid K-value is 2.12. This measurement is a
resultant mass fraction rather than a mole fraction; however, STARS requires the input
file to be expressed as a mole fraction. Therefore the K-value can be calculated as
follows:
𝑀𝑜𝑙𝑒𝑐𝑢𝑙𝑎𝑟 𝑊𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 𝑂𝑖𝑙 128
K-Value = 𝑀𝑜𝑙𝑒𝑐𝑢𝑙𝑎𝑟 𝑤𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 𝑤𝑎𝑡𝑒𝑟 × Kval lab mass fraction = 18.02 × 2.12 = 15.059

Notes:

Liquid-Liquid K values are measured in the lab and supplied for the interpretation

If the oil phase has multiple components then the total molecular weight must be
calculated and used in the above formula.

15. In the Liquid phase properties/Densities tab enter the densities and liquid compressibilities
for the components, by selecting Mass density from the Options drop-down list, as shown
below. Click Apply.

16. From the Liquid Phase Viscosities tab, select Use Viscosity Table and Single Table.
Right-click over the table and select Insert row. Enter the values shown below and click OK:

8
SWTT Modeling with STARS
17. Select Components|Reactions and then click on Add a Reaction. The reaction to be
added is:
1 Water + 1 EtF  1 Acid + 1 Ethanol.
 Enter the Frequency factor (FREQFAC) = 0.3

Note: Use this term to scale for matching the peaks of the EtF and Ethanol.

 Enter the values shown below (notice that the mass balance error has a very low value
3.9e-15):
 Click OK.

9
SWTT Modeling with STARS
18. Click over Rock-Fluid|Create/Edit Rock Types, and then click New Rock Type. Clear
the Include Capillary Pressure check box.
a. In the Relative Permeability Table, select Water-Oil Table and copy the following
values into the table:
Sw Krw Krow
0.2 0 1
0.5 1 0
Click Apply then repeat for Liquid-Gas Table in step (b) below.
b. In the Relative Permeability Table, select Liquid-Gas Table. Clear the Include
Capillary Pressure check box and copy the following values into the table:
Sl Krg Krog
0.7 1 0
1 0 1
Click OK once complete.
19. Click Initial Conditions|Initialization Settings, then select Do Not Perform Vertical
Equilibrium Calculations (VERTICAL OFF) (non-equilibrium initialization) and click OK.

Note: Additional properties have to be specified as non-equilibrium initialization is implemented.

20. Double-click Numerical Controls under the Numerical section. Then enter the following
values and options and click Ok:
i. DTMAX = 0.001 day
ii. DTWELL = 0.001 day
iii. Isothermal Option (ISOTHERMAL) = ON
iv. Model Formulation (TFORM) = ZT
v. Adaptive Implicit Method (AIM) = “Stability Switching
Criterion”.

Note: The maximum time step size of 0.001 (indicated above) is important for minimizing
numerical dispersion for this type of model.

10
SWTT Modeling with STARS
STARS will be run in isothermal mode; hence ISOTHERMAL ON and TFORM ZT is selected.

21. Save the model and close Builder

22. Open your model with CEDIT and explore the model
23. Go the numerical section and add the keyword “CONVERGE MAXRES”. Your model must
look like the figure below. Save you model and close CEDIT

11
SWTT Modeling with STARS
24. Open your model again in Builder and continuous with the following steps.
25. Click Probe mode then click the Specify Property box. Select the property from the drop-
down list in Go to Property and enter the following values for these properties (for whole
grid):
 Implicit flag (AIMSET) = 0
 Oil Mole Fraction (Dead_Oil) = 1
 Pressure (PRES) = 2750 psi
 Temperature = 163 F
 Water Mole Fraction (Water) = 1
 Water Saturation = 0.5
Click OK twice to populate these properties.
26. Click File|Save to save the dataset with these changes.

Note: based on the input of the initial water saturation of 0.5 and the relative permeability curve
it is clear that the model is initialized at residual oil saturation (Sorw). This method assumes that
the pre-flush water injection stage normally done as part of a SWCTT has swept the zone around
the well so that we are modeling Sorw.

The remaining part of the model is the Recurrent Section, where the wells and their constraints
are defined. The following points are the most important aspects related to this section:
 Volumes and injection periods of each stage of the test (also concentrations of the fluids that
are injected). The following table presents the details

12
SWTT Modeling with STARS
 Table 1. Sequence events for the SWTT
Cumulative
Stage Volume (stb) Rate (b/d) Time (days)
Time (days)
Injection - Ester 160 1000 0.16 0.16
Injection – Push bank 820* 1000 0.82 0.98
Shut-in period 0 0 2.75 3.73
Production 1327 Variable 1.3 5.03
* This volume was modified from the original paper to get a consistent profile.
 Number of wells: one injector and one producer are required to model the test but also two
additional wells must be located at the boundary to maintain constant pressure at the limit of
the reservoir.

Note: Another alternative way of modeling can be done using additional blocks and at the outer
boundary (block) use volume modifiers to keep constant boundary conditions. If this option is
implemented then the boundary wells are not required.

 Concentration of the tracers: For the injection period the ester concentration is assumed to
be constant and equal to 10072 ppm. The production period is presented in detail in the file
“Producer.fhf” (check the data required for the model).
Over the following steps it is presented the details about wells definitions and constraints.

27. Select Well|Well New.(Defining the injection well)

a. In the Create New Well screen, enter the Name as ‘Inj’.
b. Set Type to ‘INJECTOR MOBWEIGHT EXPLICIT’.
c. In the Constraints tab, select Constraint Definition.
d. In the Constraint column, select ‘OPERATE’ from the drop-down menu and ‘STW
Surface Water Rate’ from the drop-down menu under the Parameter column.
e. Enter 1000 bbl/day as ‘Operate STW Max’.
f. In the Injected Fluid tab, select Water as the Injected Fluid, with mole fraction of
Water = 0.99753, EtF = 0.00247.
Ester’s concentration is generally given in PPM (10072 ppm for this case). Given that the
simulator requires mole fraction of the fluid injected, the appropriate conversion must be
done. In the “Calculation.xlsx” Excel file you will find a spreadsheet where this
calculation is already done.
g. Check the Temperature box and enter as 163 F.
h. Click .
28. The next well to define will be a producer in the boundary ring of the model to keep constant
pressure at the boundary (Defining the boundary producer well).
a. In the Create New Well screen, enter the well Name as ‘Boundary’.
b. Set Type to ‘Producer’.

13
SWTT Modeling with STARS
 c. Set Operate BHP Min to 2750 psi (equal to initial reservoir pressure) using the same
steps as for the previous well (steps c and d) to enter this value. Click OK.
29. Under the Well & Recurrent menu, double-click Dates to open the Simulation Dates dialog
box. Click beside Add a new date. Enter ‘0.16’ under Days since Simulation Start. This
specifies the elapsed time of injection (Period when the Ester is injected). Click OK then click
Close.
30. Double-click Wells in the Wells & Recurrent menu.

a. Click on Inj and select .

b. In the Constraints tab, check the Constraint definition box and click Apply.
c. In the Injected Fluid tab, change the Water Mole Fraction to ‘1’ as shown below:

Click OK. This will start injection of water from 2001-01-01.1599 to push the ester deeper
into the formation. Prior to this, the injection involves Water 0.99753 and EtF 0.00247.
31. Double-click Dates in the Wells & Recurrent menu to open the Simulation Dates dialog box.
Click beside Add a new date. Enter ‘0.98’ in the Days Since Simulation Start box. This
specifies the start of the shut in period. Click OK then click Close.
32. Double-click on the Wells tree to open the Well Events dialog box.

a. Click on Inj and select

b. In the Options tab, make status SHUTIN.
c. Click Apply.

d. Click on Boundary well and select

e. In the Options tab, make status SHUTIN
f. Click Apply and then OK.

14
SWTT Modeling with STARS
 g. The tree view should appear as follows:

33. Double-click on Dates in the Wells & Recurrent menu to open the Simulation Dates dialog
box. Click beside Add a new date. Type ‘3.73’ in the Days Since Simulation Start box.
This specifies the start of the flow back. Click OK then click Close.

Note: For the flow back period of the test, we need to define two more wells

34. After doing the SWTT, information related to the cumulative water produced and tracer
concentrations profiles will be available. Importing this data to the model is very important
given that normally, water rate during the test is not constant and tracer profiles can be
affected by this factor. For this exercise the data is available in the file “Producer.fhf”. Next,
how to import the historical water production data:
a. In Well & Recurrent section select the option “Import Production/Injection Data”
(adding the producer well with historical data).

15
SWTT Modeling with STARS
 b. Select the option “FHF” as the format type for your data and look for the location of
your “fhf” file using the Browse option. Click Next.

c. Click Finish.

A window with the Simulation Dates will appear. Click on the option “Delete all empty dates” and
close.

16
SWTT Modeling with STARS
You will see that a new well was created called “Prod”. Now a boundary injection well must be
defined to operate at the same time of “Prod” well.
35. In the menu bar, select Well|Well New to open the Create New Well dialog box (Defining
the boundary injection well).
a. On the ID &Type tab, enter the Name as ‘BoundaryInj’ and click Ok
b. Double click on the well “BoundaryInj” in the Well & Recurrent section, look for the
date and then define the well as is shown below and click Apply

17
SWTT Modeling with STARS
 c. On the Constraints tab select Constraint definition and specify Operate BHP Max
2750 psi and click Apply.
d. In the Injected Fluid tab, select the Injected Fluid as Water, with mole fraction of
Water = 1.
e. Check the Temperature box and enter 163 F.
f. Click Ok.

36. Double-click Dates on the Wells & Recurrent tree. Click beside Add a new date. Enter
‘5.13’ in the Days Since Simulation Start box. This specifies the end of simulation. Click OK.
Select and check this date under the set STOP section and then click Close.

37. To define the well perforations for the created wells, select Well|Well Completions (PERF)
in the menu bar.
a. Select Boundary under the Well & Date drop-down list.

b. If no date is defined (inactive for entering any data) then click on the button and
select the option “Completion - Add New” and click OK (This completion will be
defined at the beginning of the simulation).
c. In the General tab, set radius (ft) to ‘0.01’.

Note: Ensure that Use layer IJK directions or XYZ geometric data is set to None.

d. Click Apply.

e. In the Perforations tab, click . Enter the block address as 490 1 1.

f. Click Apply.
g. In the upper part select BoundaryInj under the Well & Date drop-down list.
h. Under the General tab, set radius (ft) to ‘0.01’.

Note: Ensure that Use layer IJK directions or XYZ geometric data is set to None.

i. Click Apply.

18
SWTT Modeling with STARS
 j. In the Perforations tab, click . Enter the block address as 490 1 1.
k. Click Apply and then Yes.
l. Select Inj under the Well & Date drop-down list.
m. Select Yes to apply changes.
n. Click then select Completion - Add New. Click OK or Cancel when the New Well
Completion Date dialog box appears (This step is in case that you can’t see date
available for defining the completion)
o. In the General tab, set radius (ft) to 0.25.

Note: Ensure that Use layer IJK directions or XYZ geometric data is set to None.

p. Click Apply.

q. Select the Perforations tab and click . Enter the block address as 1 1 1. Click
Apply.
r. Select Prod from the Well & Date drop-down list.

s. If no date is defined (inactive for entering any data) then click on the button and
select the option “Completion - Add New” and click OK.
t. In the General tab, set radius (ft) to 0.25.

Note: Ensure Use layer IJK directions or XYZ geometric data is set to None.

u. Click Apply.

v. Select the Perforations tab and then click . Enter the block address as 1 1 1. Click
Apply, Yes and then OK.
38. Select Well & Recurrent|Open Time-Line View. Right-click on the year 2000 (displayed at
the top) and change the timeline to ‘Day’. Check that the Injection and Production wells in
timeline view and the period of the wells marked by the blue and black lines are as shown
below:

19
SWTT Modeling with STARS
39. It is required to output the EtF, and Ethanol mole fractions for plotting purposes. To enable
this option:
a. Select I/O Control|Simulation Results Output.
b. Under OUTSRF, click and select Well from the drop-down box. Click OK.

c. Select the values in the Variables selection boxes as shown below:

d. Under OUTSRF, click to open the Select New Date And Type dialog box.
Select Special from the drop-down box then click OK.
e. In the OUTSRF table, click beside Special.

20
SWTT Modeling with STARS
 f. Select MASSFRAC under Special Variable Name, highlight it with the mouse,
then, at the top left of the dialog box, click Insert|Insert and copy a row.
g. Select Prod under Well 1, and components EtF and Ethanol, one after the other,
as shown below:

h. Click OK.
i. Click over for the Grid properties and include the component composition
for oil, water, and all the phases (see figure below)

j. Click OK two times.

40. Save and run the dataset using only one processor.

Plotting the Results

One particular point related to the interpretation of the SWTT is that tracer profiles are plotted
versus cumulative water (not versus time) for the interpretation. For ideal cases some information
can be extracted from this plot to calculate the residual oil saturation. Visualizing the movement
of the tracer through the reservoir (grid) can be also useful, especially for cases where vertical
heterogeneities play an important role. In this section, the visualization of the results will be
presented.

Historical data must be prepared for doing the match and interpretation of Single well trace test.
The required file for plotting the historical data is “Producer.fhf”. This file contains the actual test
information that was measured in the field. There are four columns in the file which represent:
Time, Cumulative Water, Water Mass Fraction for Ethanol, and Water Mass Fraction for ester
(both of them in the aqueous phase).

21
SWTT Modeling with STARS
 Effluent concentration during the SWTT are reported in PPM which requires a
conversion for creating the historical data (mass fraction). Water Mass Fractions for
Ester, Ethanol or any other specie can be calculated through the following expression:
𝑃𝑃𝑀𝑖
𝑊𝑎𝑡𝑒𝑟 𝑀𝑎𝑠𝑠 𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛𝑖 =
1000000
Where i is the component (Ester or Alcohol).

To plot the results follow the procedure below:

1. Select the ‘swctt-tutorial-base.sr3’ file, then drag and drop it in Results .

2. In the Input section at the PROJECT NAVIGATION, select Data Sources and on the right
side click over Add Files. Locate and open the ‘Producer.fhf’ file.

22
SWTT Modeling with STARS
3. Go to the Plots section and click the Time Series vs. Time Series.

4. Select the option as you can see over the figure below for plotting mass fraction (Special
History) vs. Cumulative water, and then click Add to New Plot.

5. The previous plot represents the results from the simulation model. Repeat the previous step
to add the results from the historical data but for DATA SOURCES select the fhf file (see figure
below).

23
SWTT Modeling with STARS
6. Click the “ Auto hide” option and you will see the following profile

The peak’s separation between the Ethanol and Ester is used to calculate the Sor; however, there
is not a match between the historical and the simulated data. There are several factors that
controls the simulation tracer response such as: dispersion, residual oil saturation, and the
hydrolysis reaction rate. In the following sections sensitives will be done over these parameters
to see how they affect the profiles.
We can visualize the movement of the Ester inside of the reservoir and the reaction for generating
the alcohol in Results. Below you will find an example:

24
SWTT Modeling with STARS
7. In the left menu select Dashboards and click the option Create Dashboard

8. Drag from the Reservoir section an Areal View into the Dashboard.
9. Repeat the previous step. The main idea is to have the ester and alcohol profiles in the
Dashboard.
10. Click one of the plots and then the Display menu at the top. Select as the variable to visualize
the “water Model Fraction(EtF)”

11. Click over the icon to inactive the Grid Lines.

12. In the upper menu select “DASHBOARD” and deactivate the synchronize properties

13. Click on the other plot and then on the Display menu at the top. Select as the variable to
visualize the “water Model Fraction(Ethanol)”. Deactivate the Grid Lines also for this plot.
14. Change the time and see the evolution of tracer concentration profiles at different times (See
example below during the production period).

25
SWTT Modeling with STARS
15. Go to the upper menu and select: FILE|Save Project and save your session as
SWTT_results.

Sensitivity to the residual oil saturation

One of the main objectives of the interpretation of the SWTT is finding the residual oil saturation
after the water or chemical flooding. For that reason it is important to understand how the tracer
profiles behave with respect to this variable.

In this exercise different scenarios will be run where the residual oil saturation will be modified.
Changes to the model must be done over the relative permeability curve and the initial water
saturation, assuming that the SWTT is done after the water flooding (or chemical flooding).

1. Open the model “swctt-tutorial-base.dat” in Builder save as “sensitivity_sor_0.3.dat” over

in upper menu File|Save as.
2. Click on the “Rock-Fluid->Edit Rock Types…” in the upper menu (see figure below)

3. Change the maximum water saturation of the table from 0.5 to 0.7 (this means residual oil
saturation Sor= 1-0.7 = 0.3), and then click OK.

26
SWTT Modeling with STARS
4. Click on Reservoir under Model Tree View and then click on “Specify Property”
5. Set the water saturation equal to 0.7. Click Ok two times

6. Save the model and run the simulation.

7. Repeat the previous steps (1-7) and create two additional cases with residual oil saturation of
0.195, and 0.10.

Note: Change the names of the files because results for all the cases will be plotted. You can
use the following names for the new cases: “sensitivity_sor_0.195.dat” and
“sensitivity_sor_0.1.dat”

8. Run all the models and proceed to plot the results to compare the scenarios (detailed next)

9. Take the file “sensitivity_sor_0.3.dat”, and drop it on the Results icon .

10. In the Input section select Data Sources and on the right side click on Add Files. Locate
and open the following files: “sensitivity_sor_0.195.dat”, “sensitivity_sor_0.1.dat”, and
“Producer.fhf”.

27
SWTT Modeling with STARS
11. Go to the Plots section and click on the Time Series vs. Time Series.

12. Select the option you can see on the figure below for plotting mass fraction (Special History)
vs. Cumulative water, and click Add to New Plot.

28
SWTT Modeling with STARS
13. Repeat the previous step to add the results for the other cases (see figure below).

14. Click on the “ Auto hide” option and you will see the following profile (NOTE: remove the
historical from the visualization)

29
SWTT Modeling with STARS
15. Go to the upper menu and select: FILE|Save Project and save your session as SWTT_sor.

In the figure presented above one of the main principle of Single Well Tracer Test can be
explained. Separation between tracers’ peaks (ester and alcohol) is based on the residual oil
saturation and partitioning of the ester between the oil and water (K value). For this case K value
is constant, so the residual oil saturation is the only variable affecting the results. In the previous
figure it can be observed that if the residual oil saturation decreases, then less separation is
observed between the peaks of ester and alcohol.

Sensitivity to the Dispersion Coefficient

The propagation of injected tracers and chemicals employed in EOR processes are influenced by
the tortuous flow paths and (random) heterogeneities of the porous media in which they flow.
Normally this contribution to dispersion (the broadening and spreading of concentration fronts)
dominates due to molecular diffusion. In the previous exercise, it can be observed that historical
tracer profiles are wider than the simulated data which is an indication that dispersion must be
considered.

Dispersion coefficients can be added in STARS to get a better match of the peak. In this exercise
sensitivity will be done on the dispersion of the tracer components

30
SWTT Modeling with STARS
1. Open the model “sensitivity_sor_0.195.dat” in Builder.
2. Click on Probe mode
3. Then click Specify Property.
4. Select the property from the drop-down list in Go to property and enter the following values for
these properties (under Whole Grid):
Water Dispersion Coef. I (Etf) = 0.1 (ft*ft)/day
Water Dispersion Coef. J (Etf) = 0.1 (ft*ft)/day
Water Dispersion Coef. K (Etf) = 0.1 (ft*ft)/day
5. Repeat the previous step for the components: Ethanol and Acid.

6. Click OK twice to populate these properties.

7. Save the model as “Sensitivity_Disper_0.1.dat” and run the dataset.
8. Two additional cases will be created but CEDIT (CMG text editor) will be used.

9. Drag and drop the case “Sensitivity_Disper_0.1.dat” over the CEDIT icon.
10. Click on the Rock-Fluid Data and on the right side replace the dispersion coefficient from 0.1
to 0.25.
11. Save the model as “Sensitivity_Disper_0.25.dat”

31
SWTT Modeling with STARS
12. Run the previous cases that were created.
13. Take the results project from previous exercise (SWTT_sor.results) and save as
“SWTT_sens_disp.results”
14. Go to the Input|Data Sources section and right click on “sensitivity_sor_0.1.sr3”. Then
select the option Replace File and substitute with the file “Sensitivity_Disper_0.1.sr3”.

32
SWTT Modeling with STARS
15. Repeat the procedure and replace the file “sensitivity_sor_0.3.sr3” by
“Sensitivity_Disper_0.25.sr3”.
16. Go to the Time Series vs. Time Series section and see the results. Your plots should looks
like the figure below.

The conclusion that can be extracted from the previous figure is that increasing the dispersion
coefficient of the tracers, produces wider and lower profiles (a decrease in the height of the peak).
We have demonstrated that dispersion can be a history match parameter.

Sensitivity to the hydrolysis reaction rate

Given that alcohol is generated through the reaction of ester with water (hydrolysis), the profiles
of both species are related to each other. From the previous figure the alcohol simulated profile
is very low compared to the historical data while for the Ester profile the reverse trend is observed.
The previous point reflects that more ester must be reacting and thus producing more alcohol
(less Ester will be produced). The way of controlling this parameter is through the velocity of
hydrolysis reaction. In this section, sensitivity will be done on the hydrolysis reaction rate for the
ester. FREQFAC is the keyword that will be used to control this parameter

33
SWTT Modeling with STARS
1. Open the model “Sensitivity_Disper_0.25.dat” in Builder
2. Click on the “Components->Reactions…” in the upper menu (see figure below)

3. Change the Reaction frequency factor to 0.6, and then click OK.

4. Save the model as “Sensitivity_Freqfac_0.6.dat”.

5. Repeat the previous steps (1-3) and create two additional cases with Reaction frequency
factor of 0.8 and 1.0 (name these cases “Sensitivity_Freqfac_0.8.dat” and
“Sensitivity_Freqfac_1.dat”).
6. Run all of the cases
7. Now, the results of all the cases will be plotted together

34
SWTT Modeling with STARS
8. Open the template “SWTT_sens_disp.results” (from previous exercise) and save as
“SWTT_sens_freqfac.results”.
9. Go to the Input|Data Sources section and right click on the file “Sensitivity_Disper_0.1.sr3”
and replace with “Sensitivity_Freqfac_0.6.sr3”.

10. Repeat the procedure above and replace the other two models with the file names
“Sensitivity_Freqfac_0.8.sr3” and “Sensitivity_Freqfac_1.sr3”.

35
SWTT Modeling with STARS
Looking at the previous plot it is clear that by increasing the velocity of the reaction (increasing
FREQFAC value) the area and height of Ester profile decreases, while the alcohol profile
increases. This point can be explained given that as FREQFAC increases, then more Ester will
be reacting with the water and less of this tracer will be produced. For the Alcohol the trend is
inversed given that as more ester reacts, then more alcohol is generated.
From the previous results it can be concluded that dispersion and hydrolysis reaction rate affect
the height and width of the profile but not so much their separation (ester and alcohol). For
matching the position of the peaks, it is required to change the residual oil saturation which is the
main objective of SWTT.
A reasonable match of the profiles is obtained with the following parameters (see figure below
with the profiles). However, you can try with other sets of parameters.
 Residual oil saturation = 0.207
 Dispersion Coefficient = 0.22 (ft*ft)/day
 Reaction frequency factor (FREQFAC) = 0.82
11. A case will be created with the previous parameters.
12. Open the model called “Sensitivity_Freqfac_0.6.dat” with CEDIT and save as
‘Final_Model.dat’.
13. Change the dispersion, FREQFAC, residual saturation over the relative permeability and initial
water saturation.

36
SWTT Modeling with STARS
14. Save and run the case
15. Plot the historical data with the results of the model.

37
SWTT Modeling with STARS
There was limited information building this model. It could be possible to get a better match if a
more detailed vertical description is added in this model.

Thermal Modeling of SWTT (non-isothermal)

If the region sampled by the Ester and Alcohol are not kept at a constant temperature,
dependency of the K values on the temperature can complicate the estimation of residual oil
saturation. A similar condition may occur when a colder fluid is injected into a hotter formation,
adding some uncertainty to the K value to be used. This means that the match will have to be re-
iterated with a new set of parameters for a thermal model.

This exercise is designed to demonstrate that profiles can be affected when the temperature of
the fluid that is injected is colder than the formation fluids.

1. Open the model ‘Final_Model.dat’ in Builder. Click File|Save As and then save the dataset
as ‘swctt-tutorial-thermal.dat’.
2. Double-click Thermal Rocktypes in the Reservoir section to open the STARS Other
Reservoir Properties dialog box. Click beside the Rock Type box and select New
Thermal Rock Type. Select the Thermal Properties tab and enter the following values:
 Reservoir Rock= 44 Btu/(ft*day*F)

38
SWTT Modeling with STARS
  Water Phase = 8.6 Btu/(ft*day*F)
 Oil Phase = 1.8 Btu/(ft*day*F)
 Gas Phase = 0.7 Btu/(ft*day*F)

Click OK.

3. For the temperature dependence of the reaction of EtF changing to Ethanol, it is expected
that as the temperature increases, the amount of EtF that changes to Ethanol should also
increase. To add this dependency, go to the Components section, double-click on Reaction:
1 and enter the activation energy (EACT) = 20000 Btu/lbmole, and change the FREQFAC
value to 0.847e7.
4. Click OK and note that the Builder plot of Reaction Arrhenius Constants gives a value of 0.8
at a temperature of 163, which is the value that was used previously for the isothermal model
(i.e. with an EACT=20000, FREQFAC was calculated to result in the same reaction rate of
0.8 at 163 F). From this plot in Builder, observe the reaction rate increase with increasing
temperature. This value of activation energy gives reasonable temperature dependence for
this example; however, this value should be measured in the laboratory for a more accurate
estimation of this effect.

39
SWTT Modeling with STARS
5. For the temperature dependence of EtF partitioning into the oil phase using K values, it is
expected that as the temperature increases, the amount of EtF that partitions into the oil phase
should increase. Recall that the liquid-liquid K values for an aqueous component are defined
𝑙𝑖𝑞 𝑜𝑖𝑙 𝑚𝑜𝑙𝑒 𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛 𝑥
as 𝐾 ( ) = = . Therefore, to increase the partitioning into the oil phase
𝑙𝑖𝑞 𝑤𝑎𝑡𝑒𝑟 𝑚𝑜𝑙𝑒 𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛 𝑤
with increasing temperature, the K value should also increase with increasing temperature.
However, it may be that after performing laboratory measurements, this dependency is found
to be the reverse (i.e. as the temperature increases, the amount of EtF that partitions into the
oil phase decreases). In this case, the K values should decrease with increasing temperature.
For this example, assume that as the temperature increases the amount of EtF that partitions
into the oil phase will increase. To change this in Builder, go to the menu item Components|K
Values. Click the K Value tables button, change the Component to EtF, and change the
table to be as shown below. Click OK twice when done.

40
SWTT Modeling with STARS
6. Under the Numerical section, double-click Numerical Controls and select the following
options for ISOTHERMAL and TFORM as shown below. Click OK.

7. Go to the Wells & Recurrent section and double-click on Inj well. On the Injected Fluids tab,
enter the Temperature and Pressure as shown below for both the dates (2000-01-01 and
2000-01-01.166). Click Apply and OK.

A semi-analytical model will be used to calculate the pressure drop and heat transfer at the
wellbore (SAMODEL). This is very important given that temperature will be changing from the
surface to a certain point inside of the reservoir. This change in temperature will produce
variations on the K values and also in the tracer profiles. Next, the information related to the
wellbore will be defined.
8. Double-click on Inj well. On the Wellbore tab, check the box for Wellbore model (also for each
wellbore data) and enter the properties in the following table for both the dates (2000-01-01
and 2000-01-01.166).
Wellbore depth 6640 ft
Wellbore length 6640 ft
Casing length 6640 ft
inner tubing 0.125 ft
outer tubing 0.146 ft
inner casing 0.283 ft
outer casing 0.291 ft
hole 0.354 ft
Average geothermal gradient 0.016 F/ft
Surface temperature 59 F

41
SWTT Modeling with STARS
9. Select the check box for HEAD-METHOD in the section below and select GRAV-FRIC-
HLOS (see image below).
10. Click Apply

42
SWTT Modeling with STARS
11. Control-click then select wellbore and HEAD-METHOD under the Inj well event.

12. Right click and select Copy Event Using Filter. Then select Inj under the 1-Wells tab. Over
the 2.-Dates tab select dates 2000-01-01 and 2000-01-01.1599899. Click then
click OK twice. This will copy the constraints for Inj well from one date to the next selected
date.
13. To output the bottom hole temperature data vs. time data, in Builder in the I/O Control section,
double-click Simulation Results Output, then click the Select button for Special
information. On the Select/Modify Special Variables dialog box, find the Special Variable
Name “PHWELL” and change the options to the following values. Click OK when done and
close the model.

14. Open the data set in CEDIT and go to the Well & Recurrent section add the keyword
SAMINFO after the DTWELL keyword.

15. Save the dataset and run the case.

16. Once completed, open “swctt-tutorial-thermal.out” with CEDIT and for “Pressure drop and
heatloss for well Inj”. This option calculates the heat transfer within the well from surface to
the reservoir. Note the fluid temperatures along the injector and the difference in the reported
temperatures and heat-loss at different depths.

43
SWTT Modeling with STARS
17. Temperature profile will be plotted moving from the well until the limit or boundary of your
model.
18. Open “swctt-tutorial-thermal.sr3” in Results.
19. Go to the section Input|Linear Paths and on the right side insert a new linear path that goes
from block 1 1 1 to 490 1 1 (see image below)

20. Go to the section Plots|Profile and add the temperature for the following dates: ,
, , (see image
below)

44
SWTT Modeling with STARS
21. In the upper menu go to Plot Properties and click over the option of Swap Axes
(Note: Change the colors of each line). You can save this template as SWTT_thermal.results
22. . The graph should appear as follows:

23. The above graph depicts the temperature difference across the model at different times. This
is a result of the cold injected fluid going into the hotter reservoir, thereby altering the average
reservoir temperature. However, this model does not consider the pre-flush stage of this test
for calculating temperature effects. If this pre-flush stage is added to this model, the
temperature cooling effect would be more pronounced and more of the reservoir would
become cooled.
24. As an additional exercise bring the results of the model without thermal effect and compare
the profiles.

45
SWTT Modeling with STARS
25. The above plot shows a significant effect of temperature on the results when injecting cold
water during a SWCTT. From this plot, the peaks at which the chemicals arrive are different,
and therefore the calculation of Sor based on the arrival of the peak can lead to a wrong value.
From these results, it can be concluded that using numerical simulation is required when
temperature effects are involved during the test for the appropriate estimation of residual oil
saturation. Therefore, it is recommended that temperature effects should always be included
when attempting to interpret the results of a SWCTT that has cold water injection.

46
SWTT Modeling with STARS