Academic Research International Vol.

5(2) March 2014

_________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
________________________________________________________________________________________________________________________________________________________________________________
____________________________________________________________________________________________________________________________
_____________________________________________________________________________________________________________________________
___________________________________________________

Modelling the Effects of Condensate Banking on High CGR Reservoirs

Cyril Okporiri 1, Koso I. Idigbe2,
Department of Petroleum Engineering, University of Benin,
Benin city, NIGERIA.
1

[email protected], 2 [email protected]

ABSTRACT
This paper presents the results of a study undertaken to evaluate the effect(s) of
condensate banking on retrograde gas reservoirs particularly, high condensate gas
ratio (CGR) reservoirs. The effects of composition, fluid properties, liquid dropout
rate, and the gas-oil relative permeability on gas productivity, were investigated
using a scaling number, the Capillary Number, Nc. Results from the case study show
that the impact of condensate banking would be severe for these reservoirs if the
physical processes expected in these reservoirs are not correctly modelled. Also, the
impact is not likely to be serious if the reservoir conditions are favourable as are
prevalent in the Niger Delta basin of Nigeria.
Keywords: Condensate banking, Retrograde Gas Reservoirs, Liquid dropout rate,
Gas Productivity, Niger Delta basin

INTRODUCTION
Retrograde natural gas reservoirs, are populated with single phase fluid(s) natural gas, with
temperatures between the critical temperature and the cricodentherm (1, 2). The significance of
these reservoirs have grown significantly over the years as drilling targets hit deeper depths,
subsequently encountering very high temperatures and pressures which are necessary for
their presence. The cost and risk to develop these reservoirs under severe conditions of
pressure and temperature highlight the need to be able to confidently predict the recovery of
gas and liquid drop-outs from these reservoirs (2, 4). Retrograde gases exhibit dew points and
initially release increasing volumes of liquid hydrocarbons condensates, into the pore space
as the reservoir pressure is reduced below the dew point value, on production. This is the
retrograde condensation of hydro-carbon liquids from the natural gas. The amount of
condensates deposited as pressure declines depends essentially on the composition of the
hydrocarbon fluids, particularly on the amount of heavy ends in the condensate, giving rise to
two phase reservoir systems (5).
The retrograde gas fluids, single phase, are usually very rich in liquid contents such that with
pressures at/and below the dew point, large volumes of liquid are deposited. In general, rich
gas condensate reservoirs, with two phase fluids, occur close to the critical point on the
phase diagram and are sometimes referred to as near-critical Gas Condensate Reservoirs and
these confers on them certain properties which include(6.
1.

High liquid saturation in the pore spaces and around the wellbore: The condensate is
considered mobile in the pore spaces, when its saturation is greater than the critical
value.

2.

Near-Miscible conditions: due to this closeness to critical point, the physical

properties of the gas and condensate phases are very similar and they are therefore
almost miscible with each other.

3.

Low interfacial tension between the gas and condensate phases occurs for systems
particularly close to the critical point. In the development of these fluid systems,

Copyright 2014 SAVAP International

www.savap.org.pk

41

ISSN: 2223-9944, eISSN: 2223-9553

www.journals.savap.org.pk

Academic Research International Vol. 5(2) March 2014

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________
_____________________________________________________________________________________________________________________________
______________________________________________________________________________________________________ _________________________________________________________________________

these properties must be clearly modelled and understood to optimise production.

Figure 1.0 shows the liquid dropout from a rich gas condensate reservoir. The figure
clearly shows the rapid build-up of liquid as pressure declines.
Due to the large volumes of condensates deposited for small drops in pressure, and most
often with saturation values lower than the critical, condensate banking phenomenon will
occur in the reservoir resulting in gas productivity decline.
CONDENSATE BANKING
Condensate banking is a phenomenon associated with retrograde natural gas fluids. During
the production of these fluids with the reservoir pressure declining below dew point values,
hydrocarbon liquids condensates, drop out. The saturation may rapidly build up for high
CGR reservoirs quickly exceeding critical saturation for mobility and the condensates flow
towards the producing wells where a bank of condensates hydrocarbon liquids, is
generated around the wells. The condensate bank can grow as both reservoir and well flowing
bottom hole pressures decline further. This can potentially lead to impairment on well
deliverability of gas and condensate. This can be more severe for reservoirs with high initial
CGR and low permeability.
Field examples show that the effect of condensate banking could be severe leading to
productivity loss factor of between 2 to 6. An example is the Arun Field in North Sumatra,
Indonesia, which experienced a decline in productivity by a factor of 2 when the well bottom
hole flowing pressure fell below the dew point(3). The generation of a condensate bank
around producing wells and the potential impact on productivity is however dependent on a
number of reservoir, fluid and flow parameters and these parameters determine if there will
be productivity impairment or not. Therefore in the treatment of condensate banking, there is
a need to look at the type of condensate systems been described as it would be wrong to
appropriate the same behaviour of condensate banking on all types of retrograde gas
condensate systems, due to the dependence on fluid and flow properties.
Properties of the gas condensate fluids play a key role in the process of condensate banking
and mobility. For near-critical fluids, the expected trend will be a similarity between the gas
and liquid phases. For rich gas, the large amount of condensates deposited exceeds the critical
condensate saturation necessary for mobility and therefore the condensate is mobile. Also the
interfacial tension (IFT) between the condensate and gas phases will initially be very small (6).
Under these conditions, the displacement process becomes more miscible-like and straightline relative permeability curves could be used to describe the recovery process. Further
depletion below the dew point causes the phases to become more distinct and the IFT
becomes larger and immiscible-like and capillary-dominated flow begins to dominate again.
Therefore the IFT is a key parameter controlling the flow of gas and condensate for
conditions close to the critical point (6).
In general, for natural gas reservoirs; near the wellbore, viscous forces and flow velocities
can become very high. At these high velocities, two competing phenomena occur which
control the productivity of the wells
1.

Increase in relative permeability due to increase in velocity. This effect is sometimes

called VISCOUS STRIPPING or POSITIVE COUPLING.

2.

Inertial (non-Darcy) flow effects, which reduce the effective permeability at high
velocity. This is generally referred to in modelling as NON-DARCY EFFECT.

These effects tend to act in opposite directions and both need to be accounted for in flow
calculations to accurately represent well productivity.
Copyright 2014 SAVAP International

www.savap.org.pk

42

ISSN: 2223-9944, eISSN: 2223-9553

www.journals.savap.org.pk

Academic Research International Vol. 5(2) March 2014

_________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
________________________________________________________________________________________________________________________________________________________________________________
____________________________________________________________________________________________________________________________
_____________________________________________________________________________________________________________________________
___________________________________________________

The first effect of viscous stripping needs to be built into the relative permeability model. The
magnitude of the effect depends on the balance between viscous (stripping) forces and
capillary (trapping) forces. This is described below in terms of the capillary number, a scaling
number.
Productivity above the dew point pressure is controlled principally by the reservoir
permeability and thickness, and the viscosity of the retrograde gas. Below the dew point, the
degree of productivity reduction will be controlled by the critical condensate saturation and
the shapes of the gas and condensate relative permeability curves.
Capillary Number The Scaling Parameter
The capillary number is a dimensionless parameter defined as the ratio of the viscous force to
the capillary force, and is an indication of the relative strength of viscous stripping to
capillary trapping. Capillary number, Nc, is a function of fluid viscosity (), velocity (V), and
the interfacial tension (), and is expressed mathematically as:
Nc =

K dP
......................................................................................1
dx

The reservoir and fluid properties, particularly the interfacial tension, determine whether
miscible or immiscible conditions will prevail. Further, the flow rate determines the
magnitude of the viscous force and the ability to overcome capillary trapping forces. The
combined effect needs to be captured in the relative permeability and flow model. It has been
proposed by several authors that the Capillary Number, Nc, be used as a scaling number to
assign the correct relative permeability values throughout a model (7, 8, 9, 10).
It is generally accepted that for capillary numbers less than 10 -5 (commonly referred to as the
threshold value), the residual oil saturation is almost constant and the displacement process
can be adequately represented by traditional immiscible Corey type relative permeability
curves where capillary forces tend to dominate. For capillary numbers greater than 10 -5,
viscous forces or miscible effect (low IFT) begin to play a more dominate role. The residual
oil saturation decreases with increasing capillary number and ultimately, miscible relative
permeability curves should apply. It is not well established at what upper limit value of
capillary number, fully miscible (straight line) relative permeability curves should be used.
Sensitivity studies were made to test the impact of the upper limit. This is described in the
results section below.
STUDY METHODOLOGY
The focus of this case study is to investigate the near wellbore effects controlling well
productivity in rich gas condensate systems. The reservoir is a sandstone reservoir in the
Niger Delta region of Nigeria, with layers of shale in between its various sand beds. For the
study, a single well radial dynamic model was chosen to investigate the condensate banking
and dropout effects. The model is described as follows:

A 2-dimensional (r-z coordinate) system, with

A 30 block radial grid with increasing grid spacing to capture near well pressure
sink,

A single layer to nullify gravitational effects,

No aquifer, and

A fully completed well.

Copyright 2014 SAVAP International

www.savap.org.pk

43

ISSN: 2223-9944, eISSN: 2223-9553

www.journals.savap.org.pk

Academic Research International Vol. 5(2) March 2014

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________
_____________________________________________________________________________________________________________________________
______________________________________________________________________________________________________ _________________________________________________________________________

The model was populated with uniform reservoir properties. These properties are given in
Table 1. Monitors or trackers were created in the model to track the behaviour of gas and
condensate saturation, interfacial tension and the capillary number in different grid blocks;
the distance of these grid blocks from the wellbore being 1ft, 5ft, 20ft and 100ft.
The reservoir is at a depth of about 9800 Ftss. The original pressure of the reservoir is 6101
psia. As stated earlier, it is a stratified reservoir with a very rich retrograde gas having a high
level of condensate fluids, and it is compositionally graded with the top of the reservoir
known to have a leaner gas condensate fluid.
Retrograde gas, the fluid used in this investigation, has an initial average CGR of
240bbl/mmscf, determined from PVT analysis. The dew point pressure of the fluid is 5377
psi from PVT analysis. The PVT data used in this model were based on a bottom-hole sample
collected above the dew point pressure of the fluid. The Peng-Robinson 78 equation-of-state
was used to achieve an acceptable simulation of the fluid behaviour. The simulation of the
PVT data resulted in a 10-component model shown in Table 2. Compositional grading was
also introduced into the PVT model. The CCE experiment shows a maximum liquid dropout
of 44.0%. The phase envelope of the fluid is shown in Figure 2.
Two relative permeability curves were used for this study. One is the Corey function relative
permeability and the other is the parameter dependent relative permeability; the parameter
being the capillary number, Nc. The data/parameters for the two models are shown in Table 3.
RESULTS OF STUDY
The monitors in the radial model were used to determine the impact of condensate banking in
the reservoir. The model was constrained to produce at a constant initial rate of 70 mmscf of
gas per day until the rate could no longer be sustained. The model was run initially with an
absolute permeability of 1000 md and it made use of lift tables in predicting the performance
of the well.
The gas and condensate saturation for the different grid blocks were plotted against time and
this is shown in Figure 3. The plot shows that the condensate saturation quickly builds up to
almost 50% as pressures deplete and gas saturation declines thereby exceeding the critical
condensate saturation at the first dropout.
The relative permeability behaviour of the model for the first grid block plotted as a function
of gas saturation shows both the miscible and the immiscible relative permeability curves. In
Figure 4, it could be seen that during production just below the dew point, the model made
use of the miscible relative permeability to define the relative permeability behaviour of the
fluid. However as production continues, there is a transition from the miscible relative
permeability curve to the Corey function relative permeability plot due to production
termination as a result of lift die out. Also, shown, is the relative permeability behaviour of
the fluids in the first grid block in Figure 5.
This effect came as a result of the behaviour of the capillary number which can be attributed
to the interplay of the interfacial tension behaviour and the effect of viscous stripping. The
capillary number for the same four monitored grid blocks is shown in Figure 6. The grid
block closest to the well is predicted to have the highest capillary number due to a
combination of high flow rate and low Interfacial Tension. Capillary number is plotted on a
logarithmic scale, emphasising the huge variation in the value as the simulation proceeds.
The capillary number for all monitored grid blocks was initially predicted to be several orders
of magnitude larger than the value at which capillary forces dominate. (Capillary forces are
expected to dominate for Nc values up to 10-5). For the grid block closest to the well, Nc is
Copyright 2014 SAVAP International

www.savap.org.pk

44

ISSN: 2223-9944, eISSN: 2223-9553

www.journals.savap.org.pk

Academic Research International Vol. 5(2) March 2014

_________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
________________________________________________________________________________________________________________________________________________________________________________
____________________________________________________________________________________________________________________________
_____________________________________________________________________________________________________________________________
___________________________________________________

above 10-5 throughout the simulation run, indicating that the Corey function relative
permeability never applies to this grid block. Grid blocks further out in the reservoir fall to
below 10 -5 towards the end of the simulation period, at which point Corey relative
permeability functions are applicable.
Very low interfacial tension values are predicted immediately below the dew point pressure.
Interfacial tensions gradually increase with time as the pressure continues to decline and the
phases become less miscible as can be seen in Figure 7. It is noted that the plot is on a
logarithmic scale emphasising the large change in values.
In other to be able to verify the effects of this behaviour on productivity, the production rate
of both the gas and the condensate fluids at the surface equipments, were plotted as a function
of time. Figure 8 shows that just below the dew point, there was no decline in production due
to condensate dropout or banking.
It is generally accepted that the threshold capillary number below which capillary forces and
immiscible behaviour dominates, is 10 -5. The flow characteristics at these low capillary
numbers are well described by the standard Corey type relative permeability functions.
However, as the capillary number increases above 10-5, the displacement and recovery
process becomes increasingly miscible-like. At a threshold value of the capillary number, the
displacement is fully miscible. There is no industry accepted threshold value for miscibility,
although, it is generally quoted to be in the range10 -3 to 10-2. The reference case model
described above assumes 10-2.
Sensitivities to threshold values of 10 -2, 10-1 and 10 -0 are shown in Figures 9 and 10.
Productivity reduces as the threshold capillary number increases. For threshold value of 1.0,
the impact of low IFT and high flow rates is suppressed and the recovery process is
dominated by standard Corey function.
Finally, in other to investigate the effect of absolute permeability of the reservoir on
condensate banking and productivity decline, sensitivity simulations to lower permeability
values were performed and are shown in Figure 11. The case for 500 md is very similar to the
reference case of 1000 md. However productivity drops dramatically for the case with 150
md. Condensate banking and gas productivity impairment effects are more evident at lower
permeability values. For fixed gas production rate at 70 mmscf/d, the drawdown increases as
the permeability reduces. At some point, the minimum bottom-hole pressure is reached and
the production falls off plateau. For low permeability reservoirs, the effects of banking can be
more severe as in the case of the Arun field (3), which has permeability on the range 0.1 md to
1 md.
CONCLUSIONS
From this study, we infer the following conclusions:
1.
2.

3.
4.

Near-critical retrograde gas systems are characterised by high CGR and high initial
liquid dropout saturation.
There is a high rapid liquid saturation build-up around the wellbore, with low
interfacial tension and potentially high capillary numbers, Nc, exhibiting nearmiscible properties displacement.
The capillary number, Nc, is highest closer to the wellbore, indicative of viscous
stripping.
Condensate banking effects are controlled to a large extent by relative permeability
of gas and condensate, and for near-critical gas condensate systems, the relative
permeability is a function of interfacial tension and flow rate of gas.

Copyright 2014 SAVAP International

www.savap.org.pk

45

ISSN: 2223-9944, eISSN: 2223-9553

www.journals.savap.org.pk

Academic Research International Vol. 5(2) March 2014

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________
_____________________________________________________________________________________________________________________________
______________________________________________________________________________________________________ _________________________________________________________________________

5.
6.

Failure to correctly model the physical processes expected in a reservoirs could lead
to wrong prediction on the impact of condensate banking.
Simulation of the case reservoir from the Niger Delta region of Nigeria shows that
reservoirs with higher values of permeability are unlikely to be affected by
condensate banking.

REFERENCES
[1]

McCain, W. D. Jr. (1990). The Properties of Petroleum Fluids (Second Edition). Tulsa,
Oklahoma: PennWell Publishing Company, ISBN 0-87814-335-1

[2]

Saeidi, A. & Handy, L. L. (1974). Flow and Phase Behaviour of Gas Condensate and
Volatile Oils in Porous Media. Paper SPE 4891.

[3]

Afidick D., Kaczoroswki, N. J. & Srinivas B. (1994). Production Performance of a

Retrograde Gas Reservoir: A Case Study of the Arun Field Paper SPE 28749.

[4]

Kamath, J. (2007). Deliverability of Gas condensate Reservoirs Field Experiences

and Prediction Techniques Paper SPE 103433 Distinguished Author Series.

[5]

Wheaton, R. J., Zhang, H. R. & Reading, B. (2000). Condensate Banking Dynamics in

Gas Condensate Fields: Compositional Changes and Condensate Accumulation
around Production Wells Paper SPE 62930.

[6]

Barnum, R. S., Brinkman, F. P., Richardson, T. W. & Spillette, A. G. (1995). Gas

Condensate Reservoir Behaviour: Productivity and Recovery Reduction Due to
Condensation Paper SPE 30767.

[7]

Blom, S. M. P. & Hagoort, J. (1998). How to Include the Capillary Number in Gas
Condensate Relative Permeability Functions Paper SPE 49268.

[8]

Curtis, H. W., Pera, Q. F. & Aud S. (1999). Gas Condensate Relative Permeability for
Well Calculations Paper SPE 56476.

[9]

Henderson, G. D., Danesh, A., Tehrani, D. H. & Al-Kharusi, B. (2000). The Relative
Significance of Positive Coupling and Inertial Effects on Gas Condensate Relative
Permeabilities at High Velocity Paper SPE 62933.

[10] Robert, M., Andrew, C.& Mike, S. Measurements and Simulation of Inertial and High
Capillary Number Flow Phenomena in Gas-Condensate Relative Permeability Paper
SPE 62932.

Copyright 2014 SAVAP International

www.savap.org.pk

46

ISSN: 2223-9944, eISSN: 2223-9553

www.journals.savap.org.pk

Academic Research International Vol. 5(2) March 2014

_________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
________________________________________________________________________________________________________________________________________________________________________________
____________________________________________________________________________________________________________________________
_____________________________________________________________________________________________________________________________
___________________________________________________

Appendix
Nomenclature
bbls: barrels
CGR: Condensate Gas Ratio
Ftss: Feet sub-surface
Mmscf: million standard cubic feet
PVT: Pressure, Volume, Temperature
Table 1. Properties of the Single Well Radial Model
Parameter

Value

Grid size (NX x NY x NZ)

30 x 1 x 1

External radius, re

6350 ft

Well-bore radius, rw

0.75 ft

Dip

0 degrees

Reservoir Angle

PI radians (180)

Thickness, h

100 ft

NTG

1.0

Phi (Porosity)

0.22

Perm

1000 md

GIIP

3.0 Bscf

CIIP

80.0 MMstb

Table 2. 10-component PVT data showing composition at a depth of 9630 Ftss

Composition

Mole %

N2 - C1

0.69384

CO2 C2

0.07397

C3 C4

0.08448

C5 C6

0.03236

C7 C11

0.06974

C12 C16

0.02553

C17 C19

0.00755

C20 C21

0.00494

C22 C26

0.00443

C27 C45

0.00316

Copyright 2014 SAVAP International

www.savap.org.pk

47

ISSN: 2223-9944, eISSN: 2223-9553

www.journals.savap.org.pk

Academic Research International Vol. 5(2) March 2014

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________
___________________________________________________________________________________________________________________________________________________________________________________________________________________________________ _________________________________________________________________________

Table 3. Relative Permeability Data/Parameters

Parameters

Corey model
Immiscible

Straight line
Miscible

Scw

0.0431

0.0431

Krw

0.33

0.33

Nw

4.11

4.11

Sorw

0.21

0.00

Sorg

0.10

0.00

Krow

0.75

1.00

Nog

3.47

1.00

Now

2.11

2.11

Sgc

0.10

0.00

Krg

0.90

1.00

Ng

2.11

1.00

Figure 1. Liquid Dropout Curve for a near-Critical Gas Condensate System

Figure 2. Phase Diagram of the Fluid used in the Model (Reservoir Temp.: 224.2oF)
Copyright 2014 SAVAP International

ISSN: 2223-9944, eISSN: 2223-9553

Academic Research International Vol. 5(2) March 2014

_________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
________________________________________________________________________________________________________________________________________________________________________________
____________________________________________________________________________________________________________________________
_____________________________________________________________________________________________________________________________
___________________________________________________

Figure 3. Condensate Saturation for the Radial Model from Selected Grid Blocks

Figure 4. Relative Permeability Plots showing the Performance of the Reservoir in terms of
Miscibility

Figure 5. Relative Permeability to Gas and Condensate Saturation Plot against Time
Copyright 2014 SAVAP International

www.savap.org.pk

49

ISSN: 2223-9944, eISSN: 2223-9553

www.journals.savap.org.pk

Academic Research International Vol. 5(2) March 2014

___________________________________________________________________________________________________________________________________________________________________________________________________________________________________
_____________________________________________________________________________________________________________________________
______________________________________________________________________________________________________ _________________________________________________________________________

Figure 6. Capillary Number Plots for Selected Grid Blocks (Simulator generated)

Figure 7. Prediction of Interfacial Tension Behaviour for Model Fluid

Figure 8. Gas and Condensate Production Plot of the Model

Copyright 2014 SAVAP International

www.savap.org.pk

50

ISSN: 2223-9944, eISSN: 2223-9553

www.journals.savap.org.pk

Academic Research International Vol. 5(2) March 2014

_________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
________________________________________________________________________________________________________________________________________________________________________________
____________________________________________________________________________________________________________________________
_____________________________________________________________________________________________________________________________
___________________________________________________

Figure 9. Dependency of Gas and Condensate Production Rates on the threshold Capillary Number
necessary for Miscibility

Figure 10. Relative Permeability to Gas Behaviour due to Capillary Number for Miscibility Changes

Figure 11. Gas and Condensate Production Plots for different Values of Permeability
Copyright 2014 SAVAP International

www.savap.org.pk

51

ISSN: 2223-9944, eISSN: 2223-9553

www.journals.savap.org.pk