SPE 94074

The Impact of Pertinent Parameters on the Design of Hydraulic Fracturing in Gas

Condensate Reservoirs
G.A. Carvajal, A. Danesh, M. Jamiolahmady, and M. Sohrabi, Petroleum Engineering Inst., Heriot-Watt U.

Copyright 2005, Society of Petroleum Engineers

flow have been widely studied [1-3]. However, the positive
This paper was prepared for presentation at the SPE Europec/EAGE Annual Conference held coupling effect, that is, improvement of gas and condensate
in Madrid, Spain, 13-16 June 2005.
relative permeability by increasing velocity, in fractured wells
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
has not been yet fully analyzed. The positive coupling, as a
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to flow mechanism, is certainly known in flow around un-
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the SPE, their officers, or members. Electronic reproduction, distribution, or storage fractured wells [4-7]. But the new concepts related to the flow
of any part of this paper for commercial purposes without the written consent of the Society of rate and capillary number have not yet been incorporated into
Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract
of not more than 300 words; illustrations may not be copied. The abstract must contain gas condensate deliverability models for fractured wells.
conspicuous acknowledgment of where and by whom the paper was presented. Write
Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
Reports by Wang et al. [8] and Indriatti et al. [9] are the
only available literatures that explore the effect of capillary
Abstract
number in their study. They adjusted the relative
Fracturing is the most common well stimulation technique in
permeabilities with the capillary number for the effect of
gas condensate reservoirs. Flow of black oil and dry gas in
interfacial tension.
induced fracture systems has been widely studied by
numerical simulation, core-flood and well tests. However, the
The main objective of this study is to evaluate the impact
flow behavior in fractures surrounding a gas condensate well,
of positive coupling and inertial effects on design of fracturing
which could be significantly different to that of conventional
and prediction of fractured well performance. We shall
systems, has not been fully understood. Most of the available
analyze the significance and impact of several parameters,
literature has quantified the negative effect of condensate
describing rock, fracture and fluid properties that can
banking and inertial (non-Darcy) flow on well productivity,
influence the process. The ECLIPSE [10] reservoir simulator
giving a pessimistic view of the gas rate in these systems. The
was used to develop realistic models of flow in fractured
above studies have neglected improvement in the gas and
systems and to evaluate the effect of pertinent parameters on
condensate relative permeability by increasing rate due to
fracture performance.
positive coupling of the two phases.
Fractured Well Modeling
In this work, measured values of permeability, single-
Generally, vertical fractures are simulated using Cartesian
phase inertial factor and relative permeability of a propped
models rather than radial ones. It is basically due to its flow
fracture were used to evaluate the effect of pertinent
regime, which is considered to be mainly linear from the
parameters on fracture performance more realistically. A
reservoir into the fracture face, and through the fracture
reservoir simulator was used to develop realistic models of
towards the wellbore. However flow can be considered
flow in fractured systems. A comprehensive sensitivity
pseudo-radial far way from the fracture. The significance of
analysis on the impact of fracture characteristics on well
the latter flow regime depends upon how much the fracture
productivity for different fluids and rock properties was
improves the productivity index compared to an un-fractured
conducted. The results demonstrated that not only inertia was
case.
important but also positive coupling strongly affected the flow
performance. In particular for a tight formation, whereby Furthermore, geometry representation of radial model of a
fracturing is most effective, positive coupling in the matrix fracture in the simulator is unrealistic. That is, it exhibits an
can overcome negative inertia in the fracture region. The increase in thefracture width, away from the well. The width
results also reveal the important contribution of the fracture has a strong effect on velocity as it is linked directly to the
width, as wide fractures reduce the inertial effect significantly. flow cross sectional area. This was confirmed in a sensitivity
Furthermore, we will show that increasing the fracture length study whereby the effect of inertia was found to be
does not necessarily lead to higher gas well productivity as unrealistically minimal for a fractured well model in radial
there are optimum values for any given set of parameters. system due to inaccurated calculations of velocity. Although
shape factors were developed and included in simulation to
Introduction alleviate the above problem, it was concluded that a Cartesian
The mechanisms that affect fracture performance in gas coordinate system could capture flow in a fractured well
condensate systems such as condensate banking and inertial model more realistically.
2 SPE 94074
Grid Optimization.
A grid in fractured well simulation should describe reasonably
the changes in pressure, fluid saturation and velocity
especially along the fracture pathway, around the fracture tip,
the wellbore area and the fracture edge whereby these changes
are more significant. A comprehensive sensitivity study was
conduced to optimize the grid system. In this exercise it was
noticed that cell distribution was more important than the
number cells. Based on this finding we selected very small
cells near the fracture in both directions ‘y and x’,
perpendicular to the fracture face, growing by a geometry
factor of 2, with further refinement near the fracture tip,
Figure 1. The optimized grid was found to be 30x30. The
reservoir was modelled by one vertical cell, neglecting
gravitational segregation.
Reservoir properties.
The reservoir is 5,300 ft in x and y-directions and producing at
an initial reservoir pressure of 6,800 psia with a controlled
bottom hole pressure of 4,400 psia at the wellbore radius of
0.23 ft. Matrix properties are those measured on a sandstone
reservoir rock with a permeability value of 0.18 mD and
porosity of 17%, which gives a reservoir pore volume of
260MMrb. The experimentally measured single-phase
inertial factor for this core was 1.06E12 m-1.
To evaluate the impact of positive coupling and/or negative
inertia on flow performance, the Velocity Dependent Relative
Permeability (VDRP) option of ECLIPSE [10] was used. This
option allows the use of rate dependent relative permeability
correlations, which has been developed by this laboratory in
the past. The required base relative permeability and rate
dependent core exponents for this rock sample were
determined experimentally.
Fracture properties.
We considered a fracture with a propant porosity of 30% and
permeability of 15 D. The fracture width and length were
varied between 1 to 60 mm and 150 to 600 ft, respectively.
The required base relative permeability and rate dependent
core exponents were those determined experimentally for a
porous medium packed with proppant. The experimentally
measured single-phase inertial factor for this porous medium
was 3.51E5 m-1.
Fluid properties.
Three gas condensate model fluids with 16, 20 and 30% v/v,
maximum liquid dropout (MLDO) in the constant composition
expansion test at the reservoir temperature of 250oF were used
in this study. The corresponding dew point pressure values
were 5180, 5280 and 5405 psia, respectively. For the case of
single-phase gas a mixture with dew-point pressure of 3192
psia, above the controlled wellbore pressure of 4400 psia was
used.
Results
Two principal regions, i.e. the fracture pathway and the matrix
control the flow towards the wellbore. In a series of
sensitivity exercises we initially evaluated the importance of
velocity effects (i.e., negative inertia and positive coupling) in
each region.
Fracture length, particularly in tight formation, is
considered to be the most important parameter [1,14]. We
evaluated the importance of this parameter as well as that of
fracture width. In this exercise the significance of inertia and
its competition with coupling when MLDO is increased from
0 (single-phase) up to 30% was also evaluated.
Rate Effect by Region
A large of number of cases, using different fluids and
accounting for different rate effects was simulated. Examples
of some cases, which refelect the general flow behaviour are
as follows.
Effect of coupling by region.
Figure 2 shows the gas production rate versus time for four
different cases. In three of these cases coupling was
considered (a) only in the matrix, (b) only in the fracture and
(c) both in the matrix and the fracture. Inertia was ignored for
these three cases. The figure shows that the effect of coupling
is mainly significant in the matrix region. That is, there is
hardly any difference between the production curves of case
(a) and case (c). The realistic production curve, case 4, with
the effect of inertia and coupling both accounted in all regions,
highlights the negative effect of inertia on productivity.
Effect of Inertia by Region.
Figure 3 shows the gas production rate versus time for four
different cases. In all these cases coupling was ignored and
inertia was considered (a) only in the matrix, (b) only in the
fracture, (c) both in the matrix and the fracture and in (d)
inertia was deactivated both in the matrix and the fracture.
The results suggest that inertia in the matrix is negligible
whereas it is quite significant in the fracture.
Effects of coupling and Inertia by Region.
Figure 4 shows the gas production rate versus time for three
different cases. In this exercise the combined effect of
coupling and inertia was considered (a) only in the matrix, (b)
only in the fracture and (c) both in the matrix and the fracture.
The results confirm the previous findings that a realistic model
should account for the effect of inertia in the fracture and that
of coupling in the matrix.
Effect of Fracture Geometry, Single-phase
Figure 5 shows the gas production rate versus time for the
cases of un-fractured well and with and without inertia at two
different fracture lengths. It is noticed that increasing the
length increases productivity but this increase is more
pronounced when inertia is not activated. This also suggests
that the effect of inertia is more significant at longer fracture
lengths. Therefore, it can be concluded that there is an
optimum fracture length for any given set of parameters and
increasing the length does not necessarily lead to significant
increase in well productivity
Figure 6 shows the gas production rate versus time for the
cases of with and without inertia at two different fracture
SPE 94074
widths. It is noticed that increasing the fracture width
improves the productivity, which is more pronounced when
inertia is activated.
Effect of Fracture Geometry, Two-phase
Figure 7 shows the gas production rate versus time for the
cases of with and without velocity effects at two different
fracture lengths. It is noticed that similarly to the single-phase
results increasing the length increases the productivity, which
is more pronounced when velocity effects are not activated.
Since the production rate of the effected cases is lower than
those without the effect, we can conclude that here the inertial
effect is dominant to the positive coupling effect.
Furthermore, similarly to the results of single-phase flow,
Figure 5, inertia is more significant in longer fractures.
Figure 8 shows the gas production rate versus time for
two sets of data, all including velocity effects. Set-1 and Set-2
correspond to variations of production as a result of a change
in the fracture length at a fracture width of 5 and 15 mm,
respectively. For a fracture width of 5mm the effect of length
is not very significant. However, at a fracture width of 15mm
gas productivity increases significantly by increasing the
fracture length. The results indicate that increasing the fracture
length does not significantly improve productivity in narrow
fractures as the inertial effect at such conditions reduces the
improvement due to additional length.
Effect of Fluid Properties
Figure 9 shows the gas production rate versus time for three
sets of data. Sets 1, 2 and 3 correspond to data for single-
phase, 16 and 20% MLDO, respectively, with and without
velocity effects. At single-phase flow conditions the velocity
effect is only due to inertia, which results in a reduction in
productivity. The same reduction in productivity due to the
velocity effects is observed at two-phase flow conditions with
MLDO of 16%. However, as the fluid becomes richer the
cases accounting for the rate effect overcome those without,
signifying that coupling becomes dominant.
Others runs were conducted to evaluate the effect of gas
richness in narrower fractures, Figure 10. The production rate
for the case without the velocity effects is more than that with,
i.e., inertia is dominant when using the fluid with
MLDO=20%. However, the production rate for the case with
the velocity effects are higher indicating that coupling is more
dominant using the fluid with MLDO=30%. These results
demonstrate that coupling is still important even for narrower
fracture cases in rich gas condensate systems.
Conclusions
Sensitivities were conducted to evaluate the effect of flow rate
on fracture performance for different fluid and fracture
characteristics in a low permeability gas condensate reservoir.
The results indicated that:
1) The negative inertial impact is mainly limited to the
fracture region, whereas the positive coupling contribution
3
comes from the matrix region. Therefore, it was demonstrated
that an overestimation in gas profile is obtained if inertia is
ignored in the fracture region; and underestimation if coupling
is not considered in the matrix.
2) The fracture width, which which controls inertia is very
important in gas condensate systems.
3) The improvement in productivity due to increase in fracture
length becomes less significant in narrow fractures due to
increases of inertia.
4) In a rich gas condensate the coupling effect my overcome
the negative inertial effect even in narrower fractures.
Acknowledgements
The above study has been sponsored by: The UK Department
of Trade and Industry, BP Exploration Operating Company
Ltd, Gaz de France, Marathon Oil UK, Statoil A.S.A. and
Total Exploration UK plc, which is gratefully acknowledged.
Schlumerger-Geoquest is thanked for the use of ECLIPSE
simulator.
References
1. Settari, A., Bachman, R.C., Hovem K, and Paulsen, S.G.:
“Productivity of Fractured Gas Condensate Wells-A Case
Study of the Smorbukk Field,” SPE 35604, SPE Reservoir
Engineering Journal (November 1996) 11.
2. Sognesand, S.: “Long-Term Testing of Vertically Fractured
Gas Condensate Wells,” SPE 21704, presented at 1991 SPE
Production Operations Symposium, Oklahoma City,
Oklahoma, 7-9 April.
3. Carlson, M.R., and Myer, J.W.G.: “The Effects of
Retrograde Liquid Condensation on Single Well Productivity
Determined Via Direct (Compositional) Modelling of a
Hydraulic Fracture in a Low- Permeability Reservoir,” SPE
29561, presented at 1995 Low Permeability Reservoirs
Symposium, Denver, Colorado, 19-22 March.
4. Danesh, A., Khazam, M., Henderson, G.M., Therani, D.H.
and Peden, J.M.: “Gas Condensate Recovery Studies,” DTI
Improve Oil Recovery and Research Dissemination Seminar,
London, June 1994.
5. Henderson, GD, Danesh, A, Tehrani, DH, and Peden, J.M:
“The Effect of Velocity and Interfacial Tension on the
Relative Permeability of Gas Condensate Fluids in the
Wellbore Region,” Proc. of the 8th European Symposium on
Improved Oil Recovery, Austria, May 1995, pp. 201-208.
6. Jamiolahmady, M., Danesh, A., Tehrani, D. H. and Duncan,
D. B.: “A Mechanistic Model of Gas-Condensate Flow in
Pores,” Transport in Porous Media, 2000, 41 (1), pp. 17-46.
7. Jamiolahmady, M., Danesh, A., Tehrani, D. H. and Duncan,
D. B.: “Positive Effect of Flow Velocity on Gas-Condensate
Relative Permeability: Network Modelling and Comparison
with Experimental Results,” Transport in Porous Media, 2003,
52 (2), pp. 159-183.
8. Wang, X., Indriati, S., Valko, P. and Economides, M.:
“Production Impairment and Purpose-Built Design of
Hydraulic Fractures in Gas Condensate Reservoirs,” SPE
4 SPE 94074

64749, presented at 2000 International Oil and Gas

2100
Conference and Exhibition in China, 7-10 November. No Inertia in all regions
2000
9. Indriati, S., Wang, X. and Economides, M.: “Adjustment of
1900 Inertia only in matrix
Hydraulic Fracture Design in Gas Condensate Wells,” SPE
73751, presented at 2002 International Symposium and 1800 Inertia all regions

Qg /MSCFD
Exhibition on Formation Damage Control, Lafayette, 1700 Inertia Only in fract
Louisiana, 20-21 February. 1600
10. ECLIPSE Reference Manuals, version 2002A, 2002, 1500
Technical Description, Schlumberger. 1400
1300
1200
1100
wellbore 0 2 4 6 8 10
tip x-direction Æ
Time /Years
Fracture
Figure 3: Gas production rate versus time for cases isolating
inertia by region all without coupling, fracture width=5 mm,
Refinement fracture length=150 ft and maximum liquid drop-out=20%.
in fract.tip & wellbore
2100

2000 Inertia & Coupling in everywhere

Matrix
1900 Inertia & Coupling only in Fract
Qg /MSCFD 1800
Inertia & Coupling only in Matrix
y 1700

1600

Figure 1: Grid arrangement with refinement near the fracture tip 1500
and wellbore. Note, the increase of cell size with a geometry
1400
factor of 2 in both directions.
1300

1200
0 2 4 6 8 10
Time /Years
2100
Coupling only in MATRIX
2000 Coupling only in FRACTURE Figure 4: Gas production rate versus time for cases isolating
Coupling only in all regions inertia and coupling by region, fracture width=5 mm, fracture
1900
Coupling & Inertia in all regions length=150 ft and maximum liquid drop-out=20%.
1800
Qg /MSCFD

1700

1600 Fractured 600 ft

1500
wo/ effect

1400

1300

1200 Fractured
600 ft w/effect Fractured 150 ft
0 2 4 6 8 10
wo/ effect
Time /Years
Figure 2: Gas production rate versus time for cases isolating Fractured
coupling by region all without inertia, fracture width=5 mm,
150 ft w/effect
fracture length=150 ft and maximum liquid drop-out=20%.
a) Un-fractured

Figure 5: Gas rate versus time for a fractured well at single-phase

flow conditions, with and without the effect of inertia and different
fracture lengths, fracture width=5 mm.
SPE 94074 5

15 mm
wo/effects
Set 1:Single-phase condition
15 mm
w/effects * effects

effects
Set 2:
At 16% MLDO
1 mm
wo/effects
Set 2:
At 20% MLDO
1 mm Coupling overcomes
w/effects Inertia * Crossover: Coupling
becomes important

Figure 6: Gas production rate versus time for a fractured well at Figure 9: Gas production rate versus time for a fractured well with
single-phase flow conditions, with and without the effect of inertia and without velocity effects at single-phase and two-phase flow
and different fracture widths, fracture length=150 ft. conditions with two different maximum liquid drop-out values of
16% and 20%, fracture width=15 mm, fracture length=150 ft.

600 ft wo /effect

5 mm w/effect at 20% MLDO

150 ft wo/ effect
Set 1:
5 mm wo/effect at 20% MLDO
w/effects

600 ft
5 mm w/effect at 30% MLDO
150 ft

5 mm wo/effect
Figure 7: Gas production rate versus time for a fractured well at at 30% LDO
two-phase flow conditions, with and without velocity effects and
different fracture lengths, fracture width=5 mm and maximum
liquid drop-out=16%.

Figure 10: Gas production rate versus time for a fractured well
with and without velocity effects at two-phase flow conditions
with two different maximum liquid drop-out values of 20% and
30%, fracture width=5 mm, fracture length=150 ft.

600 ft
Set 2:
15 mm
150 ft
600
Set 1:
5 mm 150 ft

Figure 8: Gas production rate versus time for a fractured well at

two-phase flow conditions, different fracture lengths and widths,
all with velocity effects and maximum liquid drop-out=16%.