Spe 181684 MS

SPE-181684-MS
The Impact of Geomechanics and Perforations on Hydraulic Fracture

Initiation and Complexity in Horizontal Well Completions
George Waters and Xiaowei Weng, Schlumberger
Copyright 2016, Society of Petroleum Engineers
This paper was prepared for presentation at the SPE Annual Technical Conference and Exhibition held in Dubai, UAE, 26-28 September 2016.
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 have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written
consent of the Society of 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 conspicuous acknowledgment of SPE copyright.
Abstract
As is well known, hydraulically fractured horizontal wells have been extremely successful in the
development of low permeability reservoirs throughout the world. The vast majority of these completions
employ cased and cemented wellbores drilled approximately in the direction of minimum horizontal stress.
Multiple, relatively short perforation clusters are included within each frac stage along the lateral. This
efficiently creates many hydraulic fractures propagating orthogonal to the well, but it does not insure that
each perf cluster is effectively stimulated.
Many efforts have been made to improve the effectiveness of horizontal completions. This has mainly
focused on using lateral measurements to place perforation clusters in rock of similar stress so they are
more likely to be successfully stimulated. But this ignores the impact of formation initiation pressure and
tectonics on fracture initiation. In addition, the number, dimensions and orientation of the perforations in
each cluster can greatly influence the effectiveness of the stimulation at each initiation point.
To address these issues a near-wellbore fracture initiation calculator has been developed that predicts
whether a fracture will initiate at a perforation, the minimum initiation pressure, the fracture initiation
location and orientation at each perforation, and the injection rate into each perforation. These parameters
are a function of the casing size and orientation, the mechanical properties of the rock and cement, the
principle effective stresses, and properties of the perforations.
A series of sensitivities have been performed to quantify the impact of injection rate, tectonic setting,
stress variation between clusters, and perforation properties on hydraulic fracture creation, orientation and
complexity at each perf cluster. The sensitivities demonstrate that fractures may not initiate at many clusters
and that within an active cluster some perforations may not be accepting fluid. Incorporating the results
from this model enables engineers to design completions that insure all perforation clusters are effectively
stimulated and near-well fracture complexity is minimized.
This methodology does not just look at a single perforation, or cluster. Instead, it accounts for the stress
variation between multiple perforation clusters within a frac stage, in addition to perforation orientation,
dimensions and eccentricity, to predict the likelihood that each perforation cluster will be stimulated. By
employing this methodology one can better design a perforating system and optimize perforation placement
within a lateral to insure hydraulic fractures are created at all perforation clusters.
2 SPE-181684-MS
Introduction
Studies have shown that many perforation clusters are not producing fluids after stimulation (Miller et al.
(2011)). There are multiple causes for this behavior including:
• Unstimulated perforations because of:
◦ Perforations within a stage placed in higher stressed intervals than other perforations.
◦ Insufficient limited entry to overcome stress variations within a frac stage.
◦ Induced stress increases due to fractures spaced extremely close to each other: the so-called
"stress shadow."
◦ Tortuous near-well fracture geometries making proppant placement problematic.
• Loss of fracture conductivity due to:
◦ Over flushing during stimulation treatments and pump down operations.
◦ Perforations placed in a high stress interval resulting in post-stimulation movement of proppant

radially away from the wellbore in the hydraulic fracture.
◦ Perforations in intervals prone to proppant embedment.
◦ Post-stimulation crossflow between clusters creating unpropped fractures in the near-wellbore

area.
◦ Aggressive flowback resulting in proppant production, crushing and/or embedment.
One parameter that is frequently ignored is the near-well fracture initiation process and how perforation
cluster designs can impact this. It is well known that perforation density, phasing and orientation will impact
the geometry of the hydraulic fracture(s) at the wellbore (Behrmann and Elbel (1991), van de Ketterij
and de Pater (1997), Behrmann and Nolte (1998)). These perforation parameters, in conjunction with the
principle stress field and well orientation, will also control the initiation pressures at each perforation. It is
quite possible that the bottomhole pressure never exceeds the pressure required to initiate fractures at some
clusters. And within a cluster only a limited number of perforations may be active. Gaining an understanding
of how fractures form and (potentially) coalesce will allow one to improve the perforation design. This in
turn will improve the odds that significant fluid and proppant will be placed through a larger percentage
of the perforation clusters.
Theory
Figure 1 illustrates a cased and cemented horizontal wellbore with multiple clusters of spirally placed
perforations. The wellbore is typically oriented in the minimum horizontal stress direction, though it could
have an azimuth angle α with σhmin.
SPE-181684-MS 3
Figure 1—Diagram of multiple clusters of spiral perforations in a

horizontal well and initial cracks originated from the perforations.
When fluid is pumped into the wellbore at a high pump rate, the pressure rises rapidly until it exceeds the
fracture initiation pressure to propagate fractures from at least some of the perforations. Fracture initiation
pressure will vary for different perforations oriented at different angles and located at different depths (for
a generally heterogeneous formation). The perforation with the lowest initiation pressure will break down
first. However, the entire flow rate cannot be forced through one or a limited number of perforations due
to very high perforation friction caused by a large flow rate per perforation. Consequently, the wellbore
pressure will continue to rise with time, with the pumped fluid stored in the wellbore due to the limited flow
rate that can be accommodated by the broken perforations, subsequently leading to more fractures initiated
from additional perforations whose initiation pressure is exceeded by the rising wellbore pressure. This
process continues until all the broken perforations can take the full pump rate at a wellbore pressure below
the next initiation pressure among the remaining perforations. A peak pressure is hence reached, which is
defined as the breakdown pressure of the treatment interval and can be observed in the treating pressure data.
The process of fracture initiation from a perforation and subsequent fracture propagation from the
perforation after its initiation is very complex. It is affected by many factors, including but not limited to:
• The hoop stress surrounding the initial borehole.
• The bonding between the cement and the rock, and cement and casing, and the stress perturbation
by the cement curing process.
• The stress state surrounding the perforation tunnel when the perforation is created and potential
mechanical damage/stress cage resulting from perforating.
• The tensile strength of the rock.
• Possible fluid penetration into the cement micro annulus during the fracture initiation process.
• Rock failure and/or plastic deformation.
• Poroelastic effect due to fluid diffusion into the rock matrix.
• Influence of fluid flow into the starter fractures on the peak breakdown pressure.
• Fracture tortuosity during early propagation after the fracture initiation contributing to added near-
wellbore friction pressure.
4 SPE-181684-MS
• Interaction of multiple fractures originating from multiple perforations affecting fracture initiation
and breakdown pressure.
To compute the fracture initiation pressure, certain assumptions are made, including:
• The rock is assumed elastic and isotropic (although a correction for plasticity based on empirical
data is included).
• The effect of fluid viscosity and pressurization rate on fracture initiation pressure is ignored.
• Potential delamination of the cement bond and fluid flow into the cement micro annulus is not
considered. Similarly, potential fracture initiation from a micro-annulus is also not considered.
• The rock is assumed to have very low permeability and the poroelastic effect is neglected.
• Stress perturbation due to the presence of the neighboring perforations is neglected, as is the
interaction of the initiated fracture with the neighboring perforations.
• Additional pressure loss due to near-wellbore tortuosity is neglected.
For the estimate of fracture initiation pressure from a perforation, the perforation is considered as an
idealized cylindrical open hole. The initiation pressure is determined using the elastic solution for an infinite
open hole. Two most likely initiation sites will be considered: one at the base of the perforation, and the
other at the tip of the perforation. At each initiation site the local stresses in the rock, taking into account the
wellbore, will first be computed and then used as the "far-field stresses" applied on the perforation hole, as
proposed by Hossain et al. (2000). This approximation is reasonable since the size of the perforation tunnel
is much smaller than the size of the borehole.
Initial Stress Distribution in the Rock Around the Borehole

Figure 2 illustrates the general configuration of the problem and the coordinate system used. The formation
is assumed to be subjected to a vertical principal stress σv, the minimum horizontal stress σhmin, and the
maximum horizontal stress σhmax. A coordinate system (x1, x2, x3) with its axes aligned with the far field
stresses is shown in Fig. 2. The wellbore considered has an arbitrary deviation angle where Es and from
vertical, and an azimuth angle α measured from the minimum horizontal stress direction as shown in Fig. 2.
A local coordinate system (x, y, z) is established such that the origin is at the center of the wellbore cross-
section and the z axis is aligned with the wellbore axis. The x axis is aligned with the highest point of the
well circumference, and y axis is pointed to the side of the well.
SPE-181684-MS 5
Figure 2—Diagram showing the far field stresses and the coordinate system.
The coordinate transformation between the (x1, x2, x3) and (x, y, z) can be obtained as follows:
(1)
Projecting the far field in-situ stresses in the (x1, x2, x3) coordinate system to the coordinate system (x, y,
z) aligned with the wellbore axis, the remote stress components become
(2)
where A is the coordinate transformation matrix given in Eq. (1), and σ1 = σhmin, σ2 = σhmax, and σ3 = σv.
Figure 3 further shows a diagram of the cross-section of the wellbore. The wellbore consists of a cemented
casing assumed to be concentric with the well with inside and outside radius of Ri and Ro, and the borehole
radius is rw.
6 SPE-181684-MS
Figure 3—Diagram of wellbore cross section.
First, consider the stress distribution around the borehole in the rock under the in-situ stresses when the
borehole is drilled. Based on the elasticity solution by Kirsch, the stress distribution is given below in the
polar coordinates (r, θ) (Yew and Li (1988), and Thiercelin and Roegiers, Chapter 3, Reservoir Stimulation,
(2000)):
(3)
SPE-181684-MS 7
where pR is the pressure applied at the rock face at the borehole wall. The equation for σz is obtained based
on the assumption of plain strain condition, i.e. the incremental strain in the axial direction during borehole
creation is zero. For a rock that has an initial pore pressure p0, all the stress components and pressure shown
above are the actual quantities subtracted by p0. When a casing is cemented in the borehole, the effective
wellbore pressure is approximately equal to the hydrostatic pressure of the cement column pc. The actual
pressure, or more accurately the radial stress applied at the interface between cement and rock interface, can
evolve with time due to cement expansion/shrinkage due to chemical reaction and temperature changes as
the cement cures, and possibly rock creep. For simplicity, these effects are neglected here and the stresses
induced from the cement pressure are assumed to be permanently locked in.
When fluid is pumped into the well after it is cased and perforated, the fluid pressure pw is applied inside
the casing. This pressure acting radially on the inside of the casing wall is only partially transmitted to the
rock face since the casing is much stiffer than the rock. Li (1991) derived the radial stress transmitted to the
interface of casing and cement and it is given as follows
(4)
where Es and vs are Young's modulus and Poisson's ratio of the steel and TF is the transmission factor defined
as a fraction of the wellbore pressure transmitted as a radial stress on to the rock face.
The above calculation assumes the cement is perfectly bonded. However, if the bonding between the
cement and casing or cement and rock is poor and there exists micro annulus, fluid may penetrate into the
micro annulus, resulting in an effectively higher fluid pressure applied at the cement casing interface. In the
limiting case of a very permeable channel existing at the interface, the wellbore fluid pressure is directly
applied on the casing cement interface, in which case TF = 1. Since perforations are typically shot in spiral
fashion, it is likely that fractures are initiated from only some of the perforations first. As these fractures
grow in size and open up in width, the casing/cement interface may fail in tension or shear, allowing fluid
to penetrate into the now more permeable interface. This leads to more elevated interfacial fluid pressure
and easier fracture initiation in the remaining perforations in the vicinity of these fractures.
Stress Concentration at the Perforation Wall

Along a perforation tunnel, the rock is subjected to varying circumferential stress σθ (also referred to as
the "hoop stress"), axial stress σz, as well as radial and shear stresses, as a function of radial distance and
angle, as indicated in Eq. (3). These stresses act as remote stresses for the perforation tunnel, causing
further stress concentration in the rock around the perforation tunnel. Accurate computation of the stresses
at the tunnel wall requires a numerical simulation (Alekseenko et al. (2012)). However, since the diameter
of the perforation is generally much smaller than the size of the borehole, we can obtain a reasonable
approximation by assuming that the perforation is a cylindrical open hole, and each section of the open
hole is subjected to remote stresses equal to the local stresses in the rock around the borehole prior to the
perforating given by Eq. (3) (following an approach by Hossain et al. (2000)). Furthermore, the stress field
at the wall of the perforation tunnel can be calculated using the same Kirsch elasticity equations to assess
fracture initiation.
8 SPE-181684-MS
Initiation at the base of the perforation. At the base of the perforation, the initial stresses before perforating
can be computed from Eq. (3) by letting r = rw and θ = θp, which denotes the angle of the perforation. The
resulting stresses acting on the perforation tunnel can be denoted as (σr,wb, σθ,wb, σz,wb, σrθ,wb, σrz,wb, σθz,wb)
(note: both σrθ,wb, σrz,wb are zero). If one looks at the cross section at the base of the perforation in the plane
perpendicular to the perforation axis, the stresses acting in the plane are (σθ,wb, σz,wb, σθz,wb), as illustrated
in Figure 4.
Figure 4—Diagram illustrating the stresses acting at the base of a perforation.
The local principal stresses in this plane can be obtained as
(5)
If the shear stress σθz,wb is non-zero, it causes the local maximum stress to have an angle with wellbore
axis, which can be determined from
(6)
When the wellbore is pressurized, the perforation tunnel is exposed to the wellbore pressure pw. As pw
increases, it induces an increasing tensile hoop stress around the perforation tunnel. At a sufficiently high
pw, the tangential stress at the perforation wall, σ , where ϕ refers to the angle around the circumference
ϕ
of the perforation tunnel, exceeds the tensile strength of the rock, σT, and a fracture initiates. That is, the
fracture initiation pressure can be determined from
(7)
where σ1 is the smaller of the two principal stresses in the cross-sectional plane.
In the case where σθ,wb < σz,wb and σθz,wb = 0, σ1 = σθ,wb and σ2 = σz,wb. The initiated fracture is normal to
σθ,wb, i.e. parallel to the wellbore axis and hence a longitudinal fracture. In the case where σθ,wb > σz,wb and
σθz,wb = 0, then σ1 = σz,wb and σ2 = σθ,wb. The initiated fracture is normal to σz,wb, i.e. perpendicular to the
wellbore axis and hence a transverse fracture.
It should be mentioned that the effective tensile strength of the rock, σT, is scale dependent. In a small
hole, such as a perforation tunnel, the effective tensile strength can be much greater than the corresponding
value for a large borehole, or that measured in a laboratory tensile test that typically uses a larger rock
sample, as shown in the experimental results by Carter (1992) (Figure 5).
SPE-181684-MS 9
Figure 5—Effect of hole size on fracture initiation from a cylindrical hole under uniaxial load (Carter).
Additionally, the process of creating perforations by shape charges causes plastic deformation of the rock
and potential residual stress in the vicinity of the perforation tunnel, which can contribute to an effectively
larger tensile strength. An effective tensile strength determined from laboratory test by initiating a fracture
from a hole size representative of a perforation would give a more realistic value.
Initiation at the tip of the perforation. At the tip of the perforation, the initiation pressure can be determined
in a similar way as at the base. The remote stresses acting on the perforation can be computed from Eq. (3)
by letting θ = θp, and r = rp, where rp = rw + lp, and lp is the length of the perforation tunnel. The resulting
stresses acting on the perforation tip can be denoted as (σr,pt, σθ,pt, σz,pt, σrθ,pt, σrz,pt, σθz,pt).
Similar to at the base of the perforation, one can determine the principal stresses, σ1 and σ2, in the plane
perpendicular to the perforation, and their angles with respect to the wellbore axis, from Eqs. (5) and (6).
However, unlike at the base of the perforation, the off-plane shear components σrθ,pt and σrz,pt can be non-
zero. To determine the fracture initiation condition, the tangential stress at the perforation wall, σ , as given
ϕ
in Eq. (7), is combined with the off-plane shear stresses in an equation similar to Eq. (5) to determine the
maximum principal tensile stress. Because of the off-plane shear, the induced fracture can be at an angle
cutting across the perforation axis.
Horizontal wellbore drilled parallel to the σhmin

Perforation at the top or bottom of the borehole. At this perforation orientation, the perforation is vertical.
For the given wellbore orientation, the remote stresses acting on the borehole are
(8)
For simplicity, all the stresses shown are effective stresses (with pore pressure subtracted; same is true
for the pressure).
First consider the initiation at the perforation tip. The stress distribution around the perforation tunnel
near the tip can be computed using Eqs. (3) – (7) as outlined above. For a sufficiently long perforation length
(lp > 2rw), the remote stresses acting on the perforation tip are very close to the far-field in-situ stresses.
10 SPE-181684-MS
Therefore the initiation pressure is close to the initiation pressure for a vertical open hole. The fracture
initiation pressure is the same as that obtained from the classic Hubbert and Willis equation
(9)
where pinit, tip refers to initiation pressure from the tip of the perforation. The initiated fracture is normal to
the σhmin along the perforation tunnel, i.e. transverse to the wellbore, as illustrated in Figure 6a.
Figure 6—Illustration of different fracture initiation scenarios from a perforation.
Now consider the stress condition at the base of the perforation. From Eq. (3),
(10)
At a sufficiently high fluid pressure, a fracture will initiate at the base of the perforation if the tangential
stress at the perforation wall becomes tensile (negative value) and exceeds the rock tensile strength. That is,
the initiation pressure satisfies one of the following conditions, whichever is reached first as the wellbore
pressure pw increases,
(11a)
or
(11b)
If the Eq. (11a) is satisfied first at a lower pw than that for Eq. (11b), which corresponds to the case of
σz,wb > σθ,wb, the initiated fracture is longitudinal and parallel to the wellbore axis, as shown in Figure 6b.
This corresponds to the following condition
SPE-181684-MS 11
(12)
This condition may be met when σv is significantly larger than the horizontal stresses, or if the
transmission factor is large, leading to a longitudinal fracture. The corresponding initiation pressure can be
determined from Eq. (11a) as
(13)
where pinit, base refers to initiation pressure from the base of the perforation.
Conversely, if Eq. (11b) is satisfied first than Eq. (11a), i.e. σz,wb < σθ,wb, the initiated fracture is transverse
to the wellbore, as shown in Figure 6c. The corresponding initiation pressure is
(14)
In the special case of v = 1/3, Eq. (14) becomes
(15)
Among the scenarios depicted in Fig. 6, where the fracture initiates is dictated by which scenario requires
the lowest initiation pressure. It is interesting to note that in a cased and perforated well parallel to σhmin,
a transverse fracture can be initiated (Fig. 6c). While for an uncased hole, the tensile failure can only
occur circumferentially (at least when natural fractures are not present at the borehole wall), leading to an
unfavorable longitudinal fracture for this well orientation.
In the case of a relatively small TF (good cement and stiff casing), the initiation pressure for creating a
longitudinal fracture at the perf base (Fig. 6b) as given by Eq. (13) will be high. Therefore the most likely
initiation scenarios would be either initiating from the perforation tip (Fig. 6a) or at perforation base as in
Fig. 6c, both leading to a transverse fracture.
In the limiting case of TF ~ 1, i.e. with the wellbore pressure fully transmitted to the rock face (e.g. due to
fluid leakage into the cement micro annulus), the initiation pressure at the perforation base (Fig. 6b) given
by Eq. (13) can be less than the initiation pressure at the perforation tip (Fig. 6a) given by Eq. (8), leading
to unfavorable initiation of a longitudinal fracture. The longitudinal initiation pressure given by Eq. (13)
can be even less than the corresponding initiation pressure from an open borehole (i.e. 3 σh max – σv + σT),
due to the additional tensile stress induced by the internal pressure at the perforation tunnel wall.
Perforation oriented to the side of a borehole (i.e. θp = 90°). For a perforation oriented to the side, let
us first consider the tip of the perforation. For a perforation with a length significantly greater than the
borehole radius, the stress condition at the tip is close to a horizontal open hole parallel to σhmax. Therefore,
the initiation pressure is (assuming σv > σhmin)
(16)
and the initiated fracture is perpendicular to σhmin and transverse to the well.
12 SPE-181684-MS
Note the initiation pressure from the tip of the side perforation would be lower than that from the tip of
top/bottom perforation given by Eq. (9) if σv >σhmax.
Now consider the base of the perforation. From Eq. (3), the stresses in the rock at the borehole wall are
(17)
If σz,wb > σθ,wb, which corresponds to the following
(18)
a longitudinal fracture will initiate. This corresponds to the case of horizontal stresses greater than the
vertical stress, which occurs in formation with reverse faulting condition.
For typical stress conditions, σz,wb < σθ,wb is more likely to occur and the initiation at the base of a side
perforation tends to produce a transverse fracture. The corresponding initiation pressure is
(19)
If the initiation pressure from Eq. (19) is less than the initiation pressure at the tip given by Eq. (16), the
fracture initiates at the base of the perforation.
For v = 1/3, Eq. (15) also holds true. The initiation pressure can be less than that at the tip given by Eq.
(16) if TF is not close to 1. For Poisson's ratio v > 1/3, the initiation pressure at the base of a side perforation
would be greater than the initiation pressure from the top/bottom perforation given in Eq. (14), if σv >σhmax.
On the other hand, if σv <σhmax, the initiation pressure from the side perforation becomes less than the top/
bottom perforation.
When there is a large compressive hoop stress and a low axial stress at borehole wall, as is the case
for a side oriented perforation, it may be possible that a low wellbore pressure is sufficient to initiate a
micro-crack at the base of the perforation. However, this crack cannot extend far from the perforation if the
pressure is still below the "remote" stress acting on the perforation (i.e. the stress around the borehole before
perforating). In that case, the initiation pressure is given the value of the local minimum principal stress.
Horizontal wellbore drilled parallel to the σhmax

In this well orientation, the remote stresses acting on the borehole are
(20)
Perforation at the top or bottom. First, consider the initiation at the perforation tip. Again, assume the
perforation length is sufficiently long so that the influence of the borehole is negligible. Since the perforation
is oriented vertically, the fracture initiation pressure is the same as that given in Eq. (9), and the initiated
fracture is perpendicular to σhmin. For this well orientation, the initiated fracture is parallel to the well axis,
instead of transverse as in the previous case. This means that if the fracture is to initiate from the tip of
the top/bottom oriented perforation, the fracture will be longitudinal, a favorable initiation scenario for this
wellbore orientation.
Now considering the stress condition at the base of the perforation, one has,
SPE-181684-MS 13
(21)
If σz,wb > σθ,wb, which corresponds to the following
(22)
a longitudinal fracture will initiate. This condition is almost always met and the corresponding initiation
pressure is
(23)
Similar to the previous discussion on the well aligned with σhmin, the initiation pressure at the perf
base depends on TF; a low value of TF leads to much higher initiation pressure (and low likelihood of
occurrence), while a TF close to 1 gives an initiation pressure lower than the corresponding value for an
open borehole (3 σh min – σv + σT). Furthermore, this initiation pressure for TF = 1 is also lower than the
initiation pressure at the tip given by Eq. (9) if σv > σhmax. Similar to the earlier case, a high TF is promoting
the initiation of a longitudinal fracture, which is more favorable in this well orientation as opposed to the
transverse fracture being more favorable for the well oriented in σhmin direction.
Now consider the case of σz,wb < σθ,wb, in which the initiated fracture is transverse to the wellbore. The
corresponding initiation pressure is
(24)
For the special case of v = 1/3, Eq. (24) becomes
(25)
This is generally greater than the initiation pressure of a longitudinal fracture given by Eq. (23). Therefore,
a longitudinal fracture is more likely to initiate from a top/bottom perforation for a well aligned with σhmax.
Perforation oriented to the side of a borehole (i.e. θp = 90°). For perforation oriented to the side, and if
σv > σhmax, the initiation pressure from the tip is given as
(26)
The initiated fracture is perpendicular to σhmax and transverse to the well. This initiation pressure may or
may not be greater than the initiation pressure from the tip of a top/bottom perforation given by Eq. (9),
depending on the magnitude of the in-situ stresses. If σhmax is close to σhmin, then the initiation pressure from
the tip of a side perforation is lower than from the top/bottom perforation.
14 SPE-181684-MS
If σv < σhmax, then a horizontal fracture would be initiated along the perforation tunnel and the fracture is
longitudinal to the well, and the corresponding initiation pressure is
(27)
If the stress condition is reverse faulting, i.e. σv < σhmin < σhmax, the fracture plane will stay as a horizontal
plane and there is no fracture turning as the fracture grows in size. However, if the stress regime is strike-slip,
i.e. σhmin < σv < σhmax, the initiated horizontal fracture will have to reorient to become perpendicular to σhmin as
it propagates, resulting in severe fracture tortuosity and is hence undesirable. But in the strike-slip condition,
the initiation pressure from the tip of the side perforation given by Eq. (27) is much higher than the initiation
pressure from the tip of the top/bottom perforation given by Eq. (9). So in the case when perforations of
both orientations co-exist, the top/bottom oriented perforation will be first to initiate a fracture.
Now let us consider the initiation from the base of the side-oriented perforation. From Eq. (3), the stresses
in the rock at the borehole wall are
(28)
Again, if σz,wb > σθ,wb, which corresponds to
(29)
a longitudinal fracture will initiate. This again corresponds to the case of horizontal stresses greater than the
vertical stress, which occurs in formation with reverse faulting condition.
For typical stress conditions, σz,wb < σθ,wb is more likely to occur and the initiation from the base of the
side perforation is generally a transverse fracture. The corresponding initiation pressure is
(30)
Consider the special case of v = 1/3, Eq. (30) becomes
(31)
This initiation pressure will be high for large TF. The transverse fracture initiation may occur for relatively
small TF.
Generally speaking, for a well oriented in the σhmax direction, the fracture initiation pressure from the side
perforation typically result in an unfavorable transverse fracture for this well orientation, except in the case
of reverse faulting stress condition where a longitudinal horizontal fracture can initiate either from the tip
or at the base of the perforation. It is more desirable to put perforations at top/bottom positions for this well
orientation (except for the reverse faulting stress condition) to initiate a longitudinal fracture perpendicular
to σhmin.
The analytical model presented above has been compared to the numerical simulations by Alekseenko
et al. (2012) and good agreement has been obtained. Due to the space limitation, this comparison will not
be presented in this paper.
SPE-181684-MS 15
Computation of Overall Breakdown Pressure and Number of Open Perfs

To compute the overall breakdown pressure, the wellbore pressure will be increased incrementally. The
pressure will be checked against the initiation pressure for individual perforations. If the pressure exceeds
the initiation pressure, the corresponding perforation is assumed broken down. The corresponding flow rate
into the perforation is computed based on the equation for perforation friction and the difference between the
wellbore pressure and the fracture propagation pressure (for simplicity assumed as the local normal stress
acting on the starter fracture). If the sum of the flow rates into all open perforations is less than the total
pump rate, the wellbore pressure is increased and the process repeats until the total pump rate is reached.
The perforations that are opened are thus determined.
In determining the number of perforations broken down in the above process, a perforation whose
initiation pressure has not yet been reached but is physically intersected by a fracture initiated from a
neighboring perforation is also considered as "open", since direct fluid flow path is established between
the perforation and the neighboring fracture. To account for this connection, for each initiated fracture, its
possible intersection with its immediate neighbors are checked (note: intersection beyond the immediate
neighbors are not considered). Figure 7 shows a perforation from which a fracture is initiated at an angle
γ with respect to the wellbore axis, and the projection of its two neighboring perforations onto the plane
that contains the wellbore and is normal to the perforation. If the fracture angle falls within the range of
γperf_base ≤ γ ≤ γperf_tip, where γperf_tip is the angle of the dotted line that connects the perforation and the tip of
the neighboring perf projected on the plane with respect to the wellbore axis as shown in Fig. 7, and γperf_base
is the corresponding angle of the line connecting the adjacent perforations.
Figure 7—Diagram illustrating possibility of an initiated fracture intersecting neighboring perforations.
That is,
(32)
where θphasing is the phasing angle between the neighboring perforations and Δ is the perforation spacing. For
a fracture angle falling in the range 0 ≤ γ ≤ γperf_base, even though the fracture does not directly intersect the
neighboring perforation tunnel, we also consider the perforation as being connected by the initiated fracture,
since the opening of the fracture causes the rock to be detached from the casing and consequently allows
direct fluid communication with the neighboring perforation through the casing annulus.
16 SPE-181684-MS
Perforation Phasing and Orientation

Whether injection occurs through a perforation will be a function of not only the far field effective stresses,
and mechanical properties of the rock and cement, but also the diameter, length and orientation of the
perforation. A stressed rock perforating analysis software package was used to determine these parameters.
Perforation performance in stressed rocks has been determined to be different than API Standard RP 19B,
Section I, unstressed concrete perforation performance. Harvey et al. (2012) and Grove et al. (2016) have
provided overviews of the stressed model used in this work.
A common industry practice is to shoot horizontal wells with guns incorporating 6 shots per foot, oriented
at 60 degree angles. If the gun is centered a 60 degree phasing assures that a hydraulic fracture will be present
within 30 degrees of a perforation. This work shows that fractures will be created along the perforation
when a quality annular seal is present. Short perforation clusters of 1 to 2 ft are common since this promotes
perforation friction to try to induce fractures at multiple perforation clusters within a single frac stage. It
also minimizes the risk of multiple transverse fractures propagating from a perforation cluster. As this work
shows, short clusters do not guarantee that this will occur. All modeling in this paper utilizes a 6 shot per
foot, 60 degree phased, spiral gun.
Inputs into the stressed rock model are as follows:
Formation Compressive Strength: 10,000 psi Borehole Diameter: 8.75 in

Formation Bulk Density: 2.63 sg Casing: Centralized 5 ½" 17#, P110
Vertical Stress: 13,460 psi Cement: 16.4 lb/gal
Pore Pressure: 7,742 psi Wellbore Fluid: 8.42 lb/gal brine
Temperature: 190 F
A 3 1/8" gun with 21.6 gr charges was used. This gun is designed to provide similar hole dimensions
around the pipe when eccentered. Eccentricity on the bottom of the casing, as is common in horizontal
wells, is assumed. Since hole performance will vary depending on how the charges orient, three different
simulations were performed:
– Charges oriented at: 0, 60, 120, 180, 240 and 300 degrees
Zero degrees is referenced to the top of the casing. These three orientations will cover the range of
possible perforation orientations in a horizontal well. The perforation performance for these orientations
are shown in Figures 8, 9, 10.
SPE-181684-MS 17
Figure 8—Stress rock perforation performance for an eccentered gun with a charge orientation to the top of the hole.
Figure 9—Stress rock perforation performance for an eccentered

gun with a charge orientation 15 degrees from the top of the hole.
18 SPE-181684-MS
Figure 10—Stress rock perforation performance for an eccentered

gun with a charge orientation 30 degrees from the top of the hole.
The average hole diameter for all three gun orientations is 0.42 in. But the actual diameters vary from
0.37 in to 0.42 in (18.9% variation) when oriented to the top of the hole, 0.38 in to 0.44 in (15.8% variation)
when oriented 15 degrees from the top of the hole, and 0.39 in to 0.44 in (12.8% variation) when oriented
30 degrees from the top of the hole. This will impact the injection rate through each perforation when
limited entry is employed. This injection rate will vary depending on how the charges orient in the hole
(gun rotation), and is something that cannot be controlled from the surface.
The average formation penetration is 16.04 in, and varies by 6.3%, 6.2% and 5.6% as the While the
charges in the guns are oriented 60 degrees to each other, the holes in the pipe are not actually at 60 degrees
because of gun eccentricity. This will impact the initiation pressures and potentially the fracture initiation
point and orientation. The actual orientation of the holes in the casing are shown in Table 1, and can be seen
in Figs. 8, 9 and 10. It is recommended to use the largest diameter guns possible to mitigate this effect and
to equalize hole performance around the casing. Simulations through the rest of this paper will show results
when using the 3 1/8 inch gun for two conditions:
– Charges oriented symmetrically at 60 degrees around the casing: Charge Orientation in Table 1.
– Holes in the casing accounting for gun eccentricity: Hole Location on Casing in Table 1.
Table 1—Actual hole location in the casing due to gun eccentricity for multiple charge orientations.
Charge Hole Location Charge Hole Location Charge Hole Location

Orientation on Casing Orientation on Casing Orientation on Casing
0° 0° 15° 20° 30° 36°
60° 76° 75° 95° 90° 109°
120° 139° 135° 146° 150° 162°
180° 180° 195° 195° 210° 198°
240° 221° 255° 236° 270° 251°
300° 284° 315° 298° 330° 324°

SPE-181684-MS 19
Lateral Azimuth
The majority of hydraulically fractured horizontal wells are drilled close to the direction of σhmin. A
discussion on the initiation pressures and hydraulic fracture initiation points and orientations for wells in
this orientation will be covered in the subsequent Tectonic Setting section of this paper. But the authors
would be remiss not to address wells drilled in the direction of σhmax.
An extensional structural setting is assumed with the following properties:
TVD = 10,000 ft
Pore Pressure Gradient = 0.65 psi/ft Overburden Gradient = 1.10 psi/ft
Minimum Horizontal Stress Gradient = 0.75 psi/ft Maximum Horizontal Stress Gradient = 0.85 psi/ft
Young's modulus = 5,000,000 psi Poisson's ratio = 0.20
Tensile Strength = 1,000 psi Unconfined Compressive Strength = 10,000 psi
Hole Diameter = 8.75 in Cement Pressure = 6,500 psi
Casing OD = 5.5 in Casing ID = 4.89 in
For a well in this orientation the preferred far field frac plane is vertical and in the same axis as the
wellbore. For a well without perforations there is a strong stress concentration requiring the near-well
fracture to be vertical and axial with the well. Table 2a and 2b shows that even with 60 degree phased
perforations there is still a strong preference for the near-well induced fracture to be vertical and axial. The
pressure required to initiate fractures from perfs 60 degrees from the top and bottom of the hole is 2,372
psi higher. While unlikely, if these perforations are activated then the fractures will initiate at the tip of the
perforation tunnel and be transverse to the wellbore. There is a some reduction in the initiation pressure for
the side perforations when accounting for casing hole eccentricity (Table 2b). The perforations at the top
and bottom of the hole have the same initiation pressure. Those on the side, while different than when holes
are centered around the pipe, still have a much higher initiation pressure.
Table 2a and 2b—Initiation Pressure, Location and Orientation for 60 degree phased guns oriented to
the top of the hole. Table on right shows the location of the perf hole on the casing due to eccentricity.
Charge Initiation Fracture Fracture Hole Initiation Fracture Fracture

Orientation Pressure (psi) Location Orientation Location Pressure (psi) Location Orientation
0° 7500 Base Axial 0° 7500 Base Axial
60° 9872 Tip Transverse 76° 9600 Tip Transverse
120° 9872 Tip Transverse 139° 8980 Base Transverse
Unless designed to do so the gun may not lie in the hole with perforations oriented to the top and bottom
of the casing. Table 3a and 3b shows the initiation pressure and fracture properties if the perforations are
rotated 15 degrees from the top of the hole for centered and eccentered casing holes, respectively. Table
4a and 4b show these properties if the perforations are rotated 30 degrees. For only 15 degree rotation the
initiation pressure does not change significantly and the near well fracture orientations do not change. Once
20 SPE-181684-MS
the rotation reaches 30 degrees the initiation pressure has increased by over 1,500 psi, plus the location of
the initiated fracture has changed. Because of the increase in stress concentration at the perforation base
the fractures now initiate from the tip of all the perforations not on the sides of the hole when the holes
are symmetrical around the casing. But due to eccentering the fracture will still initiate first at the holes
closest to the bottom of the casing (18 degrees off) and create an axial frac from the perforation root. In this
scenario the perforations near the top of the hole are not likely to be active because their initiation pressure is
approximately 1,900 psi higher than on the bottom of the casing. If poor proppant transport is present in the
casing (low pump rate or Slickwater usage) there is an increased risk of screenout of the lower perforations.
Table 3a and 3b—Initiation Pressure, Location and Orientation for 60 degree phased guns oriented 15 degrees
from the top of the hole. Table on right shows the location of the perf holes on the casing due to eccentricity.

135° 9500 Base Transverse 146° 9267 Tip Axial
315° 9500 Base Transverse 298° 9775 Tip Transverse

30° 9030 Tip Axial 36° 9406 Base Transverse
150° 9023 Tip Axial 162° 7500 Base Axial
210° 9023 Tip Axial 198° 7500 Base Axial
It is clear that fracturing pressures and near-well geometries are strongly affected by how a gun lies in the
bottom of a horizontal well. Dramatically different initiation pressures may be encountered with different
size guns (variable eccentricity), or simply because of gun rotation of only 30 degrees. For wells drilled in
this orientation it is recommended to incorporate an oriented perforating strategy where holes are shot only
on the top and bottom of the casing. Long perforation clusters are acceptable although this relies on the well
being perfectly aligned with σhmax. If spiral perforating is employed then limited entry should not be applied
as this will dramatically raise treating pressures and induce near-well fracture complexity.
Tectonic Setting
Far field tectonics will impact the initiation pressure, and fracture location and orientation even when the
same perforating systems are used on horizontal wells drilled in the direction of minimum horizontal stress.
Therefore caution must be exercised when fracturing in tectonically stressed environments and applying
perforating strategies used in extensional environments such as are commonly found in North America.
Significantly different initiation pressures and near-well fracture geometries should be expected. This is
SPE-181684-MS 21
demonstrated via examples using the same perforating and casing designs in the three potential structural
settings.
Extensional: σhmin < σhmax < σv

Far field hydraulic fractures should be vertical in this structural setting. If one assumes mechanical properties
isotropy then hydraulic fractures should initiate axially and vertically at the wellbore in the absence of
perforations for wells drilled in the direction of minimum horizontal stress because overburden is the highest
tangential stress. As will be shown, when accounting for perforations the near-well fracture orientation will
change.
For simulations in this setting the following properties were assumed:
TVD = 10,000 ft
The same hole, cement, casing and perforating gun properties as previously discussed are used, and it is
assumed that the guns are laying on the bottom of the hole.
Table 5a shows that while the holes on the top and bottom of the pipe are the first to initiate a frac, there is
essentially no variation in initiation pressure around the hole. This is because of the low effective stress due
to the high pore pressure. This is commonly encountered in organic shales. If the principle stresses remained
the same and the pore pressure gradient dropped to 0.50 psi/ft the variation in initiation pressure would
increase by over 1,000 psi. All fracs initiate at the tip of the perf except for those at the top and bottom of
the hole. The presence of perforations is altering the near-well fracture geometry from axial to transverse.
This should improve the well to fracture communication since the fracture(s) will always be vertical and
transverse, although there still can be near well complexity due to multiple fracture initiation points and
because some fracs initiate at the tip of the perforations and some at the root. Short perforation clusters will
reduce the potential for multiple, near-well fractures.
When accounting for eccentricity (Table 5b) four perforation orientations have the same initiation
pressure, but there is still little variation in initiation pressure around the hole. The perforation location is
only impacted by the shots on the side of the pipe since the top and bottom perforations are still up and
down. In this case there will be preferential injection from the bottom half of the hole because of the lower
initiation pressure and presence of more holes near the bottom of the pipe. Near-well fracture geometries
should be similar to the case when eccentricity is ignored.

0° 7500 Base Transverse 0° 7500 Base Transverse

22 SPE-181684-MS
When rotating perforations 15 degrees from the top of the hole the initiation pressure does not change,
but there are now four perfs that have the same, low initiation pressure (Table 6a). The fractures continue
to be transverse to the well. Fracs initiate from the tip of two of these perfs and from the base of the other
two. This shows that how the tool orients in the casing can impact the location of the fracture initiation even
when the phasing and initiation pressure do not change.
Table 6b shows the actual hole locations in the casing when accounting for eccentricity. There is still little
variation in initiation pressure but initial fractures all initiate from the base of the perfs. This shows that
actual hole location on the pipe (gun diameter) can have an effect on the location of the fracture initiation
even when phasing does not change.

There continues to be little change in initiation pressures when the holes are rotated 30 degrees from the
top/bottom of the hole (Table 7a). In this case the initial fracs are all at the base of the perfs. The fractures
continue to be transverse to the well. Table 7b shows the actual hole locations in the casing when accounting
for eccentricity. There is still little variation in initiation pressure but initial fractures now initiate from both
the base and the tip of the perforations.

Strike – Slip: σhmin < σv < σhmax

Far field hydraulic fractures should be vertical in this structural setting. If one assumes mechanical properties
isotropy then hydraulic fractures should initiate axially and horizontally at the wellbore in the absence of
perforations for wells drilled in the direction of minimum horizontal stress because the maximum horizontal
stress is the highest tangential stress. But as was the case in the extensional environment, when accounting
for perforations the near-well fracture orientation will change.
SPE-181684-MS 23
TVD = 10,000 ft
The same hole, cement, casing and perforating gun properties as previously discussed are use, and it is
Table 8a shows that the holes on the side of the pipe (60, 120, 240 and 300 degrees) are the first
to breakdown, with an initiation pressure of 11,634 psi or 1.16 psi/ft. Fractures are initiated at the base
of the perforations and are transverse to the well, meaning they are vertical. There is about 600 psi
variation in initiation pressure. The top and bottom perfs are the last to become active and will initiate
transverse fractures at the tip of the perforations. As in the extensional case, the presence of perforations is
altering the near-well fracture geometry from axial to transverse. This should improve the well to fracture
communication since the fracture(s) will always be vertical and transverse, although there still can be near
well complexity due to multiple fracture initiation points.
When accounting for eccentricity the initiation pressure has been reduced by about 400 psi because the
holes at 60 and 300 degree phasing are now actually closer to the side of the pipe, at 76 and 284 degrees,
respectively (Table 8b). The initiation pressure for the holes at 120 and 240 degrees (139 and 221 degrees)
increases slightly because they are now further away from the side of the pipe. The fracs continue to initiate
transverse to the casing. One change is that after the initial breakdown all subsequent fracture initiations
are at the tip of the perfs.

The initiation pressure only declines by 400 psi when the holes are rotated 15 degrees from the top/bottom
of the hole (Table 9a). But fractures are initiated at only two perforations at the lowest initiation pressure,
compared to four perfs when they are oriented to the top of the hole. The variation in initiation pressure
increases to about 1,000 psi. The fractures continue to be transverse to the well. Table 9b shows the actual
hole locations in the casing when accounting for eccentricity. Only one perf is initially active. It also shows
a variety of initiation pressures since symmetry is lost. In this case the initiation pressure is only slightly
lower, but there is still over 1,000 psi variation in initiation pressure around the hole.
24 SPE-181684-MS

When the holes are rotated 30 degrees from the top/bottom of the hole the initiation pressure is about 500
psi lower than when oriented to the top of the hole (Table 10a). As with the 15 degree rotation, fractures
are initiated at only two perforations at the lowest initiation pressure, compared to four perfs when they are
oriented to the top of the hole. The variation in initiation pressure is over 1,000 psi. The fractures continue
to be transverse to the well.
Table 10b shows the actual hole locations in the casing when accounting for eccentricity. Unlike with
the perfs rotated 15 degrees, there are again two perfs initially active. It also shows a variety of initiation
pressures since symmetry is lost. With this orientation the initiation pressure increases slightly compared
to the 15 degree orientation, because the perfs are no longer on the side of the hole. There is over 800 psi
variation in initiation pressure. The last perforations to breakdown are at the top and bottom of the hole.

Reverse: σv < σhmin < σhmax

Far field hydraulic fractures should be horizontal in this structural setting since overburden is the least
principle stress. If one assumes mechanical properties isotropy then hydraulic fractures should initiate
axially and horizontally at the wellbore in the absence of perforations. But when perforations are placed
both axial and transverse fractures can be created.
TVD = 10,000 ft
SPE-181684-MS 25
The same hole, cement, casing and perforating gun properties as previously discussed are used, and it is
Table 11a shows that the perfs on the side of the casing are the first to breakdown, with an initiation
pressure of 16,620 psi or 1.66 psi/ft. Fractures are initiated at the tip of the perforations and are axial to the
well, meaning they are inclined. These inclined fractures will turn horizontal very close to the perforation
tunnel since the stress concentration will decay within four hole diameters (Hsiao (1987), Yew and Li
(1988), Yew and Schmidt (1989) and El Rabaa (1989)). The actual location of the horizontal fracture will
be a function of the perforation length. If the wellbore pressure rises by 900 psi fractures may be created
at the tip of the vertical perforations from which transverse, vertical fractures are initiated. Horizontal
fractures are most likely to form shortly beyond the tip as the stress concentration decays most quickly
there. Thus, multiple horizontal fractures may be formed in this environment. Ultimately there will likely be
one dominant horizontal fracture as the others will be arrested due to the stress shadowing of such closely
spaced fractures. It is difficult to determine which horizontal fracture will dominate. This could be near
the wellbore since a fracture here provides the easiest flow path from the perfs to the fracture. But textural
variations along bedding planes may also influence which horizontal fracture dominates. There is a very
high risk of a near-well screenout due to a very complex near-well fracture geometry in this setting. If the
structural setting changes through the vertical section of a reservoir strong consideration should be given to
landing the lateral in an interval that is extensional or strike-slip versus reverse.
Table 11b shows the actual hole locations in the casing when accounting for eccentricity. The lowest
initiation pressure has been reduced by about 1,000 psi because there are now holes closer to the side of
the pipe. These fractures initiate as vertical, transverse fracs rather than axial, and will likely turn quickly
to reorient to become horizontal fracs. Once the pressure rises just over 300 psi subsequent initiations are
axial at the tip of the perforations (139 and 221 degrees). These fractures will rotate to horizontal from the
tip of the perforations. Thus fracture complexity is likely in this environment.
the top of the hole. Table on right shows the location of the perf holes on the casing due to eccentricity.

120° 16620 Tip Axial 139° 16003 Tip Axial
The initiation pressure also declines by about 1,000 psi when the holes are rotated 15 degrees from the
top/bottom of the hole (Table 12a). This places perforations closer to the side of the hole where the stress
concentration is the lowest. The fractures corresponding to the lowest initiation pressure are now vertical
and transverse to the well and initiated at the base of the perforations. There are fewer active holes in
this scenario until the pressure rises by 400 psi to initiate fractures from the other perforations. These new
fractures will be axial from the perf tip. Near-well frac complexity is likely in this scenario.
Table 12b shows the actual hole locations in the casing when accounting for eccentricity. As in the
extensional setting, only one perf is initially active. It also shows a variety of initiation pressures since
symmetry is lost. In this case the initiation pressure is lower by about 350 psi and the first frac now initiates
from the base and axial with the well (hole at 95 degrees). This will create a sub-horizontal fracture on
one side of the well. All other fractures initiate at the tip of the perfs and continue to be axial, but it is
26 SPE-181684-MS
unlikely that all perfs will be active due to the 1,700 psi variation in initiation pressure. In this scenario there
continues to be the possibility for multiple offset horizontal fractures at a perf cluster.

75° 15622 Base Transverse 95° 15291 Base Axial
255° 15622 Base Transverse 236° 16399 Tip Axial
The initiation pressure is the lowest when the gun rotated 30 degrees from the top of the hole because
this results in perforations at 90 and 270 degrees, where the stress concentration is the lowest (Table 13a).
Fractures at this point start horizontally (axial) at the base of the perforation. Eccentricity increases the
initiation pressure by close to 1,000 psi and alters the fracture orientation from axial to transverse (Table
13b). Gun size (eccentricity) will therefore have a material impact on initiation pressure and frac orientation.

90° 15100 Base Axial 109° 15892 Base Transverse
270° 15100 Base Axial 251° 15892 Base Transverse
Summary of Perforating in Variable Tectonic Settings

Extensional.
– The initiation pressure is low because of the low effective stress due to the high pore pressure. In this
example the initiation pressure is comparable to the far field closure stress.
– There is little variation in initiation pressure for holes in different orientations around the pipe.
– In all cases the fractures will initiate perpendicular to the hole. This is different than what would occur
in the absence of perforations, for example when using port collars instead of perforations. Multiple
transverse fractures will lead to flower petal-like multiple, slightly offsetting transverse fractures,
each occupying a different quadrant of the well due to stress shadow. Short perf cluster lengths will
minimize the number of far field transverse fractures.
– Most of the time fractures first initiate at the base of the perforations. But fractures initiated at the tip
are likely since there is little initiation pressure variation around the hole.
– Accounting for the actual location of the hole on the pipe will alter the initiation pressure from what
is estimated when assuming that all holes are 60 degrees apart around the casing.
SPE-181684-MS 27
Strike-Slip.
– The initiation pressure is over 2,000 psi higher than the far field closure stress because the effective
stress is higher than in the extensional example.
– Initiation pressure can vary by 1,000 psi around the hole due to stress concentrations.
– All fractures are vertical and transverse to the well even though the near-well fracture would be axial
and horizontal in the absence of perforations.
– The initiation pressure varies by about 500 psi depending on how the perforation charges orient in
the hole when the guns are laying on the bottom of the hole.
Reverse.
– Very high initiation pressures should be expected unless the pore pressure is exceedingly high. In this
example the initiation pressure varies from 1.51 to 1.75 psi/ft depending on hole location.
– The initiation pressure can change by up to 1,000 psi when hole eccentricity is taken into account, as
compared to when it is assumed that holes are every 60 degrees in the casing.
– This case is unique in that axial fractures are possible. These axial fractures typically will rotate
horizontally from the perf tip. This is a function of how the tool lies on the bottom of the casing.
– The initiation pressure around the hole can vary by up to 1,800 psi. This means that significant limited
entry is required to initiate fractures at all perforations within a perf cluster even when the far field
stresses are the same. It may be impossible to inject through all perforations when multiple perforation
clusters are placed in areas of variable closure stress in a given frac stage.
Multi-Cluster Completion
Customizing horizontal well completions by shifting perforations within a frac stage so that perforation
clusters are placed in intervals with similar properties is commonly employed when lateral measurements are
available. The lateral is usually grouped into sections based on petrophysical and geomechanical properties:
reservoir quality and completion quality, respectively. These engineering exercises have had a positive
impact on well productivity (Cipolla et al. (2011), Slocombe et al. (2013), Walker et al. (2012), Wutherich
and Walker (2012) and Wutherich et al. (2012)).
The vast majority of horizontal wells only have a GR measurement. Consequently, geometric completions
are utilized. Limited entry is employed to increase the odds that all perforation clusters will be stimulated.
Perforation friction of approximately 1,000 psi is common (roughly 2.2 bpm/perf through an average hole
diameter of 0.40 inches). On shallow wells higher perforation friction can be tolerated as the treating
pressure is usually low. This may not be possible on deeper wells or on wells with high injection rates
through small diameter tubulars. In such cases fewer number of perf clusters in a stage should be considered.
An example from the Eagle Ford Shale demonstrates the challenge of activating all perforation clusters,
and even perforations within a cluster when non-engineered, geometric completions are employed. The
variation in minimum horizontal stress within frac stages is shown for an engineered and geometric
completion in Figure 11 (Slocombe et al. (2011)). The stresses were estimated from a memory dipole sonic
tool logged on drill pipe after drilling the lateral. The average stress variation within a stage for the geometric
completion is 615 psi, and 235 psi for the engineered completion. Stage 6 will be used to demonstrate the
variable cluster injectivity of the geometric completion. In this stage the maximum stress variation between
clusters is 835 psi.
28 SPE-181684-MS
Figure 11—Estimated stress variation within a frac stage for geometric completion (red bars) and an engineered
completion designed to minimize stress variation at stage perforation clusters (blue bars) (URTeC 1571745).
The modeled inputs are: Geomechanical Parameters:

Total Injection Rate = 80 bpm Young's Modulus = 5,000,000 psi
Perforation Clusters = 4 Poisson's Ratio = 0.20
Perforations / Cluster = 9 Tensile Strength = 1,000 psi
Perforation Phasing = 60 degrees Unconfined Compressive Strength = 10,000 psi
Perforation Density = 6 shots/foot
It is assumed that mechanical properties are constant at each perforation cluster. The stresses and pore
pressures are shown in Table 14. The maximum horizontal stress is assumed to be 0.15 psi/ft higher than
the minimum horizontal stress, a reasonable assumption for this geologic setting. The perforation properties
are shown in Figure 8.
Table 14—Stress and pore pressure gradients at Stage 6 perforation clusters.
TVD (ft) Pore Pressure Overburden Minimum Maximum

Gradient (psi/ft) Gradient (psi/ft) Horizontal Stress Horizontal Stress
Gradient (psi/ft) Gradient (psi/ft)
Cluster 1 11,913 0.65 1.13 0.81 0.96
Cluster 2 11,911 0.65 1.13 0.86 1.01
Cluster 3 11,910 0.65 1.13 0.79 0.94
Cluster 4 11,909 0.65 1.13 0.83 0.98

SPE-181684-MS 29
The variable closure stresses will result in variable initiation pressures and injection rates between the
clusters. Figure 12 shows the bottomhole pressures and the corresponding total pump rates required to
initiate fractures at each perforation cluster. Fractures are quickly initiated at three of the four clusters.
But the rate must exceed 50 bpm before a fracture is initiated at Cluster 2, which has the highest closure
stress. This may not occur until after any acid pumped at the beginning of the frac has passed through the
perforations, causing the pressure to decline and a corresponding increase in injection rate.
Figure 12—Cluster initiation pressure and corresponding injection rate at initiation for each cluster.
Even though clusters start accepting fluid at the rates shown in Figure 12, all of the perforations are not
active. When the first frac starts at Cluster 3 at 9,409 psi only the holes on the top and bottom of the pipe
are accepting fluid (Figure 13a). There are 9 shots over 1.5 ft at 60 degree phasing. The seventh, eighth
and ninth shots are shown at 360, 420 and 480 degrees, respectively. When the second initiation occurs at
Cluster 1 at 9,650 psi the fracs again initiate at the top and bottom holes. But as can been seen in Figure
13b, the majority of the injection continues through Cluster 3. At 9,885 psi a frac is initiated at the top and
bottom perfs in Cluster 4 (Figure 13c). But the majority of the injection occurs in Clusters 3 and 1. In this
case this variable injectivity is not that important because all three clusters are active at a rate of only 7.3
bpm (Figure 12). But Cluster 2 does not become active until the bottom hole pressure reaches 10,555 psi at
a rate of 50.9 bpm. At this point the injection profile is as shown in Figure 13d. Clusters 3 and 1 are now
fully active. Injection is only out the top and bottom holes of Clusters 4 and 2.
30 SPE-181684-MS
Figure 13a—Injectivity by perforation in each cluster at time of frac initiation in the first cluster (Cluster 3). Rate < 1 bpm.
Figure 13b—Injectivity by perforation in each cluster at time of frac initiation in the second cluster (Cluster 1). Rate = 3.1 bpm.
Figure 13c—Injectivity by perforation in each cluster at time of initiation in the third cluster (Cluster 4). Rate = 7.3 bpm.
Figure 13d—Injectivity by perforation in each cluster at time of

frac frac initiation in the fourth cluster (Cluster 2). Rate = 50.9 bpm.
SPE-181684-MS 31
Figure 14 shows the injection rate through each perforation in each cluster when the designed pump rate
of 80 bpm is achieved. Holes at the top of the pipe are now receiving the least amount of fluid because of
the smaller hole size, but the injection around the hole is fairly symmetrical for Clusters 1, 3 and 4. Injection
at Cluster 2 continues to be at the top and bottom perfs only. This is because the tangential effective stress
is lower at the top and bottom of the hole in an extensional environment. Consequently, they are the first
perforations to take fluid. The injection rate into this cluster is limited because the higher closure stress,
and perforation friction is not adequate to initiate fractures from perfs on the side of the hole. Injection is
diverted to other clusters prior to activation of perfs on the side of the hole.
Figure 14—Injectivity by perforation for each perforation cluster prior to erosion (0.6 Discharge Coefficient). Rate = 80 bpm.
Figure 15 shows perforation injectivity after erosion (Discharge Coefficient = 0.80). Injectivity of Cluster
4 now is similar to Cluster 2. Due to erosion injectivity into some of the perforations actually stops. This
could easily lead to a near-well screenout within these perforations. This demonstrates that limited entry may
not be effective to insure continued injectivity into clusters throughout the treatment. Fiber optic monitoring
of stimulation treatments has shown injection rate variability in completions such as this (Molenaar et al.
(2011)).
Figure 15—Injectivity by perforation for each perforation to after erosion (0.8 Discharge Coefficient). Rate = 80 bpm.
Figure 16 shows the rates on a cluster basis when considering the maximum injection rate of 80 bpm.
The second cluster is accepting 7% of the total injection rate. The other three clusters are each receiving
about 30% of the total injection. Once the holes have eroded the cluster injectivity changes as shown in
Figure 17. Cluster 2 continues to accept only limited flow, but Cluster 4 flow has now reduced to 10% of the
total injection. Clusters 1 and 3 are now receiving over 80% of the total injected fluid. Such a scenario can
easily lead to screenout of Clusters 2 and 4, and over stimulation of Clusters 1 and 3, resulting in variable
productivity within the stage. For simplicity, the modeling of injection rate per cluster ignores any Net
Pressure build or stress shadow that may alter the pressure in the far field fractures.
32 SPE-181684-MS
Figure 16—Injectivity by perforation cluster before erosion (0.6 Discharge Coefficient). Rate = 80 bpm.
Figure 17—Injectivity by perforation cluster after erosion (0.8 Discharge Coefficient). Rate = 80 bpm.
When using a 6 shots/foot, 60 degree phased gun and more than one foot of perforations are shot, then
injectivity around the hole will be asymmetric depending on how the guns orient as some orientations will
have two shots. If a 9 shot gun lies in the hole so that the first shot is oriented to the top of the hole with
subsequent shots spiraling clockwise around the hole when looking down the well then the injectivity will
be as shown in Figure 18. In this case there are two holes at 0, 60 and 120 degrees. Conversely, if the first
charge orients 60 degrees from the top of the hole then the injectivity will be as shown in Figure 19. And
if the charge orients at 120 degrees then injectivity will be as shown in Figure 20. Injection from Cluster
2 is always from the top and bottom perfs as these are the only holes active. The injection rate pattern is
not symmetrical because of variable hole sizes due to gun eccentricity. This variable injectivity around the
hole may selectively screenout perforations with low injection rates. This may result in asymmetric fracture
wing growth and can potentially induce a screenout of a fracture wing with low injection rates. This may
be one reason why proppant pumped early in the treatment is often seen during early flow back.
SPE-181684-MS 33
Figure 18—Injectivity (bpm) around the hole when the toe charge orients to the top of the hole.
Figure 19—Injectivity (bpm) around the hole when the toe charge orients 60 degrees from the top of the hole.
34 SPE-181684-MS
Figure 20—Injectivity (bpm) around the hole when the toe charge orients 120 degrees from the top of the hole.
In all cases the fractures at the perforations are perpendicular to the well. Fractures from perforations
at the top and bottom of the hole start at the base of the perforation. Those on the side of the hole start
at the tip of the perforation because the stress concentration is lower at the tip in these orientations (Table
15). This is consistent with what was previously shown (Table 5a and 5b) in the extensional geomechanical
setting discussion.
Table 15—Initiation point of fracture at each perforation.
0° 60 ° 120 ° 180 ° 240 ° 300 ° 360 ° 420 ° 480 °
Cluster 1 Base Tip Tip Base Tip Tip Base Tip Tip
Cluster 2 Base Base Base
Summary of Multi-Cluster Completions

– Stress variations between perforation clusters within a frac stage will result in variable initiation
pressures, and variable injection rates per cluster. This issue is exacerbated with hole erosion as
limited entry injection is lost.
– Higher stress at a perforation cluster may require a high injection rate before injection into the cluster
commences. Injection may not occur until after any acid used at the beginning of the treatment has
already passed through the perforations.
– All perforations within a cluster may not be active and injectivity will change with changes in
pump rate and perforation erosion. Some perfs may stop accepting fluid which may induce near-well
screenout.
SPE-181684-MS 35
– With more than 1 ft of perforations at 60 degree phasing in a cluster there is the potential for variable
injectivity from each side of the hole. Ideally this will not affect far field fracture geometry, but it
could lead to screenout of perforations accepting limited fluid.
– Asymmetric injectivity around the hole will vary depending on how the gun lies in the bottom of the
casing. This is an uncontrolled parameter with current perforating technologies.
– All fractures initiate transverse to the well. This creates the potential for multiple, competing fractures
at each perf cluster. Shorter cluster lengths will reduce the risk of this.
Conclusions
This work has shown that fracture initiation, location and orientation is strongly influenced by:
– Gun diameter, eccentricity and orientation in which the gun lies in a horizontal well.
– Structural setting: extensional, strike-slip or reverse.
– Lateral orientation relative to the stress field: in the direction of σhmin or σhmax.
Transverse fractures can be initiated even for wells drilled in the direction of σhmax. Transverse fractures
can also be created for wells drilled in the direction of σhmin, although this is a function of the stress field.
A complex near-well fracture geometry can be induced when high perforation friction is used, with
fractures at some perforations being transverse to the well while others are axial to the well. These
geometries are influenced by the location of the actual holes in the casing which is a function of gun and
casing sizes.
Stress variations between perforation clusters in a frac stage can result is variable injectivity into the
clusters. Asymmetric injectivity can occur around the clusters in addition to between the clusters. This
injectivity will change as the perforations are eroded and can cause some active perforations to cease taking
fluid. The consequences of this can be screenout of individual perforations, or potentially whole clusters,
and variable stimulation volumes between clusters.
Even though the same gun may be employed, significant differences in initiation pressures and injectivity
will occur as the structural setting changes. Consequently, what works in extensional settings such as North
American may not be appropriate for compressional setting. Structural setting should be taken into account
when designing perforating systems.
Acknowledgements
The authors thank Schlumberger for permission to publish this work.
References
Alekseenko, O.P., Potapenko, D.I., Cherny, S.G., Esipov, D.V., Kuranakov, D.S. and Lapin, V.N. 2012. 3-D Modeling
of Fracture Initiation from Perforated Non-Cemented Wellbore. Paper SPE 151585 presented at the SPE Hydraulic
Fracturing Technology Conference, The Woodlands, Texas, USA, 6 – 8 February 2012.
Behrmann, L.A. and Elbel, J.L. 1991. Effect of Perforations on Fracture Initiation. JPT (May, 1991) pp 608 – 615.
Behrmann, L.A. and Nolte, K.G. 1998. Perforating Requirements for Fracture Stimulations. Paper SPE 39453 presented at
the SPE International Symposium on Formation Damage Control, Lafayette, Louisiana, USA, 18 – 19 February 1998.
Carter, B.J. 1992. Size and Stress Gradient Effects on Fracture around Cavities. Rock Mech. Rock Eng. 25 (3): 167–186.
Cipolla, C., Weng, X., Onda, H., Nadaraja, T., Ganpuly, U. and Malpani, R. 2011. New Algorithms and Integrated
Workflow for Tight Gas and Shale Completions. Paper SPE 146872 presented at the SPE Annual Technical Conference
and Exhibition, Denver, Colorado, USA, 30 October – 2 November 2011.
El Rabaa, W. 1989. Experimental Study of Hydraulic Fracture Geometry Initiated from Horizontal Wells. Paper SPE
19720 presented at the Annual Technical Conference and Exhibition, San Antonio, Texas, USA, 8 – 11 October 1989.
Grove, B. Harvey, J. and Martin, A. 2016. Perforating System Performance at Downhole Conditions: Recent Advances
in Modeling and Prediction. Paper SPE 178979 presented at the SPE International Conference and Exhibition on
Formation Damage Control, Lafayette, Louisiana, USA, 24 – 26 February 2016.
36 SPE-181684-MS
Harvey, J., Grove, B. and Zhan, L. 2012. Stressed Rock Penetration Depth Correlation. Paper SPE 151846 presented at
the SPE International Symposium and Exhibition on Formation Damage Control, Lafayette, Louisiana, USA, 15 –
17 February 2012.
Hossain, M.M., Rahman, M.K. and Rahman, S.S. 2000. Hydraulic Fracture Initiation and Propagation: Roles of Wellbore
Trajectory, Perforation and Stress Regime. J. Pet. Sci. & Eng., 27, pp 129–149.
Hsiao, C. 1987 A study of Horizontal Wellbore Failure. Paper SPE 16927 presented at the Annual Technical Conference
and Exhibition, Dallas, Texas, USA, 27 – 30 September 1987.
Li, Y., 1991. On Initiation and Propagation of Fractures from Deviated Wellbores. Ph.D. Thesis, The University of Texas
at Austin.
Miller, C., Waters, G. and Rylander, E. 2011. Evaluation of Production Log Data from Horizontal Wells Drill in Organic
Shales. Paper SPE 144326 presented at the SPE North American Unconventional Gas Conference & Exhibition, The
Woodlands, Texas, USA, 12 – 16 June 2011.
Molenaar, M.M, Hill, D.J., Webster, P., Fidan, E. and Birch, B. 2011. First Downhole Application of Distributed Acoustic
Sensing for Hydraulic-Fracturing Monitoring and Diagnostics. SPE Drilling and Completions, March 2012, pp 32 – 38.
Slocombe, R., Acock, A., Chadwick, C., Wigger, E., Viswanathan, A., Fisher, K. and Reischman, R. 2013. Eagle
Ford Completion Optimization Strategies Using Horizontal Logging Data. Paper URTeC 1571745 presented at the
Unconventional Resources Technology Conference, Denver, Colorado, USA, 12 – 14 August 2013.
Thiercelin, M.C. and Roegiers, J.C., 2000. Chapter 3 Formation Characterization: Rock Mechanics. Reservoir Stimulation,
3rd ed., Economides, M.J. and Nolte, K.G., John Wiley & Sons.
van de Ketterij, R.G. and de Pater, C.J. 1997. Experimental Study on the Impact of Perforations on Hydraulic Fracture
Tortuosity. Paper SPE 38149 presented at the SPE European Formation Damage Conference, The Hague, The
Netherlands, 1 – 2 June 1997.
Walker, K., Wutherich, K., Terry, J., Shreves, J. and Caplan, J. 2012. Improving Production in the Marcellus Shale Using an
Engineered Completion Design: A Case Study. Paper SPE 159666 presented at the SPE Annual Technical Conference
and Exhibition, San Antonio, Texas, USA, 8 – 10 October 2012.
Wutherich, K. and Walker, K. 2012. Designing Completions in Horizontal Shale Gas Wells. Paper SPE 155485 presented
at the Americas Unconventional Resources Conference, Pittsburgh, Pennsylvania, USA, 5 – 7 June
Wutherich, K., Walker, K., Aso, I., Ajayi, B. and Cannon, T. 2012. Evaluating an Engineered Completion Design in
the Marcellus Shale Using Microseismic Monitoring. Paper SPE 159681 presented at the SPE Annual Technical
Conference and Exhibition, San Antonio, Texas, USA, 8 – 10 October 2012.
Yew, C.H. and Li., Y. Fracturing of a Deviated Well, SPE Production Engineering (November, 1988) pp 429 – 437.
Yew, C.H. and Schmidt, J.H. 1989. On Fracture Design of Deviated Wells. Paper SPE 19722 presented at the Annual
Technical Conference and Exhibition, San Antonio, Texas, USA, 8 – 11 October 1989.

Spe 181684 MS

Uploaded by

Copyright:

Available Formats

Spe 181684 MS

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Spe 181684 MS

Uploaded by

Copyright:

Available Formats

SPE-181684-MS

The Impact of Geomechanics and Perforations on Hydraulic Fracture

George Waters and Xiaowei Weng, Schlumberger

Copyright 2016, Society of Petroleum Engineers

• Unstimulated perforations because of:

◦ Insufficient limited entry to overcome stress variations within a frac stage.

• Loss of fracture conductivity due to:

◦ Over flushing during stimulation treatments and pump down operations.

◦ Perforations placed in a high stress interval resulting in post-stimulation movement of proppant

◦ Post-stimulation crossflow between clusters creating unpropped fractures in the near-wellbore

Figure 1—Diagram of multiple clusters of spiral perforations in a

• The hoop stress surrounding the initial borehole.

• Rock failure and/or plastic deformation.

• Poroelastic effect due to fluid diffusion into the rock matrix.

Initial Stress Distribution in the Rock Around the Borehole

Figure 3—Diagram of wellbore cross section.

Stress Concentration at the Perforation Wall

Figure 4—Diagram illustrating the stresses acting at the base of a perforation.

The local principal stresses in this plane can be obtained as

Horizontal wellbore drilled parallel to the σhmin

Figure 6—Illustration of different fracture initiation scenarios from a perforation.

In the special case of v = 1/3, Eq. (14) becomes

If σz,wb > σθ,wb, which corresponds to the following

Horizontal wellbore drilled parallel to the σhmax

If σz,wb > σθ,wb, which corresponds to the following

For the special case of v = 1/3, Eq. (24) becomes

Again, if σz,wb > σθ,wb, which corresponds to

Consider the special case of v = 1/3, Eq. (30) becomes

Computation of Overall Breakdown Pressure and Number of Open Perfs

Figure 7—Diagram illustrating possibility of an initiated fracture intersecting neighboring perforations.

Perforation Phasing and Orientation

Formation Compressive Strength: 10,000 psi Borehole Diameter: 8.75 in

Figure 9—Stress rock perforation performance for an eccentered

Figure 10—Stress rock perforation performance for an eccentered

Charge Hole Location Charge Hole Location Charge Hole Location

0° 0° 15° 20° 30° 36°

60° 76° 75° 95° 90° 109°

120° 139° 135° 146° 150° 162°

180° 180° 195° 195° 210° 198°

240° 221° 255° 236° 270° 251°

300° 284° 315° 298° 330° 324°

Charge Initiation Fracture Fracture Hole Initiation Fracture Fracture

0° 7500 Base Axial 0° 7500 Base Axial

60° 9872 Tip Transverse 76° 9600 Tip Transverse

120° 9872 Tip Transverse 139° 8980 Base Transverse

180° 7500 Base Axial 180° 7500 Base Axial

240° 9871 Tip Transverse 221° 8980 Base Transverse

300° 9872 Tip Transverse 284° 9600 Tip Transverse

Charge Initiation Fracture Fracture Hole Initiation Fracture Fracture

15° 7500 Base Axial 20° 7500 Base Axial

75° 9599 Tip Transverse 95° 9613 Tip Transverse

135° 9500 Base Transverse 146° 9267 Tip Axial

195° 7500 Base Axial 195° 7500 Base Axial

255° 9601 Tip Transverse 236° 10076 Tip Transverse

315° 9500 Base Transverse 298° 9775 Tip Transverse

Charge Initiation Fracture Fracture Hole Initiation Fracture Fracture

30° 9030 Tip Axial 36° 9406 Base Transverse

90° 9615 Tip Transverse 109° 9591 Tip Transverse

150° 9023 Tip Axial 162° 7500 Base Axial