Methods Improve Stimulation Efficiency of Perforation Clusters in Completions
Stephen R. Ingram, Matthew Lahman, and Stephen Persac, SPE, Halliburton
Horizontal shale wells present the challenge of generating large, high-density fracture networks, reflecting the sub-microdarcy permeability of the formations drilled by these wells. The goal is to create the largest fracture network volume to maximize ultimate recovery, because the fracture network volume in these wells has been shown to correlate strongly with the production level.
However, as the network becomes too large for a given wellbore access point, the relative benefit of size diminishes. This is because of the low fracture conductivity, which creates large pressure drops within the network and makes it difficult to drain distant portions. And the effect is exacerbated by the inability to move water or liquid hydrocarbon through a large complex network (Mayerhofer et al. 2006). Thus, it is very important to create an optimal number of conductive transverse fractures or access points that intersect the wellbore.
Today’s unconventional wells incorporate wellbore planning and completion designs that are based on the reservoir-specific characteristics needed for optimal drainage and field development. The key elements of the design and planning process must be carefully considered. They are well spacing, lateral length, the number of stages, the length of isolated stages, and the number of perforation clusters per stage. The strategies used are based in part on advancements in reservoir simulation, reservoir modeling, and production correlations from trial and error that stem from the initial work in various plays, except the relatively unique Barnett shale.
Progress in Shale Completion Designs
A good example of this progression toward more reservoir-specific completion designs was seen in the Haynesville shale. The play saw a rapid ramp-up in activity from 2009 to 2012 with peak completion activity occurring in mid-2011. By November 2011, it had reached its highest production level of 7.2 Bcf/D (EIA 2014). This dramatic rise in production was in part due to the optimization of completion and stimulation designs, particularly the reduction of the isolated length of each stage (plug-to-plug distance) and, thus, an increase in the number of stages per foot of lateral.
The average daily gross perforated interval per stage (top perforation to bottom perforation) that Halliburton completed in the Haynesville and Bossier shales from 2010 to 2013 was analyzed. The data encompasses nearly 11,000 stages for more than 30 operators. It illustrates that many operators began to reduce their gross perforated interval per stage across the play by the middle of 2011. In July 2011, it was 272 ft and by mid-2012, it declined to 150 ft, falling at a relatively constant rate as operators increasingly went to a shorter isolated stage interval.
This indicates closer stage spacing (plug to plug) or more stages per well, with lateral length remaining relatively constant. These trends continued into 2012 and a dramatic improvement was seen not only in the slope of the projected production decline curve, but also in the estimated ultimate recovery (EUR) for the wells being brought online.
The 72-month cumulative production average for wells completed in 2010 and 2011 was roughly the same at 4.5 Bcf per well, but for wells completed in 2012, it rose to 5.5 Bcf per well (Kaiser and Yu 2013). This marked a 22% improvement and was largely because of the reservoir-specific completion optimization progression, such as closer stage spacing.
Case Study: Tighter Stage Spacing
Operator A had a significant portion of acreage outside of the known Haynesville fairway. The operator began with the standard stage length design at that time, roughly 300 ft from plug to plug. In mid-2011, it began to experiment with reducing the stage spacing and completing more stages per lateral foot. By the beginning of 2012, the operator had dropped its stage interval lengths to 150 ft with a gross perforated interval of 100 ft (top perf to bottom perf) with 31-ft perforation clusters, each spaced 50 ft apart.
Operator B, which had a major block of acreage in the Haynesville sweet spot, kept essentially the same stage spacing throughout the period. The standard design was roughly a 375-ft stage length (plug to plug), with a 320-ft gross perforated interval and eight 1-ft perforation clusters spaced 50 ft apart. Because each completion had the same perforation cluster spacing of 50 ft, there were roughly the same number of clusters in each well, and this design formed a basis for comparison.
Fig. 1 shows the results of well productivity vs. stage length. The data shows all the wells for each operator that had a first production date between January 2010 and January 2013. The 12-month cumulative gas production was averaged for all the wells that were brought online in each month and was compared with the average gross perforated interval per stage for wells completed for each operator.
As expected, Operator B with prime acreage and a constant design achieved consistently good average production per well. Operator A’s initial wells fell well below the average of Operator B, indicating that the former’s well completions were in less productive areas of the play. However, well productivity improved as stage spacing was progressively reduced. Following the initiation of tighter stage spacing, the average production per well for Operator A was 46% higher than earlier. Over the 23-month period of reduced stage spacing, Operator A’s wells outperformed those of Operator B by 18%.
The upward deflection in the wells’ performance for Operator A closely matches their progressively tightening stage spacing. This indicates that by dividing the total number of clusters for the well into more isolated stages and thus stimulating fewer clusters at a time, higher cluster completion efficiency was achieved for the whole wellbore. The data is also in line with the original thesis presented that more transverse fractures per well is beneficial in this type of formation.
By reducing the stage lengths and increasing the number of isolated stages per well, it is possible to achieve effectively higher cluster efficiency for the whole lateral. The most common method, as Operator A demonstrated, is to keep cluster spacing constant but decrease the number of clusters per stage. This increases the number of stages per well but treats a smaller portion of clusters at a time. Because of the higher completion costs and additional completion time needed for additional wellbore interventions and stages, there is a point of diminishing returns.
What Is Perforation Cluster Efficiency?
Many of the progressive completion strategies, such as increased stages per lateral foot, were carried over to other shale plays, including the Eagle Ford shale. With these highly optimized completion designs, it becomes essential to effectively stimulate all the clusters designed for each stage for the best fracture coverage and optimal drainage.
There are two types of perforation cluster efficiency: cluster completion efficiency and cluster production efficiency. Cluster completion efficiency is the percentage of clusters that receive effective stimulation for a given interval. A higher efficiency value would mean a greater number of highly conductive transverse fractures with more uniform half-lengths and fewer gaps in the fracture network along the lateral. A poor cluster completion efficiency could result in gaps in the fracture network and could not only affect flow rates, but also the EUR proportionally to the volume of area missed (Mayerhofer 2006).
Cluster production efficiency is the percentage of clusters effectively contributing to production for a given interval. Attempts to solve this challenge continue to be developed (Buller et al. 2010).
There is a link between both efficiency measurements. Many diagnostic and fracturing technologies are designed to economically improve these factors. Many papers and production logging studies illustrate that on average in horizontal multiple-cluster unconventional resource wells, only 50% to 60% of the wells’ perforation clusters contribute significantly to production.
Challenges to Cluster Completion Efficiency
Stage intervals with multiple clusters present two distinct challenges to completion cluster efficiency. A primary challenge is the heterogeneous formation closure stresses and brittleness factors, such as Young’s modulus and Poisson’s ratio, that can lead to varying fracture initiation pressures along the lateral. Even within a single stage interval, the clusters can encounter different rock properties, which significantly affect the propagation pattern of fractures at the clusters (Shin and Sharma 2014). The lower-stress clusters will take the majority of the stimulation energy initially, and once a fracture is initiated in a lower-stress interval, it can be difficult to break down the higher-stress clusters.
As previously discussed, close cluster spacing can provide enhanced well production through the potential for more conductive transverse fractures. But multiple clusters closely spaced in an isolated interval naturally can cause stress interference that can result in lower cluster efficiency if the completion has not been designed for this type of cluster spacing. A phenomenon known as stress shadowing demonstrated by Cheng (2012) and Shin and Sharma (2014) results in higher induced stresses and, thus, significantly reduced width on a portion of the clusters in a multiple-cluster stage.
Essentially, simultaneous fracture initiations at multiple clusters can act as competing fractures and cause a portion of them to dilate more by receiving more stimulation fluid. Studies have demonstrated that in a homogeneous stress environment, the inner clusters will have their dilation suppressed by the edge fractures pressing in on them.
This increased stress can prevent propagation and reduce the width of a portion of the clusters, theoretically causing them to accept some fluid early but no proppant later in the treatment, or result in some clusters never being broken down. In this scenario, the outer clusters would achieve an unaffected width and normal stress, because of the mechanical interaction that drives the edge fractures to dilate openly to the outer side, where the stress shadow can be released more easily by the far-field stress (Cheng 2012). Therefore, the effect of cluster spacing has a big effect on fracture initiation and propagation.
Intrastage Diversion Method
One successful method to improve cluster efficiency, as demonstrated in multiple studies, is to selectively place isolated clusters in “like” rock or where the properties of the reservoir rock near the wellbore are the same so that the breakdown pressures are similar (Mullen et al. 2010). However, even with homogenous rock properties across a stage interval, the cluster-to-cluster stress interference will exist, as stated previously.
An alternative method is needed that allows for separation of the initially dominant clusters from the pinched group that is being ineffectively stimulated. It would also need to overcome heterogeneity in an interval by enabling elevated bottomhole pressures over initial levels so that fractures can be initiated at the clusters that encounter higher formation stresses.
The concept of an intrastage diversion, developed by Halliburton, uses advancements in diversion technology, such as the BioVert NWB chemical diverter that is biodegradable, self-removing, and residue-free. A more effective diversion material, along with improved delivery methods used in the AccessFrac stimulation service, enables the separation of an isolated stage into a series of treatments that are separated with a diversion spacer. This technique is designed to overcome the cluster completion efficiency challenges to achieve a unique fracture initiation at every perforation cluster.
The bridging and isolation of the dominant clusters taking fluid initially by means of the diversion material creates an elevated bottomhole treating pressure. This can facilitate the breakdown of subsequent clusters encountering higher formation stresses and overcome the stress shadow placed in the interval by the initial dominant fractures. The multitreatment approach to a stage also increases the chance of creating a unique fracture at every perforation cluster, rather than having them all grow into one.
The incorporation of an intrastage diversion on some or all the stages of a well is applicable to every plug-and-perf completion. It can aid in reaching higher completion cluster efficiencies and provide a more complete fracture network with more access points to the wellbore. Field results are showing improved initial production (IP) and longer sustained producing rates, thereby indicating higher recovery factors.
Case Study: Intrastage Diversion
The example plot shown in Fig. 2 is an Eagle Ford shale stage with six 1-ft clusters, spaced 45 ft apart, which incorporated a series of two equally sized treatments separated by a 200-lbm biodegradable diverter drop intrastage. A pressure change of 1,200 psi was observed when the diverter reached the interval, and good breakdown responses occurred following diversion, during the rate establishment on the second treatment. Effective diversion and fractures being created at new clusters can be inferred by the bottomhole diverter response, breakdown signatures, and bottomhole treating pressure analysis.
The intrastage diversion technique has been applied in every major unconventional basin in North America. And its use is growing in these settings and around the world, for example, with applications already completed in Argentina’s Vaca Muerta shale and in Mexico’s Eagle Ford shale. The recent advancements in diversion capability are also providing secondary benefits by enabling ever increasing lengths of laterals and numbers of clusters to be fracture stimulated with limited wellbore interventions, such as isolation plugs, or enabling the complete elimination of wellbore intervention during fracturing.
Case History: Treatment Interval Length
As an example of reducing isolation plugs and increasing treatment interval lengths, one operator in the Haynesville shale recently tested the limits on stage and cluster spacing. In a trial, half of the designed 26 stages in each of two wells were combined in pairs using an intrastage diversion that eliminated a plug for each of the combined intervals. The diversion stages combined two traditional treatment intervals for a total of six 1-ft clusters, spaced 60 ft apart along a cemented lateral, for a total designed interval length of 356 ft (plug to plug).
From a design perspective, each diversion stage (two stage lengths combined) included a series of two stimulation treatments separated by one diversion spacer with 275 lbm of BioVert NWB diverting agent. The two wells using the intrastage diversion were a part of a four-well pad and were compared with the other two wells, which used a standard completion method. Both of the wells had a total of 78 1-ft clusters with 60-ft spacing. The results were
- Only 18 fracture isolation plugs needed for each diversion well vs. 25 on the standard completions
- A 50% reduction in completion time for both of the laterals using diversion
- Equivalent high IP on all the wells, ~20 MMcf/D (<10% variance from lowest to highest well)
- Substantial wellbore intervention and third-party cost savings
- A 44% reduction in coiled tubing mill-out time, equivalent to almost a day per well
- Buller, D., et al. 2010. Petrophysical Evaluation for Enhancing Hydraulic Stimulation in Horizontal Shale Gas Wells. Paper SPE 132990 presented at the SPE Annual Technical Conference and Exhibition, Florence, Italy, 19‒22 September. http://dx.doi.org/10.2118/132990-MS.
- Cheng, Y. 2012. Mechanical Interaction of Multiple Fractures—Exploring Impacts of the Selection of the Spacing/Number of Perforation Clusters on Horizontal Shale Gas Wells. SPE J. 17 (4): 992‒1001. Paper SPE 125769-PA. http://dx.doi.org/10.2118/125769-PA.
- EIA. 2014. Natural Gas Weekly Update. Monthly Dry Shale Gas Production. US DOE, Energy Information Administration (9 January 2014).
- Kaiser, M. and Yu, Y. 2013. Haynesville Update—North Louisiana Gas Shale’s Drilling Decline Precipitous. Oil & Gas J. 111 (12) 62‒67.
- Mayerhofer, M.J., et al. 2006. Integration of Microseismic Fracture Mapping Results With Numerical Fracture Network Production Modeling in the Barnett Shale. Paper SPE 102103 presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, 24‒27 September. http://dx.doi.org/10.2118/102103-MS.
- Mullen, J., Lowry, J.C., and Nwabuoku, K.C. 2010. Lessons Learned Developing the Eagle Ford Shale. Paper SPE 138446 presented at the SPE Tight Gas Conference, San Antonio, Texas, 2‒3 November. http://dx.doi.org/10.2118/138446-MS.
- Shin, D. and Sharma, M. 2014. Factors Controlling the Simultaneous Propagation of Multiple Competing Fractures in a Horizontal Well. Paper SPE 168599 presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, Texas, 4‒6 February. http://dx.doi.org/10.2118/168599-MS.