SPE - 108075 - Horizontal Well Completion and Stimulation Techniques

Copyright 2007, Society of Petroleum Engineers This paper was prepared for presentation at the 2007 SPE Latin American and Caribbean Petroleum Engineering Conference held in Buenos Aires, Argentina, 15t18 April 2007. This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes 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 where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836, Richardson, Texas 75083-3836 U.S.A., fax 01-972-952-9435.

Abstract This paper presents a brief review of the available techniques in the oil and gas industry to complete and stimulate horizontal wells, with emphasis on low permeability carbonates. These techniques can also be applied in non-conventional reservoirs, particularly in tight formations. The paper starts by reviewing the lessons learned in some chalk fields in the North Sea (Dan, Halfdan, South Arne, Valhall and Eldfisk) and in a few pilot projects offshore Brazil (Congro and Enchova). Based on these lessons learned and in the broad literature, the paper devises some considerations on the methodology to select completion and stimulation techniques for horizontal wells. Cased and cemented horizontal wells, in addition to open hole and perforated/slotted liners wells are addressed. The macro aspects of field/area management are stressed as the completion and stimulation drivers. The key parameters for designing, implementing and evaluating horizontal completion and stimulation are presented, emphasizing the most common failures and the controversial aspects. The paper presents a summary of mature field and new scenarios that are candidate to horizontal completion and stimulation in Brazil and other Latin America countries. Then it makes a few comments on the resources available in Latin America to face the mentioned opportunities and related challenges. It is supposed that this brief review will be useful for the low permeability scenarios in Latin America and worldwide. Introduction This paper presents a brief review of the available techniques in the industry to complete and stimulate horizontal wells, with emphasis on low permeability carbonates. The emphasis on low permeability carbonates in this work is justified by the renewed importance of this scenario in Brazil and other Latin America countries. Although it does not focus on

nonconventional reservoirs, such as tight gas, it is related to them as stimulated horizontal completions have been used on their development. This paper focuses fracturing stimulations, also making a few references to matrix stimulation. It also assumes that a horizontal well has already been justified and what is being discussed is its completion and stimulation. The paper starts by reviewing the lessons learned in some chalk fields in the North Sea (Dan, Halfdan, South Arne, Valhall and Eldfisk) and in a few pilot projects offshore Brazil (Congro and Enchova). Then it devises some thought on the methodology used to select completion and stimulation techniques for horizontal wells. It address cased and cemented horizontal wells, in addition to open hole and perforated/slotted liners completions. The key parameters for designing, applying and evaluating horizontal completion and stimulation are presented, underlining the most common failures and the controversial aspects. Completion and Stimulation of North Sea Low-Permeability Carbonates The North Sea low permeability chalks are taken here as a reference due to the outstanding technological evolution verified there in the last decades. Amongst more than ten fields producing from these reservoirs in the North Sea this paper focuses on the Dan, Halfdan, South Arne, Valhall and Eldfisk fields. The main characteristics of these fields are: shallow waters (43 to 69 m), dry completion, high volumes of OOIP (1.6 to 2.9 billions barrel), low permeability carbonates (0.2 md to 10 md) with microfractures in the central areas (10 md to 120 md), high porosities (up to 48%), soft to very soft chalks, small to medium net pays (15 m to 65 m), high oil saturation (up to 97%), and light oils ( about 36o API). What most distinguishes these fields is their over-pressured soft chalks which are subjected to a high degree of compaction under pore pressure depletion, resulting in loss of drilling fluids, rapid production decline, well failures and seafloor subsidence. On the other hand the positive effects of rock compaction as a reservoir drive energy, outweigh by far the negative ones. The recovery factor under primary recovery can be as high as 30%. In general the North Sea chalks experienced an evolution from vertical/directional wells stimulated with acid treatments to multiple fractured horizontal wells.

SPE 108075

Horizontal Well Completion and Stimulation Techniques - A Review with Emphasis on Low Permeability Carbonates Valdo Ferreira Rodrigues, SPE, Luis Fernando Neumann, SPE, Petroleo Brasileiro S.A.; Daniel Torres, SPE, Halliburton Energy Services; Cesar Guimarães, SPE, Schlumberger; Ricardo Sadovski Torres, BJ Services

2 SPE 108075

The Dan and Halfdan Complex1-11 This complex is located in the Danish part of the North Sea Central Graben and is comprised by Dan (2.9 billions barrel OOIP, 12 production platforms, and 108 wells) and Halfdan (1.615 billions barrel OOIP, 4 production platforms, and 62 wells). The Dan field was discovered in 1971 and started production in 1972, making it the oldest producing oil field in the North Sea. It consists of a high porosity (18% to 40%), low permeability (1 md in average) chalk. It was initially developed with fractured vertical/deviated wells. To improve productivity the operator adopted horizontal cemented wells in 1987 associated with acid fracturing stimulation. The stimulation treatment was soon changed to hydraulic (propped) fracturing looking for medium and long term productivity. Amongst the many lessons learned in the Dan field development one can mention: 1. Water injection is fundamental to increase the recovery

factor in the low permeability carbonates; 2. To achieve high water rates in nonfractured chalks it is

necessary to inject above the fracture propagation pressure, and true stimulation is not necessary4,5;

3. Water injection management can be achieved by using sliding sleeve (SS) configuration (opened or closed), downhole fixed chokes, and surface injection pressure control 4,5;

4. Horizontal cemented wells associated with multiple hydraulic fractures, were successfully introduced in 1987;

5. A stimulation and lower completion system that allowed perforating, stimulating, and isolating the treated interval, in one trip, was introduced in 19891;

6. The controlled acid jetting liner (CAJ), introduced in 19986, allowed the extension of the horizontal section beyond coiled tubing reach (used to operate the SSs of the previous system) while permitting efficient acid treatment along the liner;

7. Whenever possible the wellbore orientation must be in the preferred fracture direction in order to avoid or minimize tortuosity in the fracture initiation/propagation process.

8. The perforation interval must be short to avoid the initiation of several fractures, and with high density and large entrance holes to ease the placement of high proppant concentration;

9. The use of injection tests prior to the main treatment can estimate the magnitude of tortuosity;

10. Sand injection tests were successfully used to remove or reduce tortuosity; increasing the treatment rate and/or the fracturing fluid viscosity also helped overcoming tortuosity;

11. It is possible to overcome poor cement isolation filling the channels with sand before pumping the fracturing fluids.

The Halfdan field was discovered in December 1998 by a 9022 m long horizontal well (horizontal section of 6324 m) drilled from a production platform in the Dan field. It went on stream in the same year through a pilot production system. The reservoir is laterally extensive and consist of a homogeneous and isotropic chalk of Maastrichtian and Danian age, with high porosity (25% to 35%), low permeability (0.5

md to 2.0 md), low thickness (5 m to 15 m) and without structural closure8,9. The Halfdan development plan was conceived at the very beginning by making use of the lessons learned in Dan. The water injection secondary recovery was the main determinant in the Halfdan development. Among the several lessons learned in Halfdan one can emphasize the injection fracturing in a densely spaced line drive water flood technology, including the fracture aligned sweep technology (FAST) concept9-11. South Arne 12,13

The South Arne field is located in the northern part of the Danish sector of the North Sea in water depth of 60 m. The installations comprise a combined wellhead, processing, and accommodation platform. The reservoir is an elongated structure, 93 km² in area, 2,800 m deep (deepest chalk in Denmark), and is comprised of high porosity/low permeability chalks from the Maastrichtian and Danian ages. The wells horizontal sections (around 1800 m) navigate through the upper portion of the Tor formation (0.2 md to 4 md). The fractures along the lateral were supposed to cover the Tor formation and at least 75% of the Ekofisk formation (0.0 md to 0.7 md) above to ensure adequate drainage. South Arne is an example of initial total failure that was overcame by proper analysis and actions. This case study12,13 revealed many important conclusions such as: 1. Using proper fracture initiation, the wellbore orientation

did not make difference in South Arne; 2. Pressure depletion can significantly affect fracture

geometry; 3. The combination of G-function and log-log pressure

decline analyses provided more reliable and consistent interpretation of fracture closure pressure;

4. The use of G-function superposition analysis of mini-frac data can assist in the recognition of pressure dependent leakoff due to fissure opening;

5. The fracture initiation should use cross-linked gel, proppant slugs to reduce complexity and sand slugs to plug open natural fractures/fissures. zones with high natural fracture density should be avoided;

6. 100-mesh sand slugs at concentrations of 3 ppg to 4 ppg may control excessive fluid loss due to the activation or dilation of natural fractures;

7. Porosity (25% to 45% in South Arne) is an important determinant in the fracturing design affecting very much the Young’s modulus (less than 500,000 psi to over 2,000,000 psi);

8. Most fracture treatment problems arose in zones with porosity less than 30%;

9. In a case of longitudinal fractures in a poorly cemented liner there were no evidences of additional problems.

Valhall 14-18 The Valhall field, located in the Norwegian sector of the North Sea, was discovered in 1975 and started production in October 1982. The Valhall complex consists of five separate steel platforms for quarters, drilling, wellheads, production, and water injection. In addition the field has two unmanned flank platforms, one in the south and one in the north, both around 6

SPE 108075 3

kilometers from the field centre. The reservoir is comprised of an over-pressured, under-saturated Upper Cretaceous chalk reservoir, which main characteristics are: depth from 2,400 m to 2,600 m, high porosity (35% to 50%), low matrix permeability (1 md to 10 md), total permeability from 1md to 300 md, high oil saturation (92% to 97%), 36 o API oil gravity, average thickness of 25 m with large lateral variation, GOR of 800 scf/STB to 1,400 scf/STB, and two oil bearing formations, Hod and Tor. The Tor formation is the primary reservoir having higher porosity and permeability and providing the greatest reservoir volume. The OOIP is 2.6 billion barrels with an original reserve of 250 million barrels. A major differential in Valhall is the high degree of compaction suffered by its weak chalk under pore pressure depletion. Valhall had about 2,900 psi original over-pressure related to an original 6,550 psi pore pressure @ 2,450 m. It means an effective stress on the chalk matrix of 650 psi equivalent to a burial depth of only 700 m to 800 m17. This explains drilling fluid loss, rapid production decline, well failures and seafloor subsidence. Well failures have been the greatest challenge throughout Valhall’s history. Failure due rock influx is the uncontrolled chalk production, caused by pressure cycling and high drawdown. Production is restored by an expensive coiled tubing cleanout17. Terminal failure is the lack of wellbore access caused by a collapsed casing or liner. The observed subsidence has been around 25 cm/year. The total subsidence was 4.91 m as of March 200317. On the other hand the positive effects (rock compaction contributes more than 50% of the reservoir drive energy) outweigh by far the negative ones20. It is important to note that pressure depletion was not a matter of negligence or lack of resources. Water injection was repeatedly studied for the Valhall field since 1989, always showing no economical gain. In 1996 a multidisciplinary study was created to evaluate waterflooding the Tor formation18. A waterflood project was installed at late 2004. The Valhall completion and stimulation history14-17, started in 1982, is summarized in the figure 1.

Figure 1 Completion phases in the Valhall field

All the Valhall completion/stimulation changes were based on comprehensive studies. The implementation of the cemented and cased horizontal well was based on a reservoir study

carried out in 1991 to 1992. The association between cemented and cased horizontal well and multiple propped fractures came from a completion study accomplished in 1994 to 1995. In 1996 a study for hydraulic fracturing optimization using an implicit reservoir simulator defined multiple fractures geometry (60 m long, 0.55 in. wide), conductivity (5 lbm/sqft to 6 lbm/sqft) and spacing (110 m to 160 m). After that the focus was to optimize operations times and fracture Tip Screen-out. An update in the beginning of 2003 revealed the execution of over 150 proppant fractures in Valhall. In face of a new development in the North and South flank areas, a study was carried out comparing acid versus proppant fracturing15. This study, justified by the expected differences in the reservoir in the flank areas, where the chalk was supposed to be thinner and less soft, was based on: • Comparison of the production history of propped and acid

fractured wells; • Modeling using an in-house analytical reservoir simulator

especially built for proppant fracture wells; • Numerical modeling using a commercial simulator15.

This study provided a Valhall stimulation strategy and a fracturing decision tree. As of 2003 reserves had increased from 250,000 STB to 1 billion STB due to: • Better reservoir description; • Drilling, completion and stimulation evolution; • Complementary development of the flank areas; • Waterflood in the crestal section. The imminent risk of well failures and other challenges led to a great emphasis on well management. This evolved to a dedicated offshore control room monitoring the wells continuously17. A group of onshore petroleum engineers monitors well performance and provides guidelines to the offshore staff. The well opening guidelines are very tight. The usual strategy is to draw the wells down quite fast during clean-up, then slow down gradually, and finally open the well slowly. A horizontal multiple propped fracture in Tor formation, for instance, may take a year to reach full opening at separator pressure17 A Valhall redevelopment is on the way to optimize recovery and deal with subsidence. It includes a new production and hotel platform, planned to be in operation during 2009. Valhall is already considered a field of the future, in addition to employing 4D seismic as a key reservoir management tool. It was also the first Norwegian field to feature a seismic receiver array permanently installed on the seabed. Eldfisk19 The Eldfisk field is located in the southern part of the Norwegian North Sea. It was discovered in 1970 and came on production in 1978. The original development consisted of three facilities. In 1999, a new water injection facility was installed, based on horizontal injection wells. Eldfisk has an OOIP of 2.8 billion barrel of oil (37o API) and 4,600 billion scf gas. The reservoir drive is comprised by 30% from rock-compaction, 30% from gas influx, 20% from oil expansion and 20% from water influx. Eldfisk suffered similar problems

4 SPE 108075

as it neighbors Valhall and Ekofisk: rapid production decline, casing deformation and solids flow. The Tor chalk in Eldfisk is a very soft one with Brinell hardness number (BHN) less than 5 kg/mm². The Upper Ekofisk BHN is generally less than 10 kg/mm². The Eldfisk wells were stimulated with massive, high-rate acidizing, multistage, limited-entry treatments on clustered perforations (2 m to 3 m intervals) separated by 10 m to 12 m. The initial rates varied from 3,000 bbl/d to 4,000 bbl/d, declining to 1,200 bbl/d to 1,500 bbl/d in one year. Production was usually restored by acid restimulations. The production decline was attributed to the closure of natural fractures and to offset-well interference. It was possible to manage solids production by periodic inexpensive acid-wash and separator cleanout operations. However in many wells solids production caused casing deformation resulting in leaks, collapses and erosion, as well as scale deposition. An update in 2001 verified that 75% of Eldfisk wells suffered some degree of casing deformation. In late 1997 four propped fractures were successfully created in a 1,524 m section in the Tor formation of a horizontal well in Eldfisk. It is important to say that despite two previous unsuccessful trials this third attempt was made based on the success of other operators in the North Sea chalk field community. The initial trials were probably unsuccessful due to mechanical problems and low proppant concentrations. The goals defined for placing four propped-fractures in this Eldifisk pilot well were achieved. The well has produced at sustained oil rates significantly greater than all the offset wells in the field. Thus, multiple propped fractures in an Eldfisk horizontal well achieved similar results as other neighboring fields in the North Sea. Campos Basin Low-Permeability Carbonates Completion and Stimulation The Campos Basin was discovered in 1974 and started production by an early production system in August 1977. The history of low permeability carbonates completion and stimulation in Campos Basin, offshore Brazil, is similar the North Sea counterparts. In the eighties the completions were vertical or directional wells stimulated with massive acid fracturing or matrix acidizing. Rapid production decline was the main problem in this period. In 1989 a huge hydraulic fracturing treatment was performed in a subsea directional well, based on a comprehensive study. Unfortunately a previously unidentified fault very close to the well took most of the pumped proppant invalidating this important pilot operation. From 1990 to 1995 a second stimulation period was characterized by low volume matrix treatments (up to 75 gal/ft of 15% HCl) with similar results. A third period started in 1995 with the drilling and completion (with preperforated liner) of the first horizontal well in the Quissamã formation. A standard stimulation treatment included alternating stages of nonviscous acid, viscous acid up to 30 gal/ft of 15% HCl and acid with diversion feature pumped below fracture rates, spotted by the use of dedicated stimulating string or coiled tubing inside the horizontal liner. In August 2001, after realizing that matrix treatments in

noncemented horizontal wells were not presenting sufficient results, the operator launched a research project searching for alternatives. Following the North Sea operator’s experience, the project team realized that the creation of a limited number of discrete fractures, with proper length and conductivity, widely separated and well distributed, could stimulate and provide sustained productivity in horizontal wells. As part of the research project an onshore horizontal cemented well was multifractured in October 2002 as a pilot application. In November 2002 and April 2004 two offshore platform noncemented horizontal wells were multifractured by sand jetting fracturing technology. Propped fractures were created in these three wells. Concluding the research project a subsea noncemented horizontal well was acid multifractured in August 2004 and a subsea cased, cemented and perforated horizontal well was hydraulically multi fractured in November 2005, both in Campos Basin. This paper summarizes these two applications. Acid Multiple Fracturing in a Subsea Horizontal Noncemented Well 20-22 The Macaé/Quissamã is a carbonate formation present in various fields (Bonito, Cherne, Congro, Enchova) in Campos basin offshore Brazil. The Congro field was discovered in 1977, in water depths from 190 m to 600 m, and started production in 1985. The Quissamã carbonate in Congro produces by a unique noncemented horizontal well, 1-RJS-512HA, drilled in 1996. Oil production had declined from the original 201 m3/d (1,264 bpd) to around 100 m3/d to 138 m3/d (629 bpd to 868 bpd). A reservoir management plan elaborated in 2002 raised the opportunity for a stimulation of the 1-RJS-512HA well. The main challenge was how to create multiple fractures in a noncemented subsea well. An extensive study including geology and stratigraphy analysis, well logs interpretation, well testing, nodal analysis, production forecast and economic evaluation achieved the following conclusions20: • Fractures propagation would follow a perpendicular

direction in relation to the wellbore; • A sand jetting fracturing process was the best technique for

this noncemented well; • Production forecast, from a reservoir simulator with

capacity to model multiple fractured horizontal well, and subsequent economic evaluation showed that 5 acid fractures should be placed in this well at 3,915 m, 3,792 m, 3,689 m, 3,580 m, e 3,435 m MD depths (Fig. 2).

Table 1 shows the parameters used in the stimulation study. Focusing only on the fracturing method used, this well

offered additional difficulties such as: • The presence of two abandoned lateral that could cause

high additional leak off reducing fracture growth; • Keeping the jetting tool in position during each fracturing

stage was a first in wells connected to floating rigs; • Fluid loss control during fracturing while pulling the

fracturing string out of the hole and when tripping the production tail string in hole was also a concern;

• Pumping stimulation fluids under high pressure

SPE 108075 5

simultaneously down the tubing and the annulus from the stimulation vessel and the floating rig required modification of the onshore procedures.

Well name 1-RJS-512HA (subsea)) Water depth 331 m Offshore Rig SS-45 (moored floating rig) Location Congro, Campos Basin Formation Macaé/Quissamã 20 Lithology Carbonate Production slotted liner 7” 26# P-110 @ 2,926 m Production casing 9 5/8 in 47# P-110 @ 2,972 m Pore pressure 3,569 psi @ 3,050 m TVD Bottom hole temperature 210 °F Fluid type / API Oil / 28 API Fluid viscosity 1.007 @ 3,000 psi (PVT) Formation thickness 99 m gross / 28 m net Avg Permeability 1 md radial – 0,5 md vertical Avg Porosity 19% to 29% Avg Water saturation 0.1% Sand jetting. tool 4.5 in OD w/8 jets ¼ in Table 1 – Quissamã reservoir parameters

The solutions to overcome these difficulties and other

operational details were presented in previous papers21,23. The only exception to the predicted operational parameters was the abnormal annular flow rate rose from 4.5 bbl/min in the first fracture to 17.0 bbl/min in the fourth fracture instead of the expected 9.0 bbl/min. As a matter of fact this behavior was hypothesized in the planning phase and was explained by the created fractures intercepting one abandoned lateral 12 m distant21,23. The great importance of this operation in a subsea well was to make the used technology available for all the operator’s scenarios

Figure 2 Five acid fractures in the 1-RJS-512HA well

The production results were above the expectations as the oil rate as of January 2006, 29 months after the treatment, was twice the rate before the stimulation. The First Multiple Fractured Subsea Horizontal Cased and Cemented Well23

The Enchova field was discovered in 1976, came on production in 1983 and is on primary recovery until today. Its main reservoir is the Quissamã Member of Macaé Formation, which is a Lower Cretaceous, Albian age carbonate reservoir. The Enchova field is elongated southwest to northeast and lays in water depth from 106 m to 130 m. The Quissamã carbonate is characterized by low to medium porosity (15% to 25%) and low permeability (1 md to 10 md), exhibiting a typical thickness of 40 m to 70 m. The drive mechanism is gas expansion combined with some water influx of a weak aquifer.

The under saturated oil had an initial pressure and temperature of about 3550 psia and 190°F, respectively and a viscosity of 11 cP. As previously stated the previous completion and stimulation methods applied in the Quissamã carbonate of Enchova field did not provide sustained production. So, it was decided to drill and complete a horizontal, cased and cemented well and place a number of discrete fractures, with proper length and conductivity, along its horizontal section.

To design properly this well a project team studied the following aspects: 1. In situ stresses and the choice of the creation of transversal

or longitudinal fractures; 2. Selection of the fracturing treatment, that is, propped

fracture, acid fracture or a combination of both; 3. Cementation of the 7 in liner in the horizontal section; 4. Method to isolate each fracture; 5. Perforation of the well and related issues regarding the

orientation of the well; 6. Best spacing of the fractures and fracture length, height

and width and characteristics of the tip screen out (TSO); 7. Best use of calibration tests to optimize the pad and

propped laden fluid volumes; 8. Methods to control proppant flow back; 9. Techniques to evaluate each fracture without flowing each

one individually. The project and the execution details were presented elsewhere23. The wellbore orientation was determined by reservoir drainage aspects, being aligned with the maximum far field horizontal stress, resulting in longitudinal fractures. The number of fractures, established as seven, was obtained using an in-house semi-analytical model. The well design comprised a 1500 m (4900 ft) horizontal section, 30 m (65 ft) bellow of the top of the Quissamã formation (≈ 2,380 m or 7,810 ft TVD) and 70 m to 90 m (230 ft to 295 ft) above oil/water contact. The fracturing type selection was based on laboratory tests on cores plugs which revealed: • Porosity and oil permeability of 16% to 22% and 0.2 md to

0.7 md, respectively, which were bellow the expectations; • Young's modulus between 2.34×106 psi to 3.34×106 psi; • BHN from 9 to 39; • Results from oedometer cell tests suggested embedment of

1% to 3%; • Acid etch testing showed low conductivity values.

Cement evaluation through sonic, and ultrasonic logs was not straightforward due to the eccentricity of the tools. However they indicated a mud channel in the lower part of the well.

An in-house semi-analytical model predicted the need to obtain fractures conductivities between 3,000 and 4,000 md-ft. Assuming a fracture half-length (xf) of 165 ft (50 m) and 1 md formation permeability a commercial hydraulic fracture simulator showed that 1 lb/ft2 to 2 lb/ft2 of high quality proppant, that is, 20/40 or 16/20 mesh ceramic or bauxite, could provide the required conductivity.

Proppant flowback was controlled either by the use of a liquid resin coated system, or by the use of deformable particles mixed into the proppant.

The SSs of the lower completion deployed could be used to

6 SPE 108075

flow each fracture selectively during cleanup. However, this is a cost prohibitive approach from a floating drilling rig. So it was decided to tag all individual stages of each of the seven fractures with oil and water-soluble tracers. Both types of tracers are relatively inert, stable at reservoir conditions, and can be analyzed to very low parts per billion levels. After trouble with screen-out in fractures 1 and 3, changes and improvements, such as higher rate, larger pad sizes, circulating crosslinked gel and 100-mesh sand to the perforations prior to the main treatment, in addition to proppant slugs and smaller-sized proppant, made the following fractures treatments successful. The well was completed and declared ready for production with six propped fractures and one acid fracture (Fig. 3). All fractures were mechanically isolated with packers and SSs.

Figure 3 Horizontal well EN-52H wellbore schematic with seven fracture stages (35% welbore coverage)

The well came on stream in November 2005 with 4 intervals opened and 3 closed to avoid cross flows due to significant reservoir pressure differences caused by depletion from offset wells. In April 2006 the 3 SSs still closed were opened with difficulty due to the presence of proppant in the horizontal section. After one year on stream the oil rate stabilized in 150 m3/d. In a recent post job study, production history match was only achieved assuming a vertical permeability in the order of micro Darcy. This surprisingly small vertical permeability had to be set to match the quite low water rate and to avoid coning of water from the transition and water zones below. Thus, ignoring proppant flowback harms, the poorer than expected reservoir quality is the cause of the lower than anticipated production24. Methodology to Select Completion and Stimulation Techniques for Horizontal Wells We focus here what most differ the process of selecting completion and stimulation techniques for horizontal wells of the same process for vertical wells. Though unstimulated horizontal wells have been very successful in naturally fractured reservoirs and in reservoirs with gas or water coning problems25 there were many cases of underachieving horizontal wells. The poor performance of a horizontal well may be attributed to one or the combination of the following: 1. Reservoir rock quality poorer than anticipated; 2. Unpredicted shale stringers or poor vertical permeability; 3. Poor positioning of the wellbore in the pay zone - large

standoff from the center and/or missed producing zones in some parts of the lateral;

4. Severe drilling/completion formation damage25,26.

Additionally, the production into a horizontal well is rarely uniform across the whole interval The process of selecting completion and stimulation techniques for horizontal wells is much more complex than in vertical wells. The main reasons for that are: 1. Each stimulated horizontal well causes a great impact in

the reservoir drainage; 2. The completion and stimulation design, execution and

evaluation are much more complex and time consuming. 3. The completion and stimulation costs are much higher,

increasing the capital investment risk. 4. Failures can have great impact in terms of deferred

production and remedial costs. 5. Often the process is not to select a good candidate, but to

find a solution for an underachieving horizontal well. 6. Often the completion was not designed for stimulation

purpose increasing the challenges to design and execute an effective treatment. The first reason represents perhaps the main differential

between completing and stimulating vertical and horizontal wells. A stimulated horizontal well can be seen as many vertical wells, significantly affecting the reservoir drainage. Thus the macro aspects of field/area management and the multidisciplinary approach are much more important. To select the optimum completion/stimulation option the engineer, in close cooperation with geologists, reservoir and production engineers, must consider the macro aspects of the reservoir drainage and the particular aspects of each well. The Halfdan field development, for instance, is a very illustrative case where the waterflood technology adopted, a macro aspect, determined that fracturing must be avoided11.

Because the history of horizontal wells, the asset teams now face a scenario of selecting completion and stimulation techniques for new wells or searching for alternatives to stimulate horizontal wells already completed, often with an inappropriate wellbore configuration. This process is based on the balance between investment, risks and possible rewards. The expected rewards are production anticipation and reserve increase.

With regard to the type of treatment the usual stimulation options comprise matrix treatments, acid fracturing, hydraulic fracturing and mixed fracturing. The matrix treatments, usually acid based, are pumped at rates below fracture closure stress and are efficient only in near wellbore region (penetration of inches to feet). These are damage removal treatments and not true stimulation. In acid fracturing, acid is pumped at pressures above fracture closure stress, creating penetrating fractures. This is true stimulation, potentially increasing production and effective drainage area. The main concerns about acid fracturing are the effective fracture conductivity length and the stability of the etched width (Will the fracture lose conductivity during production?). In propped hydraulic fracturing proppant laden fluid is pumped at pressures above fracture closure stress and at sufficient rates to create the required geometry. The fluid leaks off and the fracture closes on proppant, creating a highly conductive flow path. Although there are many issues that have to be properly considered in hydraulic fracturing design and execution, this technique is more controllable than acid fracturing. Mixed fracturing is the association between propped and acid

SPE 108075 7

fracturing. The selection of fracturing type (acid, propped or mixed) is based on specific laboratory analysis, used to estimate fracture geometry and conductivity along the production time. In the absence of these and other reservoir information it is necessary to assume certain parameters values, increasing the process uncertainty. Concerning fracturing design, following the usual approach in the last decades the stimulation engineer, using a fracturing simulator, verify the technical viability, risks and costs of placing one or more fractures in a vertical well and represents the benefits in terms of productivity index. The engineer can then estimate post-fracture production via analytical methods such as nodal analysis software. General optimization methodology steps are: • Predict well performance for the base case (unfractured)

and for different fracture lengths, conductivities and drainage areas;

• Estimate treatment costs required to create the assumed fractures;

• Perform the economical evaluation based in indicators such as net present value (NPV), rate of return (ROR) or net present value to investment ratio (NPV/IR);

• Based on optimal economical indicators and practical considerations make the final decision.

The beauty of this process is simplicity and speed and it works for vertical well in homogeneous reservoir. However, analytical modeling provides poor production forecasting for wells with complex trajectories and/or in heterogeneous reservoirs. This happens because in general the analytical models do not take into account important parameters and effects such as reservoir layering and geometry, non-Darcy effects along the fracture, fracture conductivity variation along the time, multiphase flow, condensate banking and flow convergence25,27,28. The problem is worse in horizontal wells where modeling the vertical communication provided by the fractures is crucial28, and it is necessary to model several discrete fractures in one well. An accurate modeling of a horizontal well with hydraulic fractures can be achieved with numerical simulation, that permit detailed reservoir properties, layering, pressure-volume-temperature (PVT) properties, field geometry and well geometry to be built-in into the model. It is then possible to model the fluid flow from the matrix to the fractures and along the fractures more realistically. However until recently it was very complex to build the model using the numerical simulators available even without numerical diversion. To solve these problems, new numerical fracture modeling software or templates were recently developed27,28. Using these user-friendly interfaces between fracturing and reservoir simulators, multiple fractured horizontal wells can be modeled faster. It is also easy to change parameters and regenerate the model permitting sensitivity analysis. A good example of stimulation study where there were neither cores nor special logs available is the study for the 1-RJS-512HA well in the Congro Field in Campos Basin20. A reference study on field scale, where all the phases and steps were carefully fulfilled, is the one carried out in 2003 for the Valhall field15. Summarizing, the methodology to select completion and stimulation techniques for horizontal wells comprise:

1. Adopt a macro view considering the field/area drainage; 2. Teamwork is a must; 3. Although a short period may be used in the techniques

selection phase, the final design must consider the well life cycle;

4. Use a template interface between fracturing and reservoir simulators to perform multiple fractured horizontal wells numerical modeling;

5. The well trajectory design must consider many reservoir aspects and the balance between longitudinal and transverse fractures;

6. Provide specific laboratory analysis to support the selection of the stimulation type;

7. For each area/field develop an strategy for completion/stimulation and a decision analysis tree for stimulation selection;

8. Post job evaluation, including production history match, is very important for future treatments.

It should be mentioned that the suggested approach still has to overcome many controversial points. Real teamwork is still a challenge due to the different background and cultural aspects of well engineering and reservoir engineering people. The template interface between fracturing and reservoir simulators is a recent tool. The general simulation process still require premises or provide results that seems to contradict the real world. Completion/Stimulation Alternatives for Multiple Fractures in Horizontal Wells The type of completion determines the sort of treatments that can be made and the productivity that can be realized. When designing a new well or, which is better, several wells in a project, there is a chance of selecting both the completions and the stimulations that are optimum to that area or field. When dealing with an existing well or group of wells the task of selecting the optimum stimulation for the existing completion may be a huge challenge. The stimulation selection is very dependent on the lower completion or formation completion, that can be comprised of cemented liner, noncemented liner with annulus barrier tools, slotted or preperforated liner without annulus barrier tool or simply openhole (barefoot). Thus, completion design is a fundamental step. The premise that horizontal wells would not be stimulated supported the adoption of openhole or uncemented liner (pre or post perforated) as a structural reinforcement. Horizontal wells in high permeability unconsolidated sandstone have been completed with open hole gravel pack (OHGP). More than 200 OHGP have been deployed in Campos Basin horizontal wells as of October 2006, as well as some stand alone and expandable applications. In Brazilian carbonates reservoirs slotted or perforated liner have been used like in many other areas in the world. The available stimulation techniques will be presented by the lower completion type in the next sections. New Completion in Cased and Cemented Well Cased and cemented wells provide the maximum freedom to define and control stimulation (where the fractures will be

8 SPE 108075

placed and how large each fracture should be) and oil and water management. On the other hand this completion presents high costs, is very time consuming, and may also limit the production due to the long blank sections and the possibility of cementing natural fractures and fissures29. In addition, later restimulations are not easy due to internal diameter restrictions. In this kind of completion the stimulation, in particular multiple fractures, can be staged, simultaneously created or placed with alternative techniques.

Staged multiple fractures are those placed in a wellbore in separate stages, or selectively. Since each treatment only creates a single fracture, the fracture models and pressure analysis techniques are applicable, allowing on-site redesign. This enhances the chances of achieving the design objectives and reducing job failures. The first fracture treatment is carried out at the toe of the lateral, and the following treatments are placed sequentially in the heel direction. There are various methods to carry on staged multiple fractures depending on the devices (bridge plug, composite plug, sand plug, chemical packer) used to isolate each created fracture from the next one and the type of intervention unit used (drilling rig, snubbing unit, coiled tubing unit). Some information about these methods was presented previously in this paper.

The more complete methods for staged fracturing are those that provide selectivity both during treatments and production such as the one introduced in the Dan field in 19891. This system is comprised of two basic assemblies run into the well all together. The permanent one is a downhole assembly, consisted of sealing assembly, SSs and spacing tubing, for isolating each interval after stimulation treatment (Fig. 4).

Figure 4 Wellbore schematics with selective system – removing proppant excess

After removing the proppant excess, the sealing assembly

is stabbed into a bridge plug (first fracture) or into the previous packer (second fracture onwards) and then the packer is set. The service assembly is retrieved from the permanent assembly (lower completion) after setting the packer and then moved to the next perforating interval. After perforating it is pulled out of the hole with the fired gun carrier1.

Figure 5 Wellbore schematics with selective system - seven fractures ready to open the sliding sleeves. Following upper completion installation a coiled tubing conveyed tool is run in to open the SS of each stimulated

interval. A new system was developed specially for application on

multiple fractured subsea horizontal cased and cemented wells as well as multi-zone frac pack wells. Benefits of this system are as follows: • Utilizes proven sleeve/shifting tool technology for

isolation; • Utilizes screen designed for optimum production and

proppant flowback control; • All positions are mechanically identified and

hydraulically verified; • Design provides large and unrestricted ID normally equal

to production tubing string. This system was first field tested for a multi-national

operator in Louisiana-USA in 2005, and it has been selected to be used for two multi-national operators in Indonesia. This is an alternative that provides the ability to multi-frac a horizontal cased well or frac pack multi-zone wells in a single trip, with complete zone isolation both before and after treatment and the ability to provide selective or commingled production.

Figure 6 Comparison example chart showing operation hours required for a conventional multi frac operation and the single trip system

The maximum functionality may be achieved with

remotely controlled valves (intelligent or smart completion), that need to be developed for fracturing operations conditions.

Sand jetting fracturing, although developed for noncemented liners (see section on existing wells in noncemented liner) can be applied in cemented liners for staged fracturing.

An alternative method, Casing Conveyed Perforating Technology, that fits the casing string with integral isolation devices (such as flapper valves), perforating guns external to the casing, and methods (such as external hydraulic control lines) to actuate the equipment remotely, was introduced in 1999. Recently this technology has been successfully deployed in horizontal completions in tight gas formation. As many as thirteen discreet interval have been stimulated in a single horizontal intervals, nearly 4,000 ft (1,219 m) long30. However, there are geometric limitations and operational issues to address for offshore horizontal well applications.

When the staged multiple fractures are not justified, an alternative is the simultaneous creation of the multiple

SPE 108075 9

fractures, what saves significant amount of completion cost. The main drawback in this method is that the proppant may be placed in a few dominant fractures (due to variations in in-situ stress and permeability), leaving the rest of intervals not fractured or with much smaller fractures. Simultaneous creation of multiple fractures can be achieve with limited entry technique or bullheading.

The limited entry technique, developed for vertical wells, has also been used in horizontal wells for many years31. Several cases of success have been presented in the literature. Zonal isolation (good cement) and perforation methods are key for limited-entry fracturing success31. Acid soluble cement also improved fracture effectiveness by reducing near-wellbore effects32. Biodegradable ball sealers carried in acid spearhead may be used to breakdown the perforations. After ball sealers dissolution a step down test from the main treatment rate can allow the estimation of the number of opened perforations33. However, this technique presents many drawbacks that are exacerbated in horizontal wells, such as high fluid friction along the horizontal section; high perforation erosion; and large variation of stress and near-wellbore effects along the horizontal section. Thus, there is a significant chance that the majority of the proppant-laden slurry may go into the perforation intervals near the heel, while only a little amount reaches the toe section. In this case, besides having poor or no fractures close to the toe, the fractures close to the heel can be much bigger than designed, wasting time and money and eventually hitting water or gas zones.

A bullhead treatment is a very simple technique that can be applied down production tubing or operational string. It is difficult to predict where and how many fractures will be created. The pad volume can be adjusted on site by usual minifrac pressure decline analysis which estimates the nominal fluid efficiency. But this analysis is more difficult and less reliable. Bullhead fracturing long horizontal perforated intervals in Prudhoe Bay field, Alaska34 showed, both by pressure response and radioactive tracers, that only a few fractures were propagated. However, these remedial fracture treatments resulted in significant production increase, since the wells contained formation damage or missed pay. Bullheading can be assisted with diversion techniques such as sand slugs, rock salt and fibers. Unfortunately these diversion methods can bridge the annulus and cause premature screenout. New Completion in Openhole or Noncemented Liner Openhole or barefoot completion may be an option for some scenarios. Some important parameters for selecting this kind of completion are wellbore stability, geological properties, and reservoir fluids control, as defined by hydraulic units and permeability magnitudes and anisotropy35. Noncemented liners, whether preperforated or slotted, provide some structural reinforcement, while maintaining the advantage of large well exposure. In general these completions designs offer limited flexibility in stimulation, remediation and production control. Stimulations for either completion can be done by sand jetting, limited entry, coiled tubing washing or bullheading.

In 2002, an openhole completion with cased hole functionalities for horizontal wells was introduced to the industry. The completion system enables multiple fractures of an uncemented completion in one pumping treatment. Openhole packers are run on conventional casing to segment the reservoir and maximize reservoir drainage (Fig. 7). Between each set of openhole packers are hydraulically activated SSs. During pumping, balls are dropped from surface to shift each SS open and isolate previously fractured SSs. Following the pumping treatment, the well can be immediately cleaned up to minimize any damage from the fracturing fluid. During cleanup of the well, the balls which shift the sleeves are collected at surface for disposal. A final feature is that the sleeves can be shifted throughout the life of the well to aide in reservoir management. Since 2002, this completion system has been run in over 400 wells, both in land and offshore wells, in a wide range of formations including sandstone, carbonate, shale, coal and chalk. The packers have been run to withstand 8,000 psi differential pressure, been run in a well with BHST of 380 oF and in a variety of wellbore fluids including oil, gas, condensate, H2S and CO2

29.

Figure 7 Noncemented liner with packers and fracports

In the future it may be possible to create several effective (optimum geometry and conductivity) transversal fractures in noncemented completions. Then, as explained by Surjaatmadja et al. the flow of fluids will be as follows36 (Fig. 8).

Figure 8 Flow pattern in a preperforated liner completion (Surjaatmadja, et.al., SPE 103774)

The formation fluids near the wellbore between two

contiguous fractures will flow straight into the wellbore (a). Far from the wellbore, the fluids flow is parallel with the face of the fracture (b). The flow follows a radial pattern through the fracture, and as the radial fluid velocity (c) becomes high, the pressure drop between position (d) and the wellbore also becomes high, which may cause fluids to reenter the formation (e) and from there, directly enter the perforated liner36. Thus this type of completion would add the advantages of openhole

10 SPE 108075

completion with multitransversal fractures. Sand jetting. fracturing could be used to create the

transversal fractures, adding vertical jetted holes to eliminate or to reduce flow constriction in the wellbore connection. Jetted cavity length can exceed 1.52 m (5 ft), providing an effective well diameter over 3.04 m (10 ft). This is significantly greater than effective wellbore diameters achievable with regular perforations (about 8 in length, generating 16 in effective wellbore diameter)36.

An even better improvement could be the use of dual fractures as proposed by the same authors36 (Fig. 9).

Figure 9 Flow in a dual frac completion (Surjaatmadja, et.al., SPE 103774)

This completion would have a more simple flow pattern due to the second fracture. The possibility of fracture redirection allowing the creation of a longitudinal fracture after the transversal fracture, has long been recognized37. Some questions have to be answered such as how much time is allowed between the creation of the two fractures before the relaxation of the rock eliminates the redirection effect. Existing Completion in Cased and Cemented Well Some wells that were cased and cemented and completed without considering the necessity of stimulation can, soon or later, require stimulation due to poor well performance. The difficulty in creating multiple fractures in these wells is aggravated in cases of long perforated intervals not adequately separated from each other. The following techniques can be used: bullhead treatment, sand jetting stimulation, multiple alternating treatment stages. Bullheading was presented in a previous section. Sand jetting fracturing although developed for noncemented wells can be used for cased and cemented wells (see next section). Existing completion in openhole or noncemented liner These are perhaps the most difficult completions for fracture stimulation treatments, since any weak planes such as natural fractures, faults and joints can be fracture initiation sites. Furthermore, a slotted or preperforated liner prevents mechanical isolation through packers and bridge plugs. To stimulate this type of completion one can use a bullhead treatment or a sand jetting. stimulation. To facilitate fracture initiation one can use perforating, breakdown with straddle packers, propellant fracturing and notching with a sand jet. Bullhead treatments have failed in most hydraulic fracturing in this type of completion. The main cause of failures is

excessive leak-off leading to insufficient fracture width and consequent screen-out. Another cause of failure may be slurry dehydration in the slotted liner. Sand jetting fracturing is a effective method to stimulate noncemented liner in various scenarios20-22, 38,39. This method is a combination of three distinct processes: i) cutting a slot and initiating a fracture with sand jetting, ii) fracturing (down the tubing), and iii) co-injection down the annulus. Rather than using mechanical sealing or chemical blocks, this method uses the fluids own dynamic to divert the majority of the fluid flow into a specific point in the horizontal section. This method allows the placement of several discrete fractures, sequentially, from the toe to the heel of the horizontal section. As many texts38,39 described the sand jetting fracturing process, this paper will not discuss it in detail. It should be mentioned that the method has limitations such as erosion of the jet nozzles and tool body due to back splash, fracturing string sticking and excessive loss of fracturing fluid when there is high pressure differential between fractured intervals. Key Aspects for Designing and Implementing Horizontal Completion and Stimulation Amongst the many crucial aspects for design, deployment and evaluation of horizontal completion and stimulation are: 1. In situ stresses, well placement and the choice of the

creation of transversal or longitudinal fractures; 2. Determination of fracture orientation; 3. Treatment type selection; 4. Horizontal section cement execution and evaluation; 5. Definition of the number of fractures and their

characteristics; 6. Perforating design; 7. Proppant flow back control; 8. Calibration tests to optimize the pad and propped laden

fluid volumes; 9. Procedures to opening up the well; 10. Fracturing monitoring; 11. Evaluation of each fracture.

The in situ stresses and the well placement define whether

the fractures will be transversal or longitudinal (Fig. 10). Transverse fractures are created when the wellbore is

oriented perpendicular to the fracture plane (±15º). Longitudinal fractures result when the wellbore is parallel to the fracture plane (±15º)25. When the horizontal section axis is not aligned with the fracture propagation plane the fracture will experience a reorientation from the wellbore to this plane, what can raise near-wellbore issues25,40. Thus, it is essential to know the in situ stresses and the direction of fracture propagation and to make the best effort to align the horizontal section with one of the principal horizontal stresses. It is not always possible because the macro aspects of reservoir drainage come first, when defining well placement orientation. While an undamaged horizontal well provides a nearly infinitely conductive path in the reservoir along the wellbore direction, the transverse fracture enhances the flow in the other direction. Thus, transverse fractures drain a larger section of the reservoir and have the potential for attaining even higher productivity than longitudinal fractures. Many cases histories

SPE 108075 11

have shown the good results of transverse fractures20-22,38,39. On the other hand transverse fractures may present more difficult for proppant placement, and as they only connect the wellbore at one point, flow convergence may raise some issues. These issues are more relevant in high rate wells, formations that produce both liquid and gas, what can cause non-Darcy and convergence effects, and in formations where creating transverse fractures can be difficult due to near-wellbore tortuosity. A study indicated that a high-conductivity proppant could be placed in the fracture to help overcome the additional pressure drop caused by fluid convergence at the borehole. Also, to really make a difference, the fracture must have an adimensional conductivity (Cfd) of at least 525 A longitudinal fracture across a horizontal wellbore behaves similarly to its vertical well counterpart, with one great difference: the horizontal well is usually long and the fracture height is usually short. Thus, a longitudinal fracture is generally not effective for increasing the drainage area, but to anticipate the production. It does provide a conductive path to the wellbore in the vertical direction, which is mostly useful in a laminated formation with very low vertical permeability. Thus, in general the transverse fractures are preferred whenever possible, considering the fractures creation process and the productivity and reserve increase benefits.

Figure 10 Transverse and longitudinal fractures

Fracture orientation can be identified by regional fault

analysis, using logs (sonic, resistivity images), oriented cores associated with other measurements (microfrac, anelastic strain recovery, Kaiser effect), downhole extensometer and nearby wells observation (tiltmeters or microseism instruments)41. In deepwater subsea wells offshore Brazil all these methods, except tiltmeters and microseismic, have been used 20,23,42

The detection of the fracture orientation from cores samples is sometimes difficult. The type of the rock, retrieval problems, and the fact that the associated measurements (microfrac, anelastic strain recovery) is done in a vertical section of the wellbore that is offset from the proposed horizontal section, can lead to uncertainties in determining the true directions of the horizontal stresses. Logs methods require lab and field calibration for adequate accuracy.

There were some special cases where hydraulic fractures

previously created were intercepted by new wells or by hydraulic fracturing a nearby well. These are rare opportunities to check the methods used for fracture direction prediction and to estimate fracture geometry. In the Dan field , in March 1990 a massive propped hydraulic fracture restimulation performed in a well intercepted a nearby well. Thus it was possible to determine exactly the direction of this particular fracture: 5ºNNW3. In 1996 a new producer intercepted the fracture created by an injector well which was injecting above fracture propagation pressure for 6 months6. In Halfdan a design well was used to confirm the propagation direction of a previously created fracture in a nearby well11. Images from infill horizontal wells can also identify hydraulic fractures previously created in offset wells43.

For offshore wells most of the methods are very expensive and time consuming. This aspect associated with the verification that pressure depletion and other artifacts affect the fracture orientation along a horizontal section, make this subject a controversial one. Should one rely on broad geological analysis? Is it worth to invest in direct check with a design well, for instance? Can resistivity image logs while drilling the horizontal provide all the necessary information? Selection of the treatment type focusing on true stimulation includes acid fracturing, hydraulic (propped) fracturing and mixed fracturing (acid plus proppant). As mentioned before the selection of the optimum treatment is based on specific laboratory analysis used to estimate fracture geometry and conductivity along the production time. Experience in the area and practical aspects are also important. Horizontal section cementation – for a cemented horizontal well that will be stimulated a good quality cement sheath is very important. Higher stress levels at perforations, and multiple high pressures cycles during fracturing impose tremendous stresses on cement sheaths. The fracturing fluid during early stages may cleanup cement channels, allowing proppant placement along these channels. There was a case where proppant traveled around 100 feet behind the casing, re-entering the casing, and damaging the work string9. Low quality cement may add tortuosity, which can increase the treatment pressures and can cause premature screen-out. It can also compromise zonal isolation, and, in more adverse situations, damage the fracturing string. Acid treatments can percolate through even small channels. Some keys aspects to cementing horizontal wells are: good mud rheology and drilling practices; good mud removal, dependent upon circulation, centralization, liner movement and effective spacers and washers; good liner centralization and consistent cement properties, in particular no sedimentation, no free water, minimal fluid loss and low gel strength. Field experience has shown that it has been very difficult to achieve cement isolation in horizontal sections. A promising solution is the acid soluble system (ASC), which can provide conditions similar to openhole in the perforating interval region by using an acid pre-flushing. A recent case study revealed improved fracture initiation conditions, while maintaining zonal isolation when applying this new solution44. It has also been difficult to evaluate the cement isolation even using a full set of sonic and ultrasonic tools. It is an area of insufficient knowledge and/or communication.

σv σhma

σhmi

longitudinal

transversal

12 SPE 108075

The number of created fractures and their characteristics should be defined, as mentioned before, using a fracture simulator, a template interface and a reservoir simulator, that allow an accurate and fast modeling of the effects of multiple fractures on a horizontal well productivity and long term production27,28. Fracturing simulators are used to define fracture length, height, width and agressivity of net pressure gain. The known relationship CfD=kfw/kx denotes that, to increase the dimensionless fracture conductivity CfD, it is required to increase the kfw product41. To achieve high kfw in medium to soft formations one should adopt the TSO (Tip Screen Out) technique. It should be mentioned that in fracturing deviated or horizontal wells, the width at the borehole entrance might not enlarge after initiation40.

The template interface into the fracture simulator exports a refined grid with the selected parameter to represent fracture productivity gain to the reservoir simulator. This is used to calculate the predicted the long term production for each number of fractures and their characteristics.

Perforating for fracturing design presents different requirements than perforating for production. This is particularly important in horizontal/deviated wells, in which the connection between perforations and fracture may be limited. Several parameters such as wellbore inclination, wellbore orientation relative to the fracture propagation plane, formation hardness, proppant size and loading are very important. The design should use computer simulation with parameters input at downhole conditions and performance data based on API RP 19B. Improper perforations may cause high pressures in the treatment, premature screenouts and/or hamper production due to choke effect in the wellbore-fracture interface. In horizontal/deviated wells there is a tendency to initiate an axial fracture that soon switches to following the far field stresses. El-Rabaa studied this problem and concluded that to prevent the creation of axial fractures, the perforated interval should not exceed four times the diameter of the wellbore40. For an 8.50 in. wellbore it means 34 in. (0.86 m). To avoid multiple fractures it is better not to exceed two times this diameter. While in some low permeability carbonates this criterion has been relaxed a little (perforated lengths from 1 m to 2 m) 40, a recent study on tight formation confirmed the importance of not exceeding the four times wellbore diameter criterion 44.

The perforation density should be high, which for oriented perforating may require reperforating about the same interval with a small depth difference . An experimental study showed that at least for low permeability rock and high injection rate, the maximum perforation spacing in a row (180º phasing) should not exceed 4 times the perforation diameter. This way fractures initiated at the perforations link up, regardless of wellbore inclination45. The perforation diameter should be 8 to 10 times (minimum of 6) the mean proppant diameter.

For transverse fractures, phasing should cover the full 360º of the wellbore circumference.

For longitudinal fractures it is recommended to perforate only the high and low sides of the wellbore (oriented 180º phasing), with high density46. It is based on many experiments that showed that fracture initiated at the high and low sides of the wellbore, wherever overburden stress was greater than the

other stresses (> 99% of wells). After the stimulation, an extra perforation with 360º coverage may be considered for production reasons46. Some engineers prefer 60º phasing when they don’t know the fracture propagation direction. The perforated interval may be longer (a few meters). It is recommended to design for several separate longitudinal fractures because it is not practical to obtain an entire horizontal perforated well exactly in the direction of the maximum horizontal stress. The perforation cluster distance or fracture spacing has a minimum to avoid the stress shadow phenomenon. Roughly this minimum is the previous fracture height.

Sand jetting may be an alternative for perforating. It has the potential for deep penetration and to cut a hole the full 360º of the wellbore circumference, which may be necessary to surpass the stress concentration area around the wellbore.

Proppant flowback can be a major problem over the life of a fractured horizontal well. Any proppant production becomes a problem because there is no rat hole in which it can fall. There are many methods of proppant flow-back control such as: proppant consolidation, proppant tacking agents, deformable proppants, micro-fibers, heat-sensitive plastic film, forced fracture closure47. What has been most used in low permeability carbonates is: resin-coated proppant (stress bonding or bonding to stress type), forced fracture closure and very careful bean-ups. According to the production platform engineers proppant flow is still common, so the proppant flow back control methods are still being optimized.

Calibration tests to optimize the pad and propped laden fluid volumes, the combination of G-function and square root time pressure decline analyses of both injectivity test, and minifrac, have resulted in a much more reliable and consistent interpretation of fracture closure pressure. The application of G-function and G dP/dG derivative analysis of minifrac data can aid in the identification of pressure dependent leakoff, thus providing an indication of potential fracture treatment problems. It should be mentioned that G-Function analysis is based on 2D models, which according to some engineers is an important drawback. The application of step down test analysis is very useful in identifying near wellbore tortuosity, and in defining the deployment of proppant slugs5,12,14.

Procedures to production management of a fractured horizontal well must be written and rigorously followed since the cleanup of fracturing fluids through all its producing life. The cleanup process in offshore wells is often carried out some time after the conclusion of the stimulation/completion. It may take months in some cases. So, the first well unloading is an important moment in the project life where the multidisciplinary approach may be broken. Bean up procedures, including samples collection, must be rigorously applied. This is very important to manage stimulation evaluation, proppant flow back, fines migration and other issues, mainly in well with transverse fractures. As previously mentioned, Valhall wells were drawn down quickly initially, then slowed down and then finally opened slowly to peak production17. Frequent shutdowns are a problem in offshore wells demanding frequent restarts. These restarts are more problematic in deep water wells due to the conflict between

SPE 108075 13

slow rate bean up, to minimize shocks on the completion, and the risk of hydrate formation or organic scale deposition. It also should be mentioned that the operator’s production sections should maintain the internal stimulation experts and the external partners informed of the well production history. This would be very important for the evaluation of the methods applied and their evolution. It can be said that, unfortunately, it is still a controversial point.

Fracture monitoring has experienced a significant evolution. Real time monitoring through live feed satellite communication allows the consultation of experts real-time. Besides monitoring parameters such as rates, volumes, pressures, proppant loading and chemical concentrations some of them both on the surface and downhole, engineers now have resources such as tiltmeters, microseismic and various wirelines tools48-51.

The evaluation of each fracture should ideally be done selectively. There are several methods to achieve that. However, in the offshore environment, particularly in deepwater, it is a cost prohibitive approach. One given solution is to tag all individual stages of each fracture with oil and/or water-soluble tracers. Both types of tracers are relatively inert, stable at reservoir conditions, and can be analyzed to very low parts per billion levels. Once the well is on production, collected water and oil samples can be analyzed for presence of both tracers.

Scenarios for horizontal completion and stimulation in Brazil and other Latin America countries

Information obtained in the industry events allows one perceive that there are a lot of opportunities for horizontal well completion and stimulation in several countries in Latin America. In Brazil one can say that there is a new phase of renewed interest in low permeability carbonates and other scenarios for horizontal completion and stimulation. This includes a broad program of carbonate revitalization in the mature areas of Campos Basin. A pilot project in the Bonito field was in the drilling phase on February 2007. Two projects were on technical and economical evaluation in the Congro and Cherne fields. New finds in Santos and Campos Basin opened opportunities for horizontal completion and stimulation in oil and gas reservoirs.

Regarding the resources required to face the challenges raised by the new opportunities in Latin America, it can be said that the operators and their partners are ready to provide them. It should be mentioned that there is a personnel shortage and the delivery times are long for materials and equipments. Thus, project coordinators must be aware of these difficulties and take actions to overcome them.

Conclusions and recommendations. 1. Many discoveries and technological innovations in completion and stimulation of North Sea low-permeability carbonates were presented; 2. The first applications of horizontal stimulation methods in subsea wells of Campos basin were presented. 3. The paper presented some considerations on the methodology to select completion and stimulation techniques for horizontal wells;

4. Completion and stimulation methods for cemented liner, noncemented liner with annulus barrier tools, slotted or preperforated liner without annulus barrier tool and openhole were reviewed; 5. The paper also summarized the key aspects for designing, implementing, and evaluating horizontal completion and stimulation; 6. Some controversial or unsatisfactorily solved issues were raised, which optimization can result in significant gains; 7. It was verified that there are many promising scenarios for horizontal completion and stimulation in Brazil and other Latin America countries; 8. It was also verified that personnel shortages and long delivery times for equipments are important issues in implementing horizontal completion and stimulation in Latin America

Acknowledgements The authors thank Petróleo Brasileiro S/A, Halliburton Energy Services, Schlumberger and BJ Services for their support and permission to publish this paper. Special thanks go to Marcos Antonio Rosolen and Paulo Dore Fernandes from Petróleo Brasileiro S/A, Anthony Neil Martin and Frederico Carvalho from BJ Service and Steve Deyo from Schlumberger for manuscript revision. Nomenclature

API = American Petroleum Institute API RP = API recommended practice BHN = Brinell hardness number BHST = bottom hole static temperature GOR = gas oil rate MD = measured depth OOIP = original oil in place SS = sliding sleeve STB = stock tank barrel TVT = total vertical depth 4D = four dimensions

References 1. A.P Damgaard, D.S. Bangert, D.J. Murray, R.P. Rubbo e G.W.

Stout, A Unique Method for Completing Horizontal Wells, SPE 19282, February 1992.

2. Owens, K.A. et al.: “Practical Considerations of Horizontal Well Fracturing in the ‘Danish Chalk’,” SPE 25058, SPE European Petroleum Conference, Cannes, France, 16–18 November, 1992.

3. Weijers, L., De Pater, C.J., Owens, K.A., Kogsbøll, H.H.: “Geometry of Hydraulic Fractures Induced from Horizontal Wellbores,” paper SPE 25049, SPE Production & Facilities, May 1994

4. Larsen F.P., Cowie, D.R., Ovens, J.V.: ”Using 4000 ft Long Induced Fractures to Water Flood the Dan Field,”paper SPE 38558, presented at the 1997 Offshore Europe Conference, held in Aberdeen, Scotland, 9-12 September.

5. Ovens, J.E.V., Larsen, F.P., Cowie, D.R.: “Making Sense of Water Injection Fractures in the Dan Field,” papers SPE 52669 presented at the 1997 SPE Annual Technical Conference and Exhibition, held in San Antonio, Texas, 5-8 October.

6. Hansen, J.H., Nederveen, N.: “Controlled Acid Jet (CAJ) Technique for Effective Single Operation Stimulation of 14,000+ ft Long Reservoir Sections”, paper SPE 78318, presented at the

14 SPE 108075

SPE European Petroleum Conference, Aberdeen, Scotland, 29–31 October, 2002.

7. Kogsboll, H.H., Pitts, M.J., and Owens, K.A., “Effects of Tortuosity inf Fracture Stimulation of Horizontal Wells – A Case Study of the Dan Field”, SPE 26796, Offshore Europe, Aberdeen, Scotland, 7-10 September, 1993.

8. Jacobsen, N.L., Engstrøm, F., Uldall, A., Petersen, N.W.: “Delineation of Hydrodynamic/Geodynamic Trapped Oil in Low Permeabiliy Chalk,” paper SPE 56514 presented at ,1999.

9. Albrechtsen, T.; Andersen, S.J.; Dons, T.; Engstron, F.; Jorgensen, O. and Sorensen, F.W.: “Halfdan: Developing Non-Structurally Trapped Oil in North Sea Chalk”, paper SPE 71322 presented at 2001 SPE Annual Technical Conference and Exhibition, New Orleans, Lousiana, 30 September-3 October 2001.

10. Jorgensen Ole: “Using Flow Induced Stresses for Steering of Injection Fractures”, paper SPE/ISRM 78220, presented at the SPE/ISRM Rock Mechanics Conference, Irving, Texas, 20-23 October 2002.

11. Rod, M.H. and Jorgensen O.: “Injection Fracturing in a Densely Spaced Line Drive Waterflood – The Halfdan Example”, paper SPE 94049 presented at the SPE Europe/EAGE Annual Conference, Madrid, Spain, 13-16 June 2005.

12. Cipolla, C.L., Berntsen, B.A., Moos, H., Ginty, W.R., Jensen, L.: “Case Study of Hydraulic Fracture Completions in Horizontal Wells, South Arne Field Danish North Sea,” paper SPE 64383 presented at the 2000 SPE Asia Pacific Oil and Gas Conference and Exhibition held in Brisbane, Australia, 16-18 October 2000.

13. Cipolla, C.L., Hansen, K. K. and Ginty, W.R.,”Fracture Treatment Design and Execution in Low Porosity Chalk Reservoirs”, SPE 86485, 2004 SPE International Symposium and Exhibition on Formation Damage Control, Lafayette, L.A., USA, 18-20 February 2004.

14. Norris, M.R., Bermtsen, B.A., Myhre, P., Winters, W.J.: Multiple Proppant Fracturing of a Horizontal Wellbore: an integration of two technologies”, paper SPE 36899 presented at the 1996 SPE European Petroleum Conference, Milan, Italy, 22-24 October, 1996.

15. Karen E. Olson, Eivind Olsen, Sami Haida, Alexandre Boulatsel, and Kristian Brekke: “Valhall Field: Horizontal Well Stimulations “Acid vs. Proppant” and Best Practices for Fracture Optimization, paper SPE 84392 presented at SPE Annual Technical Conference and Exhibition, Denver, Colorado, U.S.A., 5-8 October, 2003

16. Norris, M.R., Bernsten, B.A., Skartveit, L., and Teesdale, C., “Multiple Proppant Fracturing of Horizontal Wellbores in a Chalk Formation: Evolving the Process in the Valhall Field,” SPE 50608, SPE European Petroleum Conference, The Hague, Oct. 1998.

17. Olav Barkved, Pete Heavey, Roar Kjelstadli, SPE, Tore Kleppan, and Tron Golder Kristiansen, Valhall Field – Still on Plateau after 20 Years of Production, SPE, BP, paper SPE 83957 presented at the Offshore Europe 2003, Aberdeen, UL, 2-5 September 2003.

18. Martinsen, R., Kjelstadli, R.M., Ross, C., Rostad, H.: ”The Valhall Waterflood Evaluation: A Decision Analysisi Case Study,”, paper SPE 38926 presented at the SPE Annual Technical Conference and Exhibition held in San Antonio, Texas, 5-8 October 1997.

19. Cook, C.C., Brekke, K.: “Productivity Preservation Through Hydraulic Propped Fractures in the Eldfisk North Sea Chalk Field”, paper SPE 73725 presented at the 2002 SPE International Symposium and Exhibition on Formation Damage, Lafayette, Louisiana, 20-21 February 2002. Reedited as SPE 88031.

20. Surjaatmadja, J.B., Gonzales, C.M., Rodrigues, V.F., Neumann, L.F., Rosolen, M.A., Fernandes, P.D., Lima, C., Costa, M.G.F., Silveira, M.A.: “Sucessful Pin-Point Placement of Multiple

Fractures in Highly Deviated Wells in Deep Water Offshore Brazil Fields”, paper SPE 95443, presented at 2005 SPE Annual Technical Conference and Exhibition, Dallas, Texas, U.S.A., 9-12 October 2005.

21. Surjaatmadja, J.B. et al: “Selective Placement of Fractures in Horizontal Wells in Offshore Brazil Demonstrates Effectiveness of Hydrajet Stimulation Process,” paper SPE 90056 presented at the 2004 SPE Annual Technical Conference and Exhibition, Houston, Texas, U.S.A., 26-29 September 2004.

22. Rodrigues, V.F., Neumann, L.F., Rosolen, M.A., Fernandes, P.D., Lima, C., Surjaatmadja, J.B., Gonzalez, C.M., Carneiro, F.: “First Implementation of Hydrajet Fracture Acidizing in Deepwater Offshore Brazil Fields,” paper SPE 94706 presented at the SPE European Formation Damage Conference held in Scheveningen, The Netherlands, 25-27 May 2005.

23. Neumann, L.F., Fernandes, P.D., Rosolen, M.A., Rodrigues, V.F., Silva Neto, J.A., Pedroso, C.A., Mendez, A., Torres, S.A., “Case Study of Multiple Hydraulic Fracture Completion in a Subsea Well, Campos Basin, paper SPE 98277, apresentado no 2005 SPE International Symposium ande Exhibition on Formation Damage Control, Lafayette, LA, 15-17 February 2006.

24. Pinnacle Technologies Inc.: “Post Job Evaluation of the 7-EN-52H-RJS Enchova Field, Offshore Brazil, report prepared for Petroleo Brasileiro SA, January, 2006.

25. Soliman, M.Y., Boonen, P.,: “Review of Fractured Horizontal Wells Technology”, paper SPE 36289 presented at the SPE Asia Oil and Gas Conference and Exhibition held in Kuala lumpur, Malaysia, April 14-16, 1997

26. Soliman, M.Y., Hunt, J.L., Azari, J.: “Fracturing Horizontal Wells in Gas Reservoirs,” paper SPE 59096, SPE Prod. & Facilities 14 (4), November 1990.

27. Sadrpanah, H., Charles, T., and Fulton, J.: “Explicit Simulation of Multiple Fractures in Horizontal Wells, paper SPE 99575, presented at the SPE Europe/EAGE Annual Conference and Exhibition, Vienna, Austria, 12-15 June 2006.

28. Shaoul, J.R., Behr, A., Mtshedlishvili, G.: “Developing a Tool for 3D Reservoir Simulation of Hydraulically Fractured Wells”, paper SPE IPTC 10182, presented at the International Petroleum Technology Conference, Doha, Qatar, 21-23 November 2005.

29. Lohoefer, D., Athans, J.: “New Barnett Shale Horizontal Completion Lowers Cost and Improves Efficiency”, paper SPE 103046 presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, U.S.A., 24-27 September 2006.

30. Rodgerson, J.L., Ruegamer, M.L., Snider, P.M.: “External Casing Perforating Provides Optimal Treatment Coverage in Horizontal Pay,” paper SPE 97175 presented at the 2005 SPE Annual Technical Conference and Exhibition, 9-12 October, Dallas, Texas.

31. McDaniel, B.W., Willet, R.M., Underwood, P.J.: “Limited-Entry Frac Applications on Long Intervals of Highly or Horizontal Wells,” paper SPE 56780 presented at the 1999 SPE Annual Technical Conference and Exhibition, Houston, Texas, Oct. 3-6.

32. Roudakov, V., Rohwer, C.: “Successful Hydraulic Fracturing Techniques in Horzontal Wells for Sandstone Formations in the Permian Basin,” paper SPE 102370 presented at the 2006 SPE Russian Oil and Gas Technical Conference and Exhibition held in Moscow, Russia, 3-6 October 2006

33. Jennings Jr., A.R., Westerman, C.W., Taklock, D.E., Anderson, M.: “ A Systematic Approach to Improved Success With Hydraulic Fracturing Applications,” paper SPE 101837 presented at the SPE Annual Technical Conference and Exhibition held in San Antonio, Texas, U.S.A., 24-27 September 2006.

34. Sanders, G.S., Weng, X., Ferguson, K.R.: “Horizontal Well Fracture Stimulation Experience in Prudhoe Bay, “ paper SPE 38607 presented at the Annual Technical Conference and Exhibition, San Antonio, October 1997.

SPE 108075 15

35. Erwin, M.D., Ogbe, D.O.: “Predicting Openhole Horizontal Completion Success on the North Slope of Alaska,” paper SPE 97121 presented at the 2005 SPE Annual Technical Conference and Exhibition, Dallas, Texas, U.S.A., 9-12 October.

36. J.B. Surjaatmadja, B.W. McDaniel, L.E. East, and J. Pyecroft, Halliburton, Considerations for Future Stimulation Options Is Vital in Deciding Horizontal Well Drilling and Completion Schemes for Production Optimization, paper SPE 103774 presented at First International Oil Conference and Exhibition in Cancun, Mexico, 31 August-2 September 2006.

37. Warpinski, N.R., Branagan, P.T.: ”Altered Stress Fracturing, ”JPT, September, 1988, pp 990-997.

38. McDaniel, B.W., East, L., and Hazzard, V.: “Overview of Stimulation Technology for Horizontal Completions without Cemented Casing in the Lateral,” paper SPE 77825 presented at the 2002 SPE Asia Pacific Oil and Gas Conference and Exhibition, Melbourne, Australia, 8-10 October.

39. Surjaatmadja, J.B. et al: “Effective Stimulation of Multilateral Completions in the James Lime Formation Achieved by Controlled Individual Placement of Numerous Hydraulic Fractures,” paper SPE 82212 presented at the 2003 European Formation Damage Conference, The Hague, The Netherlands, 13-14 May.

40. El Rabaa, W.,: “Experimental Study of Hydraulic Fracture Geometry Initiated From Horizontal Wells, “SPE 19720, SPE Annual Technical Conference and Exhibition, San Antonio, TX, Oct. 8-11, 1989.

41. Economides, M.J. and Nolte, K.G.: Reservoir Stimulation, Third Edition, London, UK, John Wiley & Sons (2000).

42. De Lima, B.B., Viana, R.: “Application of Multiple Techniques to Determine In-Situ Directions in Marlim and Albacora Fields,”, paper SPE 28711 presented at the 1994 SPE Petroleum Conference and Exhibition of Mexico, held in Veracruz, October 10-13.

43. Water, G., Heinze, J., Jackson, R., Ketter, A., Daniels, J., Bentley, D.: “Use of Horizontal Well Image Tools to Optimize Barnett Shale Reservoir Exploitation”, SPE 103202 presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, U.S.A., 24-27 September 2006.

44. Ketter, A.A., Daniels, J.L., Heinze, J.R., Waters, G.: “A Field Study Optimizing Completion Strategies for Fracture Initiation in Barnett Shale,” paper SPE 103232 presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, U.S.A., 24-27 September 2006.

45. Van de Ketterij, R.G., de Paater, C.J.: “Impact of Perforations on Hydraulic Fracture Tortuosity,” paper SPE 56193, SPE Production & Facilities, 14 (2), May 1999.

46. Kim, C.M., Abass, H.H.: “Hydraulic Fracture Initiation from Horizontal Wellbores: Laboratory Experiments,”, paper presented at the 32nd Symposium on Rock Mechanics, The University of Oklahoma, Norman, Oklahoma, July 10-12, 1991.

47. Daneshy, A.: “Proppant Distribution and Flowback in Off-Balance Hydraulic Fracture,” paper SPE 89889 presented at the 2004 SPE Annual Technical Conference and Exhibition, Houston, 26-29 September

48. Mayerhofer, M.J., Bolander, J.L., Williams, L.I., Pavy, A., Wolhart, S.L.: “Integration of Microseismic Fracture Mapping, Fracture and Production Analysis With Well Interference data to Optimize Fracture Treatments in the Overton Field, East Texas”, paper SPE 95508 presented at the 2005 SPE Annual Technical Conference and Exhibition, Dallas, Texas, U.S.A., 9-12 October 2005.

49. Olson, K.E., O’Keefe, M., Nicholyasen, A., Cameron, D.: “Vahall Field: Microfracture Acquisition Using Wireline-Formation-Tester Tools To Validate Fracture Confinement”, paper SPE 96773, presented at the SPE Annual Technical

Conference and Exhibition, Dallas, Texas, U.S.A., 9-12 October 2005.

50. Rod, M.H., Zyweck, M., Dom-Lopez, D.W., Frederiksen, C.H., Gaucher, E.: “Permanent Microseismic Monitoring System in a Long Horizontal Well”, paper SPE 97075 presented at the SPE Annual Technical Conference and Exhibition, Dallas, Texas, U.S.A., 9-12 October 2005.

51. Minner, W.A., Du, J., Ganong, B.L., Lackey, C.B., Demetrius, S.L., Wright, C.A.: “Rose Field: Surface tilt Mapping Shows Complex Fracture Growth in 2500’ Laterals Completed with Uncemented Liners”, paper SPE 83503 presented at the SPE Western Regional/AAPG Pacific Sector Joint Meeting, Long Beach, California, U.S.A., 19-24 May 2003.