Application of a Lightweight Cement Slurry for the Naturally Fractured Mesa Verde Formations D.B. Posey and D.L. Purvis, BJ Services, SPE Members

Copyright 2004, Society of Petroleum Engineers Inc. This paper was prepared for presentation at the SPE Annual Technical Conference and Exhibition held in Houston, Texas, U.S.A., 2629 September 2004. This paper was selected for presentation by an SPE Program Committee following review of information contained in a proposal 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 a proposal of not more than 300 words; illustrations may not be copied. The proposal must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

Abstract The Mesa Verde formation is highly naturally fractured and varies in depth from 5500 ft to 6200 ft. Low recoverable reserves make wells in this area marginally economical. The lower hole is air drilled and cemented back to the intermediate at 3700 feet. Historically, the production casing has been cemented with a 50/50 Poz/H blend mixed at 13.2 ppg. Bond logs were often marginal and fall back was a problem. This paper reviews the development of a lightweight cement system with the ability to develop high compressive strength to enhance the mechanical seal. This slurry needed to have plating lost circulation material (LCM) to prevent fall back and minimize node buildup due to high fluid loss. Compressive strength, fluid loss and rheological properties for various formulations are presented. The results of over forty jobs are summarized with operation and bond log results reviewed in detail from four typical wells. Introduction The Mesaverde grouping, the largest low permeability producer of gas in the San Juan Basin, was deposited along the Cretaceous Western Interior Seaway during the late Cretaceous period. The Mesaverde group includes the Cliff House, the Menefee, and the Point Lookout sandstone. Gas in place estimates vary but have been estimated as high as 38 trillion cubic feet of gas. 1 It is a naturally fractured reservoir that has a total thickness of approximately 900 feet. The average depth to the top of formation is 5400 feet.2 Typical frac gradients are 0.43 - 0.45 psi per foot. It is a very mature reservoir with initial production dating back to the late 1920s.3

The wells being drilled are on the outer edge of the basin. The average Estimated Ultimate Recovery (EUR) is the range of 0.5 1.5 Bcf. This makes the wells marginally economical. Any additional cost will further reduce the wells ability to produce commercially. Due to pressure depletion, the formation is not able to support a full column of water. A completion and a lightweight cement system have been developed to address these reservoir issues. Area of Interest The area of this study is confined to the northwest San Juan Basin, specifically in 31N-5W to 6W Rio Arriba County, New Mexico. The area consists of wells drilled to the Mesaverde along with wells drilled to the deeper Dakota formation. Drilling Practices The standard drilling practice in this area is to drill the intermediate casing with mud down to approximately 3700 feet. This is cemented to surface. The well is then air drilled to total depth. The target zones are the Mesaverde at approximately 6,100 feet or the Dakota at approximately 8,000 feet. A 4 liner is run from 3600 to TD on wells designed as a single completion. A 5 long string is used for dual completions. Historically these wells were cemented with a 50:50:4 Poz:Class H:Gel blend mixed at a weight of 13.2 pounds per gallon (ppg). In the case of a long string, a lead and tail system was incorporated. The lead system was a 50:50:4 Poz: Class H:Gel mixed at 13.4. The tail system consisted of Class H plus additives. Bond logs were often marginal. Cement fall back often exposed the liner top or the zone of interest. This required numerous squeezes to ensure wellbore integrity prior to stimulation. These remedial treatments cost a considerable amount of time and expense relative to the overall economic life of the well. To reduce remedial expense a cement system was needed that would meet the following criteria:

Density & rheology low enough to keep the

Equivalent Circulating Density (ECD) below 13.0 ppg across the Mesaverde formation.

Compressive strengths above 1500 psi after 24 hours. Fluid loss below 100 ml.

SPE 90489

2 SPE 90489

Low gas permeability Right angle set characteristics

Due to the possibility of encountering corrosive waters from the Morrison located below the Dakota, resistance to corrosive waters was also advantageous.

Slurry Selection Many types of cement and additives were evaluated during the development of the slurry used in this study. Most of these slurries were blends utilizing API Class H cement combined with a percentage of fly ash.

Type F fly ash is a low-cost and readily available byproduct of coal combustion. Commonly known in oilfield operations as pozzolan or poz it is often used as a cement extender. The beneficial properties of pozzolan, a highly reactive silica, combined with cement have been recognized by the construction industry for many years. Laboratory tests have reported significant reductions in gas permeability when pozzolan is added to a cement slurry4. As lime is liberated during the hydration of cement, it reacts with pozzolan to form calcium silicates and aluminates. These by-products increase the density and decrease the permeability of the set cement slurry contributing to compressive strength development. This reaction also reduces the amount of lime present to react with organic acid, thus producing a more acid resistant cement5,6. The partial replacement of cement with pozzolan reduces the over-all heat of hydration, thereby reducing the propensity to develop thermal cracks at lower temperatures7. Pozzolan has a specific gravity of 2.5 as compared to 3.16 for Portland cement. This density difference results in decreased slurry density in pozzolan blends8.

Silica fume was identified as a potentially beneficial additive when combined with pozzolan. The increased rate of hydration due to the presence of silica fume significantly improves compressive strength development. This increased hydration rate is due to the silica fume pozzolanic reaction.9 Various alternatives to Class H cement were also evaluated. ASTM Type III cement has been utilized by the construction industry for many years. Its low level of tricalcium aluminate provides resistance to sulfate as well as corrosives often found in formation waters7. The typical Blaine fineness for Type III cement is 6000 cm2/gram compared to 2000 cm2/gram for API Class H. This increase of approximately 200% in surface area significantly increases the amount of water that can be added to the slurry without producing excessive free water. This in turn reduces the slurry density. Properties Based on significant laboratory testing a slurry based on Type III cement, pozzolan, silica fume and other proprietary additives was developed and optimized for the San Juan Basin. This lightweight slurry is typically mixed at a weight of 12.5 ppg. The rheological properties tested in the lab were very similar to the 50:50:4 Class H slurry. However, in actual field operations the lightweight slurry mixed considerably smoother with less variation in density.

The primary benefit of this system lies in the compressive strength, fluid loss and loss circulation properties. The Class H based system at a temperature of 150 deg F developed 24 and 48 hour compressive strengths of 980 and 1825 psi, respectively. Fluid loss was 490 cc per 30 minutes. The new system also at 150 deg F developed a 24 and 48-hour compressive strength of 1825 and 2425 psi, respectively. Fluid loss was 52 cc per 30 minutes. Chart 1 shows a comparison of these properties. Laboratory testing also indicated the slurry exhibited right angle set characteristics. This property aids in the prevention of gas migration. An improved LCM was added at 4% by weight of cement (BWOC) to control fall back. These temperature conditions are typical for a Mesaverde completion. Slurry test results for the deeper Dakota formation are similar with the exception of compressive strength, which is higher do to the elevated temperatures. As a result of the higher compressive strength of this new slurry, the tail slurry commonly utilized in Dakota completions was eliminated. This simplified the operation as well as eliminating the need for separate transports and batch mixer thus further reducing overall cost.10 Case History #1 The first case history is a 4 liner job with the 50:50 Poz:H slurry. The intermediate section was drilled to 3986 measured depth (MD). This section was cemented to surface. The well was drilled to TD at 6,160. A spacer system consisting of 10 bbls of gel with 5 lbs bentonite per bbl followed by 10 bbls of fresh water was used to coat the annulus ahead of cement. Cement was pumped at 4.5 bbl/min and displaced at 7 bbl/min. Ten barrels of cement was reversed out from the top of the 4 liner.

An acoustic bond log was run 13 days after cementing. Figure 1 shows the bond from TD to 5,750 feet. While bond is evident, the log indicates possible channeling as well as a poor bond to pipe. Figure 2 indicates that the cement fell back to approximately 5,380. The bond from this depth to the top of the liner shows little or poor cement coverage. Exceeding the fracture gradient at that depth most likely caused this cement fall back. Once the formation broke down, the formation continued to take cement until the overbalanced condition in the annulus was lost. In order to fracture the well, a squeeze was required to cover the Cliff House/ Menefee before these zones could be fracture stimulated. Cost of this squeeze work plus rig time was in excess of $15,000. Case History #2 The second case history is a 5 long string utilizing the new lightweight cement system. The intermediate section was set at a depth of 3,662 and cemented to surface. The well was then drilled to TD at 8,147. A spacer system consisting of 20 bbls of gelled water followed by 10 bbls of fresh water was used to coat the annulus ahead of the cement. The cement was

SPE 90489 3

pumped at a rate of 4 bbl/min and displaced at a rate of 5.5 bbl/min. An acoustic bond log was run 12 days after cementing. Figure 3 shows the bond from TD to 7,810 feet. Good bond to pipe is evident from this depth to 2,800 (Figure 4). No remedial squeezes were needed on this well. Case History #3 The third case history is a 4 liner using utilizing the new lightweight cement system. The intermediate section was set at a depth of 3,747 and cemented to surface. The well was then drilled to TD at 6,113. The liner top was set at 3,628. A spacer system consisting of 10 bbls of gelled water followed by 10 bbls of fresh water was used to coat the annulus ahead of the cement. The cement was pumped at rate of 4 bbl/min and displaced at a rate of 5 bbl/min. Twenty barrels of cement was reversed out from the top of the 4 liner.

An acoustic bond log was run 3 days after cementing. Figure 5 shows the bond from TD to 5,850. Good bond to pipe is evident from this depth to liner top (Figure 6). No remedial squeezes were needed on this well. Case History #4 The forth case history is a 4 liner using utilizing the new lightweight cement system. The intermediate section was set at a depth of 3,747 and cemented to surface. The well was then drilled to TD at 6,097. The liner top was set at 2,681. A spacer system consisting of 10 bbls of gelled water was used to coat the annulus ahead of the cement. The cement was mixed at 12.3 ppg instead of the original 12.5 ppg. It was pumped at rate of 4.5 bbl/min and displaced at a rate of 6 bbl/min. Fifteen barrels of cement were reversed out from the top of the 4 liner.

An acoustic bond log was run 20 days after cementing. Figure 7 shows the bond from TD to 5,835. Good bond to pipe is evident from this depth to liner top (Figure 8). No remedial squeezes were needed on this well. Summary From January 1999 to August 2001, a total of 53 wells were cemented with the 50:50 Poz:H blend. Remedial squeeze work was performed on 13 of those wells, a success ratio of only 75%. Assuming an average cost of $10,000 per squeeze (includes rig time and cement costs), an additional $130,000 was needed to complete these wells. Since the implementation of the lightweight slurry system in August of 2001, forty wells have been cemented. To date no squeeze work has been required. If the same cost and success ratio is applied, this amounts to direct savings of a $100,000 over the previous blend.

Conclusions A new lightweight cement system that has the ability to give high compressive strength, low fluid readings and good lost circulation control has been developed and successfully implemented in the San Juan Basin. The lightweight system is ideal for a naturally fractured environment. After implementing the new system wells has yielded superior bond logs and as a result zero remedial squeeze jobs have had to been performed. Operationally, the system has been easier to pump on location relative to the 50:50 Poz:H system. In the case of the deeper Dakota wells, it has eliminated the need for two systems.

References

1. Crist, T.E. and Boyer, C.M..: A Geological and Coalbed Methane Resources Analysis of the Menefee Formation in the SanJuan Basin, Southwestern Colorado and Northwestern New Mexico SPE 18945 presented at the 1989 Rockcy Mountain Regional Low Permeiability Reservoirs Symposium, Denver Co. (March 6-9)

2. Dutton, Shirley P., Clift, Sigrid J., Hamilton, Douglas S.,

Hamlin, H. Scott, Hentz, Tucker F., Howard, William E., Akhter, M. Saleem, Laubach, Stephen E.: Major Low-Permeability Sandstone Reservoirs in the Continental United States Report of Investigations No. 211

3. Four Corners Geological Society: Oil and Gas Fields of

the Four Corners Area Volume 1 1978

4. Golapudi, U. and Ramakrishnan, V., Islam, M.R.: Improvement of Properties and Microstructure of Oil-Well Cements SPE 26312 submitted June 8, 1993

5. Ostroot, G.W. and Donaldson, A.L.: Sub-Surface Disposal

of Acidic Effluents SPE 3201 presented at the 1970 Evangeline Regional Meeting, Lafayette, La. (Nov 9-10)

6. Blount, C.G., Brady, J.L., Fife, D.M., Gantt, L.L., Heusser,

J.M. and Hightower, C.M.: HCL/HF Acid Resistant Cement Blend: Model Study and Field Application SPE 19541; JPT (February 1991)

7. Beach, H.J., Frawley, F.E., EnDean, H.J. and Yates, D:

Causes and Prevention of Failures in Cement Pipe Linings SPE 2478; JPT (January 1970)

8. Purvis, D.L. and Merritt, J.W.: Economic Completion

Slurries Utilized in Partially Depleted Reservoirs SPE 80942 presented at the 2003 SPE Productions and Operations Symposium, Oklahoma City, Ok., March 22-25

9. Mueller, D.T. and Dillenbeck, R.L.: The Versatility of

Silica Fume as an Oilwell Cement Admixture SPE 21688 presented at the 1991 SPE Productions and Operations Symposium, Oklahoma City, Ok., April 7-9

10. Posey, D.B., Purvis, D.L.: Application of a Light Weight

Slurry for Naturally Fractured Formations, Presented at the 51st Annual Southwestern Petroleum Short Course, Lubbock Texas, April 21-22, 2004

4 SPE 90489

Figure 1 50-50 Blend

3.8

0.98

4.9

0.3

3.1

1.85

0.52

00

1

2

3

4

5

6

Thickening Time (hrs) CompressiveStrength (x1000

psi) 24 hr

Fluid Loss (x100ml/30min)

Free Water (%)

50:50 BlendLightweight Blend

Chart 1 Comparison of Properties

SPE 90489 5

Figure 2 50-50 Blend

Figure 3 Lightweight Blend

Mesa Verde/ Dakota Well

6 SPE 90489

Figure 4 Lightweight Blend

Mesa Verde/ Dakota Well

Figure 5 Mesa Verde Well

SPE 90489 7

Figure 6 Mesa Verde Well

Figure 7 Mesa Verde Well Mixed at 12.3 ppg

8 SPE 90489

Figure 8 Mesa Verde Well Mixed at 12.3 ppg