Finite Element Approach to Predict the Effect of Annular Pressure Buildup
on Wellbore Materials Siva Rama Krishna Jandhyala and Abhinandan Chiney, Halliburton
Copyright 2014, Offshore Technology Conference
This paper was prepared for presentation at the Offshore Technology Conference Asia held in Kuala Lumpur, Malaysia, 25–28 March 2014.
This paper was selected for presentation by an OTC program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been reviewed by the Offshore Technology Conference and are subject to correction by the author(s). The material does not necessarily reflect any position of the Offshore Technology Conference, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Offshore Technology Conference 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 OTC copyright.
Abstract Annular pressure buildup (APB) is caused by the thermal expansion of wellbore fluids enclosed in a confined annulus experiencing cyclic thermal loads resulting from well operations. In the worst cases, APB can lead to significant damage to the cement sheath, potentially compromising zonal isolation and potentially leading to casing collapse/burst. This work focuses on using an engineered approach based on finite element modeling to analyze the cement sheath response to APB loads for the life of the well. Actual field data of pressure buildup and bleeding rates are used in the analysis. The present study provides a methodology to analyze cement systems for their ability to survive APB conditions. Proper knowledge of the response of wellbore materials (cement and casing) to APB can be helpful for the design of better cement systems and in the selection of appropriate casing materials that can withstand APB conditions. It was observed in this comparative study of two cement systems that the stresses in elastic cement are less than those in conventional cement. A parametric study was also performed to analyze the effect of different cement properties. The results are quantified in terms of stresses and remaining capacities experienced by the cement sheath. This study is based on the fundamental principles of structural mechanics and thus the predictions are more accurate. They can be used to determine the appropriate cement system for a predetermined production/injection rate or can provide a recommendation of production/injection rates for a given cement system so that the risk of damage to the cement sheath and casing failure can be reduced.
Introduction APB refers to a thermal phenomenon wherein fluids trapped in the wellbore build pressure in a constrained space because of their inability to undergo expansion because of heating during well operations. When an unconstrained fluid is heated through a temperature rise of ΔT, the fluid volume increase is described by Eq. 1
V = V0 (1 + αΔT ) ...................................................................................................................................... (1)
where α = the thermal expansion coefficient of the fluid.
Because the fluids in the annulus are trapped, any attempt to change the volumetric amount of fluids leads to an increase in pressure. In such a case, the pressure increase is quantitatively given by
where V = volume after expansion, V0 = initial volume, and BN = fluid compressibility.
If there are any leak paths in the annulus or the possibility exists for thermal annular volume expansion or casing compression and ballooning, the net increase in pressure will be lower than that predicted by Eq. 2. APB is a significant concern in most wells for several reasons: 2 OTC-24837-MS
• When the trapped fluid is in contact with a porous and permeable formation, a natural bleedoff to the formation usually prevents a catastrophic rise in pressure. However, when the rise in pressure is very large, APB might cause formation damage, depending on the local fracture gradient of the rock. • In cases where the annular fluid is trapped between casings/tubings with no scope of bleedoff, a sufficiently large increase in pressure might cause casing collapse or burst. • Pressure buildup can deteriorate the cement sheath in contact with the annular fluid and might cause inner and outer debonding, which could allow fluid influx.
The above scenarios could lead to a loss of zonal isolation, production (Vargo et al. 2002), or, in extreme cases, loss of the well (Nelson 2002). Some of the techniques that the industry has been using over the years to mitigate APB include the following: • Circulating nitrogen gas in the wellbore before pumping annular fluids. This causes gas entrapment in the trapped fluids. Gases, unlike liquids, are compressible in nature. When there is a rise in pressure caused by increasing temperatures, the entrapped gas reduces in volume following its thermodynamic equation of state, providing extra volume for the trapped liquid to expand. This significantly reduces the probability of APB occurring. However, handling gas might not be a possible option always, and this technique also requires accurate calculations of gas volumes. • Crushable materials are often attached to the casing. These materials crush under pressure and provide extra volume for the trapped liquid to expand. This technique, however, works only once, unless the crushing action is reversible and the crushed volume is recovered after pressure reduction. • Use of foamed fluids, which acts in the same way as fluids with entrapped gas. • Use of vacuum insulated tubing. This popular method reduces the rate of heat transfer to the trapped fluid caused by conduction and convection and hence reduces the rise in temperature. • Use of a valve at the surface (for onshore wells) to bleed off pressure intermittently from the annulus. Although this is a very effective technique for onshore wells, bleedoff is often not possible for subsea wells.
Despite these techniques being followed to mitigate APB, it is still a safe practice to design cement and casing to withstand the APB loads that might be exerted in the absence of mitigating techniques or when these techniques do not perform as expected. Such a cement and casing design requires a thorough structural analysis of the wellbore to determine the possible APB loads. Analysis must be performed at critical locations, including the casing, cement-mud interface, and the formation. It should also include the most severe loading scenarios to determine a conservative design. The present work performs one such analysis that meets the geometrical and loading requirements stated previously. The finite element technique is used to perform this analysis for the life of the well, including the loading operations of drilling-running casing- cementing-waiting on cement (WOC)-production. Several theoretical models have been developed to estimate APB loads in wells (Klementich and Jellison 1986; Adams 1991; Oudeman and Kerem 2006). Most of these models take fluid expansion into consideration and predict the increase in pressure. These also study the ballooning of the casings, casing compression, and leakoff of the annular fluids. For prediction of fluid expansion, generally, standard thermodynamic equations of states are used. Besides these theoretical approaches, the industry also observed actual field measurements of pressure increases resulting from APB using sensing technology. These direct measurements encompass the effects of fluid expansion, annular volume expansion, casing compression, ballooning, and fluid leakoff. The present analysis uses the field measured data presented by Oudeman and Kerem (2006). The measured annular pressure vs. time data is used to analyze the effect of the increase in annular pressure on the cement sheath structural integrity for the 13 3/8-in. casing-9 5/8-in. casing to casing annular section. The details of the model are described in the next section.
Model Development As discussed in the preceding section, a finite element analysis is performed to predict the effect of APB on wellbore materials using the wellbore information provided in Oudeman and Kerem (2006). Fig. 1 shows a schematic indicating the actual limits of the wellbore section being analyzed. This schematic is a condensed version of the full-scale wellbore shown in Oudeman and Kerem (2006). The idea of condensing the model is (a) to study only the critical section, which is the mud- cement interface and (b) to generate a model that can computationally be handled in a reasonable amount of time. The model representation is axi-symmetric. OTC-24837-MS 3
Fig. 1—Schematic of the wellbore model used for simulation.
The model is divided into three parts to account for the overburden effects caused by model condensation. A 10 m region at the top named “formation overburden” captures the overburden stresses of the formation, including the water column. The section named “modified cement” captures the hydrostatic pressure caused by the cement column above the analysis section in the outer annulus (formation-cement). The section named “modified mud” captures the hydrostatic pressure caused by the mud column above the analysis section in the inner annulus (13 3/8-in. casing-9 5/8-in. casing). This way, the top 10-m column, which comprises the first part, will compensate for all of the stresses and pressures coming from above. The second part of the analysis section (2388 to 2438 m) is uncemented in the inner annulus, and the third part (2438 to 2488 m) is the cemented portion of the analysis section. The interface between the mud and cement is assumed to be clearly defined, with no cement-mud mixing. This simplifies the effort while defining the analysis problem. Mixing can be assumed in some pre- determined form. This would then require different concentrations of mud-cement samples cured and mechanically tested so that appropriate properties could be used for the mixed zone. The APB loads will be applied on the top of the cement in the inner annulus (at 2438 m) and also on the outer edge of the 9 5/8-in. casing and inner edge of the 13 3/8-in. casing. This is shown in a mesh view in Fig. 2. Here, only portions of the inner annulus, 9 5/8-in. casing, and 13 3/8-in. casing are shown. 4 OTC-24837-MS
Fig. 2—Axisymmetric model of the analysis section showing APB loading locations and direction.
A complete model zoomed at the mud-cement interface is shown in Fig. 3. The model is made of 103,240 linear quadrilateral axisymmetric elements. In Figs. 2 and 3, the gravity direction and model depth are shown along the positive Y axis.
Fig. 3—Axisymmetric model of the analysis section showing all materials, zoomed at the mud-cement interface.
Apart from the APB load, which acts during production, the wellbore also experiences usual loading stages during drilling, running casing, cementing, and WOC. The resultant stress state from all of these stages will be an input to the APB loading condition, and hence simulating these stages is also important. For the sake of analysis, 11-lbm/gal mud and 15.8- OTC-24837-MS 5
lbm/gal cement are used. The mud above the cement in the inner annulus is only used to apply a uniform APB load in all directions, along with its own weight, which varies axially. Because of the application of the large load and the assumption of good bonding between the inner cement and both of the casings, stress artifacts could exist on the top most elements of the inner cement. Hence, the results of stresses are monitored at the second mid element from the top. This location is marked in red in Fig. 3.
Test Cases The test cases selected here analyze the effect of cement properties and casing grades on stresses experienced and also the remaining capacities in cement during APB loading. Remaining capacity is a measure of the proximity of the material to the failure envelope. The more remaining capacity present, the greater the ability of the material to bear additional load before it fails. It is a mathematical representation connecting the present stress state with the stress state at failure. Hence, this representation can be used to compare two materials, even if their failure envelopes are different. Two cement systems—one conventional and the other resilient—are compared in the analysis. The properties of both cements are obtained from detailed mechanical testing. These are designed for another application but are used here for parametric analysis. The conventional cement has a Young’s modulus of 1.5 Mpsi and an average compressive strength of 6,670 psi. The resilient cement has a Young’s modulus of 0.97 Mpsi and an average compressive strength of 1,485 psi. The conventional cement usually undergoes bulk shrinkage of up to 2%. It is important to note that during the comparison, the cement properties are changed from conventional to resilient for only the inner annulus (between 9 5/8-in. casing and 13 3/8- in. casing). The cement in the outer annulus is always assumed as conventional. This is to study the effect of changing only one annular material’s properties at a time. This comparison can easily be extended for the outer annulus, if desired. Casing grade plays an important role in countering the effect of APB. Higher casing grades are often used to counter increased pressure spikes in the system. This study compares two casing grades for the 9 5/8-in. casing—a P110 43.5-lbm/ft and an N80 40 lbm/ft casing.
Results The first comparison is of two cement systems for a given casing design. In the present study, a P110 casing is analyzed with a conventional and a resilient cement system. The second invariant of deviatoric stress in both of these cements is shown in Fig. 4. This stress quantity is a measure of the susceptibility of the cement sheath failure in shear mode. The higher the deviatoric stress, the lower the remaining capacity in cement in shear mode, usually.
Fig. 4—Comparison of second invariant of deviatoric stress in cement between conventional and resilient cements.
Other than shear failure, cement can also fail as a result of excessive tensile stresses or because of debonding at the outer or inner edge of the cement sheath. In the present scenario, outer debonding occurs between the 13 3/8-in. casing and inner cement sheath, and inner debonding occurs between the inner cement sheath and the 9 5/8-in. casing. Fig. 5 shows a comparison of outer debonding stresses between the two cement systems. 6 OTC-24837-MS
Fig. 5—Normal stress at the interface between inner cement and 13 3/8-in. casing for conventional and resilient cement systems.
In addition to the stresses comparison, Figs. 6 and 7 show the remaining capacities of conventional and resilient cement, respectively, in each failure mode to which the cement can be subjected.
Fig. 6—Remaining capacity in a conventional cement sheath at the location marked in Fig. 3. OTC-24837-MS 7
Fig. 7—Remaining capacity in a resilient cement sheath at the location marked in Fig. 3.
In this scenario, the resilient cement has more remaining capacity than the conventional cement in all the four failure modes. The conventional cement undergoes tensile failure during some of the APB events because of the combined effect of cement shrinkage stresses and the APB stresses. The second comparison is that of using different casing grades for the 9 5/8-in. casing. Simple von Mises stress-based analysis is used for analyzing the casing. Fig. 8 shows a comparison of second invariant of deviatoric stress in inner cement for P110 vs. a N80 casing when a resilient cement system is used. As is obvious, no significant effect on cement stresses is observed when using N80 casing instead of P110. Similar behavior has been observed for tensile stresses in cement, both for resilient and conventional cements. However, there was a slight difference in interface stresses at the 9 5/8-in. casing-inner cement interface and the inner cement-13 3/8-in. casing interface. The interface stresses at the 9 5/8-in. casing-inner cement interface are shown in Fig. 9 for the two casing grades using a resilient cement system in the inner annulus.
Fig. 8—Comparison of second invariant of deviatoric stress in cement for two different 9 5/8-in. casings. 8 OTC-24837-MS
Fig. 9—Normal stress at the interface between 9 5/8-in. casing and inner cement for the resilient cement system.
Within the observed casing grades, the cement sheath stresses did not change much with a change in casing grade for the APB loading conditions analyzed. The interface immediate to the 9 5/8-in. casing (i.e., inner cement-9 5/8-in. casing) showed some difference, as expected. However, detailed thermo-structural analysis with different casings must be performed to check the effect of casing grade.
Summary The following conclusions are a result of this study: • This work demonstrates the usage of a scientifically proven engineered approach to analyze the cement sheath response to APB loads for the life of the well, which helps create better cement designs. • For a given casing design, it was observed that resilient cement is relatively better poised to withstand APB loads than conventional cement. The significant reduction (35%) of the Young’s modulus of the resilient cement enables it to easily adapt to (cyclic) load compared to stiff conventional cement. • Correct model generation is important, and it should capture all affected annuli over the critical sections and exert all classical loads, such as drilling, WOC, etc., along with APB loads.
Reference Adams, A. 1991. How to Design for Annulus Fluid Heat-Up. Paper SPE 22871 presented at the SPE Annual Technical Conference and Exhibition, Dallas, Texas, USA, 6–9 October. http://dx.doi.org/10.2118/22871-MS. Klementich, E.F. and Jellison, M.J. 1986. A Service-Life Model for Casing Strings. SPEDE 1 (2): 141–152, SPE-12361-PA. Nelson, K. 2002. AOGCC will Rule on Annular Pressure Management at Prudhoe Bay. Pressure Exceeded Design Burst Rating. Petroleum News 8 (4). Oudeman, P. and Kerem, M. 2006. Transient Behavior of Annular Pressure Buildup in HP/HT Wells. Paper SPE 88735 presented at the Abu Dhabi International Conference and Exhibition, Abu Dhabi, United Arab Emirates, 10–13 October. http://dx.doi.org/10.2118/88735-MS. Vargo Jr. R.F., Payne, M., Faul, R. et al. 2003. Practical and Successful Prevention of Annular Pressure Buildup on the Marlin Project. SPEDC 18 (3): 228–234, SPE-85113-PA.
