J. King Saud Univ., Vol. 9, Eng. Sci. (1), pp. 113-128 (AH. 1417/1997) A Laboratory Study of Factors Affecting Primary Cement Sheath Strength Musaed N.J. Al-Awad Petroleum Enginecring Department , Coilege of Engineering, King Saud University, P.O. Box 800, Riyadh 11421, Saudi Arabia {Received 11 November 1995; acecpted for publication 21 May 1996) Abstract, An investigation of factors affecting cement sheath strength has been carried out.. ‘This study performed on hardened (cured for seven days) cement specimens. These specimens were prepared by mixing 35% fresh water by weight of dry Portland cement (class A according to API classification or C 150 type | aecording to ASTM classification) plus various percentages of Thumama sand (5-40% by weight of dry cement). The relationship between cement tensile nd compressive strengths a3 well as cement direct and indirect tensile sirenghs have been established. Factors affecting cement-casing and cement-[ormation shear bond strength were studied. Cement compressive and tensile strengih were correlated 10 each other as well as 10 ‘the coment shear bond strength. It was found that casing surface roughness, casing surface cicaning and casing centralization have enhanced the strength af cement-casing shear bond, whereas the deposition of mud cake on the wall of the simulated borehole has reduced the strength of cement-formation shear bond. Nomenclature A = Cross-sectional area BWOC = By weight of cement D = Diameter of test specimen DF = Drilling fluid D, = Casing outside diameter Dh = Hole size Cc = Constant of cement supporting capability h = Casing length 134 Musaed N.I. Al-Awad L = Casing dead weight Li = Compressional load L2 = Tensile force Rz = Average surface roughness r = Radius of test specimen t = Thickness of test specimen. Oe = Uniaxial tensile strength 12,0; = Direct and indict tensile strength respectively % “= Shear bond strength Introduction Obtaining a successful cement job will remain one of the most important factors that determine the productive life of any well. Oil well cementing is the process of mixing cement with water and pumping the resultant slurry down the hole through the casing and up the annulus. Cementing procedures may be classified into primary and secondary phases, Primary cementing is performed immediately after the casing is ran into the hole. Secondary cementing includes plug-back to another producing zone, phigging a dry hole and formation squeeze cementing. The main goals of cementing operations are [1,2]: i) Torestrict fluid movement between formations and the surface. ii) To provide support for the casing string. iii) To prevent pollution of fresh water formations. iv) To prevent casing corrosion. v) To support the borchole especially through productive intervals. To achieve the above goals, the cement should reach downhole with the same properties Planned in the laboratory. Failure to surround the casing with continuous and strong cement sheath can lead to such problems as annular imigration of formation fluids, casing corrosion or collapse, loss of well control and high remedial cementing costs (3,4). Objectives Several researchers [1-4] have investigated factors affecting primary cement sheath strength, such as cement composition, contamination and displacement rate, pipe rotation, spacer fluid composition, etc. In this study the following factors which affect the cement sheath strength were investigated: i) Interrelationships among cement mechanical properties, such as compressive strength, direct and indirect tensile strengths and cement-casing and cement-formation shear bond strengths. it) The effect of casing surface contamination and the deposition of cake on the wall ofA Laboratory Study of Factor ns a simulated borehole on cement-casing and cement-formation shear bond strength. iii) The effect of casing surface roughness and casing surface contamination on cement- casing shear bond strength. iv) The effect of casing centralization on cement-casing shear bond strength. The fulfilment of the above objectives is very important to maintain stability for the casing string against static stresses encountered due to the dead weight of the casing string as well as against the dynamic stress resulting from the drilling operations, especially drillstring vibration. Testing Materials A series of laboratory tests has been carried out to investigate how cement mechanical properties, casing offset from the hole centre and surface contamination and degree of roughness can affect the stability of the casing in the oil and gas wells. Cement: Locally produced Portland cement (class A according to API classification or C 150 type I according to ASTM classification) was used in this study. This cement has been proven to be suitable for cementing decp oil and gas wells [5]. Sand: Local sand from Thumama area near Riyadh was used to enhunce cement strength. The chemical and x-ray diffraction analysis of this sand have shown that it is mainly composed of Quartz and Feldspar as well as little quantities of Appetite, Zircon, Pyroxene, Amphibole and Anatase. The sieving analysis of this sand is shown in Fig. 1, whereas the effect of sand addition on cement uniaxial compressive strength is presented in Fig. 2. Simulated borehole: Thick-walled cylinders cored from tocal sandstone have been used as simulated boreholes as shown in Fig, 3. The ratio of sample inner to outer diameter was 0.25, Casing roughness: Resin-sand coats of different roughness were used to provide various degrees of pipe roughness. A sensitive linear differential variable transducer connected to a chart recorder was used to measure the average casing surface roughness. Drilling fluids; Two types of drilling fluids were used in this smdy to produce a contaminated pipe surface including: water-base drilling fluid consisting of 20% Wyoming bentonite by weight of fresh water, and oil-based drilling fluid consisting of 100% crude oil of 32°API.116 Musaed NJ. Al-Awad. 2000 7100 500 300 250 Mesh size, wm Fig, 1. Granulometric analysis of Thumama sand, Cement slurry composition : 35% BWOC Fresh water. 15% BWOC Sand. T days a degree C: 1.0 atm ‘Uniaxial compressive strength, MPa ‘Quantity of sand in cement slurry, SBWOC Fig. 2, Effect of sand addition on cement compressive strength,A Laboratory Study of Factors ..... Test Theory ‘Uniaxial Compressive Strength Brazilian or Indirect Tensile Strength, Direct Tensile Strength {a) Cement-Casing Shear Bond Strength Shear Bond. Strength \(b) Cement-Formation| Fig. 3. Tests conducted to estimate cement mechanical and bonding strengths. 7118 Musaed NJ. Al-Awad Sample preparation: Thumama sand was sieved to remove any contaminants, then was washed with distilled water to remove any dust it may contain. After washing the sand was left to dry at room temperature for 24 hours. Fresh water was used in cement slurry Preparation. After pouring the cement slurry in the molds, it was left to dry at room temperature for the required curing period, Two types of molds were used to prepare cement specimens, 1.5" x 4.5" and 1.5" and 0.75" molds were cut from a PVC pipe for the Preparation of test specimens used in the measurement of cement compressive and indirect tensile strengths, Specimens for direct tensile test were made using dumbbell molds. Experimental Work Testing procedure The following testing procedures have been followed during the course of this study: Uniaxial compressive test: In this test, cylindrical cement samples with a length three (mes its diameter is loaded steadily by a compression testing machine. When the specimen fails the applied load divided by the specimen cross-sectional area is equal to the uniaxial compressive strength of the cement and given by: iu a A For comparison, one inch cubic samples made from the same slurry were tested using a similar procedure as shown in Fig. 3. Indirect (Brazilian) tensile test: The Brazilian test [6] is intended to indirectly measure the unaxia] tensile strength of a cement specimen. When a cement disk is loaded diametrically in compresion as shown in Fig. 3, combined tension and compression is generated in the central part of the specimen. A crack starting in this region propagates parallel to the axis of loading. The tensile strength of a cement specimen with a thickness approximately equal to the specimen radius is calculated from the following formula [7]: oy= 2 . Direct tensile strength test: Direct tensile tests were conducted on dumbbell-shaped cement specimens as shown in Fig. 3. These tests were used to check the validity of data obtained from indirect tensile tests, The direct tension test was conducted on the dumbbell-shaped cement specimens in a tension-compression testing machine. ‘The middie third of the specimen was subjected to tension by attaching metal hooks to the top and bottom third of, the specimen (with enlarged sections) and pulling the sections