Chemical Shrinkage Properties of Oilfield Cements (includes associated paper 23477 )
- M.E. Chenevert (U. of Texas) | B.K. Shrestha (U. of Texas)
- Document ID
- Society of Petroleum Engineers
- SPE Drilling Engineering
- Publication Date
- March 1991
- Document Type
- Journal Paper
- 37 - 43
- 1991. Society of Petroleum Engineers
- 4.5 Offshore Facilities and Subsea Systems, 5.1.1 Exploration, Development, Structural Geology, 4.3.1 Hydrates, 1.14 Casing and Cementing, 4.3.4 Scale, 5.2 Reservoir Fluid Dynamics, 1.6 Drilling Operations, 2.4.3 Sand/Solids Control, 1.14.3 Cement Formulation (Chemistry, Properties)
- 0 in the last 30 days
- 471 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 5.00|
|SPE Non-Member Price:||USD 35.00|
Summary. Chemical shrinkage was measured at elevated temperatures and pressures for 29 different cements with modified PVT equipment. Results showed that a "dilution" effect exists; i.e., the cements that had the highest yield had the lowest chemical shrinkage. For the 29 tests, the maximum chemical shrinkage measured was 4.6% and the minimum was 1.6%. Tube column tests showed that chemical shrinkage helped to reduce hydrostatic pressures.
In 1900, Le Chatelier reported that Portland cement showed 4.6% chemical shrinkage when it set. Since then, other studies have been performed on this important subject. These studies have shown that chemical shrinkage occurs in two stages. In the first stage, the clinker minerals react chemically with water and shrink as they approach hygral equilibrium. In the second stage, chemical shrinkage continues at a much slower rate as water moves into the hardened paste. Recent studies support the high level of chemical shrinkage reported by Le Chafelier and list total chemical shrinkage values as high as 7%.
Chemical shrinkage is the direct result of the formation of calcium silicate crystals as water reacts chemically with cement clinker crystalline material. Such reactions produce an overall chemical shrinkage (see Fig. 1) that some authors call "hydration volume reduction."
A major factor in cement chemical shrinkage in the first stage is the formation of the calcium silicate crystal "ettringite." During hardening, water adsorbs onto and absorbs within this crystal; as it does, the net volume of the cement decreases. Although ettringite formation does cause expansion, the net volumetric effect is chemical shrinkage because of the formation of many other shrinking-type crystals.
Numerous discussions have focused on exactly where these water movements and subsequent chemical shrinkages occur. It is generally agreed that cement chemical shrinkage is a volume-change phenomenon associated with the removal of free water from the cement mixture.
Beirute and Tragesser and Beirute suggested that some cement compositions expand during hardening. We believe that their results were particular to the "expansion meter" they used, which observed linear movements in one direction only under specific, nonhydrostatic loading conditions.
Bensted showed the importance of water in cement chemical shrinkage. He noted that increasing the water content of the cement slurry increased ettringite formation and the amount of cement chemical shrinkage. Bensted suggested that these effects were caused by the additional water aiding in the dissolution and transport of calcium and sulfate ions to the aluminate and ferrite phases.
From the viewpoint of oilwell application, other cement characteristics specific to chemical shrinkage are of interest, such as the bonding of the cement to the pipe and gelation tendencies of the cement.
Refs. 14 and 15 are particularly interesting. Sabins et al. and Sabins and Sutton discuss the concept that cement chemical shrinkage is of little concern until the cement's gel strength reaches a value that would prevent the transmittal of hydrostatic pressures to the zone of interest. "Delayed-gelation"cements have been developed that provide more time for cement to set before the gel stress advances to a level where hydrostatic pressure cannot be transmitted to the hole bottom.
The idea that the combined effects of increases in cement chemical shrinkage and gel strength can reduce bottomhole pressures (BHP's) is beyond the theoretical stage. Levin et al. showed that under laboratory conditions, hydrostatic pressures can decrease in a column of cement as it sets. In these tests, the cement columns had no outlets for filtration, thus showing that chemical shrinkage alone can reduce hydrostatic pressure. Cooke et al. performed field measurements and showed that a 57% decrease in hydrostatic pressure is possible. They used pressure transducers to measure cement pressures in a 9,000-ft well and showed that the BHP decreased from 7,200 to 4,100 psi in 7.6 hours.
We began our study when we realized the importance of cement chemical shrinkage and that very few data exist for oilwell cements under downhole conditions.
Table 1 lists the types of cement slurries studied, and Tables 2 and 3 list their compositions. Tests were run on 29 cement samples at temperatures up to 435 degrees F and at pressures up to 17,600 psi. These samples were chosen because (1) they represented new, advanced cement mixtures, (2) companion tests showed them to have 3 to 4 hours of pumping time at low temperatures(100 to 250 degrees F) and 4 to 5 hours at high temperatures (300 to 435 degrees F), and (3) several of the ingredients were believed to help reduce chemical shrinkage.
Expansive-type cements have been mentioned in the literature, but they were not included in our study.
All slurries were mixed with Class H cement from the same bag. Likewise, all chemicals and additives came from single containers.
In slurry preparation, weights of the various ingredients were calculated with the following standard assumptions and procedures: (1) the weights of dry additives were based on the weight of cement, (2) the weight of salt was based on the weight of water; and (3) additives less than 2% of the total weight were not included in the calculations.
The cement slurries to be tested were prepared and blended according to specifications described in Sec. 5 of API Spec. 10. A commercially available, standard mixing device was used.
Experimental Apparatus and Procedure
Fig. 2 is a schematic of the equipment used. The equipment includes a high-pressure cell, a high-pressure injection pump with a pressure transducer and a digital pressure gauge, a heating jacket, a vacuum pump, autoclave high-pressure tubings and valves, and an electronic thermostat with a sensing thermocouple.
For each test, a thin-walled lead tube, 1.375 in. in diameter and 4 in. in length, containing the cement slurry was capped with a brass plug and placed inside the high-pressure cell inside the heating jacket. The lead tube was completely surrounded by pressurized mineral oil throughout the test. Heat was generated around the high-pressure cell with heating coils until the cement sample reached the desired temperature. The thermostat sensed the temperature through a thermocouple inserted inside the pressure cell and controlled the cement-slurry temperature to within 1 degrees F.
The high-pressure pump was used to inject mineral oil around the cement sample and to control the pressure during the test. The pressure, displayed by a digital gauge, was sensed by a pressure transducer.
|File Size||1 MB||Number of Pages||8|