document

Summary. Dynamic material testing was used to study cement-slurry structure during early setting. The structure is very sensitive to shear strain and, to a lesser degree, to vibrational frequency. Both setting kinetics and cement structure can be affected by hydration conditions, including water content, temperature, and commonly used additives.

Introduction

The rheology of cement slurries is a manifestation of the detailed interaction between the cement particles and water molecules. The properties of hardened cement and concrete depend to a large extent on the chemical reactions and physical processes that occur during the early stages of hydration. Calorimetric and electron microscopy studies of cement hydration indicate an initial stage of rapid formation of a gelatinous hydrate coating around the cement grains within the first few minutes of mixing cement with water. An intermediate stage with a very low reaction rate then follows for several hours (the induction or dormant period). At the end of the induction period, the reaction rate accelerates again as the surface coatings break and hydration products grow away from the surfaces into the space between grains. The cement slurry develops physical strength rapidly at this stage (cement set).

Information on rheological properties of cement up to the onset of setting will facilitate better handling and transportation of cement slurries. This rheological information is important to industries that use cement in construction and casting applications. One of the most critical needs for cement-rheology data exists in the petroleum industry, however, where cements are pumped many petroleum industry, however, where cements are pumped many thousands of feet into the earth to anchor and seal tubular casing to boreholes. Oil- and gaswell cement must remain pumpable for several hours under the set-acceleration effects of high temperature and pressure. Therefore, chemical additives are used to delay thickening, while still allowing the cement to set within a reasonable time to allow resumption of operations. The cost of premature setting can be as high as the redrilling or abandonment of wells. On the other hand, the costs of an overdelayed cement set can be poor cement quality because of invasion by formation fluids, casing poor cement quality because of invasion by formation fluids, casing sinking upon premature release, and the expense of rig time while sufficient set is awaited. Quantitative and detailed rheological in-formation, in addition to the existing empirical tests-e.g., thickening time and gel strength-can help us understand the structure of hydrating cement better and therefore improve slurry design.

Almost all existing rheological data on cement slurries are concerned with the viscous (flow) properties. These data include flow curves (shear rate as a function of shear stress), yield stress, and viscosity (as a function of shear rate). While these data are useful, e.g., during the pumping of cement slurries, they cannot depict the behavior of cement slurries completely. For example, viscous properties do not apply to a quiescent slurry in place in a casing properties do not apply to a quiescent slurry in place in a casing column.

Cement slurries are viscoelastic materials; they exhibit properties characteristic of both elastic solids and viscous fluids. properties characteristic of both elastic solids and viscous fluids. Many phenomena important to oilwell cementing-e.g., gel-strength development and subsequent pressure loss in cement columns (which can allow annular fluid flow with consequently poor zone isolation) -are manifestations of the cement's viscoelastic properties. Only a small number of studies of the viscoelastic properties of cement appear in the literature. They have used a small-amplitude oscillatory technique, and the slurries studied were restricted to freshly mixed cements. Effects of hydrating conditions and additives were not investigated. In this report, a similar dynamic (oscillatory) testing technique was used to study the viscoelastic behavior of cement slurries typical of those used to cement oil and gas wells. One advantage of this technique is that it uses very small amplitudes of oscillation: peak strains on the order of the magnitude of typical cement particle sizes (10 mu m). Cement slurries are very strain-sensitive, and the dependence of cement rheology on strain may be very important in the modeling of oilfield operations.

Material properties of cement slurries that can be determined from this dynamic testing are dynamic elastic modulus, G', and dynamic viscosity 8 . These properties can provide information about the structure of the material being tested. The relative values of G' and 8 can indicate whether the material is more solid-like (elastic) or fluid-like (viscous). For example, a large G' and small 8 indicate a more solid-like than fluid-like material; a small G' and a large 8 indicate the opposite. In ideal cases, G'= 0 for a Newtonian fluid and 8 =0 for an elastic solid. The frequency dependence of G' and 8 can provide further information on the nature of the material structure. Specifically, a monotonic increase in G' with increasing frequency and a corresponding monotonic decrease in 8 indicate a mobile elastic liquid. On the other hand, an elastic modulus independent of frequency indicates a three-dimensional (3D) network structure in the sample.

During the course of this work, dynamic material testing was used to investigate several phenomena associated with cement setting.

1. The effect of vibration on cement structure was investigated by examining the effects of oscillation amplitude and frequency on cement rheology. This is important because casing vibration has been proposed as a means of improving primary cementing.

2. The effect of such hydration conditions as water content and temperature on the cement rheology and kinetics of change of material properties was also investigated.

3. Finally, the effect of additives (dispersants, retarders, and fluidloss additives) on cement rheology and setting kinetics was studied.

Materials and Methods

Dynamic Testing.

SPEPE

P. 543