Keywords

Introduction

It has been observed that leakage from oil and gas wells, in 2018, resulted in the U.S. Environmental Protection Agency Greenhouse Gas Inventory (GHGI) reporting 281 kilotons of methane gas being released into the atmosphere [1]. Further studies indicate that the amount of leaking methane could be higher than this number by 60–100 percent [2, 3]. Because of this, one of the most important parts in any wellbore construction or decommission is maintaining the well integrity. This is achieved with the use of cement to hold the casing in place, uphold zonal isolation, prevent circulation loss, and seal the reservoir for plugging and abandonment. A shortcoming of using cement is its brittle nature that can lead to fracturing and failure when subject contamination by the drilling fluid, cyclic high temperatures and pressures the well encounters in subsurface operations [4,5,6,7,8]. To endure these conditions and improve cement performance, the use of graphene nanoplatelets (GNPs), geothermally formed zeolite, and olivine are proposed as additives to the cement slurry to form a stronger and more resilient cement material.

Graphene is made of carbon atoms in a flat two-dimensional hexagonal lattice that link up to create a honeycomb-like sheet structure that is the thinnest and strongest material known at this time [9]. The intrinsic properties of these graphene-based materials, that have high strength, flexibility, surface area, thermal and electrical conductivity although being lightweight, makes them suitable for applications in new technologies and material composites [10, 11]. To help prevent fracturing and failure of the cement, it has been proposed that low percentages (<0.1%) of graphene nanoplatelets can be added to transfer some of its intrinsic mechanical properties to the cement matrix.

Zeolite has seen uses in the cement industry beginning with research on chabazite and clinoptilolite in cement [12,13,14,15]. More recent studies have included property enhancements by the use of zeolites like Ferrierite which had even shown self-healing in geothermal cements [16, 17]. We aim to achieve an improved cement blend with enhanced performance properties by the addition of these two materials.

Olivine, which is abundant in high-temperature igneous rock with forsterite and fayalite endmembers of the (Mg, Fe)2SiO4 olivine series, is a nesosilicate having equal bond strengths in all directions with a specific gravity between 3.27 and 4.37 and hardness of 6.5–7.0 [18]. The purpose of adding olivine into cement is taking the advantages of olivine’s enhanced carbonation rate at high temperature and high pressure which will convert CO2 to stable carbonate minerals mitigating the risk of CO2 leakage and reducing the concerns over long-term monitoring and liability issues [19, 20].

Thus, the objective of the study reported in this paper is to investigate the impact of graphene nanoplatelets, micronized zeolites, and olivine as additives to wellbore cement, understanding the additive impact to mechanical and microstructural properties that translate to the required field performance. Experiments were carried out to evaluate the effects of these three additives on reinforcement on the Young’s modulus, confined compressive strength, and hardness. At the same time, pore-scale measurement and imaging were carried out to trace the micromechanical origins of the observed impacts for these additives on cement properties.

Materials and Methods

Three additives used in this study were purchased from different suppliers. The graphene was derived from refined biomaterials remanence that supplied by CarbonEra Platinum. The olivine sand was ground from ultramafic rock and supplied by Reade Advanced Materials. Zeolites were received from the Trabits group as rocks and were ground into fine micron-sized powders with a particle size of ~95 μm to be added in cement.

All three additives were first examined as received for nanostructure, composition, and chemical stability by methods of scanning electron microscopy (SEM) and Energy-Dispersive X-ray Spectroscopy (EDS). They were then added to Class-H wellbore cement at 0.008, 0.016, and 0.05% by weight of cement (bwoc) for graphene and 5, 15, and 30% bwoc for zeolite, and 5, 15, and 30% bwoc for olivine, along with D-air 5000 at 0.25%bwoc, dispersant CFR-3 at 0.30% bwoc, and bentonite at 2.0%bwoc to make cement at a slurry of density 16.4ppg (1.94 g/cm3) and a water to cement ratio of 0.38.

Cement Curing and Pre-test Preparations

Cement slurry was cast in both 1 inch and 30 mm diameter molds in order to suit dimensional requirements of testing equipment. The molds were covered with cling wrap to prevent any fluid loss by evaporation and set to hydrate at ambient conditions for 24 h. After 24 h, they were demolded and immediately submerged in calcium hydroxide solution of ~pH13 and covered with aluminum foil and sealed with saran wrap. The samples in solution were placed in an environmental chamber and cured at 90 °C and 95% relative humidity to simulate subsurface wellbore conditions. Samples were cured at these conditions for 28 days. Hydrated (set) cement cores were first wet cut using a band saw to trim uneven edges and create flat parallel ends. Samples used for SEM and indentation were polished starting with a 600-grit silicon carbide (SIC) abrasive disc used for grinding to remove initial deformations. After each step, the surfaces are inspected under the microscope to ensure a uniform scratch pattern. Grinding induced deformation is removed using 6 µm diamond suspension on Gold Label polishing cloth and 1 µm diamond suspension on White Label polishing cloth, with Purple-Lube. Samples are sonicated with isopropyl alcohol in a tabletop sonicating bath for 5 min at the end of each step to remove fragmented cement, residual diamond suspension, and colloidal silica. The polished samples are then dried overnight in a drying oven at 50 °C, and further water removal is done prior to SEM, as samples are coated and analyzed.

Scanning Electron Microscopy (SEM) and Energy-Dispersive Spectroscopy (EDS)

A flat and smooth surface for the polished samples of cement was achieved by following the polishing procedure listed below in Sect. 2.2. Polished and fractured cement surfaces were iridium-coated to prevent charging effects before imaging with Thermofisher Scios2 SEM, using secondary electron (SE) at 20 keV and backscatter electron (BSE) at 12 keV, and EDS, using 12 keV and acquisition time of approximately 30 min.

Triaxial Compression Test

In this study, a triaxial test was used to determine the ultimate axial strength and stiffness of the specimen. The triaxial tests are performed in a temperature-controlled Hoek-type triaxial compression cell, which consists of three main parts: axial loading system, confining stress system, and temperature system. The experimental approach entailed applying confining pressure and temperature replicating downhole conditions while simultaneously imposing the increasing axial (deviatoric) load until the specimen fails. The deviatoric loading was controlled by an INSTRON-600DX load frame, up to 600KN. The confining stress was maintained by a high-pressure syringe pump (ISCO-260D), which also allowed precise measurement of the volume change of the specimen associated with a given confining stress up to 70 MPa. The temperature was provided by wrapping the Hoek cell with the heating tape that provided a controlled temperature up to 180 °C. After the desired system temperature was achieved and stabilized, the confining pressure and vertical load were increased to the targeted downhole pressure value, so the specimen was initially loaded isotropically, and then the deviatoric load was increased until the specimen failed. Following ASTM-D7012, the specimen was tested at a constant rate (\(3.3\times {10}^{-6}\,{\rm m}\)/s) so that specimen failed approximately 10–15 min into the testing [21,22,23]. During the test, the load frame recorded the axial position of the top piston. These data were used to derive the axial strain (\({\mathbf{e}}_{\mathbf{A}}\)) of the specimen.

Computed Tomography Scanning

Samples were scanned using a North Star Imaging M-5000 Industrial Computed Tomography (CT) scanner after triaxial compression tests. The samples were received at the National Energy Technology Laboratory submerged in the hydraulic oil that they were tested in, and they were not removed from the oil for scanning. Two-dimensional radiographs were captured with the Feinfocus FXE source at 185 kV and 200 mA, with 12 frames averaged for each radiograph. A 360-degree rotation of the sample was performed with 1440 images captured. These scans were reconstructed with North Star Imaging efX-CT® software and the resultant 3D images had a voxel resolution of (32.9 mm)3. Image segmentation of the open voids and fractures from the cement matrix was performed using pixel segmentation with ilastik [24]. Further post-processing of the images and visualization was performed using ImageJ/FIJI [25]. Analysis of these images is ongoing.

Results

Scanning Electron Microscopy of Hydrated Cement

SEM imaging of hydrated cement polished surfaces was used to observe the microstructure and to view where and how the proposed enhancing materials behaved within the cement matrix.

In graphene cement, the platelets are seen within pore spaces and crevices, protruding from the pore walls. The clusters of platelets appear to be broken up into mono or multilayered platelets during the slurry mixing or hydration process. Figure 1 shows graphene in a micropore, occupying the space in a variety of orientations to strengthen the weakest points in the cement matrix.

Fig. 1
figure 1

a As-received graphene nanoplatelets imaged before adding to cement. Mono- and multilayered platelets stack and cluster together. b Graphene nanoplatelets observed within micro- and nanopore space of cement

Full size image

For ferrierite cement, hollow ferrierite crystals were observed and identified within the cement matrix using EDS showing the elemental composition (Fig. 2).

Fig. 2
figure 2

EDS maps of cement matrix showing area rich in alumino silicate mineral (Zeolite) confirming its presence in cement matrix. High magnification at the grain boundaries needs to be conducted to understand the exact interaction of zeolite with cement

Full size image

Figure 3 shows an open-faced crystal from cutting and polishing the sample. In higher magnification, the crystal appears intact after a surface mechanical test from indentation had been conducted with the cement matrix fracturing around the crystal rather than breaking through it.

Fig. 3
figure 3

a Scanning electron micrograph of a single indent done by the indenter on 5% FER added sample at 1200 × magnification containing ferrierite. b SEM image showing the FER crystal at a magnification of ~ 6500 × next to the indenter mark highlighting failure of cement on indentation with the fracture diverting around the ferrierite crystal and could be an indication of how the overall strength is enhanced in the cement at the macroscale

Full size image

Ultimate Axial Stress in Triaxial Compression Test

To observe how the enhanced cements will perform in their downhole application, various percentages of each additive in cement were mechanically tested, simulating downhole pressure and temperature. For graphene, increasing the graphene content showed an increasing axial strength with the highest in 0.05%. It was also observed that GNP cement had a reduced brittle failure even after passing the maximum stress, indicated by a sharp reduction in axial stress. The highest axial stress of all samples was seen in 5% ferrierite, with a decreasing maximum stress with further increasing the ferrierite percentage. The same trend was observed for the olivine cement, with the 5% olivine sample having the highest ultimate axial stress.

Discussion

The addition of three different additives increases the strength of the cement sample at various levels but the mechanism by which this is accomplished is not completely resolved and the optimal additives percentages are still required the detailed studies. However, the transition from brittle to a stronger and more resilient cement in the triaxial loading test is the most likely the result of mechanisms by positioning of strong GNPs in micropores, bridging of fractures, as well as weak van der Waals forces between platelets allowing them to slide over one another Berman et al. [26]. All of these enable graphene to act as nano springs within the cement matrix which led to reducing the brittle failure behavior (Fig. 4). In our samples, we observed thin layers of received graphene stacked together under SEM images. Imaging of graphene mixed with the cement indicated that most of the graphene is distributed within the pore spaces or weakest parts of the material (Fig. 1), which alters the fracture initiation and propagation during mechanical stress and thus altering the failure behavior. In addition, all ferrierite cement samples showed improved strength with 5% having the greatest ultimate stress for all the samples tested. This possibly achieved by the hollow needle crystals of the ferrierite that are imbedded throughout the cement matrix. Olivine added showed to have reduced the cement brittleness with greater percentages added, but with only 5% resulting in withstanding a higher stress than the baseline of neat cement. The addition of these three additives shows encouraging results to improve the mechanical properties of cement with other potential benefits of reducing fracture propagations, leakage, and chemical attack.

Fig. 4
figure 4

Axial stress–strain curve of neat, graphene, and ferrierite cement at simulated wellbore conditions of 13.7 MPa confining pressure and 90 °C. Neat cement (red) shows a lower failure strength compared to graphene cement (blue) and ferrierite (yellow) and olivine (green). Post-failure CT scans are taken to view the fracture network of for each cement type (right)

Full size image

Conclusion

From this study, the following conclusions were made for the performance of wellbore cement with the addition of graphene nanoplatelets, micronized ferrierite, and micronized olivine to mitigate the risks of well leakage:

  • Graphene nanoplatelets were identified within the walls of the pore structure to reinforce weak points in the cement.

  • Ferrierite crystals were observed to strengthen the cement microstructure.

  • All cases of cement enhancement resulted in increased ultimate stress compared to neat cement when mechanically loaded at simulated downhole conditions of 13.7 MPa and 90 ˚C.

  • The addition of GNPs in 0.05% or less resulted in an increased ultimate stress of 12–25% when tested at simulating conditions.

  • All percentages of GNPs reduced the abrupt brittle failure seen in neat cement.

  • 5% ferrierite and 5% olivine cement resulted in the highest ultimate strength of all samples.