Nanoengineered solution for repairing cement leakage in deep wells

Highlights

• Wellbore integrity is considered is one of the major sources of methane emission into atmosphere.

• There are few millions of abandoned wells around the world with leakage problems.

• We proposed a novel nanoengineered cement slurry to address the well leakage problem.

• The proposed cement solution is sealing the wells with superior pumpability, mechanical strength and penetration properties.

Abstract

In addition to the sharp growth of unconventional natural gas wells in the past decade, there are few hundred thousands of abandoned and orphaned wells across the world, potentially contributing to gas emissions. Failure in the barrier and sealing capabilities of the cement and casing are of factors playing role in these leakages. Squeeze job as a remedial treatment is performed to place cement slurry from the surface to the target points to seal narrow spaces behind the casing and/or perforations placed in the casing. In this study, a novel additive for the squeeze-cement is proposed by introducing surface-modified graphite nanoplatelets (GNPs) to carefully adjust rheology and mechanical properties of the prepared cement slurry to remediate some flaws in pre-existing cement sheaths. We seek for better penetration into narrow spaces behind the casing with appropriate thickening time, low fluid-loss, and high final strengths. The process of surface modification is very important to generate a homogeneous cement composite with superior mechanical properties. We develop a chemical method for the surface treatment of GNPs to alter their intrinsically hydrophobic properties to hydrophilic, making them compatible with the aqueous medium of cement. Details of surface modification approach and the concentration of surface-modified nanoparticles are tailored to achieve a desirable slurry for squeeze treatments following API RP 10B-2 (2013) procedures. We believe that these nanoparticles due to their 2D nanostructure not only improve cement movability, but also provide appropriate sealing in the target zones. To assess the performance of the prepared cement slurry to effectively seal narrow targets, narrow-slot tests with 120 microns thickness are conducted to show the supremacy of the proposed solution in sealing narrow spots. We also investigate the effect of adding a commercial superplasticizer on the properties of nanoengineered cement slurry to better resemble field practices. Results show that the presence of surface-modified GNPs provide cement slurry with low rheological properties, zero free fluid, low fluid-loss, and deep penetration which are promising for future field trials.

Introduction

Providing energy while meeting important environmental challenges continuously requires development of novel technologies to achieve better economics, enhanced safety, and low proliferation risk. Using the database gathered by the Department of Environmental Protection (DEP) in the US state of Pennsylvania, Ingraffea (2013) reported the barrier and sealing failure of 211 ($6.2\%$) of 3391 shale gas wells drilled in Pennsylvania in 2011 and 2012. Using the same database, Considine et al. (2013) reported the failure of 1144 environmental violations issues for 3533 wells in Pennsylvania from 2008 to 2011, $8.7\%$ of which were due to the cement and casing leakages. Davies et al. (2014) using the datasets of IOGCC in 2008 reported a demand of $1.43 billion for remedial cementing of 320,202 orphaned oil or gas wells across the United States. Furthermore, depleted oil and gas reservoirs, which are not economically producing further hydrocarbons, are of available infrastructures for the breakdown of costs associated with geologic storage of CO₂ (Vidal-Gilbert et al., 2009; Rubin et al., 2015). However, lack of integrity in the cement sheath behind the casing may endanger the whole process. Considering all the above-mentioned possible impacts on greenhouse gas emissions, introducing an economic and efficient treatment for remedial cementing becomes necessary.

The most successful and economic way for remedial cement jobs is avoiding it by thorough planning, designing, and performing all completion operations. However, severe downhole conditions such as frequent pressure and temperature fluctuations can jeopardize the cement sheath integrity. These fluctuations can be especially intensified for the wells undergoing thermal shocks and hydraulic fracturing treatments (Wang and Dahi Taleghani, 2017; Liu et al., 2018; Jiang and Dahi Taleghani, 2018). These situations may urge remedial cementing to restore a well’s operation. The success of remedial cementing jobs is crucial, as its failure, not only may further damage the well, but also may lead to the total loss of wellbore. Therefore, it becomes necessary to develop a more reliable technology to make these treatments operationally and economically more effective. In general, there are two categories for remedial cementing operations, squeeze cementing and plug cementing (Economides et al., 1998). In this study, we focus on the former approach.

Squeeze cementing is the process of placing cement slurry from surface to downhole to penetrate through narrow spaces behind the casing and/or perforations placed in the casing. Mainly, a squeeze job is conducted to repair a faulty primary cement job, to isolate formation intervals, to alter formation characteristics, and to finally repair some casing problems (Economides et al., 1998). Of some reasons requiring squeeze cement treatments are the presence of micro-annuli at the casing-cement or the cement-formation interfaces, improper displacement of drilling-fluid, and gas influx into the cemented annulus. In this remedial process, enough hydraulic pressure is used to dehydrate or squeeze water out the cement slurry, leaving a filter cake on the surface of the treated area which will harden and seal the voids. Key properties determining the functionality of the squeeze slurry include density, viscosity, thickening time, and fluid-loss (Smith, 1990). Among the above-mentioned properties, fluid-loss is considered as the most important parameter (Beach et al., 1961; Binley et al., 1958). For a cement slurry without fluid-loss control (greater than $1000\text{m}\text{L}/30\text{m}\text{i}\text{n}$), dehydration happens quickly and remains a thick filter cake. Typical squeeze cement slurries possess a moderate fluid-loss, $200-500\text{m}\text{L}/30\text{m}\text{i}\text{n}$, which still results in rapid dehydration but decreases the rate of the filter cake formation (Economides et al., 1998). To achieve the maximum penetration, a slurry with $50\text{m}\text{L}/30\text{m}\text{i}\text{n}$ fluid-loss control is required. At this level of fluid-loss, very slow dehydration occurs, and a moderate filter cake develops which is adequate to seal perforation tunnels and survive hydraulic pressure (Economides et al., 1998). It is noteworthy to mention that extremely low fluid-loss ($20-50\text{m}\text{L}/30\text{m}\text{i}\text{n}$) cement slurries can also be prepared using modern additives, however they may result in too thin filter cake to seal narrow gaps and to stand the pressure exerted on the cement slurry. Therefore, designing a desirable slurry for squeeze jobs by introducing fluid-loss controlling additives would be of interest. Formation permeability is among other factors defining the allowable fluid-loss. For example, for the formations with ultra-low ($<1\text{m}\text{D}$), low, and high ($>100\text{m}\text{D}$) permeabilities, the fluid-loss is required to be $200\text{m}\text{L}/30\text{m}\text{i}\text{n}$, $100-200\text{m}\text{L}/30\text{m}\text{i}\text{n}$, and $35-100\text{m}\text{L}/30\text{m}\text{i}\text{n}$, respectively (Young, 1967). For the case of high-pressure high-temperature wells where deep penetration into narrow channels or microfractures is required, fluid-loss rate under $50\text{m}\text{L}/30\text{m}\text{i}\text{n}$ is needed (Nelson and Guillot, 2006).

Traditionally, squeeze cement is composed of API class-A, G, and H cements in conjunction with fluid-loss control agents to minimize slurry dehydration. In 1980’s, micro-cements were also introduced for squeeze cementing jobs. The diameter of largest particles of this small-particle-size cement must be smaller than $30\text{μ}\text{m}$ (rather lower than $10-15\text{μ}\text{m}$) with the median diameter in the range of $2-3\text{μ}\text{m}$. Note that, the corresponding values for the class G cement are $100-150\text{μ}\text{m}$ for the maximum particle diameter and $15-18\text{μ}\text{m}$ for the median diameter (Nelson and Guillot, 2006). Farkas et al. (1999) developed engineered micro-cement (EMC) slurry by grinding API class cements to prepare $3-10$ times smaller particles than the initial employed cement. They showed that fluid-loss of their provided EMC is smaller than $20\text{m}\text{L}/30\text{m}\text{i}\text{n}$, and it can penetrate ten times deeper into a $120$ microns channel than the conventional micro-cements (Farkas et al., 1999). Prasetyo et al. (2014) also introduced an EMC by engineering the particle size distribution of cement, in such a way that, the largest and smallest diameter of particles were in the range of $30-50\text{μ}\text{m}$ and $10-20\text{n}\text{m}$, respectively. They obtained low rheology properties for their EMC slurry, i.e. the plastic viscosity of $32\text{c}\text{P}$ and the yield point of $5\text{l}\text{b}\text{f}/\left(100{\text{f}\text{t}}^2\right)$.

In this study, we investigate the appropriateness of the cement slurry composed of surface-modified graphite nanoplatelet (GNP) for the squeeze operations. To this end, experiments based on API RP 10B-2 (2013) are conducted. It is notable that the proposed additive is not costly like many nanoadditives introduced in the literature and can produced at competitive costs. In the following in Section 2, we first describe materials, experimental setups and equipment utilized in this study. Section 3 is devoted to briefly describe characteristic properties of GNPs and discuss the methodology that we developed to chemically modify their surface properties. Then, the experimental results are reported in Section 4 and followed by a conclusion in Section 5.