Effect of Elevated Temperature on the Microstructure of Metakaolin-Based Geopolymer.

Currently, geopolymer is being considered as a future oil-well cement. For wellbore applications, geopolymers are initially tested at specific temperature conditions. However, an oil-wellbore may experience a sudden increase in temperature which may adversely affect geopolymer systems designed for low to moderate temperature conditions. In this work, the effect of elevated temperatures on the microstructure of the geopolymer was simulated. Metakaolin-based geopolymer systems cured at 163 °F for 48 h were subjected to a temperature ramp of 194 °F and 248 °F for 24 h. X-ray diffraction, Fourier transform infrared spectroscopy, and thermogravimetry analysis techniques were used to study the microstructural changes. The analytical techniques show the formation of new crystalline phases when the geopolymer cured at 163 °F was suddenly exposed to higher temperatures. These crystalline phases, for instance, gobbinsite and anorthite, observed in the microstructure have the potential to cause thermal stress, weaken the system, and ultimately affect the geopolymer's ability to effectively isolate the formation and support the casing.

Introduction

The production of ordinary Portland cement (OPC) is one of the main sources of anthropogenic carbon dioxide (CO2) output, contributing around 5–7%.[1,2] This necessitates the investigation of alternative cement systems with low CO2 emissions.[3] Geopolymer is one of the most used alternative cementitious systems.[4,5] The geopolymer is a type of cementitious material (inorganic polymeric binder) formed when aluminosilicate-rich precursors are dissolved in alkaline solutions and undergo a polycondensation reaction.[5]

The application of geopolymer systems in oil-well cementing is still in the research stage. A review study by Adjei et al.[6] discussed the investigations that have been performed to date. These studies can be classified under four main categories: (1) geopolymer application in acidic and high saline environments,[7−13] (2) geopolymer application in well plug and abandonment,[14−17] (3) compatibility of geopolymer with drilling mud,[18−22] and (4) effect of temperature on geopolymer systems.[23−26] These studies indicated that in comparison to OPC systems, a geopolymer is more durable in aggressive environments, could be an excellent plug material, and is highly compatible with drilling fluid.

In the fourth category (effect of temperature), researchers developed geopolymer systems simulating fixed temperature conditions. It has been reported that the optimum performance of geopolymer systems can be achieved when they are generally cured at 122–176 °F.[27,28] Nasvi et al.[23] observed the optimum temperature within 122 °F to 140 °F. Nasvi et al.[24] explored the feasibility of geopolymer as a binder in oil and gas wells designated for carbon dioxide storage. The study indicated higher strength for a geopolymer system cured at 144 °F, however the authors reported that a slight increase in strength has been observed up to 185 °F. Nasvi et al.[25] also noticed that the carbon dioxide permeability increased with increasing curing temperature. Even though the increase was lower compared to the OPC-based system, such a scenario would decrease the sealing ability of the geopolymer.

The above findings have shown that geopolymer systems would be idle for low to moderate temperature conditions. However, cementitious systems placed in shallow intervals may be subjected to elevated temperatures during the hardening period. For instance, drilling, completions, and workover operations and production techniques such as cyclic steam injection may cause elevation of downhole temperature that would adversely impact the cement sheath.[29−31] In the case of conventional Portland cement systems, studies have shown that above 230 °F, the calcium silicate hydrate gel produced through cement hydration is converted into a more dense phase known as alpha dicalcium silicate hydrate, which has a detrimental effect on the cement system.[32−35] The addition of a silica source modifies the C–S–H phase into more stable phases like tobermorite and Xonotlite.[34]

The macroscopic effect of elevated temperatures has been well documented. The objective of this current study is to understand the microstructural effect of elevated temperatures. X-ray diffraction (XRD), Fourier transform infrared (FTIR) spectroscopy, and thermogravimetry analysis (TGA) were used to better understand the effect of sudden temperature increase on the geopolymerization process of the geopolymer developed using metakaolin produced locally from the calcination of Saudi kaolinitic shale.

Materials and Method

Materials

In this study, powdered kaolinitic shale (PKS) was converted into metakaolin for the synthesis of the geopolymer. The PKS was obtained by grinding mudrocks collected from the Qusaiba Member of Qalibah Formation, Saudi Arabia. Ground granulated blast furnace slag (GGBFS) was combined in the synthesis of the geopolymer. The chemical composition of the raw materials as indicated by the X-ray fluorescence (XRF) and the specific gravity of these materials is given in Table . Both PKS and GGBFS are dominated by silica and alumina, which are required in geopolymerization. A Bruker X-ray diffractometer (XRD) was used for characterizing the mineral phases within a scanning range of 5–70°, 2θ. The diffractogram (Figure ) shows that the GGBFS is highly amorphous, showing a hump at 25–35°, 2θ°. The GGBFS also shows the presence of crystals of coesite (100%), a polymorph of crystalline silicon dioxide.[36] The PKS contains kaolinite (77%) and quartz (22.6%). A Nicolet FTIR device was used to collect spectra data within the 400–4000 cm–1 range. The infrared (IR) spectrum of the GGBFS and PKS is presented in Figure . The OH-stretching and Al–OH deformation bands of kaolinite appear at 3000–4000 and 912.51 cm–1, respectively. The asymmetric Si–O–Si (Al) stretching vibration occurs at 998–1034 cm–1. The Si–O-symmetrical stretching vibration of SiO2 appears at 678–763 cm–1.[37−39] The intensity of the bands in the OH-stretching region is not pronounced in the PKS, implying a low degree of orderliness, making it an excellent source for metakaolin production.[39]

Table 1

Chemical and Physical Properties of Raw Materials

raw materials	SiO2	Al2O3	Fe2O3	CaO	K2O	Na2O	MgO	SO3	TiO2	specific gravity
GGBFS	35.30	14.21	0.50	43.79	0.00	0.00	0.00	0.00	0.75	2.89
PKS	55.55	27.48	12.08	0.00	3.07	0.00	0.33	0.00	1.49	2.60

Figure 1

Diffractogram showing mineral phases present in powdered kaolinitic shale and ground granulated blast furnace slag.

Figure 2

Infrared spectrum of powdered kaolinitic shale and ground granulated blast furnace slag.

The alkaline solution was a mixture of sodium hydroxide (NaOH) and sodium silicate (Na2SiO3). The NaOH having ≥98% purity was supplied by Sigma-Aldrich. The Na2SiO3 (SiO2/Na2O = 3.375, specific gravity = 1.390) was obtained from Loba Chemie, India. Distilled water was used in all the formulations.

Methodology

Conversion of PKS to Metakaolin

The PKS (75 μm) was placed in an oven at 1562 °F for 1 h to produce metakaolin.[40] Earlier investigations had revealed this to be the optimum calcination temperature for the kaolinitic shale used in this study. XRD, FTIR, and thermogravimetry analysis (TGA) were used to study the dehydroxylation process. The XRD and FTIR tests were performed using the same equipment and conditions as discussed earlier. The powders were also placed in an SDT Q600 device manufactured by TA Instruments and heated from the ambient room temperature up to about 1634 °F at a rate of 50 °F/min in an atmosphere of air.

Determination of Minimum Required Silicate Modulus

The initial objective was to determine the minimum silicate modulus (Ms) required to have an alkaline solution with suitable gelation. A modulus silicate of 0.7, 0.9, and 1.1 was initially investigated. The NaOH was used to adjust the silicate modulus. The NaOH pellets were dissolved in the sodium silicate solution and allowed to cool down to ambient room temperature. Distilled water was used to adjust the H2O/Na2O ratio of the silicate solution. The H2O/Na2O controls the alkalinity and workability of the slurry and it has been reported that a value of 11/12 gives the optimum workability.[41,42] In this work, an H2O/Na2O of 11 was used in all systems.

Geopolymer Synthesis

A binary geopolymer system was developed using the PKS and GGBFS in the ratio of 70:30 (PKS/GGBFS) by weight of blend (BWOB). The choice of GGBFS was based on the report by several authors that it enhances the setting and contributes to strength, especially at an optimal level of 30%.[13,43−45] The sodium bentonite was dry presheared at 12 000 rpm for 5 min. The PKS and GGBFS were added to the liquid phase composed of the alkaline solution, defoamer, and dispersant. The role of sodium bentonite was to control free water and sedimentation observed in the initial trial tests. The mixing of the slurry, conditioning, rheology, and thickening time tests were according to the guidelines provided by American Petroleum Institute (API).[46,47] The recipes for the geopolymer system are given in Table .

Table 2

Mix Design for Geopolymer

material	proportion, %BWOB
PKS	70
GGBFS	30
sodium bentonite	10
defoamer	0.3
alkaline solution/binder ratio	4

Curing

The slurries were poured into 1.5 in. × 4 in. cylindrical molds. The molds (three samples for each system) were placed in aging cells containing distilled water and then placed in an electric oven. A pressure of 200 psi was applied to the aging cells. The geopolymer systems were all initially cured at 163 °F for 48 h. Then the control sample (aged at 163 °F for 48 h) was taken out of the oven, and the temperature was ramped up to 194 °F and kept at this temperature for 24 h. The experiment was repeated for a temperature rise of 248 °F. Microstructural analyses were performed at the end of each curing period.

Results and Discussion

Investigating the Conversion of PKS to Metakaolin

XRD Analysis of Raw and Heated PKS

The characterization of the PKS indicated the presence of kaolinite (K) in a high proportion (77%) (Figure ). Such a high amount of kaolinite implies the rock can be categorized under high grade, a category that possesses the potential of being highly reactive upon heat treatment.[37,48] The diffractogram in Figure shows that while the peak of quartz is present that of kaolinite disappears upon heat application. This is probably due to dehydroxylation of the structural water indicating the transformation of kaolinite into metakaolin, confirmed by the presence of a halo hump between 20 and 35°, 2θ.[49]

Figure 3

Diffractogram showing the effect of temperature on the powdered kaolinitic shale.

FTIR of Raw and Heated PKS

The FTIR can be used to study the dehydroxylation process by monitoring the behavior of the bands associated with the OH groups. When the OH groups in the clays’ structure are removed, the material loses its crystal structure and becomes amorphous.[50] When a rock contains more than one clay mineral, it is sometimes difficult to characterize it using the FTIR technique due to overlap in the spectra. However, this sample shows only kaolinite, which makes the use of FTIR efficient. Figure compares the IR spectrum before and after heat treatment. The two peaks in the 3000–4000 cm–1 range attributed to OH-stretching vanish upon heating, indicating dehydroxylation has taken place.[37,39] Dehydroxylation is further confirmed in the disappearance of the OH-deformation band at 912.51 cm–1 in the heated sample.[39,51] The change in the microstructure upon heating is also inferred from the intensity of the bands at 998–1037 cm–1 attributed to the asymmetric Si–O–Si (Al) stretching.[39,52]

Figure 4

IR spectrum showing the effect of temperature on the powdered kaolinitic shale.

Thermogravimetry Analysis of Raw and Heated PKS

Figure compares the weight loss of the raw and heated samples. The loss of structural water at certain temperature intervals could be used to infer the dehydroxylation process. In the raw sample, the weight loss up to about 230 °F is due to the removal of the adsorbed water in the interlayer of the clay mineral while the loss between 959 °F and 1634 °F is due to dehydroxylation.[37,53,54] It is confirmed that the material undergoes dehydroxylation upon heat treatment, inferred from the negligible weight loss in the regions associated with the loss of structural water. The total weight loss in each of the samples is provided in Table .

Figure 5

Thermogram of raw and heated powdered kaolinitic shale powders.

Table 3

Total Weight Loss of Samples at the End of the Test

sample	weight loss, %
raw	19.36
1562 °F	1.08

Minimum Silicate Modulus

The silicate modulus (Ms) of the sodium silicate solution controls the extent of gepolymerization and hence affects parameters like setting, viscosity, and strength.[55−57] Commonly used ratios fall within the range of 0.6 to 2 with the authors reporting various optimum values.[57−62] It was necessary to determine the minimum Ms that would help to achieve a workable solution. Figure compares the behavior of alkaline solutions with Ms of 0.7, 0.9, and 1.1. The system with a Ms of 0.7 (Figure a) quickly gels as the solution cooled down, however, when the Ms is increased to 0.9 the rate of gelling reduces. In Figure b, the solution is clear and remains in this state for about 4 h, however, precipitation begins in Figure c after about 4 h, and severe precipitation occurs in Figure d at 24 h. However, the solution designed at Ms of 1.1 remains clear and stable for the entire 24 h period. The objective of this study is not to investigate the effect of Ms and hence the minimum Ms required to achieve a workable alkaline system, which in this investigation 1.1 was selected.

Figure 6

Effect of modulus silicate on the alkaline system.

Rheology and Viscosity

The consistency plot (Figure ) of the developed geopolymer was fitted with the Bingham plastic model (BP). The model fits the data with a coefficient of determination (R2) of 99.9%. The slurry has a yield stress of about 1.06 lbf/100 ft2. The plastic viscosity is approximately 0.178 lbf/100 ft2 (85 cP). The plot of the shear rate versus viscosity plot in Figure indicates that the developed system has a shear thinning behavior, which is the desired flow behavior for oil-well cementing. The above parameters show that the geopolymer system exhibits good flow behavior.

Figure 7

Rheology of metakaolin-based geopolymer at 114 °F.

Figure 8

Viscosity of metakaolin-based geopolymer at 114 °F.

Thickening Time

Cement systems should have a reasonable setting or thickening time to allow for efficient cement placement. In conventional cement systems, the time a slurry achieves a consistency value of 70 or 100 Bearden unit of consistency (Bc) is often reported as the thickening time.[46] The consistency of the slurry at different times is shown in Figure . The investigation was done up to 70 Bc. The initial consistency of the slurry is about 4.8 Bc and it achieves 70 Bc in about 3 h and 26 min.

Figure 9

Thickening time of synthesized geopolymer system at 114 °F and atmospheric pressure.

Results and Discussion

XRD Analysis of Geopolymer Systems

The mineralogical composition of the geopolymer systems aged under different temperature conditions is presented in Figure . All geopolymer systems show an equivalent amorphous hump at 20 to 40°, 2θ°. This indicates the formation of the aluminosilicate gel and amorphous calcium silicate hydrates from the geopolymerization of the metakaolin and GGBFS, respectively.[43,63,64] The control sample cured at 163 °F only, G1, shows the presence of alite, coesite, and babingtonite. The alite is from the dissolution of both raw materials. The coesite is from the GGBFS, while the babingtonite (Ca2(Fe,Mn)FeSi5(OH)14) is associated with zeolite minerals.[65] When the temperature was ramped to 194 °F (G2), coesite, gobbinsite (Na5(Si11Al5)O32·11H2O), and gypsum are seen in the microstructure while anorthite crystallizes at 248 °F (G3). The gobbinsite and anorthite (CaO, Al2O3, 2SiO2) are related to the zeolite group.[66−68] In general, the diffractogram shows an increase in crystallinity with increasing curing temperature. Increased crystallization could lead to thermal stress in the microstructure which would degrade the system.[69,70]

Figure 10

Diffractogram of geopolymer systems cured under different temperature conditions.

FTIR Analysis of Geopolymer Systems

The FTIR technique is accurate in determining tiny changes in the microstructure.[11] The significant difference in the infrared (IR) spectra (Figure ) is a confirmation of the formation of new bonds due to the presence of new phases. The main band in Figure occurs at 929–950 cm–1, and it is a result of the asymmetric stretching of the Si–O–Si (Al) band of the geopolymer structure.[71] First of all, the shift of the asymmetric Si–O–Si (Al) peak to lower wavenumbers (1037 cm–1 to 929 cm–1 to 950 cm–1) is an indication that geopolymerization has occurred, but the reduction in the intensity of the peak in this region with increasing curing temperature would suggest a breakdown of the gel structure.[72,73] This would be responsible for the observed loss in strength reported by several authors for geopolymers subjected to higher curing temperatures.

Figure 11

IR spectrum of geopolymer systems cured under different temperature conditions.

TGA Geopolymer Systems

In general, the weight loss up to about 572 °F is because of the escape of physically and chemically bound water. The weight loss from 572 to 1202 °F is due to dehydration of the binding gel. In Figure , there is greater weight loss in the geopolymer systems exposed to elevated temperatures. Higher weight loss would suggest the presence of a greater amount of cementitious gels. However, this is not the situation. It can be explained that at elevated temperatures, new phases are formed which would be unstable or could have more water molecules.[74,75] These factors would contribute to the higher mass loss in G2 and G3 geopolymer systems.

Figure 12

Thermogram of geopolymer systems cured under different temperature conditions.

Conclusion

The application of geopolymer in oil-well cementing as an alternative to conventional Portland cement systems is currently being explored. This work focuses on the microstructural changes of geopolymer systems under elevated temperature conditions. The XRD, FTIR, and TGA techniques were used to observe this phenomenon. The findings indicate that when geopolymer is subjected to elevated temperatures, the gel structure is altered. This is due to the formation of crystalline phases. These phases may induce thermal stresses. This explains the reduction in the macroscopic properties of the geopolymer systems as reported by several authors.