The inﬂuence of TiO 2 and SiO 2 nanoparticles on filtration properties of drilling muds

During the drilling of a well, differential pressure causes filtration of the mud liquid phase into the surrounding rocks. To reduce the filtrate invasion in the rock, it is necessary to maintain the density of the mud. Also, Wellbore Strengthening Materials (WSM) can be preventively added to the mud to increase the fracture initiation pressure of the drilled formation. The solid particles from the mud create a mud cake on the wellbore wall, which, for conventional drilling, implies the use of bentonite and barite whose particle dimensions range from 0.1 to 100 µm. While drilling shales, rocks which contain pores that have an average pore size of 10 to 30 nm, it is not possible to create a high-quality mud cake, so water passes into the shale and creates wellbore instability problems. To stabilize the shale, the petroleum industry uses oil-based muds, but due to their environmental impact, it is not always possible to implement them. Nanoparticles, because of their small dimensions, can enter into the nanopores and fill this space and strengthen the rock, resulting in decreased filtration. In this paper, a review of previous laboratory research of adding nanoparticles on filtration is shown. Also, the impact of adding nanoparticles on API and HTHP filtration is examined by adding TiO 2 and SiO 2 nanoparticles in concentrations of 0.5 wt% and 1 wt% to the five water-based muds. The best result was obtained with TiO 2 nanoparti cles at a concentration of 0.5 wt%.


Introduction
Nanofluids are defined as suspensions of nanoparticles having an average particle diameter up to 100 nm (Ilyas et al., 2014). Material properties at a nanometer scale can be significantly different from a material which has larger dimensions. The nanoparticles have a much more specific surface (see Figure 1) compared to the specific surface of a material which has larger dimensions, for the same unit of mass, resulting in greater chemical reactivity and a greater influence on their physical properties (Aftab et al., 2017; Kök and Bal, 2019). Due to the larger specific surface, nanoparticles can be linked in a way to create firmer and lighter materials (El-Diasty and Ragab, 2013).
During drilling, companies face many problems that must be resolved in order to enable safe drilling progress in accordance with the project. The basic function of drilling mud is to carry out cuttings from a wellbore while maintaining the desired bottom hole pressure (BHP) and wellbore stability (Al-Zubaidi et al., 2017; Zadravec and Krištafor, 2018). Bottomhole pressure at a certain depth is primarily dependent on the density of the drilling mud and during conventional drilling, it must be higher than the expected pore pressure. During the drilling of a well, differential pressure, which is the difference between the pressure in the well and the pore pressure at a certain depth, causes the filtration of the mud liquid phase into the surrounding rocks. At the same time, the solid particles from the mud create a mud cake on the wellbore wall. By increasing the pressure in the borehole above the fracturing pressure of the rock results in the occurrence of fractures and fissures. (Pašić, 2012). In order to reduce the filtrate invasion in the rock during the drilling of wells, certain preventive measures should be taken. First, it is necessary to prepare a mud which has a certain density that will not cause a pressure Figure 1: Increasing the specific surface by dividing macroparticles to nanoparticles (Salih et al., 2016) in the wellbore higher than the fracture pressure of the formation. Additionally, Wellbore Strengthening Materials (WSM) are preventively added to the mud in order to close the existing natural cracks and thereby increase the rock strength or fracture pressure of these hardened rocks (see Figure 2). WSM are very similar to Lost Circulation Materials (LCM).
According to Figure 2, case A shows a fracture initiation pressure without added WSM, while case B shows a fracture initiation pressure with WSM. In the case when WSM are not used (A), fracture initiation pressure is lower, and for further drilling, a lower density mud is allowed than in the case where WSM are added to the drilling mud (B). Wellbore strengthening can be applied in cases where the borehole is made in areas a) where frequent mud losses occur, b) when there is a small difference between the hydrostatic pressure in the well and the fracture pressure, c) for drilling wells in depleted reservoirs, etc. (Gaurina-Međimurec et al., 2015). In recent years, the drilling of unconventional reservoirs is rapidly rising, preferably shales containing nanopores having an average size of 10 to 30 nm and they are characterized by their extremely low permeability (Sensoy et al., 2009). Due to this, it is not possible to create a high-quality mud cake on such rocks by using conventional mud additives. Conventional mud additives, such as bentonite and barite, have particle diameters ranging from 0.1 to 100 µm. Consequently, water passes into the   shale pores whose ions react with the clay minerals resulting in wellbore instability problems . To stabilize the borehole wall during the drilling of shale formations, the petroleum industry uses oilbased muds that do not react with clay minerals in the shale, but due to their environmental impact, it is not always possible to implement them (Sensoy et al.,

2009
). Therefore, the use of nanoparticles is considered, which, due to their small dimensions, can enter into the nanopores of shale, fill this space and strengthen the rock, which also results in an increased fracture initiation pressure. Nanofluids are defined as suspensions of nanoparticles which have an average particle diameter less than 100 nm (Ilyas et al., 2014). Table 1 shows dif- ferent types of rocks, common permeability, pore dimensions, and materials used for the effective plugging of the pores, while the effect of adding nanoparticles to reduce filtration and increase the stability of the wellbore is shown in Figure 3. Due to the pressure difference between hydrostatic/ hydrodynamic and pore pressure at a certain depth, there is still a penetration flow of mud from the wellbore into the rock. This problem is present during drilling unconventional as well as conventional reservoirs. Figure 3 (left) shows the case when conventional lost circulation materials were added to mud to fill the pores on the borehole walls.  As seen in Figure 3 (left), LCM has larger particle dimensions than pores in formations and cannot adequately plug them. By adding nanoparticles, they plug the space between larger particles which results in the creation of a high quality, thin, impermeable mud cake (Figure 3 (right)).

Type of nanoparticle
According to literature, the advantage of adding nanoparticles to reduce filtration was determined by laboratory tests with only a few field applications. For example, during the drilling of a 12 1/4" interval through a freshwater zone in the Reconcavo Basin, Brazil, there was a risk of contamination of the freshwater zone. To prevent contamination of the freshwater zone, Nano Fluid without chlorides was used. This type of mud also provides adequate shale inhibition as well as optimum drilling performance. After a 13 3/8″ casing was run to a TD of 515 m, drilling was resumed with the Nano Fluid from 515 m to 1600 m without any problems, so the Nano Fluid showed excellent performance while drilling the shale formation (Barroso et al., 2018). A detailed summary of the most relevant laboratory research related to nanoparticle application for reducing the filtration of drilling fluids is presented in the next section.

Review of previous laboratory tests
A summary review of the type, size and concentration of nanoparticles used to reduce filtration and improve wellbore strength is shown in Table 2. Most of presented papers indicate the possibility of nanoparticles to plug the nanopores, which reduces filtration and improves wellbore strength.
By analyzing the results of previous tests, the addition of nanoparticles to some muds gave unexpected results. For example, Vryzas et al. (2015) found that iron oxide nanoparticles significantly reduced the filtration of bentonite suspensions. In high temperature high pressure (HTHP) conditions, the largest filtration reduction (reduction was 42.5% in comparison to bentonite suspension without nanoparticles) was observed when iron nanoparticles at a concentration of 0.5 wt.% were added into bentonite suspensions. Also, API filtration was measured and at higher concentrations (1.5 wt% and 2.5 wt%), the results showed a decrease in filtration. At the same time, both API filtration and HTHP filtration were significantly increased with the addition of SiO 2 nanoparticles at all three concentrations (0.5 wt%, 1.5 wt% and 2.5 wt%). Taraghikhah et al. (2015) found that filtration properties remained unaltered at concentrations below 1 wt% of SiO 2 nanoparticles in comparison to mud without nanoparticles. Wahid et al. (2015) added SiO 2 nanoparticles in a synthetic mud at a concentration of 0.32 to 0.71 wt% and obtained a reduction in the filtration and thickness of the mud cake, with stable rheological properties.
According to the results of laboratory research presented in Table 2, it is very hard to extract specific conclusions. Despite the presented test types and conditions, in most cases there is a lack of detailed composition of the tested fluids (concentration of added additives, nanoparticle size, nanoparticle and additives manufacturer, etc.) which greatly aggravates a quality comparison of the presented results. All of this indicates that there is still room for future examinations and that the impact of nanoparticle addition into mud is still partially unclear.

Material and methods
The impact of adding nanoparticles to the water-based mud on its filtration properties was examined at the Drilling Fluid Laboratory (Department for Petroleum Engineering, Faculty of Mining, Geology and Petroleum Engineering, University of Zagreb) in Zagreb, Croatia. Data on TiO 2 and SiO 2 nanoparticles obtained from the manufacturers are shown in Table 3

(Evonik Industries, 2017).
In order to gain a better understanding of the effect of adding nanoparticles on the filtration properties of drilling mud, five water-based drilling mud samples were prepared and tested: base bentonite drilling mud (A) and four samples containing different types of nanoparticles (B-E) (see Table 4).
After the preparation of the bentonite drilling mud (A), 9 or 18 ml of Aerodisp W740X (TiO 2 ) nanoparticles were added to a base mud which approximately corresponds to the concentration of 0.5 wt% and 1 wt% of TiO 2 nanoparticles (mud B and C). Muds were stirred for 30 minutes in a stirrer to reduce the possibility of agglomeration. After the drilling mud sample was prepared, the influence of adding TiO 2 on the filtration prop-   erties and mud cake thickness were determined. In the same manner by adding 14 ml (0.5 wt%) or 28 ml (1 wt%) Aerodisp W7330N (SiO 2 ) nanoparticles into bentonite drilling mud (A), muds D and E were prepared.

Characterization of nanoparticles
To verify the data obtained from the manufacturer, granulometric analyses of the samples were made at Ruđer Bošković Institute, Zagreb, Croatia to determine particle size distribution. Two samples were analyzed, namely: 1. Aerodisp W740X -an aqueous suspension of TiO 2 nanoparticles and 2. Aerodisp W7330N -an aqueous suspension of SiO 2 nanoparticles. The selected samples were diluted with deionized water prior to the granulometric analysis to achieve the optimum suspension concentration. Granulometric analyses were done by the laser beaming method using the instrument LS 13320 (Beckman Coulter). A ULM module (universal fluid module) was used for the analysis, ranging from 0.04 to 2000 μm. Each sample was measured in triplicate. Figures 4 and 5 show the average result of two consecutive measurements.
According to the shape of the granulometric curve, it can be concluded that Aerodisp W740X -an aqueous suspension of TiO 2 nanoparticle has a narrow and uniform distribution and these particles are less than 300 nm, while the average particle size is 117 nm. Aerodisp W7330N -an aqueous suspension of SiO 2 nanoparticles has a wider but relatively symmetric distribution and these particles are less than 400 nm, while the average particle size is 116 nm. Although the manufacturer declared an average particle size of 70 nm for TiO 2 nanoparticles and 120 nm for SiO 2 nanoparticles, these measurements showed that both types of the implemented nanoparticles had similar average particle size. This means that these two types of nanoparticles can be comparable in this research.

API filtration
The API filtration properties of selected drilling muds were determined according to the API RP 13B-1 (1997) standard. The API filter press consists of a cell to which a pressure of 6.895 bar (100 psi) was applied for a period of 30 minutes to measure the volume of filtrate extracted from the drilling mud through a filter paper (Whatman No. 50) and gathered in a measuring cylinder. The filter paper surface, and thus the filtration area is 45.8 cm 2 (7.1 in 2 ).
Based on the results of the API filtration, it can be seen that, regardless of the type and concentration of nanoparticles TiO 2 and SiO 2 , their addition to the base drilling mud caused an increase in the API filtration value from 4.17% (muds B, D and E) to 10.42% (mud C) (see Table 5).
Bentonite drilling mud with added TiO 2 nanoparticles at a concentration of 0.5 wt% (mud B) has an increased API filtration value by 4.17 % in comparison to the value measured with bentonite drilling mud (mud A) and the same results were obtained with mud D and mud E. A more significant difference was made with the addition of TiO 2 nanoparticles to the bentonite drilling mud at a concentration of 1 wt% (mud C). It increased the API filtration value by 10.42% in comparison to the bentonite drilling mud (mud A).

HTHP filtration
The HTHP filter press consists of a cell in which a high pressure and high temperature are applied, and the filter medium can be filter paper (Whatman, No. 50) or a  Ceramic disc with mud cake Ceramic disc with mud cake ceramic disk. The filtration area is 22.9 cm 2 (3.5 in 2 ), twice as low as that of API filtration. In this paper, the test was carried out on a device for determining the ability of the drilling fluid to plug pores in a ceramic disc called a Permeability Plugging Tester (PPT), which represents a modification of the standard HTHP filter press. This device is used for conducting filtration tests under high pressure and high temperature conditions and is useful for assessing the ability of the drilling mud to form a high-quality mud cake due to the size and concentration of the solid particles present in drilling mud, which will prevent the further penetration of the filtrate through the filter medium. The test was carried out with the base drilling mud (mud A) and the drilling muds which contain TiO 2 (muds B and C) and SiO 2 (muds D and E) nanoparticles added in concentrations of 0.5 and 1 wt%. The permeabilities of the used ceramic discs are 0.4 μm 2 (400 mD) and 0.75 μm 2 (750 mD) and the tests were carried out at a differential pressure of 55 bar (800 psi) and a temperature of 88°C (192°F). Ceramic disc permeabilities for this study are selected based on the summary of the laboratory research presented in Table  2. Test results obtained using the PPT device are shown in Tables 6 and 7. Since the filtration surface in the API filtration is twice as large, to compare the results of the PPT test to an API filtration test, fluid volume collected after 30 minutes needs to be multiplied by 2 (see Equation 1), while the initial filtration or spurt loss can be calculated using Equation 2  Filtration through the 0.4 μm 2 (400 mD) ceramic disc showed a slight increase in filtration volume (after 30 minutes it was only 5.56% compared to the base drilling mud) for mud containing TiO 2 nanoparticles at a concentration of 1 wt% (mud C) and mud containing SiO 2 nanoparticles at a concentration of 1 wt% (mud E). Those results correspond to results observed with the API filtration measurements for muds C and E. Filtration for mud containing TiO 2 nanoparticles at a concentration of 0.5 wt% (mud B) was decreased by 22.2% in comparison to the base mud, while the addition of SiO 2 nanoparticles at a concentration of 0.5 wt% to the base mud (mud D) resulted in a filtration volume reduction after 30 minutes of only 5.6%. However, observing the spurt loss or the amount of fluid that penetrates through the disc before the mud cake is formed, it can be seen that the amount of fluid that is lost before the formation of the mud cake is significantly higher for all the tested muds than in the case of the base drilling mud (mud A). Keeping this in mind, it is obvious that mud cakes made of muds with nanoparticles are formed later than those of base muds, but after forming they are less permeable and of higher quality.
Filtration through the ceramic disc which has a permeability of 0.75 μm 2 (750 mD) showed a slight increase (6%) in filtration volume after 30 minutes for mud containing SiO 2 nanoparticles at a concentration of 1 wt% (mud E), while the filtration volume of mud containing TiO 2 nanoparticles at a concentration of 1 wt% (mud C) was the same as those measured with the base mud (mud A). The filtration volume of mud containing TiO 2 nano-particles at a concentration of 0.5 wt% (mud B) was decreased by 17.0% in comparison to the base mud after 30 minutes of filtration, while the addition of SiO 2 nanoparticles at a concentration of 0.5 wt% (mud D), the filtration volume was reduced by 10.6% in comparison to the base mud. However, observing the spurt loss, it can be seen that the amount of fluid that is lost before forming the mud cake is lower by 13.3% for muds which contain TiO 2 nanoparticles at both concentrations (mud B and C) than in the case of a basic drilling mud (mud A). On the other hand, the spurt loss of muds which contain SiO 2 nanoparticles at both concentrations (mud D and E) was higher than those in the base mud by up to 20%. Figure 6 shows a comparison of the API and HTHP filtration results for all tested drilling muds.

Discussion of results
According to Figure 6, the addition of TiO 2 and SiO 2 nanoparticles to the base mud did not have the anticipated effect on API filtration. At HTHP conditions, tests were carried out at a differential pressure of 55 bar (800 psi) and a temperature of 88°C (192°F), and a greater difference in the measured filtration was observed for the different concentrations of the added nanoparticles (0.5 wt% or 1 wt%). The test was carried out with two different discs, which have a permeability of 400 and 750 mD, but in all measurements, the same trend can be observed. As can be seen in Figure 6, the filtration of mud with TiO 2 nanoparticles added at a concentration of 0.5 wt% had the smallest filtration. Also, the filtration of mud SiO 2 nanoparticles added at a concentration of 0.5 wt% was less than those measured with mud that contained SiO 2 nanoparticles added at a concentration of 1 wt%. If these results are compared with other results presented in Table 2, similar trends can be seen, but to get a better insight on the influence of adding nanoparticles on filtration properties, comprehensive research is necessary. For further laboratory research, different nanoparticle types and sizes (for the same type) should be properly selected. Also, measurement should be conducted with a base drilling mud (eg. bentonite suspension) and gradually different additives should be added by following the proper procedure.

Conclusion
Based on the conducted laboratory research, it can be generally concluded that the TiO 2 and SiO 2 nanoparticles, which were used here, had an impact on filtration properties. The concentration of the added nanoparticles differently affects the filtration properties of the tested muds. Comparing both types of nanoparticles, even though both types have similar particle sizes, which was confirmed by granulometric analyses, better results were obtained by using TiO 2 nanoparticles. The HTHP filtration of muds was the lowest in the muds which contained TiO 2 and SiO 2 nanoparticles added at a concentration of 0.5 wt%, even lower than those measured by the addition of those nanoparticles at a concentration of 1 wt%, and it was confirmed that nanoparticles should be added at lower concentrations. Better insight will be obtained through comprehensive laboratory research which should include a wider range of different nanoparticle types and sizes added in different muds.