Effect of Novel Chelating Agent Seawater Based System on the Integrity of Sandstone Rocks

Hydrochloric acid (HCl), hydrofluoric acid (HF) and mud acid are the most commonly used acids to stimulate sandstone formations. HF acid can dissolve feldspar, carbonates, clays, micas, and quartz. The primary purpose to use HF in sandstone acidizing jobs is to remove silica and silicates that exist in quartz, clays, and feldspars. The solubility of various clay minerals (kaolinite, bentonite, chlorite and illite) in HF acid is function of concentration, reaction time, and temperature (Gdanski 1997; Zhou et al. 2013). Interaction of HCl and mud acid with minerals such as clays and feldspars may cause fine migration and damage the sandstone formations (Simon et al. 1990; Civan 2007; Mahmoud et al. 2015). As a result of that, this process will weaken the clay structure and makes it more sensitive to fluid flow. Clays and carbonates are major cementing materials in sandstone formations. Dissolving these minerals by acids will cause loosing of the sand grains and hence end up with sand production. The near-wellbore area will be exposed extensively to the injected stimulation fluid and this may cause sand production and collapse to the near-wellbore and in turn the productivity will be highly affected.

Hou et al. (2013) studied the change in rock characteristics and stress distribution of two types of sandstones that have different degree of consolidation after acidizing by mixtures of HCl and HF acids with different concentrations. They pointed out that elastic modulus and stress around the wellbore will significantly change which may cause bore hole collapse during well testing after stimulation. Nasr-El-Din et al. (2014) reported that the main problem of HF-based acids is the sand production.

Chelating agents have been widely used in oil industry to control undesirable reactions of cations with formation minerals to prevent the precipitation of solid byproducts. Moreover, they were used as a standalone stimulation fluid for sandstone and carbonate formations (Fredd and Fogler 1997; Ali et al. 2002; Parkinson et al. 2010; Nasr-El-Din et al. 2014; Mahmoud et al. 2015)

Nasr-El-Din et al. (2014) presented a successful field application of GLDA chelating agent to acidize offshore oil well drilled in sandstone formation. Good enhancement in oil production was obtained after allowing the fluids to soak for 6 hours without any fine migration or sand control problems. Mahmoud et al. (2015) stimulated sandstone cores using three different chelating agents EDTA, HEDTA and GLDA at pH of 4. Core flood experiments were conducted on the sandstone cores with illite content up to 18 wt. % at 300°F. No sand deconsolidation was noticed with any of the three fluids.

Based on the literature review, conventional acids such as mud acid and HCl used to treat sandstone formations will affect the stability of formation and cause sand production. Moreover, using seawater to prepare HCl causes not only clays instability but also calcium sulfate precipitation (He et al. 2011). Chelating agents were able to prevent scale and remove the damage caused by calcium sulfate (Mahmoud 2014). The objective of this paper is to evaluate the use of chelating agents such as EDTA and DTPA diluted in seawater along with potassium carbonate as catalysts and clay stabilizer on the integrity of sand stone rocks. Two sandstone rocks will be used, namely; Berea and Bandera sandstone cores. Mechanical properties such as Young's modulus, Shear modulus, Bulk modulus, and Poisson's ratio will be evaluated through acoustic measurements. The possibility of sand production also will be evaluated using the introduced fluids.

The primary waves (p-waves) are longitudinal waves which transfer along the same direction of the medium displacement. P-wave can travel through solids and fluids. Whereas the shear waves, S-waves, (secondary waves-because they are slower than p-waves) move perpendicularly to the wave direction and travel through solids only. The velocities of both s-wave and p-wave are useful to address the wellbore stability by monitoring the values of elastic moduli (Fjar et al. (2008)).

This group of moduli includes Young's modulus, bulk modulus, Poisson's ratio, and shear stress. These moduli are used to measure the stiffness of the acidized cores. In another words they give indications of what happened after treatment by monitoring the velocity of p-waves and s-waves. Young's modulus is a ratio of stress to strain along an axis and it is related to core stiffness.

Poisson's ratio reflects the rock compressibility and can be defined as the amount of lateral expansion relative to longitudinal contraction and can be determined as follows:

Shear modulus is the resistivity of the core against the shear deformation.

Bulk modulus known as the inverse of the compressibility and it is the resistance of the sample to hydrostatic compression.

Berea and Bandera 2-inch length and 1.5-inch diameter cores were used during the experiments. 3 wt. % KCl was used to saturate the cores and as brine for pre-flush and post-flush stage in the core flooding runs. 20 wt. % EDTA and DTPA at pH of 11 diluted with seawater combined with 3 wt. % of potassium carbonate were used to acidize the core samples.

Bandera and Berea cores were dried at 65°C for 24 hours and the dry weight was measured. The weight difference method was used to estimate the porosity after saturating the cores by 3 wt. % KCl. The cores were CT-scanned after saturation. Acoustic measurements were used to measure the velocity of p-waves and s-waves to calculate the moduli and evaluate the rock integrity. Berea and Bandera cores were loaded to a sleeve to apply overburden pressure on them. Manual pump was connected the acoustic system to control the confining pressure. Overburden pressure was increased gradually from 3, 5.5, and 8.3 to 10 MPa. At each pressure the acoustic system was used to capture the velocity of p-wave and s-wave. Then the pressure was decreased again in the same sequence. Then core flood system was used to inject the acids into the cores. The overburden pressure was 2000 psi and the back pressure was 1000 psi. The injection rate was 5 cm³/min and the temperature 250°F for all experiments. 3 wt. % KCl brine was used for pre-flush stage and post flush to measure the permeability of the core before and after the treatment with the main acid. 6 PVs of the acid were initially injected. The final permeability after treatment was estimated, the cores were CT-scanned and the acoustic measurements were done again. The same steps were repeated after injecting 10 pore volumes.

Increasing the injected pore volumes of the acid means more acid will contact the minerals inside the cores. As a result of dissolving carbonates and clays the permeability is expected to raise. For Berea sandstone, the permeability ratio improved by 15 % after injecting 6 PVs of 20 wt. % EDTA acid (pH of 11) at injection rate of 5 cm³/min at temperature 250°F. Increasing the injected pore volumes to 10 PVs increased the permeability ratio to 40 %. Bandera cores treated using EDTA solution showed 27 % improvement in permeability ratio after injecting 6 PVs. After increasing the injected pore volumes to 10 PVs, the permeability ratio increased to 32 %. Fig. 1 shows improvement in permeability ratio after injecting 6 PVs and 10 PVs of EDTA solution for the two sandstone core.

On the other hand, using DTPA at the same conditions to stimulate Berea sandstone core samples showed 23 % and 36 % enhancement in the permeability ratio after injecting 6 PVs and 10 PVs respectively. For Bandera sandstone, using 6 PVs of DTPA gave 40 % enhancement in permeability ratio. After injecting 10 PVs of DTPA the percentage increased to 44 %. Fig. 2 shows improvement in permeability ratio after injecting 6 PVs and 10 PVs of DTPA solution at 250°F anj injection rate of 5 cm³/min for Berea and Bandera sandstones. Table 1 shows the initial properties of the cores.

Elastic properties of the rock are useful indicators of the changes happened for the formation. Thus they are used to evaluate wellbore stability after acidizing jobs as well as to predict sanding problem in sandstone formations.

Fig. 3 shows Young's moduli of sandstone cores after injecting 20 wt. % DTPA and EDTA acids (pH=11) after adding 3 wt. % of potassium carbonate catalyst diluted using seawater. For Berea samples, insignificant reduction in Young's modulus was recorded after treating the cores with the two acids. In contrast, Young's modulus sharply decreased after injecting 6 pore volumes of 20 wt. % DTPA and EDTA for Bandera cores.

The reduction in Bandera cores can be related to the amount of dissolved minerals. Bandera cores contain 16 % dolomite, calcium plagioclase, and clays (14 wt. %). Chelating agents can chelate magnesium, calcium, aluminum, and ferric which are higher in Bandera than Berea cores.

Generally, when the injected pore volume increases, the Young's modulus declines indicating success of the treatment. The s-waves and p-waves move through the rock body. The velocity of these waves declined when the acid dissolved cementing minerals such as carbonates and clays. As a result of that, Young's modulus will decrease indicating the discontinuity of the pores as a result of dissolving cementing minerals and hence permeability enhancement. Bandera cores acidized with 20 wt. % DTPA combined with 3 wt.% potassium carbonate diluted using seawater showed the largest decrease in Young's modulus value as well as the highest permeability enhancement after injecting 10 PVs.

Poisson's ratio showed little increase for the two types of sandstone after treatment (Fig. 4). S-waves can move only through solids. Increasing Poisson's ratio means that amount of minerals inside the core decreased and replaced with the brine (3 wt. % KCl). Poisson's ratio cannot clearly reflect the enhancement of permeability because the increase is small. Increasing Poisson's ratio means increasing the compressibility of the cores.

Shear modulus directly proportional to the velocity of s-waves. Dissolving some minerals in sandstone cores such as carbonate and dolomite will decrease s-wave velocity because it can only travel through solids. Consequently, shear modulus will decrease as shown in Fig. 5. 

Bulk compressibility of the formation is related to the formation strength. Sand production and sand un-consolidation can be predicted early using the compressive strength of the formation. Berea and Bandera types are consolidated sandstones both have porosity around 20 % so the compaction strength of the grains is high. Table 2 summarizes the properties and velocities of Berea and Bandera core samples which were used to calculate the elastic moduli

Changes in average CT number of the cores reflects the dissolution or precipitation that took place inside the rock during the treatment. For both Berea and Bandera sandstone core samples the decrease in the average CT-number indicates that there is no precipitation inside the cores. For Berea sandstone core samples, the CT-number declined from 1730 to 1630 after injecting 10 PVs of DTPA fluid. While for the sample treated using EDTA the CT-number dropped from 1725 to 1650. Bandera sandstone has higher CT-number than Berea which reflects its stiffness and its high carbonate content. Injecting 10 PVs of EDTA decreased the CT-number from 2000 to 1920 and for DTPA case the number declined to 1820. the average CT numbers for all the treated cores are listed in Table 3 

Tixier et al. (1995) suggested a model that uses shear modulus to bulk compressibility ratio to predict the sand production after matrix acid treatment. The model was developed based on real data. Several wells have composition similar to Bandera sandstone. Khamehchi et al. (2015) validated Tixier's model to predict sanding problem using real field data. This model suggests that if shear modulus to bulk compressibility ratio is less than 0.8 × 10¹² psi² (approximately 38 ×10¹⁸ Pa²) sand problem will occur and the probability will be high if the value is around 38 ×10¹⁸ Pa². The main disadvantage of this method is that the maximum rate at which sand production may happen cannot be estimated and sand production only related to rock mechanical parameters assuming that the production rate does not affect the sand production.

Shear modulus to bulk compressibility ratio decreased for sandstone cores after injecting 6 pore volumes (Fig. 6). Nevertheless, sand production is not expected for the treated Berea and Bandera samples because the ratio is higher than the threshold limit (38 ×10¹⁸ Pa²). Although sanding problem is not expected for this scenarios, other factors such as production rate and draw down should be considered. The applied model can only predict either sand problem will happen or not but cannot specify the effect of these factors. For instance, at very high production rate sand production will occur whether we reached shear modulus to bulk compressibility ratio limit or not. Moreover, water production may trig the sand production earlier.

Rock mechanical properties were investigated after acidizing of Berea and Bandera sandstone core using 20 wt. % of EDTA and DTPA at pH of 11 combined with 3 wt. % K2CO3 diluted using seawater at injection rate 5 cm³/min and temperature 250°F. The effect of injected volume was investigated as well. The following are the conclusions that can be drawn from this study:

1. DTPA and EDTA solutions diluted using seawater enhanced the permeability of Bandera and Berea sandstone cores.

2. Young's modulus and Shear modulus decreased as injected pore volume increased. Bandera samples showed the most significant reduction for both EDTA and DTPA solutions. In contrast, Berea samples did not show high drop in moduli values.

3. Slight increase was recorded in Poisson's ratio for Berea and Bandera.

4. Shear modulus in the case of Bandera cores showed higher decrease than Berea cores. However, this reduction did not affect the rock integrity and no sand production was noticed.

5. Sand production is not expected at these conditions according to shear modulus to bulk compressibility ratio model. EDTA and DTPA chelating agent at 20 wt.% and pH 11 are safe fluid to be used for sandstone acidizing to avoid sand production problems.

This project was funded by the National Plan for Science, Technology and Innovation (MAARIFAH)-King Abdulaziz City for Science and Technology-through the Science & Technology Unit at King Fahd University of Petroleum & Minerals (KFUPM)-The Kingdom of Saudi Arabia, award number (13-Oil-151-04).

Nomenclature
AbbreviationExpansion 
• EDTA

  Ethylene diamine tetra acetic acid

 
• DTPA

  Diethylene triamine Penta acetic acid

 
• E20C3

  EDTA 20 wt. % / potassium carbonate 3 wt. %

 
• D20C3

  DTPA 20 wt. % / potassium carbonate 3 wt. %

 
• BN

  Bandera Gray sandstone.

 
• BR

  Gray Berea sandstone.

 
• P-waves

  Primary waves

 
• S-waves

  Secondary (shear) waves

 
• E

  Young's modulus, GPa.

 
• ρ

  Rock density, gm/cm³

 
• υ

  Poisson's ratio

 
• G

  Shear modulus, GPa.

 
• K

  Bulk modulus, GPa.

 
• υ_p

  Velocity of primary wave, m/s.

 
• υ_s

  Velocity of shear wave, m/s.