CO₂ Corrosion of Downhole Sand Control Screen: Experiments, Model, and Application

Abstract

Under simulated conditions typical of a high-temperature, high-pressure (HTHP) oil and gas reservoir in the South China Sea, dynamic corrosion evaluation experiments were performed on a three-layer screen structure and three types of sand retaining media. The results showed significant variations in corrosion morphology and rates among different screen components and materials. Corrosion products on the base pipe accumulated as cubic crystals, while the protective shroud showed surface needle-like corrosion products. Sand retaining media exhibited “coiled wire” corrosion products with cubic accumulations along seam edges. The 316L media showed a high corrosion risk, especially at temperatures between 140–150 °C. As CO₂ partial pressure increased, corrosion rates generally rose. A new predictive method was developed to assess and compare the corrosion resistance and life of screens, achieving a compliance rate of over 90%. This method supports evaluating the corrosion life of screens in HTHP environments. For a typical well in the South China Sea gas field with 4% CO₂, there is a high risk of screen corrosion. The screen media was identified as a failure site with a minimum corrosion life of about 5 years, while the protective shroud’s life was estimated at 11–12 years.

1. Introduction

Carbon dioxide (CO₂) corrosion, which will cause damage to the downhole pipelines, is one of the main restricting factors that disturb the normal production of the oil and gas fields [1]. In recent years, the increased occurrence of CO₂ corrosion issues can be attributed not only to the CO₂ generated during the deposition and formation of hydrocarbon source rocks but also to its presence in gas storage and development applications, such as CO₂ burial and CO₂ injection for enhanced recovery [2,3]. High CO₂ concentrations in reservoir hydrocarbons have become an undeniable reality, which is highly corrosive to metal components. This can lead to the deterioration of pipelines, wellhead equipment, and downhole tubulars [4,5].

In hydrocarbon production processes, a mechanical screen is a widely employed technique for preventing the influx of formation sand into the wellbore. This method utilizes screen tubing as a physical barrier that effectively filters out sand particles, akin to the function of the kidneys in the human body. The mechanical screen, typically made of perforated metal or woven wire mesh, is strategically placed within the wellbore to intercept the sand-laden production fluids. As the fluids flow through the screen, the sand particles are captured and retained on the exterior surface of the screen while the fluids, typically oil or gas, are allowed to pass through. This separation process prevents the sand from entering the wellbore and potentially causing a multitude of problems, including equipment damage, reduced production efficiency, and safety hazards [6,7,8]. However, as the vital component of the wellbore, mechanical screen pipes commonly experience CO₂ corrosion failure, which significantly impacts their lifespan [9,10]. In the subterranean realm, the confluence of elevated temperatures and immense pressures conspire to exacerbate the corrosive assault on the screen pipe, relentlessly compromising its integrity and structural stability [11]. Consequently, it is of paramount importance to conduct a comprehensive investigation of the various components of screen pipe.

In the 1940s, the petroleum industry began to focus on the corrosion of oil and gas well pipelines and implemented the use of coatings as an anti-corrosion measure [12]. Numerous studies have also been carried out on CO₂ corrosion behavior [13,14]. Currently, domestic and international researchers are conducting indoor experiments to simulate the corrosion caused by primary factors such as carbon dioxide, materials, flow rate, and water/gas content. These studies aim to investigate the impact of these factors on the corrosion of petroleum pipelines and identify trends in corrosion behavior [15,16,17,18,19]. It is now well established that the CO₂ corrosion behavior of steels is dominated by the precise environmental conditions such as temperature, CO₂ partial pressure, flow structure, and corrosion film, etc. [20,21,22,23,24].

For CO₂ partial pressure, studies have consistently shown that the trend of CO₂ corrosion rate under unscaled conditions is an increasing CO₂ partial pressure [25,26,27]. This is attributed to the fact that as the partial pressure of CO₂ increases, the concentration of H₂CO₃ in the aqueous phase increases, leading to a decrease in solution pH. This acidic environment accelerates the cathodic reaction, and, consequently, the overall corrosion process. However, this trend is not universally observed, and increasing the CO₂ partial pressure does not necessarily lead to accelerated corrosion [28,29]. At high partial pressures of carbon dioxide, the concentration of bicarbonate and carbonate ions in solution tends to increase, leading to supersaturation of FeCO₃. This promotes the formation of protective scales, which can hinder corrosion [30]. Therefore, the effect of CO₂ partial pressure on corrosion is complex and depends on various factors.

Temperature is also a key factor influencing CO₂ corrosion. It directly affects the electrochemical reactions occurring on the steel surface. In general, between 25 and 40 °C, the corrosion rate increases with increasing temperature [31]. At low temperatures (below 40 °C), the corrosion products (mainly of Fe₃C and some FeCO₃) formed on the steel surface are typically very loose and porous, providing little protection against further corrosion [32,33]. However, at elevated temperatures (above 60 °C), the behavior of CO₂ corrosion changes. This is due to the lower solubility of FeCO₃ at high temperatures, changing the type and protectiveness of the corrosion products formed on the steel surface [34,35,36,37]. Evidence from several experimental studies has established that the temperature range between 40 and 60 °C is a critical region where the corrosion rate transitions from being relatively high to being relatively low [38,39,40,41]. In addition, temperature has a strong influence on corrosion scale formation and determines corrosion scale properties such as density, porosity, and permeability [42].

In the oil and gas industry, the Norsok model is widely used to calculate the surface corrosion rate of carbon steel [43]. However, this model may not be directly applicable to screen, which has a more complex structure and material composition compared to conventional casing and tubing. Mechanical screens typically consist of a base pipe, sand retaining media, and an outer protective shroud. The base pipe is usually made of carbon steel, while the sand barrier media and outer protective shroud can be made of different materials, such as 13Cr, 316L, N80, and other materials of stainless steel. It is not clear how to accurately predict its corrosion rate. Additionally, the downhole environment where the screen is installed can vary significantly, making it difficult to assess its corrosion rate and evaluate its life downhole. Despite extensive research on the mechanism of CO₂ corrosion, limited attention has been paid to the corrosion behavior of downhole mechanical screen in harsh environments characterized by a high temperature, high pressure, and high production conditions.

In this work, the target object is D-X gas field, which is a typical representative of high temperature and high-pressure gas field in the South China Sea, with a well depth of 5000 m, and is a medium and low permeability reservoir. The pressure coefficient of the field is 1.82 and the temperature is 152 °C. The highest field temperature can reach 188 °C and the highest pressure coefficient can reach 2.08, which is a typical HTHP gas reservoir. At present, the development of horizontal wells and large displacement wells is being implemented, and the production rate of a single well is higher than 1.2 million m³/d, which is a kind of high-yield gas well [44]. This environment poses unique challenges for material durability and corrosion resistance, necessitating a detailed and methodical evaluation. To address this challenge, a new type of dynamic corrosion experimental apparatus for the screen was developed, and carbon dioxide corrosion experiments specifically for the mechanical screen were carried out. Combined with the results of the corrosion experiments, a life prediction model of screen is proposed, and a case study is carried out. This model, considering the unique structural and material properties of screen tubing, as well as the specific downhole conditions where it is installed, is an effective method for evaluating integrity of the screen.

2. Experimental

2.1. Apparatus

The dynamic corrosion evaluation experiments of the mechanical screen were carried out by using the dynamic corrosion evaluation experimental apparatus of screen pipe in an HTHP gas reservoir. The apparatus is shown in Figure 1, which is composed of four parts, a high pressure pump conveying system (gas pressurization supply system), two sets of main reactor, temperature and pressure sensor, and control acquisition system. The maximum temperature of the reactor is up to 350 °C and the maximum pressure is 30 MPa. The large-capacity reactor of the device can directly accommodate the local 10 cm × 10 cm samples of the mechanical screen and the short segment samples of the whole screen pipe. Its special clamping mechanism can place 6–8 samples at the same time for corrosion comparison experiments. Another feature of the apparatus is that it can simulate dynamic corrosion conditions. In addition to the dynamic rotation of the sample in the corrosive fluid through the rotating gripping mechanism, it can also replace the gas and liquid without stopping the machine to simulate the continuous output conditions of the actual downhole fluid, making the corrosion simulation results closer to the actual situation. The two reactors can be set at a different temperature, pressure, and fluid composition conditions at the same time for corrosion evaluation contrast experiments.

During the experiment, the whole or part of the sample of the screen is installed in the high-pressure reactor (or low-pressure reactor), as shown in Figure 1c. Through the multi-component corrosive gas quantitative proportioning and pressurization pumped into the high-pressure reactor, it simulates the actual production of high-pressure combination of gas on the corrosion of the screen pipe short section, and the corrosion rate and corrosion resistance of the screen short section for the evaluation of different corrosion parameters under the conditions (temperature, partial pressure, stratum water mineralization, etc.). In the experimental process, the screen samples in the reactor are driven by servo motors with high accuracy of speed measurement to rotate at a set angular velocity. When the experimental time reaches the set value or the screen samples show obvious corrosion phenomenon, the experiment can be ended.

2.2. Materials and Preparation

2.2.1. Corrosion Samples

Currently, the metal mesh screens dominate the sand control completion field, and are especially widely used in offshore oil and gas fields [45,46,47]. The metal mesh mechanical screen is usually composed of three components: base pipe, multi-layer metal mesh sand retaining media, and outer protective shroud. For this experiment, a typical multi-layer mechanical screen commonly used in offshore oil and gas fields was used. The base pipe is made of N80 steel, the outer protective shroud is made of 304 stainless steel, and the sand retaining media is made of 316L stainless steel. The corrosion performance of different types of sand retaining media of wedge wire screen (316L), metal fiber screen (P110), and foam metal screen were also studied to assist in the study. The final experimental samples obtained are shown in Figure 2. The samples of uniform height (50 mm) were made from a short section of the screen using wire cutting. The surfaces of the experimental samples were ground smooth and flat using sandpaper, and then cleaned using acetone and anhydrous alcohol before being weighed and placed in the dryer for use.

It is necessary here to clarify exactly that N80 is a carbon steel grade, 316L and 304 are stainless steel grades with different carbon contents, P110 is a high-strength carbon steel grade. Foam metal can be made from various metals with different chemical compositions. The foam metal used in this paper is made of nickel.

At the bottom of the well, the base pipe, protective shroud, and sand retaining media layer of the screen are simultaneously damaged by corrosion; due to their different materials and structures as well as their interactions, the three components corrode at different rates, and corrosion damage to any one of the components implies the total failure of the screen.

2.2.2. Fluids

The experiment directly simulates the wellbore CO₂ corrosion conditions in an HTHP gas field in the South China Sea. The simulated formation water used in the experiment was configured as a solution using analytically pure pharmaceuticals, and its composition is shown in

Table 1. Chemical compositions of simulated water.

Ions	K+ and Na+	Ca2+	Mg2+	Cl−	SO42−	HCO3−
Content (mg/L)	5652	40	17	6402	1300	2539

. The experimental temperatures were 100 °C, 140 °C, 160 °C, and 190 °C, the total pressure was 25 MPa, of which the CO₂ partial pressures were set to 1.5 MPa, 2.5 MPa, 4.5 MPa, 9.5 MPa, 14.5 MPa, and the stratum water-to-gas ratios were 0.6 m³/10⁴ m³, 0.8 m³/10⁴ m³, and 1.0 m³/10⁴ m³, and the experimental time was 15 days.

2.3. Procedure

The experimental procedure is shown in Figure 3. During the experiment, the samples to be tested were fixed on the clamping mechanism and placed into the reactor. After sealing the reactor cover, nitrogen was introduced into the reactor for 1 h to expel the oxygen. The outlet valve was then closed to warm up to the predetermined temperature. CO₂ and other gases were introduced into the reactor to reach the corresponding partial pressure value, followed by nitrogen to achieve the design pressure. Once the pressure stabilized, the rotor was turned on to start the timing. Every 2–3 days, the corrosive medium in the reactor was dynamically adjusted by discharging and replenishing it to maintain the desired conditions. The experiment continued until the set time was reached, indicating the completion of the corrosion reaction. After the experiment, the reactor was opened, and the samples were removed and dried. An electron microscope and a scanning electron microscope were used to observe the corrosion micro-morphological characteristics. The sample surface deposits were then rinsed with water, dried with medical gauze, and placed into anhydrous ethanol for about 5 min. The samples were removed, dried with cold air, and weighed after drying for 30 min to calculate the corrosion rate.

2.4. Experimental Evaluation Methods

The quantitative corrosion experimental evaluation method was used to quantitatively evaluate the corrosion rate of different structural complex media. The indoor experimental evaluation methods are specified below:

(1)

Corrosion weight loss rate is the proportion of mass lost by corrosion per unit time:

$$\alpha =\frac{∆m}{m_0·∆t}$$

(1)

where α is the corrosion weight loss rate, 1/y; ∆m is the difference in mass before and after corrosion, g; m₀ is the mass before corrosion, g; ∆t is the corrosion time, y.

(2)

The mass corrosion rate is the mass lost by corrosion per unit area per unit time:

$$v_m=\frac{∆m}{S·∆t}=\frac{∆m}{m_0·∆t}·\frac{\rho }{k}=\alpha ·\frac{\rho }{k}$$

(2)

where v_m is the mass corrosion rate in g/(cm²·y), S is the specimen surface area, cm², ρ is metal density, and k is surface-to-volume ratio, cm⁻¹.

The value of k is related to the shape and structural parameters of the material, and is calculated by the formula:

$$k=\frac{S}{V}=\frac{S·\rho }{m_0}$$

(3)

(3)

The depth corrosion rate is the corrosion depth or thickness of the media surface per unit time:

$$v_c=\frac{∆m}{S·\rho ·∆t}=\frac{v_m}{\rho }$$

(4)

where v_c is the deep corrosion rate, mm/y; ρ is the material density, g/cm³.

The conventional metal indoor corrosion evaluation method cannot be used for complex structural components. However, the evaluation method established in this thesis can be applied to the corrosion rate testing and evaluation of complex screen structures and components. The introduction of the area correction coefficient corrects the face-to-face ratio parameter for staggered structures, further improving the accuracy of the evaluation method.

3. Results and Discussion

3.1. Microscopic Morphology and Product Analysis

3.1.1. Corrosion Morphology of Basic Pipe

The base pipe samples of the same type were taken and placed in a reactor with CO₂ partial pressures of 1.5 MPa, 2.5 MPa, 4.5 MPa, 9.5 MPa, and 14.5 MPa, respectively, and the experimental time was 360 h, the temperature was 160 °C, and the water/gas ratio was 0.6 m³/10⁴ m³. The surface corrosion morphology of the base pipes and the SEM images under the different conditions at the end of the experiments are shown in Figure 4.

The increase in CO₂ partial pressure resulted in a significant rise in the number of grains within the corrosion product film. At a CO₂ partial pressure of 1.5 MPa, the grain size peaked, with a minimal quantity and non-uniform distribution. Minimal corrosion was observed on the base pipe surface, leading to a darkened appearance with sporadic greenish corrosion spots. Upon elevating the CO₂ partial pressure to 2.5 MPa, noticeable surface rust developed on the base pipe, exhibiting a mossy corrosion morphology. The grains within the corrosion product film became denser, minimizing voids. However, at a CO₂ partial pressure of 4.5 MPa, the film between crystals lacked tight bonding, presenting substantial gaps that contributed to an escalating corrosion rate. Subsequently, at a CO₂ partial pressure of 9.5 MPa, corrosion products enveloped the substrate surface entirely, displaying a more organized and compact structure. Nevertheless, as the CO₂ partial pressure soared to 14.5 MPa, crystal growth persisted, forming grouped bonds. Consequently, severe corrosion affected the base pipe, resulting in darkening at high temperatures, while layers of corrosion rust covered its surface. This signified a lack of effective suppression of CO₂ corrosion at that stage.

The functional performance of the corrosion product film on substrate protection is influenced by various factors, including film integrity, densification, internal stress, substrate metal properties, and other variables. In Figure 5, the N80 steel base pipe underwent SEM analysis at a magnification of 7000× under a carbon dioxide partial pressure of 1.5 MPa. The morphological characteristics of the corrosion products were examined by enlarging and analyzing images. The composition of the corrosion products present on the N80 steel base pipe was determined using an energy dispersive spectrometer (EDS). According to Cui’s study [1] and in combination with SEM and EDS analysis, the predominant corrosion product identified was FeCO₃. Under a CO₂ partial pressure of 1.5 MPa, the corrosion products manifested as cubic crystals adhering to the base pipe surface. Notably, at this pressure level, the grain size appeared large, with a limited quantity, non-uniform distribution, and loosely arranged corrosion products.

3.1.2. Corrosion Morphology of Sand Retaining Media

The multi-layer sand retaining media (316L Steel) were taken and placed in the reactor. The temperature inside the reactor was stabilized at about 160 °C, the total pressure was 25 MPa, the CO₂ partial pressure was 2.5 MPa, and the water–gas ratio was the base value of 0.6 m³/10⁴ m³. After the experiment, using a type of scanning electron microscope, the corrosion product film on the surface of the outer protective cover under different CO₂ partial pressures was observed, and the images shown in Figure 6 were obtained.

Under the SEM, it can be clearly seen that with the increase in CO₂ partial pressure, the more obvious corrosion phenomenon generated on sand retaining media in Figure 6. By comparing when the other conditions are the same, different CO₂ partial pressures under the multi-layer sand retaining media samples of corrosion can be obtained, although 316L stainless steel in the two cases of corrosion is not obvious, but from a relative perspective can be seen; CO₂ partial pressure is greater when the sample of corrosion occurred more seriously.

As shown in Figure 7, corrosion products attach to the single or multi-layer metal mesh in the form of cubic crystals. However, due to the special structure of the sand retaining media, a complete product film does not form over the entire medium, making it more susceptible to corrosion than the base material of the remaining structure made of the same material. Dynamic corrosion comparison experiments were carried out under the baseline experimental conditions (CO₂ partial pressure of 2.5 MPa, temperature of 160 °C, water/gas ratio of 0.6 m³/10⁴ m³, and experimental time of 30 days) for three types of sand retaining media. The corrosion microscopic morphology of each sand barrier medium after the experiment is shown in Figure 8.

The corrosion surface morphology of the wire-wrapped screen after the experiment is shown in Figure 8a. From the SEM images seen, after the experiments around the silk screen plate surface scattered with sporadic etching points, and from the scanning electron microscope that can be seen the silk plate corrosion products “along the growth of the seam” in the gap at the corrosion products increased significantly. For the wire-wrapped screen, the gap between the filament winding layer mainly plays the role of sand blocking and channeling, and the gap is most likely to corrode, so for the circulation of the filament winding screen, corrosion damage has a greater impact on it. In Figure 8b, sintered metal fiber sand blocking media in 200× scanning electron microscope magnification, showing random fiber metal wire cross longitudinal intersection state, with high reliability of sand prevention, has the advantages of strong penetration ability. After the dynamic corrosion experiment on the metal wire, as well as the metal wire intersection, appeared many corrosion products, showing disordered and chaotic distribution; corrosion products to cubic, rhombic particles are mainly locally compact, and the circulation performance of the sand blocking medium and the sand blocking performance have a certain impact on this. For the metal foam, its multilayer three-dimensional structure makes it have good circulation performance. After the experiment, the microscopic corrosion morphology is “diagonal accumulation type”, that is, lots of corrosion products attached to the polygonal structure of the metal foam; as shown in Figure 8c, the corrosion products grow longitudinally after the accumulation of the cubic structure, with polygonal diagonal products, in general, not having a significant effect on the middle of the porous structure. Therefore, from the morphological analysis, the corrosion resistance of the metal foam sand barrier medium is better.

3.1.3. Corrosion Morphology of Protective Shroud

Take the same type of outer protective shroud samples, placed in the CO₂ partial pressure of 1.5 MPa, 2.5 MPa, 4.5 MPa, and 14.5 MPa conditions of the reactor, the experimental time of 360 h, the temperature of 160 °C, the water–gas ratio of 0.6 m³/10⁴ m³, after the end of the experiment; the surface corrosion of the outer protective shroud corrosion morphology of the different conditions is shown in Figure 9.

After conducting the dynamic cyclic corrosion test, it was observed that rust spots predominantly localized along the depressed areas of the cut seam on the inner wall of the protective shroud. Moreover, the corrosion in proximity to these depressions became increasingly pronounced as the partial pressure of CO₂ escalated. Specifically, when the CO₂ partial pressure was set at 1.5 MPa in the experimental setting, only minor corrosion was evident on the outer protective shroud, with no visible signs of significant corrosion. When the CO₂ partial pressure was raised to 2.5 MPa, the specimen’s surface lost its metallic sheen, displaying a rust layer akin to the corrosion product formed at the cut seam depression of the outer protective shroud. At a CO₂ partial pressure of 4.5 MPa, a purplish-red rust coating emerged on the outer protective shroud’s surface, particularly noticeable in the depressions of the cut seam. Subsequently, at a CO₂ partial pressure of 14.5 MPa, the entire surface of the outer protective shroud exhibited a rust-like appearance. Notably, at both 4.5 MPa and 9.5 MPa CO₂ partial pressures, the crystals within the corrosion product film appeared coarse and loosely aggregated, presenting significant gaps, consequently leading to a gradual increase in the corrosion rate. As the CO₂ partial pressure reached 14.5 MPa, the crystals exhibited irregular longitudinal growth but with a uniform arrangement.

3.1.4. Corrosion Morphology of the Screen as a Whole

The screen sample was placed in the reactor and subsequently divided into three distinct sections: the base pipe, the sand retaining medium, and the outer protective shroud. The reactor was maintained at a stable temperature of approximately 160 °C, with a total pressure of 25 MPa, a CO₂ partial pressure of 2.5 MPa, and a water–gas ratio set at the base value of 0.6 m³/10⁴ m³. The surface morphology of the sample post-experimentation is depicted in Figure 10. Evidently, following the dynamic cyclic corrosion experiment, localized ulcerative corrosion was observed on certain areas of the base pipe, wherein the severity of the surface corrosion exhibited a reddish-brown rust layer, albeit unevenly distributed with areas of breakage. As a result of the pronounced corrosion on the base pipe, reddish-brown rust spots were also found on the surface of the sand-blocking medium and outer protective shroud, indicative of local corrosion occurrences.

Figure 10 shows the corrosion photos of the overall parts of the screen under the body-view microscope. It can be seen that the corrosion is especially obvious at the connection between the holes of the inner protective sleeve and the sand-blocking medium, and the reddish-brown rust layer accumulates along the surface of the metal wire of the sand retaining medium. There are localized flaking phenomena and serious blistering. The reddish-brown rust stains can also be clearly seen after magnification of the metal screen of the sand retaining medium, and there are many massive corrosion products on the surface of the base pipe.

Comparison of the individual components reveals a notable decline in overall corrosion resistance despite individual components being made of high-quality stainless steel. For instance, utilizing 316L material for the sand media, recognized as one of the most corrosion-resistant stainless-steel materials, results in relatively minor corrosion under experimental conditions when examined separately. However, when considering the screen as a whole within the sand media, the severe corrosion of the base pipe leads to mutual interaction, causing significant corrosion products on the media surface, a phenomenon also observed on the outer protective shroud. Consequently, it can be preliminarily inferred that under real downhole conditions, the simultaneous corrosion damage, mutual influence, and interaction among the three screen components render the corrosion process more intricate and severe compared to individual component scenarios.

3.2. Corrosion Rate Analysis

3.2.1. Effect of Corrosion Rate under Different CO₂ Partial Pressures

With the increase in CO₂ partial pressure value, the average corrosion rate of each part also changed. To visualize the corrosion rate of different screen parts under different CO₂ partial pressures, a histogram of the variation of mass change rate with CO₂ partial pressure was plotted as shown in Figure 11.

The corrosion mass change rate of each component also showed an overall increasing trend with increasing CO₂ partial pressure. The effect of CO₂ partial pressure on the corrosion rate was greater for both the base pipe, the outer protective shroud, and the sand retaining medium.

As shown in Figure 12, when the partial pressure of CO₂ ranges from 0 to 5 MPa, the corrosion rate exhibits an insignificant increase with the rise in partial pressure. However, beyond 5 MPa, the corrosion rate escalates rapidly as the partial pressure of CO₂ increases. In the case of the base pipe, the corrosion rate at a CO₂ partial pressure of 14.5 MPa is nearly tenfold that of the corrosion rate at 1.5 MPa. This phenomenon arises due to the dissolution of CO₂ in water, leading to the ionization of H₂CO₃ into HCO₃⁻, which subsequently reacts with iron in the steel to generate FeCO₃ rust and cause metal corrosion. As the partial pressure occupied by CO₂ increases, the production of HCO₃⁻ ions is amplified, driving the chemical reaction in the positive direction and augmenting the corrosion rate to a certain extent. In summary, it is indisputable that the corrosion rate is significantly higher at elevated CO₂ partial pressures compared to low pressures. [48,49]. However, for the 304 and N80 steel, the 304 material is obviously more resistant to corrosion at high partial pressures of CO₂ in the same area. The corrosion resistance of the 316L steel at high CO₂ partial pressures is even more impressive.

It is noteworthy that the sand retaining medium exhibits a significantly higher corrosion mass loss compared to the base pipe and protective shroud (Figure 11), despite having the lowest material corrosion rate. This discrepancy can be attributed to the smaller specific surface area of the base pipe and outer protective shroud specimens. In contrast, the sand-blocking medium, composed of finer wire screen, possesses a larger specific surface area.

3.2.2. Effect of Corrosion Rate at Different Temperatures

Temperature is one of the key factors affecting corrosion. Figure 13 shows the average corrosion rate measurements obtained by the experimental evaluation method under experimental conditions at different temperatures with a CO₂ partial pressure of 14.5 MPa.

According to Figure 13, with the increase in temperature, the average corrosion rate of each part of the first increase and then decline. The base pipe is affected by the temperature of the largest, at a temperature of 140 °C when the corrosion rate reaches a maximum, the temperature increases to 160 °C with the corrosion rate instead of dropping very quickly; outside the protective shroud and sand media curve fluctuations are smaller, but overall after 100 °C there is also an increase in temperature instead of decreasing the corrosion rate significantly. This is due to the high temperature state, the screen surface easily forms a thin and protective film, so it can reduce the corrosion rate.

The results illustrate that the corrosion rate increases to its peak as the temperature rises from 100 °C to 140 °C. Low temperatures restrict the formation of corrosion products, and corrosion intensifies as the temperature increases. Conversely, at higher temperatures, the corrosion rate diminishes with increasing temperature, which can be attributed to the impact of temperature on CO₂ solubility, H₂CO₃ ionization, electrochemical reactions, and corrosion scaling. The influence of temperature on the corrosion process is intricate. On one hand, elevated temperatures foster electrochemical reactions and species migration (molecular diffusion), thereby aggravating corrosion. On the other hand, high temperatures expedite the precipitation of corrosion products and the formation of protective scales, thereby mitigating corrosion. Corrosion products, such as iron carbonate (FeCO₃), precipitate and deposit on metal surfaces after reaching their solubility limit. Johnson [50] reported that the solubility of CO₂ decreases with increasing temperature when the temperature is below 149 °C, resulting in a decrease in the concentration of carbonic acid and an increase in pH. Generally, the severity of corrosion in carbonic acid environments is higher at low temperatures than at high temperatures [51]. However, the critical corrosion temperatures vary for different materials and structures.

3.2.3. Effect of Corrosion Rate under Different Water–Gas Ratios

Water plays an important role in CO₂ corrosion; only the presence of water can make CO₂ dissolved to generate binary acid, and then further corrode to generate rust. Figure 14, for the mechanical screen, shows different parts of the average corrosion rate with the water–gas ratio change rule.

As can be seen from Figure 14, in the range of water–gas ratio from 0.6 to 1.0 × 10⁻⁴ m³/m³, both the corrosion mass loss rate and corrosion rate are accelerated with the increase in water–gas ratio, but the magnitude of the change is relatively flat. Relatively speaking, the corrosion rate of the base pipe is more obviously affected by the water–gas ratio, while the corrosion rate of the outer protective shroud and the sand retaining medium changes relatively gently, and the growth rate slows down at high water–gas ratios. On the one hand, due to the sand retaining medium material being 316L stainless steel, its own corrosion resistance is better; on the other hand, when the water–gas ratio increases to a certain value, the dissolution of CO₂ reaches saturation, which also affects the corrosion. The experiment also found that with the same corrosion environment in the same reactor, the specimen immersed in the liquid corrosion is significantly more serious than the specimen in the gas.

4. Prediction Method of Screen Corrosion Rate

4.1. Modification of the Corrosion Rate Model

Norsok M-506 standard [52], also known as the NO model, presents a method for calculation of corrosion rates in hydrocarbon production and process systems where the corrosive agent is CO₂, which is an empirical model summarized from field data and developed with broad petroleum industry participation by interested parties. The NO model focuses on the protective properties of the corrosion product film, while being sensitive to changes in pH, shown in

Table 2. Variation of parameters with pH and temperature (NORSOK M-506 standard) [52].

Temperature °C	${\mathit{K}}_{\mathit{t}}$	pH	$\mathit{f}{(\mathit{p}\mathit{H})}_{\mathit{t}}$
20	4.762	pH < 4.6	f(pH) = 2.0676 − (0.2309 $\times$ pH)
		4.6 ≤ pH	f(pH) = 5.1885 − (1.2353 $\times$ pH) + (0.0708 $\times$ pH2)
40	8.927	pH < 4.6	f(pH) = 2.0676 − (0.2309 $\times$ pH)
		4.6 ≤ pH	f(pH) = 5.1885 − (1.2353 $\times$ pH) + (0.0708 $\times$ pH2)
60	10.695	pH < 4.6	f(pH) = 1.836 − (0.1818 $\times$ pH)
		4.6 ≤ pH	f(pH) = 15.444 − (6.1291 $\times$ pH) + (0.8204 $\times$ pH2) − (0.0371 $\times$ pH3)
80	9.949	pH < 4.6	f(pH) = 2.6727 − (0.3636 $\times$ pH)
		4.6 ≤ pH	f(pH) = 331.68 $\times$ e(−1.2618 $\times$ pH)
90	6.250	pH < 4.6	f(pH) = 3.1355 − (0.4673 $\times$ pH)
		4.6 ≤ pH < 5.6	f(pH) = 21,254 × e(−2.1811 $\times$ pH)
		5.6 ≤ pH	f(pH) = 0.4014 − (0.0538 $\times$ pH)
120	7.770	pH < 4.3	f(pH) = 1.5375 − (0.125 $\times$ pH)
		4.3 ≤ pH < 5	f(pH) = 5.9757 − (1.157 $\times$ pH)
		5 ≤ pH	f(pH) = 0.546125 − (0.071225 $\times$ pH)
150	5.203	pH < 3.8	f(pH) = 1
		3.8 ≤ pH < 5	f(pH) = 17.634 − (7.0945 $\times$ pH) + (0.715 $\times$ pH2)
		5 ≤ pH	f(pH) = 0.037

.

When t = 20 °C, 40 °C, 60 °C, 80 °C, 90 °C, 120 °C, 150 °C, the corrosion rate is calculated according to the following equation.

$$v_{corr}=K_t\times f_{C{O}_2}^{0.62}\times {\left(\frac{{\tau }_w}{19}\right)}^{0.146+0.0324log{f}_{{co}_2}}\times f_{\left(PH\right)}·{\beta }_1·{\beta }_2$$

(5)

$$f_{{CO}_2}=\alpha P_{{CO}_2}$$

(6)

$$\mathsf{\alpha }=\left\{\begin{array}{c}{e}^{\left[P\times \left(0.0031-\frac{1.4}{T}\right)\right]}, P\le 25 \mathrm{MPa}\\ e^{\left[\times \left(0.0031-\frac{1.4}{T}\right)\right]}, P>25 \mathrm{MPa}\end{array}\right.$$

where v_corr, corrosion rate, mm/y; $K_t$, temperature influence coefficient; ${\tau }_w$, shear stress on the wall, Pa;$P_{{CO}_2}$, CO₂ partial pressure, bar; P, ambient pressure, bar.

For the outer protective shroud, sand retaining media, and base pipe, the corrosion velocity/rate, corrosion damage form, likelihood, and time are simulated based on the corrosion prediction model modified based on the corrosion experimental results.

Based on the screen corrosion experiments, considerations of material and complex media structure were incorporated into the model corrections. Taking the N80 material as a benchmark, the experimental data were used to fit the corrected material influence coefficient β1, and the screen structure influence coefficient β2 for each basic model and the value are shown in

Table 3. Material influence coefficient β1, and the screen structure influence coefficient β2.

Screen pipe Assemblies	Material	β1 (Material Impact Factor)	β2 (Structural Impact Factor)
Base pipe	N80	1.0	1.0
Protective shroud	304	0.28	0.43
Sand retaining medium	Ni	0.11	0.32
	P110	0.89	0.42
	316L	0.23	0.88

.

The fitting effect according to the model is shown in Figure 15. For different experimental conditions and experimental materials, the model has good regressivity, and the prediction and experimental compliance rate is 92.3%.

4.2. Evaluation Method of Corrosion Life of Bottom-Hole Screen Pipe

4.2.1. Characteristic Calculation Elements of Screen

Based on the experimental results and evaluation methods, a corrosion calculation model for the mechanical screen is proposed. Initially, an analysis of the corrosion experimental morphology reveals that the corrosion primarily consists of two types of metal corrosion: wire and flatbed metal. The corrosion test results indicate that the corrosion area essentially covers the surface of both types of metals, while the depth of corrosion varies for different components. Assuming uniform corrosion depth of the corroded object, the schematic diagram of the two calculation element models is presented in Figure 16. The variable H is defined as the original characteristic thickness of the component in mm, such as the thickness of the base pipe/protective shroud and the diameter of the metal wire.

H_cc is defined as the critical depth of corrosion damage, unit mm; R_cc is defined as the critical thickness ratio of corrosion damage, dimensionless. Calculation formula:

$$R_{cc}=\frac{2·H_{cc}}{H}$$

(7)

When the corrosion damage depth reaches the critical depth or critical thickness ratio R_cc, it is regarded as the end of life. According to the research and experience, the R_cc of the base pipe, the protective shroud, and the filter medium are 75%, 75%, and 50% respectively. The selection of the standard will affect the life evaluation results.

4.2.2. Corrosion Evaluation Method of Screen

The degree of corrosion damage R_c is the ratio of the corrosion depth to the critical damage corrosion depth H_cc, which is dimensionless and characterizes the degree of corrosion failure. When R_c reaches 100%, the damage fails. Calculation formula:

$$R_c=\frac{v_{corr}·t}{H_{cc}}$$

(8)

Corrosion damage rate V_rc is the ratio of corrosion damage degree to time t under time t, y⁻¹, which characterizes the relative corrosion damage speed of components of different types and structures. Calculation formula:

Prediction and evaluation of gas reservoir bottom-hole screen corrosion degree

$${\mathrm{v}}_{rc}=\frac{R_{\mathrm{c}}}{t}$$

(9)

Taking Equations (7) and (8) into Equation (9), the formula for the corrosion damage rate of the screen is obtained:

$$v_{rc}=\frac{2v_{corr}}{R_{cc}·H}$$

(10)

The dimensionless corrosion damage rate V_rc is the degree of corrosion damage per unit time, unit 1/y. The higher the V_Rc value, the slower the corrosion damage of the screen or component, and the longer the service life.

Corrosion damage life T_s is the time required to reach corrosion damage. Calculation formula:

$$T_s=\frac{1}{v_{rc}}$$

(11)

According to the above evaluation method of corrosion resistance of screen tubing at the bottom of gas reservoir, the evaluation table of corrosion resistance of screen tubing is obtained as shown in

Table 4. Corrosion Evaluation index of the screen.

Qualitative Indicators for Evaluation	Excellent	Good	Moderate	Poor
Corrosion damage rate VRc (1/y)	<0.067	0.067–0.1	0.1–0.2	>0.2
Corrosion damage life Ts (y)	>15	10–15	5–10	<5

.

The evaluation system considers both the absolute corrosion rate of the different component materials of the screen and the geometrical shape and size of the different parts of the screen, which can be better applied to the prediction and evaluation of the corrosion degree of the screen. This methodology offers cost and time effectiveness by providing a reliable predictive model that reduces the need for extensive long-term field testing. This predictive method can significantly cut down on the time and resources required to assess corrosion resistance, thereby offering substantial savings and enabling quicker decision-making in material selection and maintenance planning. There are also limitations in the study, as the effect of different ionic content of the brine was not considered. This matter will be further added in future studies. Future research will be critical to continually improve predictive models using broader data sets, the possibility of applying them to other high-risk environments, and the development of advanced materials based on the results of the research.

5. Wellbore Screen Optimization Applications

5.1. Typical Gas Reservoir Conditions and Screen Data Used

Typical conditions in the South China Sea gas field were used for application, and the fluid physical conditions are shown in

Table 5. Typical gas reservoir conditions of Well X in the South China Sea.

Item	Temperature	Reservoir Pressure	CO2 Content	CO2 Partial Pressure	Water–Air Ratio	Cl− Content
Data	142–188 °C	53 MPa	3.4–4.0%	2–12 MPa	0.6 m3/104 m3	6400 mg/L

. Four gas samples were obtained from the gas wells, and their CO₂ contents were 3.41%, 3.44%, 3.61%, and 3.85%, respectively, with reservoir pressure of about 53 MPa and CO₂ partial pressures of about 1.80–2.05 MPa.

According to the base data of Well X, the production section of the typical well is about 498.4 m long, with a daily production of 106.71 × 10⁴ m³/d and a temperature of 142 °C. A multi-layer metal mesh screen was used in the well, and its base pipe, outer protective shroud, and sand retaining media are made of N80, 304, and 316L steel. The thickness of the base pipe is 9.17 mm, the thickness of the protective shroud is 1.25 mm, and the diameter of the screen wire is 0.3 mm, which corresponds to the critical damage ratios taken as 25%, 45%, and 75%, respectively.

5.2. Evaluation Results of Corrosion Resistance for Screens in HTHP Environments

The corrosion rate prediction was performed using the newly fitted model, and the prediction results are shown in Figure 17.

According to the corrosion rate with CO₂ partial pressure change curve, the corrosion rate of different materials is positively correlated with the CO₂ partial pressure. When the CO₂ partial pressure rises from 1.0 MPa to 15 MPa, the corrosion rate of N80 material base pipe rises from 0.069 mm/y to 0.37 mm/y, the corrosion rate of 304 material outer protective cover rises from 0.019 mm/y to 0.102 mm/y, and the corrosion rate of 316L steel sand blocking medium rises from 0.016 mm/y to 0.086 mm/y. Under the typical conditions, the CO₂ partial pressure is 2.0 MPa. Under the typical conditions of CO₂ partial pressure of 2.5 MPa, the corrosion rate of N80 base pipe, 304 outer protective cover, and 316L sand barrier medium is 0.122, 0.034, and 0.028 mm/y respectively.

As for temperature, when the temperature is low, the corrosion rate of different materials will be positively correlated with the ambient temperature, and reach the maximum value at the ambient temperature of about 140 °C. In the typical conditions of 142 °C, N80 material, P110 material, 304 material, 316L material, Ni material, have a corrosion rate of 0.125, 0.110, 0.035, 0.030, 0.022 mm/y. When the ambient temperature rises to 140 °C or more, the corrosion rate of the different materials will coincide with the temperature increases and decrease.

The corrosion life evaluation indexes of each part of a typical screen under different CO₂ partial pressure conditions were calculated. The lowest corrosion life of each component is selected as the overall corrosion life (Ts) of the screen. The relationship between the corrosion damage rate of each component under different CO₂ partial pressures and temperatures is shown in Figure 18.

As CO₂ partial pressure increased from 1.0 MPa to 15 MPa, the corrosion rate of the base pipe, indicated by the evaluation index Vrc, increased from 0.046 to 0.237, which is an 80.6% increase. Similarly, the corrosion rate of the outer protective shroud increased from 0.062 to 0.337, representing an 81.6% increase. The corrosion rate of the sand retaining medium also increased from 0.134 to 0.72, which is an 81.4% increase. These results demonstrate the impact of higher CO₂ partial pressure on the corrosion of the different components of the screen. The corrosion life of the screen components showed a negative correlation with the partial pressure of CO₂. When the reservoir temperature is 142 °C and the partial pressure of carbon dioxide exceeds 2.12 MPa, the predicted life of the screen tube is less than 5 years. According to these corrosion life predictions, the screen is rated as having a moderate to poor result in terms of its corrosion resistance. It can also be seen from the predicted results in Figure 18b that the evaluation of the screen corrosion life in different temperature ranges is of moderate level when the partial pressure of the CO₂ is 2 MPa.

According to the results of corrosion studies, 316L screen media, although the slowest corrosion rate in materials, its wire diameter is thin, and it has the shortest corrosion life. Therefore, the following optimization recommendations are provided: (1) Under the premise of maintaining the mesh size and sand-blocking precision, increase the wire diameter by 1.5 times (0.4–0.5 mm), to improve the corrosion resistance of the medium and extend the life. (2) Optimize the structure of corrosion products that does not affect the circulation, reducing the impact of corrosion on functionality.

Using materials with higher corrosion resistance, applying chemical inhibitors to form protective layers, and using protective coatings to act as barriers against corrosive agents can reduce corrosion. Additionally, controlling the environment by reducing CO₂ partial pressure and adjusting temperature and pressure to fewer corrosive levels can also mitigate corrosion.

6. Conclusions

In this paper, the corrosion experiments of different parts of the screen in different environments were carried out to qualitatively analyze the influence of different influencing factors on the degree of corrosion of various parts of the screen, and a complete set of corrosion evaluation methodology for the mechanical screen was obtained.

(1)

Through systematic experiments to reveal the corrosion law of screen components under HTHP environment. N80 base pipe screen corrosion rate is the highest under 140–150 °C; this condition is exactly the gas field in the South China Sea with bottoming temperature conditions, and the risk of corrosion is high. The corrosion rate of 316L screen media is the slowest, but due to the media feature size being small, corrosion damage is the fastest, and this is a high corrosion risk component.

(2)

A corrosion evaluation method for screen tubing in HTHP gas reservoirs is constructed with corrosion experimental evaluation, wellbore corrosion damage evaluation, and corrosion resistance evaluation as the core. The new evaluation method considers the corrosion rate and structural parameters of screen tubing with complex structure, and the overall compliance rate with the experiment is higher than 90%, which provides key support for the evaluation of screen tubing corrosion. It is recommended to utilize the method in similar HTHP environments, regularly update it with new data to enhance accuracy, and implement comprehensive monitoring systems for early detection and intervention.

(3)

The corrosion rate of the screen tube was predicted using actual field data. For a typical Well X, due to the high CO₂ content (4%), the risk of screen tube corrosion is relatively high, and the screen retaining media is a corrosion failure part, with a minimum corrosion life of about 5 years, and the life of the outer protective shroud is about 11–12 years. It is recommended to consider improving the corrosion resistance of the screen in terms of optimizing the structure of the outer protective shroud and the structural parameters of the screen.