Analysis for metal sheath corrosion rate in AC high voltage power cable

Analysis for metal sheath corrosion rate in AC high voltage power cable INTRODUCTIONCorrugated aluminium sheath is a typical metallic sheath used in high‐voltage XLPE (cross‐linked polyethylene) cable. It shall make good electrical contact with the underlying semiconducting layer and provide a concentric conducting path for leakage current and fault current [1, 2]. Corrugated aluminium sheath is widely used in Asian countries due to its excellent mechanical performance [3, 4]. In recent years, a number of cable failures caused by electrical contact defects, which occur at the interface between corrugated aluminium sheath and buffer layer, have been reported. According to the literatures studying the dissection of breakdown cables, pits were found at the crest of corrugated aluminium sheath; the insulated corrosion product was spotted at the surface of the buffer layer (see Figure 1) [5‐7]. It was reported by BoXue Du and Xiaoyun Du that the high‐resistivity corrosion product affects the electric field strength of the contact surface; discharge under this condition is the main cause of the failure [8‐10]. In [11], the authors proposed that corrosion product degrades the electrical contact between aluminium sheath and buffer layer; Current is concentrated at those positions that remain in good electric contact; overheat occurs at these positions and finally causes the ablation. Based on published articles, it can be concluded that insulated corrosion product spoils the electrical contact, leading to the ablation of the interface and failure of cable. However, the corrosion mechanism and factors influencing sheath corrosion rate haven't been fully investigated.1FIGUREDisserted breakdown cable sample. (a) Cross section view of high voltage XLPE power cable; (b) white spots over buffer layer of breakdown cable; (c) corroded surface on the crest of aluminium sheath; (d) ablation trace observed on extruded insulation shieldIn previous literatures, the corrosion mechanism of aluminium sheath has been proposed. In [12], the authors proposed that sodium polyacrylate (NaPA [14]), which is commonly used as superabsorbent polymers (SAP [13]) in buffer layer of high‐voltage cable, absorbs water, hydrolyzes, and creates alkaline environment. Aluminium is corroded in alkaline environment and the interface is covered with insulated corrosion product. Yidong Chen investigated the effects of AC voltage on the corrosion rate. It is concluded that AC voltage can accelerate the corrosion effectively [14]. The corrosion rate of aluminium sheath is mainly affected by corrosive environment. In [15], TiO2 short nanofibers were used to reinforce the corrosion resistance of aluminium. The corrosion resistance of the produced nanocomposites was evaluated by polarization technique in 3.5% NaCl solution. According to Hany S. Abdo et al. [16], chloride ion increases the corrosion rate of steel by breaking the passive layer and the effect of chloride concentration is related to other environment parameters. In [6], large amount of chlorine was observed in corrosion product of the breakdown cables. The source of chlorides content has not been fully investigated. Limited number of studies have been proposed to investigate the corrosive effect of chlorides content on the aluminium sheath in high‐voltage cable. Thus, there is still research gap concerning the source of chloride ion and the corrosive effect of chloride ion on aluminium sheath in high‐voltage cable.The corrosion of aluminium sheath severely affects the lifespan and sustainability of the high‐voltage cable [22]. Therefore, there exist an urgent need to fully investigate the corrosion rate of aluminium sheath in various corrosive environment. The effect of chloride ion on the corrosion rate of aluminium sheath has not been evaluated. There is no specification limiting the chlorides content in high‐voltage power cable. Therefore, an investigation on the effect of chloride ion on corrosion rate of aluminium sheath provides good reference for high quality power cable design.Here, aluminium sheath corrosion experiment is performed to evaluate the influence of chloride ion on the corrosion rate of sheath. GemniSEM 500 scanning electron microscope with EDS is used to study the corroded surface of the sheath. Pits are observed at the crest of aluminium sheath samples. Chloride‐rich phase can be observed closed to the pits and corrosion product. Titration experiment is used to quantify the weight percentage of chlorine in buffer layer. The visual effect of chloride ion on accelerating aluminium sheath corrosion is investigated by comparing the morphologies of corroded surface in different chlorides content environment. Anodic polarization curve of aluminium in different chlorides content electrolytes is measured; pitting potential Epit, corrosion current density Jcorr and polarization resistance Rp of aluminium are investigated to quantify the acceleration effect of chloride ion on aluminium sheath corrosion.THE EFFECT OF CHLORIDES ON ALUMINUM SHEATH CORROSIONThe corrosion mechanism of aluminium sheath in HVAC cableCorrosion that appears at the interface between the corrugated aluminium sheath and the buffer layer in high‐voltage cable can be explained by electrochemical mechanism, which is shown in Figure 2. To guarantee a good electrical contact, the inner crest of aluminium sheath is in direct contact with semiconductive buffer layer. The conductive material in buffer layer is carbon black composite conductive fibres or copper woven fibre glass tape [17, 18]. On the contact surface, aluminium sheath and the conductive fibres serve as two electrodes with different electrochemical potentials and the contact points between them provide an electron transfer path supporting electrochemical reaction. When water penetrates the interface, the corrugated aluminium sheath electrochemically reacts with buffer layer. The electrochemical potential difference between buffer layer and aluminium sheath drives the electrochemical reaction to proceed in a specific direction. When the electrochemical potential of the aluminium sheath is higher than equilibrium potential, aluminium acts as anode and is corroded [19, 20]. Aluminium matrix dissolves to generate aluminium ion; Conductive fibre acts as the cathode where hydrogen evolution reaction occurs. The corrosion process can be described as follows:1Al→Al3++3e−(Anthode)$$\textit{Al}\to A{l}^{3+}+3{e}^{-}(\textit{Anthode})$$22H2O+2e−→2OH−+H2↑(Cathode)$${\rm{2}}{H}_2O + {{2}}{e}^ - \to {\rm{2}}O{H}^ - + {H}_2 \uparrow (Cathode)$$2FIGUREThe electrochemical corrosion mechanism of aluminium sheath in high‐voltage power cableAluminium ion is hydrolyzed in solution to form aluminium hydroxide. When the contact surface is dried, aluminium hydroxide is precipitated and covers the contact surface of aluminium sheath and buffer layer.3Al3++3OH−→Al(OH)3↓$${\mathit{Al}}^{\mathit{3}\mathit{+}}\mathit{+}\mathit{3}{\mathit{OH}}^{\mathit{-}}\mathit{\to}\mathit{Al}{(\mathit{OH})}_{\mathit{3}}\mathit{\downarrow}$$Aluminium is a passive metal. The corrosion of aluminium and its alloys is a multi‐step process [30, 31]. Aggressive anions are firstly adsorbed on passive film on the surface of aluminium. Adsorbed anions react with aluminium ion located in the lattice of aluminium oxide or precipitated aluminium hydroxide to form soluble reaction product, which dissolves and thins the oxide film. Bare aluminium is then exposed to the environment and vulnerable to corrosion, making corrosion develops further. Due to the variation in composition and structural defects, the corrosion rate on metal surface varies from point to point. Usually, corrosion initiates from the points with structural defect.As an aggressive anion, chloride ion can significantly accelerate the electrochemical corrosion of aluminium and its alloys [32]. As shown in Figure 3, during the corrosion process, corroded area acting as anode, chloride ion is adsorbed to the corroded area and concentrated over the oxide film of aluminium. On one hand, the adsorbed chloride ion forms soluble complex product with aluminium oxide, promotes the dissolution of passive film and thus, forms active areas on the surface of aluminium sheath [23]; on the other hand, chloride ion promotes the hydrolysis of aluminium ion, which can acidify the environment. The diffusion of hydrogen ion is restricted by the corrosion product covering the pit. Thus, pH value in the pits is reduced. The acidic environment in pit accelerates the dissolve of aluminium. As the result, the pit expanses and develops deeper. This corrosion process is an autocatalytic one [21]. In different pH environments, chloride ion combines with the aluminium ion in aluminium oxide lattice to form different soluble complexes. In solution of low pH:4Al3++4Cl−→AlCl4−$$A{l}^{3+}+4C{l}^{-}\to \textit{AlC}{l}_{4}^{-}$$in solution of neutral pH:5Al3++2Cl−+2OH−→Al(OH)2Cl2−$$A{l}^{3 + } + 2C{l}^ - + 2O{H}^ - \to Al{(OH)}_2Cl_2^ -$$3FIGUREChloride ion accelerates the development of pitsThe distribution of chlorine throughout corroded areaAn aluminium sheath corrosion experiment was carried out to investigate the behaviour of chlorine in corrosion process and validate the electrochemical corrosion mechanism of aluminium sheath. The corrosion morphology and the distribution of chlorine at the interface between aluminium sheath and buffer layer were observed. Figure 4 shows the experiment setup. The experiment was carried out in an incubator, which kept ambient temperature at 50°C. Corrugated aluminium sheath with 12 cm in length was sampled from a high‐voltage power cable. 500 g load was evenly placed over the aluminium sheath sample to ensure that the contact area between aluminium sheath sample and buffer layer is sufficiently large. To simulate the multi‐layer structure of the buffer layer in cable, the buffer layer used in experiment was composed of buffer tapes with four layers of 0.8 mm thick and one layer of 1.8 mm thick. The tapes did not contain SAP. The chlorine distribution on the surface of buffer tape was investigated before corrosion experiment. As shown in Figure 5, chlorine was uniformly distributed on the tapes. Deionized water was applied to each contact surface by a syringe. In every 24 h, 0.6 mL of deionized water was applied to every contact surface. The sample of corrugated aluminium sheath has five pitches in total. Each pitch forms one contact surface with the buffer layer below. The experiment continued for 96 h. At the end of the experiment, the macroscopic morphology of the corroded surface, the chlorine distribution on aluminium sheath and the corrosion product on buffer tape were recorded using GemniSEM 500 with energy dispersive spectrometer.4FIGUREAluminium sheath corrosion experiment5FIGUREThe distribution of chlorine on buffer tape before the corrosion experiment. It can be observed that chlorine was uniformly distributed on the tapeFigure 6 shows the macroscopic morphology of corrosion area after 96‐h experiment. Corrosion traces appears at the interface between the aluminium sheath and the buffer layer. Corroded area is observed in the form of a grey oval surface, which is entirely covered with corrosion product. The corroded area is about 0.6 cm2. Polygonal brownish‐yellow patches are observed on non‐contact surface. A gap appears between the corroded surface and the brownish‐yellow patches. Corroded product precipitates and adheres to the conductive fibres of buffer tapes, forming white spots on the surface of buffer tapes. The boundary of the white spots coincides with the corroded surface.6FIGUREThe corroded surface is covered by corrosion product. White spots are observed on the surface of buffer tapeThe morphological characteristics and element contain of the experiment sample are compared to the corroded area in a defective cable. The sample was taken from defective cable with corroded aluminium sheath. Before sampling, the defective cable has been conducted a 780‐day heating cycle voltage test. Figure 7 shows SEM and EDS images of the corroded areas under 100 times magnification of scanning electron microscope. Pits appear on both the corroded surfaces of experiment sample and the defective cable sample. Pits observed on the surface of experiment sample are in the form of spherical cavities up to ≈ 230 μm in diameter, and small pits can be observed at some spots with diameter ≈ 50 μm. On the sample of defective cable, pits with diameter ≈ 125 μm are densely distributed throughout the porous, corroded surface. Tables 1 and 2 present the EDS results of the area on buffer tape where corrosion product is found. The corrosion product mainly consists of C, O, Al, Na and Cl, which is in good agreement with the investigations of early researchers [6, 14]. It is concluded that the surface of the buffer tape is covered with corrosion product, which is aluminium hydroxide. Since the experiment sample does not contain SAP, the weight percentage of sodium in EDS result of experiment sample is extremely low; the buffer tape used in the defective cable contains SAP, so sodium appeared in the EDS result. The elements detected on corroded aluminium sheath are similar to those on the surface of the corresponding buffer layer. Sulphur is found on the surface of buffer layer. The non‐water‐blocking buffer tape used in experiment does not contain SAP, thus the buffer tapes cannot create an alkaline environment in damp condition. Without alkaline environment, the chemical corrosion rate of aluminium sheath is low. However, in this experiment, the contact surface between aluminium sheath and buffer layer is severely corroded. It can be concluded that the electrochemical corrosion mainly contributes to the corrosion at the interface in experiment condition.7FIGUREThe SEM and EDS results of the corroded area in experiment sample and defective cable sample. Obvious pitting trace can be observed under 100 times magnification. (a, b) SEM and EDS images of the experiment sample. (c, d) SEM and EDS images of the defect cable sample1TABLEElement content on the buffer tape used in defective cableElementCONaAlSClwt%63.4031.120.144.430.650.272TABLEElement content on the buffer tape used in experimentElementCOAlSClwt%27.1942.5017.970.2112.13Microstructure of the corroded area of experiment and defective cable sample with chlorine distribution mapping are shown in Figure 8. Chlorine distribution on the surface of aluminium sheath (see Figure 8A) indicates that the chlorine‐rich phase presents on the corroded surface. The distribution of chloride‐rich phase is parallel to the development direction of corroded surface. On the corroded surface, an obvious boundary is observed between chlorine‐rich phase and chlorine‐free phase. In Figure 8B, chlorine‐rich phase is observed close to the edge of the pit, and the area far from the pit is chlorine‐free. This result suggests that chlorine is concentrated at the pitting spots. Figure 8C is the distribution of chlorine on the buffer layer of defective cable. Corrosion product is concentrated in the centre of the picture and chlorine is observed mainly in the corrosion product. A small amount of chlorine is detected over conductive fibres which are not covered with corrosion product. Throughout the surface of the buffer layer of experiment sample, corrosion product adheres to conductive fibres and covers the intersections of fibres (see Figure 8D). Chlorine‐rich phase is detected close to these intersections. In general, the distribution of chlorine is nonuniform and chlorine‐rich phase is observed on corroded surface and corrosion product.8FIGUREThe SEM and EDS results of the corroded areas. Chlorine is concentrated in corrosion product and corroded surface. (a) SEM and chlorine distribution images captured at the edge of corroded surface. (b) SEM and chlorine distribution images near the pit. (c) SEM and chlorine distribution images of the white spots in defect cable sample. (d) SEM and chlorine distribution images of the white spots in experiment sampleChlorine was uniformly distributed on buffer tape before the experiment. After the corrosion, chlorine is concentrated in corroded area. This result indicates that a redistribution of chlorine occurs in corrosion process. According to the electrochemical corrosion mechanism of aluminium, anodic area can attract the aggressive anions diffused in electrolyte. The electric field near the surface of anodic area pushes chloride ion to the surface of anode. Chloride ion then participates in reaction to form soluble complex, accelerates the corrosion of aluminium. Therefore, when water evaporates and corroded surface turns dried, the soluble complex is precipitated and covers the corroded area. As a result, chlorine‐rich phase is observed in the area close to corroded surface and corrosion product.The weight percentage of chloride ion in buffer tapeChlorine is attracted to the corroded area and accelerates the corrosion of aluminium sheath. To further investigate the source of chloride ion and quantify the weight percentage of the chlorine which can be ionized in damp condition in the tape, a titration experiment including four steps (see Figure 9) was designed. The experiment process is shown as follows: (1) Cut 2 g of dried buffer tape, put the tape into a beaker containing 200 mL of deionized water as the original sample. Stir the sample with glass rod to disperse the tape. Keep the solution still for 15 min to let chlorine fully ionize in water. (2) Filter out carbon black and fibre in the original sample with funnel and quantitative filter paper. This step ensures that the end point of titration can be evidently observed. (3) Drop 10 mL of the filtered sample into test glass. Drop 0.1 mL of 30% H2O2 and mix for 1 min to eliminate the interferences of sulphide ion [24]. Add two drops of 10 g/L phenolphthalein indicator solution and mix. If the sample turns red (or pink), adjust the pH with sulfuric acid (1+19) to the end point (pH 8.3). (4) Add 1.0 mL of K2CrO4 indicator, titrate the sample with 0.01 mol/L silver nitrate solution. The persistence of brick‐red colour indicates the end point of titration. Meanwhile, titrate a blank. The weight percentage of chloride ion of buffer tapes collected from 5 different manufactures were investigated in this experiment.9FIGUREThe experiment setup used to titrate the weight percentage of chloride ion in buffer tape. Buffer tape samples from five different manufacturers were investigatedIn the titration process, the following chemical reaction occurs in the solution first [25]:6Ag++Cl−→AgCl↓(white)$$A{{\rm{g}}}^ + + C{l}^ - \to AgCl \downarrow (white)$$When chloride ion is completely precipitated, the excess of AgNO3 reacts with K2CrO4 to form brick red precipitation according to the following equation:72Ag++CrO42−→Ag2CrO4↓(brick−red)$$2A{g}^ + {\rm{ + }}C{\rm{r}}{O}_4^{2 - } \to A{g}_2Cr{O}_4 \downarrow (brick - red)$$Re‐sampling the titration from the original sample and repeat the experiment process thrice. The average volume of silver nitrate required for titration is recorded. The mass of chloride ion m in 10 mL sample is calculated as follows:8m=c×(V1−V0)×35.453$$m = c \times ({V}_1 - {V}_0) \times 35.453$$where c is the concentration of standard AgNO3 solution, V1 is the average volume of AgNO3 required for titration the sample, V0 is the volume of AgNO3 required for titration the blank.The weight percentage of ionizable chlorine in 2 g buffer tape, which is immersed in water for 15 min, is given by:9wt%=20×m20×m20002000$$wt\% = {{20 \times m} \mathord{\left/ {\vphantom {{20 \times m} {2000}}} \right. \kern-\nulldelimiterspace} {2000}}$$Buffer tapes produced by five different manufacturers are investigated. The result is shown in Figure 10. Titration result shows that after immersing in deionize water for 15 min, the weight percentages of ionized chloride ion in buffer tapes vary greatly. The highest chloride ion weight percentage in buffer tape is 4.2%, produced by Manufacturer 4. The lowest chloride ion weight percentage in the tape is 0.7%, produced by Manufacturer 1. When contact surface between corrugated aluminium sheath and buffer layer is wet, the increase of the weight percentage of chloride ion in the tape will increase the chlorides content at the interface. Due to the difference in production and storage methods of buffer tape, the weight percentages of chloride ion in tape are different, which in turn affects the quality of buffer tape.10FIGUREThe weight percentage of chloride ion in the buffer tapes produced by different manufactures (Note: “M1” is the abbreviation of “Manufacturer 1″)THE CORROSIVE EFFECTS OF CHLORIDES ON ALUMINUM SHEATHAluminium sheath corrosion experiments in different chlorides content environmentThe corrosion morphologies of aluminium sheath in different chlorides content environments are compared to study the corrosive effect of chlorides content on aluminium sheath. The experiment setup shown in Figure 4 was changed to control the chlorides content in the experiment. Buffer tapes (without SAP) with chloride ion weight percentage less than 0.1% were used. Sodium chloride solutions with chloride ion weight percentages of 0%, 1.2% and 1.8% were injected into the contact surface of aluminium sheath and buffer tape. Each aluminium sheath sample has five contact surfaces with the tape. Each contact surface was injected with 0.2 mL of solution. To ensure that the initial conditions of the contact surface of each group were consistent, the surface of the aluminium sheath was pre‐treated. Each surface was polished with 400Cw‐800Cw‐1500Cw waterproof abrasive papers in turn to remove the passive film and defects on the surface of the sheath. Clean the debris with deionized water, rinse the surface with alcohol and rinse it again with deionized water, and bake the pretreated samples to remove moisture. Keep ambient temperature at 19°C. The corrosion morphologies of the contact surface in different chlorides content groups were recorded. The development process of pits on the surface of corrugated aluminium sheath sample was observed. The corrosion rate of aluminium sheaths in different chlorides content environments was compared and the development of aluminium sheath corrosion was summarized.The development of corroded surface is shown in Figure 11. After 1.5 h experiment, pits are observed on the contact surface of the 1.8 wt% group (as shown in Figure 11a). In the 1.2 wt% group, pits are observed at the edge of contact surface and in the form of stripes (as shown in Figure 11b). There is no obvious corrosion trace in the centre of the contact surface. No corrosion trace is observed on the surface of 0 wt% group (as shown in Figure 11c).11FIGUREThe development of corroded surface in different chlorides content environments. (a–c) Corrosion images of aluminium in environment with 1.8, 1.2 and 0 wt% chlorides respectively after 1.5 h experiment. (d–f) Corrosion images after 4 h experiment. (g–i) Corrosion images after 52.5 h experimentAfter 4 h experiment, the surface of the sheath in 1.8% wt group is covered with pits. The corroded surface lost its metallic lustre (as shown in Figure 11d) In 1.2 wt% group, corrosion mainly occurs at the edge of the contact surface, and pits further develop along the edge, which is shown in Figure 11e. In 0 wt% group, pits are sparsely distributed on the contact surface (as shown in Figure 11f).After 52.5 h experiment, cracks are observed on the contact surface of the 1.8% wt group (as shown in Figure 10g). In 1.2% group, corrosion pits can be observed in the centre of contact surface; At the edge of contact surface, the strip formed by pits expands (as shown in Figure 11h). In 0 wt% group, aluminium sheath sample maintains metallic lustre. Compared with the corroded surface after 4 h experiment, the number of pits on the contact surface increases significantly; pits are randomly distributed over the entire contact surface and grows larger and deeper (as shown in Figure 11i).This result indicates that aluminium sheath is vulnerable in chloride content environment [32–34]. Chloride ion accumulates in the active area, invades the defects in the lattice of passive film. This process destroys the passive film and exposes aluminium matrix. Thus, the area of passive film is reduced and the area where corrosion reaction can occur is increased. This results in higher corrosion rate of aluminium sheath. Finally, larger corroded area with obvious pits and crack can be observed in the sample with higher chloride content. The corrosion of aluminium sheath initially appears in the form of local pitting. With the development of corrosion, the number of pits continue to increase. As pitting process continues to expand and deepen, the corrosion area covers most part of the contact surface. Cracks can appear in the area severely corroded. Finally, when moisture is removed from the surface, corrosion product aluminium hydroxide is precipitated and adheres to the contact surface to form white spots, which is observed in previous studies [6, 14].The effect of chlorides on the polarization curve of aluminiumThe experiment shown in Figure 12 was conducted to further investigate the corrosion rate of aluminium in different chlorides content environments. Polarization curve of aluminium in different chloride content were measured and the characteristic parameters were used to quantitatively describe the corrosion rate of aluminium [26, 27]. The weight percentages of chloride ion in electrolytes were 0.6, 1.2 and 1.8 wt%, respectively and the solutions were neutral. An aluminium plate with a diameter of 6 mm was used as working electrode and the exposed area of the working electrode was 0.283 cm2; the rest of the electrode was covered with insulator. A calomel electrode was used as the reference electrode to reduce the interference of the ohmic resistance of solution [28]. A platinum plate was used as counter electrode. Thus, the polarization current flows between working electrode and counter electrode. The exposed aluminium surface was successively polished with waterproof abrasive papers of 400#, 800# and 1500#. Deionized water and alcohol were used to resin the polished surface. The ambient temperature was 20°C. Anodic polarization curves of working electrode in electrolytes with different chloride ion weight percentage were recorded by an AUTOLAB‐PGSTAT302 electrochemical workstation and the scanning speed was 5 mV/s.12FIGUREThe polarization curve of aluminium in different chloride ion content electrolytes were measured using a conventional tri‐electrode electrochemical cellAn anodic polarization curve includes four regions: the activation region, activation‐passivation region, passivation region and transpassivation region [29, 20]. In the activation region, polarization current increases with the increase of electrode potential. At this stage, Tafel equation is used to describes the relationship between the applied potential and polarization current. The corrosion current density obtained from the polarization curve can directly reflect the corrosion rate of the working electrode. The higher the corrosion current density Jcorr, the faster the corrosion of the surface of working electrode in the electrolytic cell environment. The polarization corrosion current Icorr and the corresponding corrosion current density Jcorr can be calculated according to the Stern‐Geary formula:10Icorr=1Rp×βa×βc2.3(βa+βc)$${I}_{corr} = \frac{1}{{{R}_p}} \times \frac{{{\beta }_a \times {\beta }_c}}{{2.3({\beta }_a + {\beta }_c)}}$$11Jcorr=Icorr/S$${J}_{corr} = {I}_{corr}/S$$where Rp is the linear polarization resistance, βa and βc are Tafel constants. The linear polarization resistance and Tafel constant can be obtained by linear fitting the activation region on the polarization curve. When electrode potential increases and reaches the passivation potential, polarization curve enters the passivation region. At this stage, the passive film formed on the surface of aluminium inhibits electrons from passing the surface of aluminium, thus prevents the corrosion of aluminium. As a result, in passivation region, the increase of electrode potential has little effect on the polarization current. As electrode potential further increases and exceeds the pitting potential Epit, the passive film on the metal surface breaks, and polarization current increases significantly with the increase of electrode potential. Parameters such as corrosion current density, pitting potential, and the length of passivation region which are related to aluminium corrosion resistance in the specific environment can be obtained from the polarization curve.The effect of chlorides content on polarization curve of aluminium is shown in Figure 13. The Tafel constant, linear polarization resistance Rp, corrosion current density Jcorr and pitting potential Epit of aluminium in different chlorides content environment are shown in Table 3. As shown in Figure 13, the increase of the weight percentage of chloride ion causes a left shift of Epit. This result indicates that aluminium can be easily corroded in environment with high chlorides content. A passivation region ranging from −0.9 to ‐0.7 V is observed in the group with 0.6 wt%. At this stage, polarization current changes slowly with the increase of electrode potential. However, when chlorides content increases to 1.2 wt%, the length of the passivation region is significantly shortened.13FIGUREThe polarization curves of aluminium in different chlorides content electrolytes3TABLECorrosion parameters obtained from the polarization curves of aluminium in different chlorides content electrolyteswt%βaβcRp (MΩ)Jcorr (nA/mm2)Epit (V/SCE)0.60.07000.05680.22648212.4−0.6891.20.25160.28330.133561531.3−0.7241.80.44570.93420.089985145.9−0.742As the weight percentage of chloride ion increases, Jcorr increases significantly. When 0.6 wt% chloride ion is applied, the Jcorr and Epit are 212.4 nA/mm and −0.657 V, respectively. When chlorides content increases to 1.2 wt%, the Jcorr of aluminium increases to 1531.3 nA/mm and the pitting potential is −0.698 V. When chlorides content increases to 1.8 wt%, Jcorr further increases to 5145.9 nA/mm and the pitting potential is −0.719 V. With the increase of chlorides content, the pitting potential of aluminium shifts left, the anode potential required to enter transpassivation region decreases which indicates that the corrosion of aluminium is easier. Therefore, the increase of chlorides content affects the corrosion resistance of aluminium.The pitting potential and the corrosion current density are highly related to the chloride concentration of the solution. A high chloride concentration leads to a high corrosion current density. The passive region on the curve disappears when the chloride concentration continues to increase. This result is attributed to the degradation of the passive films in high chloride concentration solution. The passive films are limited in the present of chloride ion, which results in deterioration of corrosion properties.CONCLUSIONCorrosion of corrugated aluminium sheath seriously affects the lifespan of high‐voltage power cable. An investigation of the influence factors affecting corrosion rate of aluminium sheath provides a good reference for the design of high‐quality cable. Here, the aluminium sheath corrosion mechanism, the corrosive effect of chlorines content and the source of chlorides content are studied. The effect of chloride ion on accelerating sheath corrosion is validated and evaluated by experiments. The result can be concluded as follows:(1) Electrochemical corrosion occurs at the interface between corrugated aluminium sheath and semiconducting buffer layer. Chloride ion can accelerate the corrosion of aluminium sheath. The EDS images of the experiment sample shows that buffer layer is the source of chlorine. The chlorides content is concentrated at corroded surface in corrosion process. The concentration of chlorine implies that electrochemical corrosion occurs at the interface between aluminium sheath and buffer layer.(2) After being immersed in deionized water for 15 min, the content of ionizable chlorine in buffer layers varies. The titration result shows that the weight percentage of chloride ion in buffer tape varies from 4.2% to 0.7%. This result indicates that the buffer layers in high voltage power cable is one of the sources of chlorides content. In damp condition, the weight percentage of ionizable chlorine in buffer tape will directly influence the concentration of chloride ion at the interface and thus, affects the corrosion rate of aluminium sheath.(3) Corrosion experiment and polarization curves in different chlorides content electrolyte show that when the concentration of chlorides increases, the corrosion rate of aluminium sheath increases. When the chlorides content increases from 0.6 to 1.8 wt%, pitting potential Epit changes from −0.689 to −0.742 V. With the increase of chloride concentration, he corrosion current density Jcorr increases significantly, and the length of passive region decreases. Aluminium sheath is susceptible to local pitting in chloride‐rich solution.Corrosion of corrugated aluminium sheath severely degrades the electric contact between aluminium sheath and buffer layer. In chloride‐rich environment, aluminium sheath is susceptible to corrosion. Buffer layer has been proved to be one of the sources of chlorides content. Nevertheless, the source of chlorine has not been fully investigated, future research investigating the source of chloride ion can contribute to the reduce of chlorides content in high‐voltage cable and therefore, reduce the corrosion rate of aluminium sheath. Methods have been proposed to improve the corrosion resistance of metal. However, the application of these methods in high‐voltage cable metallic sheath design has not been fully discussed. The relationship between corrosion of aluminium sheath and lifetime of cable is uncertain. The further study considering life assessment of aluminium sheath will help ensure the maintenance and availability of cable in service.CONFLICT OF INTERESTThe authors declare that there are no conflicts of interest regarding the publication of this paper.FUNDING INFORMATIONThere is no funding to support for this submission.DATA AVAILABILITY STATEMENTThe data that support the findings of this study are available from the corresponding author upon reasonable request.AUTHOR CONTRIBUTIONSL.L.: Software, data curation, formal analysis, validation, writing—Original draft.Z.W.: Software, formal analysis, visualization.X.L.: Methodology, resources, supervision.Y.H.: Methodology, resources, writing—Review and editing.G.L.: Conceptualization, methodology, resources, supervision, writing—Review and editing.REFERENCESStandard for extruded insulation power cables rated above 46 through 500 kV, ICEA S‐108‐720, 2018Specification for extruded insulation power cables and their accessories rated above 46kV through 345kV, AEIC Standard CS9‐15, (2015)Nelson, R.A., Daly, J.M.: Corrugated metallic sheathed cable–Design and applications. IEEE Trans. Ind. Appl. 11(2):196–203 (1975)Liu, Y., Chen, J.: Design of smooth aluminum bonded sheaths in HV XLPE cables based on the bending performance research. IEEE Access 8:102493–102501 (2020)Song, P., et al.: A case study on ablation breakdown of high voltage cable buffer layer. In: 2020 IEEE International Conference on High Voltage Engineering and Application (ICHVE), Beijing, China, pp. 1–4 (2020)Chen, Y., et al.: Failure investigation of buffer layers in high‐voltage XLPE cables. Eng. Failure Anal. 113(10), 104546 (2020)Jiang, L., et al.: Study on ablation between metal sheath and buffer layer of high voltage XLPE insulated power cable. In: 2019 2nd International Conference on Electrical Materials and Power Equipment (ICEMPE), Guangzhou, Guangdong, China, pp. 372–375 (2019)Du, B., et al.: An investigation on discharge fault in buffer layer of 220 kV XLPE AC cable. IET Sci. Meas. Technol. 15:508–516 (2021)Li, Z.J., et al.: Effects of the environmental factors on electrical properties of high voltage cable buffer layer. In: 2020 IEEE 3rd International Conference on Dielectrics (ICD), Valencia, Spain, pp. 546–549 (2020)Cao, J., et al.: Analysis on buffer layer discharges below the corrugated aluminum sheath of XLPE cables and comparison with other metal sheath structures. In: 2019 IEEE 3rd International Conference on Circuits, Systems and Devices (ICCSD), Chengdu, China, pp. 21–26 (2019)Xin, Y., et al.: Cause analysis of aging ablation on sheath of 110 kV single core high voltage cable. In: 2019 IEEE Conference on Electrical Insulation and Dielectric Phenomena (CEIDP), Richland, WA, USA, pp. 162–165 (2019)Liu, Y., et al.: The formation mechanism of white substances on water‐blocking buffer layer of HV XLPE cables. In: 2020 IEEE International Conference on High Voltage Engineering and Application (ICHVE), Beijing, China, pp. 1–4 (2020)Nordas, S., Helleso, S.M., Hvidsten, S.: Measurements and modeling of water diffusion in water blocking tapes for high voltage extruded cables. In: 2008 International Symposium on Electrical Insulating Materials (ISEIM 2008), Yokkaichi, Japan, pp. 263–266 (2008)Chen, Y., et al.: Hydrogen evolution and electromigration in the corrosion of aluminium metal sheath inside high‐voltage cables. High Volt. 7:260–268 (2022)Abdo, H.S., et al.: Electrochemical behavior of inductively sintered Al/TiO2 nanocomposites reinforced by electrospun ceramic nanofibers. Polymers 13(24):4319 (2021)Abdo, H.S., et al.: Microstructural characterization and corrosion‐resistance behavior of dual‐phase steels compared to conventional rebar. Crystals 10, 1068 (2020)Chen, Y., et al.: Effects of connection conditions between insulation screen and Al sheath on the buffer layer failures of high‐voltage XLPE cables. Eng. Failure Anal. 122, 105263 (2021)Su, C.Q.: Failure analysis of three 230kV XLPE cables. In: 2010 IEEE/PES Transmission and Distribution Conference and Exposition: Latin America (T&D‐LA), Sao Paulo, Brazil, pp. 22–25 (2010)Ibrahim, E.S: Corrosion control in electric power systems. Electr. Power Syst. Res. 52, 9–17 (1999)Zhang, B., et al.: Numerical approach of impressed potential on buried pipelines near high‐voltage DC grounding electrodes. IET Gener. Transm. Distrib. 12(5):1177–1182 (2018)Bhandari, J., et al.: Modelling of pitting corrosion in marine and offshore steel structures—A technical review. J. Loss Prev. Process Ind. 37:39–62 (2015)Korzhov, A.V.: Effects of magnetic field on cable sheath corrosion. Eng. Failure Anal. 116, 104749 (2020)Tian, H., Liu, P., Zhou, S., Wang, Q., Wu, Z., Zhang, J., Peng, Z.: Research on the deterioration process of electrical contact structure inside the ±500 kV converter transformer RIP bushings and its prediction strategy. IET Gener. Transm. Distrib. 13(12):2391–2400 (2019)Standard Test Methods for Chloride Ion in Water, ASTM D 512‐12, 2012Trejo, D., Ahmed, A.A.: Adopting auto‐titration to assess chlorides in concrete. ACI Mater. J. 116(3), 43‐52 (2019)Cui, L., et al.: Experimental study on multi‐component corrosion inhibitor for steel bar in chloride environment. Constr. Build. Mater. 313, 125533 (2021)Zhang, S., et al.: Experimental data treatment of the pipeline steel polarization curve under AC interference. Int. J. Electrochem. Sci. 14(12):10888–10906 (2020)Lee, S.E., Tang, MH. ‘Reliable reference electrodes for nonaqueous sodium‐ion batteries. Int. J. Electrochem. Sci. 166(14):3260–3264 (2019)Zhou, Y., et al.: The relationship between activation‐passivation transition and grain boundary dissolution on four steel samples in acidic solutions containing NO2−. Faraday Discuss. R. Soc. Chem. 9(41), 23589–23597 (2019)Hung, T.: On the mechanism of pitting of aluminium. J. Electrochem. Soc. 126:1855–1860 (1979)Calderón, J.P., et al.: Corrosion behavior of Al modified with Zn in chloride solution. Materials 15(12):4229 (2022)Senderowski, C., et al.: Effect of low chloride and sulfate concentrations on corrosion behavior of aluminum and zinc arc thermal sprayed coatings. Coatings 12(5):653 (2022)Calderón, J.P. et al. ‘Corrosion behavior of Al modified with Zn in chloride solution’. Materials 15(12), 4229 (2022)Senderowski, C. et al. ‘Effect of low chloride and sulfate concentrations on corrosion behavior of aluminum and zinc arc thermal sprayed coatings’. Coatings 12(5), 653 (2022)AAPPENDIXFLOW CHART FOR THE RESEARCH METHODOLOGY http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png "IET Generation, Transmission & Distribution" Wiley

Analysis for metal sheath corrosion rate in AC high voltage power cable

, Volume 16 (22) – Nov 1, 2022
12 pages

/lp/wiley/analysis-for-metal-sheath-corrosion-rate-in-ac-high-voltage-power-FVW1f2Ly0g
Publisher
Wiley
© 2022 The Institution of Engineering and Technology.
eISSN
1751-8695
DOI
10.1049/gtd2.12620
Publisher site
See Article on Publisher Site

Abstract

INTRODUCTIONCorrugated aluminium sheath is a typical metallic sheath used in high‐voltage XLPE (cross‐linked polyethylene) cable. It shall make good electrical contact with the underlying semiconducting layer and provide a concentric conducting path for leakage current and fault current [1, 2]. Corrugated aluminium sheath is widely used in Asian countries due to its excellent mechanical performance [3, 4]. In recent years, a number of cable failures caused by electrical contact defects, which occur at the interface between corrugated aluminium sheath and buffer layer, have been reported. According to the literatures studying the dissection of breakdown cables, pits were found at the crest of corrugated aluminium sheath; the insulated corrosion product was spotted at the surface of the buffer layer (see Figure 1) [5‐7]. It was reported by BoXue Du and Xiaoyun Du that the high‐resistivity corrosion product affects the electric field strength of the contact surface; discharge under this condition is the main cause of the failure [8‐10]. In [11], the authors proposed that corrosion product degrades the electrical contact between aluminium sheath and buffer layer; Current is concentrated at those positions that remain in good electric contact; overheat occurs at these positions and finally causes the ablation. Based on published articles, it can be concluded that insulated corrosion product spoils the electrical contact, leading to the ablation of the interface and failure of cable. However, the corrosion mechanism and factors influencing sheath corrosion rate haven't been fully investigated.1FIGUREDisserted breakdown cable sample. (a) Cross section view of high voltage XLPE power cable; (b) white spots over buffer layer of breakdown cable; (c) corroded surface on the crest of aluminium sheath; (d) ablation trace observed on extruded insulation shieldIn previous literatures, the corrosion mechanism of aluminium sheath has been proposed. In [12], the authors proposed that sodium polyacrylate (NaPA [14]), which is commonly used as superabsorbent polymers (SAP [13]) in buffer layer of high‐voltage cable, absorbs water, hydrolyzes, and creates alkaline environment. Aluminium is corroded in alkaline environment and the interface is covered with insulated corrosion product. Yidong Chen investigated the effects of AC voltage on the corrosion rate. It is concluded that AC voltage can accelerate the corrosion effectively [14]. The corrosion rate of aluminium sheath is mainly affected by corrosive environment. In [15], TiO2 short nanofibers were used to reinforce the corrosion resistance of aluminium. The corrosion resistance of the produced nanocomposites was evaluated by polarization technique in 3.5% NaCl solution. According to Hany S. Abdo et al. [16], chloride ion increases the corrosion rate of steel by breaking the passive layer and the effect of chloride concentration is related to other environment parameters. In [6], large amount of chlorine was observed in corrosion product of the breakdown cables. The source of chlorides content has not been fully investigated. Limited number of studies have been proposed to investigate the corrosive effect of chlorides content on the aluminium sheath in high‐voltage cable. Thus, there is still research gap concerning the source of chloride ion and the corrosive effect of chloride ion on aluminium sheath in high‐voltage cable.The corrosion of aluminium sheath severely affects the lifespan and sustainability of the high‐voltage cable [22]. Therefore, there exist an urgent need to fully investigate the corrosion rate of aluminium sheath in various corrosive environment. The effect of chloride ion on the corrosion rate of aluminium sheath has not been evaluated. There is no specification limiting the chlorides content in high‐voltage power cable. Therefore, an investigation on the effect of chloride ion on corrosion rate of aluminium sheath provides good reference for high quality power cable design.Here, aluminium sheath corrosion experiment is performed to evaluate the influence of chloride ion on the corrosion rate of sheath. GemniSEM 500 scanning electron microscope with EDS is used to study the corroded surface of the sheath. Pits are observed at the crest of aluminium sheath samples. Chloride‐rich phase can be observed closed to the pits and corrosion product. Titration experiment is used to quantify the weight percentage of chlorine in buffer layer. The visual effect of chloride ion on accelerating aluminium sheath corrosion is investigated by comparing the morphologies of corroded surface in different chlorides content environment. Anodic polarization curve of aluminium in different chlorides content electrolytes is measured; pitting potential Epit, corrosion current density Jcorr and polarization resistance Rp of aluminium are investigated to quantify the acceleration effect of chloride ion on aluminium sheath corrosion.THE EFFECT OF CHLORIDES ON ALUMINUM SHEATH CORROSIONThe corrosion mechanism of aluminium sheath in HVAC cableCorrosion that appears at the interface between the corrugated aluminium sheath and the buffer layer in high‐voltage cable can be explained by electrochemical mechanism, which is shown in Figure 2. To guarantee a good electrical contact, the inner crest of aluminium sheath is in direct contact with semiconductive buffer layer. The conductive material in buffer layer is carbon black composite conductive fibres or copper woven fibre glass tape [17, 18]. On the contact surface, aluminium sheath and the conductive fibres serve as two electrodes with different electrochemical potentials and the contact points between them provide an electron transfer path supporting electrochemical reaction. When water penetrates the interface, the corrugated aluminium sheath electrochemically reacts with buffer layer. The electrochemical potential difference between buffer layer and aluminium sheath drives the electrochemical reaction to proceed in a specific direction. When the electrochemical potential of the aluminium sheath is higher than equilibrium potential, aluminium acts as anode and is corroded [19, 20]. Aluminium matrix dissolves to generate aluminium ion; Conductive fibre acts as the cathode where hydrogen evolution reaction occurs. The corrosion process can be described as follows:1Al→Al3++3e−(Anthode)$$\textit{Al}\to A{l}^{3+}+3{e}^{-}(\textit{Anthode})$$22H2O+2e−→2OH−+H2↑(Cathode)$${\rm{2}}{H}_2O + {{2}}{e}^ - \to {\rm{2}}O{H}^ - + {H}_2 \uparrow (Cathode)$$2FIGUREThe electrochemical corrosion mechanism of aluminium sheath in high‐voltage power cableAluminium ion is hydrolyzed in solution to form aluminium hydroxide. When the contact surface is dried, aluminium hydroxide is precipitated and covers the contact surface of aluminium sheath and buffer layer.3Al3++3OH−→Al(OH)3↓$${\mathit{Al}}^{\mathit{3}\mathit{+}}\mathit{+}\mathit{3}{\mathit{OH}}^{\mathit{-}}\mathit{\to}\mathit{Al}{(\mathit{OH})}_{\mathit{3}}\mathit{\downarrow}$$Aluminium is a passive metal. The corrosion of aluminium and its alloys is a multi‐step process [30, 31]. Aggressive anions are firstly adsorbed on passive film on the surface of aluminium. Adsorbed anions react with aluminium ion located in the lattice of aluminium oxide or precipitated aluminium hydroxide to form soluble reaction product, which dissolves and thins the oxide film. Bare aluminium is then exposed to the environment and vulnerable to corrosion, making corrosion develops further. Due to the variation in composition and structural defects, the corrosion rate on metal surface varies from point to point. Usually, corrosion initiates from the points with structural defect.As an aggressive anion, chloride ion can significantly accelerate the electrochemical corrosion of aluminium and its alloys [32]. As shown in Figure 3, during the corrosion process, corroded area acting as anode, chloride ion is adsorbed to the corroded area and concentrated over the oxide film of aluminium. On one hand, the adsorbed chloride ion forms soluble complex product with aluminium oxide, promotes the dissolution of passive film and thus, forms active areas on the surface of aluminium sheath [23]; on the other hand, chloride ion promotes the hydrolysis of aluminium ion, which can acidify the environment. The diffusion of hydrogen ion is restricted by the corrosion product covering the pit. Thus, pH value in the pits is reduced. The acidic environment in pit accelerates the dissolve of aluminium. As the result, the pit expanses and develops deeper. This corrosion process is an autocatalytic one [21]. In different pH environments, chloride ion combines with the aluminium ion in aluminium oxide lattice to form different soluble complexes. In solution of low pH:4Al3++4Cl−→AlCl4−$$A{l}^{3+}+4C{l}^{-}\to \textit{AlC}{l}_{4}^{-}$$in solution of neutral pH:5Al3++2Cl−+2OH−→Al(OH)2Cl2−$$A{l}^{3 + } + 2C{l}^ - + 2O{H}^ - \to Al{(OH)}_2Cl_2^ -$$3FIGUREChloride ion accelerates the development of pitsThe distribution of chlorine throughout corroded areaAn aluminium sheath corrosion experiment was carried out to investigate the behaviour of chlorine in corrosion process and validate the electrochemical corrosion mechanism of aluminium sheath. The corrosion morphology and the distribution of chlorine at the interface between aluminium sheath and buffer layer were observed. Figure 4 shows the experiment setup. The experiment was carried out in an incubator, which kept ambient temperature at 50°C. Corrugated aluminium sheath with 12 cm in length was sampled from a high‐voltage power cable. 500 g load was evenly placed over the aluminium sheath sample to ensure that the contact area between aluminium sheath sample and buffer layer is sufficiently large. To simulate the multi‐layer structure of the buffer layer in cable, the buffer layer used in experiment was composed of buffer tapes with four layers of 0.8 mm thick and one layer of 1.8 mm thick. The tapes did not contain SAP. The chlorine distribution on the surface of buffer tape was investigated before corrosion experiment. As shown in Figure 5, chlorine was uniformly distributed on the tapes. Deionized water was applied to each contact surface by a syringe. In every 24 h, 0.6 mL of deionized water was applied to every contact surface. The sample of corrugated aluminium sheath has five pitches in total. Each pitch forms one contact surface with the buffer layer below. The experiment continued for 96 h. At the end of the experiment, the macroscopic morphology of the corroded surface, the chlorine distribution on aluminium sheath and the corrosion product on buffer tape were recorded using GemniSEM 500 with energy dispersive spectrometer.4FIGUREAluminium sheath corrosion experiment5FIGUREThe distribution of chlorine on buffer tape before the corrosion experiment. It can be observed that chlorine was uniformly distributed on the tapeFigure 6 shows the macroscopic morphology of corrosion area after 96‐h experiment. Corrosion traces appears at the interface between the aluminium sheath and the buffer layer. Corroded area is observed in the form of a grey oval surface, which is entirely covered with corrosion product. The corroded area is about 0.6 cm2. Polygonal brownish‐yellow patches are observed on non‐contact surface. A gap appears between the corroded surface and the brownish‐yellow patches. Corroded product precipitates and adheres to the conductive fibres of buffer tapes, forming white spots on the surface of buffer tapes. The boundary of the white spots coincides with the corroded surface.6FIGUREThe corroded surface is covered by corrosion product. White spots are observed on the surface of buffer tapeThe morphological characteristics and element contain of the experiment sample are compared to the corroded area in a defective cable. The sample was taken from defective cable with corroded aluminium sheath. Before sampling, the defective cable has been conducted a 780‐day heating cycle voltage test. Figure 7 shows SEM and EDS images of the corroded areas under 100 times magnification of scanning electron microscope. Pits appear on both the corroded surfaces of experiment sample and the defective cable sample. Pits observed on the surface of experiment sample are in the form of spherical cavities up to ≈ 230 μm in diameter, and small pits can be observed at some spots with diameter ≈ 50 μm. On the sample of defective cable, pits with diameter ≈ 125 μm are densely distributed throughout the porous, corroded surface. Tables 1 and 2 present the EDS results of the area on buffer tape where corrosion product is found. The corrosion product mainly consists of C, O, Al, Na and Cl, which is in good agreement with the investigations of early researchers [6, 14]. It is concluded that the surface of the buffer tape is covered with corrosion product, which is aluminium hydroxide. Since the experiment sample does not contain SAP, the weight percentage of sodium in EDS result of experiment sample is extremely low; the buffer tape used in the defective cable contains SAP, so sodium appeared in the EDS result. The elements detected on corroded aluminium sheath are similar to those on the surface of the corresponding buffer layer. Sulphur is found on the surface of buffer layer. The non‐water‐blocking buffer tape used in experiment does not contain SAP, thus the buffer tapes cannot create an alkaline environment in damp condition. Without alkaline environment, the chemical corrosion rate of aluminium sheath is low. However, in this experiment, the contact surface between aluminium sheath and buffer layer is severely corroded. It can be concluded that the electrochemical corrosion mainly contributes to the corrosion at the interface in experiment condition.7FIGUREThe SEM and EDS results of the corroded area in experiment sample and defective cable sample. Obvious pitting trace can be observed under 100 times magnification. (a, b) SEM and EDS images of the experiment sample. (c, d) SEM and EDS images of the defect cable sample1TABLEElement content on the buffer tape used in defective cableElementCONaAlSClwt%63.4031.120.144.430.650.272TABLEElement content on the buffer tape used in experimentElementCOAlSClwt%27.1942.5017.970.2112.13Microstructure of the corroded area of experiment and defective cable sample with chlorine distribution mapping are shown in Figure 8. Chlorine distribution on the surface of aluminium sheath (see Figure 8A) indicates that the chlorine‐rich phase presents on the corroded surface. The distribution of chloride‐rich phase is parallel to the development direction of corroded surface. On the corroded surface, an obvious boundary is observed between chlorine‐rich phase and chlorine‐free phase. In Figure 8B, chlorine‐rich phase is observed close to the edge of the pit, and the area far from the pit is chlorine‐free. This result suggests that chlorine is concentrated at the pitting spots. Figure 8C is the distribution of chlorine on the buffer layer of defective cable. Corrosion product is concentrated in the centre of the picture and chlorine is observed mainly in the corrosion product. A small amount of chlorine is detected over conductive fibres which are not covered with corrosion product. Throughout the surface of the buffer layer of experiment sample, corrosion product adheres to conductive fibres and covers the intersections of fibres (see Figure 8D). Chlorine‐rich phase is detected close to these intersections. In general, the distribution of chlorine is nonuniform and chlorine‐rich phase is observed on corroded surface and corrosion product.8FIGUREThe SEM and EDS results of the corroded areas. Chlorine is concentrated in corrosion product and corroded surface. (a) SEM and chlorine distribution images captured at the edge of corroded surface. (b) SEM and chlorine distribution images near the pit. (c) SEM and chlorine distribution images of the white spots in defect cable sample. (d) SEM and chlorine distribution images of the white spots in experiment sampleChlorine was uniformly distributed on buffer tape before the experiment. After the corrosion, chlorine is concentrated in corroded area. This result indicates that a redistribution of chlorine occurs in corrosion process. According to the electrochemical corrosion mechanism of aluminium, anodic area can attract the aggressive anions diffused in electrolyte. The electric field near the surface of anodic area pushes chloride ion to the surface of anode. Chloride ion then participates in reaction to form soluble complex, accelerates the corrosion of aluminium. Therefore, when water evaporates and corroded surface turns dried, the soluble complex is precipitated and covers the corroded area. As a result, chlorine‐rich phase is observed in the area close to corroded surface and corrosion product.The weight percentage of chloride ion in buffer tapeChlorine is attracted to the corroded area and accelerates the corrosion of aluminium sheath. To further investigate the source of chloride ion and quantify the weight percentage of the chlorine which can be ionized in damp condition in the tape, a titration experiment including four steps (see Figure 9) was designed. The experiment process is shown as follows: (1) Cut 2 g of dried buffer tape, put the tape into a beaker containing 200 mL of deionized water as the original sample. Stir the sample with glass rod to disperse the tape. Keep the solution still for 15 min to let chlorine fully ionize in water. (2) Filter out carbon black and fibre in the original sample with funnel and quantitative filter paper. This step ensures that the end point of titration can be evidently observed. (3) Drop 10 mL of the filtered sample into test glass. Drop 0.1 mL of 30% H2O2 and mix for 1 min to eliminate the interferences of sulphide ion [24]. Add two drops of 10 g/L phenolphthalein indicator solution and mix. If the sample turns red (or pink), adjust the pH with sulfuric acid (1+19) to the end point (pH 8.3). (4) Add 1.0 mL of K2CrO4 indicator, titrate the sample with 0.01 mol/L silver nitrate solution. The persistence of brick‐red colour indicates the end point of titration. Meanwhile, titrate a blank. The weight percentage of chloride ion of buffer tapes collected from 5 different manufactures were investigated in this experiment.9FIGUREThe experiment setup used to titrate the weight percentage of chloride ion in buffer tape. Buffer tape samples from five different manufacturers were investigatedIn the titration process, the following chemical reaction occurs in the solution first [25]:6Ag++Cl−→AgCl↓(white)$$A{{\rm{g}}}^ + + C{l}^ - \to AgCl \downarrow (white)$$When chloride ion is completely precipitated, the excess of AgNO3 reacts with K2CrO4 to form brick red precipitation according to the following equation:72Ag++CrO42−→Ag2CrO4↓(brick−red)$$2A{g}^ + {\rm{ + }}C{\rm{r}}{O}_4^{2 - } \to A{g}_2Cr{O}_4 \downarrow (brick - red)$$Re‐sampling the titration from the original sample and repeat the experiment process thrice. The average volume of silver nitrate required for titration is recorded. The mass of chloride ion m in 10 mL sample is calculated as follows:8m=c×(V1−V0)×35.453$$m = c \times ({V}_1 - {V}_0) \times 35.453$$where c is the concentration of standard AgNO3 solution, V1 is the average volume of AgNO3 required for titration the sample, V0 is the volume of AgNO3 required for titration the blank.The weight percentage of ionizable chlorine in 2 g buffer tape, which is immersed in water for 15 min, is given by:9wt%=20×m20×m20002000$$wt\% = {{20 \times m} \mathord{\left/ {\vphantom {{20 \times m} {2000}}} \right. \kern-\nulldelimiterspace} {2000}}$$Buffer tapes produced by five different manufacturers are investigated. The result is shown in Figure 10. Titration result shows that after immersing in deionize water for 15 min, the weight percentages of ionized chloride ion in buffer tapes vary greatly. The highest chloride ion weight percentage in buffer tape is 4.2%, produced by Manufacturer 4. The lowest chloride ion weight percentage in the tape is 0.7%, produced by Manufacturer 1. When contact surface between corrugated aluminium sheath and buffer layer is wet, the increase of the weight percentage of chloride ion in the tape will increase the chlorides content at the interface. Due to the difference in production and storage methods of buffer tape, the weight percentages of chloride ion in tape are different, which in turn affects the quality of buffer tape.10FIGUREThe weight percentage of chloride ion in the buffer tapes produced by different manufactures (Note: “M1” is the abbreviation of “Manufacturer 1″)THE CORROSIVE EFFECTS OF CHLORIDES ON ALUMINUM SHEATHAluminium sheath corrosion experiments in different chlorides content environmentThe corrosion morphologies of aluminium sheath in different chlorides content environments are compared to study the corrosive effect of chlorides content on aluminium sheath. The experiment setup shown in Figure 4 was changed to control the chlorides content in the experiment. Buffer tapes (without SAP) with chloride ion weight percentage less than 0.1% were used. Sodium chloride solutions with chloride ion weight percentages of 0%, 1.2% and 1.8% were injected into the contact surface of aluminium sheath and buffer tape. Each aluminium sheath sample has five contact surfaces with the tape. Each contact surface was injected with 0.2 mL of solution. To ensure that the initial conditions of the contact surface of each group were consistent, the surface of the aluminium sheath was pre‐treated. Each surface was polished with 400Cw‐800Cw‐1500Cw waterproof abrasive papers in turn to remove the passive film and defects on the surface of the sheath. Clean the debris with deionized water, rinse the surface with alcohol and rinse it again with deionized water, and bake the pretreated samples to remove moisture. Keep ambient temperature at 19°C. The corrosion morphologies of the contact surface in different chlorides content groups were recorded. The development process of pits on the surface of corrugated aluminium sheath sample was observed. The corrosion rate of aluminium sheaths in different chlorides content environments was compared and the development of aluminium sheath corrosion was summarized.The development of corroded surface is shown in Figure 11. After 1.5 h experiment, pits are observed on the contact surface of the 1.8 wt% group (as shown in Figure 11a). In the 1.2 wt% group, pits are observed at the edge of contact surface and in the form of stripes (as shown in Figure 11b). There is no obvious corrosion trace in the centre of the contact surface. No corrosion trace is observed on the surface of 0 wt% group (as shown in Figure 11c).11FIGUREThe development of corroded surface in different chlorides content environments. (a–c) Corrosion images of aluminium in environment with 1.8, 1.2 and 0 wt% chlorides respectively after 1.5 h experiment. (d–f) Corrosion images after 4 h experiment. (g–i) Corrosion images after 52.5 h experimentAfter 4 h experiment, the surface of the sheath in 1.8% wt group is covered with pits. The corroded surface lost its metallic lustre (as shown in Figure 11d) In 1.2 wt% group, corrosion mainly occurs at the edge of the contact surface, and pits further develop along the edge, which is shown in Figure 11e. In 0 wt% group, pits are sparsely distributed on the contact surface (as shown in Figure 11f).After 52.5 h experiment, cracks are observed on the contact surface of the 1.8% wt group (as shown in Figure 10g). In 1.2% group, corrosion pits can be observed in the centre of contact surface; At the edge of contact surface, the strip formed by pits expands (as shown in Figure 11h). In 0 wt% group, aluminium sheath sample maintains metallic lustre. Compared with the corroded surface after 4 h experiment, the number of pits on the contact surface increases significantly; pits are randomly distributed over the entire contact surface and grows larger and deeper (as shown in Figure 11i).This result indicates that aluminium sheath is vulnerable in chloride content environment [32–34]. Chloride ion accumulates in the active area, invades the defects in the lattice of passive film. This process destroys the passive film and exposes aluminium matrix. Thus, the area of passive film is reduced and the area where corrosion reaction can occur is increased. This results in higher corrosion rate of aluminium sheath. Finally, larger corroded area with obvious pits and crack can be observed in the sample with higher chloride content. The corrosion of aluminium sheath initially appears in the form of local pitting. With the development of corrosion, the number of pits continue to increase. As pitting process continues to expand and deepen, the corrosion area covers most part of the contact surface. Cracks can appear in the area severely corroded. Finally, when moisture is removed from the surface, corrosion product aluminium hydroxide is precipitated and adheres to the contact surface to form white spots, which is observed in previous studies [6, 14].The effect of chlorides on the polarization curve of aluminiumThe experiment shown in Figure 12 was conducted to further investigate the corrosion rate of aluminium in different chlorides content environments. Polarization curve of aluminium in different chloride content were measured and the characteristic parameters were used to quantitatively describe the corrosion rate of aluminium [26, 27]. The weight percentages of chloride ion in electrolytes were 0.6, 1.2 and 1.8 wt%, respectively and the solutions were neutral. An aluminium plate with a diameter of 6 mm was used as working electrode and the exposed area of the working electrode was 0.283 cm2; the rest of the electrode was covered with insulator. A calomel electrode was used as the reference electrode to reduce the interference of the ohmic resistance of solution [28]. A platinum plate was used as counter electrode. Thus, the polarization current flows between working electrode and counter electrode. The exposed aluminium surface was successively polished with waterproof abrasive papers of 400#, 800# and 1500#. Deionized water and alcohol were used to resin the polished surface. The ambient temperature was 20°C. Anodic polarization curves of working electrode in electrolytes with different chloride ion weight percentage were recorded by an AUTOLAB‐PGSTAT302 electrochemical workstation and the scanning speed was 5 mV/s.12FIGUREThe polarization curve of aluminium in different chloride ion content electrolytes were measured using a conventional tri‐electrode electrochemical cellAn anodic polarization curve includes four regions: the activation region, activation‐passivation region, passivation region and transpassivation region [29, 20]. In the activation region, polarization current increases with the increase of electrode potential. At this stage, Tafel equation is used to describes the relationship between the applied potential and polarization current. The corrosion current density obtained from the polarization curve can directly reflect the corrosion rate of the working electrode. The higher the corrosion current density Jcorr, the faster the corrosion of the surface of working electrode in the electrolytic cell environment. The polarization corrosion current Icorr and the corresponding corrosion current density Jcorr can be calculated according to the Stern‐Geary formula:10Icorr=1Rp×βa×βc2.3(βa+βc)$${I}_{corr} = \frac{1}{{{R}_p}} \times \frac{{{\beta }_a \times {\beta }_c}}{{2.3({\beta }_a + {\beta }_c)}}$$11Jcorr=Icorr/S$${J}_{corr} = {I}_{corr}/S$$where Rp is the linear polarization resistance, βa and βc are Tafel constants. The linear polarization resistance and Tafel constant can be obtained by linear fitting the activation region on the polarization curve. When electrode potential increases and reaches the passivation potential, polarization curve enters the passivation region. At this stage, the passive film formed on the surface of aluminium inhibits electrons from passing the surface of aluminium, thus prevents the corrosion of aluminium. As a result, in passivation region, the increase of electrode potential has little effect on the polarization current. As electrode potential further increases and exceeds the pitting potential Epit, the passive film on the metal surface breaks, and polarization current increases significantly with the increase of electrode potential. Parameters such as corrosion current density, pitting potential, and the length of passivation region which are related to aluminium corrosion resistance in the specific environment can be obtained from the polarization curve.The effect of chlorides content on polarization curve of aluminium is shown in Figure 13. The Tafel constant, linear polarization resistance Rp, corrosion current density Jcorr and pitting potential Epit of aluminium in different chlorides content environment are shown in Table 3. As shown in Figure 13, the increase of the weight percentage of chloride ion causes a left shift of Epit. This result indicates that aluminium can be easily corroded in environment with high chlorides content. A passivation region ranging from −0.9 to ‐0.7 V is observed in the group with 0.6 wt%. At this stage, polarization current changes slowly with the increase of electrode potential. However, when chlorides content increases to 1.2 wt%, the length of the passivation region is significantly shortened.13FIGUREThe polarization curves of aluminium in different chlorides content electrolytes3TABLECorrosion parameters obtained from the polarization curves of aluminium in different chlorides content electrolyteswt%βaβcRp (MΩ)Jcorr (nA/mm2)Epit (V/SCE)0.60.07000.05680.22648212.4−0.6891.20.25160.28330.133561531.3−0.7241.80.44570.93420.089985145.9−0.742As the weight percentage of chloride ion increases, Jcorr increases significantly. When 0.6 wt% chloride ion is applied, the Jcorr and Epit are 212.4 nA/mm and −0.657 V, respectively. When chlorides content increases to 1.2 wt%, the Jcorr of aluminium increases to 1531.3 nA/mm and the pitting potential is −0.698 V. When chlorides content increases to 1.8 wt%, Jcorr further increases to 5145.9 nA/mm and the pitting potential is −0.719 V. With the increase of chlorides content, the pitting potential of aluminium shifts left, the anode potential required to enter transpassivation region decreases which indicates that the corrosion of aluminium is easier. Therefore, the increase of chlorides content affects the corrosion resistance of aluminium.The pitting potential and the corrosion current density are highly related to the chloride concentration of the solution. A high chloride concentration leads to a high corrosion current density. The passive region on the curve disappears when the chloride concentration continues to increase. This result is attributed to the degradation of the passive films in high chloride concentration solution. The passive films are limited in the present of chloride ion, which results in deterioration of corrosion properties.CONCLUSIONCorrosion of corrugated aluminium sheath seriously affects the lifespan of high‐voltage power cable. An investigation of the influence factors affecting corrosion rate of aluminium sheath provides a good reference for the design of high‐quality cable. Here, the aluminium sheath corrosion mechanism, the corrosive effect of chlorines content and the source of chlorides content are studied. The effect of chloride ion on accelerating sheath corrosion is validated and evaluated by experiments. The result can be concluded as follows:(1) Electrochemical corrosion occurs at the interface between corrugated aluminium sheath and semiconducting buffer layer. Chloride ion can accelerate the corrosion of aluminium sheath. The EDS images of the experiment sample shows that buffer layer is the source of chlorine. The chlorides content is concentrated at corroded surface in corrosion process. The concentration of chlorine implies that electrochemical corrosion occurs at the interface between aluminium sheath and buffer layer.(2) After being immersed in deionized water for 15 min, the content of ionizable chlorine in buffer layers varies. The titration result shows that the weight percentage of chloride ion in buffer tape varies from 4.2% to 0.7%. This result indicates that the buffer layers in high voltage power cable is one of the sources of chlorides content. In damp condition, the weight percentage of ionizable chlorine in buffer tape will directly influence the concentration of chloride ion at the interface and thus, affects the corrosion rate of aluminium sheath.(3) Corrosion experiment and polarization curves in different chlorides content electrolyte show that when the concentration of chlorides increases, the corrosion rate of aluminium sheath increases. When the chlorides content increases from 0.6 to 1.8 wt%, pitting potential Epit changes from −0.689 to −0.742 V. With the increase of chloride concentration, he corrosion current density Jcorr increases significantly, and the length of passive region decreases. Aluminium sheath is susceptible to local pitting in chloride‐rich solution.Corrosion of corrugated aluminium sheath severely degrades the electric contact between aluminium sheath and buffer layer. In chloride‐rich environment, aluminium sheath is susceptible to corrosion. Buffer layer has been proved to be one of the sources of chlorides content. Nevertheless, the source of chlorine has not been fully investigated, future research investigating the source of chloride ion can contribute to the reduce of chlorides content in high‐voltage cable and therefore, reduce the corrosion rate of aluminium sheath. Methods have been proposed to improve the corrosion resistance of metal. However, the application of these methods in high‐voltage cable metallic sheath design has not been fully discussed. The relationship between corrosion of aluminium sheath and lifetime of cable is uncertain. The further study considering life assessment of aluminium sheath will help ensure the maintenance and availability of cable in service.CONFLICT OF INTERESTThe authors declare that there are no conflicts of interest regarding the publication of this paper.FUNDING INFORMATIONThere is no funding to support for this submission.DATA AVAILABILITY STATEMENTThe data that support the findings of this study are available from the corresponding author upon reasonable request.AUTHOR CONTRIBUTIONSL.L.: Software, data curation, formal analysis, validation, writing—Original draft.Z.W.: Software, formal analysis, visualization.X.L.: Methodology, resources, supervision.Y.H.: Methodology, resources, writing—Review and editing.G.L.: Conceptualization, methodology, resources, supervision, writing—Review and editing.REFERENCESStandard for extruded insulation power cables rated above 46 through 500 kV, ICEA S‐108‐720, 2018Specification for extruded insulation power cables and their accessories rated above 46kV through 345kV, AEIC Standard CS9‐15, (2015)Nelson, R.A., Daly, J.M.: Corrugated metallic sheathed cable–Design and applications. IEEE Trans. Ind. Appl. 11(2):196–203 (1975)Liu, Y., Chen, J.: Design of smooth aluminum bonded sheaths in HV XLPE cables based on the bending performance research. IEEE Access 8:102493–102501 (2020)Song, P., et al.: A case study on ablation breakdown of high voltage cable buffer layer. In: 2020 IEEE International Conference on High Voltage Engineering and Application (ICHVE), Beijing, China, pp. 1–4 (2020)Chen, Y., et al.: Failure investigation of buffer layers in high‐voltage XLPE cables. Eng. Failure Anal. 113(10), 104546 (2020)Jiang, L., et al.: Study on ablation between metal sheath and buffer layer of high voltage XLPE insulated power cable. In: 2019 2nd International Conference on Electrical Materials and Power Equipment (ICEMPE), Guangzhou, Guangdong, China, pp. 372–375 (2019)Du, B., et al.: An investigation on discharge fault in buffer layer of 220 kV XLPE AC cable. IET Sci. Meas. Technol. 15:508–516 (2021)Li, Z.J., et al.: Effects of the environmental factors on electrical properties of high voltage cable buffer layer. In: 2020 IEEE 3rd International Conference on Dielectrics (ICD), Valencia, Spain, pp. 546–549 (2020)Cao, J., et al.: Analysis on buffer layer discharges below the corrugated aluminum sheath of XLPE cables and comparison with other metal sheath structures. In: 2019 IEEE 3rd International Conference on Circuits, Systems and Devices (ICCSD), Chengdu, China, pp. 21–26 (2019)Xin, Y., et al.: Cause analysis of aging ablation on sheath of 110 kV single core high voltage cable. In: 2019 IEEE Conference on Electrical Insulation and Dielectric Phenomena (CEIDP), Richland, WA, USA, pp. 162–165 (2019)Liu, Y., et al.: The formation mechanism of white substances on water‐blocking buffer layer of HV XLPE cables. In: 2020 IEEE International Conference on High Voltage Engineering and Application (ICHVE), Beijing, China, pp. 1–4 (2020)Nordas, S., Helleso, S.M., Hvidsten, S.: Measurements and modeling of water diffusion in water blocking tapes for high voltage extruded cables. In: 2008 International Symposium on Electrical Insulating Materials (ISEIM 2008), Yokkaichi, Japan, pp. 263–266 (2008)Chen, Y., et al.: Hydrogen evolution and electromigration in the corrosion of aluminium metal sheath inside high‐voltage cables. High Volt. 7:260–268 (2022)Abdo, H.S., et al.: Electrochemical behavior of inductively sintered Al/TiO2 nanocomposites reinforced by electrospun ceramic nanofibers. Polymers 13(24):4319 (2021)Abdo, H.S., et al.: Microstructural characterization and corrosion‐resistance behavior of dual‐phase steels compared to conventional rebar. Crystals 10, 1068 (2020)Chen, Y., et al.: Effects of connection conditions between insulation screen and Al sheath on the buffer layer failures of high‐voltage XLPE cables. Eng. Failure Anal. 122, 105263 (2021)Su, C.Q.: Failure analysis of three 230kV XLPE cables. In: 2010 IEEE/PES Transmission and Distribution Conference and Exposition: Latin America (T&D‐LA), Sao Paulo, Brazil, pp. 22–25 (2010)Ibrahim, E.S: Corrosion control in electric power systems. Electr. Power Syst. Res. 52, 9–17 (1999)Zhang, B., et al.: Numerical approach of impressed potential on buried pipelines near high‐voltage DC grounding electrodes. IET Gener. Transm. Distrib. 12(5):1177–1182 (2018)Bhandari, J., et al.: Modelling of pitting corrosion in marine and offshore steel structures—A technical review. J. Loss Prev. Process Ind. 37:39–62 (2015)Korzhov, A.V.: Effects of magnetic field on cable sheath corrosion. Eng. Failure Anal. 116, 104749 (2020)Tian, H., Liu, P., Zhou, S., Wang, Q., Wu, Z., Zhang, J., Peng, Z.: Research on the deterioration process of electrical contact structure inside the ±500 kV converter transformer RIP bushings and its prediction strategy. IET Gener. Transm. Distrib. 13(12):2391–2400 (2019)Standard Test Methods for Chloride Ion in Water, ASTM D 512‐12, 2012Trejo, D., Ahmed, A.A.: Adopting auto‐titration to assess chlorides in concrete. ACI Mater. J. 116(3), 43‐52 (2019)Cui, L., et al.: Experimental study on multi‐component corrosion inhibitor for steel bar in chloride environment. Constr. Build. Mater. 313, 125533 (2021)Zhang, S., et al.: Experimental data treatment of the pipeline steel polarization curve under AC interference. Int. J. Electrochem. Sci. 14(12):10888–10906 (2020)Lee, S.E., Tang, MH. ‘Reliable reference electrodes for nonaqueous sodium‐ion batteries. Int. J. Electrochem. Sci. 166(14):3260–3264 (2019)Zhou, Y., et al.: The relationship between activation‐passivation transition and grain boundary dissolution on four steel samples in acidic solutions containing NO2−. Faraday Discuss. R. Soc. Chem. 9(41), 23589–23597 (2019)Hung, T.: On the mechanism of pitting of aluminium. J. Electrochem. Soc. 126:1855–1860 (1979)Calderón, J.P., et al.: Corrosion behavior of Al modified with Zn in chloride solution. Materials 15(12):4229 (2022)Senderowski, C., et al.: Effect of low chloride and sulfate concentrations on corrosion behavior of aluminum and zinc arc thermal sprayed coatings. Coatings 12(5):653 (2022)Calderón, J.P. et al. ‘Corrosion behavior of Al modified with Zn in chloride solution’. Materials 15(12), 4229 (2022)Senderowski, C. et al. ‘Effect of low chloride and sulfate concentrations on corrosion behavior of aluminum and zinc arc thermal sprayed coatings’. Coatings 12(5), 653 (2022)AAPPENDIXFLOW CHART FOR THE RESEARCH METHODOLOGY

Journal

"IET Generation, Transmission & Distribution"Wiley

Published: Nov 1, 2022

References

Access the full text.