Acta Metallurgica Sinica(English Letters), 2017, 30(4): 342-351
doi: 10.1007/s40195-017-0539-x
Galvanic Effect Between Galvanized Steel and Carbon Fiber Reinforced Polymers
Chi Zhang1, Da-Jiang Zheng1, Guang-Ling Song1,2
1.Center for Marine Materials Corrosion and Protection, State Key Laboratory of Physical Chemistry of Solid Surface, College of MaterialsXiamen UniversityXiamenChina
2.School of Mechanical and Mining EngineeringThe University of QueenslandBrisbaneAustralia
 Cite this article:
Chi Zhang, Da-Jiang Zheng, Guang-Ling Song. Galvanic Effect Between Galvanized Steel and Carbon Fiber Reinforced Polymers. Acta Metallurgica Sinica(English Letters)[J], 2017, 30(4): 342-351 doi:10.1007/s40195-017-0539-x

Abstract:

The galvanic corrosion behavior of carbon fiber reinforced polymers (CFRPs) GM-CFRP and Tepex-CFRP in contact with a Zn-coated DP590 steel in solution containing 0.9 wt% NaCl + 0.1 wt% CaCl2 + 0.075 wt% NaHCO3 was investigated. The results showed that the GM-CFRP/steel couple was initially more resistant to galvanic corrosion, but its galvanic corrosion activity gradually became higher than the Tepex-CFRP/steel couple. The different galvanic behaviors of these two couples were discussed based on the electrochemical performance of GM-CFRP, Tepex-CFRP and DP590 coupons in the testing solution.

Key words: Carbon fiber reinforced polymer (CFRP) ; Composite ; Carbon fiber ; Steel ; Galvanic effect
1 Introduction

Carbon fiber reinforced polymer (CFRP) is one of the most important composite materials, which has many advantages, such as lightweight, high strength and high modulus, and has various applications in industry [1, 2, 3, 4, 5, 6, 7, 8]. In practice, CFRP is often used together with engineering metals, such as steels, aluminum alloys and magnesium alloys. The galvanic effect between CFRP and metals has been a big concern for engineers, as these two types of materials have different electrochemical properties. Many investigations on galvanic corrosion damage have been reported [9, 10, 11, 12, 13, 14, 15, 16]. According to the well-known galvanic series in seawater [17, 18, 19], graphite has the noblest open-circuit potential (OCP). Thus, it is always a cathode if in contact with an engineering metal. Previous studies on galvanic corrosion damage of various CFRP/metal couples have enabled engineers to design better CFRP/metal joints, which has to some extent reduced the galvanic corrosion risk, and facilitated the application of CFRP in practice.

Although the measurement and evaluation of galvanic effect between metals and different types of CFRPs [10, 11, 13, 20, 21, 22, 23, 24] have clearly shown that engineering metals, including steel, aluminum, magnesium and their alloys, when coupled with carbon fiber or CFRP, can be preferentially corroded in aggressive environments due to the well-known galvanic effect, some fundamental issues remain unaddressed, such as the influence of the CFRP microstructure and environment factors on galvanic corrosion of metals in contact. It is unclear whether the traditional galvanic theory can be directly applied when CFRP is in contact with an engineering metal, as CFRP may not be a good electronically conducting material like metal.

In this study, the conductivity of two types of CFRPs was measured to examine the applicability of the traditional galvanic principle. After that, the galvanic currents from a Zn-coated DP590 steel caused by the two different types of CFRP were recorded in 0.9 wt% NaCl + 0.1 wt% CaCl2 + 0.075 wt% NaHCO3 solution to understand the influence of CFRP microstructure on galvanic corrosion.

2 Experimental
2.1 Materials and Solution

The CFRP composite materials in this study were obtained from two different sources. The samples from GM R&D, denoted as GM-CFRP, were composed of carbon fiber and epoxy resin, while those named as Tepex-CFRP obtained from Tepex Company were made of carbon fiber and nylon. Steel DP590 with a Zn coating was also provided by GM R&D. Such a galvanized steel is widely used in the auto industry. All the samples had dimensions of 100 mm × 40 mm × 2 mm. Their compositions are shown in Table 1.

Table 1

Compositions of test materials

The CFRP samples after cleaning with ethanol and distilled water were used directly. Tests were carried out on their original surfaces. For comparison, the surfaces of some CFRP samples were sanded/ground with 80# sandpaper to remove their surface polymer layers to expose the inside carbon fiber. The ground samples were also cleaned with ethanol and distilled water before test.

A GM standard test solution 0.9 wt% NaCl + 0.1 wt% CaCl2 + 0.075 wt% NaHCO3 (GM WORLDWIDE ENGINEERING STANDARD) was used in this study.

2.2 Bulk Resistance Measurement

To measure the bulk resistance of a CFRP sample, four holes with diameters of 2 mm were drilled in the sample. Bolts were screwed in the holes tightly. A constant current (0.2 A) was applied to the CFRP sample through the bolts in the outer two holes, and the potential drop between the bolts in the two inner holes was measured. Several inner holes with different distances between each other were made. According to the Ohm’s law, the measured resistance value should be linearly dependent on the distance between two inner holes, and thus, the resistivity of the CFRP can be estimated if the cross-sectional area of the sample is known. The setup for the measurement is schematically illustrated in Fig. 1.


Fig. 1

Schematic illustration of bulk resistance measurement with a constant current (0.2 A) applied to the bolts in the outer holes. The distances between the inner holes used for voltage drop measurement are 2, 4 and 6 cm, respectively

2.3 Open-Circuit Potential, EIS and Polarization Curve Measurements

A three-electrode conventional electrolytic cell as shown in Fig. 2 and an electrochemical measurement system AUTOLAB were used to measure the open-circuit potentials (OCP), electrochemical impedance spectra (EIS) and polarization curves of the steel and CFRP samples. In the cell, the counter electrode was a platinum plate and the reference was an Ag/AgCl/saturated KCl electrode. The measured surface area of the CFRP sample was around 0.78 cm2. In polarization curve measurement, the potentiodynamic scanning rate was 0.001 V/s. EIS measurement was carried out at the OCP. The AC potential amplitude was 10 mV, and the frequency range was from 10 mHz to 100 kHz with 10 frequency points per logarithmic decade.


Fig. 2

Schematic illustration of the electrolyte cell setup for open-circuit potential, EIS and polarization curve measurements. The electrolyte volume in the cell is 100 ml

2.4 Galvanic Current Measurement

The galvanic currents between CFRP samples and steel were measured in a different electrolytic cells as shown in Fig. 3, in which W1 represents the CFRP sample (a GM or Tepex coupon), W2 is the Zn-coated DP590 steel and RE is an Ag/AgCl reference electrode. The distance between W1 and W2 is 6 cm, and the exposed areas of the CFRP and steel samples are 0.20 cm2.


Fig. 3

Schematic illustration of the setup for galvanic current measurement. The electrolyte volume in cell is 100 ml

2.5 Microstructure Characterization

The surface morphology and cross-sectional microstructure of CFRP were observed by scanning electron microscopy (SEM, SU70 Hitachi and TM3000).

3 Results
3.1 CFRP Structures

Figure 4 shows the macroscopic morphologies of the two types of CFRP samples, in which the distribution of fiber bundles can be clearly seen. In the GM-CFRP sample, carbon fiber bundles are perpendicularly crossing each other on one side, while on the other side some carbon fiber bundles have a zip-zap distribution between two longitudinally parallel strands. The fiber bundles in the Tepex-CFRP are vertically knitted on both sides, just like the texture of a “woven bag.”


Fig. 4

Macroscopic morphologies of CFRP samples

The surface micromorphologies are shown in Figs. 5 and 6. The surface of GM-CFRP is smooth, covered by epoxy. The carbon fiber in the sample can be better revealed after surface grinding (Fig. 5). The diameter of single carbon fiber is about 5 μm. In the Tepex-CFRP sample, the carbon fiber texture can be clearly seen on the original surface (Fig. 6). The nylon layer covering over the fiber on the sample surface appears to be too thin to conceal the fiber texture. Surface grinding led to carbon fiber exposed completely.


Fig. 5

Surfaces of GM-CFRP sample before a, after b surface grinding


Fig. 6

Surfaces of Tepex-CFRP sample before a, after b surface grinding

The SEM cross-sectional images clearly show the microstructures of these two types of CFRP samples (Figs. 7, 8). The GM sample is like a “sandwich” with different directions of carbon fiber bundles in different layers. The thickness of these layers can be roughly measured in Fig. 7. The fiber bundles in Tepex sample displayed in Fig. 8 appear to be “knitted” (vertically overlapped) together in nylon matrix.


Fig. 7

Cross-sectional images of a GM-CFRP sample


Fig. 8

Cross-sectional images of a Tepex-CFRP sample

3.2 Bulk Resistance of CFRPs

Figure 9 shows a linear dependence of the bulk resistance on measuring distance for GM-CFRP and Tepex-CFRP samples. All the repeatedly measured resistance values are completely overlapped in the figure, indicating high reproducibility of the measurement. The bulk resistivity of GM-CFRP and Tepex-CFRP samples can be estimated to be (0.93-3.12) × 10-3 and (1.20-1.94) × 10-3 Ω/cm-1, respectively, suggesting that the CFRPs are as conductive as metals [25]. There is no difference in bulk resistance before and after surface grinding.


Fig. 9

Bulk resistances for the CFRP samples before and after surface grinding

3.3 AC Impedance of CFRPs

The differences in EIS between GM and Tepex samples measured at their OCPs are shown in Fig. 10. These results represent the impedance behavior of the surface layers of the GM and Tepex samples. They can be summarized as follows. First, the reproducibility of the EIS measurements on GM-CFRP samples was much worse than that on Tepex samples. Second, the EIS of the Tepex-CFRP sample has a Warburg diffusion characteristic, while that of the GM sample does not. Third, the resistance of GM-CFRP estimated from EIS measurement is much higher than that of Tepex-CFRP, indicating that the coverage and thickness of the surface polymer layer could be larger on the GM sample than on the Tepex.


Fig. 10

Repeatedly measured EISs of CFRP samples: a GM-CFRP, b Tepex-CFRP

An equivalent circuit as shown in Fig. 11 can be employed to estimate the resistance (R) of the surface polymer layer. In fact, the R value should also include the electrochemical reaction resistance on the carbon fiber surface in contact with solution, although it is negligible compared with that of the surface polymer layer. It should be noted that the Warburg impedance Zw was set at zero when the circuit was used to simulate the EIS of GM-CFRP. The estimated results using the equivalent circuit are listed in Table 2.


Fig. 11

Equivalent circuit for simulating the measured EISs shown in Fig. 10

Table 2

Surface resistance (R) of CFRPs

In Table 2 both GM-CFRP and Tepex-CFRP have a high surface resistance, but the former (2.06 × 107 Ω/cm-2) is nearly 3 orders of magnitude higher than the latter (4.15 × 104 Ω/cm-2). The large resistance difference suggests that the GM and Tepex samples should have different surface layers. The former surface could be fully covered by an electrically insulating layer, whereas the surface layer on the latter could be relatively thin and carbon fiber may be exposed in some areas on its surface.

Figure 12 shows the variation of EIS with time for CFRPs under -1 V/Ag/AgCl polarization. The impedance of both CFRPs in low frequencies decreased with immersion time, but their decreasing trends were slightly different; the low-frequency impedance of the GM sample gradually decreased with time, whereas that of the Tepex at the beginning decreased rapidly and then became relatively steady. Finally the resistance of GM-CFRP was even lower than that of Tepex-CFRP.


Fig. 12

Variation of EIS with immersion time for CFRPs under a polarizing voltage of -1 V/Ag/AgCl: a GM-CFRP sample, b Tepex-CFRP sample

3.4 OCPs of Different Samples

Figure 13 shows the variation of open-circuit potentials (OCPs) of galvanized DP590 steel and the two CFRPs with immersion time. Except the initial few minutes of negatively shifting, the OCP of DP590 steel was relatively steady around -1.03 VAg/AgCl/Cl. The OCPs of GM-CFRP and Tepex-CFRP were around -0.11 VAg/AgCl/Cl- and +0.02 VAg/AgCl/Cl-, respectively, more positive than that of DP590 steel. This is because the electrochemical behavior of carbon fiber in CFRP is similar to a noble metal such as gold or platinum [26]. As reported by Schneider et al. [27], if the potential difference between two materials is <0.05 V, the galvanic corrosion can be ignored. In this study, the potential difference between the CFRPs and steel is about 1 V. It means that the galvanic corrosion will be severe if the CFRPs are coupled with the steel. The OCP measurements also indicated that the Tepex-CFRP/steel couple had a higher driving force for galvanic corrosion than the GM/steel couple.


Fig. 13

OCPs of DP590 steel, GM-CFRP and Tepex-CFRP samples versus immersion time

3.5 Polarization Curves of CFRP and Galvanized DP590 steel

Figure 14 shows the polarization curves of galvanized DP590 steel, GM-CFRP and Tepex-CFRP. The Tepex has the highest OCP while the DP590’s OCP is the most negative. The intersections of their polarization curves are around -1 V Ag/AgCl/Cl-.


Fig. 14

Polarization curves of different samples

An important finding from this figure is that the cathodic current densities of the Tepex sample are higher than those of the GM sample, and the cathodic process on the Tepex sample is completely controlled by diffusion, while the diffusion characteristic on the curve of GM sample appears to be relatively less evident. This explains why the Warburg impedance appeared during the EIS measurement of the Tepex sample, whereas the EIS of the GM sample did not exhibit a diffusion feature (see Fig. 12). The diffusion characteristic appears on Tepex-CFRP sample under -1 V/Ag/AgCl polarization in the initial stage of immersion. The polarization curve predicts that at the beginning the galvanic corrosion of the steel/GM-CFRP couple is less significant than that of the steel/Tepex-CFRP couple, as the anodic polarization curve of the steel intersects at a low current density with the cathodic polarization curve of the GM sample than with Tepex. As the polarization curves of these samples can change with immersion time, the galvanic corrosion of these couples cannot be accurately predicted simply from these polarization curves. The monitoring of real galvanic current densities of couples is thus important and essential.

3.6 Galvanic Current Density Measurement

Figure 15 shows the variation of the galvanic corrosion current densities of GM-CFRP and Tepex-CFRP in electric connection with galvanized DP590 steel. In Fig. 15a, the GM-CFRP/steel couple has relatively low initial galvanic current densities (ca. 1-5 μA/cm2), which continuously increases with time in the first 6 h. After that the current densities become relatively stable in the range of 5-9 μA/cm2. In Fig. 15b, the initial galvanic current densities of the Tepex-CFRP/steel couple start at relatively high values (4-7.5 μA/cm2), but they rapidly decrease. The galvanic current densities quickly stabilize around 4 μA/cm2. The time required for the Tepex-CFRP/steel couple to reach the steady current density is considerably shorter than that for the GM-CFRP/steel. In addition, the reproducibility of the galvanic current measurement of the Tepex-CFRP/steel couple was much better than that of the GM-CFRP/steel couple.


Fig. 15

Galvanic current densities of CFRP/steel couples: a GM-CFRP/steel, b Tepex-CFRP/steel

The galvanic current results (Fig. 15) verify the polarization curve prediction that at the beginning the galvanic corrosion of the Tepex-CFRP/steel is more severe than that of the GM-CFRP/steel, and the galvanic corrosion damage of the latter may eventually become more significant than that of the former. This was confirmed by observation of the corrosion damage on the steel surfaces in the steel/GM-CFRP and steel/Tepex-CFRP couples after galvanic current measurements (see Fig. 16). The galvanic corrosion of the steel coupled with the GM-CFRP sample was slightly more localized and with slightly thicker corrosion products on the corroded areas than that with the Tepex-CFRP.


Fig. 16

Galvanic corrosion damage of DP590 steel coupled with a Tepex-CFRP and b GM-CFRP after 8 h of galvanic current measurements

4 Discussion

The bulk resistivity of GM-CFRP and Tepex-CFRP is very low (see Fig. 9), which means that they are very electrically conductive. Therefore, the traditional galvanic theory, in which the resistance of the anode and cathode is normally ignored [28, 29, 30], can be directly applied to the metal/CFRP couples. The difference in bulk resistivity between the two CFRPs is relatively small, because both the CFRPs have a similar content of carbon fiber (Table 1). The small resistivity difference (<3.6 × 10-4 Ω/cm-1) could be ascribed to the different arrangements of the carbon fiber bundles in GM-CFRP and Tepex-CFRP; the fiber bundles in the latter are more tightly knitted than those in the former, which can be seen from their cross sections in Figs. 7 and 8. Surface grinding did not affect the bulk resistance of the CFRPs (see Fig. 9). This is apparently due to the fact that the carbon fiber is the only component conducting electrons in the CFRPs. It also means that the carbon fiber strands distributed in the CFRP samples are very well electrically connected. Since the surface polymer layers are insulating, they cannot have a contribution to the conductivity of the composites and, thus, grinding did not affect their conductivity.

The obvious difference in surface layer resistance between GM-CFRP and Tepex-CFRP (Table 2) could result from their different microstructures or surface states (see Figs. 5, 6). The epoxy layer over the GM-CFRP surface must be relatively permeable and thick (see Fig. 4), and the thickness and permeability may vary from area to area. The nylon layer on the Tepex-CFRP surface could be relatively thin, and some carbon fiber bundles could be exposed through micropores in the thin layer on the original sample surface. The surface polymer layers on these two CFRPs are schematically presented in Fig. 17.


Fig. 17

Schematic illustration of different surface layers of a GM-CFRP, b Tepex-CFRP

The relatively thick surface epoxy layer on the GM sample can more effectively inhibit the electrochemical reactions on the carbon fiber in the initial stage of immersion, and thus, the surface resistance is very high (see Table 2). In other words, the cathodic reaction over the sample surface is very slow, controlling the whole cathodic process. Therefore, no Warburg impedance can be clearly detected in the EIS (Fig. 11). On the contrary, the electrochemical reactions on the Tepex sample, which has a relatively thin, are much fast, i.e., the measured surface resistance is low (Table 2). In this case, the diffusion of oxygen from the bulk solution to the carbon fiber through the micropores in the surface layer is relatively slow, becoming a reaction determining step limiting the cathodic process. Therefore, the EIS is dominated by Warburg diffusion impedance (Fig. 11).

The difference in galvanic current variation with time between the GM-CFRP/steel and Tepex-CFRP/steel couples (Fig. 15) could be caused by different breakdown behaviors of the surface polymer layers. The epoxy layer on GM-CFRP after being penetrated by corrosive solution could gradually break down, resulting in a larger surface area of carbon fiber exposed. The surface layer breakdown process was gradually accelerated, and surface area of carbon fiber in contact with the solution also gradually increased with immersion time. Thus, the galvanic corrosion became faster and current increased with immersion time (Fig. 15a). In Tepex-CFRP, some carbon fiber had already been exposed on the original surface. Thus, immersion could not further worsen the surface layer’s coverage, resulting in a relatively stable galvanic current with immersion time (Fig. 15b).

The EIS measurements (Fig. 12) under -1 V/Ag/AgCl polarization condition also support the postulation regarding the different deterioration behaviors of the surface layers on GM-CFRP and Tepex-CFRP. The gradually decreasing low-frequency impedance of GM-CFRP was obviously caused by the breakdown of the surface epoxy layer, which resulted in an increased surface area of the carbon fiber in the sample exposed to solution. The quickly dropped low-frequency impedance of Tepex sample could result from rapid establishment of steady electrochemical reactions on the carbon fiber exposed to the solution through the micropores in the surface layer, since the carbon fiber in the sample was in contact with solution quickly after immersion. As the nylon matrix in the Tepex sample could not significantly break down under the immersion condition as that on the GM sample, the low-frequency impedance was stably maintained at a relatively high value. This is different from GM-CFRP. In GM-CFRP, more and more carbon fiber was exposed to the solution after the surface epoxy layer broke down. Hence, the final impedance becomes much lower.

5 Conclusions

1. Both the GM-CFRP and Tepex-CFRP are as conductive as a metal. The traditional galvanic effect theory can be directly applied to a CFRP/metal couple.

2. The open-circuit potentials (OCPs) of GM-CFRP and Tepex-CFRP samples are more positive than that of galvanized DP590 steel. There is a big concern about galvanic corrosion if the CFRPs are used together with the steel.

3. The galvanic corrosion activity can vary with immersion time. The galvanic corrosion of a GM-CFRP/steel couple is initially slower than that of a Tepex-CFRP couple, but the galvanic corrosion of the former can gradually accelerate with immersion time, while the latter process keeps relatively stable, and eventually the Tepex-CFRP/steel becomes slightly more galvanic corrosion resistant than the GM-CFRP/steel.

4. The galvanic corrosion behavior of a CFRP/steel couple is critically dependent on the CFRP’s surface polymer layer. The presence of a resistant surface polymer layer is helpful in inhibiting the galvanic corrosion of a CFRP/steel couple.

Acknowledgements

This work was supported by GM R&D Center.

The authors have declared that no competing interests exist.

参考文献

[1]
G. Ben, N. Sugimoto, Y. Aoki, Adv. Compos. Mater. 19, 363(2010)
DOI:10.1163/092430410X504152 URL
Carbon fiber reinforced plastic (CFRP) laminates are used in various industrial fields because they have excellent properties in their specific strength and specific stiffness. The CFRP has a potential of weight reduction in the automotive structure, which can contribute to the improvement of mileage as well as the reduction of carbon dioxide. On the other hand, the safety issue in case of collision should be also clarified when employing the CFRP as automotive structures. In this paper, hybrid beams which consisted of an Al alloy beam and a CFRP laminate were examined by both experiments and numerical analyses as candidates to replace the conventional steel door guarder beam used inside the automotive door. The experimental relations of impact loading to the displacement for the Al beams with different thicknesses, widths and types of CFRP showed good agreement with those from numerical results. In order to increase the impact energy absorption, other hybrid beams were devised and calculations performed by using the same parameters in the same numerical method and their results also agreed with those of experiments executed after the calculation. These facts show that the numerical method developed here can be generally used for estimating the impact behavior of CFRP/Al hybrid beams. Furthermore, a suggestion for how to install the hybrid beam to automotive doors was proposed by the numerical method. As a result, the effectiveness of our simulation technology is demonstrated here.
[本文引用: 1]
[2]
M. Elchalakani, Thin-Walled Struct. 77, 86(2014)
DOI:10.1016/j.tws.2013.12.002 URL
61CFRP strengthened as-received and rehabilitated corroded box girders were subjected to 3-point bending to determine P–M strength.61Three levels of plate corrosion were examined, i.e., 20%, 40%, and 60%.61The P–M strength of corroded box girder can be significantly increased by bonding CFRP.61Average gain in strength was 65% for the strengthening series and 19.9% for the rehabilitations series.61The increase in strength was affected by section class where the maximum gain was obtained for slender section.
[本文引用: 1]
[3]
I.M. May, M. Roy, C. Lang, R.A. Khan, Proc. Inst. Civ. Eng. Struct. Build. 169, 141(2016)
[本文引用: 1]
[4]
T. Sharples, Aeronaut. J. 84, 177(1980)
[本文引用: 1]
[5]
C. Soutis, Prog. Aeosp. Sci. 41, 143(2005)
DOI:10.1016/j.paerosci.2005.02.004 URL
Fibrous composites have found applications in aircraft from the first flight of the Wright Brothers Flyer 1, in North Carolina on December 17, 1903, to the plethora of uses now enjoyed by them on both military and civil aircrafts, in addition to more exotic applications on unmanned aerial vehicles (UAVs), space launchers and satellites. Their growing use has risen from their high specific strength and stiffness, when compared to the more conventional materials, and the ability to shape and tailor their structure to produce more aerodynamically efficient structural configurations. In this paper, a review of recent advances using composites in modern aircraft construction is presented and it is argued that fibre reinforced polymers, especially carbon fibre reinforced plastics (CFRP) can and will in the future contribute more than 50% of the structural mass of an aircraft. However, affordability is the key to survival in aerospace manufacturing, whether civil or military, and therefore effort should be devoted to analysis and computational simulation of the manufacturing and assembly process as well as the simulation of the performance of the structure, since they are intimately connected.
[本文引用: 1]
[6]
J. Takahashi, K. Uzawa, I. Ohsawa, N. Matsutsuka, A. Kitano, K. Nagata, Applicability of recycled CFRP to secondary parts of automobile, in Proceedings of the Twelfth U.S.-Japan Conference on Composite Materials (Destech Publications, Inc, Lancaster, 2006)
[本文引用: 1]
[7]
I. Taketa, K. Yamaguchi, E. Wadahara, M. Yamasaki, T. Sekido, A. Kitano, The CFRP automobile body project in Japan, in Proceedings of the Twelfth U.S.-Japan Conference on Composite Materials (Destech Publications, Inc, Lancaster, 2006)
[本文引用: 1]
[8]
G. Williams, R. Trask, I. Bond, Compos. Pt. A Appl. Sci. Manuf. 38, 1525(2007)
DOI:10.1016/j.compositesa.2007.01.013 URL
Self-healing is receiving increasing interest worldwide as a technology to autonomously address the effects of damage in composite materials. This paper describes the results of four point bend flexural testing (ASTM-D6272-02) of T300/914 carbon fibre reinforced epoxy with resin filled embedded hollow glass fibres (HGF) which provided a self-healing functionality. The study investigated the effect of the embedded HGF on the host CFRP mechanical properties and also the healing efficiency of the laminates after they were subjected to quasi-static impact. Specimens were tested in the undamaged, damaged and healed conditions using a commercial two-part epoxy healing agent (Cytec Cycom 823). Microscopic characterisation of the embedded HGF was also undertaken to characterise the effect on the host laminate fibre architecture.
[本文引用: 1]
[9]
V.A. Katkar, G. Gunasekaran, Corrosion 72, 400 (2016)
[本文引用: 1]
[10]
M. Mandel, L. Kruger, Materialwiss. Werkstofftech. 45, 1123(2014)
DOI:10.1002/mawe.201400352 URL
In this study, the corrosion behaviour of a self-piercing rivet joint which consists of a sheet of the aluminium alloy EN AW-6060-T6 and a carbon-fibre reinforced polymer laminate was investigated by multiple corrosion cycles, according to the standard VDA 621-415. Due to the electrochemically quite different corrosion behaviour of the used materials the aluminium alloy is attacked by pitting corrosion close to and below the overlap of the joint. To investigate the amount of the corrosion attack the aluminium alloy was analysed by white-light interferometry at the most attacked region in the joint after each corrosion cycle. Based on the results of the topographical scans the pit volumes were calculated in an area of 2 × 5 mm2. The plot of the pit volume versus the corrosion cycle number revealed a sigmoidal correlation that was adequately described by the Boltzmann function. Furthermore, due to the s-shaped pit volume – cycle number correlation three stages of pitting corrosion were identified. In dieser Arbeit wurde das Korrosionsverhalten einer Halbhohlstanznietverbindung bestehend aus einem Blech der Aluminiumlegierung EN AW-6060-T6 und einem kohlenstofffaserverst01rkten Polymer-Laminat für mehrere Korrosionszyklen nach dem VDA 621-415 Standard Korrosionstest untersucht. Aufgrund des unterschiedlichen elektrochemischen Verhaltens der verwendeten Werkstoffe wird die Aluminiumlegierung nahe des 05berlappbereiches der Fügeverbindung durch Lochkorrosion angegriffen. Zur Bewertung der St01rke des Korrosionsangriffes wurde die Aluminiumlegierung mit einem Wei08lichtinterferometer im Bereich der st01rksten Sch01digung in der Fügeverbindung nach jedem Korrosionszyklus untersucht. Auf Grundlage der Ergebnisse der Oberfl01chenprofile wurde das Lochvolumen in einer Fl01che von 2 × 5 mm2 berechnet. Die Darstellung des Lochvolumens über der Korrosionszyklenzahl ergab einen sigmoidalen Zusammenhang, welcher angemessen mit der Boltzmann-Funktion beschrieben wurde. Au08erdem wurden anhand der s-f02rmigen Korrelation zwischen Lochvolumen und Zyklenzahl drei Stadien für Lochkorrosion ermittelt.
[本文引用: 2]
[11]
Z. Peng, X. Nie, Surf. Coat. Technol. 215, 85(2013)
DOI:10.1016/j.surfcoat.2012.08.098 URL
78 Galvanic corrosion between carbon fiber and steel, Al- and Ti-alloys was studied. 78 Potentiodynamic polarization and zero resistance ammeter (ZRA) tests were used. 78 The steel and Al need a protective coating while Ti exhibits no corrosion problem. 78 Oxide coatings on the Al alloy were deposited to address the galvanic corrosion. 78 Unipolar+bipolar duplex mode offers excellent corrosion resistance to the coatings.
[本文引用: 2]
[12]
B.P. Cai, Y.H. Liu, C.K. Ren, Z.K. Liu, X.J. Tian, A.B.B. Corros. Eng. Sci. Technol. 47, 289(2012)
DOI:10.1179/1743278212Y.0000000010 URL
Abstract The galvanic corrosion behaviours of carbon fibre composite coupled to aluminium are studied when the galvanic couple is just immersed in artificial seawater or connected to a closed electric circuit. The effects of grinding condition, concentration of artificial seawater, applied torque, applied current and experimental time are studied. The roughness average, weight gain of carbon fibre composite and weight loss of aluminium are investigated as a function of the above variables. The results show that the applied current can accelerate the galvanic corrosion greatly. With improving grinding condition and increasing applied torque, the roughness average, weight gain of carbon fibre specimen and weight loss of aluminium specimen increase, reach maximum and then decrease. With increasing concentration of artificial seawater, applied current and experimental time, these measured values increase. Corroded surface morphology is also investigated using scanning electron microscopy.
[本文引用: 1]
[13]
Y. Pan, G. Wu, X. Cheng, Z. Zhang, M. Li, S. Ji, Z. Huang, Corros. Sci. 98, 672(2015)
[本文引用: 2]
[14]
A.A.Torres-Acosta J. Compos. Constr. 6, 112(2002)
DOI:10.1061/(ASCE)1090-0268(2002)6:2(116) URL
Studies were conducted to understand the galvanic interactions between CFRP and steel in chloride-contaminated concrete. CFRP-pultruded rod samples (6 mm in diameter), #3 deformed plain rebar (PR, uncoated), and #4 epoxy-coated reinforcing (ECR) steel bars were tested. After 350 days, potential measurements of CFRP and steel samples in chloride-contaminated concrete were 200 and 600 mV (versus CSE), respectively. These results confirm that chloride contamination in concrete could allow galvanic corrosion between CFRP and steel. The measured galvanic current densities were up to 0.7 and 100 A/m2for the CFRP-PR and CFRP-ECR couples, respectively, raising concerns about the degradation of both CFRP and steel. The results showed that PR steel was unaffected, as the corrosion rates estimated before and after the coupling with CFRP were similar. In contrast, coupling CFRP and ECR steel showed an increase of 10 times of the estimated corrosion rate, suggesting that galvanic interaction might affect the ECR steel.
[本文引用: 1]
[15]
A.A. Torres-Acosta J. Compos. Constr. 6, 116(2002)
DOI:10.1061/(ASCE)1090-0268(2002)6:2(116) URL
Studies were conducted to understand the galvanic interactions between CFRP and steel in chloride-contaminated concrete. CFRP-pultruded rod samples (6 mm in diameter), #3 deformed plain rebar (PR, uncoated), and #4 epoxy-coated reinforcing (ECR) steel bars were tested. After 350 days, potential measurements of CFRP and steel samples in chloride-contaminated concrete were 200 and 600 mV (versus CSE), respectively. These results confirm that chloride contamination in concrete could allow galvanic corrosion between CFRP and steel. The measured galvanic current densities were up to 0.7 and 100 A/m2for the CFRP-PR and CFRP-ECR couples, respectively, raising concerns about the degradation of both CFRP and steel. The results showed that PR steel was unaffected, as the corrosion rates estimated before and after the coupling with CFRP were similar. In contrast, coupling CFRP and ECR steel showed an increase of 10 times of the estimated corrosion rate, suggesting that galvanic interaction might affect the ECR steel.
[本文引用: 1]
[16]
M. Dawood, S. Rizkalla, E. Sumner, J. Compos. Constr. 11, 659(2007)
DOI:10.1061/(ASCE)1090-0268(2007)11:6(659) URL
Due to corrosion and the continuous demand to increase traffic loads, there is a need for an effective system which can be used to repair and/or strengthen steel bridges and structures. This paper describes an experimental program, recently completed, to investigate the fundamental behavior of steel–concrete composite scaled bridge beams strengthened with new high modulus carbon fiber-reinforced polymer (HM CFRP) materials. The behavior of the beams under overloading conditions and fatigue loading conditions was studied as well as the possible presence of shear lag at the interface of the steel surface and the CFRP strengthening material. The test results are compared to an analytical model based on the fundamental principles of equilibrium and compatibility, to predict the behavior of the strengthened steel–concrete composite beams. Based on the findings of this research work, combined with other work in the literature, a design guideline is proposed for the use of HM CFRP for strengthening the steel flexural members typically used for bridges and structures.
[本文引用: 1]
[17]
W.C. Tucker, J. Energy Resour. Technol.(United States) 113, 264(1991)
[本文引用: 1]
[18]
C.H. Hare, J. Prot. Coat. Linings 14, 74 (1997)
[本文引用: 1]
[19]
M.A. Rodríguez, JOM 66, 503 (2014)
[本文引用: 1]
[20]
M. Mandel, L. Krüger, Corros. Sci. 73, 172(2013)
[本文引用: 1]
[21]
L. Arronche, K. Gordon, D. Ryu, V.L. Saponara, L. Cheng, J. Mater. Sci. 48, 1315(2013)
DOI:10.1007/s10853-012-6876-2 URL
In this article, we investigate the effect of modified carbon fiber-reinforced polymer (CFRP) composites when galvanically coupled with AISI 1018 carbon steel. Two different resins were used to manufacture the CFRPs: neat epoxy resin, and epoxy resin modified with multi-walled carbon nanotubes (MWCNTs). The specimens of composite (the cathode of the galvanic cell) and metal (the anode of the galvanic cell) were paired and immersed in electrolyte (NaCl solution, 2 % by weight) at 40 C, to simulate a corrosive environment and accelerate the electrochemical reaction. Results of corrosion rate (CR) and mass loss rate (MR) were obtained, and the electrical resistances of the CFRPs were also measured. This new study shows that the MWCNTs do not have a statistically significant impact on the corrosion and mass loss rate results, and that both types of CFRP composites have statistically the same electrical resistance. Therefore, common methods used in the engineering practice in conventional hybrid CFRP/steel joints and repairs may be sufficient to delay galvanic corrosion, as there is no increased liability.
[本文引用: 1]
[22]
I.A. Wootton, L.K. Spainhour, N. Yazdani, J. Compos. Constr. 7, 339(2003)
DOI:10.1061/(ASCE)1090-0268(2003)7:4(339) URL
The rehabilitation, repair, and strengthening of concrete structures has increased worldwide with a growing number of systems employing externally applied fiber-reinforced polymer (FRP) composites. However, the service life and effectiveness of FRP repair and strengthening techniques when applied to concrete in corrosive marine environments is still not well understood. This paper presents the results of an experimental study on the corrosion performance of embedded steel reinforcement in cylindrical reinforced concrete specimens with 13 different surface treatment options. Samples were subjected to an impressed current and a high salinity solution. Test variables included the type of epoxy, wrap fiber orientation, and the number of wrap layers. Samples were evaluated for corrosion activity by monitoring corrosion potentials and impressed current flow levels, and by examining reinforcement mass loss and concrete chloride content among samples. Test results indicated that FRP wrapped specimens had prolonged test life, decreased reinforcement mass loss, and reduced concrete chloride content. The performance of wrapped specimens was superior to that of either control samples or those coated only with epoxy. Epoxy type had a significant effect on the performance of samples regarding their resistance to corrosion. It was concluded that carbon FRP wraps were able to confine concrete, slowing deterioration from cracking and spalling and inhibiting the passage of salt water.
[本文引用: 1]
[23]
T.E. Maaddawy, K. Soudki, J. Compos. Constr. 9, 187(2005)
DOI:10.1061/(ASCE)1090-0268(2005)9:2(187) URL
This paper presents the results of an experimental study designed to investigate the viability of using externally bonded carbon-fiber-reinforced polymer (CFRP) laminates to extend the service life of corroded reinforced concrete (RC) beams. A total of 14 beams,152 254 3,200mmeach, were tested. Three beams were not corroded; two of them were strengthened by CFRP laminates, while one specimen was kept as a virgin. The remaining 11 beams were subjected to different levels of corrosion damage up to a 31% steel mass loss using an impressed current technique. Six of the corroded beams were repaired with CFRP laminates, whereas the remaining five beams were not repaired. Eventually, all specimens were tested to failure under four-point bending. Corrosion of the steel reinforcement significantly reduced the load-carrying capacity of RC beams. At all levels of corrosion damage, CFRP repair increased the ultimate strengths of the corroded beams to levels higher than the strength of the virgin beam but significantly reduced the deflection capacity.
[本文引用: 1]
[24]
Y. Mueller, R. Tognini, J. Mayer, S. Virtanen, J. Biomed. Mater. Res. Part A 82, 936 (2007)
[本文引用: 1]
[25]
H.C. Hwang, Am. J. Physiol. Regul. Integr. Comp. Physiol. 25, 79(1975)
[本文引用: 1]
[26]
M.D. Danford, R.H. Higgins, NASA Tech Briefs (ISSN 0145-319X), 12, 1, p. 64 (1988)
[本文引用: 1]
[27]
M. Schneider, K. Kremmer, C. Lämmel, K. Sempf, M. Herrmann, Corros. Sci. 80, 191(2014)
DOI:10.1016/j.corsci.2013.11.024 URL
The galvanic corrosion risk of metal/SiC-based ceramic coupling in 3.5 wt% NaCl aqueous solution is investigated. Electrochemical measurements as linear sweep voltammetry and galvanic corrosion tests according to DIN-50919 were used. The results clearly show that the resistivity of the SiC-based ceramic plays the key role in the understanding of galvanic corrosion between metal and ceramic. In all cases the ceramic represents the cathodic site of the coupling. The corrosion rate is dominated by the electrical resistance of the ceramic.
[本文引用: 1]
[28]
G.L. Song, Corros. Sci. 52, 455(2010)
DOI:10.1016/j.corsci.2009.10.003 URL
Many practical galvanic corrosion problems can be simplified into a one-dimensional mathematical equation. In this study, theoretical expressions for galvanic potentials and currents of one-dimensional systems are deduced and some critical parameters involved in the potential and current equations are systematically discussed. The developed analytical approach is then applied to some practical galvanic corrosion cases, such as a steel l joint exposed to bio-fuel E85, a galvanic couple separated by a passive spacer and a scratched organic coating. It is demonstrated that the analytical approach can examine the reliability of a computer modeled galvanic corrosion process and help understand the mechanism behind the computer modeled galvanic corrosion behavior.
[本文引用: 1]
[29]
G. Song, B. Johannesson, S. Hapugoda, D. Stjohn, Corros. Sci. 46, 955(2004)
DOI:10.1016/S0010-938X(03)00190-2 URL
An investigation was carried out into the galvanic corrosion of magnesium alloy AZ91D in contact with zinc, aluminium alloy A380 and 4150 steel. Specially designed test panels were used to measure galvanic currents under salt spray conditions. It was found that the distributions of the galvanic current densities on AZ91D and on the cathodes were different. An insulating spacer between the AZ91D anode and the cathodes could not eliminate galvanic corrosion. Steel was the worst cathode and aluminium the least aggressive to AZ91D. Corrosion products from the anode and cathodes appeared to be able to affect the galvanic corrosion process through an “alkalisation”, “passivation”, “poisoning” effect or “shortcut” effect.
[本文引用: 1]
[30]
D.B. Huang, J.Y. Hu, G.L. Song, X.P. Guo, Corros. Eng. Sci. Technol. 48, 155(2013)
DOI:10.1179/1743278212Y.0000000051 URL
The self-corrosion and galvanic corrosion of magnesium based alloys Mg-10Gd-3Y-0.5Zr (GW103) and AZ91D coupled to carbon steel and Cu in ethylene glycol (EG) solution with and without inorganic-organic inhibitor packages Na3PO4 + SDBS and Na3PO4 + benzoate were investigated through measuring their open circuit potentials, coupling potentials, potentiodynamic polarisation and galvanic current densities. It was found that these two inhibitor packages effectively inhibited the self-corrosion of GW103 and AZ91D but accelerated the dissolution of the carbon steel and Cu in the EG solution. When the Mg alloys were coupled with the carbon steel or Cu, their galvanic current densities were also inhibited if the Na3PO4 + benzoate inhibitor package was used in the EG solution. It is surprising that the addition of Na3PO4 + SDBS accelerated the galvanic corrosion. Based on the mixed potential theory, it is believed that the variation in E-corr difference between the anode and cathode and the change of their Tafel slopes after inhibitor addition should be responsible for the galvanic corrosion behaviour of these couples.
[本文引用: 1]
Resource
PDF downloaded times    
RichHTML read times    
Abstract viewed times    

Share
Export

External search by key words

Carbon fiber reinforced polymer (CFRP)     
Composite     
Carbon fiber     
Steel     
Galvanic effect     

External search by authors

Chi Zhang    
Da-Jiang Zheng    
Guang-Ling Song    

Related articles(if any):