Rubber-Induced Corrosion of Painted Automotive Steel: Inconspicuous Case of Galvanic Corrosion

Abstract

Rubber components filled with carbon black are widely used in vehicles for sealing, preventing water ingress, and reducing vibration and aerodynamic noise. However, carbon particles increase the electrical conductivity of rubber. When a carbon-filled rubber part comes into contact with the metal car body, it may act as a cathode, accelerating metal corrosion via galvanic coupling. This study combined volume resistivity and zero-resistance ammeter (ZRA) measurements, resistometric corrosion monitoring, and accelerated corrosion testing to assess the effect of rubber conductivity on the corrosion degradation of painted car body panels in defects. More conductive rubber induced a higher galvanic current and accelerated paint delamination from defects. Real-time monitoring confirmed an earlier onset of corrosion and higher corrosion rates for steel coupled with conductive rubber. These findings emphasize the importance of using low-conductive rubber with resistivity from 10⁴ Ω·m to minimize the risk of galvanic corrosion of the car body.

1. Introduction

Corrosion in the automotive sector includes, in addition to more straightforward issues, some less typical cases of corrosion damage. One such case is the increased corrosion of painted and galvanized steel in areas of contact with rubber elements, which is typically manifested as blistering and delamination of the paint layer and steel rusting after a relatively short operation time [1]. Rubber is widely used in automotive applications to insulate openings, such as doors and windows, prevent water ingress, and reduce vibration and aerodynamic noise, thus increasing passenger comfort [2]. Rubber materials are usually filled with carbon black to improve their antistatic properties, increase durability, and achieve the required appearance. The addition of carbon black increases the electrical conductivity of the material. It depends on the composition, morphology, volume concentration, and distribution of carbon black added to the rubber mixture, the type and structure of the polymer, and the fabrication procedure [3].

As they are attached to the car body, rubber elements come into direct contact with metallic parts. During vehicle operations, protective paint integrity can be affected by stone chipping or scratching. Then, in the presence of de-icing salts at high humidity and/or wet road mud, the surface electrolyte provides sufficient surface conductivity to allow electrical contact between the rubber part and the defect, forming a galvanic couple [1]. As the open circuit potential of steel and zinc is lower than that of graphite in common corrosive environments, carbon-filled rubber will become a cathode when coupled with galvanized steel [1,4] eventually accelerating anodic metal dissolution in defects [1].

The effect of galvanic coupling between bare and painted cold-rolled and hot-dip galvanized steel and aluminum with rubber of different conductivity has previously been studied by LeBozec et al. [1], Colombo [2] and Al-Hatemmi [3]. Corrosion acceleration proportional to rubber conductivity was observed in all studies. LeBozec et al. recommended rubber with resistivity higher than 10⁵ Ω·cm for automotive applications to avoid blistering and filiform corrosion [1].

In recent years, carbon-fiber-reinforced polymers (CFRPs) have become widely used in the automotive, aerospace, and construction industries due to excellent properties, such as their light weight, high strength and good stability [5]. However, electrically conductive carbon fibers act as highly efficient cathodes when connected with most metals [5,6,7]. In their review of the galvanic effect of CFRPs on engineering metals, Song et al. ranked the galvanic current densities of the metals in contact with CFRP based on potentiodynamic polarization curves as follows: AZ31 magnesium alloy > zinc coated steel > steel > aluminium alloy [4]. They concluded that any engineering metal in contact with CFRP is anodic and that preventing direct contact between carbon fibers and the metal is the most effective way to reduce galvanic corrosion [4]. Yang et al. studied the initiation and mechanism of galvanic corrosion in carbon steel/CFRP coupling under atmospheric conditions and reported that CFRP enhanced the cathodic reaction and induced mainly localized pitting-like corrosion of steel near electrical contact points [8]. In another study, the authors observed galvanic corrosion acceleration with temperature [9]. Zhang et al. pointed out a high risk of severe galvanic corrosion of zinc-coated DP590 steel coupled with conductive CFRPs [6]. Wu et al. studied the galvanic coupling between aluminum alloys and CFRPs and observed that the conductivity of CFRPs increased with the concentration of carbon fibers, resulting in a more active corrosion behavior [10]. Adapala et al. also reported corrosion damage of automotive aluminum alloys in contact with CFRPs [11].

Coelho and Olivier used the scanning vibrating electrode technique (SVET) to monitor the inhibitor efficiency to prevent galvanic corrosion of the AA2024 aluminum alloy coupled with graphite [12]. Peng and Nie studied the effect of the galvanic connection between CFRP and steel, aluminum, or titanium alloys and reported that, while steel and the A356 aluminum alloy were severely corroded, the titanium alloy remained almost intact [7]. Liu et al. observed that CFRP acted as an efficient cathode when connected to an uncoated 7075 aluminum alloy. A significant decrease in the galvanic current of up to 90% was achieved through the application of a PEO coating [13]. Pan et al. studied the accelerative effect of CFRP on the corrosion of magnesium alloys AZ31, LZ91, and LZ141, reporting a higher galvanic activity of CFRP/Mg–Li alloys and its increase with lithium concentration [14]. Jun et al. used polymer insulation between CFRP and the magnesium alloy AZ31B to mitigate galvanic corrosion in bolted joints [15].

Zhou et al. reported that the magnesium alloy AZ31 could not be directly connected to silicone rubber containing conductive nickel-coated graphite or nickel-coated carbon fibers due to high galvanic current densities [16]. Hakimian et al. identified galvanic corrosion induced by the use of graphite-containing gaskets as one of the main causes of failure in flanged gasketed joints used in piping systems and pressure vessels [17].

Strategies for preventing galvanic corrosion of engineering metals coupled with conductive CFRPs or gaskets typically involve connection design, material selection, and enhancement of the barrier between the carbon fibers and metal through increasing the thickness of the non-conductive polymer matrix on the composite surface, or the application of a protective coating or inhibitor to the metal. However, these approaches are not fully applicable to preventing the corrosion of pained automotive steel sheets coupled with conductive rubber. Galvanic coupling between car body and conductive rubber parts remains an issue in the automotive industry, suggesting that a deeper understanding of the phenomenon and greater awareness of associated risks are needed. Therefore, this study focused on the following four goals:

• To confirm that galvanic coupling between rubber and galvanized automotive steel can accelerate corrosion.
• To connect and put into context results obtained using electrochemical methods and accelerated corrosion testing.
• To assess the risk of corrosion of painted automotive steel in contact with a series of rubber materials.
• To recommend an optimal approach to minimize the risk of accelerated corrosion of car parts.

2. Materials and Methods

2.1. Samples

A car maker supplied a selection of ten types of rubber stops marked A–J (Figure 1). No information on their properties and dielectric behavior was provided to avoid any confirmation bias. For electrochemical tests, i.e., resistivity and zero-resistance ammeter (ZRA) measurements, as well as resistometric corrosion monitoring, the specimens were cut into circular slices 2–3 mm thick, approximately 20 mm in diameter. For accelerated corrosion testing, the specimens were used as received.

2.2. Volume Resistivity Measurement

The initial evaluation of the supplied rubber specimens was carried out according to the IEC 62631-3-1:2023 standard [18]. It describes the determination of volume resistance and volume resistivity to categorize rubber materials in terms of their potential to induce galvanic corrosion. The exact procedure may vary depending on the tested materials, type of electrodes, voltage, and sample thickness. The setup used in this study is based on a preliminary test described in the standard.

A slice of rubber of known thickness was placed between two electrodes with a defined area of 55 mm² made of polyimide foil with conductive graphite layers. The foil was applied to a copper plate to provide mechanical support and ensure reliable connectivity (Figure 2). The proper connection between the electrodes and the rubber sample was secured using a conductive gel consisting of 200 g polyethylene glycol, 50 mL water, 2.5 g KCl and 0.5 g soft soap. The samples were conditioned at ambient temperature and relative humidity (RH) for 24 h prior to measurement. A constant voltage of 10 V was applied using a potentiostat SP-200 from BioLogic, France, and the electric current that passed through the sample was recorded until stabilization. Volume resistivity ρ_V was determined as follows:

$${\rho }_V={\displaystyle \frac{U\times S}{I\times h}},$$

(1)

where U is the applied voltage, I is the stabilized current passing through the sample, and h and S are the thickness and contact area of the sample, respectively. Five measurements were made for each rubber specimen tested, which was discharged after each measurement.

2.3. ZRA Measurement

ZRA was applied to compare galvanic corrosion current densities in rubber × carbon steel and rubber × zinc galvanic couples. ZRA measurements were performed using the same potentiostat as for the volume resistivity measurement described above. EN 10,130 carbon steel (minimum 97.8% Fe, maximum 0.2% C) and EN 988 zinc (minimum 99.8% Zn) were used. Four types of rubber specimens with different resistivity levels were selected. A rubber sample was connected as a working electrode (WE) in a three-electrode setup, while carbon steel or zinc were connected as a counter electrode (CE). Both electrodes had the same area of 1.5 cm². The setup was completed with a saturated calomel electrode (SCE) as a reference electrode and a 1 wt.% NaCl solution as an electrolyte. This solution was selected as a reasonable approximation of conditions on automotive body parts contaminated with deicing salts during winter [19]. The measurement consisted of a 30-min open circuit potential (E_OC) stabilization without galvanic coupling, subsequent galvanic connection, and 5 min corrosion current density (j_galv) and galvanic potential (E_galv) recording using the potentiostat. Five measurements were performed for each rubber sample tested.

2.4. Resistometric Corrosion Monitoring

Resistometric corrosion monitoring is based on applying a low electric current to a thin metal track exposed to a corrosive environment and recording its electrical resistance [20]. As the track corrodes, its thickness decreases, leading to increased resistance. Since electrical resistance also depends on temperature, one part of the track is covered during exposure to protect it from corrosion, acting as a reference. The corrosion depth (CD) of the uncovered sensing track can then be calculated according to [21,22]

$$CD=t_{init}\left({\displaystyle \frac{R_{ref, init}}{R_{sens, init}}}-{\displaystyle \frac{R_{ref}}{R_{sens}}}\right)$$

(2)

where t_init is the initial thickness of the metal track; R_{ref, init} and R_{sens, init} are the initial reference and sensing track resistance, respectively, and R_ref and R_sens are the actual reference and sensing track resistance, respectively. The real-time corrosion rate can then be assessed as the derivative of the corrosion depth over time [23]. In this study, CorrSen loggers (Gema Ltd., Unhošť, Czech Republic) with sensors consisting of three sensing tracks and one reference tracks were used [24]. Details about the measurement principle and the monitoring system can be found elsewhere [20,25].

The steel corrosion rate was monitored in contact with rubber B. For comparison, coupling of steel with rubber without carbon filler (Inert) was also tested. The measurement was carried out using carbon steel sensors with an original thickness of 50 µm. The sensor reference tracks were covered with two-component epoxy glue. Then, a NaCl solution in methanol was applied on the sensing tracks to achieve a uniform surface contamination of 1 g m⁻² Cl⁻ [26]. After evaporation of the solvent, the rubber pieces were attached to the sensing tracks and fixed using inert strips, covering one half of each sensing track as shown in Figure 3. The rubber pieces were tightly attached to ensure good electrical contact between rubber and steel and to avoid crevices. Finally, the sensors were inserted into electronic loggers.

The loggers were placed horizontally in a climatic chamber and exposed to a climatic cycle consisting of 4 h wet and 4 h dry phases at 95% RH and 50% RH, respectively, separated by 2 h wetting and drying periods. The temperature was kept constant at 35 °C. The experiment was carried out for 7 days (14 climatic cycles). Data were recorded every 10 min and continuously transmitted to a server, where the actual corrosion depth and corrosion rate were evaluated.

2.5. Accelerated Corrosion Test

To confirm electrochemical results and assess the extent of delamination, specimens of painted automotive steel in contact with rubber were exposed to the VW PV 1209 accelerated corrosion test for 4 weeks. The VW PV 1209 test is a combination of the VW PV 1210 accelerated corrosion test and the VW PV 1200 climatic test. The VW PV 1210 part of the test (Figure 4) is a 24 h cycle consisting of a 4 h neutral salt spray (NSS, 35 °C, 5 wt.% NaCl), a 4 h dry phase (18–28 °C, 40–60% RH) and a 16 h wet phase (40 °C, 100% RH). The VW PV 1200 part (Figure 5) consists of 12 h cycles of rapid temperature and RH variations from −40 to 80 °C and 30 to 80% RH, respectively, with constant conditions for 4 h at 80 °C and 80% RH and at −40 °C with uncontrolled RH. Five cycles of the VW PV 1210 test followed by four cycles of the VW PV 1200 test form one week of the combined VW PV 1209 test.

To better simulate service conditions, samples cut from galvanized steel car bodies with three slightly different paint systems denominated as White, Beige and Grey were used in the test; see Figure 6. Rubber B, C, and Inert were selected for the test based on volume resistivity and ZRA data to represent different application properties. Each type of rubber material was placed in four samples of coated metal sheets, three of which (No. 1–3) had an artificial defect. The defect was created by removing the coating layer from the cut edge of the inner hole using P180 emery paper so that the uncovered metal was in direct contact with rubber. Artificial defects were introduced to accelerate the corrosion attack. Paint on cut edges of the samples denominated with I was kept intact. A set of samples without rubber parts was exposed in parallel for comparison.

The average and maximum delamination distance was evaluated according to EN ISO 4628 after the removal of non-adherent paint around defects with a knife [27].

3. Results

3.1. Volume Resistivity

The resistivity of ten rubber stops is given in

Table 1. Volume resistivity of rubber stops.

Rubber Material	Volume Resistivity [Ω·m]	Order of Magnitude [Ω·m]
C	1,064,859	106
H	1699	103
J	275	102
A	213	102
E	202	102
B	133	102
F	101	102
D	23	101
I	20	101
G	12	101

. The data are sorted by decreasing resistivity in view of the potential risk of inducing galvanic corrosion and are color-coded from red (highest potential risk) to green (lowest potential risk). Sample C showed the highest volume resistivity in order of 10⁶ Ω·m, which is 1000 times higher than that of sample H, second in order. In contrast, resistivity of rubber G was the lowest in order of 10¹ Ω·m. For comparison, the volume resistivity of Inert rubber without carbon filler was >10⁸ Ω·m.

3.2. ZRA

Based on the data in Table 1, four materials A, B, C and H were selected for ZRA. Rubber materials D, I, and G with the lowest resistivity were deemed unacceptable for use in contact with automotive steels and were therefore excluded from further evaluation. Since the volume resistivity of rubber parts in the 10² Ω·m range was similar, only materials A and B were tested to reduce the number of samples.

Table 2. ZRA data for rubber × carbon steel galvanic couple.

Rubber	EOC-R [V/SCE]	Egalv [V/SCE]	EOC-Fe [V/SCE]	jgalv [A·m−2]
A	−0.15 ± 0.02	−0.48 ± 0.02	−0.48 ± 0.01	0.46 ± 0.13
B	−0.06 ± 0.02	−0.52 ± 0.01	−0.53 ± 0.01	1.33 ± 0.20
C	0.05 ± 0.00	−0.49 ± 0.01	−0.49 ± 0.01	(2.1 ± 1.3)·10−6
H	−0.22 ± 0.07	−0.49 ± 0.01	−0.49 ± 0.02	0.47 ± 0.10

shows the average values of the open circuit potential of rubber (E_OC-R) and carbon steel (E_OC-Fe), and the galvanic potential (E_galv) and the galvanic current density (j_galv) in the rubber × carbon steel galvanic couple. Identical information for the rubber × zinc galvanic couple is given in

Table 3. ZRA data for rubber × zinc galvanic couple.

Rubber	EOC-R [V/SCE]	Egalv [V/SCE]	EOC-Zn [V/SCE]	jgalv [A·m−2]
A	−0.23 ± 0.00	−0.97 ± 0.00	−0.99 ± 0.00	768 ± 37
B	−0.20 ± 0.03	−0.98 ± 0.01	−1.00 ± 0.00	537 ± 54
C	0.31 ± 0.05	−1.00 ± 0.01	−1.00 ± 0.01	(5.3 ± 3.2)·10−6
H	0.05 ± 0.01	−1.03 ± 0.01	−1.03 ± 0.01	5 ± 1

.

As expected, all rubber materials were significantly more cathodic than the metals. The potential difference between steel and rubber and zinc and rubber was 0.3–0.5 and 0.8–1.3 V, respectively. Rubber was a cathode in all galvanic couples.

For both carbon steel and zinc, there was no significant shift in the potential of the unconnected metal and galvanic potential. Indeed, the metals have much higher conductivity and electrochemical activity and are thus difficult to polarize. However, it should be noted that the rubber and metal samples had equal areas of 1.5 cm². In real automotive applications, the bare metal surface will come into electrical contact with conductive rubbers in small defects, resulting in coupling of a large cathode and small anode, where the polarization can be expected to be stronger. Measurable galvanic current was recorded in all cases, proving that the galvanic connection was established and rubber served as a cathode accelerating anodic dissolution of both metals. The galvanic current density was the lowest in the couples with low-conductive rubber C, being several orders of magnitude lower than for the other three pairs.

3.3. Corrosion Monitoring

On the basis of the volume resistivity and ZRA measurements, rubber B with higher conductivity was selected to evaluate the real-time effect of galvanic coupling together with a reference non-conductive Inert rubber. The average corrosion depth for three measurement tracks of steel sensors with attached rubber B and Inert after seven days of exposure was 16 ± 3 and 6 ± 1 µm, respectively. The coupling with conductive rubber resulted in a 2.6-fold higher corrosion depth.

Figure 7 shows average corrosion rates of the resistometric steel sensors coupled to rubber during each 12 h wet/dry cycle. The corrosion rate per cycle was calculated as the difference in the corrosion depth at the beginning and end of the cycle divided by the duration. In the early stage of exposure (first to third wet/dry cycles), the corrosion rates of the three sensors were similar. From the fourth climatic cycle onward, the corrosion rate of sensor B continued to increase. The average corrosion rate of the Inert sensor remained constant at 0.04 µm h⁻¹ during the first 12 cycles (six days) of exposure and then gradually increased. The experiment showed that the coupling with conductive rubber induced earlier activation and accelerated corrosion of the steel component.

The real-time corrosion rates recorded during the 11th wet/dry cycle are seen in Figure 8 (complete corrosion rate records are available elsewhere [23]). They were the lowest during the dry phase at 50% RH. With wetting, the corrosion rate of sensor Inert increased reaching the maximum of 0.06 µm h⁻¹ at 95% RH and remained about stable throughout the wet phase. Sensor B showed stronger corrosion activation during wetting up to 0.22 µm h⁻¹ and a gradual decrease in the corrosion rate under stable wet conditions. Galvanic coupling with the conductive rubber thus accelerated corrosion, particularly during the wetting process.

3.4. Accelerated Corrosion Test

Figure 9 shows sample appearance after the four-week exposure in VW PV 1209. As shown in detail in Figure 10, corrosion damage was only negligible around defects in samples without rubber (No rubber) and samples with lower conductivity rubber C. On the contrary, significant corrosion damage occurred around defects in samples connected to more conductive rubber B. The red circle indicates the onset of corrosion even on the intact sample with Rubber B. These observations highlight the acceleration effect of galvanic coupling with conductive rubber.

Table 4. Average and maximum delamination distance on painted steel samples No. 1–3 with artificial defects after four weeks of VW PV 1209 test.

Sample	Average Delamination Distance [mm]	Maximum Delamination Distance [mm]
No rubber	0.7 ± 0.2	1.4 ± 0.8
Rubber B	6.9 ± 0.9	9.5 ± 1.7
Rubber C	0.2 ± 0.2	0.4 ± 0.3
Rubber Inert	0.4 ± 0.3	0.6 ± 0.2

provides an evaluation of the average and maximum delamination distances around the inner holes with removed paint at the cut edges according to EN ISO 4628. Since there were no systematic differences between paint systems White, Beige and Grey, they are treated together.

Average and maximum delamination distances around the artificial defects were similar for the samples with inert and low-conductive rubber C. They were slightly higher for the sample without rubber showing on some shielding of the defect area by rubber stoppers. A significantly larger average delamination distance of 6.9 ± 0.9 mm was measured for samples coupled with highly conductive rubber B, which is consistent with previous results.

4. Discussion

The present study aimed to confirm that galvanic coupling between galvanized automotive steel and conductive rubber can accelerate steel corrosion and to establish a link between simple volume resistivity and galvanic current measurements and results of accelerated corrosion testing.

First, volume resistivity was measured to categorize rubber materials according to their potential to accelerate corrosion. The resistivity of the supplied materials used by a car manufacturer varied significantly by up to five orders of magnitude from 10¹ to 10⁶ Ω·m. For subsequent ZRA measurements, rubber parts with volume resistivity in order of 10² Ω·m (A and B), 10³ Ω·m (H) and 10⁶ Ω·m (C) were coupled with carbon steel and zinc. All rubber parts anodically polarized steel, but, while the galvanic current density was in the range of 10⁻¹ to 10³ A∙m⁻² in couples with more conductive rubber materials A, B, and H, it was significantly lower at around 10⁻⁶ A∙m⁻² for the least conductive rubber C. The difference was greater for zinc than for carbon steel.

Real-time resistometric measurements and exposure of painted galvanized steel samples in the accelerated corrosion test showed that galvanic coupling with conductive rubber indeed accelerated steel corrosion. The monitoring revealed that the average corrosion rate of carbon steel pre-contaminated with sodium chloride increased within a particular wet/dry cycle. A similar effect was previously observed during cyclic corrosion tests by Prošek et al. [21,26], Van den Steen et al. [28], Popova and Prošek [29] and Mizuno [30]. Connecting steel sensors with conductive rubber sped up this effect, as the corrosion rate increase was observed after 4 wet/dry cycles for the sensor coupled with rubber B, but only after 12 cycles for the sensor coupled with Inert rubber. The average corrosion rate was 2.6 times higher for steel coupled to conductive rubber B, accelerating especially during the wetting process. These results are consistent with those of Adapala et al., who measured the galvanic current between aluminum alloys AA6××× and CFRPs in the Ford CETP 00.00-L-467 accelerated test [11]. The authors reported that the highest galvanic current was measured during the chloride spray phase and transition phases with a maximum at 75% RH due to an increase in the NaCl concentration in the thin electrolyte close to the deliquescence point [11]. Yang et al. observed an increase in the galvanic current density in the carbon steel/CFRP coupling as the thickness of the NaCl electrolyte layer decreased from 0.5 mm to 0.3 mm [8].

Exposure of painted automotive steel panels with artificial defects coupled with low-conductive rubber C in the VW PV 1209 accelerated test led to negligible corrosion degradation with the delamination distance similar to that for the steel–Inert rubber combination. Coupling with highly conductive rubber B enhanced paint delamination by an order of magnitude.

Based on the results of the real-time resistometric measurement and the accelerated corrosion test, the following corrosion acceleration mechanism of automotive steel panels in contact with conductive rubber can be proposed (Figure 11). On the top and bottom sides, steel is protected against corrosion by a zinc coating and organic paint, but the cut edges in contact with rubber are only painted (Figure 11a). During wetting, adsorption of water molecules results in formation of a thin electrolyte layer with a high concentration of corrosive ions and a rapid supply of oxygen to the surface, creating particularly corrosive conditions (Figure 11b) [21,25,31,32]. As a result of combined mechanical and environmental degradation factors, the paint film at the cut edge is likely to fail locally during service. In such a defect, zinc and steel come into contact with carbon-filled rubber. The electrolyte layer ensures the electrical connection between the metal and rubber, enabling efficient galvanic coupling (Figure 11c). Carbon particles in rubber become cathodes where oxygen reduction reaction proceeds, enhancing the active dissolution of zinc during wet phases [12]. Once the protective zinc layer degrades in a sufficiently large area, the steel substrate is not galvanically protected anymore and becomes the principal anode in the couple, corroding rapidly. The proposed mechanism is consistent with the assumption that during vehicle operation, the corrosion acceleration effect of galvanic coupling with conductive rubber will be mainly pronounced when road salt, mud and high humidity form a conductive surface electrolyte connecting rubber and bare metal [1].

LeBozec et al. studied blistering and filiform corrosion on painted cold-rolled steel, hot-dip galvanized steel, and aluminum alloy 6016 with defects coupled with conductive rubber and exposed at 35 °C and 85% RH after being covered with a thin layer of artificial mud containing 1 wt.% NaCl [1]. The authors reported that corrosion was accelerated in the presence of rubber with resistivity of up to 10² Ω·m for all studied materials. These observations are consistent with the results of this study. Rubber B with resistivity in the same range caused intense paint delamination in VW PV 1209 and induced a high galvanic current in connection with carbon steel and zinc. Based on their results, LeBozec et al. recommended rubber materials with resistivity higher than 10³ Ω·m to avoid corrosion risks [1]. Rubber H complied with this requirement, but the galvanic current in the steel-rubber couple was identical to that of rubber B, which accelerated paint delamination in the cyclic test. For the zinc-rubber couple, rubber H induced a galvanic current two orders of magnitude lower than for rubber B. Although these findings are not fully conclusive due to the missing exposure of painted samples coupled with rubber H in the cyclic test, we propose moving the threshold value defining the risk of galvanic corrosion above 10³ Ω·m. Indeed, rubber C with resistivity 10⁶ Ω·m was fully safe. Since no rubber with resistivity in the range from 10³ to 10⁶ Ω·m was included in the testing programme, an exact value cannot be provided. However, it is probable that rubber with resistivity from 10⁴ Ω·m will be safe in view of activation of galvanic corrosion.

The minor discrepancy between this study and the work of LeBozec et al. is most probably linked to the different exposure conditions, although the sample geometry and corrosion stability of the selected paint system can also play a role.

It should be pointed out that the resistivity of all rubber parts supplied by the industrial partner for testing, except for rubber C, was lower than the safe threshold. This suggests that, although the risk of galvanic corrosion between rubber and steel is well documented, it still remains an issue of practical concern in the automotive industry.

5. Conclusions

Volume resistivity and ZRA measurements, real-time corrosion monitoring, and accelerated corrosion testing were carried out to study the galvanic coupling of painted galvanized steel car body sheets and conductive carbon-filled rubber parts. On the basis of the results obtained, the following conclusions can be drawn:

• In corrosive environments, the car body is susceptible to the acceleration of corrosion in defects due to galvanic coupling with carbon-filled rubber. The extent of paint delamination decreases with increasing rubber resistivity. Rubber with resistivity from about 10⁴ Ω·m is considered safe in view of galvanic corrosion and corrosion acceleration.
• Laboratory galvanic current and even simple volume resistivity measurements correlated well with results of the accelerated cyclic corrosion test of painted automotive steel samples coupled with rubber and can therefore be used to estimate the risk of corrosion damage.
• The strict specification and verification of the electrical volume resistivity of rubber materials by automotive manufacturers is required to eliminate the potential detrimental effect of galvanic corrosion.