Printer Friendly

Electrochemical characterization of coated self-piercing rivets for magnesium applications.

ABSTRACT

This work reports on measurement and analysis of the galvanic interaction between steel self-piercing rivets (SPRs) having several different surface conditions and magnesium alloy substrates under consideration for use in automotive structural assemblies. Rivet surface conditions included uncoated steel, conventional Zn-Sn barrel plating and variations of commercial aluminizing processes, including supplemental layers and sealants. Coating characteristics were assessed using open circuit potential (OCP) measurement, potentiodynamic polarization scanning (PDS), and electrochemical impedance spectroscopy (EIS). The degree of galvanic coupling was determined using zero-resistance ammeter (ZRA) and the scanning vibrating electrode technique (SVET), which also permitted characterization of galvanic current flows in situ.

CITATION: Upadhyay, V. , Qi, X., Wilson, N., Battocchi, D. et al., "Electrochemical Characterization of Coated Self-Piercing Rivets for Magnesium Applications," SAE Int. J. Mater. Manf. 9(1):2016,

INTRODUCTION

Self-Piercing Riveting (SPR) is an established mechanical fastening technique with a history dating over 100 years [1]. The technology gained prominence during the 1990s in the automotive industry when the manufacturer Audi adopted SPR to join body parts in its all-aluminum A8 vehicle [2], [3]. The Jaguar XJ [4], and the 2015 Ford F-150 [5] are also recent examples of the use of SPR in aluminum vehicle body construction. Steel SPRs are finding increased use in the automotive industry because they enable the joining of alternative materials which are often difficult or impossible to join using more traditional resistance spot welding. Additionally, SPR may enable the joining of mixed materials such as steel to Al [4, 6, 7, 8, 9]. The "self-piercing" aspect also allows for fastening of materials without the extra step of drilling a hole in a substrate, although it does require two-side access to the joint of interest.

Rivets used in SPR are predominately made from hardened and tempered steel in order to accomplish both the piercing function as well as having sufficient ductility to deform in order to create the clinching effect by the rivet. Boron steels permit a high degree of hardenability with only modest degrees of alloying. The rivets are generally treated with a barrier or sacrificial coating prior as part of the manufacturing process, since the typical grades of steel employed are susceptible to corrosion in an uncoated condition. However, when the rivet is used to join non-ferrous materials such as Al or Mg, the applied coating also needs to act as a buffer to prevent galvanic coupling between the more cathodic steel and the anodic light metal being fastened. This is an inversion of the usual concern in protection of steel fasteners (i.e. where the steel of the fastener is the anode or corroding species). For the case where a nominally steel fastener is used in joining of more anodic metals (e.g. magnesium or aluminum), the approach becomes one of effectively starving the cathodic (usually hydrogen reduction) reaction occurring on the steel surface and thereby limiting the anodic dissolution of the light metal in contact. Iron and magnesium are a particularly insidious couple due to both the large differential in corrosion potentials for the dissimilar metals in a common electrolyte, but also the large proclivity for hydrogen reduction on iron surfaces in aqueous environments.

A number of studies have examined the corrosion behavior of SPRs and the impact of corrosion on the fastener performance [10, 11, 12, 13, 14, 15]. A recent project, "Magnesium Front End Research and Development (MFERD)," undertaken by the United States Automotive Materials Partnership, LLC (USAMP), with support by the U.S. Department of Energy, has emphasized the importance of understanding issues in joining Mg to itself and other materials for automotive applications [16, 17, 18, 19, 20]. Although SPRs have been identified by this project as a candidate for Mg joining, corrosion at the joint areas is an issue that must be addressed. Further, a paucity of established quantitative measurement techniques has made it challenging to assess the effectiveness of potential rivet coatings in mitigating corrosion at the interface between the rivet and the base materials. The work reported here was undertaken to explore the viability of several candidate electrochemical techniques in characterizing the efficacy of selected coating to steel SPRs and their impact on the subsequent behavior of the rivet as the cathode of a galvanic cell including adjacent magnesium alloy.

Electrochemical techniques namely open circuit potential (OCP), electrochemical impedance spectroscopy (EIS), potentiodynamic polarization scan (PDS), galvanic corrosion (GalCorr) in zero resistance ammeter (ZRA) mode, and scanning vibrating electrode technique (SVET), were used in this work with the aim of obtaining information about the rivets, and their interaction with magnesium. OCP measurement was used to obtain information on the tendency towards reactivity/corrosivity of the rivet under study. PDS added to the information by providing corrosion rate, corrosion current density etc. EIS provided information on protective properties of coatings on the rivets. These techniques can therefore provide information during rivet selection process in their early stage before their use as a galvanic couple. SVET was used to obtain precise information of the anodic and cathodic regions along with their current density, whereas GalCorr was used to provide an understanding of the overall galvanic interaction of the rivet-Mg couple. Both SVET and GalCorr measurement are extremely useful in studying the coupling interaction between dissimilar materials.

EXPERIMENTAL

Materials

Rivets and Sample Preparation

The 5mm head diameter self-piercing rivets used in this study were produced by Henrob Corp. (1) and supplied through the United States Automotive Materials Partnership LLC (USAMP) (2). The rivets were made of 10B37 boron steel and heat treated to hardness of 47 RC. A micrograph of self-piercing rivet in its applied form is given in Figure 1.

Coatings

Among the four rivets studied, one was uncoated 10B37 whereas three of the rivets were surface modified/coated. The detail of the surface modifications is listed in Table 1 along with their photograph in Figure 2.

The 3 layer coating with polymeric top coat* are as follows:

1. 'AlumiPlate' electrolytic aluminum coated to 0.001" (25 microns). The AlumiPlate process is a non-aqueous electroplating of aluminum tonominal thicknesses of approximately 25 microns (Performed by AlumiPlate Inc. Minneapolis, MN, USA).

2. Henkel "Alodine[R] [EC.sup.2] [TM]" - electroceramic coating applied to the first aluminum layer. This is an inorganic electrophoretic coating from HenkelCorp, Madison Hts., MI, USA.

3. Henkel "29000[TM]" sealant/topcoat applied over the [EC.sup.2] treatment. This is a proprietary polymer sealant from Henkel Corp, Madison Hts., MI, USA.

The self-piercing rivets (Figure 3a) were properly masked to facilitate electrochemical measurements. The inside surface of the tubular portion of the rivet was abraded mechanically to ensure exposure of the base metal. A thin copper conducting wire was then placed coaxially with the tubular rivet shank, and completely filled with a conductive epoxy resin, purchased from MG Chemicals, Canada (CAT. No. 8331-14G, part A and part B). This ensured adequate electrical conductivity through the wire in contact with the rivet, and facilitated rivet connection to the measuring instrument. The resin was cured for 24 h. After resin cure, the neck of the SPR, extending to a certain length of the wire, was masked using a heat shrink tube. This ensured isolation of the wire and SPR/wire coupling region from electrolyte exposure (Figure 3b and 3c).

Magnesium

For the galvanic experiments, die-cast AM60B Mg plate was used as the coupling agent whereas Mg alloy AM30 was used for SVET experiments. For SVET, the rivet of interest was potted into a pre-machined flat piece of AM30 wrought magnesium, in order to produce a coplanar surface with rivet head and surrounding magnesium. Wrought AM30 was selected in this case for greater homogeneity in comparison to the die-cast AM60B.

Method

The following electrochemical measurements were performed on the rivet samples:

1. Open circuit measurement (OCP): rivets only

2. Electrochemical impedance spectroscopy (EIS): rivet only

3. Potentiodynamic polarization scan (PDS): rivet only

4. Galvanic corrosion measurement (GalCorr): rivet- AM60B matrix

5. Scanning vibrating electrode techniques (SVET): rivet-AM30 matrix

Electrochemical impedance spectroscopy (EIS), and potentiodynamic polarization scan (PDS) was performed using a 3 electrode set-up as shown in Figure 4a and 4b. The SPR acted as the working electrode, a platinum mesh as the counter electrode, and a saturated calomel electrode (SCE) was used as the reference electrode. The electrolyte used was aqueous 3.5 wt. % NaCl.

An IFC 1000 Potentiostat/Galvanostat/ZRA, from Gamry Inc. (3) was used for the electrochemical measurements. Data were acquired using Gamry Framework[TM] software version 6.21. A Perspex[TM] cylinder mounted on a non-reactive plastic board and clamped with an O-ring insert was used to hold the electrolyte, aqueous 3.5 wt. % NaCl. For the EIS measurement, impedance response from [10.sup.-2] Hz to [10.sup.5] Hz was measured and the data were acquired at the rate of 10 points per decade. In EIS a small AC potential perturbation, typically a sine wave of amplitude ~ [+ or -]10mV is applied on a system with respect to its open circuit potential over a wide range of frequency (typically from [10.sup.5]-[10.sup.-2] Hz) and the response of the current is measured at each frequency. Barrier properties, water uptake and diffusion rate of organic coating, or the various processes involved during corrosion can be obtained [21, 22, 23, 24, 25, 26].

PDS scans were performed by scanning the potential of the working electrode from -0.5 volts to +1.5 Volts with respect to the open-circuit potential (OCP) at a scan rate of 1mV/sec. In PDS, the potential of the working electrode is scanned at a selected rate over a relatively large potential range by applying a current through the electrolyte. It is a very commonly used testing technique for measuring corrosion resistance, corrosion current density, etc. of metals.

The experimental arrangement for the galvanic corrosion measurements consisted of two compartments separated by a salt bridge as shown in Figure 5. One compartment was a glass beaker filled with the electrolyte of interest (aqueous 3.5 wt. % NaCl) into which the shrouded SPR assembly wasimmersed as working electrode 1 (WE-1), as well as the saturated calomel reference electrode (SCE). The second compartment consisted of a die-cast magnesium plate (AM60B) as working electrode 2 (WE-2) mounted with a Perspex[TM] cylinder and clamped with O-ring insert to hold the electrolyte. An agar salt bridge was used to connect the two compartments. Salt bridge completes the electrical circuit, maintain charge balance and allow the cell to function. The experiment was performed in the zero resistance ammeter (ZRA) mode of the Gamry potentiostat previously described. The ZRA mode ensures that the two working electrodes (WE-1 and WE-2) behave as if they were directly connected by a 'zero-resistance' wire, such that the galvanic current between cathode (rivet) and anode (magnesium), and potential can be measured/recorded. Material loss rate at the anode can be then determined by application of the Faraday law once the current flow is measured.

In a largely different style of experiment involving the coated rivets and their galvanic interaction with surrounding magnesium, the scanning vibrating electrode technique (SVET) was employed, permitting an "in situ" determination of the coupling between any particular rivet and adjacent magnesium as might be encountered in an actual riveted joint. For these experiments, however, the rivets as acquired were mechanically pressed into pre-drilled holes in the substrate of interest - magnesium alloys, in this case. This technique uses a vibrating probe/electrode to scan the sample surface that detects the gradient in ohmic potential produced by local currents generated from actively corroding surfaces immersed in the electrolyte. From the variation in potentials, areas of varying current densities can be located and measured. Details of the SVET methodology can be found elsewhere [27, 28, 29].

SVET experiments were performed to measure the current density distribution between the rivet head of interest and surrounding magnesium for the geometric situation illustrated in Figure 6 (a), throughout the volume defined by the bi-metal/electrolyte interface at its free corrosion potential (i.e. there were no externally applied potentials). The SVET instrument used in these experiments was purchased from Applicable Electronics, USA (4). The scanning probe was a Pt-Ir microelectrode with a 10[micro]m diameter tip that was platinized to a 20[micro]m sphere. A 2D schematic of the SVET process is shown in Figure 6a. The probe was placed 150 [micro]m above the sample surface and vibrated with an amplitude of 20 [micro]m along the X and Y direction. A pair of platinized Pt wires were used as a reference and bath ground electrode respectively. Scans were initiated immediately after immersion in an equal volume mixture of 0.05 wt. % NaCl and 0.05 wt. % MgSO4. This particular solution, determined after a number of trials, displayed less corrosivity for the rivet-Mg couple under study, when compared to electrolytes of higher ionic concentration where the galvanic reaction was vigorous with copious hydrogen evolution, and wherein the SVET measurements were difficult due to this disruption. For the results illustrated here, the defined probing area included a matrix of 20x20 individual point measurements within an approximately 3x3 mm field, wherein a scan of the entire data field lasted approximately 6 minutes. Two such scans were performed, with an 80 minute interval between. The experimental probing arrangement for the rivet and surrounding magnesium is illustrated in Figures 6 (b) and 6 (c), and consisted of portions of the rivet head and adjacent magnesium. Figure 6 (c) illustrates the image of the probed array seen through the instrument microscope, with overlay image of the probe locations in the test array. Areas of the galvanic couple that were isolated from contact with the electrolyte were defined using a combination of acetate solution (e.g. nail polish) with carefully overlayed transparent tape. This then confined the volume of measured activity as well as the insulating boundary conditions for electric fields generated by the couple.

RESULTS AND DISCUSSION

Conceptual Development

Coatings to steel rivets for the purpose of thwarting galvanic corrosion with adjacent magnesium in assembled structures is an "inverse" problem to that of providing barrier or sacrificial coatings to steel fasteners as a means of "self-protection" from corrosion. There is a vast industry devoted to self-protection of steel fasteners and dozens of alternative processes are commercially available for this purpose. One aim of the work reported here was to explore likely electrochemical characterization tools for the purpose of determining suitable approaches to assess performance of various coatings to hardened steel self-piercing rivets with express purpose of minimizing or eliminating local galvanic contact. As indicated above, a variety of characterization tools were available to approach the problem, results of which are described in the following sections.

Open-circuit potential and DC polarization studies are principally directed to the more metallic coatings and their characteristics as electrodes. In particular, the "open-circuit" or free corrosion potential provides an indication of the coated rivet's behavior relative to known points of reference such as the baseline steel, zinc and its alloys, aluminum or polymeric insulations (which often do not readily yield stable metallic-like electrode characteristics and must be treated as dielectrics using AC methods such as impedance spectroscopy). Using a "mixed-potential" approach may indicate whether the coating behaves more like the base element of the coating (e.g. zinc) or reflects a potential and electrode behavior more like iron. The latter case is particularly devastating for galvanically-inducted corrosion of magnesium since Fe ions catalyze Mg oxidation. OCP testing was selected as an initial look at the coated electrode and comparison to known elements. In the OCP tests, the free corrosion potential of IVD-Al coated rivets, Zn/Sn-coated rivets and uncoated steel rivets were monitored as shown in the Figure 7a, and the influence of different coatings was evident from the different OCP observed.

The galvanic coupling current density between coated steel rivets and bare Mg (AM60B) as using the ZRA configuration of Figure 5 is shown in Figure 7 (b) and indicates clearly the differences in coupling current with the selected magnesium alloy observed for the several rivet coatings.

Electrochemical Impedance Spectroscopy (EIS)

The incorporation of impedance spectroscopy data in characterization of rivet coatings is exemplified by Figure 8, which depicts the initial impedance modulus (i.e. |Z|) vs test frequency commonly referred to as a 'Bode plot' for rivet-only samples exposed to 3.5% NaCl solution, using the experimental arrangement of Figure 4(a). Testing using AC methods such as EIS was intended to provide some insight into the dielectric nature of the coatings and its susceptibility to breakdown in a corrosive environment. Ideally, the more insulating (i.e. better dielectric) the coating, the greater opportunity to minimize the galvanic coupling. The 3 layer coated sample, (PC - Table 1), displays the maximum observed impedance at low frequency, whereas the uncoated sample, UC, displays much lower impedance, typical of an exposed metal. Among the two metallic coatings ZS (Zn/Sn coated) and IA (IVD-Al coated), IA shows a greater low-frequency dielectric response compared to ZS. This could be due to the propensity for formation of an aluminum oxide or hydrated layer in the IA. Both aluminum and oxygen were detected on the surface of IA (Table 2) using Energy-dispersive X-ray spectroscopy (EDX) in conjunction with scanning electron microscopy (SEM). Figure 9 illustrates regions of the IA sample examined for topography and elemental composition by SEM-EDX. Table 2 indicates values for elemental composition of the near-surface layers as deduced from EDX. Impedance measurement of sample PC had not reached a steady-state at the initial time of measurement and hence a second measurement impedance scan was conducted after 24 hours of immersion. Data for this scan is illustrated in Figure 10. High impedance was observed after 24 hours as well at a steady OCP. Figure 11 compares the barrier/insulating property of the rivets, evidenced by the low-frequency impedance, which is effectively a measure of the dielectric strength of the protective layer. Measurement indicate that sample PC provides maximum barrier protection and hence isolation, based on this criterion. Among the metallic coatings, the data indicated that IA provided a superior dielectric coating than did the baseline ZS, although this measurement did not consider the resultant dielectric coupling between the nominally steel rivet and surrounding magnesium.

Potentiodynamic polarization curves from the four sample types of Table 1 are compared in Figure 12. The uncoated steel rivet sample (UC) is the control and is the basis for comparison of the coated samples PC, ZS and IA. It can be seen that the 3 layer coated rivet, PC, displayed lowest corrosion current density, with lowest [i.sub.corr] value of the group, suggesting that this coating had provided a superior physical barrier to the underlying steel interface, thereby obscuring reactive sites and retarding the apparent cathodic reaction more effectively than the other coatings examined. The [i.sub.corr] values, in each case, were obtained using the Tafel ft subroutine of the Gamry Echem analyst software (5), considering the potential range of -0.3 to +0.5 V with respect to the free corrosion potential, [E.sub.corr], established at steady state. Sample PC also displayed a low rate of anodic current density rise, and resistance to high anodic polarization. The coating was observed to be intact following the polarization run, with no visual sign of damage. Coatings ZS and IA displayed rapid rise in current density above [E.sub.corr] indicative of low polarization resistance and implied greater ease of the metal corrosion. However some amount of protection to the bare steel rivet (UC) was afforded by the aluminizing (IA) as compared to coating ZS in terms of barrier properties. Sample ZS displayed maximum corrosion current density of all samples (Figure 13), whereas sample IA displayed corrosion current intermediate between PC and ZS, but less than UC. From the figure it is clear that the polymer coated rivet, PC, was the most resistant to corrosion.

Galvanic Experiments

Results from the galvanic (ZRA) experiments, as shown in Figure 14a, reveal distinct differences between the various galvanic interactions of the coated rivets with the Mg alloy. The galvanic interaction of the 3 layer coated rivet, (PC), with the Mg alloy resulted in a very low galvanic current ([approximately equal to] [10.sup.-9] A.[cm.sup.-2]), compared to the galvanic interaction of uncoated rivet, UC, ([approximately equal to]4.74x[10.sup.-4] A.[cm.sup.-2]), Zn/Sn coated rivet, ZS, ([approximately equal to]2.18 x [10-.sup.4] A.[cm.sup.-2]) and IVD-Al coated rivet, IA, ([approximately equal to]4.20x[10.sup.-4] A.[cm.sup.-2]) with Mg. This implies that the galvanically induced corrosion of the particular magnesium alloy (in this case AM60B), due to the reaction with the steel rivet was at a minimum when the 3 layer rivet coating was employed. The sample PC acts to suppress the cathodic activity of the rivet and the couple behaves largely as that of the Mg alloy, as can be observed from the mixed potential value which is similar to the OCP of pure AM60B (Figure 14b and c). The ZS coated steel-Mg couple displayed higher galvanic current compared to the PC-Mg couple, its surface being metallic and acting as a cathodic site, facilitating the anodic dissolution reaction of the Mg. The galvanic couple UC-Mg, and IA-Mg displayed similar current but still higher than the couple ZS-Mg at a more positive mixed potential. The IVD aluminizing of the rivet did not greatly change its potential as a cathode from that of the bare steel.

Figure 15 illustrates the visual appearance of the rivet head and magnesium coupon at the conclusion of the galvanic test runs. The visuals, particularly appearance of dissolution of the magnesium component appeared to correlate with the current densities experienced. The rivet-metal galvanic interaction (PC-Mg) that displayed the lowest current density also experienced the lowest visible corrosion damage to the magnesium anode. In contrast, the galvanic interaction between the IVD aluminum-coated steel rivet (IA-Mg), yielding the maximum galvanic current also displayed maximum visible corrosion damage to the magnesium. Interestingly, the bare-steel rivet (UC-Mg) couple, while displaying greater corrosion of magnesium than the PC-Mg couple, actually showed less galvanic effect than for the aluminized (IA) rivet. Visible corrosion at Mg due to the galvanic interaction of Zn-Sn coated steel with Mg lay somewhere between that of the UC-Mg and PC-Mg. Based on the photographic images of the Mg after the galvanic experiment, a ranking for Mg corrosion was made using the ImageJ (6) software package, used to distinguish areas of corrosion with non-corroded areas based on the color differences between the two. A scale of 0 to 100% was used for comparison with 0 being no visible corrosion and 100% being the maximum visible corrosion. The measurement had an error of [+ or -]10%. Values from this analysis are shown in Figure 15.

The mass loss of Mg following passage of the galvanic coupling current in the ZRA experiments can also be estimated using the Faraday Law of Electrolysis, stated as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

where, in this instance, Q is the total charge transfer required to dissolve mass m, z is the charge of the ion, M is the molecular mass and F is Faraday's constant. Total charge, Q, was obtained by integrating the total current over time from the galvanic experiment data (Figure 14a) using the integration program from Origin software (7). M for pure Mg is 24.3 g/mole, z is +2 and F = 96500 C/mole, assuming corrosion of pure Mg at valence +2. The mass was then converted to the more commonly stated corrosion rate (CR) in mils per year (MPY), using the formula

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

where W is the weight change in milligrams, D is the density in g. [cm.sup.-3], A is the area in [inch.sup.2], and T is the exposure time in hours. Corrosion is assumed to be uniform.

Table 3 shows the calculated corrosion loss of the AM60B Mg test plate and corresponding corrosion rate in typical units. It is observed that Mg corrosion in the couples UC-Mg and IA-Mg displayed almost similar amount of Mg loss and corrosion rate, which is consistent with the measured galvanic current values (Figure 14a). The couple PC-Mg displayed lowest Mg mass loss and corrosion rate followed by the couple ZS-Mg. The visual inspection of Mg corrosion is also very consistent for couples PC-Mg and ZS-Mg. However higher level of Mg corrosion is visually observed for the couple IA-Mg compared to UC-Mg, though their mass loss and corrosion rate are almost similar. A possible explanation for this could be the higher amount of pitting in the UC-Mg couple compared to more uniform corrosion in the IA-Mg couple.

Scanning Vibrating Electrode Technique (SVET)

Figure 16 (a-h) shows the SVET- measured in situ current density distribution maps on the surfaces of the various rivets-Mg couples studied in this investigation. Positive values indicates anodic current (oxidation/corrosion) and negative values indicates cathodic current (reduction). Additionally, contour maps of the current density are also plotted at the bottom of the 3D maps. Mapping was performed both upon instant immersion and after 80 minutes of continued immersion, for all the rivet-Mg couples. From the variation in potentials, areas of varying current densities can be located and measured, or various anodic and cathodic regions can be observed and quantified [30], [31], [32].

For the PC-Mg couple it can be seen that current is nearly zero towards the rivet side with an almost flat plateau with no positive or negative peaks and valleys. Electrochemical activity, mostly cathodic, is observed at the interfacial and Mg region with a few anodic peaks at the Mg side. After 80 minutes of immersion, it was observed that the electrochemical activity at the rivet side was again almost zero. Cathodic activity is observed at the interface with few anodic peaks at the Mg. The 3 layer coated rivet greatly reduced anodic activity at the rivet as well as the interface. For the couples UC-Mg and ZS-Mg, initially, a clear distinction between the rivet and Mg can be observed from the current density mapping in Figures 16c-f. The rivet acts mostly as a cathode, whereas the rivet-Mg interface and Mg acts as the anode of such local cell. Anodic activity is also observed on the surface of the rivet, indicating that the rivet is not completely cathodic and anodic spots where anodic activity can take place exists. Since the scanning area is very small compared to the total area of the assembly, more such anodic spots may exist. After 80 minutes, it was observed that for both UC-Mg and ZS-Mg couples, the anodic areas (oxidation of the metal) at the rivet side had increased even though the magnitude of the anodic activity is less than the activity at the Mg side. For the couple ZS-Mg, a distinction between rivet and rivet-Mg interface based on the mapping could not be made.

For the couple IA-Mg, a distinction between interface and Mg was impossible even during the initial scan. Anodic activities were observed on most part of the rivet, in addition to Mg, which showed higher anodic activity. Very few cathodic regions at the rivet side were observed. After 80 minutes the cathodic activity at the rivet increased. For the 3 layer coated rivet (PC), results in SVET were similar to those of the galvanic (ZRA) observation in that this coating resulted in the least amount of activity between rivet and adjacent magnesium. The measure of the total anodic current (shown in Figure 17 also shows similar results to the ZRA experiments. The PC-Mg couple displayed the least anodic current. The rest of the couples displayed almost similar values with the couple ZS-Mg displaying slightly lower values than couples UC-Mg and IA-Mg, as also observed in the galvanic experiment.

CONCLUSIONS

Selected electrochemical characterization tools (open-circuit potential (OCP), potentiodynamic polarization scan (PDS), electrochemical impedance spectroscopy (EIS), Galvanic corrosion (GalCorr) measurement using zero-resistance ammeter (ZRA) and scanning vibrating electrode technique (SVET) have been used to characterize several coatings to steel self-piercing rivets for assessment of their galvanic coupling interaction with adjacent magnesium alloys. EIS and PDS proved to be useful tools for characterizing the corrosion resisting properties of the rivets and coatings themselves, while ZRA and SVET allowed the study of galvanic coupling of coated rivets and magnesium in situ. Among the rivet coatings studied, the 3 layer coated rivet (PC) demonstrated the greatest resistance to galvanic interaction with Mg. SVET was used to provide both qualitative and quantitative information on the local galvanic interaction between the SPR's and adjacent Mg alloy as may be encountered in an actual joint. The PC-Mg couple displayed low total current density, and the total anodic current observed had a similar trend to that observed in the galvanic coupling measurements. These techniques complements each other and together can provide a more detailed understanding of the interaction between the SPR's and Mg alloys in assemblies, as well as tools for comparing efficacy of various rivet coatings in suppressing galvanic corrosion in magnesium-containing structures.

REFERENCES

[1.] Danyo, M.W., "Self-piercing riveting (SPR) in the automotive industry: an overview," in: Chrysanthou, A. and Sun, X. B. T.-S.-P. R., eds., Self-Piercing Riveting: Properties, Processes and Applications, Woodhead Publishing, ISBN 978-1-84569-535-4: 171-180, 2014, doi:10.1533/9780 857098849.2.171.

[2.] Chrysanthou, A., "Introduction," Self-Piercing Riveting: Properties, Processes and Applications, ISBN 9780857098849: 1-7, 2014, doi:10.1533/9780857098849.1.

[3.] Kelkar, A., Roth, R., and Clark, J., "Automobile Bodies : Can Aluminum Be an Economical Alternative to Steel?," JOM 53(8):28-32, 2001, doi:10.1007/s11837-001-0131-7.

[4.] Mortimer, J., "Self-Piercing Rivets: The Key to Future Joining Technology," Auto Technol. 3(1):42-43, 2003, doi:10.1007/BF03246751.

[5.] Priddle, A., "Ford F-150 engineer provides stick-to-it attitude for the rivets team," Detroit Free Press. July 5, 2014.

[6.] He, X., Pearson, I., and Young, K., "Self-pierce riveting for sheet materials: State of the art," J. Mater. Process. Technol. 199(1-3):27-36, 2008, doi:10.1016/j.jmatprotec.2007.10.071.

[7.] Marko, I., Thomala, W., Haldenwanger, H.-G., Kudliczka, H., and Schmid, G., "Self-piercing rivet," Patent US6325584 B1, USA, 2001.

[8.] Cotterill, A., Blacket, S.E., and Singh, S., "Self-piercing riveting method and apparatus," Patent US5752305 A, 1998.

[9.] Budde, L. and Wilhelm, L., "Stanznieten ist zukunfittrachtig in der Blechverarbeitung (Piercing riveting has a promising future in sheet metalprocessing)," Bander Bleche Rohre 5:94-100, 1991.

[10.] Chrysanthou, A., "Corrosion behavior of self-piercing riveted joints," in: Chrysanthou, A. and Sun, X., eds., Self-Piercing Riveting: Properties, Processes and Applications, Woodhead Publishing, Cambridge, UK, ISBN 9780857098849, 2014, doi:10.1533/9780857098849.1.41.

[11.] Chen, Y.K., Han, L., Chrysanthou, A., and O'Sullivan, J.M., "Fretting wear in self-piercing riveted aluminium alloy sheet," Wear 255(7-12): 1463-1470, 2003, doi:10.1016/S0043-1648(03)00274-6.

[12.] Howard, R. and Sunday, S., "The Corrosion Performance of Steel Self-Piercing Rivets When Used with Aluminum Components," SAE Technical Paper 831816, 1983, doi:10.4271/831816.

[13.] Esfahani, M., Durandet, Y., Wang, J., and Wong, Y.C., "Effect of Joining Process on the Coatings of Self-Piercing Rivets," Adv. Mater. Res. 488-489:1501-1505, 2012, doi:10.4028/www.scientific.net/AMR.488-489.1501.

[14.] Calabrese, L., Bonaccorsi, L., Proverbio, E., Bella, G. Di, and Borsellino, C., "Durability on alternate immersion test of self-piercing riveting aluminium joint," Mater. Des. 46:849-856, 2013, doi:10.1016/j.matdes.2012.11.016.

[15.] Mandel, M. and Kruger, L., "Determination of pitting sensitivity of the aluminium alloy EN AW-6060-T6 in a carbon-fibre reinforced plastic/aluminium rivet joint by finite element simulation of the galvanic corrosion process," Corros. Sci. 73:172-180, 2013, doi:10.1016/j.corsci.2013.03.033.

[16.] Forsmark, J.H., Li, M., Su, X., Wagner, D.A., Zindel, J., Luo, A.A., Quinn, J.F., Verma, R., Wang, Y.-M., Logan, S.D., Bilkhu, S., and McCune, R.C., "The USAMP Magnesium Front End Research and Development Project - Results of the Magnesium 'Demonstration' Structure," Magnesium Technology 2014, John Wiley & Sons, Inc., ISBN 9781118888179: 517-524, 2014, doi:10.1002/9781118888179.ch93.

[17.] Luo, A., Lee, T., and Carter, J., "Self-Pierce Riveting of Magnesium to Aluminum Alloys," SAE Int. J. Mater. Manuf. 4(1):158-165, 2011, doi:10.4271/2011-01-0074.

[18.] Luo, A.., Quinn, J.., Verma, R., Wang, Y.., Lee, T.., Wagner, D.., Forsmark, J.., Su, S., Zindel, J., Li, M., Logan, S.., Bikhu, S., and McCune, R.., "The USAMP Magnesium Front End Research and Development Project: Focusing on a Demonstration Structure," Light Met. Age, 2012.

[19.] Luo, A.A., Nyberg, E.A., Sadayappan, K., and Shi, W., "Magnesium Front end Research and Development: A Canada-China-USA Collaboration," Essent. Readings Magnes. Technol. 41-48, 2014, doi:10.1002/9781118859803.ch6.

[20.] McCune, R., Forsmark, J., Upadhyay, V., and Battocchi, D., "Characterization of Coatings on Steel Self-Piercing Rivets for use with Magnesium Alloys," in: Manuel, M. V., Singh, A., Alderman, M., and Neelameggham, N. R., eds., Magnesium Technology 2015, ISBN 978-1-119-08243-9, 2015.

[21.] Murray, J.N., "Electrochemical test methods for evaluating organic coatings on metals: an update. Part I. Introduction and generalities regarding electrochemical testing of organic coatings," Prog. Org. Coatings 30(4):225-233, 1997, doi:10.1016/S0300-9440(96)00677-7.

[22.] Leidheiser Jr., H., "Electrical and electrochemical measurements as predictors of corrosion at the metal-organic coating interface," Prog. Org. Coatings 7(1):79-104, 1979, doi:10.1016/0300-9440(79)80038-7.

[23.] Mansfeld, F., "Use of electrochemical impedance spectroscopy for the study of corrosion protection by polymer coatings," J. Appl. Electrochem. 25(3):187-202, 1995, doi:10.1007/BF00262955.

[24.] Bierwagen, G., Tallman, D., Li, J., He, L., and Jefficoate, C., "EIS Studies of coated metals in accelerated exposure," Prog. Org. Coatings 46(2):149-158, 2003, doi:10.1016/S0300-9440(02)00222-9.

[25.] Barsoukov E. and Macdonald J. R., Impedance Spectroscopy: Theory, Experiment, and Applications. Wiley, 2005.

[26.] Loveday, D., Peterson, P., and Rodgers, B., "Evaluation of Organic Coatings with Electrochemical Impedance Spectroscopy Part 1 : Fundamentals of Electrochemical Impedance Spectroscopy," JCTR 8(August):88-93, 2004.

[27.] Yan, M., Gelling, V.J., Hinderliter, B.R., Battocchi, D., Tallman, D.E., and Bierwagen, G.P., "SVET method for characterizing anti-corrosion performance of metal-rich coatings," Corros. Sci. 52(8):2636-2642, 2010, doi:10.1016/j.corsci.2010.04.012.

[28.] Santana, J.J., Gonzalez-Guzman, J., Izquierdo, J., Gonzalez, S., and Souto, R.M., "Sensing electrochemical activity in polymer-coated metals during the early stages of coating degradation by means of the scanning vibrating electrode technique," Corros. Sci. 52(12):3924-3931, 2010, doi:10.1016/j.corsci.2010.08.010.

[29.] Isaacs, H.S., "The Use of the Scanning Vibrating Electrode Technique for Detecting Defects in Ion Vapor-Deposited Aluminum on Steel," Corrosion 43(10):594-596, 1987, doi:10.5006/1.3583835.

[30.] Mansfeld, F. and Kenkel, J.., "Laboratory studies of galvanic corrosion I. Two-Metal Couples," Corrosion 31(8):298-302, 1975, doi:http://dx.doi. org/10.5006/0010-9312-31.8.298.

Vinod Upadhyay, Xiaoning Qi, Nick Wilson, Dante Battocchi, and Gordon Bierwagen

North Dakota State University

Joy Forsmark

Ford Motor Company

Robert McCune

Robert C. McCune & Associates, LLC

CONTACT INFORMATION

Vinod Upadhyay

Center for Surface Protection

North Dakota State University 1735 Research Park Drive Fargo, ND 58102

vinod.upadhyay@ndsu.edu

ACKNOWLEDGEMENTS

This work was supported by United States Council for Automotive Research LLC (USCAR). This material is based upon work supported by the Department of Energy, National Energy Technology Laboratory under Award Number No. DE-EE0005660. This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. Such support does not constitute an endorsement by the Department of Energy of the work or the views expressed herein.

APPENDIX

A brief description of various electrochemical techniques used in this study

1. OPEN CIRCUIT POTENTIAL (OCP) MEASUREMENT

Open circuit potential, also called rest potential or corrosion potential, is the potential of a working electrode when it is at rest. OCP provides information on the tendency of a metal towards reactivity/corrosivity. A metal with more negative potential will always corrode preferentially. OCP information is useful in designing sacrificial galvanic protection. For example, zinc, which is electrochemically more negative compared to steel is used to protect steel galvanically. The information obtained from OCP measurements are however thermodynamic in nature and provides information only on the possibility but do not provide information on the rate of reaction/corrosion.

2. ELECTROCHEMICAL IMPEDANCE SPECTROSCOPY (EIS)

In EIS a small AC potential perturbation, typically a sine wave of amplitude ~ [+ or -]10mV is applied on a system with respect to its open circuit potential over a wide range of frequency (typically from [10.sup.5]-[10.sup.-2] Hz) and the response of the current is measured at each frequency. Barrier properties, water uptake and volume fraction, diffusion rate in organic coating, or the various processes involved during corrosion can be obtained. Quantitative results are obtained. A three electrode configuration is normally used. The metal substrate acts as the working electrode. Platinum mesh, graphite rods, or any other noble metal can act as the counter electrode. A saturated calomel electrode or other electrodes such as silver/silver chloride or mercury/mercury sulfate can also be used as the reference electrode. Very often modeling of EIS data is performed to obtain quantitative information [21, 22, 23, 24, 25, 26].

3. POTENTIODYNAMIC POLARIZATION SCAN (PDS)

In PDS, the potential of the working electrode is scanned at a selected rate over a relatively large potential range by applying a current through the electrolyte. The corrosion behavior is interpreted from the nature of the polarization plot. Corrosion rate, corrosion resistance, corrosion current, corrosion potential, passivity, pitting susceptibility, and corrosion mechanisms can be predicted. It is a very commonly used testing technique and find wide application for the corrosion study of metals and alloys.

4. GALVANIC CORROSION (GALCORR)

Galvanic corrosion measurement measures the galvanic current and potential between two electrodes. The experiment in the zero resistance ammeter (ZRA) mode ensures that the two working electrodes (WE-1 and WE-2) behave as if they were directly connected by a 'zero-resistance' wire, such that the galvanic current between cathode and anode, and the mixed potential between them can be measured/recorded.

5. SCANNING VIBRATING ELECTRODE TECHNIQUE (SVET)

SVET can be used for localized study of a surface behavior. In SVET a vibrating probe/electrode scans the sample surface and detects the gradient in ohmic potential produced by local currents generated from actively corroding surfaces immersed in the electrolyte. From the variation in potentials, areas of varying current densities can be located and measured. In other words, localized corrosive areas can be separated with inert areas [27, 28, 29].

(1.) Henrob Corporation, 30000 South Hill Road, New Hudson, Michigan MI 48165. USA.

(2.) 1000 Town Center Drive, Suite 300, Southfield, MI 48075, USA.

(3.) Gamry Instruments Inc, 734 Louis Drive, Warminster, PA 18974, USA. www.gamry.com

(4.) 22 Buckingham Drive, Sandwich, MA 02563, USA. http://www.applicableelectronics.com

(5.) Gamry Instruments Inc, 734 Louis Drive, Warminster, PA 18974, USA. http://www.gamry.com

(6.) http://imagej.nih.gov/ij/

(7.) One Roundhouse Plaza, Suite 303, Northampton, MA 01060, USA. www.originlab.com

Table 1. Rivet Coating type used in this study.

S. No.  Rivet coating type                         Sample Code

1       Uncoated steel rivet (10B37 boron steel)        UC
2       3 layer coating with polymeric top coat*        PC
3       Henrob Zn-Sn baseline barrel coated steel
        rivet                                           ZS
4       Ion vapor-deposited Aluminum (IVD-A1)
        coated steel rivet                              IA

Table 2. EDS of IVD-Al coated sample, IA.

                       Position 1    Position 2    Position 3
Element [down arrow]   WP1   AP1     WP2    AP2    WP3    AP3

Carbon                22.38  38.12  20.32  34.45  22.18  37.92
Oxygen                 7.69   9.83  13.73  17.47   6.95   8.92
Aluminum              67.31  51.05  61.40  46.33  68.82  52.38
Chromium               1.48   0.58   3.09   1.21   0.78   0.31
Iron                   1.14   0.42   1.47   0.53   1.28   0.47

                      Position 4
Element [down arrow]   WP4    AP4

Carbon                16.57  30.41
Oxygen                 5.65   7.78
Aluminum              73.51  60.06
Chromium               1.96   0.83
Iron                   2.31   0.91

WP= Weight percent AP= Atom percent

Table 3. Corrosion mass loss and corrosion rate of Mg after
galvanic experiment

Galvanic  Total charge  mass loss (m) of Mg/  Corrosion rate of Mg/
Couples   (Q)/coulomb   milligrams            Mils per year (MPY)

PC-Mg       0.00148     00.0001                 0.0008
UC-Mg     732.512       92.2282               415.40
ZS-Mg     341.299       42.9718               193.54
IA-Mg     727.436       91.5892               412.52
COPYRIGHT 2016 SAE International
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Upadhyay, Vinod; Qi, Xiaoning; Wilson, Nick; Battocchi, Dante; Bierwagen, Gordon; Forsmark, Joy; McC
Publication:SAE International Journal of Materials and Manufacturing
Article Type:Report
Date:Jan 1, 2016
Words:6879
Previous Article:Damage initiation and fatigue behavior of carbon-fiber composite disk springs.
Next Article:On practical implementation of the Ramberg-Osgood model for FE simulation.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters