Introduction

Lead-acid batteries together with lithium-ion batteries are the backbone of the global rechargeable battery market [1, 2]. In recent years, due to the development of renewable energy sources, there has been an increasing demand for energy storage systems, including modern lead-acid batteries [3,4,5]. One of the most promising direction for the development of lead-acid batteries involves employing porous carbon materials as a current collector in electrodes, replacing lead grids [6,7,8,9]. This research path was initiated by Czerwiński in the 90 s of the twentieth century [10, 11]. Since then, numerous studies have been conducted on the utilization of carbon materials such as reticulated vitreous carbon (RVC) [12,13,14,15,16] pitch-based carbon foam [17,18,19], carbonized corrugated paper [20,21,22], graphite [23, 24] and conductive porous carbon (CPC) [25, 26].

Another type of lead-acid battery current collector modification is the use of copper. Copper expanded metal, covered with lead, is used in negative electrode grids in stationary batteries OCMS (O – Ortfest, ger. Stationary Copper-Stretch-Metal) [4, 27, 28]. A similar solution based on composite grids made of alternating copper and lead layers is obtained by the galvanic method [29, 30]. The main advantages of both technologies are higher electric conductivity, grater mechanical strength of the grid and possibility of using thinner plates.

There are also research studies exploring incorporation of copper into current collectors based on polymer materials. Soria et al. [31] proposed PNS (polymeric network structure) current collector made of polymeric matrix modified with copper and lead. After optimization, the obtained collector showed increased mechanical strength, less weight (approx. 30%) and similar electric conductivity compared to standard grid. Another example is acrylonitrile–butadiene–styrene (ABS) polymer modified successively with layers of copper (10 µm), lead (100 µm) and polyaniline (1 µm) presented by Martha et al. [32]. Tests carried out on positive and negative electrodes of the experimental 6 V battery showed the stable performance over 120 cycles and the specific energy close to 50 Wh/kg. Ji et al. demonstrated that a current collector based on polyurethane foam modified with copper and lead exhibits similarly good electrochemical properties [33]. The results obtained for each polymer-based collectors are promising. However, their practical application strongly depends on the final production costs that are comparable to standard batteries, while maintaining the desired levels of quality and reliability.

In turn, Dai et al. [34] studied current collector for the negative electrode based on lead-coated copper foam without polymer matrix. The performance test results revealed good cyclic stability and 25% improvement in the utilization of active mass compared to conventional cast grids.

In this work, carbon foam grids based on reticulated vitreous carbon (RVC®) modified with a metallic copper-lead bilayer (Pb/Cu/RVC) were prepared and examined as current collectors for lead-acid battery positive plate. Their performance was investigated using a combination of techniques, including scanning electron microscope (SEM), cyclic voltammetry, electrochemical impedance spectroscopy and gravimetric corrosion rate test. This work serves as a follow-up to a previous research study [35] that explored the feasibility of using the described collector in the negative electrode. Development of a positive plate current collector containing carbon and copper elements is more difficult due to the highly corrosive working conditions of this electrode and the possibility of direct reaction of the carbon matrix with PbO2 present in the positive active mass.

Experimental

Electrode preparation

Experimental electrodes were prepared based on reticulated vitreous carbon (RVC®) with 20 ppi (pores per inch) porosity grade, purchased from ERG Material and Aerospace Corporation (USA). Reticulated vitreous carbon is an inexpensive carbon material characterized by low density, open pore structure and good electrical conductivity. The carbon matrix has been modified by galvanic deposition of the copper and lead layers. Before electroplating, the carbon matrix was chemically and electrochemically pretreated. The copper electroplating process was conducted in a sulfate bath with the RVC as the cathode, and a cooper plate as the anode. Copper coatings (20 µm thick) were deposited for 48 min at a current density of 2.0 A/dm2. In this work, copper coatings with a thickness of 20 µm were used, similarly to the previous studies [35]. As the tests have shown, a copper layer with a thickness of 20 µm allows for a significant increase in electrical conductivity and ensures the mechanical durability of the current collector. Following, the lead electroplating process was conducted in a methane sulfonate bath with the Cu/RVC as the cathode, and a lead plate as the anode. Lead coatings (80 µm thick) were deposited for 246 min at a current density of 0.6 A/dm2. Electrodes with different lead layer thicknesses (used for corrosion rate tests) were obtained by adjusting the Pb layer deposition time to 123 min for the 40 µm coating, and to 370 min for the 120 µm coating. The copper and lead coating depositions were carried out by a method developed in previous studies, described in the patents [36, 37].

Characterization and electrochemical tests

Morphological properties and elemental composition analysis of copper and lead coating were performed by the JEOL JSM-6490LV scanning electron microscope (SEM) coupled with an X-ray dispersion analyzer (EDS).

The electrochemical properties and the tightness of the outer lead coating were tested using the cyclic voltamperometry method. The cyclic voltammetry was undertaken on carbon matrix of 10 mm × 10 mm × 5 mm dimension modified with the 20 μm Cu and 80 μm Pb thick double metal layers (Pb80/Cu20/RVC) as a working electrode at a scan rate of 0.05 V/s using the Voltalab PGZ 301 electrochemical workstation. The reference electrode was Hg|Hg2SO4|1.0 M H2SO4 (MSE) and the counter electrode was a large piece of lead sheet. For comparison, the cyclic voltammetry was also performed on a carbon matrix with the same dimensions modified with lead (100 µm thick) (Pb100/RVC) and a metallic lead electrode with a geometric area of ca. 5 cm2. The dimensions of the electrodes were selected so that both reticulated (based on RVC) and metallic electrodes had a similar geometric surface.

Additionally, the effect of the copper layer on the electrochemical properties of the carbon based current collector was investigated by electrochemical impedance spectroscopy. The electrode set-up was identical as used during the cyclic voltammetry measurements. A corrosion film was formed on the studied electrode before the impedance measurements. The corrosion film on the lead surface was formed by polarizing it at a potential of 1.4 V for 20 min. Then, an impedance spectrum was recorded at the same potential. The frequency band was between 1 kHz and 10 mHz, with amplitude 0.01 V. For comparison, the impedance spectroscopy was also performed on a carbon matrix modified with lead (100 µm) (Pb100/RVC). The recorded impedance spectra were processed by EIS Spectrum Analyser software.

The corrosion resistance of the examined current collectors was estimated by the weight loss of the samples after their oxidation in the galvanostatic conditions and the oxide film removal. The measurements were carried out for a series of carbon matrices with dimensions of 20 × 10 × 5 mm modified with a double metal layer of 20 µm copper and lead layers of various thicknesses: 40 µm (Pb40/Cu20/RVC), 80 µm (Pb80/Cu20/RVC) and 120 µm (Pb120/Cu20/RVC), respectively. A comparative measurement was also made on a matrix of the same dimensions modified with a 100 µm lead layer (Pb100/RVC). The gravimetric corrosion resistance tests were carried out by galvanostatic polarization of the samples with a current density of 20 mA/cm2 for 24 h (1 day) at a temperature of 50 °C. The counter electrode was a lead sheet with dimensions of 150 × 50 × 2 mm. After polarization the electrodes were thoroughly washed with distilled water. The oxide film formed was removed after 30 min in solution of the following composition: 100 g/dm3 of sodium hydroxide and 20 g/dm3 of sucrose. Based on the literature review [38, 39], it was found that this solution does not dissolve the lead components, but only the corrosion layer. Further, the electrodes were washed with distilled water, dried, and weighed. The corrosion weight loss in mg/cm2 was calculated by the following formula:

$$\Delta m= \frac{{m}_{o}-{m}_{x}}{A}$$

where m0 is the initial weight of the electrode, mg; mx is the weight of the electrode after removal of the corrosion products, mg; A is the surface area of the electrode before the test, cm2.

After weighing, the cleaned electrode was used in subsequent polarization cycles (of 20 mA/cm2 for 24 h). The described procedure was repeated cyclically 4 times (or until the integrity of the electrode was lost) for each sample. The analysis of the electrode structure after corrosion tests was carried out using a scanning electron microscope.

For all electrochemical measurements 4,9 M H2SO4 solution was utilized as the electrolyte. All experiments, except for the corrosion rate tests, were carried out at a temperature of 25 ºC.

Result and discussion

Structure of metallic coatings

The microscopic analysis of the coating involves: the individual surface inspections of the copper and the lead layers, and the examination of the metallic copper-lead bilayer cross-section. In addition, the SEM image of the unmodified RVC carbon matrix is included. The obtained SEM images are presented in Fig. 1.

Fig. 1
figure 1

SEM image of: RVC carbon matrix (A), copper layer (B), lead layer (C) and metallic copper-lead bilayer–cross-section (D). Magnifications according to the scale in the images

Full size image

The results of microstructural analysis show that in both cases (for the copper and lead layers – Fig. 1B, C) a tight coatings were obtained. They completely covered the previous substrate, and they form a uniform and crack-free structure. At the same time, the obtained copper layer is smoother, while the lead layer is characterized by a more irregular, coarse-grained structure. The crystal structure of the obtained coatings is significantly influenced by properties of the electroplating bath used. This dependence primarily stems from factors such as the properties of the metal ion being reduced, the anions present in the solution, the concentration of metal ions in the bath and the type of the basic electrolyte. From the evaluation of the metallic coating structure, it can be concluded that during the deposition of the copper coating, the rate of the new crystallization nuclei formation is relatively higher compared to the lead coating deposition. Consequently, a large amount of new, slow-growing crystals are formed, leading to the formation of a fine-crystalline, smooth deposit of metallic copper layer. Importantly, the tests focus on a thin copper coating with a 20 µm thickness and a four times thicker lead loading. For this reason, when comparing those structures, the metal crystal size in the coating should be considered as they are not the same throughout their thickness. With an increase in the thickness of the layer, one can observe a simultaneous increase in the deposited metal crystal size, leading to the disappearance of smaller crystals and the formation of a coarser crystalline structure layer. This effect may further explain the more irregular structure of the lead coating. Nevertheless, a lead coating provides a good coverage and protects the copper substrate against damage caused by electrolyte penetration.

In addition, an analysis of the individual metallic layer composition of the obtained coating was performed using the EDS method in the cross-section area (Fig. 2).

Fig. 2
figure 2

SEM image, EDS spectra and EDS maps (copper — yellow, lead — violet) form the selected areas of the cross-section of the metallic copper-lead bilayer

Full size image

The obtained EDS spectra confirm the two-layer character of the metallic coating. The spectrum obtained from the inner coating (deposited directly on the carbon substrate) contains only typical copper signals (Fig. 2.1) whereas, the spectrum recorded from the outer coating shows individual lead signals (Fig. 2.2).

Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) measurements

Furthermore, cyclic voltammetric curves were recorded to study general characteristics of the electrochemical behaviour of the Pb/Cu/RVC electrode in sulfonic acid and compare it with Pb/RVC and solid Pb electrodes. Figure 3 presents cyclic voltammetric curves of the tested electrodes within a potential range from hydrogen evolution to oxygen evolution.

Fig. 3
figure 3

Cyclic voltammogram for Pb80/Cu20/RVC, Pb100/RVC and solid Pb electrodes in potential area from -1.5 V to 1.9 V (vs. MSA); scan rate 0.05 V/s; cycle number — 25

Full size image

All recorded curves are characterized by the following processes: oxidation of Pb to PbSO4 (anodic peak a at ca. -0.8 V), oxidation of PbSO4 to PbO2 and the oxygen evolution (anodic peak b at ca. 1.7 V), reduction of PbO2 to PbSO2 (cathodic peak c at ca. 1.0 V), oxidation of inner layers of Pb and PbO to PbSO4 (peak d at ca. 0.9 V), reduction of PbO to Pb (cathodic peak e at ca. -0.9 V) and reduction of PbSO4 to Pb (cathodic peak f at ca. -1.2 V). The number and distribution of current peaks for the curves obtained on reticulated electrodes (Pb80/Cu20/RVC and Pb100/RVC) are the same and, at the same time, very similar to the results obtained for a solid lead electrode (Pb). A noticeable difference is the slight broadening of the oxidation peak a present at the potential of 0.7 V for carbon matrix based electrodes. This effect was also observed by Gyenge et al. [12]. They proposed that the Pb to PbSO4 oxidation occurring on the reticulated electrode and the PbO sublayer generation arise as two consecutive reactions instead of parallel processes. Moreover, in the case of RVC-based electrodes, the oxidation peak b appears at slightly more positive potentials and consequently, the reduction peak c for these electrodes is smaller. This effect is a result of shifting the PbSO4 to PbO2 oxidation potential towards more positive values. Under the experimental conditions, this shift leads to the formation of a smaller oxide layer, subsequently resulting in a reduced peak (c) of PbO2 reduction. The curve obtained for the electrode modified with a copper-lead double layer shows a typical characteristics for metallic lead. Current signals associated with copper redox processes are not observed. This indicates that the obtained lead layer is tight and the inner copper layer is not in direct contact with the electrolyte.

The influence of the copper layer on the electrochemical properties of the RVC-based electrodes was investigated using the EIS method. In the study, the EIS spectrum of the Pb80/Cu20/RVC electrode was analyzed in comparison to the analogous spectrum obtained for the electrode without an inner copper layer (Pb100/RVC). The obtained Nyquist plots (dependence of the imaginary component of the impedance as a function of the real one) and the fitted equivalent circuit are shown in Fig. 4. The equivalent circuit presented in this study was applied to characterize the impedance behavior of the Pb electrode under similar conditions in other research works [40,41,42,43,44].

Fig. 4
figure 4

The Nyquist plots for Pb80/Cu20/RVC and Pb100/RVC electrode with corrosion films formed on their surface

Full size image

The equivalent circuit consists of the resistor R1 and constant phase element CPE1 connected in parallel with the resistor R2. The resistor R1 corresponds to the ohmic resistance of the electrolyte. In turn, resistor R2 describes charge transfer resistance and the CPE1 element can be attributed to the capacity of electrical double layer. Nyquist plots for both electrodes exhibit an excellent fit to the equivalent circuit. The calculated values of the elements of the proposed model are shown in Table 1. Under this conditions (high, positive electrode potential), it was assumed that the electrode is mainly covered with PbO, and the presence of PbSO4 is negligible.

Table 1 Equivalent circuit parameters for the examined electrodes
Full size table

Nyquist plots obtained for both examined electrodes (Pb80/Cu20/RVC, Pb100/RVC) show semicircles, although with different diameters. Such a course of dependence indicates the activation control of electrode processes. As shown in Table 1, the electrolyte resistance (R1) for both systems is similar. Regarding the parameters related to the capacity of the electrical double layer (CPE1), two effects are observed. For both electrodes, the values of the parameter n are similar. However, the electrode modified only with lead (Pb100/RVC) exhibited a higher value of the parameter P, which suggests a larger specific area of this electrode. There is a clear difference in the case of parameter R2. The electrode modified with metallic double layer (Pb80/Cu20/RVC) showed a clearly lower value of the R2 parameter, indicated by the smaller radius of the Nyquist plot semicircle. This result shows that a lower charge transfer resistance is observed in the case of the copper-modified electrode.

Gravimetric corrosion rate tests

The results of gravimetric measurements of the corrosion rate for all tested current collectors (Pb40/Cu20/RVC, Pb80/Cu20/RVC, Pb120/Cu20/RVC, Pb100/RVC) are presented in Fig. 5. The parameter expressing the corrosion rate was determined as the slope of the obtained curves, which represents the loss of sample mass per time unit relative to the unit area of the electrode (Table 2).

Fig. 5
figure 5

Weight loss of the examined electrodes in successive cycles of corrosion rate testing

Full size image
Table 2 Calculated corrosion rate parameters
Full size table

During these measurements, the electrode modified with a 40 µm lead layer (Pb40/Cu20/RVC) lost its integrity already on the 2nd measurement cycle, therefore the value obtained in the first measurement was adopted as the corrosion rate. Similarly, electrode Pb80/Cu20/RVC lost its integrity on the 4th measurement cycle, thus this data point was not included in the determination of the corrosion rate. In both cases, a deposition of metallic copper on the lead counter electrode was observed in the last cycle of the test (in which the tested electrode was damaged). This evidences the unsealing of the lead coating and the rapid dissolution of the inner copper layer.

SEM images of all samples after the corrosion tests are presented in Fig. 6. In the case of the electrode modified with a double metal layer, which was not mechanically damaged (Pb120/Cu20/RVC), the coating cross-section was also tested using the EDS mapping method (Fig. 7).

Fig. 6
figure 6

SEM images of electrodes (collectors) after corrosion rate tests

Full size image
Fig. 7
figure 7

SEM image and EDS map of the distribution of elements: copper (yellow), lead (violet) for the cross-section of the Pb120/Cu20/RVC electrode coating after corrosion tests

Full size image

As show in Table 2, the calculated corrosion rate parameters for all electrodes are very similar. Until the tightness of the lead coating is maintained, the corrosion rate of all tested systems is comparable. On the other hand, cracking of the lead coating leads to a rapid oxidation of the copper sublayer and the carbon matrix resulting in a damage of the current collector. The time at which this effect was observed is directly correlated with the thickness of the outer lead coating. For electrodes that have not been damaged (Pb120/Cu20/RVC and Pb100/RVC), pits and roughening resulting from corrosion processes are visible on the surface of the lead coating. Moreover, the cross-sectional examination of the Pb120/Cu20/RVC electrode after 96 h of polarization indicates that the applied coating preserved its metallic bilayer character. Deep cracks and delamination are not observed.

Conclusions

The current collector of the positive plate of a lead-acid battery obtained on the basis of reticulated vitreous carbon (RVC) modified with a metallic copper-lead bilayer was presented and examined. The microscopic and electrochemical measurements revealed that the obtained coatings are dense metallic layers with electrochemical characteristics similar to those of the corresponding solid lead electrodes. The confirmed tightness of the lead coating is a parameter of particular importance in the context of applying of the tested current collector in the positive plates as it significantly influences the electrode's durability. The use of an internal copper layer allows to reduce the charge transfer resistance of the electrode processes, which in practice can reduce the polarization and energy losses in the cell and increase the efficiency of the battery. Corrosion rate tests carried out on diverse composition of the examined collector showed that until the tightness of the lead layer is maintained, the influence of the copper sublayer presence on the corrosion rate of the tested current collectors is not observed. The corrosion stability of the electrode depends on the thickness of the outer lead layer. Under the working conditions of the positive plate (protective) lead coating shields the electrode against oxidation of the carbon matrix as well as the copper sublayer. The presented study provides compelling evidence that RVC modified with metallic copper-lead bilayer can be successfully implemented as a current collector for the positives plate of a lead-acid battery.