1. 1. Introduction As a prospective candidate energy storage system for hybrid-electric vehicles (HEVs), lead-acid batteries have many advantages over other competitive technologies, including low initial cost, mature manufacturing technology, good safety performance and high recycling efficiency [1, 2]. However, in HEVs, the battery is operated continuously in high-rate partial-state-of-charge (HRPSoC) mode. In this HRPSoC working mode, the active mass of batteries can not be utilized sufficiently 1

and the discharged product, PbSO4, cannot be efficiently converted back to its original form resulting in progressive accumulation of irreversible PbSO4 on negative active material (NAM) [3-6], and therefore, the cycle life of batteries is reduced. To address this problem, a new type of lead-acid battery, called as UltraBattery, was developed [7-9], which combines an asymmetric super-capacitor and a lead-acid battery in one unit cells. Other studies reported that the introduction of carbon materials such as carbon black [10], carbon nanotube [11, 12], active carbon [13] and graphite [14, 15] to the negative plates of lead-acid batteries can improve the HRPSoC cycle life of the battery due to the utilization of the high specific surface and super-capacitance characteristics of these carbon materials. Shiomi et al. [10] proved that the addition of carbon black to the NAM improved the charge acceptance significantly and retarded the sulfation of negative plates during the simulated HRPSoC test of HEV batteries. Pavlov et al. [13] reported that addition of electrochemical active carbon in the negative paste improved the charge efficiency and slowed down the sulfation of negative plates. At the same time, carbon materials can supply double-layer capacitance during the processes of high rate charge and discharge or pulse discharge, and also reduce the damage of the negative plates, resulting from their buffer actions [16, 17]. However, the introduction of carbon materials in the negative plates accelerates the evolution of hydrogen in charging process so as to increase the inner pressure and the water loss of lead-acid batteries. Some researches try to overcome this problem by modifying activated carbon materials [18-20] and adding inhibitors in the negative paste or the electrolyte to depress the hydrogen evolution process on the negative plate [21]. Jiang et al. [18] proved that nitrogen group-enriched activated carbon demonstrated lower hydrogen evolution rate and higher specific capacitance, which can inhibit hydrogen evolution and improve the battery charge acceptance and charge retention ability. Pavlov et al. [21] demonstrated that adding poly-aspartic acid to the negative paste or to the electrolyte can retard the sulfation of the negative plates and increase the overpotential of hydrogen evolution on the lead electrode. In general, the carbon additives in the negative plate of lead-acid batteries need to have good super capacitance performance and low hydrogen evolution ability. Currently, conductive polymers/ carbon materials (carbon nanotubes and graphene) composites have been widely used as electrode materials for super capacitors due to their high capacitance [22-24]. Because polypyrrole (PPy) has good biological compatibility, high electrical conductivity and high pseudocapacitance [25-27], while grapheme oxide (GO) has high accessible surface area and conductivity [28], PPy/ GO composites for super capacitors have excited a lot of interest in recent years [29-31]. More importantly, PPy has high hydrogen evolution overpotential and good stability in acid media [32, 33], so PPy/GO composites may be a suitable additives for the negative plate of lead-acid batteries. In this paper, different PPy/GO composites were prepared using in situ polymerization synthesis method, and characterised by fourier transformed infrared (FTIR), scanning electron microscopy (SEM), the X-ray photoelectron spectroscopy (XPS) and specific surface area and porosity analyzer. PPy, GO and PPy/GO composites were added in the NAM of battery to form negative plates and simulated 2

test cells. The influence of these additives on the electrochemical performance of the negative plate and on the HRPSoC cycle performance of the simulated test cell were studied by polarisation curves, cyclic voltammetry (CV), electrochemical impedance spectrum (EIS) and galvanostatic charge-discharge methods. The influence of these additives on the microstructure and crystal morphology of the negative plate were also analysed using SEM and specific surface area and pore analyzer. The mechanisms involved were discussed.

2. 2. Experimental

2.1. Reagents and materials Pyrrole (99%, Aldrich Chemical Co.) used for the synthesis of PPy was distilled and kept refrigerated in the dark before use. Other reagents and chemicals were analytical reagents (AR) grade and used as received. Solutions were prepared with deionized water.

2.2. Preparation of GO, PPy and PPy/GO composites GO was synthesized from graphite powder with a modified Hummer’s method [34, 35]. PPy was prepared by a chemical polymerization method using FeCl3·6H2O as oxidant [36, 37]. The Fe3+/Py molar ratio ( nFe3 / n py ) was controlled at 1 and the

polymerization temperature was 5 . The obtained PPy was filtered and rinsed with deionized water and ethanol, and then was dried under vacuum at 60 for 24 h. The PPy/GO composites were prepared by in situ chemical polymerization of pyrrole on GO [36]. The weight feed ratio of pyrrole (Py) to GO ( mpy / mGO ) was varied as 0.5:1, 1:1 and 20:1, and the resulting PPy/GO composites were denoted as PG0.5, PG1 and PG20, respectively. GO was firstly dispersed in deionized water by ultrasonication for 30 min. A certain amount of pyrrole was added to the dispersion of GO and ultrasonication was continued for another 30 min. Then, the FeCl3 solution was added dropwise into the Py/GO mixture ( nFe3 / n py =1) and the in situ polymerization reaction was conducted under vigorous stirring at 5 for 24 h. The obtained PPy/GO composites were rinsed with deionized water and ethanol, and then the composites were dried under vacuum at 60 for 24 h.

2.3. Preparation of negative plates The negative plates were prepared using lead oxide power (oxidation degree = 76%), acetylene black (0.2 wt.%, versus the lead oxide, the same as below), short fibers (0.05 wt.%), barium sulfate (0.4 wt.%), humic acids (0.7 wt.%), H2SO4 (12.5 wt.%, 1.25 g cm−3), H2O (12.5-15 wt.%) and PPy, GO or PPy/GO composites (1.0 wt.%). In this paper, the negative plate without PPy, GO and PPy/GO composites is considered as the blank (NAM) plate, while PPy, GO and PPy/GO composites are considered as additives in the NAM. The negative plates were prepared according to the above formulation as following steps: firstly, acetylene black, short fibers, barium sulfate, humic acids and a PPy/GO composite were added in an agate mortar and ground uniformly; then, the lead oxide was added to the agate mortar and continually ground to form an uniform mixture. Secondly, deionized water (12.5 wt.% versus the 3

lead oxide) was added to the mixture and mixed well, and then H2SO4 was added dropwise to the mixture to form lead paste while deionized water was used to adjust the consistency of the lead paste. Finally, the lead paste (1.0 g) was coated on a Pb-Sn-Ca grid with a geometric area of 1.0×1.5 cm2 to prepare the negative plates. All as-prepared negative plates were cured under carefully controlled conditions, and then these cured negative plates were formed in H2SO4 (1.05 g cm−3) solution with the current of 8.5 mA·g−1 and 17 mA·g−1 under a commercial formation procedure. After formation, the negative plates were washed with deionized water to remove the formation acid, and then they were used for electrochemical tests or cycle life tests.

2.4. Electrochemical measurements All electrochemical experiments in this work were performed using a CS350 electrochemical workstation (Wuhan Corrtest, China) with a conventional three-electrode system in a single compartment cell at room temperature (~ 25 ). The prepared negative plate with a geometric area of 1.0×1.5 cm2 was used as the working electrode. Two platinum electrodes with relatively large area were used as the counter electrode and a Hg/Hg2SO4 electrode as the reference electrode. The electrolyte was H2SO4 with a density of 1.28 g cm−3. The cathodic polarization curves were measured at a scan rate of 1.0 mV s−1, starting from the open circuit potential (EOCP) to −1.5 V vs. Hg/Hg2SO4. The EIS was measured at EOCP. The amplitude of the AC perturbation was 10 mV and the frequency range was 10 kHz-0.005 Hz. CV measurements were performed in the potential range from −1.5 V to 0.4 V vs. Hg/Hg2SO4 at a scan rate of 10 mV s−1, starting from 0.4 V vs. Hg/Hg2SO4, and 5 cycles were recorded. The specific capacitance (C, F·g−1) was calculated from the CV curves of the fifth cycle according to Eq. (1) [38]: V2

C



V1

i (V )dV

mv(V2  V1 )]

(1)

where i (V) is the scanning current function (A), V1 and V2 are the low limit potential and upper limit potential (V), i.e., −1.5 V and 0.4 V, v is the scanning rate (V s−1), and m (g) is the mass of the active material. All the above electrochemical tests were repeated at least three times to ensure the reliability of the results.

2.5. Preparation of simulated test cells and HRPSoC cycle tests A simulated flooded lead-acid test cell was assembled comprising a prepared negative plate and two commercial positive plates with a geometric area of 2.8×2.8 cm2 separated by a 2 mm thick glass mat (AGM) separator. Then, the simulated test cell was flooded with 65 mL H2SO4 (1.28 g cm−3). The capacity of the simulated test cell was determined under different currents so as to obtain the currents of 1C and 2C from the Peukert plot, then the cell was fully charged at C10 rate and discharged to 50% state-of-charge (SoC) at 1C rate before the HRPSoC cycle test. The charge/discharge cycle processes were carried out according to the following schedule: charge at 1C or 2C rate (relative to the charge-discharge rate of the blank cell) for 30 s, rest for 10 s, discharge at 1C or 2C rate for 30 s, rest for 10 s. The cell voltage was measured at the end of each charge and discharge processes, i.e. end-of-charge voltage (Vcharge) and end-of-discharge voltage (Vdischarge), and the test was stopped when the Vdischarge fell

4

down to 1.70 V (Vdischarge limit) or the Vcharge rose up to 2.90 V (Vcharge limit). All the tests were conducted at room temperature (~ 25 ) and repeated at least three times to ensure the reliability of the results.

2.6. Physical characterizations The morphology of various additives and the negative plates containing different additives before and after the HRPSoC cycle test were analysed using SEM (JSM-7600F, Japan). The PPy/GO composites were analysed using a Vertex 70 (Germany) FTIR spectrophotometer in reflectance mode. The FTIR spectra were collected with a resolution of 4 cm−1, and 64 scans were averaged for each sample. The specific surface and porosity analysis for the various additives and the negative plates were performed using a specific surface area and pore analyzer (ASAP2020M, America). XPS analysis of GO was performed using a VG Multilab 2000 system. The base pressure in the experimental chamber was below 10−9 mbar. The spectra were measured with Al Ka (1486.6 eV) radiation and the overall energy resolution was 0.45 eV. The binding energies were calibrated relative to the C 1s peak from GO sample at 284.6 eV.

3. Results

3.1. Characterization of PPy, GO and PPy/GO composites Fig. 1 presents XPS survey spectra and high-resolution spectra of C 1s for a GO additive sample. It can be seen that the distinct C and O signals appear at the binding energy of 286 eV and 532 eV, respectively, as shown in Fig. 1a. The high-resolution spectrum of C 1s can be deconvoluted into three single peaks, as shown in Fig. 1b, corresponding to the C–C bond (284.6 eV) of sp2 carbon in the basal plan of GO sheets, C–O (286.5 eV) and C=O (288.7 eV) resulting from carboxylic groups [39]. It can be observed that peaks related to the oxidized carbon species including C–O and C=O are predominant. Table 1 gives the relative contents of functional groups obtained by quantitative analysis of C 1s spectra. It is seen that the content of C–O groups (32.73%) and C=O groups (26.90%) is 59.63%, which is larger than that of C–C groups (40.37%). Fig. 2 shows the FTIR spectra of PPy, GO and PPy/GO composites. In the FTIR spectrum of GO, the peaks at 1720.96, 1382.10 and 1059.8 cm−1 are assigned to the carbonyl (C=O) stretching, the O–H deformation and C–O stretching vibrations, respectively [40]. Based on the XPS and FTIR analysis, it can be proved that GO species has been successfully synthesized according to the reported modified Hummer’s method. Peaks at 1542.65, 1459.65 cm−1 in the PPy spectrum are attributed to the pyrrole ring skeletal vibrations and the peaks at 1168.59 cm−1 and 900.58 cm−1 are associated with the C–N stretching vibration and C-H out-plane deformation vibration, respectively [41]. The FTIR spectra of PPy/GO composites display that the band corresponding to C=O stretching vibration decreases and shows red shift compared to that of the GO, suggesting that the hydrogen bond is formed between the carboxyl groups of GO and the –NH groups of PPy [42]. In addition, the peaks at 1542.65 cm−1 and 900.58 cm−1 show red shift at low GO content but change to blue shift at high GO content. In general, the FTIR spectra of PPy/GO composites are more 5

similar to that of PPy. These results indicate that there is interaction between PPy and GO in the PPy/GO composites. The SEM morphology of PPy, GO and PPy/GO composites are presented in Fig. 3, in which Figs. 3a-3b show typical surface morphology of pure GO and PPy, respectively, while Figs. 3c-3e give the morphology of PPy/GO composites with different mpy/mGO ratios. From Figs. 3c-3e, it is seen that with an increase in the mpy/mGO ratio the surface of GO is gradually covered by spherical PPy particles, suggesting that PPy grows on the surface of GO sheets. Table 2 gives the specific surface and pore analysis results for the powders of the lead oxide, PPy, GO and PPy/GO composites. Compared with the lead oxide, the additive powders have much higher BET surface area, larger total pore volume and similar average pore width (except GO), indicating that they have more porous microstructure. Among these additives, GO has the largest BET surface area, the smallest average pore width and medium pore volume, which demonstrates that GO contains much more number of pores in smaller size. PPy also has a porous microstructure, but compared with GO, it has much larger average pore width and slightly smaller pore volume resulting in much less BET surface area. For the three PPy/GO composites, their average pore width is about 1.98-2.56 times larger than that of GO but close to that of PPy and increases slightly with an increase in the mpy/mGO ratio, suggesting that their pore size is similar to PPy. In addition, the mpy/mGO ratio has significant influence on their BET surface area and pore volume. With a lower PPy content, such as PG0.5 and PG1, their pore volume increases with the mpy/mGO ratio and becomes 1-2 times larger than that of PPy, resulting in relatively large BET surface area. With a higher PPy content, such as PG20, its pore volume and BET surface area reduces significantly and its microstructure changes to be similar to PPy. Among the three PPy/GO composites, PG1 has the largest BET surface area and pore volume, suggesting that a medium PPy content in the PPy/GO composites is suitable. The change of the microstructure of the PPy/GO composites with the mpy/mGO ratio may be related to the interaction between PPy and GO. In fact, apart from the hydrogen bond between the carboxyl groups of GO and the –NH groups of PPy, GO can also be doped in PPy chains [30]. In this case, both of their microstructures are changed.

3.2. Electrochemical performance of the negative plates containing different additives Fig. 4 shows the typical cathodic polarisation curves in H2SO4 solution (1.28 g −3 cm ) for the negative plates containing different additives, i.e., blank (NAM), NAM+GO, NAM+PPy, NAM+PG0.5, NAM+PG1, and NAM+PG20. These curves all show a similar diffusion-limited current before H2 evolution which should be related to the reduction of PbSO4, suggesting that the reduction of PbSO4 may be controlled by the diffusion process of Pb2+ ions or the dissolution of PbSO4 crystals. Then, because of the occurrence of H2 evolution the polarisation current density (I) rises again, where the potential corresponding to the start of the current rise is determined as the H2 evolution potential (EH) and the I at EH ( I EH ) can be approximately 6

considered as the current density resulting from the reduction of PbSO4, as shown in Fig. 4. When H2 evolution occurs, the polarisation current density at a potential E (IE) is the combination of the H2 evolution current density (IH, E) and the PbSO4 reduction current density I EH . So there is IH, E=IE  I EH . In order to quantitatively compare the H2 evolution ability of the negative plates containing different additives, the EH, and IH at −1.4 V vs. Hg/Hg2SO4 (IH, −1.4V), which is a medium potential for the negative plates during a charging process, are obtained from the cathodic polarisation curves, as shown in Fig. 4. Table 3 lists the average values of EH, I EH as well as IH, −1.4V ( EH , I E H , I H,-1.4V ) and their standard deviations. Compared with the EH of the blank plate ( E H

NAM

, −1.226 V vs.

Hg/Hg2SO4), the EH of the negative plates containing different additives ( EH does not display an apparent change tendency, in which EH

EH

PG1

decrease slightly (−32 mV~ −21 mV), while EH

PG 0.5

PPy

, EH

GO

and EH

PG 20

additive

)

and increase

slightly (6 mV~ 17 mV). Taking into account the errors from the EH measuring method, it may be considered that the EH of the negative plates is not altered obviously with the addition of the different additives. However, the IH of the negative plates shows quite different changes. Compared with the blank plate, the NAM+GO plate has larger I E H and I H,-1.4V , while the NAM+PPy plate has smaller values. These results suggest that adding GO in the NAM accelerates the reduction of PbSO4 and H2 evolution processes in the negative plate, while adding PPy plays an opposite role. When adding PPy/GO composites (PG0.5, PG1 and PG20) in the NAM, the values of I E H and I H,-1.4V decrease with an increase in the PPy content of PPy/GO composites, which may be due to the inhibition effect of PPy. But among the studied samples, the NAM+PG0.5 plate has the largest I E H and I H,-1.4V (larger than those of the NAM+GO plate), while the NAM+PG20 plate has the smallest I E H (less than that of the NAM+PPy plate). Apparently, these results cannot be explained based on the action of PPy and GO in the negative plate and will be discussed below. Fig. 5 shows the typical Nyquist diagrams measured at EOCP in H2SO4 solution (1.28 g cm−3) for the negative plates containing different additives. At EOCP, the additives and Pb particles in the negative plate will form galvanic cells, in which the additives act as cathodes to produce H2 evolution and the Pb particles as anodes to dissolve. In this case, the EIS results show the features of these two processes. All the Nyquist plots in Fig. 5 generally exhibit two depressed capacitive loops, and the order of the total system impedance of the negative plate (Rplate) is: RNAM+GO < RNAM+PG0.5 < 7

RNAM+PG1 < RNAM+PPy < RNAM+PG20 < RNAM. Apparently, Rplate decreases with the addition of the additives. Among these additives, GO has the most obvious effect on Rplate, while PPy in the PPy/GO composites generally weakens the effect of GO and makes Rplate increase. Fig. 6 shows the typical CVs in H2SO4 (1.28 g cm−3) for the negative plates containing different additives. In order to observe the stability of the various additives, the CV measurements were performed in a wider potential range (−1.5 V ~ 0.4 V vs. Hg/Hg2SO4). It is seen that there are two pairs of redox peaks in all the CVs, where the first corresponds to the redox processes between Pb and PbSO4 (Pb  PbSO4) and the second corresponds to those between Pb and PbO·PbSO4, as shown in Fig. 6 [13, 43]. These results indicate that in the negative plate the NAM is still only the Pb and the redox processes between Pb and PbSO4 are the main processes. From the redox peaks (Pb  PbSO4) in Fig. 6, it can be seen that the oxidation peak currents and the reduction peak currents all increase with the addition of the additives and their numerical order is: NAM+GO > NAM+PG0.5 ~ NAM+PG1 > NAM+PG20 > NAM+PPy > NAM, suggesting that the redox processes (Pb  PbSO4) are accelerated with the addition of the additives. Among the additives, GO still shows the largest acceleration effect, while PPy in the PPy/GO composites weakens the effect of GO. In order to observe the capacitance performance of the negative plates containing different additives, their specific capacitance values (Cadditives) are calculated from their CVs as shown in Fig. 6 using Eq. (1). The average value of Cadditives ( C additives ) and its standard deviation are listed in Table 4. It is seen that the addition of the additives increases the plate capacitance. The C NAM + GO is the largest (42.23 F·g−1), which increases about 2.53 times compared with the C NAM (16.72 F·g−1), while the

C NAM+ PPy (20.26 F·g−1) just increases slightly. For the plates containing PPy/GO composites, their specific capacitance values decrease obviously with an increase in the PPy content. The additives show the same influence on the plate capacitance as that on the redox peak currents (Pb  PbSO4). It should be noted that these electrochemical performance of the negative plates may change during the charging and discharging process. However, through these results we still find the effect of these additives on the electrochemical processes of the negative plate.

3.3. HRPSoC cycle performance of the simulated test cells Fig. 7 shows the typical change of Vdischarge as a function of cycle number under HRPSoC conditions for the simulated test cells with the negative plates containing different additives at discharge rates 1C and 2C within the first cycle-set, in which the cycle number when Vdischarge falls down to 1.70 V (i.e. Vdischarge limit) is defined as the HRPSoC cycle life of the test cell (CLadditive). The average value of CLadditive 8

( CL additive ) and its standard deviation are listed in Table 5. It is seen that the CL additive value of each cell at 1C rate is longer than that at 2C rate, which may be due to the different DoD at each cycle [44], but they show the same change with the addition of the additives. The CL NAM + GO values are the largest and increase about 7.35-7.56 times than the CL NAM values, while the addition of PPy just slightly increases the CL NAM+ PPy values under the same condition. Among the test cells containing various PPy/GO composites, the CL NAM + PG1 values are the largest, while the CL NAM + PG20 values are the smallest that are close to the CL NAM+ PPy values. Apparently, a medium

mpy/mGO ratio of the PPy/GO composites, such as PG1, is good for the HRPSoC cycle life of the test cell. Fig. 8 presents the typical change of Vcharge as a function of cycle number under HRPSoC conditions for the simulated test cells with the negative plates containing different additives at 2C rate within the first cycle-set. Because the same oversized positive plates were used in the test cell, their polarisation should be much less than that of the negative plate and would be approximate in different test cells during the cycle test. In this case, the change in Vcharge generally reflects the difference of the cathodic polarisation for the negative plates containing different additives. The Vcharge of the blank test cell increases quickly from 2.17 V to 2.76 V within the first 300 cycles, and the test cells containing NAM+PPy and NAM+PG20 have the similar behavior within the first 400 cycles. The Vcharge of the test cells containing NAM+GO, NAM+PG1 and NAM+PG0.5 increases much more slowly, and their cathodic polarisation degree decrease by 540mV, 511mV and 496 mV, respectively, within the first 300 cycles compared with that of the blank test cell, suggesting that the cathodic polarisation of these negative plates are significantly reduced.

3.4. Microstructure and crystal morphology of the negative plates before and after the HRPSoC cycle test The negative plates containing different additives (NAM, NAM+GO, NAM+PPy and NAM+PG1) before and after the HRPSoC cycle test at 2C rate were cut from the middle, and their cross sections were analysed using SEM to observe the microstructure and crystal morphology in the interior of the negative plate, as shown in Fig. 9. According to the HRPSoC cycle performance of the simulated test cells, PG1 is the proper additive among the three PPy/GO composites, so we just selected the NAM+PG1 plate for the SEM analysis. The SEM images of the negative plates before the cycle test in Figs. 9 (AI-DI) show that with the addition of different additives (GO, PPy and PG1) the plate microstructure seems to become more porous and shows different porosity, while Pb particles also display different shape and distributions. These results may suggest that 9

GO, PPy and PG1 have different distributions in the NAM and Pb particles may grow on their surface. Table 6 gives the specific surface and pore analysis results for these negative plates before the cycle test. It is seen that when adding GO, PPy and PG1 in the negative plates, their BET surface area increases about 2.50-3.00 times compared with that of the blank plate, while their average pore widths decrease obviously and the total pore volumes increase slightly, indicating that these plates contain more number of pores in smaller size, which further proves that the plate microstructure has been changed with the addition of the additives in NAM. These results are similar to the effect of carbon black in nanosizes on the structure of the NAM [13]. According to the Refs. [13, 44], the additives may be absorbed on the surface of NAM or be incorporated into the bulk of the skeleton branches of the NAM depending on their contents in NAM. However, GO, PPy and PPy/GO composites are materials without fixed shape, which is quite different from those materials with carbon particles, see Fig. 3 and Table 2. According to the results in Table 2 and Table 6, it would be presumed that during the preparation of the negative paste, Pb aggregates would be formed and the additives (GO, PPy and PPy/GO composites) would be attached on their surfaces or distributed between them to prevent them from forming large aggregates. After formation, small Pb particles grow on the surface of the additives and Pb aggregates, and then the porous microstructure are maintained, as shown in Figs. 9 (BI-DI). The difference of the microstructure among the plates containing different additives may be due to the different features of various additives and their different interactions with the Pb particles. After the cycle test at 2C rate, the interior of the blank plate and the plate containing PPy is full of large PbSO4 crystals with a compact microstructure, as shown in Figs. 9 (AII and CII), while those containing GO and PG1 generally have much smaller PbSO4 crystals and more porous microstructure, as shown in Figs. 9 (BII and DII). Apparently, the addition of GO and PG1 impedes the growth of PbSO4 crystals in the negative plate during the HRPSoC cycle test.

4. Discussion

4.1 Effect of the additives on the H2 evolution process in the negative plate With the addition of various additives (GO, PPy and PPy/GO composites) in the NAM, a part of surface area of Pb particles (SPb) will be occupied by the additives and the microstructure of the NAM also be changed, as shown in Fig. 9 and Table 6. At meantime, the surface of the additives has electrochemical activity and also supplies reaction area (Sadditive) for the H2 evolution process. The electrochemical activity of the various additives may be quite different from each other and different from the NAM, so the kinetics of the H2 evolution process in the negative plates may be changed with the addition of different additives. Therefore, in the same H2SO4 solution, the H2 evolution rate is mainly determined by the features of the reaction surface and the real reaction area (Sreal). The BET surface area of the NAM+GO plate is about 3 times that of the blank (NAM) plate, see Table 6, while its I H,-1.4V (22.60 mA cm−2) increases 1.75 times compared 10

with that of the blank plate (12.95 mA cm−2), see Table 3. Because the Sreal (= SGO + SPb) may not be equal to the BET surface area and the ratio of SGO / SPb is also not clear, it is difficult to clearly judge whether the H2 evolution process on the GO surface is accelerated or not. But according to the above results, it could be speculated that the acceleration of the H2 evolution in the NAM+GO plate may be mainly due to the increase of the Sreal. In contrast, the BET surface area of the NAM+PPy plate is about 1.50 times that of the blank plate, see Table 6, but its I H,-1.4V decreases to 4.47 mA cm−2. Apparently, the H2 evolution process on the PPy surface is depressed significantly. This can be ascribed to the low electrochemical activity of the PPy in a deep cathodic polarisation state [45]. The effect of the PPy/GO composites on the H2 evolution process should be determined by the comprehensive action of PPy and GO depending on the mpy/mGO ration of the PPy/GO composites, such as the NAM+PG1 and NAM+PG20 plates. However, the NAM+PG0.5 plate has larger H2 evolution rate than that of the NAM+GO plate, which does not reflect the inhibition effect of PPy. This would be due to the increase of the Sreal of the NAM+PG0.5 plate which is related to its microstructure. Apparently, too lower mpy/mGO ratio is not adequate for the inhibition of the H2 evolution process. It should be noted that the above discussion is based on the steady-state polarisation behavior of the H2 evolution process. Under the HRPSoC conditions, the transient process is important and may affect the H2 evolution performance of the negative plate.

4.2 Effect of the additives on the HRPSoC cycle performance of the simulated test cells Generally, the anodic process on the negative plate can be described as following reactions: Pb − 2e → Pb2+ (1) Pb2+ + HSO4− → PbSO4 + H+ (2) In the cathodic process of the negative plate, following reactions will occur: PbSO4 + H+ → Pb2+ + HSO4- (3) Pb2+ + 2e → Pb (4) 2H+ + 2e → H2 (5) Under HRPSoC cycle conditions, the negative plate will suffer short charging/discharging time with high currents, resulting in the insufficient utilization of the NAM (Pb) and the discharged product (PbSO4) [5]. Because the PbSO4 crystals cannot be effectively reduced to Pb within the short charging process (reaction 3-4), the recrystallization process of PbSO4 crystals occur and make them gradually grow to big size crystals that are difficult to dissolve (irreversible PbSO4), which eventually leads to the sulfation of the negative plates [5]. The characteristics of the CV diagrams in Fig. 6 indicate that the reduction rate of PbSO4 is obviously smaller than that of the oxidation of Pb, suggesting that the accumulation of PbSO4 crystals is inevitable. In this case, the growth rate of PbSO4 crystals in the negative plate will determine the HRPSoC cycle life of the simulated test cell. 11

From Fig. 8, it can be seen that the cathodic polarisation degree of the negative plates containing GO and PG1 increased slowly during the HRPSoC cycle test. It is generally considered that large PbSO4 crystals in the negative plate lead to the quick increase in its cathodic polarisation degree during charging process [13]. So the lower increase of the cathodic polarisation degree means the lower growth rate of PbSO4 crystals in these negative plates. It is also seen that smaller PbSO4 crystals distributed in the interior of the negative plates containing GO and PG1 after the HRPSoC cycle test in Figs. 9 (AII-DII). These results verify that the addition of the GO and PG1 in the NAM indeed suppresses the growth rate of PbSO4 crystals in the negative plates and then significantly prolongs the HRPSoC cycle life of the simulated test cells, as shown in Fig. 7. The effect of the additives on the growth of PbSO4 crystals may be related with the change of the microstructure of the negative plates. The addition of PPy, GO and PPy/GO composites in the negative plate will increase its BET surface area and total pore volume and decrease its average pore width, as shown in Table 6. The increase in the BET surface area and total pore volume will largely increase the Sreal of the negative plate so as to provide extra nucleation sites for PbSO4 crystals to facilitate the formation of the small-sized PbSO4 crystals with higher solubility (reversible PbSO4 crystals) [46]. In this case, the reaction (3) is promoted so that the reduction of PbSO4 crystals is accelerated. The obviously increased reduction peak currents with the addition of the additives, as shown in Fig. 6, can prove this acceleration effect. In addition, the increase in the Sreal will also result in the increase of the specific capacitance of the negative plate, as shown in Table 4. It is generally considered that the negative plate containing a mixture of carbon black and activated carbon can act as a buffer to share discharge and charge currents so as to prevent it from discharging and charging at high rate [46]. In this case, the negative plates containing GO and PG1, which have larger specific capacitance, will have relatively stronger buffer function for large charging/discharging currents so as to prevent the formation of irreversible PbSO4 crystals. It is reported that the decrease of the average pore width of the negative plate (<1 μm) may result in the formation of PbO [13], but its influence on PbSO4 crystals is not clear and needs further investigations. According to the HRPSoC cycle performance of the simulated test cells, the GO has the best performance among the studied additives, and the incorporation of PPy with GO to form PPy/GO composites will damage the HRPSoC cycle performance of the test cells. However, it should be noted that the simulated test cells are flooded during the HRPSoC cycle test, so the influence of the H2 evolution performance of the negative plate may not be reflected. Considering the effects of the H2 evolution performance of the negative plate, the PPy/GO composites with a medium mpy/mGO ratio, such as PG1, may be the appropriate additives for the negative plate of lead-acid batteries.

5. Conclusions In order to improve the HRPSoC performance of lead-acid batteries for hybrid-electric vehicles, the additives (GO, PPy and PPy/GO composites) were added in the negative plate to form simulated test cells. The effects of the additives on the 12

electrochemical performance of the negative plate and on the HRPSoC cycle performance of the simulated test cell were investigated by various electrochemical methods. Their influences on the microstructure and crystal morphology of the negative plate were also analysed using SEM and specific surface area and pore analyzer. The conclusions are drawn from the results of experiments: (1) Among the studied additives, GO has the largest BET surface area, the smallest average pore width and medium pore volume, while PPy has much less BET surface area, slightly smaller pore volume and much larger average pore width compared with GO. In the PPy/GO composites, PPy covers on the surface of GO and their microstructure changes with the mpy/mGO ratio, which may be due to the interaction between PPy and GO. Among the three PPy/GO composites (PG0.5, PG1 and PG20), the PG1 has the largest BET surface area and pore volume, suggesting that a medium PPy content in the PPy/GO composites is suitable. (2) GO in the negative plate largely increases the H2 evolution ability of the plate, while PPy has the opposite effect. When adding the PPy/GO composites (PG0.5, PG1 and PG20), the H2 evolution of the negative plate decreases with an increase in the mpy/mGO ratio. The incorporation of a certain amount of PPy with GO significantly inhibits the H2 evolution of the negative plate. Moreover, the addition of the additives in the negative plate also decreases its total impedance, accelerates the redox processes (Pb  PbSO4) on it and increases its specific capacitance. (3) The HRPSoC cycle life of the simulated test cells containing different additives are all prolonged compared with that of the blank cell. The cells containing NAM+GO display the longest cycle life, which increases about 7.35 time at 2C rate than that of the blank cells, while those containing NAM+PG1 have the second-longest HRPSoC cycle life. Through the change of Vcharge with the cycle number, it can be found that the cathodic polarisation degree of the negative plates containing GO and PG1 increases slowly during the HRPSoC cycle test. (4) With the addition of PPy, GO and PPy/GO composites, the microstructure of the negative plate will be changed by increasing its BET surface area and total pore volume and decreasing its average pore width. The SEM images of the negative plates after the HRPSoC cycle tests show that GO and PG1 impede the growth of PbSO4 crystals in the negative plate. (5) The influence of the H2 evolution ability of the negative plate on the HRPSoC cycle life may not be displayed in the flooded test cells. Considering the H2 evolution performance and the HRPSoC cycle performance, the PPy/GO composites with a medium mpy/mGO ratio, such as PG1, may be the appropriate additives for the negative plate of lead-acid batteries. Acknowledgements The authors thank for the financial support of this work by the special Funds for basic research of national universities of Huazhong University of Science and Technology (HUST) under Grant No. 2015TS147 and appreciate the analysis support of state key Laboratory of material Processing and Die & Mould Technology and Center of Forecasting and Analysis, Huazhong University of Science and Technology 13

as well as the Key Laboratory for Large-Format Battery Materials and System, Ministry of Education. References [1] A. Cooper, J. Furakawa, L. Lam, M. Kellaway,;1; The UltraBattery-A new battery design for a new beginning in hybrid electric vehicle energy storage, J. Power Sources 188 (2009) 642-649. [2] M. Conte, G. Pede, V. Sglavo, D. Macerata,;1; Assessment of high power HEV lead-acid battery advancements by comparative benchmarking with a European test procedure, J. Power Sources 116 (2003) 118-127. [3] P. Moseley, B. Bonnet, A. Cooper, M. Kellaway,;1; Lead–acid battery chemistry adapted for hybrid electric vehicle duty, J. Power Sources 174 (2007) 49-53. [4] K. Nakamura, M. Shiomi, K. Takahashi, M. Tsubota,;1; Failure modes of valve-regulated lead/acid batteries, J. Power Sources 59 (1996) 153-157. [5] L. Lam, N. Haigh, C. Phyland, A. Urban,;1; Failure mode of valve-regulated lead-acid batteries under high-rate partial-state-of-charge operation, J. Power Sources 133 (2004) 126-134. [6] L. Lam, N. Haigh, C. Phyland, T. Huynh,;1; Novel technique to ensure battery reliability in 42-V PowerNets for new-generation automobiles, J. Power Sources 144 (2005) 552-559. [7] J. Furukawa, T. Takada, D. Monma, L. Lam,;1; Further demonstration of the VRLA-type UltraBattery under medium-HEV duty and development of the flooded-type UltraBattery for micro-HEV applications, J. Power Sources 195 (2010) 1241-1245. [8] J. Furukawa, T. Mangahara, L. Lam,;1; Development of the UltraBattery for Micro-and Medium-HEV Applications, in: Meeting Abstracts, The Electrochemical Society, 2008, pp. 654-654. [9] L. Lam, R. Louey, N. Haigh, O. Lim, D. Vella, C. Phyland, L. Vu, J. Furukawa, T. Takada, D. Monma,;1; VRLA Ultrabattery for high-rate partial-state-of-charge operation, J. Power Sources 174 (2007) 16-29. [10] M. Shiomi, T. Funato, K. Nakamura, K. Takahashi, M. Tsubota,;1; Effects of carbon in negative plates on cycle-life performance of valve-regulated lead/acid batteries, J. Power Sources 64 (1997) 147-152. [11] S.W. Swogger, P. Everill, D. Dubey, N. Sugumaran,;1; Discrete carbon nanotubes increase lead acid battery charge acceptance and performance, J. Power Sources 261 (2014) 55-63. [12] S. Logeshkumar, R. Manoharan,;1; Influence of some nanostructured materials additives on the performance of lead acid battery negative electrodes, Electrochim. Acta 144 (2014) 147-153. [13] D. Pavlov, T. Rogachev, P. Nikolov, G. Petkova,;1; Mechanism of action of electrochemically active carbons on the processes that take place at the negative plates of lead-acid batteries, J.

Power Sources 191 (2009) 58-75. [14] M. Fernández, F. Trinidad, J. Valenciano, A. Sánchez,;1; Optimization of the cycle life performance of VRLA batteries, working under high rate, partial state of charge (HRPSOC) conditions, J.

Power Sources 158 (2006) 1149-1165. [15] D. Boden, D. Loosemore, M. Spence, T. Wojcinski,;1; Optimization studies of carbon additives to negative active material for the purpose of extending the life of VRLA batteries in high-rate partial-state-of-charge operation, J. Power Sources 195 (2010) 4470-4493. [16] L. Lam, R. Newnham, H. Ozgun, F. Fleming,;1; Advanced design of valve-regulated lead-acid battery for hybrid electric vehicles, J. Power Sources 88 (2000) 92-97.

14

[17] M. Fernandez, J. Valenciano, F. Trinidad, N. Muñoz,;1; The use of activated carbon and graphite for the development of lead-acid batteries for hybrid vehicle applications, J. Power Sources 195 (2010) 4458-4469. [18] B. Hong, X. Yu, L. Jiang, H. Xue, F. Liu, J. Li, Y. Liu,;1; Hydrogen evolution inhibition with diethylenetriamine modification of activated carbon for a lead-acid battery, RSC Adv 4 (2014) 33574-33577. [19] L. Zhao, B. Chen, J. Wu, D. Wang,;1; Study of electrochemically active carbon, Ga 2O3 and Bi2O3 as negative additives for valve-regulated lead-acid batteries working under high-rate, partial-state-of-charge conditions, J. Power Sources 248 (2014) 1-5. [20] L. Zhao, B. Chen, D. Wang,;1; Effects of electrochemically active carbon and indium (III) oxide in negative plates on cycle performance of valve-regulated lead-acid batteries during high-rate partial-state-of-charge operation, J. Power Sources 231 (2013) 34-38. [21] D. Pavlov, P. Nikolov,;1; Lead-carbon electrode with inhibitor of sulfation for lead-acid batteries operating in the HRPSoC duty, J. Electrochem. Soc 159 (2012) A1215-A1225. [22] G. Ning, T. Li, J. Yan, C. Xu, T. Wei, Z. Fan,;1; Three-dimensional hybrid materials of fish scale-like polyaniline nanosheet arrays on graphene oxide and carbon nanotube for high-performance ultracapacitors, Carbon 54 (2013) 241-248. [23] D. Zhang, X. Zhang, Y. Chen, P. Yu, C. Wang, Y. Ma,;1; Enhanced capacitance and rate capability of graphene/polypyrrole composite as electrode material for supercapacitors, J. Power Sources 196 (2011) 5990-5996. [24] H.-H. Chang, C.-K. Chang, Y.-C. Tsai, C.-S. Liao,;1; Electrochemically synthesized graphene/polypyrrole composites and their use in supercapacitor, Carbon 50 (2012) 2331-2336. [25] M. Mattioli-Belmonte, F. Gabbanelli, M. Marcaccio, F. Giantomassi, R. Tarsi, D. Natali, A. Callegari, F. Paolucci, G. Biagini,;1; Bio-characterisation of tosylate-doped polypyrrole films for biomedical applications, Mater. Sci. Eng., C 25 (2005) 43-49. [26] R. Sharma, A. Rastogi, S. Desu,;1; Pulse polymerized polypyrrole electrodes for high energy density electrochemical supercapacitor, Electrochem. Commun 10 (2008) 268-272. [27] D. Dubal, S. Patil, A. Jagadale, C. Lokhande,;1; Two step novel chemical synthesis of polypyrrole nanoplates for supercapacitor application, J. Alloys Compd 509 (2011) 8183-8188. [28] M.J. Allen, V.C. Tung, R.B. Kaner,;1; Honeycomb carbon: a review of graphene, Chem. Rev 110 (2009) 132-145. [29] J. Li, H. Xie, Y. Li,;1; Fabrication of graphene oxide/polypyrrole nanowire composite for high performance supercapacitor electrodes, J. Power Sources 241 (2013) 388-395. [30] K. Qi, Y. Qiu, X. Guo,;1; Pulse electrochemical incorporation of graphene oxide into polypyrrole films for supercapacitor electrode materials, Electrochim. Acta 137 (2014) 685-692. [31] S. Konwer, R. Boruah, S.K. Dolui,;1; Studies on conducting polypyrrole/graphene oxide composites as supercapacitor electrode, J. Electron. Mater 40 (2011) 2248-2255. [32] K. Qi, Y. Qiu, Z. Chen, X. Guo,;1; Corrosion of conductive polypyrrole: Effects of possibly formed galvanic cells, Corros. Sci 80 (2014) 318-330. [33] Y. Tian, F. Yang, W. Yang,;1; Redox behavior and stability of polypyrrole film in sulfuric acid,

Synth. Met 156 (2006) 1052-1056. [34] W.S. Hummers Jr, R.E. Offeman,;1; Preparation of graphitic oxide, J. Am. Chem. Soc 80 (1958) 1339-1339.

15

[35] K. Zhang, L.L. Zhang, X. Zhao, J. Wu,;1; Graphene/polyaniline nanofiber composites as supercapacitor electrodes, Chem. Mater 22 (2010) 1392-1401. [36] S. Bose, T. Kuila, M.E. Uddin, N.H. Kim, A.K. Lau, J.H. Lee,;1; In-situ synthesis and characterization of electrically conductive polypyrrole/graphene nanocomposites, Polymer 51 (2010) 5921-5928. [37] P. Wang, Y. Zheng, B. Li,;1; Preparation and electrochemical properties of polypyrrole/graphite oxide composites with various feed ratios of pyrrole to graphite oxide, Synth. Met 166 (2013) 33-39. [38] Y. Zhang, Y. Wang, S. Xu, J. Wang, Z. Wang, S. Wang,;1; Polypyrrole nanowire modified graphite (PPy/G) electrode used in capacitive deionization, Synth. Met 160 (2010) 1392-1396. [39] H.A. Becerril, J. Mao, Z. Liu, R.M. Stoltenberg, Z. Bao, Y. Chen,;1; Evaluation of solution-processed reduced graphene oxide films as transparent conductors, ACS nano 2 (2008) 463-470. [40] Y. Han, Y. Lu,;1; Characterization and electrical properties of conductive polymer/colloidal graphite oxide nanocomposites, Compos. Sci. Technol 69 (2009) 1231-1237. [41] J. Deng, Y. Peng, C. He, X. Long, P. Li, A.S. Chan,;1; Magnetic and conducting Fe3O4-polypyrrole nanoparticles with core-shell structure, Polym. Int 52 (2003) 1182-1187. [42] Z. Gu, C. Li, G. Wang, L. Zhang, X. Li, W. Wang, S. Jin,;1; Synthesis and characterization of polypyrrole/graphite oxide composite by in situ emulsion polymerization, J. Polym. Sci., Part

B: Polym. Phys 48 (2010) 1329-1335. [43] P. Ruetschi,;1; Ion Selectivity and Diffusion Potentials in Corrosion Layers Films on Pb in H2SO4, J. Electrochem. Soc 120 (1973) 331-336. [44] D. Pavlov, P. Nikolov,;1; Capacitive carbon and electrochemical lead electrode systems at the negative plates of lead-acid batteries and elementary processes on cycling, J. Power Sources 242 (2013) 380-399. [45] K. Qi, Y. Qiu, Z. Chen, X. Guo,;1; Corrosion of conductive polypyrrole: effects of continuous cathodic and anodic polarisation, Corros. Sci 69 (2013) 376-388. [46] J. Xiang, P. Ding, H. Zhang, X. Wu, J. Chen, Y. Yang,;1; Beneficial effects of activated carbon additives on the performance of negative lead-acid battery electrode for high-rate partial-state-of-charge operation, J. Power Sources 241 (2013) 150-158.

Fig. 1. (a) XPS survey spectra and (b) high-resolution spectra of C 1s for a GO additive sample.

Fig. 2. FTIR spectra of GO, PPy and PPy/GO composites.

Fig. 3. SEM images of GO, PPy, and PPy/GO composites: (a) GO; (b) PPy; (c) PG0.5; (d) PG1; (e) PG20.

Fig. 4. Cathodic polarization curves in H2SO4 solution for the negative plates containing different additives: (1) Blank (NAM); (2) NAM+GO; (3) NAM+PPy; (4) NAM+PG0.5; (5) NAM+PG1; (6) NAM+PG20 (H2SO4 density = 1.28 g cm−3, scan rate: 1 mV s−1).

Fig. 5. Nyquist diagrams measured at EOCP in H2SO4 solution for the negative plates containing different additives: Blank (NAM), NAM+GO, NAM+PPy, NAM+PG0.5, NAM+PG1, and NAM+PG20. (H2SO4 density = 1.28 g cm−3). 16

Fig. 6. CVs in H2SO4 (1.28 g cm−3) for the negative plates containing different additives: Blank (NAM), NAM+GO, NAM+PPy, NAM+PG0.5, NAM+PG1, and NAM+PG20 (scan rate: 10 mV s−1, the 5th cycle).

Fig. 7. Change of the Vdischarge as a function of cycle number under HRPSoC conditions for the simulated test cells with negative plates containing different additives (NAM, NAM+GO, NAM+PPy, NAM+PG0.5, NAM+PG1, and NAM+PG20) at discharge rates (a) 1C and (b) 2C within the first cycle-set.

Fig. 8. Change of Vcharge as a function of cycle number under HRPSoC conditions for the simulated test cells with negative plates containing different additives (Blank (NAM), NAM+GO, NAM+PPy, NAM+PG0.5, NAM+PG1, and NAM+PG20) at 2C discharge rate within the first cycle-set.

Fig. 9. SEM images of the negative plates before (I) and after the HRPSoC cycle test (II) at 2C rate containing different additives: (A) Blank (NAM); (B) NAM+GO; (C) NAM+PPy; (D) NAM+PG1 (The negative plates were fully recharged after the first cycle-set). Table 1 Relative contents of functional groups obtained by quantitative analysis of C 1s spectra Additive GO

C 1s / % C–C

C–O

C=O

40.37

32.73

26.90

Table 2 Specific surface and pore analysis results for the lead oxide powder and various additives. Pore volume Average pore width BET surface area Additives 2 −1 3 −1 （m g ) (cm g ) (nm) Lead oxide 1.91 0.0084 17.49 GO 49.78 0.062 4.98 PPy 13.25 0.051 15.52 PG0.5 29.32 0.11 14.84 PG1 35.98 0.15 16.79 PG20 9.07 0.040 17.73

Table 3 Comparison of the hydrogen evolution ability for the negative plates. Samples

EH (V, vs. Hg/Hg2SO4)

I EH (mA/cm2)

IH,-1.4V (mA/cm2)

Blank (NAM) NAM+GO NAM+PPy NAM+PG0.5 NAM+PG1

-1.226 ± 0.013 -1.247 ± 0.007 -1.250 ± 0.003 -1.220 ± 0.005 -1.258 ± 0.008

5.52 ± 0.44 8.00 ± 0.98 4.98 ± 0.49 13.57 ± 1.29 5.06 ± 0.71

12.95 ± 0.96 22.60 ± 0.85 4.47 ± 0.39 30.79 ± 1.59 8.72 ± 1.03

17

NAM+PG20

-1.209 ± 0.007

1.65 ± 0.45

7.52 ± 1.06

Table 4 Specific capacitance of the negative plates containing different additives (F·g−1). CNAM CNAM+GO CNAM+PPy CNAM+PG0.5 CNAM+PG1 CNAM+PG20 16.72 ± 42.23 ± 20.26 ± 27.17 ± 33.11 ± 2.67 18.08 ± 1.12 1.47 1.83 2.18 1.69

Table 5 HRPSoC cycle life of the test cells at discharge rates 1C and 2C within the first cycle-set. Discharge Blank NAM+GO NAM+PPy NAM+PG0.5 NAM+PG1 NAM+PG20 rate (NAM) 1C 2234 ± 16890 ± 11854 ± 2781 ± 232 9089 ± 363 2760 ± 211 191 368 397 2C 1986 ± 14593 ± 2144 ± 68 4987 ± 558 8905 ± 190 2105 ± 148 127 754

Table 6 Specific surface and pore analysis results for the negative plates before the cycle test. Negative BET surface area Pore volume Average pore width 2 −1 3 −1 plates （m g ) (cm g ) (nm) NAM 0.48 0.0031 260.15 NAM+GO 1.44 0.0059 164.30 NAM+PPy 1.20 0.0039 131.05 NAM+PG1 1.39 0.0051 185.81

TDENDOFDOCTD

18

×

Report "GO composites on the negative plate and on the high-rate partial-state-of-charge performance of lead-acid batteries"

Your name

Email

Reason -Select Reason- Pornographic Defamatory Illegal/Unlawful Spam Other Terms Of Service Violation File a copyright complaint

Description

Close Send

Our partners will collect data and use cookies for ad personalization and measurement. Learn how we and our ad partner Google, collect and use data. Agree & close