- Email: [email protected]
Enhanced performance of Zn(II)-doped lead-acid batteries with electrochemical active carbon in negative mass
Journal of Power Sources 328 (2016) 8e14
Contents lists available at ScienceDirect
Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour
Enhanced performance of Zn(II)-doped lead-acid batteries with electrochemical active carbon in negative mass Jiayuan Xiang a, *, Chen Hu b, **, Liying Chen a, Dong Zhang c, Ping Ding a, Dong Chen a, Hao Liu b, Jian Chen a, Xianzhang Wu a, Xiaokang Lai b a b c
Research Institute of Narada Power Source Co., Ltd, Hangzhou 311305, China China Electric Power Research Institute, Beijing, 100192, China Hangzhou City of Quality and Technical Supervision and Testing Institute, Hangzhou 310019, China
h i g h l i g h t s Gas evolution is reduced by 58e73% via adding Zn(II) additives. Low-temperature and high-rate capacity is improved by Zn(II) additives. Cycle life is prolonged under the co-effect of carbon and Zn(II) additives. The cell exhibits 90% reversible capacity after 2100 cycles at partial-state-of-charge duty.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 8 April 2016 Received in revised form 29 July 2016 Accepted 30 July 2016
The effect and mechanism of Zn(II) on improving the performances of lead-acid cell with electrochemical active carbon (EAC) in negative mass is investigated. The hydrogen evolution of the cell is significantly reduced due to the deposition of Zn on carbon surface and the increased porosity of negative mass. Zn(II) additives can also improve the low-temperature and high-rate capacities of the cell with EAC in negative mass, which ascribes to the formation of Zn on lead and carbon surface that constructs a conductive bridge among the active mass. Under the co-contribution of EAC and Zn(II), the partial-state-of-charge cycle life is greatly prolonged. EAC optimizes the NAM structure and porosity to enhance the charge acceptance and retard the lead sulfate accumulation. Zn(II) additive reduces the hydrogen evolution during charge process and improves the electric conductivity of the negative electrode. The cell with 0.6 wt% EAC and 0.006 wt% ZnO in negative mass exhibits 90% reversible capacity of the initial capacity after 2100 cycles. In contrast, the cell with 0.6 wt% EAC exhibits 84% reversible capacity after 2100 cycles and the control cell with no EAC and Zn(II) exhibits less than 80% reversible capacity after 1350 cycles. © 2016 Elsevier B.V. All rights reserved.
Keywords: Lead-acid battery Carbon Hydrogen evolution Partial-state-of-charge Zinc additive
1. Introduction In order to cope with the challenge of high-rate-partial-state-ofcharge (HRPSoC) duty in micro-hybrid electric vehicles and partialstate-of-charge (PSoC) duty in energy storage system for microgrid, new technology in lead-acid batteries (LABs) should be developed to provide higher levels of charge acceptance to enhance system efficiency and delay common failure mechanism such as lead sulfate accumulation on surface of negative plate [1].
* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (J. Xiang), [email protected] (C. Hu). http://dx.doi.org/10.1016/j.jpowsour.2016.07.113 0378-7753/© 2016 Elsevier B.V. All rights reserved.
Introducing carbon in types of activated carbon, carbon black and expanded graphite to negative active mass (NAM) during paste preparation has been proved to be an effective way to improve charge acceptance and prolong cycle life under PSoC duty [2e4]. The understanding of possible mechanism by which a carbon component enhance the performances of LABs has also been developed from eight different functions [5] to just three candidates that are most likely to have a significant individual effect [6]: (i) capacitive buffer to absorb charge current in excess of which can be accommodated by the Faradic reaction [7]; (ii) the extension of the surface area on which the electrochemical charge and discharge processes can take place [8]; (iii) the physical processes that hinder the crystallization of lead sulfate and thus help to maintain a high surface area for the discharge product [9].
J. Xiang et al. / Journal of Power Sources 328 (2016) 8e14
However, a risk of hydrogen evolution at the end of charge process maybe also brought in when adding carbon to LABs due to the low hydrogen evolution overpotential of carbon materials under the negative plate working conditions [10]. Attention needs to be paid to avoid harmful influences like more gas release, thermal runaway or even electrolyte dry-out. Introducing metal ions with high hydrogen evolution overpotential in LABs has been proved to be an effective way to retard hydrogen-evolution rate. For instance, In2O3, Bi2O3 and Ga2O3 were reported to be beneficial for suppressing hydrogen evolution and prolong HRPSoC cycle life of valve-regulated-lead-acid (VRLA) batteries with carbon [11,12]. But as is well known, residual metal element is always “harmful elements” for LABs since they may cause hydrogen and/or oxygen gassing and which, therefore, must be restricted or avoided. The influence of seventeen elements namely Sb, As, Bi, Gd, Gr, Co, Cu Ge, Fe, Mn, Ni, Se, Ag, Te, Tl, Sn and Zn in lead on hydrogen- and/or oxygen-gassing rates of LABs has been investigated [13], only Bi, Cd, Sn and Zn are considered as “beneficial elements” for VRLA batteries since they have little effect on gassing. The remaining metal elements are “harmful” since they will promote the gassing rate. Therefore, the study of doping “beneficial elements” in LABs with carbon in negative mass and their influence on battery performances is valuable and urgent. But to date, there is few research about it. In the present work, the focus lies on effect of Zn(II) at the form of ZnO doping in negative mass or ZnSO4 doping in electrolyte on hydrogen evolution and other performances of VRLA batteries with carbon. The purpose is to understand the function mechanism of high hydrogen-evolution-overpotential metal ions, provide a promising way to eliminate the water-consumption risk when adding carbon in negative mass, and improve the other performances of the LABs. 2. Experimental 2.1. Materials and characterization Carbon added to negative mass is provided by Cabot Corporation, with BET surface area of 1350 m2 g1 and particle size (D50) of 13 mm. Due to its high surface area and great contribution to the electrochemical reactions of charge, we labelled the carbon as electrochemical active carbon (EAC) as some previous literature named [8]. Leady oxide (80% PbO) is manufactured in Narada Company from 99.99% pure lead by ball milling. Vanisperse-A and BaSO4 is added in NAM as expander. ZnO and ZnSO4 is analytical reagent from Xilong Chemical Co., Ltd. Battery grade sulfuric acid (1.40 sp. gr.) is used during paste mixing. X-ray diffraction (XRD-7000, Shimadzu) was used to detect the phases of negative mass. Scanning Electron Microscope (SEM, Hitachi S-3400N) and BET Surface Analyzer (Gemini VII 2390, Micromeritics) were used to characterize the morphology and active surface area of carbon and negative mass. Particle Size Analyzer (Malvern-2000) was used to measure the particle size of carbon material. Mercury Intrusion Porosimetry (Autopore 9500, Micromeritics) was used to test the porosity of negative mass. Inductive Coupled Plasma-Atomic Emission Spectrometry (ICPAES, Optima 7000 DV) was used to measure the content of Zn element in negative mass and electrolyte. Gas chromatographmass spectrometer (GC-MS, Agilent 5977) is used to determine the hydrogen and oxygen ratio of the accumulated gas from the cells during 2.40Vpc charging. 2.2. Cell design and manufacturing Experimental groups of EAC and Zn(II) additives doped to
9
negative pastes are listed in Table 1. Based on the negative plate containing 0.6 wt% EAC, ZnO and ZnSO4 as Zn(II) additives were added in the cells, respectively. ZnO with doping level of 1.0 wt% of EAC was firstly mixed with EAC and then added in negative mass (0.006 wt% of NAM), and ZnSO4with doping level of 1 mg ml1 was directly added in electrolyte. For comparison, the cell containing no carbon and Zn(II) additives was also prepared. In order to describe clearly, the four groups are simply named as control, EAC, EAC þ ZnO, EAC þ ZnSO4. Each negative plate contains 0.05 wt% sodium lignosulphonate (Vanisperse-A) and 0.45 wt% BaSO4. 2 V prototype cell with 10h-rate capacity (C10) of 15 A h was fabricated and tested. The cell characteristic is shown in Table 2. 2.3. Electrochemical measurement Linear sweep voltammogram was performed on CHI660C electrochemical workstation to characterize the hydrogen evolution behavior of negative plate in sulfuric acid (1.30 sp. gr.) at scanning rate of 0.1 mV s1 from 1.0 V to 1.7 V versus Ag/Ag2SO4. Cyclic voltammogram was also performed on CHI660C electrochemical workstation to study the electrochemical behavior of EAC electrode in different electrolyte at scanning rate of 0.1 mV s1 from 0.6 V to 1.4 V versus Ag/Ag2SO4. Nyquist plots of the electrodes was measured on ModuLab XM electrochemical workstation from 10 kHz to 0.01 Hz with amplitude of 5 mV to investigate the conductivity of the electrodes. Discharge capacity was test at three conditions: (i) regular duty that discharge current was 1.5 A with cut-off voltage of 1.80 Vpc and ambient temperature was 25 C; (ii) low temperature duty that discharge current was 1.5 A with cut-off voltage of 1.80Vpc and ambient temperature was 10 C; (iii) high rate duty that discharge current was 3C (45 A) with cut-off voltage of 1.50 Vpc and ambient temperature was 25 C. Gas evolution property was evaluated under floating charge condition. After the cells were floating charged at 2.40 Vpc for 72 h, collected the gas generated during the following 192 h with also 2.40 Vpc floating charging. Charge acceptance was evaluated as follows: (i) discharged the cells with 1.5 A until the SoC level reached to 50%; (ii) rest for 24 h at temperature of 0 C; (iii) charged the cells with constant voltage of 2.40 V and no current limitation; (iv) recorded the current I10min as charged for 10 min, and calculated the ratio of I10min/I10 (I10 is the 10 h-rate current). PSoC cycling performance was evaluated using a simplified profile imitating the energy storage mode: (i) firstly discharged the cells from fully-charged-state to 30%SoC at 5 A; (ii) charged to 80% SoC at 2 A with 2.35 V limited; (iii) discharged to 30%SoC at 5 A; (iv) repeated step (ii) to step (iii) for 150 times; (v) fully recharged the cells at 2 A; (vi) determined the discharge capacity at 2 A; (vii) fully charged the cells and then went back to step (i). The cell voltage was measured at the end of discharge, the discharge capacity was recorded, and the test was stopped when the cell discharge voltage fell down to 1.80 V or the cell capacity fell below 8 Ah. 3. Results and discussion The XRD patterns of the formed negative mass are presented in Fig. 1a. The diffraction peaks of the control plate is mainly indexed to Pb (PDF No. 65e2873), only some minor peak indexed to a-PbO (PDF No. 85e1739) is detected. EAC doping in negative mass makes the XRD pattern change remarkably that the peaks of a-PbO become much stronger while the peak intensity of Pb decrease accordingly. It means the surface activity of spongy lead is enhanced with EAC, which might cause more lead oxidation reaction happen when the negative plate placed in air. As EAC added
10
J. Xiang et al. / Journal of Power Sources 328 (2016) 8e14
Table 1 Experimental groups of EAC and Zn(II) additives in cells. No.
Signature
Negative mass formulation
Electrolyte
1 2 3 4
control EAC EAC-ZnO EAC-ZnSO4
e EAC 0.6 wt% EAC 0.6 wt% þ ZnO 0.006 wt% EAC 0.6 wt%
H2SO4 H2SO4 H2SO4 H2SO4
(1.28 (1.28 (1.28 (1.28
sp. sp. sp. sp.
gr.) gr.) gr.) gr.)þ ZnSO4 (1 mg ml1)
Table 2 Characteristics of 2 V prototype cell (C10 ¼ 15 A h). Parameter
Positive
Negative
Grid alloy Gird size (mm mm) Plate thickness (mm) Plate number AGM compression ratio
Pb-Ca-Sn 70 80 4.49 4 70%
Pb-Ca 70 80 2.57 5
together with Zn(II) additive, the peak intensity of a-PbO is a little weaker than that of EAC plate without Zn(II), which indicates the effect of EAC on surface activity of spongy lead is reduced by Zn(II) doping. The semi quantitative calculation by the XRD-7000 analysis software shows the content of a-PbO phase of the control, EAC and EAC-ZnO plates are 5.25%, 12.20% and 11.12%, respectively. The SEM images of the formed negative mass in Fig. 1bed shows similar porous and spongy morphology. The BET surface area of NAM increases significantly from 1.41 m2 g1 to 5.54 m2 g1 and 6.22 m2 g1 after doping EAC and EAC þ ZnO. The porosity of the EAC þ ZnO electrode is also improved from 54.4% to 55.6% (Table 3). Generally, the theoretical BET surface of NAM including 0.6 wt% EAC should be 8.24 m2 g1 assuming all the surface of EAC (1350 m2 g1) is exposed and active in NAM. However, the actual BET surface of NAM is 5.54 m2 g1. The loss in surface area might be attributed to the absorption of sodium lignosulphonate on carbon [14]. In conventional lead-acid battery, sodium lignosulphonate as a typical organic expander are partially soluble in battery electrolyte and can adsorb on the negative plate, which increase the polarization of the positive plate to a value that allows the negative plate to become less polarized [15]. Therefore, it is understandable that sodium lignosulphonate is firstly dissolved and then adsorbs on the surface of PbO as well as carbon during the pasting process of negative mass with carbon, making some active surface of EAC is not exposed any more (Fig. 2). When EAC doping together with ZnO, Zn(II) will also combine with sodium lignosulphonate then decrease the absorption on the carbon surface, which results in the increasing BET surface area of negative mass from 5.54 m2 g1 to 6.22 m2 g1. The linear sweep voltammogram curves of different negative plates were displayed in Fig. 3a. The current density is always a key factor to indicate the hydrogen evolution rate at over-charge condition. As can be seen, the current density of the negative plate containing EAC is 0.150 mA cm2 at the potential of 1.40 V, much higher than that of the control plate (0.054 mA cm2). It is thus confirmed that carbon with high surface area doping in negative mass accelerates the hydrogen evolution. When EAC doping together with ZnO in negative mass or ZnSO4 in electrolyte, the electrode current density at 1.40 V significantly decreases to 0.077 mA cm2 and 0.103 mA cm2, respectively, which means Zn(II) is an effective additives to depress the hydrogen evolution. The gas generated from 2 V to 15 A h cells during 2.40 Vpc floating charging for 192 h is collected (according to section 2.3) to further verify the effect of Zn(II) on reducing hydrogen evolution. As shown in Fig. 3b, the total amount of gas generated from the control cell during 192 h charging is 0.30 ml h1 Ah1. When EAC
Fig. 1. XRD patterns and SEM images of the formed negative active mass: (a) XRD patterns; SEM images of (b) control; (c) EAC; (d) EAC þ ZnO.
J. Xiang et al. / Journal of Power Sources 328 (2016) 8e14 Table 3 BET surface and porosity of formed NAM for each test group. No.
Group
BET (m2 g1)
Porosity (%)
1 2 3
Control EAC EAC þ ZnO
1.41 5.54 6.22
54.4 54.1 55.6
11
0.83 ml h1 Ah1 and 0.54 ml h1 Ah1 during the 192 h charging process, which is reduced by 58% and 73%, respectively. The electrochemical reaction happened in negative electrode of LABs during charging is efficient transformation of lead sulfate to spongy lead (Eq. (1)).
PbSO4 þ Hþ þ 2e /Pb þ HSO4
(1)
Due to the high porosity of the negative mass, the charge acceptance of negative electrode is much better than the positive one, so the negative electrode is considered as over charged only until the state of charge is above 90%. At over-charge condition, Hþ in electrolyte combines with electron that makes hydrogen evolution (Eq. (2)). The reversible potential of this side reaction is about 350 mV higher than that of Pb/PbSO4. In a VRLA cell, the typical “oxygen cycle” would cause negative potential shift positively, and thus reduce or even eliminate the hydrogen generation [16]. When EAC is doped in negative mass, the hydrogen evolution rate will be no doubt accelerated since the reaction of Eq. (2) becomes much easier on carbon surface rather than on lead surface.
2Hþ þ 2e /H2 [
(2)
Fig. 2. BET surface change of NAM after doping EAC and ZnO.
doped in negative mass, the generated gas is 2.00 ml h1 Ah1, nearly 6e7 times more than the control one. But, the gas quantity decreases remarkably from the cells containing both EAC and Zn(II). The two cells, one containing EAC and ZnO in NAM, the other containing EAC in NAM and ZnSO4 in electrolyte, exhibit gas of
Fig. 3. (a) Linear sweep voltammogram curves of different negative plates; (b) gas evolutions of different cells under 2.40 Vpc floating charging.
Table 4 Composition of the accumulated gas in the last 24 h during the 192 h float-charging with 2.40 Vpc.
H2% O2%
Control
EAC
EAC þ ZnO
EAC þ ZnSO4
79.33 20.67
95.79 4.21
90.76 9.24
86.28 13.72
Fig. 4. Cyclic voltammogram curves of EAC electrode in (a) 1.28 sp. gr. H2SO4 and (b) 1.28 sp. gr. H2SO4 with 1 mg ml1 ZnSO4.
12
J. Xiang et al. / Journal of Power Sources 328 (2016) 8e14
Table 5 The quantity of Zn element in electrolyte and active mass after cell was fully charged. Group
Electrolyte (mg kg1)
NAM (mg kg1)
PAM (mg kg1)
EAC þ ZnO EAC þ ZnSO4
1.3 14.7
0.6 2.8
not detected not detected
The gas generated from the cells during 2.40 Vpc charging involves hydrogen as well as oxygen. GC-MS is used to determine the hydrogen and oxygen ratio of the accumulated gas as shown in Table 4. The hydrogen percent of the gas gathered from the control cell is 79.33%, while that of the EAC cell is as high as 95.79%. By doping ZnO and ZnSO4, the hydrogen percent reduced to 90.76% and 86.28%, respectively. According to the gas volume data in Fig. 4, the hydrogen volumes accumulated from the cells with ZnO and ZnSO4 is indeed reduced by comparing to the EAC cell. Similar to the Hþ/H2 reaction, when Zn(II) is added in negative mass or electrolyte, the dissolved Zn2þ will also combine with electron and become Zn during charging (Eq. (3)).
Zn2þ þ 2e /Zn
(3)
In order to investigate the mechanism of Zn(II) on electrochemical behavior of the cell containing EAC in NAM, especially in the over-charge process, the cyclic voltammogram curves of EAC electrode in bare H2SO4 (1.28 sp. gr.) and H2SO4 with 1 mg ml1 ZnSO4 are presented in Fig. 4. In bare H2SO4 electrolyte, EAC electrode exhibits a pair of cathodic and anodic peaks at potential of 0.95 to 1.00 V. As scanning below 1.20 V, hydrogen evolution begins. According to Bullock's research [17], the peaks at 0.95 to 1.00 V should correspond to the dissolve of carbon in aqueous solution to form carboxyl group or formic acid (Eq. (4)).
C þ 2H2 O/HCOOH ðaqueousÞ þ 2Hþ þ 2e
(4)
However, in the electrolyte containing Zn2þ, EAC electrode shows different CV curve that involves two cathodic peaks at potential of 0.92 to 0.98 V, which should be indexed to electrochemical dissolution of carbon as well as Zn (Eq. (5)). And, the peak density is smaller than that in bare H2SO4, indicating the existence of Zn(II) could reduce the oxidation of carbon during anodic polarization process.
Zn/Zn2þ þ 2e
(5) 2þ
In this case, the reversible reaction of Zn /Zn should happen on negative electrode surface though there is no distinct peak can be observed during cathodic scanning. Moreover, the electrode potential of Zn2þ/Zn (Eq. (3)) is more positive than that of Hþ/H2 (Eq. (2)), indicating the electrochemical reaction of Zn2þ to Zn is prior to hydrogen evolution. According to the above discussion, it is supposed there are two aspects of reasons responsible for the reducing hydrogen evolution. Firstly, the hydrogen evolution on carbon surface is reduced due to the formation of Zn on carbon surface. The electrode potential of Zn2þ/Zn is more positive than that of Hþ/H2 according to the CV curve in Fig. 4b, indicating the electrochemical reaction of Zn2þ to Zn is prior to hydrogen evolution during charge process. The deposition of Zn from Zn2þ during charge process occurs both on the surface of Pb and carbon. Then, at the end of charge process, hydrogen generates from the surface of negative mass, which is much easier on carbon surface rather than on lead surface due to the lower Hþ/H2 overpotential of carbon [10]. The covering of Zn layer on part of the carbon surface makes the hydrogen evolution on carbon surface be reduced. Secondly, the hydrogen evolution on lead surface is also reduced due to the increased porosity of negative mass. The electrochemical reaction happened in negative electrode of LABs during charging is efficient transformation of lead sulfate to spongy lead. Only at overcharge condition, Hþ in electrolyte combines with electron that
Fig. 5. Capacity and charge acceptance of cells with EAC and Zn(II) additives: (a) C10 discharge capacity at 25 C; (b) C10 discharge capacity at 10 C; (c) 3C discharge capacity at 25 C; (d) charge acceptance.
J. Xiang et al. / Journal of Power Sources 328 (2016) 8e14
13
Fig. 6. Schematic diagram for beneficial effect of Zn(II) on low-temperature and high-rate capacity.
Fig. 7. Nyquist plots of the EAC þ ZnO and EAC electrodes after fully charged.
makes hydrogen evolution. The porosity of the EAC-ZnO negative plate is apparently increased than those of the EAC plates. Usually, higher porosity of negative mass indicates better charge acceptance. It means the hydrogen generates on the surface of EAC-ZnO electrode is delayed comparing to the EAC electrode under the same charging condition. In other words, at the same over-charge potential, the current density of the EAC-ZnO electrode is lower than that of EAC electrode, which indicate the low hydrogen evolution rate of EAC-ZnO electrode. It is consistent with the linear sweep voltammogram curves in Fig. 3. According to Fig. 3 and Table 4, the inhibition effect on hydrogen evolution is mostly related to the concentration of Zn2þ in the surface region of negative plate. As presented in Table 5, the completed dissolution of ZnSO4 in electrolyte leads to more Zn2þ absorption on the NAM surface, making more reaction of Eq. (3) happen during charging. Although Zn2þ in electrolyte may migrate toward positive electrode, the influence on the performance of positive electrode is unclear yet. Fig. 5a shows the C10 discharge capacity at 25 C. The cell with EAC in NAM shows higher C10 capacity than the control one, ascribing to the high surface activity of EAC that promotes the reaction kinetics of Pb and HSO 4 and improves the utilization of
Fig. 8. (a) End-of-discharge voltages of the control, EAC and EAC þ ZnO cells during cycling; (b) capacity retention after cycling.
active mass. Doping ZnO together with EAC has little effect on capacity, while doping ZnSO4 in electrolyte contributes more to the capacity. As is known to all, the capacity of VRLA cell is closely related to electrolyte density. Higher electrolyte (in range of 1.20e1.30 g cm3) results in higher capacity [18]. Doping ZnSO4 in electrolyte increases the concentration of SO2 4 , that's the reason why C10 capacity of “EAC þ ZnSO4” cell is higher the other ones. The C10 discharge capacity at 10 C, and 3C (45 A) discharge capacity at 25 C is presented in Fig. 5b and c. Doping EAC in NAM causes the capacities to be reduced by 6% and 20%. Generally, the diffusion rate of ions and electrons inside the negative plate become more critical under low temperature and high current discharge conditions. The conductivity of EAC is lower than lead, which makes EAC dispersed in NAM like “insulated island” to block
14
J. Xiang et al. / Journal of Power Sources 328 (2016) 8e14
the electron transfer among the spongy lead (Fig. 6a). The blocking effect is not serious at room temperature and low rate, but becomes significant at low temperature and high rate, which results in the decreasing discharge capacity at 10 C and 3C discharge. However, doping ZnO in negative mass and ZnSO4 in electrolyte can improve the low-temperature and high-rate capacity since the formation of Zn from Zn2þ on lead and carbon surface constructs a conductive bridge among active mass, which improve the conductivity of the negative electrode (Fig. 6b). Fig. 5d shows doping EAC in NAM increases the charge acceptance of the cell by 40% due to the optimization of NAM structure and porosity [4]. Introducing Zn(II) additives has no negative effect on the charge acceptance but even has a little promotion function. This phenomenon is consistent with the specific surface area of negative active mass in Table 3, indicating the NAM surface area is a critical factor to influence the charge acceptance. Larger the surface area of negative mass, better the charge acceptance of the cell. Fig. 7 presents the Nyquist plots of the fully charged negative plates to further confirm the improved conductivity after doping Zn(II). The Z0 -intercept of the Nyquist plots of EAC þ ZnO electrode is smaller than that of the EAC electrode. As is known to all, the value of Z’-intercept is indexed to the ohmic resistance of the electrode. Therefore, it is considered that the intrinsic electric conductivity of negative mass is improved after ZnO doping, which is due to the deposition of Zn from Zn(II) on the surface of lead and carbon. The distinct semicircle in the medium-low frequency region usually reflects the charge-transfer resistance on the active mass interface. As can be seen, the EAC þ ZnO electrode exhibits the smaller semicircle diameter than the EAC electrode. It is thus indicated the charge transfer on the lead surface of the EAC þ ZnO electrode is facilitated, which should be attributed to the improved porosity and increased active surface after ZnO doping. Fig. 8a shows the end-of-discharge (EoD) voltages of the control, EAC and EAC þ ZnO cells during cycling. The EoD voltage of the control cells decreases from 1.97 V to 1.91 V during the 150 PSoC cycles. However, for the cells with EAC and EAC þ ZnO in negative mass, the EoD voltage stays at about 1.94 V and 1.97 V after cycling. The cells are fully charged back every 150 cycles and the capacity retention is shown in Fig. 8b. The capacity retention of the EAC and EAC þ ZnO cells keep 84% and 90% after 2100 cycles while that of the control cell fades quickly below 80% only after 1350 cycles. The improved PSoC cycling performance should be attributed to the cocontribution of EAC and Zn(II). The EAC optimizes the NAM structure and porosity to enhance the charge acceptance and retard the lead sulfate accumulation on the surface of negative plate. The Zn(II) additive reduces the hydrogen evolution during charge process and improves the electric conductivity of the negative electrode, especially at a discharge state. 4. Conclusions The hydrogen evolution of the cell with EAC in negative mass is significantly reduced by 58e73% via adding Zn(II) additives. The NAM structure and composition characterization and electrochemical measurements indicates the deposition of Zn on carbon surface and the increased porosity of negative mass are the main
reasons for the inhibition of H2 generation during charging. The existence of Zn(II) on the negative mass surface could also reduce the oxidation of carbon during anodic polarization process. The Zn(II) additives can also improve the low-temperature and high-rate capacity of the cell with EAC in negative mass, which ascribes to the formation of Zn on lead and carbon surface that construct a conductive bridge among the active mass. The cell with 0.6 wt% EAC and 0.006 wt% ZnO in negative mass exhibits 90% reversible capacity of the initial capacity after 2100 cycles at partialstate-of-charge duty, much better than the cell with 0.6 wt% EAC and the control cell with no EAC and Zn(II). The greatly improved PSoC cycling performance is attributed to the co-effect of EAC and Zn(II). The EAC optimizes the NAM structure and porosity to enhance the charge acceptance and retard the lead sulfate accumulation on the surface of negative plate. And the Zn(II) additive reduces the hydrogen evolution during charge process and improves the electric conductivity of the negative electrode, especially at a discharge state. Acknowledgements This work was supported by the Science and Technology Program of State Grid Corporation of China (Program Name: Study on preparation of new carbon materials for lead carbon batteries). The authors appreciate the professional assistance from Dr. Herbert K. Giess. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jpowsour.2016.07.113. References [1] L.T. Lam, N.P. Haigh, C.G. Phyland, A.J. Urban, J. Power Sources 133 (2004) 126e134. [2] A. Cooper, J. Furakawa, L. Lam, M. Kellaway, J. Power Sources 188 (2009) 642e649. [3] E. Ebner, D. Burow, A. Borger, M. Wark, P. Atanassova, J. Valenciano, J. Power Sources 239 (2013) 483e489. [4] J.Y. Xiang, P. Ding, H. Zhang, X.Z. Wu, J. Chen, Y.S. Yang, J. Power Sources 241 (2013) 150e158. [5] P.T. Moseley, J. Power Sources 191 (2009) 134e138. [6] P.T. Moseley, D.A.J. Rand, K. Peters, J. Power Sources 295 (2015) 268e274. [7] V. Srinivasan, G.Q. Wang, C.Y. Wang, J. Electrochem. Soc. 150 (2003) 316e325. [8] D. Pavlov, T. Rogachev, P. Nikolov, G. Petkov, J. Power Sources 191 (2009) 58e75. [9] P. Beca, K. Micka, P. Krivik, K. Tonar, P. Toser, J. Power Sources 196 (2011) 3988e3992. [10] Y. Chen, B.Z. Chen, X.C. Shi, H. Xu, W. Shang, Y. Yuan, L.P. Xiao, Electrochem. Acta 53 (2008) 2245e2249. [11] L. Zhao, B.S. Chen, J.Z. Wu, D.L. Wang, J. Power Sources 248 (2014) 1e5. [12] L. Zhao, B.S. Chen, D.L. Wang, J. Power Sources 231 (2013) 34e38. [13] L.T. Lam, J. Power Sources 195 (2010) 4494e4512. [14] P. Atanassov, A. DuPasquier, M. Oljaca, P. Nikolov, M. Matrakova, D. Pavlov, Carbon additives for advanced lead acid battery applications, in: 14th European Lead Battery Conference, Edinburgh, UK, Sept. 2014, pp. 9e12. [15] D. Boden, J. Arias, F.A. Fleming, J. Power Sources 95 (2001) 277e292. [16] D.A.J. Rand, P.T. Moseley, J. Garche, C.D. Parker, Valve-regulated Lead-Acid Batteries, Elsevier B.V., Amsterdam, 2004, pp. 243e244. [17] K.R. Bullock, J. Power Sources 195 (2010) 4513e4519. [18] D. Pavlov, V. Naidenov, S. Ruevski, J. Power Sources 161 (2006) 658e665.
Contents lists available at ScienceDirect
Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour
Enhanced performance of Zn(II)-doped lead-acid batteries with electrochemical active carbon in negative mass Jiayuan Xiang a, *, Chen Hu b, **, Liying Chen a, Dong Zhang c, Ping Ding a, Dong Chen a, Hao Liu b, Jian Chen a, Xianzhang Wu a, Xiaokang Lai b a b c
Research Institute of Narada Power Source Co., Ltd, Hangzhou 311305, China China Electric Power Research Institute, Beijing, 100192, China Hangzhou City of Quality and Technical Supervision and Testing Institute, Hangzhou 310019, China
h i g h l i g h t s Gas evolution is reduced by 58e73% via adding Zn(II) additives. Low-temperature and high-rate capacity is improved by Zn(II) additives. Cycle life is prolonged under the co-effect of carbon and Zn(II) additives. The cell exhibits 90% reversible capacity after 2100 cycles at partial-state-of-charge duty.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 8 April 2016 Received in revised form 29 July 2016 Accepted 30 July 2016
The effect and mechanism of Zn(II) on improving the performances of lead-acid cell with electrochemical active carbon (EAC) in negative mass is investigated. The hydrogen evolution of the cell is significantly reduced due to the deposition of Zn on carbon surface and the increased porosity of negative mass. Zn(II) additives can also improve the low-temperature and high-rate capacities of the cell with EAC in negative mass, which ascribes to the formation of Zn on lead and carbon surface that constructs a conductive bridge among the active mass. Under the co-contribution of EAC and Zn(II), the partial-state-of-charge cycle life is greatly prolonged. EAC optimizes the NAM structure and porosity to enhance the charge acceptance and retard the lead sulfate accumulation. Zn(II) additive reduces the hydrogen evolution during charge process and improves the electric conductivity of the negative electrode. The cell with 0.6 wt% EAC and 0.006 wt% ZnO in negative mass exhibits 90% reversible capacity of the initial capacity after 2100 cycles. In contrast, the cell with 0.6 wt% EAC exhibits 84% reversible capacity after 2100 cycles and the control cell with no EAC and Zn(II) exhibits less than 80% reversible capacity after 1350 cycles. © 2016 Elsevier B.V. All rights reserved.
Keywords: Lead-acid battery Carbon Hydrogen evolution Partial-state-of-charge Zinc additive
1. Introduction In order to cope with the challenge of high-rate-partial-state-ofcharge (HRPSoC) duty in micro-hybrid electric vehicles and partialstate-of-charge (PSoC) duty in energy storage system for microgrid, new technology in lead-acid batteries (LABs) should be developed to provide higher levels of charge acceptance to enhance system efficiency and delay common failure mechanism such as lead sulfate accumulation on surface of negative plate [1].
* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (J. Xiang), [email protected] (C. Hu). http://dx.doi.org/10.1016/j.jpowsour.2016.07.113 0378-7753/© 2016 Elsevier B.V. All rights reserved.
Introducing carbon in types of activated carbon, carbon black and expanded graphite to negative active mass (NAM) during paste preparation has been proved to be an effective way to improve charge acceptance and prolong cycle life under PSoC duty [2e4]. The understanding of possible mechanism by which a carbon component enhance the performances of LABs has also been developed from eight different functions [5] to just three candidates that are most likely to have a significant individual effect [6]: (i) capacitive buffer to absorb charge current in excess of which can be accommodated by the Faradic reaction [7]; (ii) the extension of the surface area on which the electrochemical charge and discharge processes can take place [8]; (iii) the physical processes that hinder the crystallization of lead sulfate and thus help to maintain a high surface area for the discharge product [9].
J. Xiang et al. / Journal of Power Sources 328 (2016) 8e14
However, a risk of hydrogen evolution at the end of charge process maybe also brought in when adding carbon to LABs due to the low hydrogen evolution overpotential of carbon materials under the negative plate working conditions [10]. Attention needs to be paid to avoid harmful influences like more gas release, thermal runaway or even electrolyte dry-out. Introducing metal ions with high hydrogen evolution overpotential in LABs has been proved to be an effective way to retard hydrogen-evolution rate. For instance, In2O3, Bi2O3 and Ga2O3 were reported to be beneficial for suppressing hydrogen evolution and prolong HRPSoC cycle life of valve-regulated-lead-acid (VRLA) batteries with carbon [11,12]. But as is well known, residual metal element is always “harmful elements” for LABs since they may cause hydrogen and/or oxygen gassing and which, therefore, must be restricted or avoided. The influence of seventeen elements namely Sb, As, Bi, Gd, Gr, Co, Cu Ge, Fe, Mn, Ni, Se, Ag, Te, Tl, Sn and Zn in lead on hydrogen- and/or oxygen-gassing rates of LABs has been investigated [13], only Bi, Cd, Sn and Zn are considered as “beneficial elements” for VRLA batteries since they have little effect on gassing. The remaining metal elements are “harmful” since they will promote the gassing rate. Therefore, the study of doping “beneficial elements” in LABs with carbon in negative mass and their influence on battery performances is valuable and urgent. But to date, there is few research about it. In the present work, the focus lies on effect of Zn(II) at the form of ZnO doping in negative mass or ZnSO4 doping in electrolyte on hydrogen evolution and other performances of VRLA batteries with carbon. The purpose is to understand the function mechanism of high hydrogen-evolution-overpotential metal ions, provide a promising way to eliminate the water-consumption risk when adding carbon in negative mass, and improve the other performances of the LABs. 2. Experimental 2.1. Materials and characterization Carbon added to negative mass is provided by Cabot Corporation, with BET surface area of 1350 m2 g1 and particle size (D50) of 13 mm. Due to its high surface area and great contribution to the electrochemical reactions of charge, we labelled the carbon as electrochemical active carbon (EAC) as some previous literature named [8]. Leady oxide (80% PbO) is manufactured in Narada Company from 99.99% pure lead by ball milling. Vanisperse-A and BaSO4 is added in NAM as expander. ZnO and ZnSO4 is analytical reagent from Xilong Chemical Co., Ltd. Battery grade sulfuric acid (1.40 sp. gr.) is used during paste mixing. X-ray diffraction (XRD-7000, Shimadzu) was used to detect the phases of negative mass. Scanning Electron Microscope (SEM, Hitachi S-3400N) and BET Surface Analyzer (Gemini VII 2390, Micromeritics) were used to characterize the morphology and active surface area of carbon and negative mass. Particle Size Analyzer (Malvern-2000) was used to measure the particle size of carbon material. Mercury Intrusion Porosimetry (Autopore 9500, Micromeritics) was used to test the porosity of negative mass. Inductive Coupled Plasma-Atomic Emission Spectrometry (ICPAES, Optima 7000 DV) was used to measure the content of Zn element in negative mass and electrolyte. Gas chromatographmass spectrometer (GC-MS, Agilent 5977) is used to determine the hydrogen and oxygen ratio of the accumulated gas from the cells during 2.40Vpc charging. 2.2. Cell design and manufacturing Experimental groups of EAC and Zn(II) additives doped to
9
negative pastes are listed in Table 1. Based on the negative plate containing 0.6 wt% EAC, ZnO and ZnSO4 as Zn(II) additives were added in the cells, respectively. ZnO with doping level of 1.0 wt% of EAC was firstly mixed with EAC and then added in negative mass (0.006 wt% of NAM), and ZnSO4with doping level of 1 mg ml1 was directly added in electrolyte. For comparison, the cell containing no carbon and Zn(II) additives was also prepared. In order to describe clearly, the four groups are simply named as control, EAC, EAC þ ZnO, EAC þ ZnSO4. Each negative plate contains 0.05 wt% sodium lignosulphonate (Vanisperse-A) and 0.45 wt% BaSO4. 2 V prototype cell with 10h-rate capacity (C10) of 15 A h was fabricated and tested. The cell characteristic is shown in Table 2. 2.3. Electrochemical measurement Linear sweep voltammogram was performed on CHI660C electrochemical workstation to characterize the hydrogen evolution behavior of negative plate in sulfuric acid (1.30 sp. gr.) at scanning rate of 0.1 mV s1 from 1.0 V to 1.7 V versus Ag/Ag2SO4. Cyclic voltammogram was also performed on CHI660C electrochemical workstation to study the electrochemical behavior of EAC electrode in different electrolyte at scanning rate of 0.1 mV s1 from 0.6 V to 1.4 V versus Ag/Ag2SO4. Nyquist plots of the electrodes was measured on ModuLab XM electrochemical workstation from 10 kHz to 0.01 Hz with amplitude of 5 mV to investigate the conductivity of the electrodes. Discharge capacity was test at three conditions: (i) regular duty that discharge current was 1.5 A with cut-off voltage of 1.80 Vpc and ambient temperature was 25 C; (ii) low temperature duty that discharge current was 1.5 A with cut-off voltage of 1.80Vpc and ambient temperature was 10 C; (iii) high rate duty that discharge current was 3C (45 A) with cut-off voltage of 1.50 Vpc and ambient temperature was 25 C. Gas evolution property was evaluated under floating charge condition. After the cells were floating charged at 2.40 Vpc for 72 h, collected the gas generated during the following 192 h with also 2.40 Vpc floating charging. Charge acceptance was evaluated as follows: (i) discharged the cells with 1.5 A until the SoC level reached to 50%; (ii) rest for 24 h at temperature of 0 C; (iii) charged the cells with constant voltage of 2.40 V and no current limitation; (iv) recorded the current I10min as charged for 10 min, and calculated the ratio of I10min/I10 (I10 is the 10 h-rate current). PSoC cycling performance was evaluated using a simplified profile imitating the energy storage mode: (i) firstly discharged the cells from fully-charged-state to 30%SoC at 5 A; (ii) charged to 80% SoC at 2 A with 2.35 V limited; (iii) discharged to 30%SoC at 5 A; (iv) repeated step (ii) to step (iii) for 150 times; (v) fully recharged the cells at 2 A; (vi) determined the discharge capacity at 2 A; (vii) fully charged the cells and then went back to step (i). The cell voltage was measured at the end of discharge, the discharge capacity was recorded, and the test was stopped when the cell discharge voltage fell down to 1.80 V or the cell capacity fell below 8 Ah. 3. Results and discussion The XRD patterns of the formed negative mass are presented in Fig. 1a. The diffraction peaks of the control plate is mainly indexed to Pb (PDF No. 65e2873), only some minor peak indexed to a-PbO (PDF No. 85e1739) is detected. EAC doping in negative mass makes the XRD pattern change remarkably that the peaks of a-PbO become much stronger while the peak intensity of Pb decrease accordingly. It means the surface activity of spongy lead is enhanced with EAC, which might cause more lead oxidation reaction happen when the negative plate placed in air. As EAC added
10
J. Xiang et al. / Journal of Power Sources 328 (2016) 8e14
Table 1 Experimental groups of EAC and Zn(II) additives in cells. No.
Signature
Negative mass formulation
Electrolyte
1 2 3 4
control EAC EAC-ZnO EAC-ZnSO4
e EAC 0.6 wt% EAC 0.6 wt% þ ZnO 0.006 wt% EAC 0.6 wt%
H2SO4 H2SO4 H2SO4 H2SO4
(1.28 (1.28 (1.28 (1.28
sp. sp. sp. sp.
gr.) gr.) gr.) gr.)þ ZnSO4 (1 mg ml1)
Table 2 Characteristics of 2 V prototype cell (C10 ¼ 15 A h). Parameter
Positive
Negative
Grid alloy Gird size (mm mm) Plate thickness (mm) Plate number AGM compression ratio
Pb-Ca-Sn 70 80 4.49 4 70%
Pb-Ca 70 80 2.57 5
together with Zn(II) additive, the peak intensity of a-PbO is a little weaker than that of EAC plate without Zn(II), which indicates the effect of EAC on surface activity of spongy lead is reduced by Zn(II) doping. The semi quantitative calculation by the XRD-7000 analysis software shows the content of a-PbO phase of the control, EAC and EAC-ZnO plates are 5.25%, 12.20% and 11.12%, respectively. The SEM images of the formed negative mass in Fig. 1bed shows similar porous and spongy morphology. The BET surface area of NAM increases significantly from 1.41 m2 g1 to 5.54 m2 g1 and 6.22 m2 g1 after doping EAC and EAC þ ZnO. The porosity of the EAC þ ZnO electrode is also improved from 54.4% to 55.6% (Table 3). Generally, the theoretical BET surface of NAM including 0.6 wt% EAC should be 8.24 m2 g1 assuming all the surface of EAC (1350 m2 g1) is exposed and active in NAM. However, the actual BET surface of NAM is 5.54 m2 g1. The loss in surface area might be attributed to the absorption of sodium lignosulphonate on carbon [14]. In conventional lead-acid battery, sodium lignosulphonate as a typical organic expander are partially soluble in battery electrolyte and can adsorb on the negative plate, which increase the polarization of the positive plate to a value that allows the negative plate to become less polarized [15]. Therefore, it is understandable that sodium lignosulphonate is firstly dissolved and then adsorbs on the surface of PbO as well as carbon during the pasting process of negative mass with carbon, making some active surface of EAC is not exposed any more (Fig. 2). When EAC doping together with ZnO, Zn(II) will also combine with sodium lignosulphonate then decrease the absorption on the carbon surface, which results in the increasing BET surface area of negative mass from 5.54 m2 g1 to 6.22 m2 g1. The linear sweep voltammogram curves of different negative plates were displayed in Fig. 3a. The current density is always a key factor to indicate the hydrogen evolution rate at over-charge condition. As can be seen, the current density of the negative plate containing EAC is 0.150 mA cm2 at the potential of 1.40 V, much higher than that of the control plate (0.054 mA cm2). It is thus confirmed that carbon with high surface area doping in negative mass accelerates the hydrogen evolution. When EAC doping together with ZnO in negative mass or ZnSO4 in electrolyte, the electrode current density at 1.40 V significantly decreases to 0.077 mA cm2 and 0.103 mA cm2, respectively, which means Zn(II) is an effective additives to depress the hydrogen evolution. The gas generated from 2 V to 15 A h cells during 2.40 Vpc floating charging for 192 h is collected (according to section 2.3) to further verify the effect of Zn(II) on reducing hydrogen evolution. As shown in Fig. 3b, the total amount of gas generated from the control cell during 192 h charging is 0.30 ml h1 Ah1. When EAC
Fig. 1. XRD patterns and SEM images of the formed negative active mass: (a) XRD patterns; SEM images of (b) control; (c) EAC; (d) EAC þ ZnO.
J. Xiang et al. / Journal of Power Sources 328 (2016) 8e14 Table 3 BET surface and porosity of formed NAM for each test group. No.
Group
BET (m2 g1)
Porosity (%)
1 2 3
Control EAC EAC þ ZnO
1.41 5.54 6.22
54.4 54.1 55.6
11
0.83 ml h1 Ah1 and 0.54 ml h1 Ah1 during the 192 h charging process, which is reduced by 58% and 73%, respectively. The electrochemical reaction happened in negative electrode of LABs during charging is efficient transformation of lead sulfate to spongy lead (Eq. (1)).
PbSO4 þ Hþ þ 2e /Pb þ HSO4
(1)
Due to the high porosity of the negative mass, the charge acceptance of negative electrode is much better than the positive one, so the negative electrode is considered as over charged only until the state of charge is above 90%. At over-charge condition, Hþ in electrolyte combines with electron that makes hydrogen evolution (Eq. (2)). The reversible potential of this side reaction is about 350 mV higher than that of Pb/PbSO4. In a VRLA cell, the typical “oxygen cycle” would cause negative potential shift positively, and thus reduce or even eliminate the hydrogen generation [16]. When EAC is doped in negative mass, the hydrogen evolution rate will be no doubt accelerated since the reaction of Eq. (2) becomes much easier on carbon surface rather than on lead surface.
2Hþ þ 2e /H2 [
(2)
Fig. 2. BET surface change of NAM after doping EAC and ZnO.
doped in negative mass, the generated gas is 2.00 ml h1 Ah1, nearly 6e7 times more than the control one. But, the gas quantity decreases remarkably from the cells containing both EAC and Zn(II). The two cells, one containing EAC and ZnO in NAM, the other containing EAC in NAM and ZnSO4 in electrolyte, exhibit gas of
Fig. 3. (a) Linear sweep voltammogram curves of different negative plates; (b) gas evolutions of different cells under 2.40 Vpc floating charging.
Table 4 Composition of the accumulated gas in the last 24 h during the 192 h float-charging with 2.40 Vpc.
H2% O2%
Control
EAC
EAC þ ZnO
EAC þ ZnSO4
79.33 20.67
95.79 4.21
90.76 9.24
86.28 13.72
Fig. 4. Cyclic voltammogram curves of EAC electrode in (a) 1.28 sp. gr. H2SO4 and (b) 1.28 sp. gr. H2SO4 with 1 mg ml1 ZnSO4.
12
J. Xiang et al. / Journal of Power Sources 328 (2016) 8e14
Table 5 The quantity of Zn element in electrolyte and active mass after cell was fully charged. Group
Electrolyte (mg kg1)
NAM (mg kg1)
PAM (mg kg1)
EAC þ ZnO EAC þ ZnSO4
1.3 14.7
0.6 2.8
not detected not detected
The gas generated from the cells during 2.40 Vpc charging involves hydrogen as well as oxygen. GC-MS is used to determine the hydrogen and oxygen ratio of the accumulated gas as shown in Table 4. The hydrogen percent of the gas gathered from the control cell is 79.33%, while that of the EAC cell is as high as 95.79%. By doping ZnO and ZnSO4, the hydrogen percent reduced to 90.76% and 86.28%, respectively. According to the gas volume data in Fig. 4, the hydrogen volumes accumulated from the cells with ZnO and ZnSO4 is indeed reduced by comparing to the EAC cell. Similar to the Hþ/H2 reaction, when Zn(II) is added in negative mass or electrolyte, the dissolved Zn2þ will also combine with electron and become Zn during charging (Eq. (3)).
Zn2þ þ 2e /Zn
(3)
In order to investigate the mechanism of Zn(II) on electrochemical behavior of the cell containing EAC in NAM, especially in the over-charge process, the cyclic voltammogram curves of EAC electrode in bare H2SO4 (1.28 sp. gr.) and H2SO4 with 1 mg ml1 ZnSO4 are presented in Fig. 4. In bare H2SO4 electrolyte, EAC electrode exhibits a pair of cathodic and anodic peaks at potential of 0.95 to 1.00 V. As scanning below 1.20 V, hydrogen evolution begins. According to Bullock's research [17], the peaks at 0.95 to 1.00 V should correspond to the dissolve of carbon in aqueous solution to form carboxyl group or formic acid (Eq. (4)).
C þ 2H2 O/HCOOH ðaqueousÞ þ 2Hþ þ 2e
(4)
However, in the electrolyte containing Zn2þ, EAC electrode shows different CV curve that involves two cathodic peaks at potential of 0.92 to 0.98 V, which should be indexed to electrochemical dissolution of carbon as well as Zn (Eq. (5)). And, the peak density is smaller than that in bare H2SO4, indicating the existence of Zn(II) could reduce the oxidation of carbon during anodic polarization process.
Zn/Zn2þ þ 2e
(5) 2þ
In this case, the reversible reaction of Zn /Zn should happen on negative electrode surface though there is no distinct peak can be observed during cathodic scanning. Moreover, the electrode potential of Zn2þ/Zn (Eq. (3)) is more positive than that of Hþ/H2 (Eq. (2)), indicating the electrochemical reaction of Zn2þ to Zn is prior to hydrogen evolution. According to the above discussion, it is supposed there are two aspects of reasons responsible for the reducing hydrogen evolution. Firstly, the hydrogen evolution on carbon surface is reduced due to the formation of Zn on carbon surface. The electrode potential of Zn2þ/Zn is more positive than that of Hþ/H2 according to the CV curve in Fig. 4b, indicating the electrochemical reaction of Zn2þ to Zn is prior to hydrogen evolution during charge process. The deposition of Zn from Zn2þ during charge process occurs both on the surface of Pb and carbon. Then, at the end of charge process, hydrogen generates from the surface of negative mass, which is much easier on carbon surface rather than on lead surface due to the lower Hþ/H2 overpotential of carbon [10]. The covering of Zn layer on part of the carbon surface makes the hydrogen evolution on carbon surface be reduced. Secondly, the hydrogen evolution on lead surface is also reduced due to the increased porosity of negative mass. The electrochemical reaction happened in negative electrode of LABs during charging is efficient transformation of lead sulfate to spongy lead. Only at overcharge condition, Hþ in electrolyte combines with electron that
Fig. 5. Capacity and charge acceptance of cells with EAC and Zn(II) additives: (a) C10 discharge capacity at 25 C; (b) C10 discharge capacity at 10 C; (c) 3C discharge capacity at 25 C; (d) charge acceptance.
J. Xiang et al. / Journal of Power Sources 328 (2016) 8e14
13
Fig. 6. Schematic diagram for beneficial effect of Zn(II) on low-temperature and high-rate capacity.
Fig. 7. Nyquist plots of the EAC þ ZnO and EAC electrodes after fully charged.
makes hydrogen evolution. The porosity of the EAC-ZnO negative plate is apparently increased than those of the EAC plates. Usually, higher porosity of negative mass indicates better charge acceptance. It means the hydrogen generates on the surface of EAC-ZnO electrode is delayed comparing to the EAC electrode under the same charging condition. In other words, at the same over-charge potential, the current density of the EAC-ZnO electrode is lower than that of EAC electrode, which indicate the low hydrogen evolution rate of EAC-ZnO electrode. It is consistent with the linear sweep voltammogram curves in Fig. 3. According to Fig. 3 and Table 4, the inhibition effect on hydrogen evolution is mostly related to the concentration of Zn2þ in the surface region of negative plate. As presented in Table 5, the completed dissolution of ZnSO4 in electrolyte leads to more Zn2þ absorption on the NAM surface, making more reaction of Eq. (3) happen during charging. Although Zn2þ in electrolyte may migrate toward positive electrode, the influence on the performance of positive electrode is unclear yet. Fig. 5a shows the C10 discharge capacity at 25 C. The cell with EAC in NAM shows higher C10 capacity than the control one, ascribing to the high surface activity of EAC that promotes the reaction kinetics of Pb and HSO 4 and improves the utilization of
Fig. 8. (a) End-of-discharge voltages of the control, EAC and EAC þ ZnO cells during cycling; (b) capacity retention after cycling.
active mass. Doping ZnO together with EAC has little effect on capacity, while doping ZnSO4 in electrolyte contributes more to the capacity. As is known to all, the capacity of VRLA cell is closely related to electrolyte density. Higher electrolyte (in range of 1.20e1.30 g cm3) results in higher capacity [18]. Doping ZnSO4 in electrolyte increases the concentration of SO2 4 , that's the reason why C10 capacity of “EAC þ ZnSO4” cell is higher the other ones. The C10 discharge capacity at 10 C, and 3C (45 A) discharge capacity at 25 C is presented in Fig. 5b and c. Doping EAC in NAM causes the capacities to be reduced by 6% and 20%. Generally, the diffusion rate of ions and electrons inside the negative plate become more critical under low temperature and high current discharge conditions. The conductivity of EAC is lower than lead, which makes EAC dispersed in NAM like “insulated island” to block
14
J. Xiang et al. / Journal of Power Sources 328 (2016) 8e14
the electron transfer among the spongy lead (Fig. 6a). The blocking effect is not serious at room temperature and low rate, but becomes significant at low temperature and high rate, which results in the decreasing discharge capacity at 10 C and 3C discharge. However, doping ZnO in negative mass and ZnSO4 in electrolyte can improve the low-temperature and high-rate capacity since the formation of Zn from Zn2þ on lead and carbon surface constructs a conductive bridge among active mass, which improve the conductivity of the negative electrode (Fig. 6b). Fig. 5d shows doping EAC in NAM increases the charge acceptance of the cell by 40% due to the optimization of NAM structure and porosity [4]. Introducing Zn(II) additives has no negative effect on the charge acceptance but even has a little promotion function. This phenomenon is consistent with the specific surface area of negative active mass in Table 3, indicating the NAM surface area is a critical factor to influence the charge acceptance. Larger the surface area of negative mass, better the charge acceptance of the cell. Fig. 7 presents the Nyquist plots of the fully charged negative plates to further confirm the improved conductivity after doping Zn(II). The Z0 -intercept of the Nyquist plots of EAC þ ZnO electrode is smaller than that of the EAC electrode. As is known to all, the value of Z’-intercept is indexed to the ohmic resistance of the electrode. Therefore, it is considered that the intrinsic electric conductivity of negative mass is improved after ZnO doping, which is due to the deposition of Zn from Zn(II) on the surface of lead and carbon. The distinct semicircle in the medium-low frequency region usually reflects the charge-transfer resistance on the active mass interface. As can be seen, the EAC þ ZnO electrode exhibits the smaller semicircle diameter than the EAC electrode. It is thus indicated the charge transfer on the lead surface of the EAC þ ZnO electrode is facilitated, which should be attributed to the improved porosity and increased active surface after ZnO doping. Fig. 8a shows the end-of-discharge (EoD) voltages of the control, EAC and EAC þ ZnO cells during cycling. The EoD voltage of the control cells decreases from 1.97 V to 1.91 V during the 150 PSoC cycles. However, for the cells with EAC and EAC þ ZnO in negative mass, the EoD voltage stays at about 1.94 V and 1.97 V after cycling. The cells are fully charged back every 150 cycles and the capacity retention is shown in Fig. 8b. The capacity retention of the EAC and EAC þ ZnO cells keep 84% and 90% after 2100 cycles while that of the control cell fades quickly below 80% only after 1350 cycles. The improved PSoC cycling performance should be attributed to the cocontribution of EAC and Zn(II). The EAC optimizes the NAM structure and porosity to enhance the charge acceptance and retard the lead sulfate accumulation on the surface of negative plate. The Zn(II) additive reduces the hydrogen evolution during charge process and improves the electric conductivity of the negative electrode, especially at a discharge state. 4. Conclusions The hydrogen evolution of the cell with EAC in negative mass is significantly reduced by 58e73% via adding Zn(II) additives. The NAM structure and composition characterization and electrochemical measurements indicates the deposition of Zn on carbon surface and the increased porosity of negative mass are the main
reasons for the inhibition of H2 generation during charging. The existence of Zn(II) on the negative mass surface could also reduce the oxidation of carbon during anodic polarization process. The Zn(II) additives can also improve the low-temperature and high-rate capacity of the cell with EAC in negative mass, which ascribes to the formation of Zn on lead and carbon surface that construct a conductive bridge among the active mass. The cell with 0.6 wt% EAC and 0.006 wt% ZnO in negative mass exhibits 90% reversible capacity of the initial capacity after 2100 cycles at partialstate-of-charge duty, much better than the cell with 0.6 wt% EAC and the control cell with no EAC and Zn(II). The greatly improved PSoC cycling performance is attributed to the co-effect of EAC and Zn(II). The EAC optimizes the NAM structure and porosity to enhance the charge acceptance and retard the lead sulfate accumulation on the surface of negative plate. And the Zn(II) additive reduces the hydrogen evolution during charge process and improves the electric conductivity of the negative electrode, especially at a discharge state. Acknowledgements This work was supported by the Science and Technology Program of State Grid Corporation of China (Program Name: Study on preparation of new carbon materials for lead carbon batteries). The authors appreciate the professional assistance from Dr. Herbert K. Giess. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jpowsour.2016.07.113. References [1] L.T. Lam, N.P. Haigh, C.G. Phyland, A.J. Urban, J. Power Sources 133 (2004) 126e134. [2] A. Cooper, J. Furakawa, L. Lam, M. Kellaway, J. Power Sources 188 (2009) 642e649. [3] E. Ebner, D. Burow, A. Borger, M. Wark, P. Atanassova, J. Valenciano, J. Power Sources 239 (2013) 483e489. [4] J.Y. Xiang, P. Ding, H. Zhang, X.Z. Wu, J. Chen, Y.S. Yang, J. Power Sources 241 (2013) 150e158. [5] P.T. Moseley, J. Power Sources 191 (2009) 134e138. [6] P.T. Moseley, D.A.J. Rand, K. Peters, J. Power Sources 295 (2015) 268e274. [7] V. Srinivasan, G.Q. Wang, C.Y. Wang, J. Electrochem. Soc. 150 (2003) 316e325. [8] D. Pavlov, T. Rogachev, P. Nikolov, G. Petkov, J. Power Sources 191 (2009) 58e75. [9] P. Beca, K. Micka, P. Krivik, K. Tonar, P. Toser, J. Power Sources 196 (2011) 3988e3992. [10] Y. Chen, B.Z. Chen, X.C. Shi, H. Xu, W. Shang, Y. Yuan, L.P. Xiao, Electrochem. Acta 53 (2008) 2245e2249. [11] L. Zhao, B.S. Chen, J.Z. Wu, D.L. Wang, J. Power Sources 248 (2014) 1e5. [12] L. Zhao, B.S. Chen, D.L. Wang, J. Power Sources 231 (2013) 34e38. [13] L.T. Lam, J. Power Sources 195 (2010) 4494e4512. [14] P. Atanassov, A. DuPasquier, M. Oljaca, P. Nikolov, M. Matrakova, D. Pavlov, Carbon additives for advanced lead acid battery applications, in: 14th European Lead Battery Conference, Edinburgh, UK, Sept. 2014, pp. 9e12. [15] D. Boden, J. Arias, F.A. Fleming, J. Power Sources 95 (2001) 277e292. [16] D.A.J. Rand, P.T. Moseley, J. Garche, C.D. Parker, Valve-regulated Lead-Acid Batteries, Elsevier B.V., Amsterdam, 2004, pp. 243e244. [17] K.R. Bullock, J. Power Sources 195 (2010) 4513e4519. [18] D. Pavlov, V. Naidenov, S. Ruevski, J. Power Sources 161 (2006) 658e665.