Enhancing the performance of lead-acid batteries with carbon - In

Journal of Power Sources 295 (2015) 268e274

Contents lists available at ScienceDirect

Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour

Short communication

Enhancing the performance of leadeacid batteries with carbon e In pursuit of an understanding Patrick T. Moseley a, *, David A.J. Rand b, Ken Peters c a

International Lead Zinc Research Organization, 1822 East NC Highway 54 Suite 120, Durham, NC 27713, USA CSIRO, Private Bag 10, Clayton South, Victoria 3169, Australia c Glen Bank, Worsley, Manchester M28 2GG UK b

h i g h l i g h t s Conflicting notions of the beneficial effect of carbon are reconciled. The evidence underpinning mechanisms is reviewed and rationalised. A combination of different forms of carbon offers maximum benefit.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 8 April 2015 Received in revised form 16 June 2015 Accepted 5 July 2015 Available online 14 July 2015

The inherently poor dynamic charge-acceptance of the leadeacid battery can be greatly improved by the incorporation of additional carbon to the negative plate. An analysis is undertaken of the various ways by which the carbon may be introduced, and of the proposed mechanisms whereby its presence proves to be beneficial. It is intended that such an investigation should provide a guide to the selection of the optimum carbon inventory. © 2015 Elsevier B.V. All rights reserved.

Keywords: Carbon Dynamic charge-acceptance Leadeacid batteries Mechanisms Negative plate

1. The challenge of high-rate partial state-of-charge (HRPSoC) duty The emerging application of leadeacid batteries for the storage of energy from regenerative braking in various types of electric vehicle requires the best possible recovery of charge during highrate partial-state-of-charge (HRPSoC) duty. Micro-hybrid vehicles, which are fitted with stop‒start features both to improve fuel economy and to reduce emissions, employ batteries that operate under PSoC conditions and are charged by energy recuperation from the regenerative braking system. The schedule is a micro-cycle of short bursts of discharge and charge, each at a high rate, with the charge from regenerative braking supplementing that from the generator. The PSoC amplitude is just

* Corresponding author. E-mail address: [email protected] (P.T. Moseley). http://dx.doi.org/10.1016/j.jpowsour.2015.07.009 0378-7753/© 2015 Elsevier B.V. All rights reserved.

a few percent. The current accepted by the battery during the charge periods should be sufficient, when efficiently used, to convert discharge product to the active state and thereby ensure availability of stop‒start and other essential functions. This charge current has been defined [1] as the ‘dynamic charge-acceptance’, DCA, and is the average current over the initial charging period, which is usually between 3 and 20 s. The parameter is quoted in terms of amperes (A) per ampere-hour (Ah) of battery capacity, i.e., A Ah1. Standardized industry specifications are in the course of preparation (CENELEC Standard EN50342-6: Lead‒acid Starter Batteries for Micro-cycle Applications). The DCA of new leadeacid batteries generally lies between 0.5 and 1.5 A Ah1, although some applications may require higher values. For effective fuel savings and low-emission features, this value should be sustained. After relatively short service under PSoC conditions, however, recent work [2,3] has found the DCA to decrease to around 0.1 A Ah1 and, after brief periods of parking, to even lower levels. Resolution of this problem is essential if the

P.T. Moseley et al. / Journal of Power Sources 295 (2015) 268e274

intended benefits are to be achieved. Factors that influence these inconsistencies in battery performance have been investigated, namely: local current distribution, state-of-charge, temperature, and variable acid strength due to stratification [4]. The progressive decrease in DCA can be attributed to rate-limiting processes that include the accumulation of lead sulfate in the negative plates with subsequent effects on concentration and potential gradients within the porous matrix. New leadeacid batteries can be recharged effectively at high rates of charge because the freshly-discharged product, lead sulfate, has a small crystallite size which facilitates rapid dissolution d a requirement that is fundamental to subsequent recharge via the socalled ‘solution‒precipitation’ mechanism (reaction [3] in Fig. 1). On the other hand, if the battery is left at a PSoC for a significant length of time after discharge from top-of-charge, the lead sulfate crystals have the opportunity to grow progressively via the ‘Ostwald Ripening’ process. Consequently, the charge-acceptance of the battery declines, particularly at the negative plate which offers less surface-area (0.5e1 m2 g1) than the positive (4e5 m2 g1). The immediate history of the battery affects the chargeacceptance quite markedly when micro-cycling is performed over a narrow range of PSoC, with no excursions to top-of-charge [5,6]. A small discharge produces some new (small) lead sulfate crystals, which can support a high rate of recharge so that a relatively healthy DCA can be achieved. Immediately after a charge event, however, the small crystals of lead sulfate will have been consumed so that only material of low surface-area remains and poor DCA is observed. Lacking a construction that is purpose-designed for HRPSoC duty, valve-regulated lead‒acid (VRLA) batteries typically lose at least 50% of their initial capacity after operating in a simulated hybrid electric vehicle (HEV) mode for a relatively short time. The loss of capacity has been attributed to a progressive build-up of lead sulfate on the negative plate, especially at the bottom. Since the battery is not brought to a full state-of-charge in PSoC duty, there is no routine method available to remove this lead sulfate. On charging a leadeacid cell, the fundamental reaction at the positive electrode may be accompanied by oxygen evolution and that at the negative by hydrogen evolution. All four reactions are independent. The only requirement is that the currents at the two electrodes are equal. The end-point of a completed charge is always water electrolysis during which gas evolution predominates. The

269

rates of other possible side-reactions, such as grid corrosion, ozone formation and the decomposition of organic additives, are low and usually neglected. With new freshly-formed cells, negative electrodes can be charged efficiently over a range of current densities and temperatures with little gas evolution, whereas under similar conditions positive electrodes evolve oxygen from the outset [7]. At high rates of charge, it becomes difficult to sustain the mass and charge balances that are necessary for the solution‒precipitation charging mechanism to proceed. While the lead sulfate crystals are small, the fluxes of ions to the reaction sites may be ratedetermining [8]. With increasing crystallite size, however, the discharge product may become resistant to the charging and, as a consequence, the potential rises and charge current is diverted to the parasitic reaction of hydrogen evolution. Ultimate battery failure occurs when the negative plate remains in the discharged state. It has been found [9] that some forms of carbon, when present in or on the negative active-material, can be very beneficial in minimizing the irreversible formation of lead sulfate. 2. Previous notions about the function of carbon It is standard practice in the manufacture of leadeacid batteries to add to the negative plate a combination of barium sulfate, an organic extract of wood, and carbon. Collectively known as ‘expanders’ and usually in total comprising less than 1 wt.% of the negative mass, the additives not only increase the available capacity (especially when cells are discharged at high current densities and low temperatures) but also reduce capacity loss during cycling. The mechanism by which expanders prove effective was studied extensively during the last century [10‒14]. Barium sulfate is considered to act as a seed for the precipitation of lead sulfate due to the isomorphous structure of the two salts. The organic additives d usually sulfonated derivatives of lignin (a natural product extracted from wood pulp) d are surfactants that absorb both on the surface of lead and on the lead sulfate crystals as they are formed during discharge. This action inhibits crystal growth during the dissolutioneprecipitation reaction and thereby helps to maintain a high surface-area by preventing contraction and solidification of the ‘spongy’ lead active-material. The function of carbon additions to the negative active-material has been the subject of much debate. At one time, the presence of finely-divided amorphous carbon, otherwise known as ‘lampblack’

Fig. 1. Discharge and charge reactions at the negative plate of a leadeacid cell.

270


(or ‘gas carbon black’) was thought to improve the conductivity of the discharge product and to assist in the formation of pasted plates during manufacture. On the other hand, no benefits of carbon addition were found during a study in which flooded cells were cycled with different levels of overcharge [15]. In summary, the combination of expander additives counteracts the tendency for the spongy lead to densify during cycling and thereby imparts sustainable low-temperature and high-rate capabilities to the negative plate. The benefits of including additional carbon in the negative active-material were eventually demonstrated by the work of Nakamura and Shiomi [16,17], who made negative plates that contained up to ten times the customary level of carbon. The actual amounts were not disclosed but, based on normal practice, were believed to be about 2.0 wt.% of the negative mass. The trials, which were directed towards both electric vehicle and photovoltaic power applications, were undertaken with VRLA batteries that operated under PSoC conditions to minimize overcharge effects. Each of the applications was likely to have entailed relatively short bursts of charge at high current densities. Batteries with standard levels of carbon failed quickly due to the build-up of lead sulfate in the negative plate. In contrast, the associated positive plate was fully charged. Batteries with extra carbon enjoyed appreciably longer operating lives. A study conducted by CSIRO under the auspices of the Advanced LeadeAcid Battery Consortium (ALABC) showed [18] that HEV duty cycles cause the exterior layers of negative plates to become almost completely sulfated. With, presumably, further crystal growth during open-circuit, the eventual result is pore blocking and acid starvation inside the plates. As pointed out much earlier [19] and given the high surface-area of positive active-material compared with that of the negative (v.s.), a full discharge results in a 0.025 mm thick layer of lead sulfate, overall, on the positive and a 0.3 mm layer on the negative. Little wonder that there is pore blocking, a muchreduced flux of sulfate ions during high-rate charging, acid starvation and poor charge-acceptance. More recent developments have revealed that the additional carbon enhances charge efficiency under high-rate charging conditions such as occur in vehicles with regenerative braking. It has also become clear that supplementary carbon can enable cell voltages to remain far more uniform in battery banks used for the storage of renewable energy where the charge is limited, for example, in remote-area power supplies which employ photovoltaic arrays and/or wind turbines. Whereas the advantages have been demonstrated, an unequivocal understanding of the means by which carbon assists the recharge of leadeacid batteries has yet to be reached. Such knowledge will provide a deeper appreciation of the functional relationship between product and required duty, which is essential for the future development and design of cells for new and emerging applications. Possible mechanisms for the ‘extra-carbon effect’ are discussed below in Section 4. 3. Types of battery configuration The configuration of the negative plate in a conventional leadeacid battery is compared in Fig. 2 with alternatives in which additional carbon has been incorporated in various ways, as follows. (i) Conventional leadeacid batteries (see Fig. 2(a)) with no additional carbon, i.e., above that typically included in the expander formulation, exhibit a sharply-declining DCA during HRPSoC operation. This restriction of the charge reaction is due to the limited solubility of lead sulfate, as described

(ii)

(iii)

(iv)

(v)

earlier in Section 1. Individual lead‒acid cells in long strings are also likely to suffer a divergence in state-of-charge during PSoC cycling at any charge/discharge rate. The simplest way to incorporate additional carbon in a conventional leadeacid battery is to mix it with the basic ingredients of the negative plate and then paste in the normal way (see Fig. 2(b)), although it must be acknowledged that supplementary water is required to maintain a satisfactory rheology. Batteries with such plates exhibit a somewhat improved DCA but small carbon particles may become isolated and lose efficacy, as described below. Axion Power International Inc [20]. offers the PbC® battery that has a negative plate with carbon as the sole active material; see Fig. 2(c). All other components are similar to those found in a conventional leadeacid battery. Given that there is no lead sulfate to limit charge-acceptance at the negative plate (the reaction involves the storage of protons, Hþ), the technology sustains DCA well. The carbon acts as a capacitor to provide not only a high degree of DCA but also to enable self-balancing of series-connected cells during PSoC cycling. The PbC® battery does, however, suffer from two disadvantages, namely, a specific energy lower than that of the conventional counterpart and a voltage that varies with the state-of-charge. The UltraBattery® d a CSIRO invention that is under commercial development by Furukawa Battery Co. (Japan), Ltd., East Penn Manufacturing (USA), and Ecoult (Australia) d has a compound negative plate in which one section consists of the usual sponge lead active-material and the other is composed of supercapacitor-grade carbon. As with the Axion PbC® design, all other components of the battery are conventional; see Fig. 2(d). In addition to providing a sustained DCA during high-rate HRPSoC operation, the UltraBattery® provides self-balancing of individual cells in long seriesconnected strings in the same manner as does the Axion PbC®.. A road test [21] of a Honda Insight medium-hybrid in which the original nickelemetal-hydride (Ni‒MH) battery had been replaced by an UltraBattery® of the same voltage (144 V) was continued for 100 000 miles at the Millbrook Proving Ground in the UK. At the end of the test, the battery was still fully-functional. Moreover, the 12 individual 12-V modules were matched together better than at the start of the test and, moreover, without the intervention of any external equalization [21]. A similar project, which involved the replacement of a Ni‒MH battery in a Honda Civic with an UltraBattery®, ran for 150 000 miles in Arizona [22,23] and again the individual modules remained fully-balanced without any electronic support. Self-balancing of the UltraBattery® has also been observed in stationary energy-storage applications by the Ecoult Company. A concept evolved by ArcActive Limited [24] has the lead grid replaced by a porous carbon material which has been ‘activated’ by an electric-arc process; see Fig. 2(e). The result is a configuration that might be viewed as analogous to the UltraBattery® but with the carbon layer under the negative active-material instead of above. Similar designs have been proposed by Kirchev et al. [25], Czerwinski et al. [26] and Firefly [28], respectively. The ArcActive cell sustains DCA well, as shown in Fig. 3, but its behaviour in long strings has yet to be reported.

4. Progress in understanding the ‘carbon effect’ In recent years, there has been much discussion over the


271

Fig. 2. Schematics of negative-plate configurations without and with additional carbon.

mechanism by which a carbon component of the negative plate can enhance the performance of a leadeacid battery. As many as eight different functions have been proposed [27] but the growing body of evidence points to just three candidates that are the most likely to have a significant individual effect; the degree of influence varies according to the design, particularly in the case of the alternatives depicted in Fig. 2 above. The situation remains complicated,

however, because all three can be operative simultaneously. The first function of carbon is to serve as a capacitive buffer to absorb charge current in excess of that which can be accommodated by the Faradaic (i.e., electrochemical) reaction; see Fig. 1. A conventional negative electrode will itself have an attendant double-layer but the capacitive function (normally in the range 0.4e1.0 F per Ah) only becomes noticeable when the surface-area is

272


both the zinc and the nickel electrode systems. By contrast, the kinetics of the electrode reactions in a lead‒acid cell are much slower, namely: 4.0 107 A cm2 at the positive plate, and 4.96 106 A cm2 at the negative [29]. With an augmented carbon inventory, the negative plate can contribute a specific capacitance of up to 30 m F cm2 whereby the time constant can be of the order of seconds and the charge equilibration (shown as reaction 3 in Fig. 4) can be regarded as a significant process that follows the removal of the external charging voltage. Under HRPSoC conditions, charge from external events, such as regenerative braking in vehicle applications, is taken up by the double-layer and thus boosts DCA efficiency. When the external input is discontinued, this charge is re-equilibrated between the double-layer and the Faradaic reaction. 4.2. Extending the conducting surface-area for the electrochemistry Fig. 3. Dynamic charge-acceptance (DCA) of the ArcActive cell is sustained through PSoC cycling in contrast to that of a conventional VRLA cell (absorptive glass-mat technology).

magnified appreciably by the addition of an appropriate form of carbon. The second effect of carbon is to extend the area of the electrode microstructure on which the electrochemical charge and discharge reactions can take place. For carbon to perform either of these two functions to best effect, it should be in an sp2 hybridized form, e.g., graphite (v.i.). The third way in which carbon can modify the behaviour of the negative active-material is by means of physical effects, for instance by obstructing the growth of lead sulfate crystals and/or by maintaining channels for irrigation of the electrode by the electrolyte. In both these cases, there is no need for the carbon to be in a conductive form. Each of the above three actions of carbon is considered in more detail, as follows. 4.1. Capacitance: current flows during and after a charge event

(i) a capacitive system, which involves high-rate charging and discharging of the electric double-layer; (ii) the conventional lead electrochemical system, which comprises the oxidation of lead to lead sulfate during discharge and the reverse process during charge. It has been found that the capacitive process dominates during cycling with charge and discharge processes each of 5-s duration, whereas and the electrochemical reactions dominate during cycling with longer pulses. These observations are consistent with the above calculation of the time constant for the transfer of charge between the two component materials of the negative electrode. 4.3. Physical processes

When a charge event applied to a leadeacid cell is discontinued, although the external current is zero, the double-layer remains charged and this results in a local current between the component materials of the negative active-mass [29]. The current is caused by the discharging of the double-layer via the Faradaic reaction and thereby the electrode potential changes to its equilibrium value with a characteristic time constant. The amount of charge involved can be substantial if the surface-area of conducting material is augmented by the inclusion of an appropriate form of carbon. The time constant, t, for the equilibration process can be estimated by equating the double-layer discharge current to the charging current that is going into the Faradaic reaction and can be approximated as:

t ¼ RTC=Fio

Cycle tests under HRPSoC conditions of cells that contain extra carbon have provided strong evidence [30] that the electrochemical and chemical processes can take place not only on the surface of lead metal but also on the surface of carbon; see Fig. 5. Subsequent research [9] has confirmed that two electrical systems are operating on carbon at the negative plate, namely:

[1]

where: R is the universal gas constant (8.3145 J mol1 K1); T is the absolute temperature (K); C is the specific capacitance (F cm2); F is the Faraday constant (96 458 As mole1); io is the exchange-current density (A cm2). Reactions that proceed relatively quickly have a small time constant, whereas those that are kinetically sluggish result in a large time constant. By way of example [29], the zinc electrode has rapid kinetics with an exchange current of around 102 A cm2 that gives rise to a time constant of the order of 50 ms. Alternatively, the nickel hydroxide electrode has a time constant of around 5 ms. Since both these time constants are so short, the involvement of the equilibration process outlined above has been largely ignored for

A widely-proposed suggestion [31,32] is that the carbon added to the negative active-material acts as a steric hindrance to the crystallization of lead sulfate and thus helps to maintain a high surface-area for the discharge product. In support of this theory, it has been observed [33] that HRPSoC cycle-life is enhanced when titanium dioxide (a poorer electronic conductor), rather than graphite, is used as the additive. Whereas cycle-life is improved by using either titanium dioxide or graphite (to different extents), the effect of the latter is not necessarily due to its conductivity alone. It has also been advocated [34] that ‘activated’ carbon increases the porosity of the negative electrode by providing an additional structural skeleton that facilitates diffusion of the electrolyte solution from the surface to the interior of the plate. As a result, sufficient sulfuric acid is supplied to keep pace with the electrode reaction during HRPSoC operation. The same work has demonstrated [34] that longer cycle-lives under HRPSoC duty are achieved with carbon of larger particle size (i.e., micron-size rather than nanometre). This observation has given rise to the view [35] that small particles of carbon can become progressively buried within crystals of lead sulfate such that their effectiveness is lost. Other studies have revealed [9] that carbon additives can alter the pore structure of the negative electrode and one consequence is that the access of SO2 4 ions to the innermost pores is impeded. On the other hand, Hþ ions may still diffuse out of the pores and allow the pH to rise locally to a value at which a-PbO is formed. This phase, which is clearly visible in X-ray diffraction


273

Fig. 4. Schematic of the positive current fluxes during and after a short charge event on a negative plate that contains carbon.

Fig. 5. Faradaic reactions of the negative plate can take place on both the lead and carbon surfaces [30]. EAC stands for electrochemically active carbon.

records, is deleterious to the continued function of the negative active-material because the formation of the oxide is irreversible. 5. Properties of carbon Carbon can be found in various forms with a very wide range of physical properties, which depend very strongly on the respective electronic properties of the atoms. The principal allotropes of the element are (i) diamond, in which the carbon atoms are sp3 hybridized so that the material is very hard and electronically resistive, and (ii) graphite, in which the atoms are sp2 hybridized and invest the material with a softer, layered structure that exhibits significant conductivity along the planes of its hexagonal structure. The physical properties of diamond and graphite are listed in Table 1. In the context of the materials present in the negative plate of a leadeacid cell, it is worth noting that the thermal conductivity of graphite is approximately four times that of lead (35.3 W m1 K1) so the presence of graphite assists heat distribution within the negative active-material. The resistivity of graphite (both parallel to and perpendicular to the basal plane, see Table 1) is greater than that of lead (2.08 107 U m). Consequently, early theories that the benefits gained by the addition of carbon are due to an improvement in the conductivity of the negative active-material appear to be groundlessdexcept when the electrode is so discharged that almost all of the sponge lead is replaced by lead sulfate.

Table 1 Physical properties of diamond and graphite. Property

Diamond

Graphite

Crystal structure Orbital hybridisation Covalent radius, pma Density, g cm3 Mohs hardness Heat capacity, J mol1 K1 Thermal conductivity, W m1 K1 Resistivity, U m

Cubic sp3 77 3.515 10 6.155 ~2200 ~1012

Hexagonal sp2 73 2.267 ~1 8.517 ~150 ~3 103 (c axis) ~4 106 (a axis)

a

picometres; 100 pm ¼ 1 angstrom (Å).

A wide range of amorphous, or poorly crystalline, substances can be prepared in which both sp2 and sp3 carbon atoms are present. Such materials exhibit physical properties that are intermediate between those of the diamond and graphite ‘end-members’, but also are strongly influenced by other parameters associated with materials. Particle size can be anywhere between a few nanometres and tens of microns, whereas surface-areas can vary from a few m2 g1 (graphite) to over 2000 m2 g1 (activated carbons and carbon blacks). Activated carbons are mainly amorphous with a fine pore structure. Carbon blacks are composed of agglomerates of interconnected clusters within which there are regions that are ordered and have the graphite structure. In the

274


absence of other factors, electrical conductivity is likely to follow the sequence graphite > carbon blacks > activated carbons. The surfaces of these materials, however, can accommodate a range of atoms or groups of atoms [36] that exercise considerable influence on properties such as wettability, double-layer formation, and chemical reactivity. Impurity levels are also important. Depending on the production process, industrial carbons can contain up to 10,000 ppm of ‘foreign’ elements that can include various amounts of iron, nickel, copper, zinc, silicon, potassium, and sulfur. In view of the need to restrict the evolution of hydrogen during charging, it is, of course, particularly important to prevent or minimize the presence of those impurities that would promote such gassing. 6. The best choice of carbon? Empirical evidence suggests that there are at least three ways by which the presence of carbon can modify the performance of the negative plate of a leadeacid battery, namely, via (i) a capacitive contribution, (ii) the extension of the surface-area on which the electrochemical charge and discharge processes can take place, and (iii) physical processes. The capacitive process (i) is favoured by carbon that has large surface-area, is conductive, and is in contact with the currentcollector (grid) via the ‘skeleton structure’ sponge lead that is present at all states-of-charge. It is not necessary, however, for the carbon to be mixed intimately with the sponge lead component of the negative electrode. The surface-area effect (ii) also requires the carbon to be conductive and in contact with the current-collector. On the other hand, given that carbon promotes a bulk rather than a surface process, the surface-area can be less than that for the capacitive process. Carbon deployed to take advantage of the physical processes (iii) does not have to be conducting, but it must be intimately mixed with the sponge lead and should not be very finely-divided or its efficacy will wane over time. In view of the conflicting requirements for the carbon to function in these several ways, it is not surprising that workers who are seeking to optimize the HRPSoC performance of lead‒acid batteries have resorted to evaluating combinations of different types of carbon. Whilst this experimental approach may be beneficial in the short term, a thorough investigation of the mechanism of the electrochemical influence of added carbon is desirableda complete understanding would encourage further enhancement of the performance of lead‒acid batteries in a variety of energy-storage applications. References [1] E. Karden, P. Shinn, P. Bostock, J. Cunningham, E. Schoultz, D. Kok, J. Power Sources 144 (2005) 505e512. [2] H. Budde-Meiwes, D. Schulte, J. Kowal, D.-U. Sauer, R. Hecke, E. Karden, J. Power Sources 207 (2012) 30e36. [3] J. Kowal, D. Schulte, D.-U. Sauer, E. Karden, J. Power Sources 191 (2009)

42e50. [4] J. Kowal, Spatially-resolved Impedance of Nonlinear Inhomogeneous Devices Using the Example of Lead‒acid Batteries, RWTH Aachen University, Institute for Power Electronics and Electrical Drives (ISEA), 2010. PhD Thesis. [5] D-U. Sauer, E. Karden, B. Fricke, H. Blanke, M. Thele, O. Bohlen, J. Schiffer, J.B. Gerschler, R. Kaiser, J. Power Sources 168 (2007) 22e30. [6] M. Thele, J. Schiffer, E. Karden, E. Surewaard, D.-U. Sauer, J. Power Sources 168 (2007) 31. [7] K. Peters, A.I. Harrison, W.H. Durrant, in: D.H. Collins (Ed.), Power Sources 2, Pergamon Press, UK, 1970, pp. 1e16. [8] P.T. Moseley, D.A.J. Rand, B. Monahov, J. Power Sources 219 (2012) 75e79. [9] D. Pavlov, P. Nikolov, J. Power Sources 242 (2013) 380e399. [10] O. Scarpa, Chimie Ind. 17 (408) (1927) 408. [11] E. Willihngantz, Trans. Electrochem. Soc. 92 (1947) 281. [12] E.J. Ritchie, Trans. Electrochem. Soc. 92 (1947) 229. [13] E.J. Ritchie, J. Electrochem. Soc. 100 (1953) 53. [14] C. Zachlin, J. Electrochem Soc. 98 (1951) 325. [15] H. Bode, in: R.J. Brodd, K.V. Kordesch (Eds.), LeadeAcid Batteries, John Wiley & Sons, Inc., New York, USA, 1977, p. 243. [16] K. Nakamura, M. Shiomi, K. Takahashi, M. Tsubota, J. Power Sources 59 (1996) 153e157. [17] M. Shiomi, T. Funato, K. Nakamura, K. Takahashi, M. Tsubota, J. Power Sources 64 (1997) 147e152. [18] A.F. Hollenkamp, W.G.A. Baldsing, S. Lau, O.V. Lim, R.H. Newnham, D.A.J. Rand, J.M. Rosalie, D.G. Vella, L.H. Vu, ALABC Project N1.2, Overcoming Negativeplate Capacity Loss in VRLA Batteries Cycled under Partial-state-of-charge Duty, Advanced LeadeAcid Battery Consortium, Research Triangle Park, NC, USA, 2002. Final Report, July 2000 to June 2002. [19] A. Simon, S. Caulder, J. Electrochem. Soc. 117 (1970) 987e992. [20] Axion Power International Inc., 3601 Clover Lane, New Castle, PA 16105, USA. http://www.axionpower.com. [21] A. Cooper, J. Furakawa, L. Lam, M. Kellaway, J. Power Sources 188 (2009) 642e649. [22] P.T. Moseley, D.A.J. Rand, Global warming and leadecarbon batteries, in: 14th European Lead Battery Conference, 9e12 September 2014 (Edinburgh, UK). [23] B. Monahov, Hybrid electric vehicles d challenge and future for advanced lead‒acid batteries, in: 9th International Conference on LeadeAcid Batteries, LABAT’2014, Albena, Bulgaria, June 2014, pp. 10e13. [24] J. Abrahamson, S. McKenzie, S. Christie, E. Heffer, H. Out, G. Titelman, H. Wong, Properties of full-scale leadeacid negative plates built around carbon felt micro-scale current collectors, in: 14th European Lead Battery Conference, 9e12 September 2014 (Edinburgh, UK). [25] A. Kirchev, L. Serra, S. Dumenil, A. de Mascarel, G. Brichard, M. Alias, L. Vinit, M. Perrin, AGMeVRLA batteries with carbon honeycomb grids for deepcycling applications, in: 9th International Conference on LeadeAcid Batteries, LABAT’2014, Albena, Bulgaria, June 2014, pp. 10e13. ski, Z. Rogulski, J. Lech, J. Wro bel, K. Wro bel, P. Podsani, [26] A. Czerwin M. Bystrzejewski, New high capacity leadeacid battery with carbon matrix. Carbon leadeacid battery (CLAB), in: 9th International Conference on LeadeAcid Batteries, LABAT’2014, Albena, Bulgaria, June 2014, pp. 10e13. [27] P.T. Moseley, J. Power Sources 191 (2009) 134e138. [28] Firefly International Energy Co., 6533 North Galena Road, Peoria, Illinois, USA 61614. [29] V. Srinivasan, G.Q. Wang, C.Y. Wang, J. Electrochem. Soc. 150 (2003) A316. [30] D. Pavlov, T. Rogachev, P. Nikolov, G. Petkova, J. Power Sources 191 (2009) 58e75. [31] P. Ba ca, K. Micka, P. Krivík, K. Tonar, P. Toser, J. Power Sources 196 (2011) 3988e3992. [32] K. Micka, M. Cal abek, P. Ba ca, P. Krivík, R. Labus, R. Bilko, J. Power Sources 191 (2009) 154e158. [33] P. Krivík, K. Micka, P. Ba ca, K. Tonar, P. Toser, J. Power Sources 209 (2012) 15e19. [34] J. Xiang, P. Ding, H. Zhang, X. Wu, J. Chen, Y. Yang, J. Power Sources 241 (2013) 150e158. [35] K. Kogure, M. Tozuka, T. Shibahara, S. Minoura, M. Sakai, Development of leadeacid batteries for idling stopestart system (ISS) use, in: 9th International Conference on LeadeAcid Batteries, LABAT’2014, Albena, Bulgaria, June 2014, pp. 10e13. [36] H.P. Boehm, Carbon 32 (1994) 759e769.