David Boden takes a closer look at Carbon atomic number 6, the key to solving the problems of negative plate sulfation.
Carbon black has been used in the negative plates of lead acid batteries for many years. It is used as a component of the expander and the amount used is generally in the range of 0.1% to 0.2% of the negative active material weight.
There has been much speculation over the years about the role of this material but it is generally believed that it gives the pasted, cured plate improved conductivity to aid in initial formation. At the commencement of formation the negative plate is essentially composed of lead sulfate, which has very poor electrical conductivity. The presence of carbon is supposed to aid in the initial conversion of lead sulfate to lead thereby providing a conductive matrix for further conversion. Once a conducting network of lead is produced the carbon plays a much reduced role in formation or in subsequent battery operation. Carbon black is also a useful colorant for the negative plates to help distinguish them from positive plates during the assembly process. The belief that a low concentration of carbon in the negative plate is all that is required has prevailed for many years but recent work has indicated that the role of carbon black may be more complex, and important, than was previously thought.
A re-examination of the role of carbon has occurred recently as an outcome of work to develop lead acid batteries for hybrid and electric vehicles. In these applications, the battery is cycled in a partial state of charge which results in sulfation of the negative active material to the point where it no longer will accept charge. In consequence the battery progressively loses its capacity. In recent work to overcome this problem it has been shown that increased amounts of carbon black in the negative plate considerably increase the life of the battery. It has also been shown that graphite gives similar benefits. These findings have given rise to a great deal of research on the true role of carbon (and graphite) and also new ideas on the mechanisms by which these improvements occur.
There are hundreds of different types of carbon black and graphite available for investigation, and since research has only recently begun little is known about which properties are necessary to produce the most beneficial results. Therefore the first challenge will be to choose the right carbon black from the wide choice available. The choice can be narrowed by understanding their principal properties so that intelligent decisions can be made rather than the trial-and-error methods used until the present time.
Carbon black is a conductive material with an intrinsic resistivity of about 0.01 – 0.1 ohm.cm. It can be obtained as powder or pellets ranging in size from a few microns to a few millimeters. When pelletized carbon black is used force must be applied to break down the particles and high shear forces are necessary to separate the primary particles into aggregates of 20 – 100 nanometers. These aggregates consist of a few thousand crystallites enabling electrical conductivity. This is the reason why ball milling is the primary method used for expander manufacturing. However care must be taken to avoid over-milling since this can break down the crystallites and reduce conductivity.
If it is assumed that the principal benefit of carbon black is conferring conductivity in the plate we need to consider how its chemical and physical properties affect conductivity. In this regard, there is considerable data since carbon blacks are widely used in the rubber and plastics industry for this purpose. The mechanism by which carbon produces conductivity in a non-conducting matrix is well known and will be reviewed briefly.
When the concentration of carbon in the non-conductor is low, the aggregates are separated and electron transfer is inhibited. As the concentration of carbon is increased the inter-aggregate distance becomes short enough for electrical conductivity to begin. The aggregates do not need to touch since electrons can jump between the aggregates due to the tunneling effect. No increase in conductivity takes place until the concentration is high enough for tunneling to begin. Therefore there is a threshold level necessary to produce some conductivity. The point where an observable increase in conductivity takes place is called the “percolation point.” Any further increase in carbon black concentration beyond the percolation point results in a rapid increase in conductivity. At a high concentration the matrix becomes saturated and no further increase in conductivity takes place. The conductivity (resistivity) versus concentration curve has a characteristic shape as shown in Figure 1. Note that these percolation points usually occur at carbon black loadings considerably higher than are currently used or have been proposed for lead acid batteries.
The level of conductivity is dependent on several factors, the most important of which are:
• The choice of carbon black
• Surface area
• Structure
• Surface chemistry
• The mixing process
• Wettability
• Dispersivity
• Shear
It is valuable to review each of these factors to understand how they affect the behavior of the carbon black in a non-conducting matrix.
A carbon particle that has a large surface area and high microporosity is capable of contacting or wetting a relatively large amount of active material. In general, the higher the surface area the lower the concentration needed to reach the percolation point. The higher the surface area (very small particle size) the greater the number of aggregates per unit weight, and the greater the amount of shear required to obtain good dispersion and wettability. Figure 2 shows the effect of surface area on the electrical conductivity of two carbon blacks. For a given electrical conductivity a considerably reduced amount of the high surface area material is required.
The term structure is used to characterize the tendency of the carbon crystallites to agglomerate into branched or fibroid structures. A schematic representation of this is shown in Figure 3. The length of the aggregate chains and the number of branches affect both surface area and porosity. A high pore volume results in more aggregates per unit weight, hence less carbon black is needed for electrical conductivity. The branched structure prevents close packing of the aggregates which makes them easier to disperse. This is of importance in expander manufacturing and in paste mixing. It should be noted that highly branched carbons are more susceptible to shear. A certain amount of shear is necessary to properly disperse the material in the non-conducting matrix but too much shear can result in breakdown of the aggregates into smaller fragments resulting in a reduction in conductivity. Therefore different carbon blacks will require different dispersing conditions to develop their optimum performance. Both too little shear and too much shear should be avoided.
In the carbon industry the degree of structure is determined by the dibutyl phthalate (DBP) absorption method and is expressed as milliliters DBP per 100g carbon black. A high DBP number denotes a high pore volume and thus a high structure.
Carbon black suppliers provide information on the structure and surface areas of their materials. An example of one supplier’s information is shown in Figure 4.
During the production of carbon black, oxygen-containing groups like lactones, quinones, phenolics and carboxylates can be formed on the surface. These are known as “volatiles” and can act as a barrier to electron tunneling, hence hampering conductivity. Carbons with low volatiles content are generally preferred in electrical conductive applications.
A wettable material will allow more carbon to wet the surface of the non-conducting substrate before carbon-to-carbon interactions take place. This results in a higher percolation threshold, i.e. higher carbon black loadings are necessary for electrical conductivity. A parameter to judge compatibility is surface tension. Carbon black has a high surface tension which increases with high surface area. The surface tension of lead sulfate is not known therefore the compatibility between lead sulfate and carbon cannot be assessed. Nor do we have any knowledge of how wettability is affected by wet mixing of the materials as is the case during paste mixing. High shear forces, and the duration and residence time at the force level, are the predominant factors that determine good dispersion in a mixture. When other than optimal conditions are used, an irreversible structure breakdown can occur leading to a reduction in conductivity. This is more important to the expander manufacturer than the battery maker since the shear forces will be much greater during the ball milling process.
Figure 4 shows a summary of the effect on electrical conductivity of the chemical and physical properties that have been reviewed.
We now need to consider which of the properties described above are important in the performance of the negative plate and how these properties can be used to select carbon blacks for optimum behavior. Since, until recently, virtually no research had been done on the effect of carbon black on negative plate performance, we can only theorize why it has beneficial effects.
Let us consider the processes taking place in the negative plate during cycling. A formed negative plate is composed almost entirely of sponge lead with a small amount of lead sulfate, usually about 5%. Embedded in the sponge lead are crystals of barium sulfate and carbon black from the expander. The usual concentration range of these materials is 0.5%-0.8% and 0.1%-0.2% respectively. Lignosulfonate is also present, which is strongly absorbed on the formed negative active material surface. When the plate is discharged the lead sponge is oxidized to lead sulfate and the active material is converted from a conductor into a non-conductor. As the depth of discharge of the plate increases, its electrical conductivity decreases.
Pb + SO4- PbSO4 + 2e
Conductor Non-conductor
A supply of sulfate ions from the electrolyte is required to support this process. At low to moderate discharge rates the distribution of lead sulfate in the plate is relatively uniform because diffusion of sulfate ions into the active material does not limit the reaction. Under these conditions, lead sulfate nucleates on the barium sulfate crystals that are distributed uniformly in the active mass. At high rates of discharge, the reaction becomes diffusion controlled and becomes localized at the surface where the concentration of SO4- is highest. Under these conditions the concentration of lead sulfate exceeds the concentration of barium sulfate resulting in the formation of new crystal nuclei. Consequently, the concentration of lead sulfate is much higher at the surface of the plate than in the interior. This is the principal reason for the buildup of the dense lead sulfate films on the surface of the negative plate that have been observed during partial-state-of-charge cycling. In partial state of charge operation the plate is always partially discharged, a considerable fraction of the lead sulfate crystals do not get charged, and they grow due to Ostwald ripening. The result is a high concentration of non-conducting lead sulfate with a low concentration of carbon black particles embedded in it. Since the usual concentrations of carbon black used in negative plates are well below the percolation point and the surface film is non-conductive, except where it comes into contact with the grid, the conditions for effective charging are poor.
Since the charging process requires a flow of electrons, electronic conductivity is essential and the reaction can only occur at sites where lead sulfate is in contact with an electronic conductor. High levels of carbon black can provide this conductivity and allow charge to take place.
We should also be aware that the charge process proceeds through a solution-precipitation mechanism:
PbSO4 Pb++ + SO4- + 2e Pb + SO4-
Dissolution Precipitation
Dissolved lead ions will be absorbed on to the surface of the carbon particles and will penetrate into the carbon microstructure. This places them in contact with a conductor facilitating their reduction to lead. This process will be enhanced by carbons with high
surface area and high internal pore structure. However, sufficient carbon must be present to allow electron transfer through the negative active material. The concentration of carbon should be at or near the percolation point. This is generally somewhere in the range of 1%-2% for the carbon blacks most frequently used in expanders. As lead sulfate becomes reduced to lead the conductivity of the active material increases and charging is further facilitated.
This mechanism suggests that the most important properties of the carbon black for facilitating negative plate recharge are:
• High surface area
• High structure
• High porosity
• Low volatile content.
Improvement in the conductivity of the negative active material can be expected to yield a number of benefits in lead acid batteries and these should not be confined to partial state of charge operation. Increased conductivity should improve formation charge efficiency and charge acceptance in general. It should be particularly beneficial in recharging from deep discharge conditions. It may also improve the cold cranking performance of automotive batteries since this is limited by the performance of the negative electrode.
An example of the effect of high carbon black loading on the initial capacity of VRLA cells is shown in Figure 6. These cells were designed to be strongly negative limited to demonstrate the differences in the negative plate performance. The cells were formed and then cycled ten times to a depth of 100% of the 1-hour rate. The lines on the chart are the average of four cells of each type. The cells with the high concentration of carbon black in the negative plates showed significantly improved capacity and the capacity continued to increase with further cycling. The most reasonable explanation for this result is that initial formation was improved because of higher negative active material conductivity resulting from the high carbon black loading in the plate. The further increase in capacity over the first six cycles can be attributed to progressive conversion of residual lead sulfate to lead.
High dosage levels of carbon black have also been shown to improve cycle life during partial state of charge operation. In a recent study conducted by the Advanced Lead Acid Battery Consortium (ALABC) cells were cycled that contained several different carbon blacks and one type of graphite. The carbon black and graphite loading levels ranged from 0.4% to 2%. A summary of the results is shown in Table 1.
In this study cell CB1 can be considered to represent a design with a conventional amount of carbon black added to the negative active material while the other cells had significantly higher loadings. All were cycled using a protocol where they were first discharged to 50% state of charge and then subjected to a rapid high rate discharge/charge sequence simulating hybrid electric vehicle operation.
The cell with the conventional carbon black loading gave 750 cycles before failure while those with the higher loading had significantly increased cycle life. A wide range of cycle lives was obtained from the cells with the high carbon black concentrations, ranging from 1850 – 4850. This illustrates how differences between the types of carbon can affect the performance. The correlation between surface area and cycle life is poor. For example, CB3 and GR2 have the same surface area but gave 2750 and 4750 cycles respectively. The explanation for this is most probably that the cycle life is a function of several variables, including surface area, structure, wettability and shear as discussed earlier. Since all forms of carbon black have the same intrinsic conductivity, the differences observed are most probably due to differences in these physical characteristics. For the best cycle life the optimum combination of physical characteristics needs to be chosen.
The finding that graphite is also an effective additive to increase partial state of charge cycle life has led to further work on this material. It was also of interest to investigate combinations of carbon black with graphite to determine whether there are synergistic interactions. This is under investigation in an additional ALABC research project. Figure 6 shows the results of cycle tests from this work carried out on VRLA cells containing both carbon black and graphite. These were cycled using the same protocol as the previous ALABC study.
Excellent cycling performance is being obtained and it should be noted that all of the cells are still in service. It is remarkable that no loss of service capacity has been seen after over 100,000 cycles compared to 4,000-5,000 cycles from cells containing carbon black only. It is too early to claim that this is the solution to the problem of premature failure of lead acid batteries in PSOC operation since the results need to be repeated and confirmed. However, it appears to be an exciting development.
Why graphite is so effective is not known at this time. However, it is well known that it has the ability to intercalate ions between the graphene planes resulting in enhanced electrical conductivity. For example, intercalation of graphite with bisulfate ions from sulfuric acid results in an order of magnitude increase in electrical
conductivity in the plane parallel to the graphene sheets. It may be that this behavior results in enhanced conductivity in the negative active material resulting in improved charge acceptance of the lead sulfate formed during PSOC cycling. Dr. Pat Moseley of ALABC has suggested that the principal roles of the carbon are: to separate the individual crystallites of lead sulfate thereby facilitating access of electrolyte for the dissolution phase of the charge reaction and, secondly, increasing electrical conductivity during the second stage oxidation/precipitation process.
It is clear that the discovery that high loadings of carbon black and/or graphite in the negative active material offers considerable potential for improving the life of lead acid batteries in hybrid vehicle applications. Carbon black, however, comes in many varieties which differ in surface area, particle size, structure and volatile content. Each of these may have an effect on the behavior of the negative plate. Considering the likely effects of these variables it appears that the most effective materials will have a combination of high surface area and structure with a low volatile content.
Graphite, although fairly new to lead acid batteries, also seems to offer great potential to overcome negative plate sulfation during PSOC operation. It appears to be effective because of its ability to intercalate ions resulting in a significant increase in conductivity. Like carbon black, graphite is available in many forms and grades and it is to be expected that there will also be a considerable range of electrochemical outcomes from these various materials.
The author has drawn extensively from the published literature of carbon black manufacturers including S. D. Richardson Carbon Company, Inc. and Akzo Nobel Polymer Chemical Company, LLC. Grateful thanks are given to Mr. Michael Widmor and Dr. Leszek Nikiel of S.D. Richardson Carbon Company for many useful discussions and to Dr. Patrick Moseley of ALABC for permission to use data from research reports.
