No soft porn in these pages as Mike McDonagh returns to explain the trials and tribulations of lead oxide production.
Lead Oxides
The methods used to turn lead from its inert metallic state into a reactive material suitable for cyclic battery manufacture have been with us for decades. Yet we still have an incomplete understanding of the reactions and processes occurring between the production of oxide and the formation of the battery plates.
Because our processes have largely developed over time, they are mostly a response to the problems encountered when we wanted to make changes. In other words, we have production methods and designs which we know work, (mostly), but which have evolved over time rather than as a result of scientific or technical design.
The good news is that within the last couple of decades, targeted research has been carried out, predominantly helping us to understand the chemical and physical processes which we use to manufacture our positive and negative active material.
Within our limited understanding of the processes, there is certainly a lot more grey than there is black and white. This goes beyond referring to the colour of lead sub oxide. It applies to our overall understanding of how the materials behave during these processes and it also applies to the effect of process parameters on the performance of the final product.
The basic problem is that there are so many variables within each sub process of the manufacturing operation, the consequences of small alterations or variability in operating conditions are difficult to predict.
This is the area of lead-acid battery manufacturing where a lot more can go wrong than can go right. If you have ever been next to a 30 tonne grey oxide silo glowing red as its contents obligingly convert its remaining free lead into lead oxide, so you won’t have the bother, you will know that we are not yet truly in control of these processes.
Equally, mysterious viral ailments, such as sudden loss of cold cranking, pcl effects, low hydrogen overvoltage in agm, poor shelf life, etc.can result from small variations during processing.
Lead oxides have been used in lead-acid batteries as starting materials for plates from the early part of the last century. Lead monoxide (PbO) and red lead (Pb3O4) were used extensively, mostly as a mixture. The development of the Ball Mill and Barton Pot processes for oxide production considerably reduced the cost of lead monoxide. The downside however, was the inefficiency of the process, leaving high levels of unreacted free lead in the resulting leady oxide dust. New processes evolved to remove the free lead before the formation stage in order to improve the efficiency of the charging process.
The advantages of low production cost and universal adaptation to all battery types meant that some additional treatment was found to reduce the lead content before formation. The current curing and paste mixing processes were the result of this.
The main disadvantage, other than the additional processes, is the variability in the product. Normally, between 25 and 35% of unreacted free lead is remaining in the oxide. This variation can be evident within the same shift, the same working day and from day to day. It is at this point, due to the variability in the manufacturing processes, that the consequential variability of the semi -finished product begins to have an effect on the subsequent paste mixing and curing operations.
There are many variables to control within the processes, but intensive research by organisations and laboratories worldwide is making the processes and resultant physical and chemical structures of the plate active masses better understood. The challenge for the battery manufacturer is to ensure that they have the technical and engineering capability to take advantage of this increasing body of knowledge.
As discussed in the previous article, lead has the capability of forming stable oxides over a range of stoichiometric ratios and different allotropic forms. It is becoming increasingly evident that the individual properties of these different forms can affect the performance of batteries, either beneficially or adversely, depending upon the application. Although both Ball Mill and Barton Pot methods produce predominantly lead monoxide and unreacted free lead in similar proportions, they produce very different behavioural characteristics in the following processes and battery performance. Even within the same process these chemical and physical characteristics can be varied by altering the process parameters. It is therefore critical to ensure that the effects of operating conditions during manufacture are understood and controlled within realistic limits.
The most common forms of lead oxide used for lead-acid battery manufacture are leady oxide, in which lead is partially converted to PbO and contains around 20–30% of unreacted or free lead; litharge PbO2 and red lead or minium Pb3O4.
The Barton Pot and Ball Mill methods of production tend to produce forms of lead oxide specific to that process, due to their different operating conditions. Table 1 shows the free energy of formation for the 3 principal oxide forms, PbO, PbO2 and Pb3O4. This has been calculated from thermodynamic data, assuming the following chemical processes for lead oxidation:
It is evident that the lowest formation energy is that of PbO which is formed in both of the popular methods. By increasing temperature and or airflow, PbO2 and Pb3O4 can also be produced in limited amounts by the Barton Pot method, but not by Ball Milling. In fact, Ball Mill oxide can be used as feedstock for the Barton Pot reactor to produce litharge and red lead. However, there are companies who include a second stage of either calcining Barton Pot oxide to produce higher oxides of minium and litharge, or more commonly, a second grinding process from Barton Pot through a Ball Mill or other attrition process. This reduces particle size and increases the reactivity of the leady oxide.
This can be a particularly useful method for the manufacture of oxides suitable for traction and AGM positive plates, which will be discussed later. In general terms it can be seen from Fig. 1 that the energy of formation of Pb3O4 is closer to that of PbO2 than that of PbO. This can reduce formation energies for traction positive plates, particularly dry filled tubular positives. This, and the benefits of a smaller particle size for material utilisation are reasons why red lead is often used in varying proportions in the positive active mass.
The majority of lead-acid battery companies, for all product types, manufacture the leady oxide and buy in red lead for blending where required. The most common and popular methods of producing oxide are the Barton Pot and the Ball Mill processes. Both of these use air to react with lead to give dust particles predominantly consisting of lead monoxide. However, the reaction is incomplete and the particles formed contain a high proportion of unreacted (free) lead. The lead dust produced is called leady oxide or lead suboxide. A simplified equation for the reaction can be given as
3Pb + O2 = 2PbO +Pb
Although the reaction is the same for both methods, that is, air oxidation of refined lead, the process routes are very different. This leads to differences in chemical and physical properties which have a profound effect on the subsequent processes and the performance of the battery. It is very important to understand the process routes and the consequences of variation in operating conditions. Knowledge of these factors will assist in both optimising the oxide properties for battery performance and avoiding the dangers associated with a freshly produced reactive material. It is best to examine each process in turn and discuss the optimum process conditions before making comparisons between the processes.
Barton Pot
Process
In the Barton Pot process, molten lead is poured from a melting pot into a reactor vessel. A rotating paddle agitates the lead to maximise exposure to the air. This action probably forms droplets caused by a splashing action resulting from the action of the paddle stirrers on the molten lead surface. Fig. 2 is a photograph of the inside of reaction vessel when not operating. This reactor processes around 15 tonnes per day.
The agitated, molten lead reacts with the oxygen in the air to form lead oxide. This is essentially splashing, which leads to particle shapes which are roughly spherical due to surface tension forces within the molten lead droplets (Fig 3). The particles are not completely oxidised, the centre of the particles containing mostly unreacted free lead. The degree of oxidation is largely an issue of economical throughput. The lead oxide powder is removed from the reactor vessel using an air extraction and filter system.The oxide
particles have to travel up the almost vertical exhaust duct against gravity in order to exit the reaction pot. This process acts as a particle size classification system. Particles of certain sizes will have enough velocity to get to the top of the duct leaving the reaction vessel, whereas the larger size particles will return to the reaction vessel until they are small enough to be removed. Fig 4 shows a schematic of the operation.
Properties
The round shape of the particles results in a lower surface area than that of oxide from Ball Mills. Consequently this produces an oxide with lower reactivity and lower absorption values than that of Ball Mill oxide.
An interesting line of investigation would be to look at additives in the metal at low concentrations, which could reduce the surface tension forces and perhaps produce more irregular, flatter particle shapes. If anyone has any information on any such tests, I would be grateful to know what the results are.
Table 2 compares the physical and chemical properties associated with the Barton Pot leady oxide with that of Ball Mill oxide.
Process variables
As can be seen from the equation on page 110, varying the conditions of lead loading and air flow can influence the ratio of Pb to PbO. Normal practice is to provide 22 to 28% free lead in the finished oxide. Higher lead feed rate will increase the free lead content.
By varying the operating conditions of the reactor, not only the degree of lead oxidation, but also particle size distribution and the ratio of αPbO to βPbO in the product can be modified.
Although the higher temperatures of the Barton Pot reactor favour mostly formation of βPbO, there is some degree of control over the ratios of αPbO to βPbO. A higher reaction temperature increases the amount of βPbO, reducing the temperature increases the ratio of αPbO to βPbO. Due to the importance of these process parameters, the reactor temperature, air flow rate and lead input are monitored and recorded on a chart for quality control purposes. Table 3 gives a summary of the operating conditions and their effect on the material properties.
One of the key differences between Ball Mill and Barton Pot oxide is the shape and size of the particles. Fig 5 compares particle size distribution of Barton Pot and Ball Mill oxide. Although the particle size and surface area of Barton Pot oxide produce a material less active than that from Ball Mill oxide, there are methods of improving the properties of Barton oxide. These are based on attrition processes which grind the oxide into finer particles, thereby increasing its surface area and increasing its reactivity and absorption properties.
An important physical property of the oxide is particle size distribution. This influences both battery performance and life. Whilst the Barton process is capable of a great degree of flexibility through process parameter adjustments, addition of a further process using a hammer mill would provide greater control of particle size. It is also reported that this process, if suitably controlled, can induce a phase transformation and increase the proportion of αPbO. The reduction in particle size increases the surface area. This improves the reactivity allowing increased water and acid absorption as well as improving cold cranking performance.
This combination of processes would give a great deal of control over the chemical and physical attributes of the oxide and should provide a higher degree of consistency than is currently experienced in most lead-acid battery factories.
Quality monitoring
Critical is the control of the reactor parameters which should be recorded at least every 30 minutes. Temperatures and flow rates should not vary by more than 5%. If automatic recording and control is not available then monitoring using appropriate statistical methods, and making appropriate adjustments, is essential.
Oxide quality must also be checked at least twice during the shift. The simplest method is to check the free lead content using an acetic acid solution diluted with demineralised water. Accurately weigh out about 10 grams of oxide into a conical flask, then pour in an excess of the solution (half a litre is sufficient). Boil this on a hot plate for 30 minutes, then filter. There should be a small button, perhaps slightly porous, of unreacted lead left in the filter paper. Wash this with demineralised water then ethanol and allow it to dry for several minutes. Weigh the button to obtain the free lead content. This should be around 22 to 28% consistently. This operation is easily carried out by the machine operator.
Less routine monitoring will include particle size analysis, acid and water absorption for which there are standard methods. These can be done weekly or monthly for peace of mind rather than an essential check. The degree of free lead remaining in the oxide is a reasonably accurate indicator of the other chemical and physical properties.
Ball Mill
Process
As with Barton Pot, the starting point is the melting of refined lead ingots in a pot. The molten lead is fed continuously into a rotary caster which produces small cylinders of lead around 15-20 cm in diameter and height. These nuggets are stored in a silo then fed into a rotating drum with an air stream. As the lead nuggets are tumbled, the surfaces are oxidised. This is an exothermic process which mostly provides the heat to drive the oxidation reaction. The oxide layer, which is more brittle than the lead substrate, is abraded through friction and flakes off the nuggets.
The oxidation of the nuggets is the result of attrition friction and the heat generated by the exothermic 2Pb + O2 = 2PbO and the air flow. Operating conditions are normally within the temperature range of 120-145°C and a constant airflow.
Whether oxide particles result from lead flakes produced from the attrition action of the lead nuggets, or, in fact, are formed on the ingot surface, then removed by abrasion, is not entirely clear. Most likely both mechanisms occur, with dislodged oxide particles being further ground and oxidised during the tumbling action. Once the oxidised flakes are small and light enough they are carried in the air stream out of the drum chamber to filters then stored in silos. The process schematic is shown in Fig. 6.
Properties
This method of oxide production results in flakes with a high aspect ratio (flat and long) rather than spherical. As with Barton oxide the oxidation is not complete and up to 30% of the particles can remain as unreacted lead. Due primarily to shape factor of the oxide particles this leady oxide has a higher surface area and is generally finer than Barton oxide.
Additionally, due to the lower operating temperature, the oxide consists almost entirely of αPbO. The finer particle size and higher surface area also mean that Ball Mill oxide is more reactive than that of Barton Pot. The acid and water absorption values are higher.
Table 2 shows the physical and chemical properties normally found with Ball Mill and Barton Pot grey oxide. It is evident that this method produces a more reactive oxide with a greater surface area. This has an important influence on the subsequent processes and performance of the battery.
Process variables
The principal parameters to control are the airflow and the loading weight of the drum. During the process the rotating drum, which is normally on load cells, is continuously fed with nuggets from a storage silo. Once the predetermined weight is reached, the nugget addition ceases.
Increasing the drum weight can lead to higher temperatures and increased free lead content. Increasing the air flow will decrease the free lead content but increase the average particle size. The leady oxide produced in this way will always be virtually 100% the α form of the oxide due to the lower operating temperatures.
Another process variable which is difficult to control is the air humidity. The reaction of leady oxide with air is catalysed by moisture. During start up and for the first hour of operation, there is an increased risk of initiating an exothermic reaction within the oxide mass. Anyone having witnessed a 50 tonne silo glowing red for a couple of days due to this particular effect will understand the importance of keeping all variables under control. Fig. 7 is an example of a modern, well controlled system, with continuous, automatic, monitoring and control of all parameters.
Quality monitoring
This will be virtually identical to those methods described for Barton Pot oxide. However, temperature monitoring of the storage silos is also advisable.
Production of red lead
The use of leady oxide is common to almost all lead-acid battery companies, but the use of red lead, in positive plates at least, is still normal practice in the traction and AGM markets. It is unusual however, for a lead-acid battery factory to manufacture red lead. It is more normal to buy in the powder and blend it with internally produced leady oxide. For this reason the description of the production methods will be brief. The most common method is calcining of Barton Pot leady oxide in a reaction vessel at fairly high temperatures.
Both PbO and red lead are produced in calcining furnaces, in which the raw material is agitated while being heated at the optimum temperature for oxidation to the required product. For PbO, this is around 600°C. For red lead, 450°C to 500°C is used. For red lead, the furnace is discharged when the specified level of Pb3O4 is reached. The longer the time in the furnace the higher the red lead content.
Effect of oxide properties on subsequent processes and battery performance
Of the three forms of lead discussed, it is the leady oxide Pb/PbO which undergoes the most modification before ending up in the final product. The majority of lead-acid batteries consist of a negative and positive plate containing leady oxide. Of these, all have negative plates which have undergone paste mixing and curing. There is a difference between flat plate positives, which are treated similarly to the negative, and tubular traction positives which are either dry powder or slurry filled.
Table 4 gives a picture of which oxide types are popularly used in the different battery categories. As a general rule SLI batteries perform better with the α modification of PbO. Traction batteries have enhanced life when higher proportions of β PbO are present in the oxide. The reasons for this will be briefly discussed in this article, but more fully examined in the next article dealing with the two critical processes of paste mixing and curing.
Paste mixing and curing
The first stage of paste mixing is to mix the oxide with water then add acid in a controlled flow. The overall reactions are probably as follows:
4PbO + H2SO4 =
PbSO4.3PbOH2O
(tribasic lead sulphate) <70°C
PbSO43PbOH2O + PbO = PbSO4.4PbO + H2O
(tetra basic lead sulphate) >70°C
The chemical nature of the lead sulphate in the paste is important since it affects the plate formation process, its electrochemical performance and its durability under discharge/charge cycling. Tribasic is preferred for engine-starting battery plates as is readily converted to PbO2 during formation and confers high initial performance.
Tetrabasic is considerably more difficult to form, particularly in sulphuric acid solutions of concentrations greater than 15%, but it contributes to the formation of a strong and stable morphology in the crystal structure of the plate. This gives improved cycle life and deep discharge capability suitable for traction and deep cycle stationary applications, including AGM. In battery types where tetrabasic is preferred, it is more common to produce this during the curing process than during paste mixing.
Leady oxide made by the Barton process contains up to 10% orthorhombic (β) PbO. It has lower stability than tetragonal (α) modification and is partially converted to the tetragonal phase during mixing. The finished paste, at the point of completion of pasting, will contain the following compounds:
αPbO, βPbO, PbSO4.4PbO, Pb(OH)2, PbSO4.3Pb H2O, H2O
The content of the orthorhombic form of PbO would be insignificant in the case of paste made with Ball Mill oxide.
The amount of βPbO is known to influence the crystal structure of the sulphates formed during the curing process. It is also possible that it may increase the amount of tetrabasic lead sullphate formed during pasting. It could promote formation of tetrabasic sulphate below 70°C. However, the overriding influence on the basic sulphate structures formed during paste mixing has been shown to be temperature and oxide/acid/water ratios.
One very significant factor to be considered and one which has very practical consequences is the water absorption values of the two oxide forms. Barton Pot has a lower value and it is difficult to get a usable paste with moisture contents above 8%. Subsequent flash drying and storage before the curing process can ensure that the moisture levels drop significantly below 7% in the wet plates. This can have serious consequences for the curing process, particularly where natural curing is practised. This will be discussed in more detail in the next article.
Red lead and litharge
These are usually blended as separate oxides mostly in the dry state with leady oxide, predominantly for positive plate production. In the case of industrial tubular plate construction, red lead is blended in various amounts from 15 to 50% by weight, sometimes with the addition of around 10 to 15% litharge. This has two benefits: it reduces the formation times and gives higher initial capacities and also promotes better filling densities with fewer problems such as voids and density strata effects. This is also found to be true for wet filling systems, although, in these cases, a choice can be made between rapid drying and curing. In the latter case care must be taken to ensure that the oxide blends promote the right sulphate structure for cyclic duty. The benefits are the same for flat plate applications, ensuring lower formation energy and giving better initial capacities and longer cycle life.
This article has attempted to show that the process of oxide manufacture can significantly increase the reactivity of lead and that this has to be done under controlled conditions with frequent monitoring to ensure consistency. It is also possible, by controlling these conditions, to influence the properties of the material in order to improve the performance of final products and to minimise problems in subsequent processes. The next article will examine the methods used to optimise battery performance by control of the parameters and of the choice of materials used in the paste mixing, plate production and curing processes.
