|VoxSolaris: The Voice of the Sun|
|General Battery Considerations|
The common parameters
The common performance parameters used for comparing different types of batteries, in no particular order of importance, are as follows.
The term 'acceptable level' is usually stated as a percentage of the original charge capacity. However it should be noted that battery deterioration applies to most of the electrical properties, not just how much charge it can accept and return. Note also that the self discharge rate detracts from colombic and electrical efficiency but in a way that depends on the application. The stated values for colombic and electrical efficiency are often arived at by test charge and discharge cycles with little or no delay. Other factors due consideration are operating temperature range, ruggedness of construction, proneness to leaks and tolerance of abuse. Certain types of batteries are damaged by being over charged, too deeply discharged or being left in a discharged state for too long. A perfectly healthy lead acid battery for example, can be ruined in a few weeks by being left fully discharged. And certain types of lithuim ion batteries have aquired a reputation for exploding or at least catching fire if overcharged.
All batteries provide electric current by means of a chemical reaction. All chemical reactions occur because of differences in electron affinity between the reactants but only in some reactions can the difference in electon affinity be exploited to create an electric current. Many but not all of these reactions, can be run in reverse by passing in a current. And where they can, cells based on the reaction are rechargable. All battery cells are constructed as a pair of electrodes (anode and cathode) which take part in the reaction and an electrolyte to separate them. But the batteries performance parameters differ very widely depending on what chemical reaction is being exploited. The differnce between the reactants in electon affinity affects the voltage of the cells. Some reations exchange one electron while others exchange two and some rarer recation such as those involving aluminum exchange three. The available energy from the reaction is related to the product of the voltage and the number of electrons exchanged while the energy density will be directly affected by the ratio of this product to the combined atomic weight of the reactants.
There are several ways of classifying batteries according to general application area. Many of the terms used are marketing terms and are of little if any real meaning. Batteries should always be chosen on their performance parameters. There are though, broad application classes as follows:-
In addition to the above application classes there are many specialist medical, military and space applications which may require one or more parameters in extremis. A battery for a heart pacer device, for example does not need to deliver much power but it needs to last for many years on end and survive inside the body. As interesting as these areas are we are focused on more mainstream uses as these are of much greater significance in terms of energy deployment and environmental impact. In all main application classes energy capcaity and power output characteristics are either important or of paramount importance and these factors are determined primarily by the battery chemistry.
The Decline of Lead-Acid
The lead-acid battery cell comprises a cathode of spongy lead and an anode of lead dioxide. The electrolyte is a strong (6 molar) sulfuric acid solution. As the battery discharges, the lead in the cathode reacts with the sulfuric acid to form lead sulphate and hydrogen whilst the lead dioxide in the anode reacts with the sulfuric acid and the hydrogen to produce lead sulphate and water. During recharge, the lead sulfate converts back to lead at the cathode and lead dioxide at the anode while the sulfate ions are driven back into the electrolyte solution to form sulfuric acid. The cells develop an open circuit voltage of 2 volts with a favorable (near linear) voltage curve. Lead acid offers an electrical efficiency of 75 percent or better but has specific energies of only 30 to 50Whr/Kg which limited its use in vehicle propulsion. It has better specific power of some 150W/Kg and came to dominate the starter battery market, a position it still holds, by being among the cheapest on a per watt basis.
A weak point of lead acid is the physical strength of the electrodes, particularly the anode. Both electrodes are supported by a grid structure made from a lead alloy as a pure lead grid is not strong enough by itself to stand vertically while supporting the active material. The most commonly alloyed metals are antimony, calcium, tin, and selenium, none of which do much to improve the electrical properties of the battery. Lead acid batteries can be either optimized for short bursts of high output (starter batteries) or for lower but more sustained output (deep cycle). The difference lies in the construction. Normal car batteries have thin plates to maximize surface area of the electrodes while deep cycle batteries use thicker plates. The plates are weakened and deformed by the charge/discharge cycle and the deeper the cycle the worse the weakening and deformation. Car batteries are usually only slightly discharged, typically by no more than 20 percent of total capacity and in this way their life is prolonged. Deep dicharging a car battery (to 80 percent of capacity) will usually kill it within 50 cycles. By contrast, the thicker plated deep cycle battery can withstand 300 to 400 deep cycles.
Recent research not surprisingly focused on the electrodes, replacing the grid with various foam substrates, particularly carbon. This dramatically improved the specific power as the foam increased the surface area of the electrodes, giving a battery that could deliver and accept high currents and withstand deep discharge. The approach also improved the number of charge cycles but not by enough. Carbon melts at 2500°C, so is not the easiest material to bubble presurized gas through to make foam and as a consequence, the foam is expensive. As a consequence, attempts to commercialize lead acid batteries stabalized by carbon foam have enjoyed little success. Lead acid is still the starter battery market leader and is still the stalwart in low end stationary solutions but faces formidable competition from lithium ion in both these application classes. As a consequence, the battery has lost research momentum. It is likely that manufacture of lead acid batteries will wind down as electric cars become more mainstream.
The Rise of Lithium-ion
The lithium ion cells have a large number of possible configurations but typically consists of a carbon anode, a metal oxide cathode and an electrolyte of a lithium salt in an organic solvent. Lithium is the lightest of all the metals and as an alkali metal, is very reactive. These properties give lithium ion a a high energy density (150 to 200 Whr/Kg) and depending on the design, high power densities. Lithium ion batteries exhibit burst power densities of 1500W/Kg and sustain outputs of the order of 300W/Kg. Lithium ion easiliy came to dominate the mobile device market where high energy and power density are paramount and far outweighed the high cost and short lifespan (then more than $1000/kWhr and around 2 years). The short lifespan, due to a deterioration process that begins as soon as the cells are manufactured and is largely independant of use, mattered little in a market where the mobile devices themselves were rapidly rendered obsolete and where their was no other battery chemistry that could actually deliver the performance nessesary to make the devices small and light enough for mass appeal.
Since then lifespans have improved dramatically and the price has fallen. Tesla Motors report that after 10 years of use in the Tesla Roadster, most of the cells are still operating within 80 percent of their original capacity. Lithium ion now outlives lead acid and at a price of some $250/KWhr lithium ion is very comparable to lead acid on a total cost of ownership basis. With prices continuing to fall it is reasonably obvious that unless it is surpassed by another chemistry, lithium ion is set to dominate all main application classes except perhaps for grid scale batteries which we think will be dominated by Sodium Sulphur Flow Batteries .
Is there enough Lithium?
As an energy think tank we always consider how much of a global answer any given battery chemistry is. There might be a battery out there with unprecedented electrical prowess but are the materials involved abundant enough for us all to have one? If the answer to that question is no the battery should not become a cornerstone of national or international energy policy even if it has a bright commercial future. There are questions surrounding lithium in this regard. With very few exceptions, actual abundances of materials vastly exceed figures quoted as 'known deposits' because the latter is a measure of what is deemed ecomonically recoverable. Lithium is extracted from subsurface brines and some salt flats rich in lithium chloride (LiCl) but also mined from ores such as petalite (LiAl(Si2 O5 )2 ), lepidolite (K(Li,Al)3 (Al,Si,Rb)4 O10 (F,OH)2 ) and spodumene (LiAl(SiO3 )2 ). From these established sources, known deposits globally are estimated at a mere 14 million tonnes. This is about 2Kg for each person on the planet so we are not facing any issues with mobile devices which can easily outbid other application classes, but this is only enough for around a billion electric cars. By comparison, there is something like 500 times as much lithium in the sea but to extract each tonne, you would have to process over two thousand tonnes of sea salt. That isn't a cheap option and it isn't going to become a cheap option anytime soon.
A number of reviews of manufacturing processes have suggested that a future price target for lithium ion batteries of $100/KWhr is realistic. At this price, electric cars can dominate the car market without subsidies or legislative coersion. But this price is based on current lithium prices which do not appear to be sustainable. Our estimates is that lithium ion will drop to $100/KWhr but then start rising again as miners are forced to use lower quality ores and invest heavily in seeking out less accessible ores. Most geologists we have consulted consider it reasonable to expect that reserves under these conditions would be revised upwards by an order of magnitude if not more. Lithium ion is thus a viable global solution.
Instead of having cells in which all the materials are in situ, flow batteries work by pumping active materials from one resevoir to another via cells where they are reacted to produce electricity or accept charge. This arrangement, similar to a fuel cell, confers important advantages such as rock steady voltage throughout the charge/dischage cycle as there is no build up of reaction products in the cell. And subject to minimal maintenence, flow batteries exhibit no capacity reduction over time and have very long, essentially indefinite lifespans. And they can be fully discharged and left in an uncharged state without risk of deterioration. The key to longevity and thus to lower total cost of ownership, is to consider ease of maintenence in the design. Although the flow motion mitigates dendrite formation it does not completely eliminate it. Over time, dendrites will destroy the medium separating the reactants so a prudent design allows for cells to be drained off allowing the separator to be easily replaced.
Only a few battery chemistries lend themselves to flow battery operation as the reactants must be fluid. And not all of these offer the combination of good electrical performance, room temperature operation and usage of cheap and abundant materials. One battery chemistry that does is Zinc-Bromine. The battery uses carbon plastc electrodes and a microporous separator, typically a microporous polyethylene coated with sulfonated polysulfone, a cationic polyelectrolyte. During the charge cycle, zinc is deposited at the cathode while bromine is collected at the anode. When discharging the zinc at the cathode is recombined with the bromine to form zinc bromide. The electrolyte on both sides of the cell is an aquaous solution of zinc bromine but on the anode side the electroyte also contains bromine. Bromine is a liquid between the temperatures of -7C and +59C and is only slightly soluble in water. This enables bromine to be removed by fractional distillation from the cathode side electrolyte and returned to the anode side, mitigating a key problem of zinc bromine cells in which bromine escapes across the seperator reducing columbic efficiency. The cell voltage is a high quality (very steady) 1.6 volts. The electrical efficiency of zinc bromine is 75 percent and the energy density is 75 Whr/Kg or better. Power densities vary from around 20 W/Kg to around 60 W/Kg.
As is obvious, solar panels only work when the sun shines and wind turbines only work when the wind blows. And if you want power whilst becalmed on a cloudy day or at night, you will either have to draw it from the grid, generate it or draw it from storage. If you are connected to a grid you can sell power when you have a surplus and buy power when you have a deficit, in effect you can use the grid as a storage medium. And in many cases it can be very economic to do this. In jurisdictions with net metering laws the grid is forced to act as storage facility free of charge. Other tarriffing arrangements were even more absurd with governments mandating payments for solar generated electricity way above prevailing grid prices, let alone wholesale prices. Going forward, it is safer to assume you won't be selling at a premium to wholesale whereas you will be buying at retail and this gives you a budget for a battery. And if you are not connected to a grid you'll need one anyway. The question is, what size?
If you have a battery capable of storing say 5KWhr of electricity, to what extent if any, would you be forced to resort to the grid or generator? How might this change if that capacity were to double? To determine optimal battery capacity, draw out a chart for estimated power consumption, another for estimated power production and then superimpose the two. The charts need to cover a whole year. Ideally the battery should be able to accept charge at the rate your chart predicts as a maximum surplus and delivery the nessesary power to cover the predicted maximum deficits. But to arrive at energy storage capacity you have to run cumulatives for the charts, bearing in mind the batteries only give back 80 or 85 percent of the energy put in.
Much the same battery right-sizing applies to electric vehicles with or without a range extender as electrically, a vehicle can be considered an example of an off grid scenario. Here though, the power demand is much easier to calculate.