Lithium Ion

Introduction

Lithium ion batteries are ubiquitous in portable electronics and electric vehicles, but are now competitive in stationary storage applications where lead-acid and nickel-metal hydride technologies once dominated.

“Lithium-ion” may refer to a number of technologies which use an electrolyte composed of a lithium-salt dissolved in an organic solvent. A graphite (carbon) anode is typically used, though alternative anode technologies are being widely investigated.

The name of the technology generally relates to the cathode in use. As an example, a lithium cobalt oxide (LCO) battery has a carbon (C6) anode and a lithium cobalt oxide (LiCoO2) cathode. Li+ ions move between the two during charging and discharging, as electrodes swell and contract to accept/give-up ions.

If energy density is thought of as the amount of energy stored per unit of weight/volume, and power density is the maximum power delivery per unit of weight/volume, then the following can be said of each lithium-ion chemistry:

  • Lithium Cobalt Oxide (LCO): commonly used in portable electronics as they offer the highest energy density of the commercially-mature lithium-ion technologies described here. However, power density, lifetime, and thermal stability are relatively poor, and hence they are unsuited to large-format EV and stationary storage applications.
  • Lithium Manganese Oxide (LMO): once the chemistry of choice for EV manufacturers, the use of manganese in place of cobalt increased power density and thermal stability when compared to LCO. However, lifetime remained short, and energy density was reduced. As a result, the technology has been largely superceded by NMC/NCA.
  • Lithium Nickel Manganese Cobalt Oxide (NMC): an “all-rounder” chemistry that offers good performance across all metrics discussed thus far. The technology holds significant market share in both EV and stationary storage applications.
  • Lithium Nickel Cobalt Aluminium Oxide (NCA): competes with NMC and LFP for majority market share in EV powertrains. NCA cells exhibit high energy density, high power density, and long shelf life, but tend to degrade faster with use than NMC cells.
  • Lithium Iron Phosphate (LFP): competes with NMC and NCA for majority market share in EV powertains, and with NMC for majority market share in stationary storage. Despite having lower energy density than NMC cells, LFP cells are widely utilised in EV’s in China due to higher thermal stability and high power capability.
  • Lithium Titanate (LTO): with LTO nanocrystals used in the anode, these batteries display unparalleled thermal stability and cycle life. Low energy density and higher capital costs (when assessed on a $/kWh basis) make them less suited to EV’s, but their ability to deliver this energy over an extremely short period makes them competitive in stationary, high power applications.

For low-rate stationary storage applications such as on-grid and off-grid solar storage, LFP and NMC chemistries are the market leaders from the lithium-ion family. Traditionally, lead-acid batteries have been more cost-effective than lithium-ion in this market, owing to lower capital costs per unit of usable storage capacity. However, the rapidly falling lithium-ion battery pack costs seen across the industry are increasingly eroding lead-acid market share.

While the cost per usable unit of capacity should be a significant criteria in any investment decisions, the following should also be considered when comparing the two technologies:

Discharging

Lithium-ion batteries fail irreversibly if over-discharged. If the cathode cannot accept any further lithium ions, metallic lithium plating may occur, resulting in irreversible capacity loss. Further, if the voltage of the battery is allowed to fall below its minimum threshold, the copper electrode can dissolve into the electrolyte. When the battery is subsequently recharged, a short-circuit can form between electrodes. These vulnerabilities mean that a BMS and a power source are essential to protect battery cells from over-discharge.

The rate at which the battery is discharged will affect the total capacity available for both lead-acid and lithium-ion chemistries. Higher rates also lead to higher cell temperatures that increase parasitic reactions and accelerate capacity fade. Generally, lithium-ion batteries lose less capacity and generate less heat than lead-acid batteries for the same discharge rate. This allows them to deliver more power per unit of energy capacity, making them better suited to high-rate applications such as wind/solar farm “smoothing”, and voltage/frequency support.

Charging

Lithium-ion batteries can generally be charged more quickly than lead-acid batteries. However, excessive charging currents can cause lithium plating on the anode when it is unable to accommodate (intercalate) lithium ions quickly enough. The BMS may provide protection against excessive charge currents to reduce this risk.

Metallic lithium plating may also occur if a lithium-ion battery is over-charged. In this case, the anode, already full with lithium ions, cannot accept more ions, which instead deposit on the surface of the electrode. As well as reducing capacity (due to active material loss), this plating can cause dendrite growth that may lead to a short-circuit between electrodes. To mitigate this risk, most BMS will monitor individual cell voltages, and dissapate excess charge from each cell as necessary to avoid over-voltage conditions.

As is the case for lead-acid batteries, lithium-ion battery packs are charged via a constant-current “bulk charging” phase and a constant-voltage “absorption” phase. The bulk phase continues until the cell/pack voltage reaches a maximum, and this voltage is then maintained through the absorption phase as the charge current falls.

For a lead-acid battery, “float” (or “trickle”) charging proceeds once the absorption phase amperage falls to some setpoint. This trickle charge opposes any self-discharge to keep the battery fully charged, reducing sulfation effects and maximising lifespan. For a lithium-ion battery however, trickle charging is undesirable, as the longer the battery spends at high SOC, the higher the rate of capacity fade. It is for this reason that lithium-ion batteries tend to be stored at 20-30% SOC.

The preference of lithium-ion technology for “partial SOC operation” is a significant advantage over lead-acid technology in many applications.

Efficiency

The ratio of the energy required to charge a battery compared to the available energy during discharge is referred to as the “round-trip efficiency”. A typical lithium ion battery will lose only 3-8% of energy round-trip, compared to 15-25% losses for lead-acid systems.

Both lead-acid and lithium-ion technologies perform well with regards to self-discharge, with losses of around 5% of capacity per month. In frequent cycling applications this loss is of little consequence.

Energy & Power Density

Where space restraints exist, lithium ion batteries are often preferred to lead-acids owing to higher energy and power density (in terms of both size and weight). This characteristic is reflected in the widespread deployment of lithium ion batteries in portable electronics and electric vehicles.

Lithium-ion battery packs are already on the market that provide energy densities of 144 Wh/L and 107 Wh/kg, compared to ~80 Wh/L and ~34 Wh/kg for lead-acid.

Toxicity & Disposal

Though primary (non-rechargeable) lithium batteries possess toxic metallic lithium, the lithium compounds in secondary (rechargeable) lithium-ion batteries are much more stable. Some chemistries do possess small traces of toxic metals, but the amounts are sufficiently small that they are most often disposed of in traditional waste streams at present, while recycling initiatives are still in their infancy.

Safety

Though lead-acid batteries consist of large amounts of highly toxic lead and corrosive sulphuric acid, they are generally considered safer than lithium-ion batteries owing to a lower risk of “thermal runaway”. Thermal runaway refers to the positive feedback loop that can cause battery swelling, fire and/or explosion when the catalytic effect of heat released during battery malfunction accelerates the reactions causing the release of the heat.

Due to the higher energy density of lithium-ion cells, the reactivity of lithium, and the flammability of their organic solvent electrolyte, thermal runaway can be more dangerous than in lead-acid battery cells. Without protection systems in place, the likelihood also tends to be greater (with thermally-stable LTO batteries the exception).

In practice, however, lithium-ion batteries are manufactured with a number of safety measures to offset this risk, and typically include:

  • Battery management systems (BMS) that isolate battery packs in the event of over/under-voltage or over/under-temperature conditions
  • Cell balancing systems that equalise the SOC of battery cells connected in series, to avoid over/under-voltage conditions
  • Battery fusing to arrest short-circuit currents
  • Thermal management systems to carry heat away from cells via air or liquid cooling

The widespread use of lithium-ion batteries in EV’s is indicative of the effectiveness of these protection systems. In stationary applications, where temperatures are less variable, the likelihood of thermal runaway is reduced even further.

Lifetime & Depth-of-Discharge

In ‘float’ mode, lead-acid batteries can have long shelf lives of up to 20 years. With respect to cycle life, claims vary, but top-end deep-cycle lead-acid datasheets state cycle lives of 3,000+ cycles for discharging down to a 40% state of charge, equivalent to 1,800+ full cycles. The capacity fades roughly linearly as the battery is cycled, until the remaining capacity has fallen to ~80% of its initial capacity (an 80% “state-of-health”). At this point, the likelihood of imminent failure increases and, for this reason, this point is generally accepted as the end-of-life point (or replacement point) for lead-acid batteries.

Lithium-ion capacity also fades roughly linearly, but this typically continues down to ~60% SOH before falling away quickly. For this reason, 60% SOH is generally accepted as lithium-ion’s end-of-life point, with top-end lithium-ion manufacturers claiming lifetimes in excess of 4,000 full cycles before this point is reached.

If manufacturers’ claims are accurate, the superior cycle life of such lithium-ion battery packs would make the additional capital expense (per kWh) worthwhile when the battery is to be frequently cycled. The primary purpose of the Lithium-ion Battery Trial is to determine the accuracy of these claims when batteries are subject to ambient temperatures typical of the Australian climate.

Costs

There are many factors to consider when comparing the cost of lead-acid and lithium-ion batteries. A simple comparison of cost per unit of capacity (ie. $/kWh) is not sufficient to make the best decision, not least because the kWh’s stored in a battery depends on the rate at which you discharge it, or that this capacity will fade at different rates for different batteries at different temperatures. The overall efficiency must also be factored in, and then the value of partial SOC operation must be considered.

What can be said is that, at low discharge rates (24 – 120 hours) lead-acid batteries tend to have a lower capital costs per kWh of storage. However, this capacity will fade faster than lithium-ion batteries with frequent use (daily cycling etc.), and the system will have to be designed so as to ensure batteries are fully charged regularly.

As the required charge/discharge rate increases, the capital costs of lead-acid batteries (on a $/kWh basis) increase, and hence the economics begin to favour lithium-ion chemistries. Similarly, in heavy-cycling applications, the slower capacity fade and higher efficiency of lithium-ion cells may offset the increased capital cost to the point that the total levelised cost is lower with a lithium-ion solution than a lead-acid solution.

Assessment of the total cost of ownership must take all these factors into account, alongside project constraints (ie. space and weight limits, serviceability etc.) and financial parameters (debt-to-equity, cost of debt, terms of debt, cost of equity, tax, etc.).

Future Deployment

The most significant obstacles to lithium-ion deployment in stationary storage applications are capital costs and product maturity. The lead-acid battery has been in use for over 100 years, deployed across many applications, and is as predictable and well-understood as is possible for an electrochemical technology. The same cannot be said of lithium-ion batteries where example projects of many years’ duration do not and cannot exist on account of the infancy of the technology.

While lead-acid technology is mature and prices are unlikely to fall significantly in future, lithium-ion prices are falling rapidly due to increasing demand from the EV sector propagating technological improvements and streamlining of manufacturing processes.

Currently, manufacturers’ data sheets provide most of the available information, which is derived only from laboratory testing. ITP’s Lithium-Ion Battery Trial is one attempt to scrutinise manufacturers’ claims and understand how this new technology performs in real-world conditions.