Research in lithium-ion batteries has produced many proposed refinements of lithium-ion batteries. Areas of research interest have focused on improving energy density, safety, rate capability, cycle durability, flexibility, and reducing cost.
Since the beginning of 2024, major smartphone companies have been switching to silicon-carbon batteries which utilize silicon anodes rather than traditional graphite anodes in lithium-ion batteries, which allows for higher energy density and longer run time.
Despite the advantages the technology has, companies like Samsung, Apple, and Google are still cautious about the use of it due to concerns about its silicon expansion, which may cause faster battery degrading, and even explosions under high pressure.
Artificial intelligence (AI) and machine learning (ML) is becoming popular in many fields including using it for lithium-ion battery research. These methods have been used in all aspects of battery research including materials, manufacturing, characterization, and prognosis/diagnosis of batteries.
Negative electrode Materials that are taken into consideration for the next generation lithium-ion battery (LIBs) negative electrode share common characteristics such as low cost, high theoretical specific capacity, and good electrical conductivity, etc. Carbon- and silicon- based materials have shown to be promising materials for the negative electrode. However, along with the desired characteristics from some of the materials, a number of weaknesses have also been shown. For example, although silicon has a theoretical specific capacity that is 10 times higher than graphite, it has low intrinsic electrical conductivity. Current research focuses on engineering materials so that their characteristics are retained and their weaknesses are accommodated.
Lithium-ion battery negative electrodes are most commonly made of graphite. Graphite anodes are limited to a theoretical capacity of 372 mAh/g for their fully lithiated state.G. Shao et al. Polymer-Derived SiOC Integrated with a Graphene Aerogel As a Highly Stable Li-Ion Battery Anode ACS Appl. Mater. Interfaces 2020, 12, 41, 46045–46056 At this time, significant other types of lithium-ion battery anode materials have been proposed and evaluated as alternatives to graphite, especially in cases where niche applications require novel approaches.
Si@void@C microreactor Dr. Leon Shaw's research group from Illinois Institute of Technology has developed the Si@void@C microreactors which show exceptional test results to be LIBs anode. The process of creating Si@void@C microreactors begins with the production of nanostructured silicon particles through a high-energy ball milling process with micron-sized silicon powder. The nanostructured Si particles are then encapsulated with carbon through carbonization of a carbon precursor containing nitrogen element. Finally, the particles are etched with NaOH to create voids with nano-channel morphology inside the Si core to form the Si@void@C microreactors.
Tests from the Si@void@C microreactors demonstrated high Coulombic Efficiency of 91% during the first lithiation process, which is significantly higher than other reported silicon anodes. The design also enabled high Coulombic Efficiency of 100% after 5 cycles, indicating no discernible SEI layer formation beyond 5 cycles. Additionally, the specific capacity increased in subsequent cycles due to the activation of more electrode material, suggesting robust electrochemical stability.
The Si@void@C(N) electrode was tested to be capable of ultrafast charging and durability over 1000 cycles, the specific capacity maintained high levels (~800 mAh g−1) even at very high current densities (up to 8 A g−1). No lithium plating was observed for the Si@void@C(N) electrode even after 1000 cycles at 8 A g−1, indicating their capability for ultrafast charging without compromising safety and capacity retention.
Intercalation oxides Several types of metal oxides and sulfides can reversibly intercalate lithium cations at voltages between 1 and 2 V against lithium metal with little difference between the charge and discharge steps. Specifically the mechanism of insertion involves lithium cations filling crystallographic vacancies in the host lattice with minimal changes to the bonding within the host lattice. This differentiates intercalation negative electrode from conversion negative electrode that store lithium by complete disruption and formation of alternate phases, usually as lithia. Conversion systems typically disproportionate to lithia and a metal (or lower metal oxide) at low voltages, < 1 V vs Li, and reform the metal oxide at voltage > 2 V, for example, CoO + 2Li -> Co+Li2O.
Titanium dioxide In 1984, researchers at Bell Labs reported the synthesis and evaluation of a series of lithiated titanates. Of specific interest were the anatase form of titanium dioxide and the lithium spinel LiTi2O4 Anatase has been observed to have a maximum capacity of 150 mAh/g (0.5Li/Ti) with the capacity limited by the availability of crystallographic vacancies in the framework. The TiO2 polytype brookite has also been evaluated and found to be electrochemically active when produced as nanoparticles with a capacity approximately half that of anatase (0.25Li/Ti). In 2014, researchers at Nanyang Technological University used a materials derived from a titanium dioxide gel derived from naturally spherical titanium dioxide particles into nanotubes In addition, a non-naturally occurring electrochemically active titanate referred to as TiO2(B) can be made by ion-exchange followed by dehydration of the potassium titanate K2Ti4O9. This layered oxide can be produced in multiple forms including nanowires, nanotubes, or oblong particles with an observed capacity of 210 mAh/g in the voltage window 1.5–2.0 V (vs Li).
Niobates In 2011, Lu et al., reported reversible electrochemical activity in the porous niobate KNb5O13. This material inserted approximately 3.5Li per formula unit (about 125 mAh/g) at a voltage near 1.3 V (vs Li). This lower voltage (compared to titantes) is useful in systems where higher energy density is desirable without significant SEI formation as it operates above the typical electrolyte breakdown voltage. A high rate titanium niobate (TiNb2O7) was reported in 2011 by Han, Huang, and John B. Goodenough with an average voltage near 1.3 V (vs Li).
Niobium tungsten oxide Niobium tungsten oxide is a Wadsley–Roth structure that is under investigation. It is a rhenium trioxide (ReO3)-type structure periodically interrupted by crystallographic shear planes. The result is continuous multi-dimensional tunnels that support rapid lithium-ion diffusion. At the shear planes, the niobium atoms share edges. These edges are structural and buffer volume expansion/contraction as lithium ions migrate.
Transition-metal oxides In 2000, researchers from the Université de Picardie Jules Verne examined the use of nano-sized transition-metal oxides as conversion anode materials. The metals used were cobalt, nickel, copper, and iron, which proved to have capacities of 700 mAh/g and maintain full capacity for 100 cycles. The materials operate by reduction of the metal cation to either metal nanoparticles or to a lower oxidation state oxide. These promising results show that transition-metal oxides may be useful in ensuring the integrity of the lithium-ion battery over many discharge-recharge cycles.
Lithium Lithium anodes were used for the first lithium-ion batteries in the 1960s, based on the cell chemistry, but were eventually replaced due to dendrite formation which caused internal short-circuits and was a fire hazard. Effort continued in areas that required lithium, including charged cathodes such as manganese dioxide, vanadium pentoxide, or molybdenum oxide and some polymer electrolyte based cell designs. The interest in lithium metal anodes was re-established with the increased interest in high capacity lithium–air battery and lithium–sulfur battery systems.
Research to inhibit dendrite formation has been an active area. Doron Aurbach and co-workers at Bar-Ilan University have extensively studied the role of solvent and salt in the formation of films on the lithium surface. Notable observations were the addition of LiNO3, dioxolane, and hexafluoroarsenate salts. They appeared to create films that inhibit dendrite formation while incorporating reduced Li3As as a lithium-ion conductive component.
In 2021, researchers announced the use of thin (20 micron) lithium metal strips. They were able to achieve energy density of 350 Wh/kg over 600 charge/discharge cycles.
Non-graphitic carbon Various forms of carbon are used in lithium-ion battery cell configurations. Besides graphite poorly or non-electrochemically active types of carbon are used in cells such as CNTs, carbon black, graphene, graphene oxides, or MWCNTs.
Recent work includes efforts in 2014 by researchers at Northwestern University who found that metallic single-walled carbon nanotubes (SWCNTs) accommodate lithium much more efficiently than their semiconducting counterparts. If made denser, semiconducting SWCNT films take up lithium at levels comparable to metallic SWCNTs.Nanotubes make for better lithium-ion batteries , Nanotechweb.org, 3 March 2014.
Hydrogen treatment of graphene nanofoam electrodes in LIBs was shown to improve their capacity and transport properties. Chemical synthesis methods used in standard anode manufacture leave significant amounts of atomic hydrogen. Experiments and multiscale calculations revealed that low-temperature hydrogen treatment of defect-rich graphene can improve rate capacity. The hydrogen interacts with the graphene defects to open gaps to facilitate lithium penetration, improving transport. Additional reversible capacity is provided by enhanced lithium binding near edges, where hydrogen is most likely to bind. Rate capacities increased by 17–43% at 200 mA/g. In 2015, researchers in China used porous graphene as the material for a lithium-ion battery anode in order to increase the specific capacity and binding energy between lithium atoms at the anode. The properties of the battery can be tuned by applying strain. The binding energy increases as biaxial strain is applied.
Silicon
Silicon is an earth abundant element, and is fairly inexpensive to refine to high purity. When alloyed with lithium it has a theoretical capacity of ~3,600 milliampere hours per gram (mAh/g), which is nearly 10 times the energy density of graphite electrodes, which exhibit a maximum capacity of 372 mAh/g for their fully lithiated state of LiC6. One of silicon's inherent traits, unlike carbon, is the expansion of the lattice structure by as much as 400% upon full lithiation (charging). For bulk electrodes, this causes great structural stress gradients within the expanding material, inevitably leading to fractures and mechanical failure, which significantly limits the lifetime of the silicon anodes. In 2011, a group of researchers assembled data tables that summarized the morphology, composition, and method of preparation of those nanoscale and nanostructured silicon anodes, along with their electrochemical performance.
Porous silicon nanoparticles are more reactive than bulk silicon materials and tend to have a higher weight percentage of silica as a result of the smaller size. Porous materials allow for internal volume expansion to help control overall materials expansion. Methods include a silicon anode with an energy density above 1,100 mAh/g and a durability of 600 cycles that used porous silicon particles using ball-milling and stain-etching. In 2013, researchers developed a battery made from porous silicon nanoparticles. Below are various structural morphologies attempted to overcome issue with silicon's intrinsic properties.