Na-ion versus Li-ion Batteries: Complementarity Rather than Competitiveness

Jean-Marie Tarascon (1953) is a professor at Collège de France, “Solid-State Chemistry and Energy” chair. His early career was spent in the United States, where he developed the plastic Li-ion technology. Back in France in 1995, he created the European network of excellence ALISTORE-ERI (2001) and the French network on electrochemical energy storage (RS2E) (2012). His present research is devoted to Li(Na)-ion batteries and other chemistries with emphasis on the development of new eco-efficient synthesis processes, novel reactivity concepts, and passive sensing approaches. He is the author of more than 650 scientific papers and retains around 100 patents. Jean-Marie Tarascon (1953) is a professor at Collège de France, “Solid-State Chemistry and Energy” chair. His early career was spent in the United States, where he developed the plastic Li-ion technology. Back in France in 1995, he created the European network of excellence ALISTORE-ERI (2001) and the French network on electrochemical energy storage (RS2E) (2012). His present research is devoted to Li(Na)-ion batteries and other chemistries with emphasis on the development of new eco-efficient synthesis processes, novel reactivity concepts, and passive sensing approaches. He is the author of more than 650 scientific papers and retains around 100 patents. Rechargeable electrochemical batteries, as one of the most versatile energy storage technologies, play a central role in the ongoing transition from fossil fuels to renewable energy for achieving a greener planet. They are the key tools to lower the CO2 footprint of both vehicle transportation and power grid sectors and are essential in a broad range of strategic industries. Our increasing dependence to batteries that are becoming the heart of our society raised several questions about materials abundance, so sustainability is becoming an overriding factor. This has led to a booming diversification of the battery research on new chemistries, such as Li(Na/Al)-air, Li-S, Mg(Ca)-ion, Na(K)-ion, or aqueous Zn-MnO2 systems. Interestingly, none of these technologies are truly new but solely brought back to the scene by key papers published in high-impact journals stating their capabilities for solving the planet’s energy storage issues. Oftentimes, the media creates a hype, few scientists keep excitements, programs are launched at the national level, and new startup companies relying on private funding are launched, but too frequently, promises turn into disillusion, as recently demonstrated by the Li-air technology. Distinguishing between hype and reality within the outgrowing literature on a societal and business-driven research topic such as batteries is becoming a real burden for both young scientists and industrial entities seeking to access the market. Therefore, pressure is on specialists with decades of knowledge on batteries to clarify certain facts. None of the aforementioned “beyond-Li-ion” battery technologies have reached the maturation stage yet, but Na-ion is the closest to this goal, having given birth to companies: Faradion (UK), Novasis (USA), HiNa (China), and Tiamat (France), to name a few. Herein, based on experimental results and facts, I will discuss the benefits and weaknesses of this Na-ion technology with the hope to simplify researchers’ and investors’ minds to make decisions. The Na-ion technology enjoyed a speedy development in the past 8 years simply by learning from the Li-ion chemistry that it mimics. We must recall that, back to 1970s, fundamental research on insertion compounds was divided between Li and Na-based ones.1Rouxel J. Danot M. Bichon J. Les Composés Intercalaires NaxTiS2. Étude Structurale Générale des Phases NaxTiS2 et KxTiS2.Bull. Soc. Chim. Fr. 1971; 11: 3930-3935Google Scholar, 2Delmas C. Fouassier C. Réau J.-M. Hagenmuller P. Sur de nouveaux conducteurs ioniques a structure lamellaire.Mater. Res. Bull. 1976; 11: 1081-1086Crossref Scopus (50) Google Scholar, 3Whittingham M.S. Chemistry of intercalation compounds: Metal guests in chalcogenide hosts.Prog. Solid State Chem. 1978; 12: 41-99Crossref Scopus (1139) Google Scholar It is only because of the outstanding performance provided by Li-based materials, owing to a greater redox potential of Li+ while being less heavy, that Na+ research nearly fell into oblivion. Because of increasing societal demands toward sustainability in the 2010s, scientists have reconsidered Na-ion batteries with the eagerness to identify the best positive and negative electrode materials together with the most suitable electrolyte for achieving stable solid electrolyte interface (SEI) while minimizing parasitic reactions; it is a recurrent problem that has been driving the Li-ion research for last 30 years. The Na-ion research, driven by sustainability, has also caught the eyes of industry, as the raw material Na2CO3 is much cheaper and less price volatile than Li raw materials like Li2CO3. A wide survey of electrode materials inspired from studies on Li-ion materials was launched, and numerous Na-based insertion compounds were identified as having similar structural types as their Li-ion counterparts. These belong to similar families of layered oxides or polyanionic compounds for the positive electrode and of carbonaceous materials or intermetallic alloys for and negative electrode, respectively.4Hwang J.-Y. Myung S.-T. Sun Y.-K. Sodium-ion batteries: present and future.Chem. Soc. Rev. 2017; 46: 3529-3614Crossref PubMed Google Scholar,5Yabuuchi N. Kubota K. Dahbi M. Komaba S. Research development on sodium-ion batteries.Chem. Rev. 2014; 114: 11636-11682Crossref PubMed Scopus (3481) Google Scholar Such findings, combined with the feasibility to use Na-based electrolytes like those of Li-ions in terms of solvents, have been a gift to rapidly assemble full Na-ion cells based on either of the polyanionic/hard carbon (HC) or layered/HC chemistries and identify the potential of Na-ion technology. In both systems, hard carbon (HC) negative electrodes with capacities of ∼300 mAh/g were preferred over alloying (NaxM) or conversion electrodes, which suffer from large volume expansion, poor reversibility, and large voltage hysteresis. Moreover, the Na+ uptake in HC, made of disordered graphene layers and nanopores, proceeds via a sloping voltage region followed by a low-voltage plateau that corresponds to the insertion of Na+ into the disordered layers and to the filing of the nanopores by Na+, respectively, which enables it to achieve higher rates without Na-plating compared to the uptake of Li+ into graphite. The choice was not as straightforward to select the Na-ion-positive electrodes, but in light of the established supremacy of the layered oxides over polyanionic ones in today’s commercial Li-ion batteries, many scientists hastily decided to pursue the development of the former (Figure 1). The Na-based layered oxides show an even richer crystal chemistry than the Li-ion ones owing to the ability of sodium to reside in both octahedral (O) and prismatic (P) environment as opposed to a single octahedral environment for Li.5Yabuuchi N. Kubota K. Dahbi M. Komaba S. Research development on sodium-ion batteries.Chem. Rev. 2014; 114: 11636-11682Crossref PubMed Scopus (3481) Google Scholar However, not only does such a structural difference account for changes in Na-stoichiometry and therefore in capacity, but it also equally explains why layered oxides are more prone to Na-driven structural phase transitions that reduce their lifetime and limit the power density (charging time). In practice, to alleviate the capacity limitation of the P-type layered oxides associated with non-stoichiometry in sodium (∼0.7–0.8 per formula unit), research has been shifted toward the stabilization of stoichiometric O3- Na1MO2 phases via the modification of the nature and content of the transition metals.6Wang Q. Mariyappan S. Vergnet J. Abakumov A.M. Rousse G. Rabuel F. Chakir M. Tarascon J.-M. Reaching the Energy Density Limit of Layered O3-NaNi0.5Mn0.5O2 Electrodes via Dual Cu and Ti Substitution.Adv. Energy Mater. 2019; 9: 201901785Crossref Scopus (40) Google Scholar An ultimate step toward high capacity lies in the synthesis of an anionic redox-active O3 NaLi1/3M2/3O2 phase. However, this phase could suffer from sluggish kinetics like its Li-rich counterpart. Faradion conceived Na-ion pouch cells with O3 or O3-P2 intermixture (or inter-growths) of NaNi(1−x−y−z)MnxMgyTizO2 phases where more than 0.8 Na can be used, and they claimed to reach a specific energy of 150 Wh/kg for the total cell weight at C/3 rate with an average voltage of 3.2 V.7Bauer A. Song J. Vail S. Pan W. Barker J. Lu Y. The Scale-up and Commercialization of Nonaqueous Na-Ion Battery Technologies.Adv. Energy Mater. 2018; 8: 1702869Crossref Scopus (86) Google Scholar For comparison, today’s commercial Li-ion batteries based on layered oxides offer a much higher cell-level specific energy, i.e., in excess of 250 Wh/kg. Though the usage of abundant earth elements and the absence of Co is attractive, power performance of layered oxides (1,000–1,500 W/kg in discharge) is not spectacular, and in-depth analyses of high temperature performance and thermal stability are still to be realized. Deviating from layered structures, the three-dimensional (3D) Na-based polyanionic phases, such as phosphates (NaFePO4), sulfates Na2Fe2(SO4)3, and flurophosphates (NaVPO4F), were also heavily studied. By exploring the huge parameter space enlisting the material elemental composition and its crystal-electronic structure that determines the cathode’s electrochemical performances, Na3V2(PO4)2F3 (NVPF) compound, with its specific crystal structure consisting of open channels for fast Na+ ions diffusion, turns out to be the most attractive electrode from a practical point of view. It can reversibly release two Na+ per formula unit (i.e., 128 mAh/g) at an average potential of 3.9V, thus offering a material-level specific energy of ∼507 Wh/kg, comparable to ∼580Wh/kg for LiFePO4 (LFP), which is widely used as a positive electrode in Li-ion batteries. This gap between NVPF and LFP has almost been filled by harnessing the electrochemical activity of the third sodium of NVPF via the formation of a disordered NaxV2(PO4)2F3.8Yan G. Mariyappan S. Rousse G. Jacquet Q. Deschamps M. David R. Mirvaux B. Freeland J.W. Tarascon J.-M. Higher energy and safer sodium ion batteries via an electrochemically made disordered Na3V2(PO4)2F3 material.Nat. Commun. 2019; 10: 585Crossref PubMed Scopus (101) Google Scholar The NVPF/HC chemistry has been successfully implemented in Tiamat’s 18,650 prototype cells delivering 122 Wh/kg at 1C rate. The cells exhibit high power rate capability (90% at 1C) with long cycling life (>4,000 cycles); however, the inherent drawback is rooted in the use of toxic vanadium, which scientists are trying to alleviate by replacing with less toxic and other abundant 3d metals (Mn). Lastly, the Prussian blue phases, which have been known for decades and long considered as electrodes for electrochromic devices that were never commercialized, are regarded as reversible host positive electrodes for Na-ion batteries. They adopt a 3D structural framework whose compositions Na2−δMnFe(CN)6.yH2O have been the most attractive for Na-ion batteries, as demonstrated by Novosis.7Bauer A. Song J. Vail S. Pan W. Barker J. Lu Y. The Scale-up and Commercialization of Nonaqueous Na-Ion Battery Technologies.Adv. Energy Mater. 2018; 8: 1702869Crossref Scopus (86) Google Scholar The positive attributes of Prussian blue phases lies in their non-toxicity, high-rate capability, and easy synthesis, provided we can reproducibly master their particle morphologies and water content, which raises difficulties for achieving long lifetime in non-aqueous electrolytes. To curb this difficulty, a water-free Prussian blue electrode was recently reported with capacity of 140 mAh/g at 1C rate with an average voltage of 3.4 V with still a modest energy density (volumetric) owing to the low material density (<1.8 g/cm3 as compared to 5.1 and 3.5 g/cm3 for LiCoO2 and LiFePO4, respectively).5Yabuuchi N. Kubota K. Dahbi M. Komaba S. Research development on sodium-ion batteries.Chem. Rev. 2014; 114: 11636-11682Crossref PubMed Scopus (3481) Google Scholar,9Hurlbutt K. Wheeler S. Capone I. Pasta M. Prussian Blue Analogs as Battery Materials.Joule. 2018; 2: 1950-1960Abstract Full Text Full Text PDF Scopus (121) Google Scholar Besides electrodes, another key to battery technologies is the electrolyte, which mainly governs their performances and lifetime. Initially, scientists blindly transposed the commonly used electrolytes for Li-ion to Na-ion by simply replacing LiPF6 with NaPF6. This strategy offered a rapid way to screen paired positive and negative electrodes in coin cells, but it fell short in achieving the figures of merits required for commercial applications10Ponrouch A. Dedryvère R. Monti D. Demet A.E. Ateba Mba J.M. Croguennec L. Masquelier C. Johansson P. Palacín M.R. Towards high energy density sodium ion batteries through electrolyte optimization.Energy Environ. Sci. 2013; 6: 2361-2369Crossref Scopus (310) Google Scholar We experienced this scenario with our first assembled 18,650 prototypes based on NVPF//1M NaPF6 in EC-DMC//HC chemistry, which show good electrochemical performance at 25°C but mediocre performances at 55°C in terms of self-discharge and capacity retention. We identified the linear carbonate component of the electrolyte (DMC), which is well used in Li-ion batteries, as the troublemaker. This could have been anticipated, bearing in mind the milder Lewis acidity of Na+ as compared to Li+, which renders the Na-based carbonates more soluble than their Li counterparts, therefore leading to a poor stability of the protecting SEI layer on the C electrode. This backtrack was rapidly overcome via a trial-error approach relying on empirical synergy rules between various additives established for Li-ion electrodes, hence the discovery of an optimized electrolyte enabling to perform over a wide range of temperature with limited parasitic reactions and therefore better kinetics at the carbon electrode.11Cometto C. Yan G. Mariyappan S. Tarascon J.-M. Means of Using Cyclic Voltammetry to Rapidly Design a Stable DMC-Based Electrolyte for Na-Ion Batteries.J. Electrochem. Soc. 2019; 166: A3723-A3730Crossref Scopus (11) Google Scholar In light of such improvements dealing with either positive or negative electrodes and electrolytes, Na-ion batteries has been assembled either as 18650 or Pouch cells, and companies such as Tiamat, Faradion, and Novasis hosting the NVPF, layered oxide, and Prussian blue chemistry, respectively, were created.7Bauer A. Song J. Vail S. Pan W. Barker J. Lu Y. The Scale-up and Commercialization of Nonaqueous Na-Ion Battery Technologies.Adv. Energy Mater. 2018; 8: 1702869Crossref Scopus (86) Google Scholar,10Ponrouch A. Dedryvère R. Monti D. Demet A.E. Ateba Mba J.M. Croguennec L. Masquelier C. Johansson P. Palacín M.R. Towards high energy density sodium ion batteries through electrolyte optimization.Energy Environ. Sci. 2013; 6: 2361-2369Crossref Scopus (310) Google Scholar They are offering to users, via prototyping, several opportunities of application markets and therefore are providing momentum for the Na-ion technology acceptance. However, some skepticism prevails, in which one of the most frequent questions is what are the chances of success, and more so, in which market sector could this technology be the winning one? To answer, we next highlight some of the identified advantages that this technology has against the Li-ion that is invading most of the energy storage related markets. We should reiterate here that, though Na-ion technology mimics the Li-ion with similar types of electrodes and electrolytes, Na is three times heavier than Li and has redox potential 300 mV lower, which inherently reduces the energy density of Na-ion technology by at least ∼30% compared to Li-ion, with such volumetric numbers out of question for Prussian blue. We should also realize that this gap will prevail forever, because progress that could be made at the materials level for Na will always be mirrored with progresses on Li, since we are dealing with the same family of materials. So straightforwardly, the usage of Na-ion technology in applications requiring high energy density, such as vehicle transportation sector, is partly eliminated. Besides energy density, another point of paramount importance for batteries in the transportation sector is power rate capability for fast charging, regenerative braking, and start-stop functions as well as within the grid sector for frequency adjustment. Within this context, Na-ion chemistries relying on the use of open 3D structures perform quite well, as demonstrated for NVPF (Figure 2), when compared to their Li counterparts. This can be illustrated by benchmarking Tiamat’s NVPF/C 18650 batteries against the super-fast-charging lithium ion battery (SCIB) from Toshiba (Figure 2). Note that Tiamat’s cells nearly compare with SCIB’s in terms of power rate while having a higher voltage (3.7 V versus 2.7 V). Moreover, they compete favorably with high-power LiFePO4/C by sacrificing less energy density, thus resulting in a lower €/kWh and €/kW cost. Thus, the Na-ion technology definitively has a key role to play in the automotive industry, where power-hungry functions (48 V, regenerative breaking) are needed, or for applications requiring power with the minimum compromise in autonomy, such as fast charging e-bus. Another great asset of Na-ion against Li-ion is rooted in its ability to be discharged or maintained at 0 V without any risks of altering its subsequent performances.7Bauer A. Song J. Vail S. Pan W. Barker J. Lu Y. The Scale-up and Commercialization of Nonaqueous Na-Ion Battery Technologies.Adv. Energy Mater. 2018; 8: 1702869Crossref Scopus (86) Google Scholar,12Roberts S. Kendrick E. The re-emergence of sodium ion batteries: testing, processing, and manufacturability.Nanotechnol. Sci. Appl. 2018; 11: 23-33Crossref PubMed Scopus (39) Google Scholar Such advantage is simply rooted in the feasibility of using aluminum current collectors that do not alloy with Na in contrast to Li, which forms an alloy (LixAl). Such alloying reaction has forced the use of a more expensive Cu negative current collector for Li-ion cells, which present risks of oxidation (Cu → Cun+) when cells stand at 0 V. This difference offers a serious advantage to Na-ion cells that can be transported in their discharge state and therefore are free of legislative transportation rules. Further dealing with safety aspects, both systems are comparable; however, the likelihood of having dendrites and explosions at high charging rates is lower for Na than for Li owing to the chemically softer nature of Na. When dealing with sustainability, the Na-ion technology hosts environmentally friendly electrodes that are made of Mn, Fe, Ni, etc. (Prussian blue and layered oxides) and are Co-free. Bearing in mind that the abundance of chemical elements resonates with lower prices and the use of cheaper Al than Cu current collectors, a 10% cost reduction of the stored kWh is foreseen with the Na-ion technology over the years if volume production is made possible.13Larcher D. Tarascon J.-M. Towards greener and more sustainable batteries for electrical energy storage.Nat. Chem. 2015; 7: 19-29Crossref PubMed Scopus (3597) Google Scholar Such a target should be feasible, as manufacturing of Na-ion cells can utilize the Li-ion assembly lines without the need for further investment. Such sustainability and cost advantage represent a serious asset for grid applications where space is not limited; thus, volumetric energy density is not any longer an overriding prerequisite. In short, we hope that this commentary, based on real and trackable facts, will raise the awareness in the battery community about the state of performance that the Na-ion technology can achieve while bearing in mind that only 8 years of research has been devoted to it compared to 30 years for Li-ion. Cells can be made for either power applications (Tiamat) or autonomy (Faradion, which just received its first order for the Australian market).14https://www.faradion.co.uk/faradion-receives-first-order-of-sodium-ion-batteries-for-australian-market/Google Scholar Achievements have been spectacular, but further progress is still needed, in particular at the electrolyte level, to reach the perfection of Li-ion in terms of lifetime and durability at various temperatures but also at the carbon negative electrode to improve packing density. We are confident that this objective will be rapidly reached based on the numerous research groups worldwide entering this research field and the numerous institutions that have integrated the Na-ion technology in their future road maps. To conclude, the Na-ion technology is becoming a reality, but please do not take this as a revolutionary new idea. It is not, and here is proof: remember that in 1869, our French visionary writer Jules Verne had already identified the benefit of this battery technology in his novel Twenty Thousand Leagues under the Sea, quoting “Sir, sodium alone is consumed and the sea provides it itself. I will also tell you that sodium batteries should be considered as the most energetic.” By highlighting the strength and weakness of the various Na-ion battery chemistries against Li-ion ones, we hope that users will now easily identify the benefits that such a technology could bring to their business not as replacement of Li-ion but for providing added values to Li-ion for specific applications requiring power. It could also complement Li-ion for massive storage applications where energy density is not an overriding factor, thus minimizing the fears of Li shortage. We are at the early stage of the Li-ion age, and we must act now to not repeat the polymer history, which was in its booming age in the 1900s and is now becoming the 2000s’ major global planet problem with plastic pollution. To avoid repeating such a mistake, the Na-ion technology, long predicted but not realized yet, stands as an attractive option toward greener and more sustainable batteries, which is the only way to electrify the world without creating a new environmental burden for the coming generations. The author wants to thank M. Morcrette, S. Mariyappan, and G. Assat for comments as well as J. Barker (Faradion) and L. Hubard (Tiamat). J.M.-T. is a shareholder of Tiamat and member of its development committee (without any compensation).