Effect of doping and crystallite size on the electrochemical performance of Li4Ti5O12

Defect spinel phase lithium titanate (Li4Ti5O12) has been suggested as a promising negative electrode material for next generation lithium ion batteries. Flame spray pyrolysis has been shown to be a viable fast, one-step process for synthesis of nanoparticulate Li4Ti5O12. However, due to the rapid quenching that is integral to the process the crystallite size remain very small and non-uniform. To overcome this shortcoming a vertical flow tube furnace was used to increase the high-temperature residence time. This resulted in an increase in the crystallite size and crystallinity of the product. As a result of this increase the electrochemical performance of the Li4Ti5O12 was markedly improved. Furthermore, silver doping of the Li4Ti5O12 material can be carried out simultaneously with its synthesis in the FSP process. The resulting nanosized silver particles on the surface of the Li4Ti5O12 particles further improve the electrochemical performance during high current operations. The specific capacities of these high-temperature synthesised pure and silverdoped Li4Ti5O12 nanoparticles were found to increase by up to 6% and 19%, respectively, compared to a commercial reference. Thus the technique provides a simple method for synthesising superior quality Li4Ti5O12 for battery appli∗Correspondence to tommi.karhunen@uef.fi Preprint submitted to Journal of Alloys and Compounds May 26, 2015 *Manuscript Click here to view linked References


Introduction
Concern over the worldwide carbon dioxide emissions, as well as the dwindling oil resources, have led to increasing interest in efficient energy storage solutions for both automotive and renewable energy supply applications. At the same time advances in the miniaturization and mobility of consumer elec- 5 tronics are placing ever increasing pressure on the capabilities of rechargeable batteries.
Currently lithium ion batteries with their high energy density and number of charge cycles are, perhaps, the best available technology to meet these demands [1]. However, for the widespread utilization of Li-ion batteries in high-demand 10 applications such as full and hybrid electric vehicles, a number of improvements are still required. These include price, safety, specific energy and power, and the cycle life [2].
Lithium titanium oxide (Li 4 Ti 5 O 12 , LTO) is recognised as a promising material for the negative electrode of the next generation Li-ion batteries as it is 15 low cost and safe, and has an excellent cycle life [2,3]. However, it also has a few drawbacks, the major one being its low electronic conductivity.
Two solutions for this problem have been proposed. First, reducing the primary particle size of the electrode materials reduces the length of the electron diffusion paths and the local current density [4]. The high-rate capabilities of 20 electrodes made with LTO samples from different suppliers were studied by Kavan et al. [5]. They showed that the specific capacity fell off rapidly for very small crystallite size (< 20 nm). The fall was attributed to a decrease in the Li-ion diffusion coefficient due to either a shrinking of the crystallite lattice or to increased Li + -Li + repulsion, or a combination thereof. 25 However, large primary particle size also leads to low active surface area and, thus, a decreased charge transfer rate between the electrode and the electrolyte.

2
As such an optimal crystallite size range exist for LTO nanoparticles. This was reported by Kavan et al. [5] as 20-80 nm.
Second, improved conductivity and rate capacity have been observed for 30 LTO doped with transition metals such as silver [6] or copper [7]. However, the standard LTO production method of solid-state chemical reactions typically produces LTO particles with a diameter on the order of 1 µm and the doping often requires a separate process thus adding to the complexity, and consequently to the cost, of production.

35
Single-step gas phase processes are more efficient than the solid-state ones in both the energy and raw materials required which is of great importance when scaling the process up to industrial scale. Gas phase synthesis also produces particles of high purity composed of non-porous primary particles with small size, and relatively narrow size distribution [8]. 40 Karhunen et al. [9] showed that flame spray pyrolysis (FSP) [10] can be used to synthesise pure and doped LTO nanoparticles. The particles were found to be about 100 nm agglomerates consisting of primary particles with high elemental and phase purity. The primary and crystallite sizes were observed to be about 10 nm. Furthermore, when silver precursor was added to the synthesis a uniform 45 distribution of silver nanoparticles (about 1 nm in size) was observed on the surface of the LTO primary particles.
However, when these LTO nanoparticles were tested in a Li-ion half cell the specific capacity fell significantly short of the a commercial reference LTO.
Based on the findings of Kavan et al. [5] it was decided that the most likely 50 reason for this was the very small crystallite size of the nanoparticles.
It is worth noting that there is a trade-off between capacity and current in the utilisation of intercalation electrodes in Li-ion batteries. Smaller crystallite sizes promote better performance at high current densities due to the shorter Li-ion diffusion path and the higher active surface area. However, they also lead 55 to less efficient utilisation of the intercalation sites due to higher ratio of surface to bulk atoms and stress caused by higher curvature at the crystallite surface.
As one of the advantages of the FSP synthesis is its simple one-step design a post-synthesis heat treatment step was undesired. Instead a high-temperature flow reactor was added to the synthesis apparatus to encourage further growth 60 of the crystallites. In this article the modified FSP set-up as well as the morphology, composition and electrochemical behaviour of the resulting nanoparticles are described. The chemicals were supplied by Sigma Aldrich and Strem Chemicals (see Table   1 in [9]) and used as supplied.
The standard FSP set-up for the synthesis of LTO was described by Karhunen et al. [9]. A premixed methane-oxygen flamelet ignites the aerosolised precursor 75 solution resulting in the formation of a high-temperature flame, with temperatures in excess of 2000 K [10]. The metallic precursor components will then nucleate and condensate to form primary particles of pure metals (e.g. silver) or oxides (e.g. Li 4 Ti 5 O 12 ).
In the standard FSP process the quenching of the particle sintering is very 80 efficient [11]. As such primary particle size remains small. However, it also stops the growth and ordering of the crystallite structures. This can lead to retardation of the Li-ion diffusion within the crystallites reported by Kavan et al. [5]. In order to promote further crystallisation a high-temperature vertical flow tube furnace (ID 25 mm, length 800 mm) was inserted above the FSP flame 85 ( Figure 1).

5
The furnace set-point was maintained at 1000 • C. This increased the hightemperature residence time from a few milliseconds to about 1 s. However, it was noted that the furnace heating element turned on only occasionally during stable operation. This indicates that the heat generated by the flame was almost 90 enough to maintain the set-point temperature at the mid point of the furnace.
The outflow from the furnace was rapidly quenched within an axial diluter (AXD) with 35 L/min of dry, particle free dilution air. The dry product powder was then collected using a Teflon bag filter (Industri-Textil Job Oy). All material characterisation as well as the electrochemical testing was conducted on the 95 collected powder as is.

Results and discussion
Pure and doped LTO nanoparticle were synthesised with the standard FSP method [10] by Karhunen et al. [9]. These particles (S-LTO) were found to consist of primary particles with high elemental and phase purity. The primary and crystallite sizes were observed to be about 10 nm, indicating single crystalline 130 composition. In the doped sample independently nucleated silver nanoparticles of about 1 nm size were observed on the surface of the LTO as primary particles.
The performance of these particles in a Li-ion secondary cell was, however, found to be inferior to that of a commercial reference LTO. The specific capacity for the pure LTO nanoparticles ranged from 93% (at 0.2C) to 4% (10C) of the 135 capacity measured for the reference LTO, while the range was 85% to 32% for the silver doped sample.
The specific capacity measurements are summarised in Figure S1 in the supplement. It is worth noting that these capacity values are not directly comparable to those obtained for the LTO synthesised with the modified FSP set-up 140 (HT-LTO) as the measurements were carried out by a third party using a different cell assembly and cycling program. However, it is possible to compare the performance by studying the capacities relative to the commercial reference The poor performance was probably due to the extremely small crystallite size of the S-LTO, detemined to be about 10 nm. It is also worth noting that this crystallite size determined from XRD data is probably an overestimate; the XRD measurement tends to weight larger particles more due to the way the x-ray radiation interacts with the crystallite planes. 150 Kavan et al. [5] showed that the optimum LTO crystallite size for Li-ion cell applications is about 20 nm, about twice that obtained for the S-LTO.
However, this optimum is only relevant for current densities (> 100C) greatly exceeding the typical operating conditions of the cells (< 10C). They also found that for smaller particles sizes the the specific capacity falls off rapidly. Thus, it 155 was deemed likely that the electrochemical performance of the FSP synthesised       14 lengths within the nanoparticles.

220
For the silver doped HT-LTO the increase in the capacity is even more evident ( Figure 7). This improvement increases with increasing current densities from about 3% at 0.2C to about 19% at 10C. This enhancement in performance can be attributed to a better electrical conductivity of the LTO material provided by the Ag-doping. The higher conductivity becomes particularly im-225 portant for high current operation.

Conclusion
It was shown by [9] that pure and silver doped Li 4 Ti 5 O 12 nanoparticles can be synthesised using the standard flame spray pyrolysis method. However, these particles exhibited poorer than expected electrochemical performance when used 230 in Li-ion half cells. The likely cause of the poor performance was determined to be the extremely small crystallite and primary particle size of the product.
To overcome the performance limitation a modified FSP system was designed to encourage growth of the crystallites. This was achieved by adding a vertical flow furnace into which the combustion gases from the FSP flame were 235 drawn. The furnace was maintained at 1000 • C to extend the high-temperature residence time of the nanoparticles.
This high-temperature synthesised LTO was found to exhibit significantly improved specific capacity compared to LTO synthesised with the standard FSP method, especially at high current rates. Furthermore, the capacity of the 240 HT-LTO matched or exceeded the performance of a commercial reference LTO powder.
With a silver doping the HT-LTO nanoparticles provided even better electrochemical performance, exceeding that of the reference material by 19% at a C-rate of 10. This high performance at high currents can be attributed to a 245 combination of the short diffusion paths due to the small primary particle size and to the improved conductivity of the material due to the silver doping.
Thus it can be concluded that to obtain the best performance for LTO nanoparticles by FSP synthesis steps need to be taken to achieve a crystallite and primary particle size optimised for the expected operational current densi-250 ties. For very low current operation larger particles can provide more lithium intercalation sites. However, as the current densities increase the sites with in the particle core may become inaccessible due to the relatively long diffusion path lengths. Thus, decreased particle size is required for the higher C-rate operation.

255
A vertical flow furnace added to the FSP system can provide a convenient way to extend the high-temperature residence time of the particles and thus encourage growth of the crystallites and primary particles. The residence time and furnace temperature can then be optimised to achieve particles of the desired size.

260
Finally it was observed that doping the LTO nanoparticles with a highly conductive additive can increase the specific capacity of the material, especially at high current conditions. Thus it can be concluded that a doped HT-LTO could provide a good alternative for battery application where high-power performance and long cycle-life is required.  is poorer compared to BF-TEM images, the enhanced mass contrast was used to highlight the heavier elements in the image.