Patent Application
20090183650
This invention relates generally to formation of non-amorphous carbon films, and, more specifically, to methods for forming structured carbon films on electrode material particles and the composite materials that result therefrom.
In 11997, LiFePO4 (triphylite) was introduced as a possible cathode for rechargeable Li-ion batteries. The abundant availability of Fe at a relatively low price and the good environmental compatibility make this an attractive alternative to currently used LiCoO2 or LiMn2O4 cathodes. LiFePO4 has a flat voltage plateau at about 3.4 V against Li and a theoretical capacity of 170 mAh/g. The practical discharge capacity of LiFePO4 electrodes, however, is often much lower than the theoretical value, especially when high current densities are used. This has been attributed to low mobility of Li ions across the FePO4/LiFePO4 interface and a low electronic conductivity of ˜10−9 S/cm.
LiFePO4 is of interest as a cathode material for Li-ion batteries intended for large-scale applications such as hybrid electric vehicles (HEVs) because of its potential for low cost and improved safety. To fulfill this promise, however, the power capability of this material needs to be improved. A factor limiting the performance of LiFePO4 is its low electronic conductivity, calculated to be about 10−9 S/cm at room temperature.
Several approaches have been used to address the rate limitations of LiFePO4 electrodes. One of the more promising of these involves producing very small particles and coating them with conductive carbon. For example, a capacity of 160 mAh/g at C rate during cycling at 80° C. for a sample with 1 wt. % carbon coating has been reported. A strong correlation between particle size and performance has also been reported. The best performing materials were synthesized below 600° C., where particle growth is inhibited. Capacities of 120 mAh/g at 5C rate were recently reported for carbon-coated LiFePO4 with particle sizes ranging from 100-200 nm.
It has been shown that carbon-coating LiFePO4 particles results in greatly improved room-temperature electrochemical performance. One of the most promising avenues is the addition of conductive carbon either postsynthesis (e.g., by cogrinding) or by cofiring with organic or polymeric additives to produce coated particles. It is useful to keep the amount of carbon low in order to avoid decreasing the energy density unduly. An interesting observation is that electrode performance does not always track with the amount of carbon in LiFePO4/C composites. The structure of the carbon, particularly the sp2/sp3 and disordered/graphene (D/G) ratios, strongly influences the electronic conductivity of the electrode. Samples containing smaller amounts of high quality carbon (i.e., those having high sp2/sp3 and low D/G ratios) can outperform those containing larger amounts of less conductive carbon. It has also been shown that the structure of the carbon in the composites produced by cofiring depends upon the source (i.e., type of organic or polymeric precursor) as well as the processing conditions. Careful attention to the carbon structure may make it possible to optimize an electrode by using a minimum amount of carbon while maximizing the conductivity of the LiFePO4 composite. Other useful cathode materials include LiFePO4, LiNi1/3Co1/3Mn1/3O2, LiCoO2, LiNiO2, LiNi0.8Co0.2O2, and LiNi0.8Co0.15Al0.05O2.
Higher electronic conductivity scales with lower D/G ratios and increased sp2/sp3 ratios in carbon. There is a correlation between the structure of carbon in LiFePO4 samples and the utilization upon discharge in lithium cells at room temperature. Significantly, some materials with low amounts of carbon and low D/G ratios outperformed those with more carbon having a more disordered structure. While optimizing the carbon structure may be key to obtaining good performance, it is difficult to produce highly structured (graphitic) coatings at the relatively low temperatures (600-800° C.) used for synthesis of LiFePO4. Thus there is a need for methods and materials to produce structured carbon coatings at temperatures low enough to use with LiFePO4 without risk of degrading the LiFePO4. Such structured carbon coatings would be useful also in a number of other applications.
The electrochemical performance of LiFePO4 can be greatly enhanced when the structure of the in situ carbon covering the particles is improved. As disclosed in the embodiments of the invention herein, a method is provided for achieving this improvement whereby small amounts of pyromellitic acid and a graphitization catalyst such as iron nitrate, ferrocene or a derivative thereof are added during processing. The overall carbon content of the resulting product, though still below 2 wt %, has a higher graphene content and the hydrogen to carbon (H/C) ratio is reduced compared to materials prepared without the additives. Optimization of the carbon structure makes it possible to maximize the conductivity of LiFePO4 particles with a minimum amount of carbon.
The figures are for illustrative purposes only and are not drawn to scale.
The low temperatures (typically ˜600-750° C.) used in the preparation of LiFePO4 have presented a challenge for the co-production of well-ordered, graphitic carbon in-situ. Nevertheless, as will be disclosed herein, it is possible to manipulate the synthesis conditions to produce carbons with desirable characteristics. In particular, the judicious selection of carbon sources and graphitization catalysts can result in markedly improved coatings.
LiFePO4 may be prepared by a number of different routes, including hydrothermal synthesis, carbothermal reduction, sol-gel, or aqueous precipitation routes, microwave processing, and solid-state synthesis under an inert or reducing atmosphere. Samples made from precursors with organic moieties (oxalates, acetates, etc.) or processed in plastic containers typically contain small amounts of residual (in situ) carbon from pyrolysis of the organics or polymers.
The amount of residual carbon present in samples varies in complex ways with the furnace conditions; some is lost as CO or CO2 during carbothelimal reduction, particularly if Fe(III) species are present. Near or above 800° C., these processes may result in the formation of iron carbide, iron phosphocarbides and/or iron phosphides from reaction with LiFePO4 itself. Nevertheless, the presence of carbon or carbonaceous materials during synthesis is beneficial as grain growth is inhibited and oxidation of iron by adventitious oxygen is slowed or prevented.
In one exemplary embodiment, samples were prepared as follows: LiFePO4 powder was produced using a sol-gel method starting from Fe(NO3)3.9H2O (Sigma Aldrich, 98+%), C2H3O2Li.2H2O (Sigma Aldrich), and H3PO4 (EMP). Precursors were dissolved in water in a stoichiometric ratio of 1:1:1 and then combined with two stoichiometric equivalents of HOCH2CO2H (glycolic acid, Sigma Aldrich, 70% solution in water). The pH was adjusted to between 8.5 and 9.5 using NH4OH (EMD) and the water evaporated on a hot plate. The sample was then calcined at 500° C. for 10 hours under a flow of nitrogen before being planetary ball milled with 6 wt. % C6H2(CO2H) (pyromellitic acid, Sigma Aldrich) and 1 wt. % ferrocene (Alfa Aesar, recrystallized) for 1 hour in acetone. The solvent was then evaporated and the resulting powder fired to 600° C. under flowing nitrogen for an additional 10 hours. Further details on synthesizing LiFePO4 can be found in the publications, M. M. Doeff, Y. Hu, F. McLarnon, and R. Kostecki, Electrochem. and Sol. State Lett. 6 (2003) A207-A209 and Y. Hu, M. M. Doeff, R. Kostecki, and R. Fifiones, J. Electrochem. Soc., 151 (2004) A1279-1285, both of which are included by reference herein.
Up to 8 wt. % pyromellitic acid (C6H2(CO2H)4, Sigma Aldrich, abbreviated PA from this point forward) and some or all of iron nitrate, ferrocene (C10H10Fe, Alfa Aesar, recrystallized), and ferrocenecarboxylic acid (C11H10O2Fe, Aldrich, 97%, abbreviated FCA from this point forward), ranging from 0.001-1 wt % were added to various samples. The mixtures were ground using a planetary ball mill in an appropriate solvent (ethanol or acetone) for one hour. The grinding solvent was then evaporated under a flow of nitrogen and the resulting powder was thoroughly mixed and fired to 600° C. for ten hours.
Phase purity was determined by X-ray diffraction (XRD) on the resulting powders using a Philips X'Pert diffractometer using an X'Celerator detector with Cu Kα radiation (λ=1.54 Å). Particle size distributions were resolved by means of a Beckman Coulter particle size analyzer (model LS 230) using Darvan*C (ammonium polymethacrylate, aqueous solution, R.T. Vanderbilt Company Inc.) as a dispersant. Particle morphology studies were conducted using a field emission-scanning electron microscope (FE-SEM, Jeol JSM-6340F). Luvak Inc. (Boylston, Mass.) was used to conduct the elemental analyses (carbon and hydrogen) of several samples, reported as H/C ratios in Table I.
An integrated confocal Raman microscope system, “Labram,” made by ISA Group Horiba was used to analyze the structure of the active materials. Raman spectroscopy measurements were carried out at room temperature in ambient atmosphere using an internal He—Ne 632 nm 10 mW laser as the excitation source. The power of the laser beam was adjusted to 0.1 mW with neutral filters of various optical densities. The size of the laser beam at the sample was ˜1.2 μm, and the average acquisition time for each spectrum was 25 seconds. The resolution of this instrument is approximately 1.7 cm−1. Baseline correction and deconvolution analysis of Raman spectra were performed with a commercial software package (PeakFit, version 4.05, SPSS Inc.).
Pressed pellets for conductivity studies were fabricated by uni-axially pressing 0.5 g of active material to 10 kpsi in a 0.5 inch (1.27 cm) stainless steel die. The pellets were then transferred into balloon holders and cold isostatically pressed to 180 kpsi achieving a final density of 70% of the theoretical LiFePO4 density (3.6 g/cm3). Thin gold electrodes were then sputtered on to each face of the pellet using a Bal-Tec SCD 050 sputter coater. AC impedance spectra were obtained using a Solartron Instruments 1260 impedance/gain-phase analyzer at selected temperatures between 25 and 200° C. Conductivities were derived from the intercepts of the capacitative arcs with the z′-axes in the Nyquist plots.
Electrode samples were formed from 80 wt % active material, 8 wt % Kynar poly(vinylidene fluoride) (PVDF) (Elf Atochem North America Inc., Technical Polymers Department), 6 wt % SFG-6 synthetic flake graphite (Timcal Ltd., Graphites and Technologies), and 6 wt % acetylene black. This fabrication process has been described previously in M. M. Doeff, Y. Hu, F. McLamon, and R. Kostecki, Electrochem. and Sol. State Lett. 6, A207 (2003), which is included by reference herein. Electrodes were cast as a slurry in 1-methyl-2-pyrrolidinone (Sigma Aldrich, 99%) onto aluminum current collectors and dried for 24 hours in air followed by 12-24 hours in a vacuum oven at 120° C. Cathodes with an area of 1.8 cm2 were punched from the cast electrode and typically had loadings of about 1 mAh/cm2. Assembly of lithium half-cells in 2032 coin cells was performed in a helium filled glove box using 1 M LiPF6 in 1:2 ethylene carbonate/dimethylcarbonate (EC/DMC) electrolyte solution and a Celgard 3401 separator. At least two cells of the same type were tested for each material to ensure reproducibility. Electrochemical studies were undertaken galvanostatically using an Arbin BT/HSP-2043 and/or a Macpile II (Bio-Logic, S.A., Claix, France) automated cycling data recorder between 2.0 and 3.9 V at room temperature. Cells were charged at a current density corresponding to C/25 and allowed to rest 15 minutes between half-cycles.
All powders were determined to be phase-pure by XRD analysis. The primary particle sizes found in the powders of all samples were highly variable, ranging from less than 100 nm to more than 1 μm, as can be seen in the SEM image in
Raman spectroscopy is a particularly useful tool for characterizing the near-surface structure (i.e. disorder and crystallite formation) of carbon films because carbon is a relatively strong scatterer with two E2g modes predicted to be Raman active.
It is a common practice to use polymeric or organic additives as carbon sources during synthesis of carbon coated LiFePO4. Some organic or polymeric precursors do not decompose completely at the low synthesis temperatures used to make LiFePO4. Residual hydrogen and functional groups on carbon lower the electronic conductivity, resulting in electrode materials with poor electrochemical performance. Raman spectra and C, H, and N elemental analyses of LiFePO4 powders processed with poly(acrylonitrile), perylenetetracarboxylicdianhydride or other well-known graphite precursors, show that these additives do not decompose sufficiently at the relatively low synthesis temperatures (˜700-800° C.) to form a carbon coating with good conductivity. In some cases, the addition of the precursors actually results in electrode materials with electrochemical performance inferior to that of samples processed without additives. In contrast, pyromellitic acid (PA) decomposes readily, as evidenced by lower H/C ratios in the
resulting products when compared to those processed with the above-mentioned additives. The quality of the carbon for samples processed with PA also improves over that for those synthesized with no additives, as can be seen by the increased sharpness and intensity of the G-band in the Raman spectrum of the former (compare middle and bottom spectra in
Table I shows carbon contents, H/C ratios, D/G and sp2/sp3 ratios for selected LiFePO4 samples processed with pyromellitic acid (PA), both with or without graphitization catalysts.
aS.D. = standard deviation.
As shown in Table I, the overall carbon content in the final products generally increases somewhat as more PA is used, although this is very dependent upon the furnace conditions. H/C ratios also rise, particularly above 8 wt %, indicating that complete decomposition becomes more difficult for large amounts of PA. The addition of PA during synthesis results in a modest increase in the carbon content and a general decrease in the H/C ratio with some sample variation, close to that of PA itself (0.05). The latter indicates the degree of decomposition of the organic components in the synthesis mixture, and suggests that better quality carbons are produced from the additive, due, in part, to its lower hydrogen content. Also included in Table 1 are carbon structural parameters determined from analysis of the Raman spectra obtained on the various LiFePO4 samples.
It should be noted that the D/G (disordered/graphene) and sp2/sp3 ratios determined by analysis of the Raman spectra and shown in Table I do not yield the actual ratios but rather values that can be correlated to these structural parameters. Thus, the ratios are useful for comparing samples to each other but not as quantitative measures of the graphene or sp3 contents. Based on this analysis, the data in Table I show that D/G ratios decrease and sp2/sp3 ratios increase as more PA is added, up to about 6 wt. %. Thus, the structure of carbon produced when PA is present during calcination is markedly different from that produced from the precursors alone.
It is well known that some iron compounds can catalyze the formation of graphite at relatively low temperatures. Graphite may precipitate upon decomposition of Fe3C (cementite) near 650° C. during the production of cast iron, in a process known as “dusting”. Furthermore, carbon nanotubes, which consist of curled graphene sheets, can be made at temperatures as low as 600-700° C. using organic or polymeric carbon sources and iron compounds as promoters.
These observations explain the variability in the in situ carbon structure found in LiFePO4 samples processed similarly, since iron oxides are common surface impurities. Anything more than trace oxidation of LiFePO4 samples during synthesis is clearly undesirable, severely limiting the options for producing graphitic carbon this way. Instead, addition of small amounts of graphitization catalysts such as iron nitrate, ferrocene, or ferrocene derivatives along with PA during LiFePO4 synthesis can be used to improve the carbon structure, as shown in
The structure of the in situ carbon influences the electrochemical behavior of LiFePO4 samples. Electrode utilization rises as D/G ratios and the amorphous carbon content decreases (i.e., the electronic conductivity increases). The observation that some samples with low carbon contents outperform those with larger amounts of poor-quality carbon is significant, and suggests that the amount of coating necessary to ensure good high-rate performance can be minimized provided that the structure is optimized. A considerable challenge is the temperature limitation (<750-800° C.) imposed by LiFePO4 synthesis conditions. For example, the graphene content and electronic conductivity are low for carbons prepared from polymeric precursors at temperatures below about 700° C., but increase dramatically above this temperature. However, the considerable variations found in the in situ carbon of LiFePO4 samples suggest that much can be done to manipulate the structure, even considering the temperature constraints.
In one embodiment, good rate behavior is obtained when LiFePO4 is processed with approximately 4-8 wt % PA, which yields materials with in situ carbon content below 1 wt %. In another embodiment, good rate behavior is obtained when LiFePO4 is processed with approximately 1-10 wt % PA. In another embodiment, good rate behavior is obtained when LiFePO4 is processed with approximately 1-20 wt % PA. There is a correlation between pressed pellet conductivities measured by AC impedance and the rate performance, but not necessarily with the amount of carbon. Secondary particles in these samples have a lava rock-like appearance, and size distributions are wide. Processing with PA does not appear to change the primary carbon particle size (˜200 nm) significantly, indicating that the observed rate effects are indeed due to the improved C structure.
The improvement in carbon structure, pressed pellet conductivities and electrochemical performance seen in samples prepared with PA is striking, rate limitations are still evident. Optimum performance may still be achieved with further improvements in the carbon structure and/or increases in the carbon content.
In an exemplary embodiment, samples calcined with small amounts of iron nitrate (0.001-0.01 wt. %) and PA produced powders with C contents below 1 wt. %, similar to those calcined with PA alone (see Table I). The D/G ratios are not significantly changed from those of samples processed with similar amounts of PA, but the sp2/sp3 ratios are higher. This suggests that, while the graphene domain sizes are not changed from materials prepared with PA only, there is a greater proportion of carbon with a graphitic nature. It is possible to assess the effect of the increased sp2 character on the electronic conductivity by comparing the results for two composites with identical carbon contents (0.71%), one processed with iron nitrate, and one without. As shown in
LiFePO4 processed with PA and higher concentrations of iron nitrate performed worse than the materials made with 0.01% iron nitrate or less as shown in Table I. When iron nitrate (or the oxides that form from thermal decomposition thereof) is located in carbon-rich areas on the surfaces of the powders, graphitization can occur. But too much iron nitrate can result in an overabundance of resistive iron oxide, which can affect the electrochemical performance adversely. Conversely, carbon-rich areas not in contact with the catalyst tend not to graphitize at the relatively low temperatures used to synthesis LiFePO4. A low concentration of iron nitrate is preferable to prevent excessive formation of iron oxide. Homogenous dispersion within the carbon source is useful in order to achieve maximum graphitization. For some materials processed with PA and iron nitrate, considerable spot-to-spot variations in the Raman spectra were observed, indicating that the quality of the carbon was not uniform throughout the powder. Consequently, such materials do not show any appreciable improvement in the electrochemical performance and are not considered further here.
Ferrocene and its derivatives have graphitization advantages over iron nitrate as they contain both catalyst centers and carbon sources within the same molecules, which helps to overcome difficulties in ensuring close contact between the two entities. They have been used extensively in the synthesis of nano-structured carbon materials, and as soot control agents for cleaner burning fuels. A composite prepared with 1 wt. % ferrocene, however, has a low carbon content (see Table I), only slightly higher than that of LiFePO4 prepared without any additives. Ferrocene sublimes at about 175° C., so much of it is lost during calcination under the synthesis conditions described herein. The quality of the resulting carbon is poor, judging from the structural factors listed in Table I, although its thorough decomposition is evidenced by a low H/C ratio. In contrast, LiFePO4 prepared with either ferrocene or FCA as well as PA has significantly higher carbon content than materials prepared with the same amount of PA alone (see Table I), although the final carbon content is still less than 2 wt. %. This suggests that an interaction between ferrocene and PA occurs upon heating, which results in improved retention of elemental carbon. The H/C ratios of these composites are lower than that of the starting materials, ferrocene (0.08) and FCA (0.075) themselves, indicating nearly complete thermal decomposition. The carbon structural factors, as determined by Raman spectroscopy, are similar to those of composites made with PA alone, although the sp2/sp3 ratios are less than that of the materials processed with iron nitrate. Thus, the several orders of magnitude increase in conductivity of a ferrocene/PA-treated LiFePO4 pressed pellet compared to the others in
The results presented here show that the composite conductivity, and therefore, the electrochemical performance, is related to the amount, structure, and distribution of the carbon in the coatings on the LiFePO4 particles. In some cases, composites containing less carbon are more conductive than those with more, due to a higher sp2 character. This translates directly into improved electrochemical performance. The best results in this study, however, were obtained for composites with carbon contents above 1 wt. %. This may be due, in part, to the fact that it is easier to obtain complete coverage over particle surfaces when more carbon is present. Provided that a relatively homogeneous coating can be produced by, for example, better mixing prior to calcination, what is the lowest carbon content needed to produce a high-rate LiFePO4? How much improvement in the carbon structure is necessary to allow a substantial decrease in the content? The pressed pellet conductivity data, although fairly limited, offers some insights into these questions.
According to an embodiment of the invention, a method for forming a structured carbon film on particles includes providing precursors for cathode material. The precursors include one or more Li sources, one or more metal sources, and an anion source. Cathode materials include any materials of interest for use in Li ion batteries, such as phosphates, sulfates, silicates, and oxides. When the cathode materials are heated in the presence of carbon, reduction of oxides can occur. It is useful to try to avoid reduction of oxides during processing. Precursors for cathode materials include Li sources, such as Li nitrates and acetates, metal sources, such as nitrates, acetates and oxylates of Fe, Mn, Co, Ni etc. or combinations of these, anion sources, such as nitrates, acetates, oxides phosphates, and oxylates.
The precursors are mixed with pyromellitic acid and, optionally, a graphitization catalyst and the composite mixture is fired in an inert atmosphere between about 500° C. and 700° C.
In another embodiment of the invention a method for forming a structured carbon film on particles includes providing substrate particles, milling the substrate particles with pyromellitic acid and, optionally, a graphitization catalyst and firing the composite powder in an inert atmosphere between about 500° C. and 700° C. In one embodiment, the pyromellitic acid comprises between approximately 1 and 10 wt % of the composite powder. The graphitization catalyst can be iron-based; it can be any of ferrocenecarboxylic acid, ferrocene, and iron nitrate. In one embodiment, wherein the graphitization catalyst comprises between approximately 0.001 and 5 wt % of the composite powder. In another embodiment, the graphitization catalyst comprises between approximately 2 and 10 wt % of the composite powder. In yet another embodiment, the graphitization catalyst comprises between approximately 4 and 6 wt % of the composite powder. In one arrangement, the firing is done at approximately 600° C.
In another embodiment of the invention, a method of forming a composite electrode, includes providing active electrode material particles, milling the active electrode material particles with pyromellitic acid and ferrocene or iron nitrate to produce a composite powder, firing the composite powder between 500° C. and 700° C., thus forming a carbon-coated electrode powder, and applying the composite powder to a current collector. The active electrode material particles can be any of LiFePO4, LiMnPO4, LiCoPO4, LiNiPO4, and combinations thereof. In another arrangement, the active electrode materials can be any of LiNi⅓Co⅓Mn⅓O2, LiCoO2, LiNiO2, LiNi0.8Cu0.2O2, LiNi0.8Cu0.15A10.05O2, and combinations thereof.
In another embodiment of the invention, a material composition includes substrate particles and a carbon coating on the particles, wherein, the carbon coating has a structure with an sp2/sp3 ratio greater than about 0.12, and wherein the overall composition contains no more than 10 wt % carbon. In some arrangements, the carbon coating has a structure with a D/G ratio less than about 1.19. In one embodiment the overall composition contains no more than 5 wt % carbon. In another embodiment, the overall composition contains no more than 2 wt % carbon. The substrate particles can be phosphates, sulfates, silicates, and oxides. In one embodiment, the substrate particles comprise LiFePO4.
Further experimental details can be found in the Journal of the Electrochemical Society, 154 (5) A389-A395 (2007), “Factors Influencing the Quality of Carbon Coatings on LiFePO4”, which is included by reference herein.
This invention has been described herein in considerable detail to provide those skilled in the art with information relevant to apply the novel principles and to construct and use such specialized components as are required. However, it is to be understood that the invention can be carried out by different equipment, materials and devices, and that various modifications, both as to the equipment and operating procedures, can be accomplished without departing from the scope of the invention itself.
This application claims priority to PCT Application PCT/US2007/071054, filed Jun. 12, 2007, which in turn claims priority to U.S. Provisional Patent Application 60/804,560, filed Jun. 12, 2006, both of which are incorporated by reference herein.
The invention described and claimed herein was made in part utilizing funds supplied by the U.S. Department of Energy under Contract No. DE-AC02-05CH11231 and in part utilizing funds provided by DARPA grant HR0011-04-1-0029. The Government has certain rights in this invention.
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/US07/71054 | 6/12/2007 | WO | 00 | 11/25/2008 |
| Number | Date | Country | |
|---|---|---|---|
| 60804560 | Jun 2006 | US |
