KR100477400B1 - Method for manufacturing Lil + xMn2-xO4 for secondary battery electrode - Google Patents

Abstract

The present invention relates to a method for preparing a lithium manganese oxide intercalation compound represented by the formula Li _{1 + x} Mn _2-x O ₄ , the method of the present invention is to sufficiently mix lithium hydroxide or lithium salt and manganese oxide or manganese salt Supplying sufficiently mixed salts to the reactor 5, continuously stirring the mixed salts in the reactor 5, and stirring the mixed salts in the reactor 5 in the range of about 650 ° C to about 800 ° C. Heating in an oxygen containing atmosphere for a time not exceeding about 4 hours at a temperature of and cooling the reaction product to about 200 ° C. or less for a time not exceeding about 2 hours in an oxygen containing atmosphere.

Description

Method for manufacturing L i 1 + x M n 2-x O 4 for secondary battery electrodes

The present invention relates to a continuous process for producing fine powders and / or films of lithium containing ternary oxides. More specifically, the present invention relates to the synthesis of Li _{1 + x} Mn _2-x O _4, an intercalation compound of interest for secondary batteries.

Conventionally, lithium-containing ternary hydroxides were prepared by mixing carbonates and oxides of constituent compounds and heating the mixture at high temperature. This method produces an effective cell material but is not commercially viable due to the long reaction and cooling times.

FIELD OF THE INVENTION The present invention relates to secondary, rechargeable lithium and lithium ion batteries, and more particularly Li _{1 + x} Mn _2-x O, wherein x is from about 0.022 to about 0.20 for use in the positive electrode of such cells. _It relates to a continuous process for preparing _four interlayer compounds.

Lithium cobalt oxide is commonly used as the positive electrode material in commercial 4-volt lithium ion batteries. Based on low cost, abundant raw materials, incidental safety, environmental acceptability and electrochemical performance, Li _{1 + x} Mn _2-x O ₄ interlayer compounds have been spotlighted as positive materials in such cells. However, in order for Li _{1 + x} Mn _2-x O ₄ to be commercially successful as a cathode material, it is necessary to develop a method for rapidly and economically producing a material having electrochemical performance characteristics, which has not been found before. The present invention addresses this problem.

LiMn ₂ O ₄ (x = 0 Li _{1 + x} Mn _2-x O ₄ ) is a sufficient mixture of Li ₂ CO ₃ and any manganese oxide as early as 1958 with a Li / Mn molar ratio of 0.50. The mixture was reacted under air at 800-900 ° C., then ground again and the mixture was synthesized by reacting the mixture at this temperature until the sample reached a constant weight [DG Wickham and WJ Croft, J. Phys. Chem. Solids 7 (1958) 351-360. The method of eluting LiMn ₂ O ₄ with acid to produce λ-MnO ₂ having a LiMn ₂ O ₄ crystal backbone, and subsequently using λ-MnO ₂ as a positive electrode material in a lithium battery, has been reported by Hunter [JC. Hunter (Union Carbide), US Patent No. 4,246,253, 20 January 1981; JC Hunter (Union Carbide), US Patent No. 4,312,930, January 26, 1982; JC Hunter, J. Solid State Chem. 39 (1981) 142-147. Hunter electrochemically reduced the λ-MnO ₂ to LiMn ₂ O ₄ , which reaction occurs at 4V, but they do not cycle the cell. Hunter also found that lithium and manganese compounds other than those specified in the literature of Wickham and Croft can be used in the synthesis as long as the compounds can degrade lithium or manganese oxide under the reaction conditions used. . Tachray et al., M. Tucker Ray, P. Johnson, L, De Fitchauto, P Bruce and J. Guduff, Mat. Res. Bull. 19 (1984) 179-187; M. Tuckerley, L. de Pichotto, A. De Cork, P. Johnson, V. Nicholas and K. Adendorff, J. Power Sources 21 (1987) 1-8, LiMn ₂ O ₄ spinels The intercalation of Li into the structure is electrochemically reversible and provides two voltage plateaus under Li about 4.1 V and 3.0 V for Li, where the first Li ion and the second Li ion are λ -MnO _{2, respectively.} Corresponding to intercalation / deintercalation into.

Many researchers have studied the method of synthesizing LiMn ₂ O ₄ by thermal reaction of lithium and manganese compounds, and found that the method can be carried out over a wide range of temperatures, ie from 300 to 900 ° C. It was. The possibility of the product inserting and detaching Li was also investigated. The so-called "cold" materials produced below about 550 ° C have a low crystallinity and twisted spinel structure and cycle at about 3V for Li but not at 4V [WJ McClean, RJ Knit and RJ Powell, J. Power Sources 34 (1991) 39-49; T. Nagaura, M. Yokogawa and T. Hashimoto (Sony Corporation), US Patent No. 4,828,834, May 9, 1989; MM Tackerray and A. De Cork (CSIR), US Pat. No. 4,980,251, December 25, 1990; V. Manev, A. Momkylov, A. Nazalevska and A. Gozawa, J. Power Sources, 43-44 (1993) 551-559]. Such materials are not the subject of this application.

So-called "hot" materials produced under air at about 600-900 ° C. have high crystallinity. They exhibit cycling capacity at 4 V for Li but do not cycle well at 3 V for Li, resulting in a dramatic loss of capacity [JM Tarascon, E. Wang, JK Shokuhi, WR McKinnon and S. Colson, J. Electrochem. Soc. 138 (1991) 2859-2868]. As with the sol-gel process, even when LiMn ₂ O ₄ is synthesized at low temperatures, the material can be cycled according to the 4 V scheme when first calcined / annealed at a high temperature, for example 600 to 800 ° C. [P. Barbooks, FK Shokuhi and JM Tarascon (Belcore), US Patent No. 5,135,732, 4 August 1992]. The high temperature LiMn ₂ O ₄ material will be the subject material in the remainder of this application.

In general, many researchers take many hours, or days of reaction time, to synthesize a single phase product in a (static) indirect heating tract, together with regrinding the heated product and reheating the regrind powder. Found that it was necessary [P. Bar Books, FK Shokuhi and JM Tarascon (Belcore), US Pat. No. 5,135,732, August 4, 1992; WJ McClean, RJ Knit and RJ Powell, J. Power Sources 34 (1991) 39-49; A. Mosva, A. Burberry and M. Tourux, Mat. Res. Bull. 18 (1983) 1137-1381; T. Ozuku, M. Kitagawa and T. Hirai, J. Electrochem. Soc. 137 (1990) 769-775. If this is not the synthetic procedure by this experiment, various by-products other than LiMn ₂ O ₄ are generated, namely Mn ₂ O ₃ , Mn ₃ O ₄ and Li ₂ MnO ₃ . These materials are undesirable for lithium batteries because of their low capacity and high fade ratio.

In addition to the generation of undesirable by-products, the synthesis parameters also affect the molecular / crystal structure and physical properties of LiMn ₂ O ₄ , and the properties of these materials greatly affect the cell capacity and the cycling potential of the material. Momkilov, Manev and others are described in A. Mumkyloff, V. Manev and A. Nasalevska, J. Power Sources 41 (1993) 305-314] changed the lithium reactant, MnO ₂ reactant, reaction temperature and reaction time before cooling in air. They found it advantageous to make the spinel from the lowest possible melting point lithium salt and the highest possible surface area MnO ₂ sample. The advantage is that the reaction time is much faster and the porosity of the product is higher, providing greater capacity and better cell cycling potential (ie, capacity fade decreases over cycle number of cycles). However, the reaction time is a few days in any case. The researchers also found that the optimum reaction temperature was about 750 ° C. [V. Manev, A. Momkylov, A. Nasalevska and A. Gozawa, J. Power Sources, 43-44 (1993) 551-559; A. Momkylov, V. Manev and A. Nasalevska, J. Power Sources 41 (1993) 305-314]. At higher temperatures, the material loses capacity, presumably because the surface area decreases and the loss of oxygen reduces some of the manganese in LiMn ₂ O ₄ . At lower temperatures, longer time is required for the synthesis, and evidence of spinel torsion appears, which appears to reduce capacity. The researchers also demonstrated the advantage obtained by preheating the reaction mixture at a temperature just above the melting point of the lithium reactant and then reacting at the final temperature.

Tarascon and co-workers found that Li _{1 + x} Mn _2-x O because of high capacity and high cycle life (1) Li / Mn molar ratios of 1/2 or more (ie, Li / Mn = 1.00 / 2.00 to 1.20 / 2.00). X at ₄ is 0.0 to 0.125) using a mixture of reactants, (2) heating the reactant at 800 to 900 ° C. for a long time (eg 72 hours), and (3) reacting the reaction product to an oxygen containing atmosphere High capacity by cooling the product to ambient temperature relatively quickly by cooling to about 500 ° C. under very low rates in the air, ie from 2 ° C./h to 10 ° C./h and finally (4) shutting down the urinary tract. And long cycle life can be obtained in the best form [JM Tarascon, WR Mackinon, F. Kuwar, TN Baumer, G. Amatucci and D. Giomar, J. Electrochem. Soc. .141 (1994) 1421-1431; JM Tarascon (Belcore) International Patent Application WO 94/26666: US Patent No. 5,425,932, June 20, 1995]. If oxygen is rich in the atmosphere, the rate of cooling from 800 ° C to 500 ° C can be increased to 30 ° C / h. The researchers found that the lattice parameter a ₀ of the product is a measure of the product's efficacy in the cell and that a should be about 8.23 kPa or less. In contrast Li / Mn = 1.00 / 2.00 and a = 8.247 kPa for compositions using air cooling.

Manef et al. Also found that a Li / Mn molar ratio of 1.00 / 2.00 or greater is advantageous in terms of capacity and cycling potential [V. Manev, A. Momkylov, A. Nasalevska and A. Sato, J. Power Sources 54 (1995) 323-328]. They adopted 1.05 / 2.00 as the optimum ratio. The researchers also found that the dose decreased significantly when the amount of premix / reactant increased from about 10 g to about 100 g in an indirect heating tract. They therefore planned to consume air in the urinary tract, resulting in a partial reduction of the product. This problem was alleviated by flowing air through the urinary tract. If there is too much air flow, the volume of the product is also reduced, so the air flow must be optimized to benefit. Manef et al. Have the most advantageous cooling rate of several tens of degrees per minute, which is more than 100 times faster than that used by Tarascon. After optimizing all conditions, including the use of lithium nitrate and manganese dioxide, a highly porous chemical, as a reactant, Manef et al. (Li _{1 + x} Mn _2-x O ₄ (a product of very high capacity and low fade ratio) x = 0.033). The use of lithium nitrate adversely affects the process because it releases toxic NO _x fumes during synthesis. Manefe has developed a successful synthesis method using lithium carbonate instead of lithium nitrate [V. Manef, Paper given at 9th IBA Battery Materials Symposium, Cape Town, South Africa, March 20-22, 1995. (Available abstract)], these new methods likewise require several days of reaction time.

Howard mainly discussed alternative LiMn ₂ O ₄ production facilities in terms of economic cost [WF Howard, Jr., in Proceedings of the 11th Int'1 Seminar on Primary and Secondary Battery Technology & Application, 1994 February 28-March 3, Deerfield Beach, Florida, SP Ulski and N. Marine Foods]. Howard did not develop / provide any data at all, but he suggested that the rotary tract delivers heat faster than a static oven, thus reducing the reaction time.

Using long thermal reaction times with desirable slow cooling rates is difficult to perform on a large scale, such as in pilot plants or commercial operations. Therefore, it is desirable to reduce the reaction time and cooling time while at the same time excluding unwanted by-products and to retain the desired Li _{1 + x} Mn _2-x O ₄ stoichiometry and structure as evidenced by smaller lattice parameters.

Summary of the Invention

According to the present invention, lithium salt / hydroxide and manganese oxide are mixed, and the mixture is continuously stirred while heating at a temperature of about 650 ° C. to about 800 ° C. for up to about 2 hours under an atmosphere rich in air, oxygen or oxygen. Thereafter, the product is cooled in a time of up to about 2 hours using a similar stirring means under air, oxygen or oxygen rich atmosphere, whereby the formula is Li _{1 + x} Mn _2-x O ₄ , where x is from about 0.022 to About 0.20) and a lattice parameter of about 8.235 Pa or less is provided.

Hereinafter, the present invention will be described with reference to the accompanying drawings, in which like components are designated by like reference numerals.

1 is a partial cross-sectional view of a preferred embodiment of a continuous reactor used in the process of the present invention.

FIG. 2 is a cross-sectional view of the reactor shell of FIG. 1.

3 is a cross-sectional view of a nonaqueous laboratory cell.

[Sample B] of FIG. 4 is Li _{1 + x} Mn _2-x prepared from EMD and LiOH, heated in a rotary urinary furnace at 725 ° C. in air (2 hours) and slowly cooled under O ₂ in an experimental rotary urinary furnace (24 hours). X-ray diffraction pattern of O ₄ spinel is shown.

[Sample C] of FIG. 5 is a Li _{1 + x} Mn _2-x O ₄ spinel prepared from EMD and LiOH, heated in a rotary urinary furnace at 725 ° C. in air (2 hours) and slowly cooled in the urinary furnace under air (1½ hour). The X-ray diffraction pattern of is shown.

[Sample D] of FIG. 6 shows X-ray diffraction of a Li _{1 + x} Mn _2-x O ₄ spinel prepared from EMD and Li ₂ CO ₃ heated in a rotary urinary furnace (1½ hour) under N ₂ at 725 ° C. The pattern is shown.

[Sample E] of FIG. 7 shows an X-ray diffraction pattern of a Li _{1 + x} Mn _2-x O ₄ spinel prepared from EMD and Li ₂ CO ₃ heated in a stationary layer at 725 ° C. in air (2 hours) and air cooled. It is shown.

[Sample I] of FIG. 8 shows the X-ray diffraction pattern of a Li _{1 + x} Mn _2-x O ₄ spinel prepared from EMD and LiOH heated in a stationary layer at 725 ° C. in air (2 hours) and air cooled will be.

[Sample K] of FIG. 9 is a control Li _{1 + x} Mn _2-x O ₄ prepared from EMD and Li ₂ CO ₃ , heated in a stationary bed at 725 ° C. in air (24 hours) and slowly cooled (36 hours). The X-ray diffraction pattern of the spinel is shown.

[Sample C] in FIG. 10 is a graph showing the cycling curve (time versus voltage) for the spinel over a 2 hour cycle.

[Sample K] of Fig. 11 is a graph showing the relationship between the discharge capacity and the number of cycles showing the manner of obtaining a fade ratio of 50 cycles (minimum square).

The present invention comprises the steps of sufficiently mixing the lithium hydroxide or degradable lithium salt and manganese oxide or degradable manganese salt in a stoichiometric amount based on the formula of lithium manganese oxide; Feeding the sufficiently mixed mixture to the reactor; Continuously stirring the mixed salt in the reactor; Flowing air, oxygen or oxygen rich gas through the reactor; Heating the mixed compound stirred in the reactor at a temperature of about 650 ° C. to about 800 ° C. for a time not exceeding about 4 hours, preferably for a time not exceeding 2 hours; And cooling the reaction product to below about 100 ° C. under controlled conditions, wherein the formula Li _{1 + x} Mn _2-x O ₄ , wherein 0.022 ≦ _x ≦ 0.20, advantageously 0.022 ≦ x ≦ 0.125 It relates to a continuous process for preparing a lithium-containing manganese oxide interlayer compound represented by a single phase. The present invention is also represented by the formula Li _{1 + x} Mn _2-x O ₄ , wherein 0.022 ≦ _x ≦ 0.20, advantageously 0.022 ≦ _x ≦ 0.125, and is essentially single having a cubic spinel type crystal structure. A method for synthesizing a commercial lithium manganese oxide is provided. In particular, the present invention relates to a process for synthesizing said oxides to produce oxides suitable for use as cathodes in electrochemical cells having an anode comprising lithium or a suitable lithium containing alloy. In addition, the present invention relates to an oxide produced by the above method and an electrochemical cell using the oxide as a cathode thereof.

According to the present invention, a method for synthesizing a lithium manganese oxide having a spinel crystal structure includes at least one lithium hydroxide or lithium salt as described herein and at least one manganese oxide or manganese salt as described herein. Forming a mixture of finely divided solid forms, and heating the mixture at a temperature in the range of about 650 ° C. to about 800 ° C. to react the compounds with one another by co-decomposition to form a spinel crystal structure and a cubic closest packed oxygen lattice. Obtaining the lithium manganese oxide having a structure.

In the present specification, the lithium salt refers to a lithium compound that decomposes when heated in air to form lithium oxide. Likewise, the manganese salt herein refers to a manganese compound that decomposes when heated in air to form manganese oxide.

The lithium compound may be a component selected from the group consisting of LiOH, Li ₂ CO ₃ , LiNO ₃ and mixtures thereof, and the manganese compound may be MnO ₂ (prepared by electrolysis or chemically), MnO, Mn ₂ CO ₃ , May be a component selected from the group consisting of Mn ₃ O, Mn ₃ O ₄ , manganese acetate, and mixtures thereof. Since the mixture can be produced in stoichiometric ratios, the Li: Mn molar ratio is at least about 1: 1 and lithium is present in a slight excess, ie, the ratio is from about 1.022: 1 to about 1.200: 1. desirable. The mixture can be prepared by mixing in a rotary drum mixer, vibratory mill, jet mill, ball mill, and the like, as long as the salts are mixed intimately intimately.

The intimately mixed compounds are then transferred to a hopper and transferred to the reactor by a screw feeder, a compressed air conveyor, a pulsed air jet, or the like.

The reactor is advantageously a horizontal rotary calcination apparatus, a horizontal calcination apparatus with a rotary screw, a fluidized bed, a heated vibratory conveyor belt, or a cascade of vertical rotary tracts. The choice of reactor type will depend on the process parameters and the salt used.

Referring to FIG. 1, according to one embodiment of the invention, the starting material 1 is injected into the feed hopper 2. This material falls into the screw conveyor 3 by the action of gravity, which is used to control the feed rate of the starting material into the reactor. The screw conveyor 3 discharges the starting material into the rotary shell reactor 5. The shell 5 can be rotated by conventional rotary drive means. This solid moves downward along the length of the rotatable shell 5 and first passes through the separately heated regions 6a, 6b, 6c surrounded by the outer shell. This solid exits the rotatable shell 5 via the valve 9 and is fed into the product drum 10. The reactor is airtight between the feed screw conveyor 3 and the discharge valve 9, in which the pressure in the shell 5 is almost atmospheric due to the exhaust to the bag filter 11 through the pipe 8. Although shown to be, it may also be present under some amount of pressure from the atmosphere that comes into contact with the product before exiting the pipe 8. The optional gas scavenging inlet pipe 7 allows the countercurrent flow of air or oxygen rich gas to be continuously flowed over the reactants. The scavenging gas is exhausted from the reactor together with the waste gas which has passed through the pipe 8. The drive means for the screw conveyor and the rotatable shell are not shown, as they are well known to those skilled in the art.

Referring to FIG. 2, the rotatable shell 220 optionally has four equally spaced lift vanes 210 attached to the interior of the shell. The lift vane may be welded to the rotatable shell 220 or otherwise suitably attached. The wings are each spaced at equal intervals from their adjacent wings, each extending axially over the entire length of the rotatable shell 220. The number and size of such wings can vary considerably as long as the wings maintain solids in intimate contact with the atmosphere inside the shell. The wings also help to move the solid through the shell 220. In order to make it easier to move the solid downstream through the shell 220, the shell is inclined down along the flow direction at a ramp of up to ¼ inch per foot of reactor length, preferably 1/16 inch per foot. Can be achieved. The size of the reactor may vary depending on the desired capacity.

The experimental unit used in the examples consisted of a horizontal rotary tube tract with a 2 "diameter and 3 'heating zone. The gas flow rate was set at 50 cc / min to 500 cc / min and the rotational speed was set at 30 RPM. It was.

A similar pilot unit consisted of a 6 "diameter reactor with an 8 'heating zone. The gas flow rate was set at 20-40 SCFH. The rotation speed was set at 3-10 RPM.

It is important to keep the reactants stirred during the treatment. The fluidization action may enable rapid heat transfer and provide continuously regenerated gas / surface interface exposure. By combining these conditions, the kinetics of the process can be greatly increased compared to the stationary bed process.

The step of heating the mixture is advantageously carried out to a temperature ranging from about 650 ° C. to about 800 ° C. in an atmosphere which is continuously evacuated by a countercurrent flow of air, oxygen or oxygen enriched atmosphere, the mixture being at the highest temperature, Preferably at a margin of error of ± 10 ° C. for up to about 4 hours, preferably for up to about 2 hours. The heating step may be followed by cooling by quenching in air or by cooling at a natural urine cooling rate.

The heating step of the invention is carried out at a temperature of about 650 ° C. to about 800 ° C. for a time not exceeding about 4 hours. The temperature of the heating step is preferably about 700 ° C. to about 750 ° C. for about 1½ hour to about 2 hours.

After the heating step, it is advantageous to cool the reaction product to about 200 ° C. or less within a time of about 2 hours or less. The product is cooled to a temperature below about 100 ° C. and the cooling step is preferably performed within about 1 hour to 1½ hours. In the case where the cooling step is carried out for a time of 1½ hours or less, it is advantageous to anneal the product so that the product absorbs oxygen, causing distortion in the lattice. If the cooling step is carried out in a time of about 1½ hours or less, the cooling step is advantageously carried out in two or more zones with progressively lower temperatures. This cooling step is performed in three or more discontinuous regions, where the temperature of each region is preferably about 90 ° C. or more lower than the immediately preceding region. Most preferably, the temperatures in the three cooling zones are about 725 ° C, 625 ° C and 525 ° C.

Li _{1 + x} Mn _2-x O ₄ products are characterized analytically in a number of ways using, for example, standard chemical and spectroscopic methods of providing Li / Mn ratios and Mn oxidation numbers. This method is applied to the sample to identify the chemical formula used to represent the material.

One of the most useful analytical methods for the analysis of these materials is X-ray diffraction (XRD) analysis using powder techniques. XRD provides two types of useful information: (1) product purity and (2) lattice parameters. Since all materials have unique and clearly defined XRD patterns, comparing XRD patterns with standard patterns can determine whether a single phase product has been obtained. Many researchers have found a slight correlation between XRD patterns and cell performance. For example, the spinel should have a clear LiMn ₂ O ₄ XRD pattern without valid peaks derived from Li ₂ MnO ₃ , Mn ₂ O ₃ and Mn ₃ O ₄ . This material does not cycle and can not be used as a good cathode material by eluting from the cathode and causing harmful action.

4, 5, 8 and 9 show the clear LiMn ₂ O ₄ XRD patterns for samples B, C, I and K, respectively. LiMn ₂ O ₄ XRD peaks are identified by their well known 2θ positions, which are labeled with integers 501 to 508. These 2θ values, of course, can be converted to well known “d” values using standard methods with the recognition that x-rays are CuKα rays. XRD patterns appear to be nearly identical for the materials, but these four materials can be distinguished by the way they are synthesized as detailed in Tables 1A and 1B below.

6 and 7 show XRD patterns for spinel contaminated with by-products. The peak for the main product Li _{1 + x} Mn _2-x O ₄ is the same integer as the corresponding peak for the pure Li _{1 + x} Mn _2-x O ₄ in FIGS. 4, 5, 8 and 9. Labeled. Peaks labeled 700 integers belong to Li ₂ MnO ₃ , peaks labeled 600 integers belong to Mn ₂ O ₃ or Mn ₃ O ₄ (the latter two compounds are difficult to distinguish from very few peaks) Do).

The second type of XRD information, i. E. Grating parameter, is one that cannot be obtained by visual observation of the scanning state as shown in the figure. Instead, very precise investigations of the exact location of all peaks by special techniques of "lattice parameter refinement" well known to those skilled in the art of crystal analysis and XRD, and from this information the "a" axis in the crystal analysis The best cubic spinel unit cell dimension of the phase can be calculated, which is the lattice parameter a ₀ .

Many researchers have found that lattice parameters can be a very useful diagnostic tool because they directly correlate with the capacity fade rate in the form of a reduction in discharge capacity with cycle number. The lattice parameters change according to the stoichiometry of the cubic Li—Mn spinel (ie _{x in} Li _{1 + x} Mn ₂ O ₄ ) and the degree of oxidation of the spinel. The lattice parameter a ₀ of Li _1.00 Mn _2.00 O ₄ is 8.2476 μs (Standard X-ray Diffraction Powder Patterns, Part 21--Data for 92 Substances, MC Morris, HF McMuddy, EH Evans, B. Paretkins, HS Parker, W.Wong-Ng, DM Gladhill and CR Hubbard, National Bureau of Standards, US Monograph 25, 21 78 (1984)). The value of a ₀ decreases with the degree of Li removal (oxidation degree), and reaches a value of 8.03 kPa for the cubic Mn ₂ O ₄ (λ-MnO ₂ ) phase. When lithium is added to LiMn ₂ O ₄ (x> 0 in Li _{1 + x} Mn _2-x O ₄ ), manganese becomes more oxidized, a ₀ decreases to about 8.2 kPa [JM tarascon, WR McKinnon, F. Kuwar, TN Baumer, G. Amatucci and D. Giomar, J. Electrochem. Soc, 141 (1994) 1421-1431], the rate of dose fade in spinel decreases. The added oxidation reaction between lithium and manganese reduces the discharge capacity, and when the theoretical maximum value of Mn oxidation number is determined to be 4.00, the theoretical maximum 4V discharge capacity is Li _{1 + x} Mn _2-x in molecular units. (1-3x) lithium ion / electron per O ₄ .

A sample of lithium manganese oxide prepared according to the above technique is intimately mixed with a small amount of graphite (10-40 wt.%) And a binder (about 5 wt.%) To form a cathode mixed material, and the cathode mixed material is converted into a conductive support. After being pressed onto the phase, the cathode mixed material / support assembly (named positive) was dried by heating in a dry air stream to make a secondary battery positive electrode. This electrode was then tested in a conventional manner in a flat electrochemical test cell. One type of such battery is a removable battery as shown in FIG. 3. The cell was saturated with an electrolyte comprising a mixture of 1 mol lithium hexafluorophosphate (LiPF ₆ ) in a 50/50 (by weight) ethylene carbonate (EC) and dimethyl carbonate (DMC) solution under a dry argon atmosphere. Li _{1 + x} with conductive support 302 separated from lithium thin film cathode 303 with stainless steel conductive support 304 by porous glass fibers and / or polypropylene and / or polyethylene separators 305 and 306. It assembled using the Mn _2-x O ₄ containing positive electrode 301. The active battery components were pressed in intimate contact and insulated from the atmosphere. In the removable battery of FIG. 3, this insulation was achieved using two flat cylindrical half cells 307 and 308 constituting the cell body. Two polypropylene pieces (with an active cell between them) were bolted (not shown in FIG. 3) to squeeze the cell components together. A polypropylene-polyethylene “O” ring 309 around the circumference of the cell between two cylindrical half cells 307 and 308 seals the cell from the discharge of the electrolyte or the air inlet, while at the same time After tightened together, they have two functions to absorb the excess pressure of the bolts. The two half cells are configured with metal bolts sealed at their centers 310 and 311, whereby the bolts conduct current into and out of the active cell component. The O ring was also used to tightly seal around the current collecting bolts 312 and 313. Nuts 314 and washers 315 were used to secure the bolts against the "0" ring.

The test cells were then evaluated to investigate the behavior of the cell voltage during the charge-discharge cycle as a function of the amount of change of lithium ions per formula unit during the gradual reversible conversion step of Li _{1 + x} Mn _2-x O ₄ . When charging is initiated (ie, battery voltage about 3.1 volts to 3.5 volts), manganese begins to oxidize and lithium ions are released from Li _{1 + x} Mn _2-x O ₄ and migrate through the electrolyte into the lithium thin film. This process continues until the voltage reaches 4.3 volts, which is when most of the lithium atoms have moved to the lithium anode. The cell was then discharged to 3.0 volts and recharged several times at a rate of 0.5 mA per cm 2 of cathode area. Two such charge and discharge cycles are shown in FIG. 10.

The cell typically passed a charge of 120 mAh per 1 g of Li _{1 + x} Mn _2-x O ₄ , the active cathode material, in each ½ cycle. This amount of charge decreased with the number of cycles, as is typical for any battery system. This reduction phenomenon, termed cycle fade, is one of the most important cell performance characteristics of the cathode material along with the initial discharge capacity. The maximum discharge capacity for the cell was recorded. This is usually the discharge capacity for the first cycle, but for a few cells the capacity is maximized for the second or even third cycle. The fade ratio for the cell was calculated as the least square slope of the straight line through a graph showing the discharge capacity against the number of cycles after 30 and 50 cycles (see FIG. 11). This slope, expressed in mAh per 1 g of Li _{1 + x} Mn _2-x O ₄ , was converted to the fade rate as percent capacity loss per cycle by dividing the slope by the initial discharge capacity and multiplying by 100.

Spinels were synthesized by the proposed method as described above in a pilot scale rotary urinary tract (reactor). It was also synthesized by various other published or patented methods for the purpose of comparison with the method of the present invention. These methods include (a) a standard laboratory procedure, i.e., a method of cooling in an ambient atmosphere after heat reaction in an indirect heating furnace, and (b) a standard laboratory procedure using a slow cooling step (using a computerized controller on the urinary tract). And (c) air cooling followed by special annealing at 850 ° C. followed by cooling very slowly (10 ° C./h) to 500 ° C. or room temperature. In addition, for the purpose of evaluating various processing parameters, the synthesis was carried out using various modified means in a pilot reactor.

During the “very slow cooling / annealing” step in air or oxygen rich atmospheres, the spinels are known to absorb oxygen into the crystal lattice. This phenomenon can be observed by monitoring weight gain using thermal gravimetric analysis (TGA).

When using the slow cooling step according to the invention, the cooling rate can be greatly increased by continuously stirring the spinel material. This results in greater exposure of the solid / gas interface. In addition, continuous scavenging with air or oxygen makes it possible to continuously use oxygen at the surface interface for rapid absorption into the lattice.

As an example, the capacity and cycle fade ratios for the spinels produced by the method of the present invention are approximately equivalent to those for the spinels produced by conventionally discussed optimal experimental methods, including heating and cooling times over several hours or days. It was. On the other hand, when the experimental method was performed for only a few hours comparable to the time used in the method of the present invention, the capacity was considerably lowered and, in some cases, the fade rate was higher.

Several samples were evaluated to increase the reliability of the correlation. Each sample was a cell that was cycled in repeated testing to provide uncertainty in each test. Therefore, the difference in dose and fade ratio is the subject of statistical investigation. Tables 1a and 1b below describe the method of making the material, and Table 2 below shows all the average results and standard deviations (σ).

The most important features of the method of the present invention are explained mostly through pairwise comparison tests (unlike linear multiple regression of all data). When performing a pair comparison, it is desirable to keep all parameters except the parameter under investigation constant. For example, when comparing different heating / cooling treatments, it is preferable to compare the same materials. In addition, when comparing different materials for a given compositional difference (eg lithium / manganese ratio), the materials are preferably made from the same precursor EMD and lithium compound. In this way as possible. Spinels of all examples were synthesized from cell grade EMD commercially available from Ker McGee Chemical Corporation with reproducible properties. Where possible, materials synthesized from the same lithium compound, ie lithium hydroxide or lithium carbonate, were compared.

Example 1

Examples of the present invention were equally better than or better than any other material tested. The method used in the embodiments of the present invention consists of reacting in a rotary urinary tract for about 2 hours under air, cooling in the rotary urinary tract for about 2 hours under air, prepared using excess lithium. That is, the Li / Mn ₂ ratio is 1.05 (not Li / Mn ₂ = 1.00). Examples of the invention are samples B and C. These two samples were treated identically, except that the cooling step was carried out in a small laboratory rotary urinary furnace (after heating the sample to 725 ° C. in the first step) under O ₂ atmosphere for sample B, and for sample C Cooled zone by zone in a pilot tract under air. Table 2 below shows that Samples B and C represent the lowest discharge rate (or equivalent) of suitable discharge capacity and the results presented in the table. In addition, their XRD patterns (FIGS. 4 and 5) were clear and the grating parameters were the minimum of the grating parameters presented.

Specifically, an embodiment of the present invention provides a reaction time of 20 hours and (a) a cooling time of 12 hours (60 ° C./h), so-called “slow cooling of the laboratory”, or (b) annealing the sample above 800 ° C. And very slowly, i.e., equivalent to the material produced by the long-term experimental route (stop layer urinary) comprising any one of the so-called [annealing / very slow cooling] comprising cooling to 10 ° C./h. It was excellent. Specific comparison targets from Table 2 are as follows.

a. Comparing B and C (invention) and H ("same" pilot material but performing [annealing / very slow cooling] after synthesis), the material of the invention is [annealing / very slow cooling] Although not as good as comparable, it is shown to be comparably superior.

b. Sample I is a sample prepared in the same laboratory with the same precursor and composition as the sample of the present invention, but Sample I was subjected to annealing / very slow cooling. Material I is not superior to B and C on average. The 50 cycle fade rate of I was statistically equivalent to the fade rate of the other two materials, since the standard deviation for the “I” value is quite large. Sample I showed a clear XRD pattern (FIG. 8).

c. Sample K is identical to B and C in terms of precursor and composition, but K is “slowly cooled in laboratory” after 20 hours of reaction in a laboratory stationary tract instead of annealing / very slow cooling. Sample K is one of the best materials, but not as good as B and C in terms of capacity and fade ratio. The XRD pattern for K was clear (FIG. 9).

d. Sample L 'is a pilot material that was subjected to [annealing / very slow cooling] after preparation from Li ₂ CO ₃ . (Sample H subjected to [annealing / very slow cooling] treatment in comparison to that prepared from LiOH). Embodiments of the invention are not made from Li ₂ CO ₃ . However, sample O is a sample prepared in the same laboratory as L ', except that it was slowly cooled in the laboratory. Sample O was comparable to Sample L ', suggesting that the [Annealing / Very Slow Cooling] treatment did not provide any improved results compared to the procedure of slowly cooling in the laboratory after reaction at 725 ° C.

e. Sample G was synthesized the same as L ', but was reheated and slowly cooled in the laboratory. Sample G exhibits a slower fade rate than L ', which, as in the case of (d), may be inferior to the procedure for annealing / very slow cooling, even slowly cooling in the laboratory after reacting at 725 ° C. Imply that there is.

g. Comparing samples I and I 'and J and J', it can be seen that the annealing / very slow cooling treatment of the material produced at 725 ° C. improves performance.

Example 2

As shown in Table 2 below, the present invention permits a reaction time of only about 2 hours by (1) a rotary urinary tract with airflow, whereas the reaction for 2 hours in a stationary tract provides a completely unsuitable product. (2) It is shown that a reaction time of about 20 hours is required in a laboratory stationary tract to produce the same effect as a reaction of about 2 hours in a rotary urinary tract using air. In this example, air cooling was used because, in fact, slow cooling in the laboratory would prolong the reaction time beyond 2 hours, causing the test to fail.

a. Sample M was air cooled after synthesis in a stationary urinary tract for only 2 hours. This method contaminated the material with harmful byproducts, which is confirmed by XRD scan. The initial capacity was only 76.8 mAh / g, and two of the four assembled batteries could not even cycle 10 times. This problem is due to the high concentration of impurities. This material is completely unsuitable as a battery cathode. A comparable starting material prepared from LiOH with a Li / Mn ₂ ratio of 1.00 and reacted for 20 hours in a stationary urinary tract is sample N. This material showed good XRD patterns and initial capacity, but the fade ratio was normal in view of good material criteria. No material was produced using the Li / Mn ₂ ratio 1.00 in the rotary urinary tract.

b. Samples A and F are two materials starting from an equivalent premix and then air cooled after reacting with airflow in a rotary urinary tract. Samples I ′, identical to A and F in terms of precursors and composition, were prepared using a reaction time of 20 hours in a laboratory stationary urinary tract. Rotating samples (A and F) were superior in terms of discharge capacity and fade ratio with material (I ') reacted for 20 hours in a laboratory urinary tract. This Example and 2.a show that the reaction time of 2 hours in the rotary urinary tract has an effect approximately equal to the reaction time of 20 hours in the stationary urinary tract, and is substantially more effective than the reaction time of 2 hours in the stationary urinary tract. Shows.

c. Three samples, prepared from equivalent (Li ₂ CO ₃ ) precursors and having equivalent composition, were prepared by using E (air cooled after using only 2 hours of reaction time in a laboratory stop tract), L (2 in a rotary pilot tract) Prepared by air cooling after using a reaction time of time) and J '(prepared by air cooling after using a reaction time of 20 hours in a lab stop urine). The treatment process of E produced contaminated material (compare the XRD pattern in FIG. 7) and showed a somewhat lower discharge capacity (112 mAh / g). J 'also showed a somewhat lower capacity, but its XRD pattern was clear. The pilot sample L showed the best capacity among the three materials, and the XRD pattern was also clear. The fade rate was moderate or inferior in all cases, but the lab sample treated surprisingly 2 hours showed the best fade rate.

Example 3

Oxygen-containing gases in rotation are essential to the process of the invention. This is shown as Sample D in Table 2, which was prepared using N ₂ flowing through the urinary tract during the reaction and cooling. The capacity was unacceptably low (101 mAh / g), which corresponds to a contaminated XRD scan (FIG. 6). Fade rates were also moderate or inferior. The equivalent sample with air in the rotary urinary tract was L, which showed an acceptable capacity and a clear XRD scan and proved to be superior to D. The fade rate of L was comparable to that of D, but the fade rate of L proceeds at higher doses. Comparisons of Examples B and C of the present invention have shown that oxygen or air atmospheres provide satisfactory results.

Example 4

The method of the invention is advantageous in that it gradually cools zone by zone in a rotary urinary tract (2 hours). As shown in Table 2 below, this is demonstrated by comparing Samples B and C, cooled in the same manner as above, and Sample A, which is made from the same premix and is the same reaction product but air cooled. Samples B and C show better capacity and especially fade ratio than A.

Example 5

Table 2 below shows that it is advantageous for the Li / Mn ₂ ratio to be greater than 1.00 in the method of the present invention. No sample was prepared and cooled in a rotary urinary tract for a Li / Mn ₂ ratio of 1.00. That is, there is no direct comparison with Samples B and C. This is because the use of excess lithium using materials synthesized in the laboratory has confirmed that there is a clear advantage. Thus, there was no need to produce inferior materials in the pilot plant. In (1.c) above, the synthesis method of the present invention was found to be as good as the best laboratory synthesis method, i.e., a method of reacting for 20 hours and cooling slowly in the laboratory (ie, cooling at 60 ° C / hour or 12 hours). have. Therefore, if the Preferred experimental material Li / Mn ₂ = and has proven to be superior to the same material, except that 1.00, Li / Mn _2> 1.00 and the pilot (the method of the present invention), the material is Li / Mn ₂ = 1.00 It can be deduced that phosphorus will be superior to the material of the process of the invention.

Samples K equivalent to materials B and C by the method of the present invention were compared to K 'which is equivalent except that the Li / Mn ₂ ratio is 1.00. Sample K 'showed a larger capacity than K, which is the theoretically expected result. However, the capacity of K was still large enough to be suitable. For the fade ratio, which is an essential feature, sample K was better than sample K '. Inferring, the material of the process of the invention must be better than the rotary material with Li / Mn ₂ = 1.00.

Example 6

This example demonstrates the advantage of the Li / Mn ₂ molar ratio of 1.033 (x = 0.022 for Li _{1 + x} Mn _2-x O ₄ ) compared to the Li / Mn ₂ molar ratio of 1.00. Sample P was prepared in the laboratory using the Li / Mn ₂ molar ratio 1.033 (x = 0.022 for Li _{1 + x} Mn _2-x O ₄ ) by the method of the present invention. The average discharge capacity of this material was 124.7 mAh / g, and the fade rates for 30 and 50 cycles were 0.33% and 0.26% per cycle, respectively (see Table 2). For comparison, sample K 'was prepared in the laboratory by the same method, except that the Li / Mn ₂ molar ratio was 1.00. Sample K 'had an average discharge capacity of 132 mAh / g and the 30 cycle and 50 cycle fade rates were 0.61% and 0.47% per cycle, respectively. The discharge capacity for sample K 'was slightly larger than for P as expected, but the discharge capacity for sample P was suitable. It is important that the fade ratio for sample P is significantly greater than the fade ratio for sample K '.

Example 7

This example shows a case where the Li / Mn ₂ molar ratio is 1.112 (x = 0.072 for Li _{1 + x} Mn _2-x O ₄ ) compared to the case where the Li / Mn ₂ molar ratio is 1.00 (at x = 0). The advantages are explained. Sample Q was prepared in a rotary urinary tract by the method of the present invention and then slowly cooled in the urinary tract. Table 2 below shows that the average discharge capacity of this sample is 126.3 mAh / g and the 30 cycle and 50 cycle fade ratios are 0.084% and 0.086% per cycle, respectively. Sample K 'acts as a control material with a Li / Mn ₂ molar ratio of 1.00. The discharge capacity for sample K 'is 132 mAh / g and the 30 cycle and 50 cycle fade rates are 0.61% and 0.47% per cycle. The discharge capacity for sample K 'is slightly larger than Q as expected, but the discharge capacity for sample Q is suitable. On the other hand, the fade ratio for sample Q is significantly larger than that for sample K '. Sample K 'was synthesized using LiOH as a lithium source in a laboratory stationary urinary tract (not a rotary urinary tract as in the case of Sample Q). The difference in this synthesis does not cause a difference in the fade ratio of sample K 'and Q, as evidenced by other examples. Comparing the results of samples I 'and J' and K and O (Table 2), it can be seen that the product synthesized from LiOH yields a better or equivalent fade ratio than the product synthesized from Li ₂ CO ₃ . have. Comparing Sample K with Samples B and C, a 20-hour reaction time in the lab yields a fade ratio equivalent to the product synthesized in the urinary tract and slowly cooled (inventive method) synthesized in the labyrinth. It can be seen. Therefore, a lower fade ratio of sample Q than that of sample K 'is caused because of a larger Li / Mn ₂ ratio for sample Q.

Claims (52)

Continuously carrying a single-phase lithium-containing manganese oxide interlayer compound represented by the formula Li _{1 + x} Mn _2-x O ₄ , wherein x is 0.022 to 0.125, having a spinel-like structure and having an a-axis lattice parameter of 8.235 μs or less As a manufacturing method, (a) sufficiently mixing a predetermined amount of lithium hydroxide or lithium salt with manganese oxide or manganese salt based on the lithium-containing manganese oxide compound, (b) feeding a sufficiently mixed compound to the reactor, (c) continuously stirring the mixed compounds in a reactor, (d) heating the stirred and mixed compound in a reactor at a temperature of 650 ° C. to 850 ° C. under air or oxygen enriched atmosphere for a time not exceeding 4 hours, (e) forming an interlayer compound of Li _{1 + x} Mn _2-x O ₄ spinel structure, and (f) cooling the reaction product to 200 ° C. or less within 2 hours or less Method comprising a. The method of claim 1, further comprising cooling the reaction product to a temperature of 100 ° C. or less. The method of claim 1 wherein the atmosphere is continuously evacuated to airflow, oxygen or oxygen rich air while performing step (c). The method of claim 1 wherein the atmosphere is continuously evacuated to a backflow of an oxygen rich atmosphere while performing step (c). The method of claim 1, wherein the reactor is a horizontal rotating tube having a discharge end lower than the injection end. The method of claim 1 wherein the reactor is a horizontal calcination apparatus with a rotating screw. The method of claim 1 wherein the reactor is a fluidized bed. The method of claim 1 wherein the reactor is a heated vibratory conveyor belt. The method of claim 1, wherein the reactor is a cascade of vertical rotary furnaces. The method of claim 1 wherein said heating step is performed within a time of 2 hours or less. The method of claim 1 wherein said heating step is performed within a time of 1.5 hours or less. The method of claim 11, wherein the temperature of the heating step is between 700 ° C. and 800 ° C. 13. The method of claim 1 comprising continuously evacuating with airflow, oxygen or an oxygen rich atmosphere while performing step (d). The method of claim 13, wherein said continuous scavenging step is performed using backflow of air. The method of claim 1 wherein said cooling step is performed in a horizontal calcination device with a rotating screw. The method of claim 1 wherein said cooling step is performed in a fluidized bed. The method of claim 1 wherein said cooling step is performed in a rotary urinary tract. The method of claim 1 wherein said cooling step is performed in a heated vibratory conveyor belt. The method of claim 1 wherein said cooling step is carried out in a cascade of vertical rotary tracts. The method of claim 1 wherein said cooling step is performed within a time of 1.5 hours or less. The method of claim 1 wherein said cooling step is performed in less than 1.5 hours. The method of claim 1, wherein said cooling step is performed in areas where said temperature is gradually decreasing. The method of claim 22 wherein said cooling step is performed in two or three zones. The method of claim 23, wherein the temperature of each gradual cooling zone is at least about 90 ° C. lower than the immediately preceding zone. The method of claim 24, wherein the temperatures in the gradual cooling zones are 725 ° C., 625 ° C., and 525 ° C. 25. The method of claim 24, wherein the temperatures in the gradual cooling zone are 800 ° C., 650 ° C. and 500 ° C. 25. The method of claim 24, wherein the temperatures in the gradual cooling zones are 750 ° C., 600 ° C. and 450 ° C. 25. The method of claim 24, wherein the cooled product is annealed in a state capable of absorbing oxygen to prepare a lithium-containing interlayer compound having a spinel structure represented by Formula Li _{1 + x} Mn _2-x O ₄ . . The method of claim 28, wherein said cooled product is annealed in a state capable of absorbing as much oxygen as possible. The method of claim 1 wherein the atmosphere is continuously evacuated by an airflow, oxygen or oxygen rich atmosphere during the cooling step. 31. The method of claim 30, wherein said atmosphere is continuously evacuated by backflow of air during said cooling step. The method according to claim 1, wherein the manganese oxide or manganese salt is composed of MnO ₂ (prepared by electrolysis or chemically), MnCO ₃ , Mn ₂ O ₃ , Mn ₃ O ₄ , MnO, manganese acetate and mixtures thereof The method selected from the group. The method of claim 1, wherein the manganese oxide is heat treated prior to step (a). The method of claim 1, wherein the lithium hydroxide or lithium salt is selected from the group consisting of LiOH, Li ₂ CO ₃ , LiNO ₃ and mixtures and hydrates thereof. The method of claim 33, wherein the electrochemically prepared MnO ₂ is neutralized with LiOH or NH ₄ OH during its synthesis. A method of synthesizing a lithium manganese oxide having a spinel crystal structure having an a-axis lattice parameter of 8.235 Pa or less, At least one lithium hydroxide or lithium salt selected from the group consisting of LiOH, Li ₂ CO ₃ , LiNO ₃ and mixtures thereof and MnO ₂ , Mn ₂ O ₃ , MnCO ₃ , Mn ₃ O ₄ , MnO, manganese acetate and Forming a sufficiently mixed mixture in the form of a finely divided solid consisting of at least one manganese oxide or manganese salt selected from the group consisting of mixtures, and Reacting the salts with each other by stirring and heating the mixture in a reactor under continuous scavenging by backflow of air for a time not exceeding 4 hours at a temperature in the range of 650 ° C. to 800 ° C. Method comprising a. The method of claim 36, wherein heating the mixture is carried out in air at a temperature in the range of 700 ° C. to 800 ° C. and maintaining the mixture at the highest temperature for a period of up to 2 hours. 37. The method of claim 36, wherein the manganese oxide is heat treated prior to forming the mixture. The method of claim 36, wherein the lithium manganese oxide is represented by the formula Li _{1 + x} Mn _2-x O ₄ , wherein x is 0.022 to 0.2. The method of claim 39, wherein the Li _{1 + x} Mn _2-x O ₄ has a (Mn ₂ ) O ₂ (Mn ₂ ) O ₄ ^n-1 framework structure, wherein the amount of Mn cation in the structure is from stoichiometric values How to change. A single phase lithium-containing manganese oxide interlayer compound represented by the formula Li _{1 + x} Mn _2-x O ₄ (wherein x is 0.022 to 0.200) having a spinel-type structure and having an a-axis lattice parameter of 8.235 μs or less, is subjected to a batch process. As a method of manufacturing by (a) sufficiently mixing the lithium hydroxide or lithium salt with manganese oxide or manganese salt in a stoichiometric amount based on the lithium-containing manganese oxide compound, (b) feeding a sufficiently mixed compound to the reactor, (c) stirring the mixed compounds in a reactor, (d) heating the stirred and mixed compound in a reactor at a temperature of 650 ° C. to 800 ° C. under air or oxygen enriched atmosphere for a time not exceeding 4 hours, (e) forming a Li _{1 + x} Mn _2-x O ₄ spinel interlayer compound, and (f) cooling the reaction product to 100 ° C. or less within 2 hours or less. Method comprising a. 42. The method of claim 41 wherein the cooling vessel is a horizontal calcination device having a rotating screw. 42. The method of claim 41 wherein the cooling vessel is a fluidized bed. 42. The method of claim 41 wherein the cooling vessel is a heated vibratory conveyor belt. 42. The method of claim 41 wherein the cooling vessel is a cascade of vertical rotary urinary tracts. 42. The method of claim 41 wherein the cooling vessel is a rotary urinary tract. 42. The method of claim 41 wherein said cooling step is performed within a time of 1.5 hours or less. 42. The method of claim 41 wherein the cooling step is performed within a time of 1 hour or less. The method according to claim 1, wherein said heating step (d) is carried out in regions in which the temperature gradually increases. The method of claim 49, wherein said heating step is performed in four zones. 51. The method of claim 50, wherein the temperature of each gradual heating zone is at least 25 ° C higher than the immediately preceding zone. The method of claim 49, wherein the stirred and mixed compound is maintained in each zone for at least 30 minutes.