U.S. patent application number 12/517990 was filed with the patent office on 2010-12-23 for a method for preparing a particulate cathode material, and the material obtained by said method.
Invention is credited to Guoxian LIANG.
Application Number | 20100323245 12/517990 |
Document ID | / |
Family ID | 39491630 |
Filed Date | 2010-12-23 |
United States Patent
Application |
20100323245 |
Kind Code |
A1 |
LIANG; Guoxian |
December 23, 2010 |
A METHOD FOR PREPARING A PARTICULATE CATHODE MATERIAL, AND THE
MATERIAL OBTAINED BY SAID METHOD
Abstract
Disclosed are a method for preparing a complex oxide particle
composition, the so-prepared particle composition and its use as
electrode material. This composition comprises complex oxide
particles having a non powdery conductive carbon deposit on at
least part of their surface. Its method of preparation comprises
nanogrinding complex oxide particles or particles of complex oxide
precursors, wherein an organic carbon precursor is added to the
oxide particles or oxide precursor particles before, during or
after nanogrinding, and pyrolysing the mixture thus obtained; a
stabilizing agent is optionally added to the oxide particles or
oxide precursor particles before, during or after nanogrinding; and
the nanogrinding step is performed in a bead mill on particles
dispersed in a carrier solvent.
Inventors: |
LIANG; Guoxian; (Quebec,
CA) |
Correspondence
Address: |
SOFER & HAROUN LLP.
317 MADISON AVENUE, SUITE 910
NEW YORK
NY
10017
US
|
Family ID: |
39491630 |
Appl. No.: |
12/517990 |
Filed: |
December 7, 2007 |
PCT Filed: |
December 7, 2007 |
PCT NO: |
PCT/CA07/02236 |
371 Date: |
May 12, 2010 |
Current U.S.
Class: |
429/231.5 ;
252/506; 427/77; 429/231.8 |
Current CPC
Class: |
C01P 2004/04 20130101;
B82Y 30/00 20130101; C01P 2004/52 20130101; H01M 4/136 20130101;
H01M 4/625 20130101; C09C 1/00 20130101; H01M 4/5825 20130101; Y02E
60/10 20130101; C01P 2004/61 20130101; C01P 2002/72 20130101; C01P
2004/62 20130101; C01P 2004/64 20130101; C01P 2004/45 20130101;
C09C 3/08 20130101; C01B 25/45 20130101; H01M 2004/021 20130101;
C01P 2004/03 20130101; C01P 2004/51 20130101; C01P 2004/53
20130101; C09C 1/22 20130101; C01P 2004/50 20130101; C01P 2002/32
20130101 |
Class at
Publication: |
429/231.5 ;
252/506; 427/77; 429/231.8 |
International
Class: |
H01M 4/583 20100101
H01M004/583; H01B 1/08 20060101 H01B001/08; B05D 5/12 20060101
B05D005/12 |
Foreign Application Data
Date |
Code |
Application Number |
Dec 7, 2006 |
CA |
2569991 |
Jul 12, 2007 |
CA |
PCT/CA2007/002236 |
Claims
1. A method for preparing a complex oxide particle composition,
wherein the complex oxide particles have a non powdery conductive
carbon deposit on at least part of their surface, said method
comprises nanogrinding complex oxide particles or particles of
complex oxide precursors, wherein an organic carbon precursor is
added to the oxide particles or oxide precursor particles before,
during or after nanogrinding, and pyrolysing the mixture thus
obtained a stabilizing agent is optionally added to the oxide
particles or oxide precursor particles before, during or after
nanogrinding the nanogrinding step is performed in a bead mill on
particles dispersed in a carrier solvent, and the size of the
particles to nanogrind, the size of the beads used to nanogrind,
and the size of the resulting particles are selected such that:
0.004.ltoreq.MS(SP)/MS(B).ltoreq.0.12
0.0025.ltoreq.MS(FP)/MS(SP).ltoreq.0.25 wherein MS(SP) represents
the mean size diameter of the particles before nanogrinding
(starting particles), MS(FP) represents the mean size diameter of
the particles after nanogrinding (final particles), and MS(B) is
the mean size diameter of the nanogrinding beads.
2. The method of claim 1 characterized in that the carrier solvent
is a reactive solvent.
3. The method of claim 2 characterized in that the carrier solvent
is water or isopropanol.
4. The method of claim 1, which further comprises a step which is
performed after grinding and before pyrolysis, said further step
comprising conditioning the reaction mixture in order to adsorb the
carbon precursor on the complex oxide precursors or on the complex
oxide, or to polymerize or cross link a carbon precursor which is a
monomer.
5. The method of claim 1, which further comprises a step consisting
in aggregating the reaction mixture comprising the carbon precursor
and the complex oxide precursor after grinding.
6. The method of claim 5, wherein aggregation is performed by
flocculating, by spray drying, or by charge effect.
7. The method of claim 1, wherein an organic carbon precursor
selected from fatty acid salts of a transition metal cation is
added to the oxide particles or the oxide precursor particles.
8. The method of claim 7 wherein the transition metal is Ni, Co or
Fe.
9. The method of claim 7 wherein the fatty acid contains at least 6
carbon atoms.
10. The method of claim 9 wherein fatty acid is selected from
stearate, oleate, linoleate, linolenate, ricinolenate.
11. The method of claim 1, wherein the mean size diameter of the
grinding beads is from 100 to 500 .mu.m.
12. The method of claim 1, wherein the organic carbon precursor is
added to oxide precursor particles.
13. The method of claim 12, wherein an organic stabilizing agent is
added before grinding.
14. The method of claim 13, wherein the organic stabilizing agent
is selected from organic electrostatic or electrosteric
stabilizers, surfactants, dispersant agents, self adsorbing agents
and encapsulant agents.
15. The method of claim 13, wherein the organic stabilizing agent
is a conductive carbon precursor.
16. The method of claim 12, wherein pyrolysis is performed before,
or during the synthesis of the complex oxide starting from the
precursors thereof.
17. The method of claim 12, wherein the organic carbon precursor
also acts as the stabilizing agent.
18. The method of claim 1, wherein the organic carbon precursor is
added to complex oxide particles.
19. The method of claim 18, wherein the complex oxide is prepared
by a solid state reaction of precursors under reducing or inert
atmosphere.
20. The method of claim 18, wherein the complex oxide is prepared
by co-precipitation or sol-gel synthesis.
21. The method of claim 18, wherein the complex oxide is prepared
by hydrothermal reaction.
22. The method of claim 18, wherein the particle size of the
particles before grinding is in the range from 1 .mu.m to 50
.mu.m.
23. The method of claim 18, wherein the complex oxide is prepared
by reacting the precursors in molten state in an inert or reducing
atmosphere, the complex oxide being pre-ground after synthesis and
solidification.
24. A particle composition comprising particles having a complex
oxide core and a conductive carbon deposit on at least part of the
surface of the core, wherein: the particles comprise elementary
nanoparticles having a nanoscale size and agglomerates or
aggregates of elementary nanoparticles having a submicron to micron
scale particle size; said conductive carbon deposit is a non
powdery deposit, and is present on at least part of the surface of
the elementary particles and on the surface of the aggregates
25. A particle composition of claim 24, wherein the complex oxide
is at least one compound of formula AMXO.sub.4 having an olivine
structure wherein: A is Li, optionally partly replaced with not
more than 10 atomic % Na or K; M represents Fe.sup.II, or Mn.sup.II
optionally partly replaced with not more than 50 atomic % of at
least one metal selected in the group consisting of Mn, Fe Ni et
Co, and optionally replaced with not more than 10 atomic % of at
least one aliovalent or isovalent metal different from Mn, Ni or
Co; XO.sub.4 represents PO.sub.4, optionally partly replace with
not more than 10 mol % of at least one group selected from SO.sub.4
and SiO.sub.4.
26. A particle composition of claim 25, wherein the aliovalent or
isovalent metal different from Fe, Mn, Ni or Co in the complex
oxide is at least one metal selected from the group consisting Mg,
Mo, Nb, Ti, Al, Ta, Ge, La, Y, Yb, Sm, Ce, Hf, Cr, Zr, Bi, Zn, Ca
et W.
27. A particle composition of claim 26, wherein the complex oxide
is LiFe.sub.1-xMn.sub.xPO.sub.4, 0.ltoreq.x.ltoreq.0.5,
LiFePO.sub.4 or LiMnPO.sub.4.
28. A particle composition of claim 24, wherein the complex oxide
is a titanate which has a spinel structure and the formula
A.sub.aM.sub.mO.sub.oN.sub.nF.sub.f wherein A represents an alkali
metal; M represents Ti alone, or Ti partly replaced with another
metal, preferably a transition metal; a>0, m.gtoreq.0, o>0,
n.gtoreq.0, f.gtoreq.0, and coefficients a, m, o, n and f are
selected to provide electroneutrality of the complex oxide.
29. A particle composition of claim 28, wherein A is Li, optionally
partly replaced with another alcali metal.
30. A particle composition of claim 29, wherein the titanate has
one of the following formulae Li.sub.4+xTi.sub.5O.sub.12 or
Li.sub.4+x-2yMg.sub.yTi.sub.5O.sub.12, wherein 0.ltoreq.x et
y.ltoreq.1, or LiTi.sub.5O.sub.12.
31. A particle composition of claim 24, further comprising powdery
carbon particles.
32. A particle composition of claim 24, wherein the particles
comprise elementary nanoparticles and micron size agglomerates or
aggregates of elementary nanoparticles, wherein said elementary
nanoparticles have dimensions ranging from 5 nm to 1.0 .mu.m and
comprise primary nanoparticles and secondary particles, said
primary particles are made of a complex oxide with or without C,
said secondary particle is an agglomerate or an aggregate of
primary particles, an aggregate of primary nanoparticles is a
micron-size assembly of primary nanosize particles held together by
physical or chemical interaction, by carbon bridges, or bridges of
locally sintered complex oxide containing of internal open porosity
and carbon deposit lower than 30%, an agglomerate is an assembly of
particles loosely held together by low forces.
33. A particle composition of claim 32, which further contains at
least one element selected from internal or external C-deposit or
carbon bridging or particulate carbon, inert or conductive phases
or sintering necks.
34. A particle composition of claim 32, which further has a
porosity.
35. A particle composition of claim 24, wherein the carbon deposit
is in the form of carbon nanotubes.
36. A nanocomposite electrode material comprising a particle
composition of claim 24 as the active electrode material.
37. A nanocomposite electrode material of claim 36, wherein at
least 50% of the elementary nanoparticles of the particle
composition have a size between 5 nm and 900 nm diameter, said
nanoparticles being not aggregated or sintered.
38. A nanocomposite electrode material of claim 36, which comprises
a particle composition wherein the elementary nanoparticles are
aggregated to form agglomerates having a size from 0.2 .mu.m and 10
.mu.m.
39. A nanocomposite electrode material of claim 36, wherein the
conductive carbon deposit attached to the complex oxide crystal
structure on at least part of the surface of the nanoparticle has a
nanoscale thickness.
40. A nanocomposite electrode material of claim 36, wherein the
conductive carbon is present on part of the surface of the complex
oxide nanoparticles, and the nanoparticles are sintered at the
complex oxide surface thereof.
41. A nanocomposite electrode material of claim 36, where the major
part of the surface of the complex oxide nanoparticles is covered
with the conductive carbon deposit, and the nanoparticles are
aggregated via carbon bridges.
42. A nanocomposite electrode material of claim 36, which further
contains at least a binder or an electron conduction additive.
43. A nanocomposite electrode material of claim 36, wherein the
particle composition comprises secondary particles and or
aggregated of elementary nanoparticles and has an open porosity,
the volumetric fraction of the pores ranging from 0.30 to 0.05.
44. A nanocomposite electrode material of claim 36, wherein the
conductive carbon deposit is at least partly graphitized carbon
obtained by pyrolysis of an organic carbon precursor that contains
elements such as N, P, Si that can be covalently bound to
carbon.
45. A cathode comprising a nanocomposite electrode material of
claim 36 on a current collector, wherein the complex oxide is a
LiMPO.sub.4 oxide.
46. An anode comprising a nanocomposite electrode material of claim
36 on a current collector, wherein the complex oxide is a
titanate.
47. An electrochemical cell comprising a electrolyte, an anode and
a cathode, wherein the cathode is a cathode of claim 45.
48. An electrochemical cell comprising a electrolyte, an anode and
a cathode, wherein the anode is an anode of claim 46.
Description
FIELD OF THE INVENTION
[0001] The invention relates generally to a method for preparing
particle compositions, as well as the particles compositions
obtained by said method, and uses thereof as electrode
material.
BACKGROUND OF THE INVENTION
[0002] Lithium-ion batteries have known a phenomenal technical
success and commercial growth since the initial work by Sony in the
early 90's based on lithium insertion electrodes: essentially the
high voltage cobalt oxide cathode invented by J. B. Goodenough and
the carbon anode using coke or graphitized carbonaceous
materials.
[0003] Since then, lithium-ion batteries have progressively
replaced existing Ni--Cd and Ni-MH batteries, because of their
superior performances in most portable electronic applications.
However, because of their cost and intrinsic instability under
abusive conditions, especially in their fully charged state, only
small cell size and format have been commercialized with
success.
[0004] In the mid 90's, Goodenough (See U.S. Pat. No. 5,910,382 and
U.S. Pat. No. 6,391,493) suggested that polyanionic phosphate
structures, namely nasicons and olivines, could raise the redox
potential of low cost and environmentally compatible transition
metals such as Fe, until then associated to a low voltage of
insertion. For example LiFePO.sub.4 was shown to reversibly
insert-deinsert lithium ion at a voltage of 3.45 V vs a lithium
anode corresponding to a two phase reaction. Furthermore,
covalently bounded oxygen atom in the phosphate polyanion
eliminates the cathode instability observed in fully charged
layered oxides, making an inherently safe lithium-ion battery.
[0005] As pointed out by Goodenough (U.S. Pat. No. 5,910,382 &
U.S. Pat. No. 6,514,640), one drawback associated with the
covalently bonded polyanions in LiFePO.sub.4 cathode materials is
the low electronic conductivity and limited Li.sup.+ diffusivity in
the material. Reducing LiFePO.sub.4 particles to the nanoscale
level was pointed out as one solution to these problems as was the
partial supplementation of the iron metal or phosphate polyanions
by other metal or anions.
[0006] One significant improvement to the problem of low electronic
conductivity of complex metal oxide cathode powder and more
specifically of metal phosphate was achieved with the use of an
organic carbon precursor that is pyrolysed onto the cathode
material or its precursor to improve electrical field at the level
of the cathode particles [Ravet (U.S. Pat. No. 6,963,666, U.S. Pat.
No. 6,855,273, WO/0227824 and WO/0227823)].
[0007] It is also known to improve conductivity of a phosphate
powder when used as a cathode material, by intimately mixing
conductive carbon black or graphite powder with the phosphate
powder or the phosphate precursors before synthesis Such addition
of carbon blakc or graphite powder involves usually relatively
large quantities of C to achieve good connectivity and does not
result in a good attachment of the C to the metal phosphate crystal
structure, said attachment being a characteristic judged essential
to maintain contact despite volume variations during long term
cycling.
[0008] Such recent improvements have led several battery
manufacturers and users to undertake the development of safe mid to
large size lithium-ion batteries based on transition metal
phosphates cathode materials for use in portable power tools,
Hybrid Electric Vehicle (HEV) and Plug-in HEV as well as for large
stationary batteries for backup power and energy storage from
intermittent sources.
[0009] Problems remain however to optimize the processability, cost
and performance especially when power, energy and cyclability are
required simultaneously.
[0010] Composite electrode optimization, for example, requires
short distances for Li.sup.+ diffusion in the solid state and the
presence of an electronically conductive phase at the level of each
nanoparticle of LiFePO.sub.4. Manipulation and processing (coating
and compacting) of elementary nanoparticles or their dispersion is
more complex than manipulation and processing micron-size
particles, given their large surfaces and low compaction. In the
present text, nanoparticle means a particle having dimensions
ranging from 5 nm to submicrons (defined as less than 1.0 .mu.m),
preferably between 20 and 600 nm that can be primary or secondary
particles. A primary particle is made of a complex oxide. A
secondary particle is an aggregate of primary particles, and may
also contain other constituants such as internal or external
C-deposit or carbon bridging or particulate carbon, other inert or
conductive phases or sintering necks. A secondary particle may also
have a porosity.
[0011] The present inventors found that the use of agglomerates of
primary and secondary nanoparticles which are elaborated at a
micron-size scale or larger (by spray drying for example), instead
of elementary nanoparticles as such, facilitates ions and electron
diffusion and the electrochemical reaction This is the result of
using nano dimensions at the level of the active material nano
particles while, benefiting from the ease of manipulating
micron-size agglomerates.
[0012] As a general rule, electrochemical performance optimization
of such agglomerates of nanoparticles or nanocomposite material
requires a material having a high proportion of active metal
phosphate, a low proportion of electrochemically inert conductive
carbon and a controlled degree of open porosity of the agglomerates
or the nanocomposite material. Furthermore, pore channel dimensions
must be designed to allow solvated lithium's ion of the electrolyte
to penetrate and reach elementary nano sized particles to support
high charge or discharge rate currents.
[0013] Designing such agglomerates of nanoparticles or
nanocomposite materials, as well as attaching efficiently
nanolayers of conducting carbon to single or agglomerated nano
particle internal or external surface becomes a challenge in order
to avoid using too much dead weight carbon. The present invention
addresses this problem at the level of presynthetised transition
metal phosphates as well as at the level of the metal phosphate
precursors.
[0014] It is known that metal phosphate agglomerated precursors
have great impact on the structure and the properties of lithium
metal phosphate final product (WO/0227824 and WO/0227823). For
example, most of the commercially available FePO.sub.4, 2H.sub.2O
which is a precursor for LiFePO.sub.4, is prepared by a wet
chemistry process and has large dense aggregates having a mean
particle size in the range of 40-200 .mu.m and composed of fine
elementary particles having a mean particle size in the range of
0.1-1 .mu.m. Synthesis of lithium metal phosphate using large
agglomerated particle precursors requires long sintering times and
sometimes leads to large particle size, sintered material and
impurity phases due to incomplete reaction between the
reactants.
[0015] Pre-synthesis grinding of FePO.sub.4.2H.sub.2O by jet
milling can reduce the size of secondary particles to micron size,
for example D50 at 2 .mu.m and D100 at 10 .mu.m. The
electrochemical performance of a carbonated Li--Fe-phosphate
(designated by LiFePO.sub.4/C) can be significantly improved by
using air jet milled FePO.sub.4.2H.sub.2O precursor. However,
sintering still occurs inside the large agglomerates and leads to
limited power capability of an electrode made of said
LiFePO.sub.4/C.
[0016] It is known that when organic carbon precursors are used in
the process for preparing Li metal phosphate materials, the
non-agglomerated nanosize FePO.sub.4.2H.sub.2O particles used as
the precursor remain un-agglomerated even at the optimized
synthesis temperatures required to obtain lithium metal phosphate.
In contrast, dense or close porosity large particles made of
agglomerates or aggregates tend to sinter to a large degree even
when an organic precursor is used (WO/0227824 and WO/0227823). Such
dense and large particles made of agglomerates or aggregates lower
the rate performance of the final products because of low Li.sup.+
diffusion and/or lack of conductive carbon inside the
particles.
[0017] It is therefore a critical step to prepare the metal
phosphate precursor so as to achieve non agglomerated and well
dispersed fine particles in the nanometer and sub-micron range
before sintering synthesis. In another aspect of the invention, it
is also possible to create precursor agglomerates having the right
structure, porosity and carbon precursor localization from said
well dispersed nano particles in order to design optimized
agglomerates of nanoparticles or nanocomposites of the final
product. There are many ways and technologies available to obtain
non-agglomerated fine particles depending on the physical
properties of the available metal phosphate. For examples, if the
metal phosphate is not made of hard agglomerates or aggregates,
ultrasounds can be used to break the secondary particles and
disperse the elementary particles or smaller agglomerates and
stabilized the liquid suspension of those by using and organic
stabilizer or dispersant. Grinding or comminuting is one of the
most used processes allowing the production of fine particles
and/or to de-agglomerate. More recently, industrial ultra fine wet
grinding equipment have been made available commercially that can
be used to reduce particle size down to 10 to 20 nm. However, with
time the nano particles tend to re-agglomerate due to strong van
der Waals interaction or electrical double layer interaction.
[0018] Various processes have been used to make lithium metal
phosphate or carbon-coated lithium metal phosphate materials. One
of them is solid state reaction of various precursors under
reducing or inert atmospheres. Depending on the nature and particle
size of the reactants, various reaction temperatures and times are
required to achieve high purity lithium metal phosphate. In most
cases, the reaction temperature required to achieve complete
reaction is high and is accompanied with sintered aggregates or
sintering necks.
[0019] Wet chemistry methods like co-precipitation and sol-gel
synthesis have been widely investigated to make homogeneous
sintering precursors at atomic scale and in principal, a low
pyrolysis temperature is needed to achieve fine particle size of
final products. However, in practice, a segregation of reacting
species occurs, and then long reaction times or higher reaction
temperatures are required to achieve high crystallinity and high
phase purity. In consequence, the particle size and particle
morphology are complex to control.
[0020] Hydrothermal reaction is one of the most elegant methods to
synthesize lithium metal phosphate. The lithium metal phosphate
particles with various well controlled particle sizes and
morphologies under moderate hydrothermal conditions can be made.
Depending on the precursors and hydrothermal conditions, various
particle size and shapes have been reported such as submicron size
ellipsoids, micron size hexagonal plate and heavily agglomerated
nanospheres or nano-rods. Difficulties are often associated with
the control of stoichiometry, crystallinity, phase purity and
particle size.
[0021] In many of the processes reported so far, difficulties
associated with the control of particle size, phase purity and
carbon coating are the bottleneck to scale up the process. To avoid
abnormal particle growth, a low sintering temperature is required.
On the other hand, to achieve high phase purity and high carbon
conductivity, a higher sintering temperature is desired. It is
difficult to achieve all optimized parameters in one single
synthesis step.
[0022] In an earlier work the applicants have also developed a low
cost synthesis process to prepare a phosphate cathode material
which has been patented (See WO 2005/062404) but said process
results in solid crystalline ingots or micron size powders as made
by conventional grinding process.
[0023] Grinding or comminuting is one of the most used processes
allowing the production of fine particles and/or to de-agglomerate
in ceramic and paint industries. More recently, industrial wet
nanogrinding bead mill equipment have been made available
commercially, that can be used to reduce particle size down to 10
to 20 nm (See for example WO 2007/100918 for lithium metal
phosphate ultrafine grinding).
[0024] During wet nanogrinding in isopropyl alcohol solvent,
preliminary experiments on pure LiFePO.sub.4, obtained from a melt
process, inventors were drawned to conclusion that such mechanical
treatment present deleterious effects that affect the use of said
pure LiFePO.sub.4 as a cathode material. Indeed, after nanogrinding
LiFePO.sub.4 in the range of 20-30 nm, only a 4% reversible
capacity was realized in a lab-cell instead of the expected >80%
as shown and discussed in a following example. After that point, it
was concluded that wet nanogrinding a lithium metal phosphate was
altering the product. However, when nevertheless a batch of this
nanoground LiFePO.sub.4 was subsequently heat treated and used for
a pyrolysis carbon-deposit experiment the inventors surprisingly
discovered that electrochemical properties of such carbon-deposit
LiFePO.sub.4 were restored as 94% of the reversible capacity was
realized. This unexpected effect of a deterioration of pure
LiFePO.sub.4 by wet nanogrinding followed by restauration of
electrochemical properties through thermal treatment and carbon
deposition by pyrolysis is a main object of the present invention
as well as the use of different organic surfactants, adsorbant and
carbon precursors that facilitate the wet grinding process and that
are converted to non contaminating and conductive carbon to make
and optimize the C deposited nano particle or nanostructured
lithium metal phosphates particle or agglomerates.
[0025] The present invention provides a method for preparing
carbon-deposited cathode nano materials, including a from molten
lithium metal phosphates process and ingots in an easy way and
resulting in a high performance cathode material.
SUMMARY OF THE INVENTION
[0026] In one aspect, the present invention provides a method for
preparing a complex oxide particle composition, wherein the complex
oxide particles have a non powdery conductive carbon deposit on at
least part of their surface. The method comprises nanogrinding
complex oxide particles or particles of complex oxide precursors,
wherein: [0027] an organic carbon precursor is added to the oxide
particles or oxide precursor particles before, during or after
nanogrinding, and pyrolysing the mixture thus obtained; [0028] a
stabilizing agent is optionally added to the oxide particles or
oxide precursor particles before, during or after nanogrinding;
[0029] the nanogrinding step is performed in a bead mill on
particles dispersed in a carrier liquid, and; [0030] the size of
the particles to nanogrind, the size of the beads used to
nanogrind, and the size of the resulting particles are important
process characteristics and are selected such that:
[0030] 0.004.ltoreq.MS(SP)/MS(B).ltoreq.0.12,
0.0025.ltoreq.MS(FP)/MS(SP).ltoreq.0.25 wherein MS(SP) represents
the mean size diameter of the particles before grinding (starting
particles), MS(FP) represents mean size diameter of the particles
after grinding (final particles), and MS(B) is mean size diameter
of the grinding beads.
[0031] As a very empiric rule, optimizing can be started with a D90
(SP) to (B) ratio of 1/10, and a (B) to D90(FP) ratio of 1000. A
preferred mean size of beads ranges from 100-500 .mu.m.
[0032] In another aspect, the invention provides a particle
composition. The particle composition comprises particles having a
complex oxide core and a conductive carbon deposit on at least part
of the surface of the core, wherein: [0033] the particles comprise
nanoparticles having a nanoscale size and agglomerates of
nanoparticles having a submicron to micron scale particle size;
[0034] said conductive carbon deposit is a non powdery deposit, and
is present on at least part of the surface of the elementary
particles and of the surface of the agglomerates.
[0035] In a further aspect, the invention is related to the use of
the particle composition as an active electrode material, a
nanocomposite electrode material comprising said particle
composition as the active electrode material, and an
electrochemical cell wherein at least one electrode comprises said
nanocomposite electrode material.
BRIEF DESCRIPTION OF THE DRAWINGS
[0036] FIG. 1 illustrates the XRD pattern for molten LiFePO.sub.4
powder before nanogrinding (a), after nanogrinding in IPA (b) and
after thermal treatment and pyrolysis (c).
[0037] FIG. 2 illustrates a TEM micrography of ground agglomerated
particles of the molten LiFePO.sub.4 after wet milling in IPA in
TEM with a low magnification.
[0038] FIG. 3 illustrates an agglomerated particle of FIG. 2 at
high magnification showing 20-40 nm nanocrystallites reagglomerated
from grinding and some alteration layer on the surface.
[0039] FIG. 4 is a SEM image of large secondary particles from
nanogrinding in IPA after thermal treatment and pyrolysis.
[0040] FIG. 5 is a SEM image representing a detail of the secondary
particles of FIG. 4 at higher magnification with partially sintered
nano-aggregates, carbon bridging and evidence of porosity in the
secondary particles.
[0041] FIG. 6 represents a Ragone plot (mAh vs discharge rates) for
the secondary particles of Example 1.
[0042] FIG. 7 is a SEM image of the sherical secondary particles
(aggregates) obtained by spray drying in Example 3.
[0043] FIG. 8 is a SEM image of a detail of FIG. 7, representing
nano-sized aggregated particles with evidence of porosity.
[0044] FIG. 9 is a SEM image of the aggregated 250-350 nm particles
of example 4 using lactose and water.
[0045] FIG. 10 illustrates a particle size distribution (PSD) vs
time, C--LiFePO4 without, with Triton and Triton+Unithox 550 as per
Example 5.
[0046] FIG. 11 represents: [0047] a) a SEM image of the
Li.sub.2CO.sub.3+FePO.sub.4 2H.sub.2O+Unithox copolymer precursor
mixture of Example 6 showing the carbonate and the phosphate
particles; [0048] b) for a SEM image of an equivalent mixture on
which the adsorbed copolymer is visible on the particles due to Bi
complexation and visibility in SEM, [0049] c) a SEM image
equivalent to a) but after pyrolysis in which C deposit on
LiFePO.sub.4 and C bridging are now visible [0050] d) a RD-EDS
diagram confirming the presence of Bi on all the surface.
[0051] FIG. 12 is a representation of the apparent particle sized
and distribution evolution upon wet grinding without, with Triton
and Triton+Unithox copolymer from Example 7.
[0052] FIG. 13 is a SEM image of C-deposited nanostructured
LiFePO.sub.4 of Example 7 after pyrolysis.
[0053] FIGS. 14 and 15 are SEM images of C--LiFePO.sub.4 aggregated
particles (200-500 nm) obtained after nanogrinding a
Li.sub.2CO.sub.3+FePO.sub.4 2H.sub.2O mixture with Triton as a
surfactant (FIG. 14) and without surfactant (FIG. 15) and pyrolysis
at 700.degree. C. in the presence of Unithox and IPA.
DETAILED DESCRIPTION
[0054] The method of the present invention is particularly useful
for the preparation of a particle composition wherein the complex
oxide is at least one compound having an olivine structure and
formula AMXO.sub.4 wherein: [0055] A is Li, optionally partly
replaced with not more than 10 atomic % Na or K; [0056] M
represents Fe.sup.II or Mn.sup.II, optionally partly replaced with
not more than 50 atomic % of at least one metal selected in the
group consisting of Fe, Mn, Ni et Co, and optionally replaced with
not more than 10 atomic % of at least one aliovalent or isovalent
metal different from Fe, Mn, Ni or Co; [0057] XO.sub.4 represents
PO.sub.4, optionally partly replace with not more than 10 mol % of
at least one group selected from SO.sub.4 and SiO.sub.4.
[0058] In a specific embodiment, the aliovalent or isovalent metal
different from Fe, Mn, Ni or Co in the complex oxide is at least
one metal selected from the group consisting of Mg, Mo, Nb, Ti, Al,
Ta, Ge, La, Y, Yb, Sm, Ce, Hf, Cr, Zr, Bi, Zn, Ca et W.
[0059] Complex oxides LiFe.sub.1-xMn.sub.xPO.sub.4,
0.ltoreq.x.ltoreq.0.5, LiFePO.sub.4 and LiMnPO.sub.4 are
particularly preferred.
[0060] These lithium transition metal phosphates are all of the
olivine structure and tend to behave similarly mechanically upon
grinding, especially when made by a melting process.
[0061] Furthermore, the method of the present invention is useful
to prepare a particle composition wherein the complex oxide is a
titanate which has a spinel structure and the formula
A.sub.aM.sub.mO.sub.oN.sub.nF.sub.f wherein A represents an alkali
metal; M represents Ti alone, or Ti partly replaced with another
metal, preferably a transition metal; a>0, m.gtoreq.0, o>0,
n.gtoreq.0, f.gtoreq.0, and coefficients a, m, o, n and f are
selected to provide electroneutrality of the complex oxide.
[0062] A is preferably Li, optionally partly replaced with another
alkali metal.
[0063] In a preferred embodiment, the titanate is of formula
Li.sub.4+xTi.sub.5O.sub.12 or Li.sub.4+x-2y
Mg.sub.yTi.sub.5O.sub.12, wherein 0.ltoreq.x et y.ltoreq.1,
preferably LiTi.sub.5O.sub.12.
[0064] The conductive carbon precursor is preferably a low
viscosity, optionally crosslinkable, polymerizable, or
polycondensable compound which is able to wet, penetrate and adsorb
on reconstituted agglomerates of nanoparticles of the precursors or
the complex oxide. The carbon precursor may be any liquid, solid
(in solution) or gaseous organic compound that leaves carbon upon
pyrolysis. It can combine different functionalities (surfactant,
self adsorbing, lubricant, catalyst, for example) and consist in
one or several products. The conductive carbon precursor is
preferably intimately mixed with the complex oxide particles or the
complex oxide precursor particles, in order to achieve impregnation
of the nanoparticle surface and of the agglomerates of
nanoparticles, so that after pyrolysis, the carbon deposit is in
intimate contact with the complex oxide.
[0065] The method of the invention preferably comprises a further
step which is performed after grinding and before pyrolysis, said
further step comprising conditioning the reaction mixture in order
to adsorb the carbon precursor on the complex oxide precursors or
on the complex oxide, or to polymerize or cross link a carbon
precursor which is monomer. The method may also comprise a further
step consisting in aggregating the reaction mixture comprising the
carbon precursor and the complex oxide precursor after grinding.
Aggregation can be performed by flocculating, by spray drying, or
by charge effect.
[0066] In a preferred embodiment, the method of the present
invention comprises the following steps: [0067] Grinding or
de-agglomerating a complex oxide or complex oxide precursors to
nano size or sub-micron size in the presence of an Organic
Precursor (OP) that might include a surfactant or not. [0068]
Adsorbing and distributing OP on the external surface of elementary
particles to stabilize their dispersion or on the external and
internal surface of the to particles of the agglomerates to
distribute the future C deposit through its precursor before,
during or after grinding. [0069] Pyrolysing the reaction mixture to
convert the adsorbed and localised OP to C before, during or after
final phosphate synthesis, when a phosphate precursor is used.
[0070] Alternatively similar pyrolysis treatment can be done after
synthesis and grinding to obtain similar electrochemically active
electrode material, i.e., nanosized metal phosphate on which a
conductive carbon deposit is attached to the crystal structure.
[0071] In one embodiment, the method provides a particle
composition, starting from compounds which are precursors for the
complex oxide LiMPO.sub.4. In this embodiment, an organic
stabilizing agent is preferably added to the suspension of initial
particles before grinding. The organic stabilizing agent modifies
the surface charge and increases the repulsive force between
particles to counter the Van der Waals force. Thus, the dispersion
is stabilized and the degree of agglomeration of nanoparticles is
controlled.
[0072] The inventors demonstrated that precursors of the complex
oxide can be nanoground together with the carbon precursor to
directly obtain a particle composition according to the invention
having high level size and shape properties, wherein said size and
shape are substantially similar to those of the nanoground
precursor particles. The attached examples demonstrate that
nanogrinding is easier to perform on precursor particles than on
complex oxide particles, possibly because of the presence of the
carbon precursor which might have a tension active effect and which
adsorbs on the others precursors.
[0073] The organic stabilizing agent can be selected from organic
electrostatic or electrosteric stabilizers, surfactants, dispersant
agents and encapsulant agents, many of them being available
commercially.
[0074] The organic stabilizing agent is preferably selected from
compounds which, upon grinding or pyrolyzing, do not generate side
effects such as highly toxic gas or compounds which would be
detrimental to the performance and cyclability of an
electrochemical cell comprising the particle composition of the
invention.
[0075] The organic stabilizing agent may additionally act as a
precursor for the conductive carbon, depending on its carbon
content.
[0076] The Li precursor may be selected from lithium salts such as
Li.sub.2CO.sub.3, LiOH, LiH.sub.2PO.sub.4, Li.sub.3PO.sub.4 for
example.
[0077] The Fe (or equivalent Mn analogs) precursor may be selected
from FePO.sub.4 nH.sub.2O, Fe.sub.3(PO.sub.4).sub.2nH.sub.2O, iron
sulphate, different iron oxides or hydroxides, iron salts of
inorganic and organic compounds.
[0078] The P precursor may be selected from derivatives of
phosphoric acid or P.sub.2O.sub.5 or mono or di ammonium acid
phosphate salts or salts that combining phosphates or
polyphosphates with the Li or Fe precursor.
[0079] Some of the aforementioned compounds can be used as the
precursor for more than one element.
[0080] When the particle composition of the present invention is
prepared starting from complex oxide precursor particles, pyrolysis
to convert the adsorbed organic carbon precursor to conductive
carbon can be performed before, or during the synthesis of the
complex oxide.
[0081] In one embodiment, the method provides a particle
composition, starting from complex oxide LiMPO.sub.4 particles. In
this embodiment, the carrier liquid is preferably a reactive
liquid, selected from water or alcohol [for example isopropyl
alcohol (IPA) or butanol].
[0082] The complex oxide can be prepared by well known prior art
method, for instance by a solid state reaction of precursors under
reducing or inert atmosphere if M is Fe, or in oxydizing atmosphere
(for instance ambient air) when M is Mn, by co-precipitation or
sol-gel synthesis, or by a hydrothermal reaction. The particle size
of the particles before grinding is preferably in the range from 1
.mu.m to 50 .mu.m. The complex oxide can also be prepared by
reacting the precursors in molten state in an inert or reducing
atmosphere, the complex oxide being pre-ground after synthesis and
solidification.
[0083] A particle composition according to the present invention
comprises particles having a complex oxide core and a conductive
non powdery carbon deposit. The particles comprise elementary
nanoparticles and micron size agglomerates or aggregates of
elementary nanoparticles.
[0084] Elementary nanoparticles have dimensions ranging from 5 nm
to 1.0 .mu.m, preferably between 10 and 600 nm and comprise primary
nanoparticles and secondary particles.
[0085] A primary nanoparticle is made of a complex oxide with or
without C.
[0086] A secondary particle is an agglomerate or an aggregate of
primary particles, and may also contain other constituants such as
internal or external C-deposit or carbon bridging or particulate
carbon, other inert or conductive phases or sintering necks. A
secondary particle may also have a porosity.
[0087] An "aggregate of primary nanoparticles" herein means a
micron-size assembly of primary nanosize particles held together by
physical or chemical interaction, by carbon bridges, or bridges of
locally sintered complex oxide containing a minimum degree (0-30%)
of internal open porosity and carbon deposit. An "agglomerate"
means a assembly of particles loosely held together by low forces.
Agglomerates are formed continuously during the grinding process,
during which there is a balance between comminution and
re-agglomeration. The method of the invention allows changing the
balance by the use of surfactants, by varying the milling
conditions or by adsorbing or encapsulating the carbon precursor on
the elementary nanoparticles. When adsorbing or encapsulating the
carbon precursor on the elementary nanoparticles, the elementary
nanoparticles are encapsulated by the carbon precursor, aggregates
are formed using the precursor to bond elementary nanoparticles,
said bonds being converted upon pyrolysis to carbon or sintering
neck to form aggregate particles which are like bigger elementary
particles. When agglomeration of elementary nanoparticles is formed
faster or before the surfactant or precursor can penetrate into the
agglomerate and this agglomerate is treated during pyrolysis, an
important sintering of the elementary nano particles agglomerate
into a larger particles that is only carbon deposited on the
surface can be observed.
[0088] The conductive carbon deposit is attached to the complex
oxide crystal structure in the form of nanosize layers of carbon,
preferentially of graphitized or "graphene" carbon. "Carbon deposit
attached to the complex oxide" means that there is an intimate
contact through physical or chemical bonding between the complex
oxide crystal and the carbon deposit. It is believed but not
limitatively, that the pyrolysis mechanism involves radical
formation, and gaseous species can result in chemical as well as in
physical bonding between the sp3 C and the PO.sub.4 entity. Such an
attached carbon deposit is used not only to induce conductivity and
homogeneity of the electrical field of the particles, but also to
partially or totally avoid extremely favourable nano particle
sintering. When the amount of carbon deposit is high enough and
covers most of the surface of the complex oxide particles, it may
achieve interparticle bounding by carbon bridges. When only part of
the surface of the complex oxide is covered by the carbon deposit,
it allows local sintering of the complex oxide nanoparticles,
providing bonded agglomerates of nanoparticles having an open pore
structure, which allows solvated lithium penetration into the
micron size agglomerates.
[0089] It has been found that the organic carbon precursor
pyrolysis mechanism in contact with transition metal of the complex
oxide allows the growth of at least partially graphitized layers on
the complex oxide crystal in a mechanism that might, although non
limitatively, involve a gas phase mechanism of carbon growth. The
nature, ratio and localization of the carbon deposit can be
controlled by selecting the appropriate organic precursor, by the
mechanism of adsorption of the carbon precursor on the surface of
the complex oxide particle or the complex oxide precursor particle,
and by pyrolysing on the external surface of the elementary
nanoparticles or on the internal or the external surface of the
agglomerates. The thickness of the carbon deposit is selected
depending on the application It can be between a few nm and a few
tenth of nm, (for instance 0.5 nm-50 nm) thus requiring very low
weight ratio carbon/complex oxide (for instance 0.5 to 10%,
preferably 1 to 5%) in order to achieve an efficient electronic
conduction by the carbon film (in case of a continuous deposit or
by short distance tunnelling mechanism (in case of a discontinuous
deposit or on a particulate deposit.
[0090] The particle composition of the present invention can be
used as an active material for an electrode. An electrode according
to the present invention comprises a nanocomposite material applied
on a current collector, said composite material comprising said
particle composition, pores, a binder and optionally an agent
providing electronic conductivity.
[0091] In a particle composition of the invention to be used as an
active material for a cathode, the complex oxide is preferably a
LiMXO.sub.4 oxide. In, a particle composition to be used as an
active material for an anode, the complex oxide is preferable a
titanate.
[0092] In an embodiment, the nanocomposite material comprises a
particle composition of the invention, wherein at least 50% of the
elementary nanoparticles have a size between 5 nm and 900 nm
diameter, preferably between 10 nm and 300 nm, said nanoparticles
being not agglomerated or sintered. It is pointed out that, if the
particle composition is prepared from precursors of the complex
oxide, the method of the invention produces particles which have
substantially the same dimension and shape as the precursor
particles, if the starting material is made of precursors of the
complex oxide.
[0093] In another embodiment, the nanocomposite material comprises
a particle composition of the invention, wherein the elementary
nanoparticles are agglomerated to form aggregates having a size
from 0.2 .mu.m and 10 .mu.m, preferably from 0.5 .mu.m to 5
.mu.m.
[0094] In another embodiment, the conductive carbon deposit
attached to the complex oxide crystal structure on at least part of
the surface of the nanoparticle has a nanoscale thickness.
[0095] In another embodiment, the conductive carbon is present on
part of the surface of the complex oxide nanoparticles, and the
nanoparticles are sintered at the complex oxide surface
thereof.
[0096] In another embodiment, where the major part of the surface
of the complex oxide nanoparticles is covered with the conductive
carbon deposit, the nanoparticles are aggregates via carbon
bridges.
[0097] In another embodiment, the nanocomposite material further
contains at least a binder or an electron conduction additive. The
binder is preferably a fluorinated or partially fluorinated
elastomer, water or an organic soluble or dispersable binder
including latex, or SBR The electron conduction additive. is
preferably selected from the group consisting of carbons, carbon
blacks, conductive polymers and graphite, intermetallic or metallic
powder, fibers or platelets. Nanoparticles of other cathode
materials might also be included.
[0098] In a further embodiment, the particle composition of the
nanocomposite material comprises secondary particles and or
agglomerates of elementary nanoparticles and has an open porosity,
the volumetric fraction of the pores ranging from 0.30 to 0.05,
preferentially from 0.2 to 0.1.
[0099] In another embodiment, the conductive carbon deposit is at
least partly graphitized carbon obtained by pyrolysis of an organic
carbon precursor that contains elements such as N, P, Si that can
be covalently bound to carbon.
[0100] The process of the invention to make nano-sized particles or
agglomerates themselves coated at a nanoscale with conductive
carbon makes such molten process viable for high power cathode
materials.
[0101] The present invention addresses the preparation and
optimisation of complex oxide-carbon cathode materials in which a
conductive carbon is chemically fixed to the complex oxide crystal
structure by pyrolysing an organic carbon precursor that is in
intimate contact with nanoparticles or with nanocomposite
aggregates of the complex oxide or the complex oxide precursor,
preferably by an adsorption process or a chemical linkage
process.
[0102] An appropriate selection of the amount, nature, thickness
and distribution of said carbonaceous deposit can be used to
control the characteristics of the particle composition of the
invention. For example, nanoparticle sintering can be avoided while
leaving open porosity between nanoparticles in the aggregates, or
on the contrary, partial bridging between nanoparticles of
agglomerates can be provided by allowing local interparticle
sintering necks to form or by creating inter nano particle carbon
bridges, or by selecting in order to allow inter nano particle
partial of the aggregates or to establish carbon bridges.
[0103] In one preferred mode of realization, the nanoparticle
suspensions are obtained by micromedia bead mill grinding or
deagglomeration of a suspension of solid particles of the complex
oxide or of the complex oxide precursor in a liquid media.
[0104] The wet grinding device may be selected from bead mills that
can reduce the particles size down to nanometer range.
Particularly, mention may be made of Ultra APEX Mill by Kotobuki
Industries Co. Ltd of Japan, High speed Netzsch Zeta agitator bead
mill by Netzsch of Germany, Hosokawa Alpine AHM mill by Hosokawa of
Japan, and MicroMedia.RTM. P1 & MicroMedia.RTM. P2 bead mill by
Buehler of Switzerland. The grinding beads. may be made of alumina,
zirconia or carbides for example.
[0105] The organic carbon precursor is preferably selected from
organic compounds which are able to form conductive carbon deposit
upon pyrolysis in the presence of the complex oxide or precursors
thereof, and to wet, impregnate, encapsulate and preferably adsorb
and/or self-organize on at least part of the surface of the complex
oxide or its precursor in order: [0106] to stabilize the
nanoparticle dispersion [0107] to help controlling the
re-agglomeration or aggregation of the complex oxide or the
precursor thereof during or after the grinding/desagglomeration
step in solution or during spray/evaporation techniques. [0108] to
leave by pyrolysis a carbon deposit localized on the surface of the
nanoparticles or on the external or internal surface of the
agglomerates or aggregates of nanoparticles [0109] to create C
bounding between elementary particles or complex oxide bounding
between elementary particles to form micron size aggregates or
nanocomposites.
[0110] The organic carbon precursor is advantageously selected from
crosslinkable monomers or oligomers, polymers, copolymers
(especially block copolymers) and high carbon content surfactants.
However liquid organic compounds in solution or solid organic
compounds are also possible. The organic precursor can combine more
than the C-source function and can also be selected to act as a
surfactant, a stabilizer etc.
[0111] Numerous products commercially available can be used as the
organic stabilizing agent. They include surface active agents (also
known as surfactants) It is an important aspect of the invention to
use low-cost and registered stabilizers. Most of these organic
compounds are amphiphilic products containing an hydrophilic part
which is ionic or non-ionic, and hydrophobic part allowing
modification of particle/solvent surface tension, wettability and
more efficient dispersion of the particle in the carrier liquid.
These products and mixtures thereof are often characterized by
their HLB number corresponding to balance between hydrophobic and
hydrophilic moieties. A large set of possible surfactant is
provided in Stepan Global Product Catalog incorporated herein by
its reference. Many others are available from worldwide specialty
chemicals manufacturers.
[0112] The surfactant may be selected for example from fatty acid
salts (for example oleic acid or lithium oleate), fatty acid
esters, fatty alcohol esters, alkoxylated alcohols, alkoxylated
amines, fatty alcohol sulfate or phosphate esters, imidazolium and
quaternary ammonium salts, ethylene oxide/propylene oxide
copolymer, ethylene oxide/butylene oxide copolymer and from
reactive surfactants.
[0113] The fatty acid esters surfactants can be prepared through
esterification. Numerous cost-effective combinations exist,
allowing for fine-tuning of surfactants properties in terms of
solubility/insolubility in various solvents, dispersibility of
submicron or nanosize complex oxide cathode material. A major
advantage of using fatty acid esters is that they can be used as a
carbon precursor which provides a high quality carbon deposit
generated upon pyrolysis of the fatty acid chains. Non-ionic fatty
acids are mainly obtained by esterification of a fatty acid with
glycol products (glycerol, glymes, . . . ). The carbonization ratio
depends on the fatty acid content, the surfactant and the fatty
acid weight. To avoid low carbonization ratio and generation of a
large amount of ashes during carbonization process, fatty acid with
molecular weight >250 are preferred. Mention may be made of
caprylate, undecylenate, palmitate, laurate, myristate, oleate,
ricinoleate, linoleate, linolenate, and stearate. Oleate, stearate,
linoleate, linolenate, and ricinoleate are preferred, more
particularly oleate and stearate, if handling/safety is considered.
If to a high carbonization ratio is a concern, glycerol monooleate
or monostearate are of particular interest. With a fatty acid
containing the same amount of carbon, the presence of insaturation
modify the solubility properties. For example, glycerol monooleate
is soluble in isopropyl alcohol (IPA) whereas glycerol monostearate
is poorly soluble. Lower solubility of glycerol monostearate is
also of interest when such compounds are processed with high-shear
mixing, especially in bead mills according to the method of the
invention. Such high-energy mixing will allow efficient and
homogeneous dispersion in IPA of low-solubility glycerol
monostearate which is further stabilized by adsorption on complex
oxide particles. Optimization of surfactant formulation is also
easily obtained by esterification of fatty acid with glymes to
produce surfactants such as the following oleate derivatives
C.sub.17H.sub.33--COO(CH.sub.2CH.sub.2O).sub.2OH or
C.sub.17H.sub.33--COO(CH.sub.2CH.sub.2O).sub.9OOC--C.sub.17H.sub.33.
[0114] Length of the glyme part and choice of the fatty acid allow
preparation of surfactant with suitable HLB value and desirable
melting point, boiling point, solubility/insolubility, wettability
in carrier solvent in view to obtain high quality carbon coating
after pyrolysis. An important point to consider from an industrial
perspective is that optimization of formulation is done at almost
constant cost of an already cost-effective solution.
[0115] Some derivatives of fatty acid are also of particular
interest. First of all, sugar-ester compounds composed of an
hydrophilic sugar part, especially sucrose, sorbitol and sorbitan,
an hydrophobic fatty acid part, and optionally a polyethylene oxide
segment depending on the desired HLB value. For example, mention
can be made of the Tween.RTM. surfactants produced by Uniquema, and
especially Tween.RTM. 80 and 81 (polyoxyethylenesorbitan
monooleate), or Tween.RTM. 85 (polyoxyethylenesorbitan trioleate).
Polyoxyethylene sorbitol hexaoleates are also important
surfactants.
[0116] Tall oil obtained as a by-product of wood pulp manufacture
is also an interesting source of fatty acid derivatives, especially
grades obtained after fractional distillation tall oil rosin and by
further distillation tall oil fatty acid which is a low cost,
consisting mostly of oleic acid, source of fatty acids. Tall oil
and tall oil fatty acid are available from numerous supplier (for
example Arizona Chemical) including in the form of ester with
glycerol or glymes.
[0117] As carbonization ratio depends on molecular weight/boiling
point of fatty acid, it is also of particular interest to use fatty
acid oligomers obtained from unsaturated fatty acid (oleate,
linoleate, . . . ). For example, mention can be made of the
Unidyme.RTM. fatty acid oligomers available from Arizona Chemical.
Dimerized product, especially dimerized oleic acid, used in the
form of polyamide in ink industry are also of interest and are
produced for example by Henkel or Arizona Chemical.
[0118] In a particular embodiment, a fatty acid salt of a
transition metal cation is used as the surfactant and the organic
carbon precursor, and the carbon deposit which is generated by
pyrolysis is in the form of carbon nanotubes. The transition metal
cation acts as a catalyst for the nanotube formation. The
transition metal is preferably selected from Ni, Co or Fe. The
fatty acid contains preferably at least 6 carbon atoms, more
preferably at least 10 and most preferably 14. The fatty acid is
preferably selected from stearate, oleate, linoleate, linolenate,
ricinolenate, preferably oleate and stearate. The use of nickel
stearate as a precursor for carbon in the form of nanotubes
precursor is described for example in J. Mater. Chem., 2005, 15,
844-849, which. By a proper choice of the fatty acid of the fatty
acid salt of a transition metal, the solubility of said salt can be
easily adjusted to suitable solvent carrier.
[0119] Alcoxylated alcohols may be selected from those which are
obtained from ethylene oxide and/or propylene oxide. Most common
alcohol precursors are fatty alcohols and alkyl-phenols (for
example octyl or nonylphenol), especially the alkoxy alcohols
available under the trade name Igepal.RTM., from Rhodia Inc or
Brij.RTM. surfactants. Alkoxylated amines are provided by Huntsman
under the trade names Jeffamine.RTM. and Surfonamine.RTM..
Surfonamine.RTM. is an EO/PO amine of particular interest as
dispersant and carbon precursor, the PO part allowing carbon
generation during pyrolysis.
[0120] Fatty alcohol sulfate or phosphate esters, including their
zwiterrionic form, are available for example from Stepan Company.
In the method of the present invention, the phosphate esters are
preferred. Special attention should be drawn to the
styreneoxide-based phosphorylated polyether available form Degussa
and of the following formula.
##STR00001##
[0121] Imidazolium and quaternary ammonium based surfactants are
available from Degussa under the trade name Tego Dispersant, for
example the compounds of followings formulae
##STR00002##
[0122] Ethylene oxide/propylene oxide copolymer surfactants are
mainly known as Pluronic.RTM., the poly oxypropylene part allowing
carbon generation during pyrolysis (see for example Chem. Commun.,
2003, 1436-1437). Modification of the EO/PO ratio and of the
molecular weight provides a large choice of cost-effective
tension-active agents with tunable properties in terms of
solubility, surface-tension, wettability, and carbonization ratio.
Important physico-chemical data on the Pluronic.RTM. products is
provided by BASF on
http://www.basf.com/performancechemical/bcperfdata_tables.html.
[0123] Polyanhydride resins obtained by alternate copolymerization
of maleic anhydride with an alkylene are also an important class of
compounds effective as surfactant and/or carbon precursor. Of
particular interest is poly(maleic anhydride-alt-1-octadecene)
produce by Chevron Phillips Chemical Company. This high molecular
weight polymer is soluble in IPA due to esterification of
anyhydride part by isopropanol.
[0124] As IPA is an important solvent in the method of the present
invention, mention be also made of use of alkylated sucrose,
especially sucrose acetate isobutyrate produce by Eastman.
Polyvinyl butyrals IPA soluble polymers available form Wacker under
the trade name Pioloform.RTM. can also be used as surfactant and/or
carbon precursors.
[0125] Reactive surfactants so called "Surfmer", are non-ionic,
cationic and anionic compounds (see Acta Polym 95, 49, 671).
"Reactive surfactant" means a surfactant containing a polymerizable
group through anionic, cationic or radical polymerization (for
instance an epoxyde, allyl, vinyl, acrylate, methacrylate,
vinylether, or maleimide group), a condensable group (for example
an amine, carboxylic acid, or alcohol group) or a chemically
reactive group (for example an isocyanate, blocked isocyanate,
carbodiimide, or epoxy group). Typical examples are the compounds
of the following formulae
##STR00003##
[0126] Noigen is available from DKS, Japan, and Hitenol is
available from DKS Japan. Other suitable compounds are available
from Uniquama under the trade name Maxemul. Use of reactive
surfactant, in whole or in part, is an important way to induce
nano-encapsulation of the complex oxide cathode material during or
at the end of grinding process. In a specific embodiment of the
invention, such additives are used only at the end of the grinding
process to encapsulate nano-powder.
[0127] The above products which are industrial compounds, combined
with a grinding process to lower particles sizes, allow
optimization of the production of battery grade nanosize and
submicron carbon coated complex oxide, in particular LiFePO.sub.4,
especially for high-power batteries, in term of cost-effectiveness,
safety (low hazard, low VOC, . . . ). The organic precursors are
preferably selected to form a thin nanoscale carbon deposit that
will be intimately contacted to the complex oxide crystal
structure, that will be at least partially graphitized during
pyrolysis. The organic precursor may contain elements such as N, P,
Si that may remain in the carbonaceous deposit after pyrolysis.
[0128] Optionally these organic precursors may be present in at
least the gas phase in equilibrium with surface distributed organic
precursor during the pyrolysis stepand able to grow graphite or
graphene layers on the surface of the metal phosphate. Optionally,
iron, cobalt or nickel catalyst can be present during the pyrolysis
process to promote a conductive C deposit of graphene or graphitic
nature. The metal catalyst may be introduced and distributed also
as a metal containing surfactant such as Fe, Co or Ni stearate or
oleate.
[0129] Preferably the nature, distribution and amount of organic
precursor or the carbon deposit after pyrolysis are adjusted to
avoid nanoparticle sintering (by carbon coating) or to control
partial sintering `with limited carbon quantities) on the complex
oxide particles or in contrast to form interparticle carbon bridges
at the nanoscale level.
[0130] Thermal treatment promotes sintering of primary
nanoparticles inside an aggregate of nanoparticles, providing
secondary particles. The size of said secondary particles is
influenced by the nature of solvent, concentrations, organic
precursor properties (adsorbed or not, for example) and thermal and
pyrolysis processes. In some cases, the aggregates have some
porosity or they are spherical if there is an important sintering
for example. These aggregates might retain the primary nanoparticle
shape or only an element of the nanoparticle like the C-deposit or
some open or closed pores. Since the invention has shown that the
shape and size of the complex oxide precursor is maintained if the
particles are properly coated with a carbonaceous deposit, both the
final synthesized product or the metal precursor can be
processed.
[0131] The following examples are intended to illustrate the
present invention more practically, but not to limit the invention
in any way.
[0132] In all examples, each grinding experiment was performed
using a wet grinding machine available from NETZSCH, Inc. under the
trade name Netzsch LABSTAR (Z) model LABSTAR LS1.
EXAMPLE 1
Synthesis
[0133] In first step, LiFePO.sub.4 was synthesized by melt casting
using the process described in WO05.062404. 2 FePO.sub.4.2H.sub.2O
and 1 Li.sub.2CO.sub.3 were mixed at nominal LiFePO.sub.4
composition with an excess of 0.5 mole of EBN-1010 (graphite
powders), and then heated to 1050.degree. C. in a graphite crucible
under inert atmosphere in a furnace. The melt was held at
1050.degree. C. for 1 h and then cooled down in the furnace. X-ray
analysis has confirmed that the obtained ingot has a LiFePO.sub.4
main phase and minor Li.sub.4P.sub.2O.sub.7 and
Fe.sub.2P.sub.2O.sub.7 phases as shown in curve a of FIG. 1. The
impurity phase accounts for less than 3% of the total
materials.
[0134] In a second step, the ingot was crashed into millimeter
sized particles by using a jaw crusher with ceramic liner to avoid
metal contamination. The millimeter sized particles are further
ground by using ball milling to achieve micrometer sized particles.
Finally, the micrometer sized powders were dispersed in IPA
solution at 10-15% of solid concentration and then ground on a bead
mill using 0.2 mm zirconia beads to achieve nanometer sized
particles.
[0135] X-ray analysis shows that the olivine structure is still
preserved, but the diffraction peaks become very broad due to small
crystallite size as shown on curve b in FIG. 1. The crystalline and
microstrain calculated form the peak width are 30 nm and 2%
respectively.
[0136] TEM observations show that wet milling leads to nanometer
sized primary particles in the range of 20-30 nm (FIG. 4) and these
primary nanoparticles are re-agglomerated to submicron sized
particles (FIG. 2, 3) in the range of 100-500 nm depending on the
concentration of solid in liquid, the nature of the liquid media
and the surfactant being used. There is a thin mechanically and
chemically distorted surface layer on the particle surface (see
FIG. 3).
[0137] In a third step, a solution of poly(maleic
anhydride-1-alt-octadecene) dissolved in IPA is mixed with the
LiFePO.sub.4 in IPA, in a ratio of 5 wt. % poly(maleic
anhydride-1-alt-octadecene) over LiFePO.sub.4. The mixed solution
was stirred thoroughly and then dried at room temperature by
blowing with dry air while stirring.
[0138] In a final fourth step, the dried powder is heated to
727.degree. C. at 6.degree. C./min and held for 1 h at 727.degree.
C. in a rotary kiln, and the cooled in a furnace at a cooling rate
of 2.degree. C./min. The heat treatment is protected by using argon
flow. After said treatments, large aggregates of carbon coated
nanoparticles having a diameter of 50-200 nm carbon are obtained.
The pyrolytic carbon content is 1.4% measured by a C, S analyser
LECO method. The product thus obtained is designated by
C--LiFePO.sub.4.
Characterization
[0139] X-ray analysis (see curve c of FIG. 1) shows that the
sintered product contains mainly LiFePO.sub.4 as a main phase with
a small proportion of Fe.sub.2P.sub.2O.sub.7 minor phase. The
diffraction peaks become sharper in comparison to that of the
as-milled product due to grain growth and structure restoration
through the thermal treatment. The crystallite size is about 190 nm
after thermal treatment. The strain is sharply reduced from 2% to
0.38%. SEM and TEM observations show that a thin layer of carbon
deposition is formed on the nanoparticle surface and that the
nanoparticles are bonded together by carbon-bridge forming
aggregates showing evidence of open porosity (see FIG. 4, 5). The
crystallite size of the thermal treated materials is roughly the
same as the primary particle size. This indicates that the each
primary particle is a single crystal after thermal treatment. TEM
observation indeed confirm that each primary particle is single
crystal.
[0140] C--LiFePO.sub.4, as prepared in the present example was used
to prepare a cathode for a liquid electrode battery.
C--LiFePO.sub.4, PVdF-HFP copolymer (from Atochem) and EBN1010
graphite powder (from Superior Graphite) were thoroughly mixed in
N-methylpyrolidone (NMP) with zirconia balls for 1 hour on a
turbula shacker, such as to obtain a 80/10/10 wt % proportion of
the components. This slurry was then coated on a carbon-coated
aluminum foil (from Intellicoat) with a Gardner coater, the coating
was dried under vacuum at 80.degree. C. during 24 hours prior to
storage in a glove box. A button type battery has been assembled
and sealed in a glove box using the cathode thus obtained, a 25
.mu.m microporous separator (from Celgard) impregnated with 1M/1
LiPF.sub.6 salt in EC:DEC electrolyte and a lithium foil as
anode.
[0141] Electrochemical measurement shows that the as-milled nano
LiFePO.sub.4 with surface distortion layer gives very low
electrochemical activity. Only 4% of reversible capacity was
realized. In contrast, after the heat treatment under inert
conditions and carbon deposition, 94% of reversible capacity was
realized. The thermal treatment of the as-milled LiFePO.sub.4 is an
essential step of the method to restore the structure and/or
chemistry and to apply carbon deposition in order to achieve good
electrochemical performance.
[0142] A Ragone plot as represented on FIG. 6 shows that the
agglomerates of nanoparticles can deliver very high power at 10 C
rate and confirms that a molten LiFePO.sub.4 ingot can lead to a
high power material after grinding down to at a nano level (20-30
nm) more or less re-agglomerated, thermal treatment and pyrolysis
of the organic precursor to make submicron composite particles made
of carbon coated and carbon bonded nanoparticles.
EXAMPLE 2
[0143] A suspension in IPA of nanometer sized particles of LiFePO4
obtained after step 2 of Example 1 was dried at room temperature by
blowing dry air. The obtained LiFePO.sub.4 was the re-dispersed in
a water-lactose solution by using ultrasonic action. The ratio of
lactose to LiFePO.sub.4 was 10 wt. %. After drying, lactose coated
LiFePO.sub.4 particles are obtained.
[0144] Thermal treatment LiFePO.sub.4 and carbonization of the
lactose were performed in a rotary kiln as described in example 1.
SEM and TEM observation have revealed that the nanoparticles
obtained after thermal treatment are bigger when lactose is used as
the carbon precursor, even starting from to same wet milled
particle precursors.
EXAMPLE 3
[0145] The LiFePO.sub.4 was synthesized by a melt casting process
and then milled to nanometer sized using a beads mill as described
in example 1.
[0146] Poly(malic anhydride-1-alt-octadecene) was dissolved in IPA
and then mixed with the LiFePO.sub.4 in an IPA suspension at a
ratio of 5 wt %. After that, the solution was spray dried and
spherical aggregates are obtained.
[0147] The spray dried aggregates containing LiFePO.sub.4
nanoparticles and Poly(malic anhydride-1-alt-octadecene) organic
precursor were thermal treated in a rotary kiln as described in
example 1.
[0148] SEM analyses show that spherical micron sized
C--LiFePO.sub.4 aggregates of nano particles are obtained (see
FIGS. 7 & 8). A thin layer of conductive carbon is deposited on
the surface of the nanoparticles.
EXAMPLE 4
[0149] LiFePO.sub.4 was synthesized by solid state reaction as
described in WO-0227823. Stoichiometric amounts of
FePO.sub.4.2H.sub.2O and Li.sub.2CO.sub.3 were mixed first, then
added to an IPA solution premixed with Unithox 550 polymer
solution. The ratio of polymer to Fe and Li precursor is 4.7%. The
mixture was dried and then heat treated in a rotary kiln as
described in example 1 (4.sup.th step). After cooling, a
LiFePO.sub.4 with 1.7% carbon deposition (designated below by
C--LiFePO.sub.4) was obtained. Unithox 550.RTM. consists of a
hydrocarbon chain and a polyethylene oxide chain of a Molecular
weight of .about.1000 corresponding to formula
CH.sub.3--(CH.sub.2).sub.37--(CH.sub.2CH.sub.2O).sub.12--H.
[0150] C--LiFePO.sub.4 was dispersed in water at 20% of
concentration and then milled in a beads mill as described in
example 1. PSD analyses of the final milled products gives a D50 of
276 nm.
[0151] The wet milled product in water suspension was further mixed
with lactose at different ratio of Lactose to LiFePO.sub.4 and then
heat treated in a rotary kiln as 4.sup.th step of in example 1.
Different carbon contents was achieved by adding various amount of
lactose.
[0152] SEM analyses indicates that nanoparticle size is in the
range of 250-350 nm (see FIG. 9). This example shows that water can
be used as a liquid carrier in the beads milling process.
EXAMPLE 5
[0153] A sample of nanoparticles of C--LiFePO.sub.4 obtained after
step 4 of example 1 is nanoground in IPA in the presence of a
Triton X-100 surfactant agent. Proportions of the materials used
are: 200 g C--LiFePO4, 1 g TritonX100 and 2b830 ml IPA.
[0154] FIG. 10 illustrates the evolution of particle size vs time
after 30 and 90 minutes. These results are also summarized in Table
1 hereunder.
TABLE-US-00001 TABLE 1 Particle size (.mu.m) Time/min d10 d50 d90
Comments on PSD 30 0.2828 1.2108 4.250 Large envelope of 2 peaks:
large & smaller 60 0.2050 0.7330 5.155 1 peak and 2 small peaks
90 0.1604 0.3470 2.483 1 peak + small peak
[0155] After 30 min, a large flat envelope formed with a maximum
around 3 .mu.m (envelope and peaks at the left). At 60 min, the
peaks envelope moved towards a smaller particle size with maximum
between 1 .mu.m and 0.3 .mu.m. When the wet milling is extended to
90 min, a formation of well defined peak at 0.25 .mu.m with small
flat peak at 2 .mu.m. The d50 was dropped from to 0.347 .mu.m after
90 min in IPA. (envelope and peak at the right).
[0156] SEM and TEM examination are about the same as in example 1
confirming aggregates of nano sized crystals but with no strong
benefit from the surfactant in the case of this already synthetized
C--LiFePO.sub.4 product by opposition to the much stronger effect
observed with a material prepared from C--LiFePO.sub.4 precursors
in the presence of a surfactant.
EXAMPLE 6
[0157] In this example a copolymer is used as the carbon precursor
with a mixture of LiFePO.sub.4 precursors dispersed to form a
slurry. The precursor adsorption on the solid particles is used to
efficiently coat carbon on the surface of the LiFePO.sub.4 through
the adsorption process.
[0158] A stoichiometric mixture or FePO.sub.4, 2H.sub.2O and
Li.sub.2CO.sub.3 is dispersed, first mixed mechanically, and then
progressively mixed with a Unithox 550 copolymere solution in IPA.
Adsorption is confirmed by the fact that the remaining IPA solution
does not contain significant amount of copolymer after mixing the
slurry.
[0159] FIG. 11 a is a SEM picture that shows the product mixture in
which both FePO.sub.4, 2H.sub.2O particles and Li.sub.2CO.sub.3 can
be distinguished in SEM because of the mass difference of the
elements while the copolymer is not visible. On FIG. 11a, the
carbonate and the phosphate are visible, but not the copolymer. On
FIG. 11b, not only the carbonate and the phosphate are visible, but
also the polymere with which the bismuth is complexed. A polymer
film covers both the carbonate and the phosphate. FIG. 11d shows
LiFePO.sub.4, carbon spots and carbon bridges resulting form
pyrolysis of the organic carbon precursor that can be seen in low
voltage SEM.
[0160] To reveal the polymer localization, a similar mixture has
been dried and treated with a soluble bismuth sulfonate salt in
IPA. In these conditions, the bismuth cation is complexed by the
POE segment of the copolymer and insolubilized. Dried bismuth
treated mixture is observed in SEM on FIG. 11 b). At that point all
particles are coated by the adsorbed copolymer and cannot be
distinguished. Chemical analysis shows in FIG. 11 c) that Bi is
present on all particles surface revealing the thin adsorbed
copolymer coverage of the mixed particles. This shows that the
adsorption is very strong since a solution of Unithox in IPA in
which FePO.sub.4 2H.sub.2O powder is added will adsorb most of the
copolymer and let only IPA solvent after one hour. Same phenomenon
is observed in the same condition using LiFePO.sub.4 powder also in
IPA.
[0161] It is interesting to note in view of these observations,
made on micron-sized particles for ease of observation, that upon
heat treatment, at 700.degree. C. of the mixture under inert or
reducing atmosphere, the C-coated LiFePO.sub.4 (1.8% wt C) is
formed with a high yield, >90%. Visualisation in SEM of the
carbon coating is illustrated in FIG. 12 d) at a higher
magnification.
[0162] In a similar experiment, a stoichiometric mixture of
FePO.sub.4.2H.sub.2O and Li.sub.2CO.sub.3 is nanoground in a bead
mill with IPA as carrier solvent in presence of Unithox 550.RTM.
copolymer. Adsorption has been confirmed by the fact that the
remaining IPA solution does not contain a significant amount of
copolymer after nanogrinding the slurry. After Bismuth treatment,
chemical analysis have shown that Bi is present on all particles
surface revealing the thin adsorbed copolymer coverage of the mixed
particles. This shows that the adsorption is very strong since a
solution of Unithox 550.RTM. in IPA in which FePO.sub.4.2H.sub.2O
powder is nanogrind will adsorb most of the copolymer and let only
IPA solvent after one hour.
EXAMPLE 7
[0163] This example illustrates the method of the invention,
starting from C--LiMPO.sub.4 precursors.
[0164] Samples were produced, starting from 200 g of a
FePO.sub.4,2H.sub.2O+Li.sub.2CO.sub.3 stoichiometric mixture in
2830 ml of IPA. The particle sizes of the starting precursors are
as follows
TABLE-US-00002 Initial particle size (.mu.m) d10 d50 d90 FePO4:
0.878 2.9 5.5 Li2CO3 1.74 3.64 6.16
[0165] The particle size of the final product is studied as a
function of grinding time.
[0166] In the first case, the reaction mixture only contains IPA
and the precursors. In the second case the reaction mixture
contains additionally Triton 100X, (0.5% wt to solids). Triton is a
surfactant which contains only C, H and O elements and is used to
help make nanoparticles or small agglomerates. In the third case,
the reaction mixture contains additionally a Unithox 550 copolymer
(4.7% wt to solid).
[0167] Results are summarized in the following tables showing that
after 90 minutes, the Triton X100 surfactant has a favourable
effect on grinding and PSD with a nearly monomodal D50 of 200 nm
versus a remaining bimodal PSD when no additive is used.
[0168] Furthermore the addition of a Unithox copolymer leads to a
well defined monomodal PSD and a D50 of about 180 nm. Also a better
fluidity of the circulating slurry with Triton is observed. The
fluidity is much better when both Triton and Unithox are used.
[0169] Unithox 550 polymer as a carbon precursor was added to the
precursor milled with 0.5% Triton surfactant in the ratio of 4.7%
polymer over ion phosphate and lithium carbonate precursor. The
precursor with polymer additive was dried slowly while stirring and
then heat treated in a rotary kiln as described in example 1 under
argon.
[0170] X-ray analyse shows that the final product is LiFePO.sub.4
with about 3-4% of Fe.sub.2P.sub.2O.sub.7 impurity phase (see FIG.
13). The crystallite size calculated from peak width is 300 nm and
the microstrain is 0.68%.
[0171] FIGS. 14 and 15 are SEM images of C--FePO.sub.4 particles
(200-500 nm) as partially spherical aggregates obtained after
nanogrinding a Li.sub.2CO.sub.3+FePO.sub.4 2H.sub.2O mixture with
Triton as a surfactant (FIG. 14) and with surfactant (FIG. 15) and
pyrolysis at 700.degree. C. in the presence of Unithox and IPA.
[0172] SEM observation shows aggregates made of from nanoparticles
size is in the range of 200-500 nm A majority of the naoparticles
have particle size in the range of 200-300 nm. Carbon deposition
can also be observed. LECO measurement shows that the carbon
content is 1.87 wt. %. This is slightly higher than the carbon
content of 1.6-1.7 wt. % of carbon when no Triton is added during
processing. It is anticipated that the Triton surfactant is
absorbed on the particle surface and carbonized in the thermal
treatment. But Triton alone does not yield much carbon.
[0173] Wet gringing all precursors together can also solve the
mixing problem for fine particles. As is known, it is very
difficult to mix submicron sized precursor particles together by
conventional mixing method. Wet milling could achieve fine particle
reactants and homogeneous mixing of the reactants, and then
C--LiFePO.sub.4 product with higher purity is expected after
synthesis as demonstrated in the present example.
[0174] Overall, surfactant and copolymer addition during the wet
grinding have improved processability and lead to fine and
monodispersed particles, probably more than C--LiFePO.sub.4 wet
grinding. Polymer adsorption shown in Example 6 might have
contributed to processability in addition to help to the carbon
deposition process.
FePO.sub.4+Li.sub.2CO.sub.3
TABLE-US-00003 [0175] Particle size (.mu.m) Time/min d10 d50 d90
Comments on PSD 30 0.1315 0.2182 4.119 2 peaks: large & smaller
60 0.1451 0.3334 18.23 1 peak and 2 small peaks 90 0.1186 0.1727
0.4044 1 peak + small peak
FePO.sub.4+Li.sub.2CO.sub.3+0.5% Triton X-100
TABLE-US-00004 [0176] Particle size (.mu.m) Time/min d10 d50 d90
Remark 30 0.1916 0.8539 5.1675 2 equal peaks 60 0.141 0.2479 1.6525
1 peak and 1 small peak 90 0.1279 0.1987 0.4447 1 peak and 1 very
small
FePO.sub.4+Li.sub.2CO.sub.3+0.5% Triton X-100+Unithox 550
TABLE-US-00005 [0177] Particle size (.mu.m) Time/min d10 d50 d90
Remark 30 0.1754 0.8512 6.1164 2 equal peaks 60 0.1410 0.2740
20.1855 1 peak and 2 small peaks 90 0.1211 0.1771 0.3064 1 peak
* * * * *
References