U.S. patent application number 12/923623 was filed with the patent office on 2011-03-31 for mixture of basic lead sulfates.
Invention is credited to George E. Mayer.
Application Number | 20110076562 12/923623 |
Document ID | / |
Family ID | 43780747 |
Filed Date | 2011-03-31 |
United States Patent
Application |
20110076562 |
Kind Code |
A1 |
Mayer; George E. |
March 31, 2011 |
Mixture of basic lead sulfates
Abstract
A microporous lead-containing solid material is produced, which
can serve as a carrier for desired materials into a reaction for
various desired purposes. For example, if the microporous solid is
impregnated with borax it tends to inhibit the growth of unduly
large crystals of tetrabasic lead, which is useful in producing
batteries having improved functional qualities.
Inventors: |
Mayer; George E.; (Venetia,
PA) |
Family ID: |
43780747 |
Appl. No.: |
12/923623 |
Filed: |
September 29, 2010 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61272490 |
Sep 29, 2009 |
|
|
|
Current U.S.
Class: |
429/227 ;
252/182.1; 427/126.3 |
Current CPC
Class: |
Y02E 60/10 20130101;
H01M 4/362 20130101; C01P 2006/40 20130101; H01M 4/5825 20130101;
C01P 2006/10 20130101; H01M 4/20 20130101; C01P 2006/14 20130101;
C01G 21/20 20130101 |
Class at
Publication: |
429/227 ;
252/182.1; 427/126.3 |
International
Class: |
H01M 4/58 20100101
H01M004/58; H01M 4/86 20060101 H01M004/86; B05D 5/12 20060101
B05D005/12 |
Claims
1. A composition of matter comprising lead sulfates which has high
internal porosity and density <3.5 g/cc.
2. The composition of matter of claim 1 further comprising
predominantly tetrabasic lead sulfate and tribasic lead
sulfate.
3. The combination of matter of claim 1, wherein the pores of the
material comprising basic lead sulfates are impregnated with a
material capable of altering the pattern of crystal growth of other
substances.
4. The combination of matter of claim 3, wherein the material
capable of altering the pattern of crystal growth of other
substances is selected from the group consisting of borax or
another combination of an alkaline anion with a suitable cation,
selected from the group consisting of borates, phosphates, acid
phosphates, silicates, and fluoborates.
5. A method of manufacture of a composition of matter comprising
basic lead sulfates which has high internal porosity and density
<3.5 g/cc, said method comprising the steps of mixing leady lead
oxide, water and sulfuric acid into a doughy mass, said mixing step
being performed very rapidly and at elevated temperature, followed
by grinding, such that the doughy mass is reduced to a fine powder
of lead salts.
6. The method of claim 5, where the ratio of oxide to water to
sulfuric acid is 100 parts by weight of oxide, to from 10 to 30
parts water, to from 10 to 30 parts of 50 wt % sulfuric acid (1.400
sp.gr.), preferably 100 parts by weight of oxide to 10 parts by
weight of water to 15 parts by weight of acid.
7. The method of claim 5, where the water is heated to
>70.degree. C. (preferably >90.degree. C.) and mixed with the
oxide, this mixture then being mixed with the acid, which is added
rapidly (<5 minutes), preferably as a fine spray in <2
minutes.
8. The method of claim 5, wherein the powder is ground until it
achieves a degree of fineness such that >99% of the particles
thus produced will pass through a 325 mesh sieve.
9. The method of claim 8, comprising the further step of collecting
the smaller or lighter particles (<325 mesh) in a wet
scrubber.
10. The method of claim 8, comprising the further step of
collecting the smaller or lighter particles (<325 mesh) in a
baghouse.
11. The method of claim 8, comprising the further step of
collecting the smaller or lighter particles (<325 mesh) in a
dropout system.
12. The method of claim 8, comprising the further step of
separating the heavier, coarser grinder throughput from the finer
particles by an air classifier, and collecting any additional finer
particles.
13. The method of claim 5, comprising the further step of mixing
the fine powder with a modifying additive that will modify the
growth of tetrabasic lead sulfate and/or promote the growth of
other small lead salt crystals, where the low-density porous lead
salt particles act as a carrier for the modifying additive.
14. The method of claim 13, wherein the modifying additive is
provided in an aqueous solution, and comprising the further step of
suspending the fine microporous powder of lead salts in the aqueous
solution, boiling the solution to expel air from the pores of the
fine powder, and wherein heating is continued until the water is
driven off and the modifying additive is dispersed substantially
within each microporous grain.
15. The method of claim 13, wherein the modifying additive is
selected from the group comprising borax and other combinations of
an alkaline anion with a suitable cation such as borates,
phosphates, acid phosphates, silicates or fluoborates.
16. A method of making a battery paste, comprising the steps of
mixing a conventional battery paste comprising leady oxides, water
and sulfuric acid with a nucleant material for tetrabasic lead
sulfate and with the fine porous powder and modifying additive of
claim 13.
17. The method of claim 16, wherein the nucleant material is mixed
intimately with the oxide at the start of paste mixing and the fine
porous powder and modifying additive is added near the end of paste
mixing, after acid addition is complete.
18. The method of claim 16, wherein the nucleant is added after
mixing leady oxide with water, before acid addition and mixture,
and the fine porous powder which contains the modifying additive is
added at the end of acid addition and before final mixing.
19. A method of making a battery plate, comprising the steps of
making the battery paste of claim 16, applying the paste to battery
plate grids and curing the paste at >70.degree. C., such that
tetrabasic lead sulfate crystals are grown that are more numerous
and/or smaller than would be grown without the fine porous powder
and modifying additive.
20. A method of making a battery paste, comprising the steps of
making the battery paste of claim 16, applying the paste to battery
plate grids and curing the plates at <70.degree. C. such that no
or limited secondary nucleation of tetrabasic lead sulfate
occurs.
21. A lead-acid battery, which contains positive and/or negative
plates made according to the method of claim 19.
22. A lead-acid battery, which contains positive and/or negative
plates made according to the method of claim 20.
23. A composition of matter comprising a microporous inorganic
solid mixed with a modifying additive that, in a process for
forming lead-containing compounds, will modify the growth of
tetrabasic lead sulfate and/or promote the growth of other small
lead salt crystals, where the microporous solid acts as a carrier
for the modifying additive.
Description
CROSS-REFERENCE TO RELATED APPLICATION
[0001] This application claims priority from provisional
application Ser. No. 61/272,490, filed Sep. 29, 2009.
FIELD OF THE INVENTION
[0002] The present invention relates to a novel, low density, high
porosity chemical mixture of, primarily, basic lead sulfate salts
which is useful as a carrier to convey other materials into a
chemical process, to processes for its use, and to products made
thereby. For example, the material of the invention can be used as
a carrier for other materials which act to inhibit excessive
crystal growth in lead-acid battery paste, which yields improved
battery performance.
BACKGROUND OF THE INVENTION
[0003] Significant background material on the manufacture of
lead-acid batteries can be found in the inventor's prior U.S.
patent application Ser. No. 11/234,077 and corresponding PCT
application PCT/US2005/034214, published as WO 2006/034466 and
incorporated herein by this reference. Some of this material is
reproduced below.
[0004] An important and time-consuming aspect of manufacture of
lead-acid batteries is the curing of wet active paste material
precursor into a dry porous mass. The paste precursor typically is
in the form of flakes of "leady oxide", i.e. flakes of solidified
lead particles which have a PbO coating. The leady oxide is made
into a wet, pliable dough ("paste") by mixing it with water and
then with sulfuric acid. The dough then is extruded onto
mechanically rigid, electrically conductive grids in a process
called "pasting". The resulting pasted grids are cured at elevated
temperature and humidity to react PbO with sulphuric acid, to form
lead sulfate salts, and to oxidize the lead core of the leady oxide
to PbO to form additional lead sulfate salts.
[0005] Lead sulfate salts which provide mechanical strength and
porosity to the leady oxide paste, and ultimately become part of
the active material, include tribasic lead sulfate
3PbOPbSO.sub.4H.sub.2O ("3BS") and tetrabasic lead sulfate
4PbOPbSO.sub.4 ("4BS").
[0006] The 3BS typically forms at low temperature and low humidity,
whereas 4BS typically forms at higher temperatures (>70.degree.
C.) and higher humidity. The 3BS typically forms as small
needle-like crystals which measure about 3 microns long and less
than about 1 micron in each of width and thickness. The 4BS
crystals are larger, and grow in length from several microns to
several hundred microns. The longer 4BS crystals have width and
thickness in proportion to length. For example, a 300 micron long
4BS crystal might have a width of 60 microns and a thickness of 50
microns. A 4BS crystal that measures 300 microns long by 60 microns
wide and 50 microns thick has a surface area of 72000 square
microns, and a volume of 900,000 cubic microns. This volume, if
tightly packed with the smaller 3BS crystals, would hold about
10.sup.7 3BS crystals which have a total surface area of about
7.2.times.10.sup.6 square microns, i.e. 1000 times greater surface
area. The size and shape of the crystals in the cured paste can be
measured by scanning electron microscopy (SEM). The amounts of the
3BS and 4BS crystals may be determined by x-ray diffraction
(XRD).
[0007] In general, the useful capacity of a battery, particularly
its ability to deliver high current for short periods of time, as
required for example in the starting of internal combustion
engines, is improved by increase in the surface area of the lead
compounds in the cured paste. Accordingly, large crystals of 4BS
are less desirable than smaller crystals, although the presence of
4BS is itself desirable.
[0008] The production of 4BS generally requires very careful
control of temperature and humidity during cure of the pasted
plates. Premature dryout and/or cooling of the plates during the
curing process inhibit the formation of 4BS. Some battery
manufacturers specify control and uniformity to .+-.2.degree. C.
and .+-.1% relative humidity (RH) compared to setpoints during
cure. It has generally been observed that addition of red lead
(Pb.sub.3O.sub.4) allows adequate processing flexibility to allow
production of 4BS over a larger range of temperature and relative
humidity relative to setpoints; however, in controlled experiments,
the red lead addition appears to interfere with free Pb
oxidation.
[0009] Production of 4BS entails nucleation of crystals and their
subsequent growth. Nucleation is effected by exposing the pasted
plates to temperatures of about 70.degree. C. or higher at high
humidity at the beginning of cure. Nucleation and crystal growth
during cure can require an induction period of about 10 hours since
4BS forms as molecules, which slowly coalesce by diffusion into
seeds. These seeds can react with additional nearby material to
grow into crystals. The growth rate of the 4BS crystals depends on
various factors such as the composition of the lead oxide used, the
oxide to sulphuric acid ratio in the paste mixture, the mixer type,
mixing time, mixing temperature, temperatures between process
steps, flash drying conditions, as well as the temperature and
humidity inside the curing chamber, as mentioned above.
[0010] The growth of 4BS can proceed by two mechanisms. Large
isotropic and "regular", i.e. uniaxial, crystals, can be prepared
by preferential deposition of a material onto one face of a seed
crystal by a screw-dislocation or a slip-plane mechanism. Since
only one face of each crystal grows, the process is slow. Crystals
grown by this mechanism have smooth crystal faces and sharp angles
between adjacent faces. Alternatively, crystals that grow
anisotropically may be produced faster by fractal growth. Fractal
growth entails growing the crystals at many locations and in many
different directions simultaneously, i.e. multiaxially. The
resulting fractal crystals are irregular and smaller in size.
Fractal crystal growth may be confirmed by plotting "quantity
produced" vs. time. Regular growth provides a straight-line linear
plot. Fractal growth provides a straight-line log-log plot. Fractal
crystal growth may produce greater mechanical strength in a paste
pellet because the multiaxial crystals interlock better than the
uniaxial crystals do. Fractal growth also may produce better
electrical conductivity when it occurs in the paste.
[0011] Some battery manufacturers prefer 3BS rather than 4BS for
engine cranking (also known as SLI, for "starting, lighting, and
ignition") batteries, which may be flooded, gelled, or absorptive
glass mat (AGM) in design. To some degree this is because
traditional curing processes tend to create highly variable yields
of 4BS which have a large crystal size and a small amount of very
large pores, and hence low surface area per unit weight. This
variability in yield and the undesired crystallinity and porosity
tends to cause variable (and generally poor) battery cranking
performance in SLI batteries. Accordingly, many manufacturers
prefer to avoid formation of 4BS insofar as possible.
[0012] When curing conditions are adjusted to preclude nucleation
and growth of 4BS, a predominance of 3BS is produced. The 3BS has
uniform crystal shape and size (3 microns X approximately 0.5
microns X approximately 0.5 microns). When a plate is pasted with
3BS the plate has uniform porosity and high cranking
performance.
[0013] Historically, free Pb was a desired component of battery
paste. The free Pb was thought to generate heat during cure of the
pasted battery plates to enhance production of 3BS, 4BS and
porosity. This heating, however, was uncontrolled and erratic, and
the resultant plates did not always have the composition and/or
porosity desired. Free Pb is now considered undesirable. A high
amount (more than about 2 wt. %) of free Pb at the end of curing
can lead to shedding and spalling failure of the positive plates
and/or high self-discharge of "formed" PbO.sub.2 plates.
[0014] The amount of free Pb in leady oxides typically is about 25
wt. %, but can form in amounts of 20 to 40 wt. % free Pb depending
on the apparatus and the process settings. It is difficult and
costly to produce a leady oxide with about 15 wt. % or less free
Pb, and even more costly to produce a nonleady oxide. This latter
usually requires a subsequent thermal processing, in small batches.
Discharge capacity of a battery depends on the porosity and surface
area of the porous battery electrode. Both the positive electrode,
which for a lead-acid battery is the lead dioxide electrode, and
the negative electrode, which for the lead-acid battery is the
sponge Pb electrode, need porosity. The porosity of negative
plates, during battery use, is improved by the well-known use of
"expander" additives, which are comprised of barium sulfate, carbon
black and lignosulfonic acid salts. All lead-acid electrodes which
have a larger surface area have a higher discharge capacity, and
higher utilization of the active material at any rate of discharge.
In high discharge rate batteries such as SLI batteries, 3BS has
been the preferred active material precursor, but if the
undesirable growth of large crystals of 4BS can be inhibited, as
provided by the invention, 4BS would also be desired in SLI
applications. 4BS is the preferred material precursor for deep
cycle and long-life stationary batteries. 4BS is also now the
preferred precursor for use in modern nonantimonial grid batteries,
the so-called "maintenance-free" batteries for SLI, float or
cycling application, because the 4BS helps prevent PCL (premature
capacity loss), i.e. short battery life. According to the present
invention, the best features of 3BS and 4BS can be obtained
simultaneously, without any apparent shortcomings, as described
below.
[0015] Curing promotes adhesion of battery paste to the grids. The
battery paste, which has an alkaline pH, reacts with (corrodes)
lead alloy in the grid to partially convert the lead alloy to Pb
compounds and ultimately to 3BS and 4BS. Generally, the higher the
temperature employed during cure, the better the adhesive bond
produced.
[0016] As mentioned above, production of 4BS depends on nucleation
and growth of 4BS crystals. One way to get 4BS nuclei immediately
into a battery paste is to use 4BS seed crystals, such as those
prepared by grinding large crystals of pure 4BS. Large 4BS crystals
can be made by any of several well-known aqueous slurry processes.
These processes, however, are slow, and yield only a small amount
of 4BS in copious amounts of liquid. Accordingly, this is very
costly. Another way to produce 4BS is to use an Eirich mixer
wherein 4BS is made into a more concentrated slurry, and then
removing excess water by vacuum and heat. A pyrometallurgical
reactor (Barton pot) also may be used to make 4BS. A slurry reactor
and reactive grinding may be also be used to make 4BS. These
methods, however, do not produce multiaxial crystals of 4BS, or
seed crystals which can grow multiaxial crystals in battery plate
pastes.
[0017] Ser. No. 11/234,077 discloses a paste curing additive
("PCA") for battery paste for use in, for example, lead acid
battery positive plates, and its methods of manufacture and use.
The PCA limits production of larger 4BS crystals by nucleating
growth of numerous 4BS crystals, so that more and therefore smaller
4BS crystals are present in the final product, as well as to grow
the 4BS in multiaxial crystal groupings. PCA, which itself contains
little 4BS, can be used to make more 4BS, and may be used to reduce
the cure time of active material paste, as well as to reduce the
amount of energy required during curing.
[0018] PCA also may be used to enhance production of 3BS during
mixing and cure of battery paste. The PCA may be used to enhance
curing of pasted battery plates, especially pasted battery plates
intended for SLI lead-acid batteries.
[0019] PCA also may be used to achieve greater porosity in the form
of higher numbers of pores as well as larger sizes of pores in the
cured plate. PCA also may be used to speed oxidation of free lead
residue in pasted plates during cure. PCA also may be used to
enhance adhesion of the cured paste to the grid. These may provide
greater utilization of the active material and easier conversion
from the non-active "paste" state to the "active material"
state.
[0020] PCA in amounts of about 1 wt. % to about 12 wt. % based on
the weight of leady oxide may be used to speed the cure of battery
plates at temperatures of about 56.degree. C. to about 100.degree.
C. at RH of about 10% to about 100%.
[0021] Lead acid battery plates which include PCA may also cure
faster and may show improved performance. In side-by-side tests for
lead-acid traction battery positive plates, PCA outperformed a
commercial ground 4BS crystal seed material: <2% free Pb was
achieved in <20 hr with PCA, in 24 hr with the competitor
material and in >40 hr with no additive (control). When these
cured plates were tested in a cycling (charge/discharge) regime,
controls operated <1000 cycles, the competitor material operated
approximately 1500 cycles and PCA operated to >2500 cycles to
end of life. The PCA cells were removed from test to allow testing
of other cells; from the data trend it appears that PCA should have
operated to >3000 cycles, which is twice the industry standard
requirement of 1500 cycles.
[0022] The use of PCA may improve development of crystals of lead
sulfate such as 3BS and 4BS, and may enhance more rapid development
of porosity and the oxidation of free lead.
[0023] In one exemplary process, PCA is produced as the reaction
product formed by heating a battery paste to a temperature of about
80.degree. C. to about 90.degree. C. for about 5 min. to about 10
min., wherein the battery paste includes sulfuric acid in an amount
of about 5 wt. % to about 6 wt. %, water in an amount of about 12
wt. % to about 16 wt. %., and balance leady oxide, all amounts
based on total weight of sulfuric acid, water and leady oxide. The
additive may be then be used in any of its dried or undried
states.
[0024] In a second exemplary process, the PCA is produced as the
reaction product formed by heating a battery paste to a temperature
of about 70.degree. C. to about 90.degree. C. for about 10 min. to
about 90 min., wherein the battery paste includes sulfuric acid in
an amount of about 3 wt. % to about 10 wt. %, water in an amount of
about 10 wt. % to about 20 wt. %., and balance leady oxide, all
amounts based on total weight of sulfuric acid, water and leady
oxide.
[0025] As mentioned above, starting-lighting-ignition (SLI)
lead-acid batteries have as their major mission to provide high
current to the starter motor, which crank the internal-combustion
engines of motor vehicles until they start. There is a market need
to improve cranking ability while simultaneously reducing size,
weight and cost. Cranking requires a high current (hundreds of
amperes) battery discharge over a short period (<1 min.) of
time. This requires that the active materials of the battery have a
large and intimate contact area: the larger the surface area, the
greater the cranking performance, all other things being equal.
[0026] Lead-acid batteries contain three electrochemically active
ingredients, two being sets of solid members forming numerous
opposed pairs of oppositely polarized "plates" and the third being
the sulfuric acid electrolyte, which is disposed within and between
the plates. The plates are connected together in two groups; the
positive polarity plates contain PbO.sub.2 and the negative
polarity plates contain sponge Pb. As mentioned above, and as
further detailed below, plates typically comprise conductive lead
alloy frames called "grids" which at manufacture are filled with a
"paste" of the approximate consistency of a plastic dough mixed
from leady oxide (i.e., flakes of solidified lead particles with a
PbO coating), water and sulfuric acid, cured, and then
electrochemically "formed" or charged. Pastes intended for positive
and negative plates are generally similar, except that pastes for
negative plates contain other ingredients called "expanders". One
assembly of two sets of opposite-polarity plates, along with
electrically insulating but porous separators, the electrolyte, and
the containment for all of these comprise a single cell. Two or
more (usually 3 or 6) series-connected cells comprise a
battery.
[0027] The surface of each plate, both positive and negative,
formed by pasting as described above, consists of two intermeshed
parts: a discontinuous solid phase formed of the various lead
compounds described above, interspersed with a discontinuous void
space, which will ultimately be filled with electrolyte. In order
to provide adequate surface area, the solid phase should consist of
moderately large crystals with a "fuzzy" surface, or more numerous
smaller crystals. The latter approach has traditionally been used
by battery manufacturers, where the small crystals (approximately 3
microns long) are tribasic lead sulfate ("3BS") the chemical
composition of which is 3PbOPbSO.sub.4 or, if hydrated,
3PbOPbSO.sub.4H.sub.2O. 3BS can be produced during the mixing of
the battery paste or during the subsequent recrystallization or
"curing" step, typically carried out at <70.degree. C. During
mixing and curing, different basic lead sulfate salts can be
produced under different conditions. These range from non-basic
lead sulfate (0BS, PbSO.sub.4) through 1BS (PbOPbSO.sub.4) and 2BS
(2PbOPbSO.sub.4) to 3BS as above and ultimately to 4BS
(4PbOPbSO.sub.4). 0BS, 1BS and 2BS are generally undesirable in
cured battery plates. 4BS, which is nucleated at >70.degree. C.
but which can subsequently grow after nucleation at any temperature
from nearly 0 C to 100.degree. C. (below this range H.sub.2O
freezes and above this range H.sub.2O boils, of course), is
desirable for use in long-life deep-cycling applications such as
fork lift truck batteries, and may be useful to some degree in SLI
battery applications as well, if the 4BS crystals can be inhibited
from growing overly large. Thus, there is a desire to provide as
much 4BS in the battery paste as possible, especially for batteries
intended for deep-cycle applications, if certain problems inherent
in the use of 4BS as mentioned can be overcome.
[0028] More specifically, and as also noted above, while 3BS
crystals stop growing at 3 microns, 4BS crystals can grow to become
several hundred microns in length. This size is mechanically and
electrically desirable, but as these large crystals provide
comparatively little surface area, they are electrochemically
undesirable. Accordingly, smaller 4BS crystals are desirable to
maximize the ratio of surface area to weight.
[0029] Generally, one 4BS crystal grows from a single seed, the
seed consisting of ground, macro-crystalline 4BS, as is now
commercially available. To grow small 4BS crystals, one approach
(for example in U.S. Pat. Nos. 7,118,830 and 7,517,370) is to use
very finely ground (0.1 to 5 micron) 4BS seeds, but this finely
ground material is difficult to handle and provides a significant
dust hazard which is undesirable. Another approach is to use a
larger quantity of seed (20% rather than the usual 1 to 2%), but
seed is costly and larger amounts appear to be preferentially
consumed before leady lead oxide in the paste mixing and/or curing
processes.
[0030] Higher growth temperatures for the growth of 4BS, performed
with or without 4BS seeds, impart secondary nucleation of 4BS
(desirable) and also generally give better grid-paste adhesion
(also desirable) by way of enhanced grid corrosion by the alkaline
battery paste. Thus, curing the paste at higher temperatures will
generally result in the formation of more 4BS, but manufacturers
may not be able or willing to cure the paste at such elevated
temperatures.
SUMMARY OF THE INVENTION
[0031] The approach taken by the present invention is to provide
seed of relatively large size, e.g. 20-40 micron, or <325 mesh,
to nucleate the growth of 4BS crystal, so as to avoid the problems
inherent in use of extremely fine powders, but to take precautions
to interfere with the growth of the 4BS crystals, so as to limit
the size of these crystals. Even though the seed particles were 20
to 40 microns in size, in the course of being mixed into paste
and/or cured, the particles disintegrated and the resultant 4BS was
acicular and generally under 40 microns in length, even without
other precautions to limit the size. As mentioned, finely powdered
4BS seeds, such as are described in U.S. Pat. Nos. 7,118,830 and
7,517,370, are costly to make and difficult to handle.
[0032] In order to prevent or limit growth of over-large crystals
of 4BS, one could drastically change the curing conditions, but
this would preferentially affect the outsides of each plate, and,
worse, perhaps affect only the outside plates of a stack of pasted
plates stacked up for curing in an oven.
[0033] The present invention recognizes that as crystal growth
generally proceeds in the direction of a single preferred
crystallographic plane of a seed, one possibility for limiting
crystal growth might be to add materials to the paste that would
tend to modify the "crystal habit", i.e. activate or deactivate
growth of crystals along various planes.
[0034] The literature, for example Perry's Chemical Engineer's
Handbook, 7th edition (1997) at Table 18-4 contains several
examples of the use of borax to modify the crystallization of
soluble alkaline earth and transition metal sulfates such as
MgSO.sub.4 and ZnSO.sub.4. Borax in a quantity of 5% is reportedly
used to aid growth of crystals of these salts, although the
original papers show no insight as to how this is accomplished. As
materials such as MgSO.sub.4 and ZnSO.sub.4 are very soluble in
water whereas basic lead sulfates are nearly insoluble in water,
such that the chemical behavior of the respective substances are
generally nonanalogous, there is no reason to expect that borax
would be helpful in controlling the growth of 4BS crystals. Neither
borate ion nor sodium is deleterious to battery operation, but
unfortunately even as little as 0.1% borax can make battery paste
soupy and unprocessable. Other salts have various other problems.
For solubility and minimal materials costs, sodium salts are
preferred to Li, K, Sr, etc. Some other sodium salts could be used,
such as acidic phosphates, silicates, fluoborates, etc. Soluble
peroxysalts (perborate, persulfate, percarbonate) have been used in
batteries, but here the salt additive is intended to provide the
peroxy oxygen to oxidize some Pb(O) to Pb.sub.2O.sub.3,
Pb.sub.3O.sub.4 or PbO.sub.2. Even nonalkaline salts such as boric
acid might be useful.
[0035] Assuming that borax or perhaps some other material would be
useful in limiting 4BS crystal growth, the next concern is to find
a way to convey such a crystal growth-modifying ingredient into a
workable paste mix. Any organic carrier can degrade ultimately to
acetic acid, which undesirably corrodes the positive grids. Any
inorganic carrier would need to be extremely porous, such that the
additive, e.g. borax, would slowly dissolve and diffuse from the
carrier to the 4BS growth plane, would also need to be extremely
inexpensive, and would also need to be inert so as not to interfere
with the battery operating conditions. This precludes use of most
materials except titania, that is, TiO.sub.2, and lead compounds.
Titania, used widely as a white paint pigment, is much cheaper than
various nonstoichiometric titanium oxides, which have rough
surfaces, but has no significant internal porosity. Titania, if
used, would remain inert within each plate, that is, would not
participate in charge or discharge, and would thus add undesired
weight. Therefore, no titanium-oxygen material is particularly
suitable as a carrier.
[0036] Considering the possibility of making a porous lead compound
to contain the borax or other crystal growth inhibitor, it would be
difficult to create porosity by adding water, air, or a
gas-producing salt, as most lead compounds are made by a pyrolytic
process at temperatures above the melting point of Pb
(327.46.degree. C.) and the heat would drive off or decompose the
additive. Soluble anions such as chloride, acetate, perchlorate, or
nitrate are deleterious to battery operation because they enhance
the undesirable grid corrosion if a lead-based carrier were to be
attempted using a hydro (low temperature aqueous solution) process.
SEM microscopy shows little if any porosity in 3BS, 4BS or any of
the other basic lead sulfates made by any of the solution/slurry
methods. It would seem that a highly porous lead-based carrier
would be impossible.
[0037] The simplest selection criterion for high porosity of a lead
material is low material density. Although lead and its compounds
generally have densities >9 g/cc, the inventor has discovered a
lead compound with density approximately one third of this and a
corresponding increase in porosity. SEM micrographs show that each
grain of this novel lead mixture (described in detail below, and
hereinafter called "Fluff") is microporous, and that the low
density is not merely imparted by a rough surface. When the Fluff
was mixed with a borax solution, the borax was absorbed in the
pores of the Fluff, as shown by the unchanged golden color and by
an increase in material density. Summarizing the process of
producing a borax-containing Fluff, the process being further
detailed below: in order to introduce the borax into grains of the
porous Fluff lead material, the most soluble borax form, borax
decahydrate, was dissolved in water, mixed with the Fluff, and
dried, leaving borax within the pores of the Fluff, probably in the
form of a lower, less soluble hydrate, e.g., borax5H.sub.2O. The
final product, in a batch consisting of 3 kg of Fluff lead material
having been impregnated with 5 kg of borax decahydrate solid
material, fits within a one gallon can after drying.
[0038] The improvements and advantages provided by using such a
combination of borax with the novel Fluff microporous lead sulfate
mixture carrier (the combination hereinafter being referred to as
"BF") include that it has no effect on the processing
characteristics of battery paste containing this material, and that
its use, like PCA, limits the size of the 4BS crystals formed to 40
to 60 microns, presumably by causing more crystals to form and
thereby reduce their size. Other crystals of large surface area and
fine crystal structure are also formed, as shown by SEM, also
providing improved high rate battery performance. XRD analysis
shows the crystal mixture consists of 4BS and some various lead
borate salts. The greater alkalinity of paste containing a BF
combination due to the presence of the sodium ion also improved
grid-paste adhesion, as shown by pellet-puncture mechanical tests.
Initial tests of experimental plates made with a paste containing
4BS nucleant seeds plus BF showed a 10% improvement in moderately
fast discharges (20 minute, the "reserve capacity") versus controls
which did not include BF. The inventor extrapolated, based on the
well-known Peukert law, that cranking performance should have been
improved by at least 10%, but because of the design of the battery
test cells, this was not tested in the intial experiments.
[0039] Subsequently, the use of BF in battery paste together with a
4BS nucleant was tested at a SLI battery factory. These tests
revealed minimal effect on paste mixing (slight initial or final
water addition), better/faster curing, and better crystallinity and
porosity of the cured plates (as seen in SEM examination). The
latter manifested as faster formation (i.e., initial battery
charging), less gassing, less water loss and a 12% increase in
cranking performance. A preferred process for using PCA and BF was
developed, but the first and three subsequent companies where tests
were performed were unwilling or unable to fully comply with the
inventor's recommendations. Nonetheless, each of 8 factory tests to
date have shown not only improved high-rate battery performance,
but also the necessary development of fine-particle
high-surface-area 4BS that would explain such improved performance.
Data from these tests is summarized in Table 1, below.
[0040] In all the tests described in Table 1, a combination of PCA
and BF gave improved high-rate battery performance: at companies 1,
2, and 3 this was shown as a >12% increase in CCA despite
processing below the recommended values (especially for curing and
drying, paste mix peak temperature and oxide quality). Company 4
found higher capacity using PCA and BF in golf car batteries. These
have an unpredictable current draw and duty cycle, so that capacity
cannot be measured easily, but these approximate a 75 amp discharge
and a 90-100 minute discharge time. It was observed that the PCA/BF
batteries need recharge only every 2 to 3 days compared with 1 to 2
days for controls.
[0041] The ability of the Fluff to convey a material that might
otherwise be deleterious into a process so as to be gradually
diffused or dispersed may be useful in other applications. For
example, the Fluff might be useful to convey colorants into glasses
and ceramics. The Fluff might also be useful to convey catalysts
for fuel cells or other applications. In another possible battery
application, the Fluff might carry barium into plates to act as an
"expander" which enhances discharge capacity. In this application,
the Fluff could be expected to disperse Ba slowly, during curing,
so that any BaSO.sub.4 produced would be minimal in size and would
not undesirably enhance shedding. This might allow use of barium as
an expander within positive as well as negative plates.
[0042] The microporous basic lead sulfate Fluff material of the
invention is distinctly different from the high surface area lead
oxide (HSAO) made by treating an aqueous suspension of lead oxide
with ozone, as described by G. Anthony Parker in U.S. Pat. No.
4,388,210. First, the HSAO material is made without sulfate, i.e.
merely from oxide., in an aqueous slurry which contains various
sodium salts but not specifically sodium tetraborate, which is
treated by addition of ozone to create a mixture of lead oxide and
lead dioxide. Second, the HSAO material seen under SEM examination
has merely a coarse surface texture: it does not have internal
microporosity, and consequently cannot store any significant
quantity of, for example, a habit-modifying ingredient. Such an
ingredient would be coated at best only on the outside of the
coarse HSAO, where it would act like bulk, unabsorbed
habit-modifier, that is, would dissolve as quickly as
habit-modifier without HSAO. Furthermore, the density of the Parker
HSAO material is substantially more than 3.5 g/cc.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
Manufacture of Lead Fluff
[0043] The Fluff material of the invention can be made in the same
process as the PCA material. Fluff was initially identified as a
nuisance and undesirable byproduct of the manufacture of PCA, as
described in application Ser. No. 11/234,077, and its production
was minimized in order to maximize the production of PCA. Any Fluff
still produced was separated during the grinding operation and was
scrapped out. The value of Fluff was recognized later.
[0044] As disclosed in Ser. No. 11/234,077, PCA can be made by
using a range of processing steps carried out on a typical battery
paste of, e.g., 5-6 wt % sulfuric acid, 12-16 wt % water, balance
leady oxide. Typically this paste is heated to 80-90.degree. C. for
about 5-10 minutes. Other examples are given in Ser. No.
11/234,077. Those of skill in the art will recognize this range of
temperatures in particular as being considerably higher than the
usual ranges used to cure such a paste in the typical battery plate
pasting process. If the mixture is mixed of pre-heated ingredients,
and especially when the acid is rapidly added, the resultant PCA
comprises 3BS and 4BS and also some low density material which is
composed of a small residue of lead oxide, some various basic lead
sulfate salts including lead sulfate hydroxide and lead sulfate
carbonate. By comparison, a typical XRD analysis of Fluff is: 63.7%
4BS, 10.5% 3BS, 7.8% 1BS, 13.1% hydrocerrusite, 3.4% litharge form
of PbO and 1.1% leadhillite.
[0045] If the paste mixing is continued, especially with continued
external heating, so that the water is driven off the paste quickly
(over on the order of 30 minutes), followed by grinding, the PCA
paste breaks down to a powder with grains in the size range from 50
down to <500 mesh. The heating and mixing create drafts such
that the finest and lowest-density portion is thrown off, and
further fine particles are similarly created during grinding. This
low-density material, which is the novel Fluff of the invention,
can be collected, e.g., with a vacuum source discharging into a
baghouse, in both the mixing and grinding steps, as desired. Fluff
might also be collected in a known "dropout" system, or by an air
classifier, known in the art for separating fine particles
according to their relative sizes. The coarser particles can be
returned to the grinder for further reduction, or can be used "as
is" as PCA. The larger particles of PCA are primarily 3BS and
unreacted lead oxide, with a small residue of free lead. As above,
Fluff is a mixture of lead oxide with various basic lead sulfates
and a smaller amount of free lead. More significantly, Fluff has a
low density (<3.5 g/cc) and correspondingly high porosity, and
is insoluble in and does not react with water.
[0046] Fluff, unlike PCA, will not by itself nucleate paste to grow
4BS, despite being substantially composed of 4BS. Hence, PCA is
still preferably also used in making battery paste, together with
BF. Without limiting the invention, it is theorized that the lack
of free PbO probably is what prevents Fluff from acting like PCA to
nucleate growth of 4BS. However, when Fluff is subsequently
incorporated into battery paste, as described below, it is
incorporated into the cured paste materials (3BS and/or 4BS) and
structure, so that its lead content contributes to the battery
function.
[0047] The heating, mixing, drying, crushing, and separation of
Fluff from battery paste which has been formulated and treated to
form PCA and Fluff can be performed over a wide range of
conditions, to be optimized depending on starting oxide composition
and paste processing characteristics. The yield of Fluff can be
improved by modification of these conditions. The yield of Fluff
also depends heavily on the processes and apparatus used to collect
it.
[0048] In one example, starting with two paste mixes totaling 1999
lb of leady oxide, 100 lb of water and 128 lb of 50 wt % (1.400
sp.gr.) aqueous sulfuric acid, the leady oxide and acid were
unheated and the water was preheated to 47.6 deg C. The ingredients
were mixed, forming a paste which was subsequently dried in the
mixer to a residual moisture content of <1%. At this moisture
content, the PCA material can be ground readily; at >2% moisture
content the grinding operation is less efficient. In the example,
the dried powder was ground slowly (500 lb/hr) in a Pulva model B
hammermill grinder, and there was no collection of dust during
mixing; the yield was 2090 lb of the high-density PCA residue and
15 lb of Fluff. As the PCA passed through the grinder into a
collection hopper, Fluff was ejected through the vent port at the
top of the grinder into a dropout box which in turn was connected
to a ventilation system.
[0049] Since the reaction stoichiometry provides a weight gain from
oxidation of most of the free Pb in the Fluff, and from the sulfate
in the aqueous sulfuric acid, the inventor calculates that a few lb
of Fluff was lost but could have been collected if dust collection
steps had been performed at the mixer. The goal of this experiment
was to maximize the yield of the high-density PCA material for
another application, and therefore was to minimize the yield of
Fluff; it appears clear that the processing conditions can be
optimized for the production of Fluff. More specifically, in
previous experiments in making PCA that were carried out on a
kilopound scale, hotter water, more rapid acid addition and more
rapid grinding were used, and a greater yield of Fluff was
achieved. Therefore, for any particular starting combination of
oxide material, acid, and water, it is logical to presume that the
yield of Fluff can be optimized or maximized responsive to
variation in the mixing and/or grinding steps.
[0050] Neither the patent nor technical literature known to the
applicant describes any method to produce a low density, highly
porous form of any basic lead sulfate. Mere mixing of 3BS, 4BS and
the other ingredients, even when in a water slurry dried using an
open-flame gas heater to make hydrocerrusite and leadhillite in
situ, does not yield a low density product, nor a particularly
microporous product.
Manufacture of Slow-Dissolving Habit-Modifier/4BS Growth
Inhibitor
[0051] Fluff will cake together when wet, but, upon drying, is
readily crushed back to a fine dust. This can be done repeatedly,
as shown by its unchanged color, density, and ability to absorb
other materials. It appears that neither the 4BS nor 3BS are
recrystallized in this process.
[0052] In use of Fluff for inhibiting growth of large 4BS crystals
in battery manufacture, the study next turned to identification of
a suitable material to be absorbed by the Fluff, which would
diffuse slowly out of the Fluff carrier to modify crystal habits
and growth, as discussed briefly above.
[0053] Organics are generally precluded for prevention of large 4BS
crystal growth. Although organics are reported to prevent formation
of 4BS, experiments with battery paste containing 4BS seeds and
organics (such as the lignosulfonates used as negative plate
expanders) show that 4BS can be grown. Therefore, the organics
prevent nucleation, not growth.
[0054] Most inorganic salts (halides, oxyhalides, sulfates,
nitrates, nitrites and transition metal oxyanions such as chromate,
manganate, etc.) are too expensive, too insoluble, and/or are
electrochemically incompatible with the battery reactions.
[0055] Carbon-containing anions (acetate, citrate, etc) and the
usual organic habit-modifying materials (urea, surfactants, etc)
can degrade to acetic acid which will rapidly and undesirably
corrode battery grids after oxidation on the PbO.sub.2 in the
formed positive plates.
[0056] It being desired to provide the crystal growth inhibitor as
a salt, so as to provide chemical neutrality, a few salts are left
as possible candidates for use as inhibitors of growth of large 4BS
crystals: hydroxides, borates, oxides, carbonates, bicarbonates,
silicates, phosphates, and possibly fluoro-derivatives of these.
The alkaline paste will readily dissolve most of these. The
counterion (cation) of choice (which is needed so that a salt is
added) is probably sodium, based on minimal cost (as compared to,
e.g., Li, K, Rb, etc.). Of possible sodium compounds, NaO and NaOH
react rapidly, and the carbonate and bicarbonate react more slowly
with lead oxide to generate the blood red salt sodium plumbate, and
change (reduce) the paste viscosity, so these are inappropriate
choices.
[0057] This leaves as candidates sodium borates, silicates, and
phosphates, and perhaps the peroxysalts mentioned above, although
the latter are much more expensive and much less stable than their
non-peroxygenated analogues. The literature referred to above shows
that borax (this term being used loosely here, as generally in the
art, to refer to, e.g., sodium borates of various specific
compositions, such as anhydrous borax (Na.sub.2B.sub.4O.sub.7),
borax pentahydrate (Na.sub.2B.sub.4O.sub.75H.sub.2O), and borax
decahydrate (Na.sub.2B.sub.4O.sub.710H.sub.2O)) has been used to
modify the crystal habits of various soluble transition metal
sulfate salts such as MgSO.sub.4 and ZnSO.sub.4, when used in a
concentration of 5 weight percent of the transition metal sulfate.
Mixing 5 wt % borax with battery paste changed its characteristics
rapidly from a useable paste to a low viscosity suspension. Even as
little as 0.1 wt % borax caused the paste to liquefy. Hence, it
appears that any direct addition of borax to paste renders the
paste useless.
[0058] As above, the inventor has found that borax could be
absorbed within lead Fluff, and that the resultant BF material
would not affect paste characteristics. Moreover, the borax
diffused out of the Fluff during curing, such that it eventually
modified 4BS growth and reacted to form various lead borate salts
which were small in size and high in surface area. The inventor
briefly considered other possible microporous carriers for borax,
such as TiO.sub.2 and ZrO.sub.2, but these are generally more
expensive than lead Fluff and would remain inert within the plates
during the battery life, since these are electrochemically inactive
in a lead-acid battery. By comparison, the lead in Fluff
contributes to the battery's function. Furthermore, at end of life,
these other carriers might cause issues with battery recycling.
[0059] Borax is commercially available cheaply in many different
forms which vary in terms of amount of hydration from anhydrous
borax through 2, 4, 5, and up to 10 waters of hydration. Borax
decahydrate (-10 H.sub.2O) was chosen, although the pentahydrate
works equally well. Manufacturer's literature shows the solubility
of borax decahydrate is 170 grams per 100 cc H.sub.2O at
100.degree. C. This was interpolated to an estimated 150 g/100 cc
solubility at 80.degree. C. In a first trial, 2 kg of the
decahydrate was dissolved in 5 liters of water but this was
insufficient since addition of the salt to the hot water cooled the
water off, so that some material would not dissolve, or, worse,
would crystallize out. 6 liters of water was sufficient to dissolve
2 kg of borax decahydrate. After the borax was dissolved, with the
solution maintained at >90.degree. C., Fluff was added. Heating
was continued until all the air in the pores of the Fluff had been
driven off and the borax solution fully penetrated the pores, and
continued until the remaining free water had been boiled off.
Although 2 kg of Fluff will absorb 6 liters of the borax solution,
there is probably some on the outside of each Fluff grain, as seen
by the difficulty in grinding this composition after it has been
dried. 3 kg of Fluff fully absorbs the borax solution, forming the
desired BF material, and is easy to dry and grind, so the preferred
ratio of materials is 6 liters of water, 5 kg of borax decahydrate
and 3 kg of Fluff.
[0060] The lab-scale production of BF described above was carried
out in enameled steel or stainless steel cooking pots heated over
electric hot plates, but it is obvious that this can be scaled up
for quantity. The water was measured out and heated to
>/90.degree. C., and borax decahydrate was weighed out and was
added in incremental portions to the pot, such that temperature was
maintained above 90.degree. C. so that no borax precipitated out of
the solution. After all the borax had been dissolved, the Fluff was
weighed out and also added in portions to the pot to maintain the
temperature above 90.degree. C., to maintain the complete
solubility of the borax. Fluff does not dissolve, so the solution
becomes a suspension of the Fluff in the aqueous borax solution.
The suspension is heated with the pot partially uncovered such that
the volume of the suspension is reduced 10-20%, and the suspension
reaches the consistency of a thick mud. In order to control the
subsequent drying of the mud, it was found convenient to transfer
the mud to smaller, shallower stainless steel pans and then dry
these in a laboratory oven at between 80 and 140.degree. C. If the
mud is dried too quickly, such that borax separates from the mud
rather than being absorbed by the porosity of the Fluff carrier,
the batch can be redissolved and drying can be repeated at a lower
temperature or at a lower drying rate (for example, by providing
less ventilation).
[0061] Other materials that might also be useful are compounds
which, like borax, that is, sodium borate, include an alkaline
anion or hydrogen ion with a suitable cation, such as sodium
phosphates, acid phosphates, sulfates, acid sulfates, silicates,
fluoborates, or peroxysalts such as perborate, persulfate or
percarbonate.
Use of BF Material in Battery Paste
[0062] The efficacy of BF as an inhibitor of growth of 4BS crystals
requires, first, that it be used with 4BS seed crystal materials,
which are available from several commercial sources, and second,
that both these seed crystals and the habit-modifier additive BF be
mixed intimately into the battery paste. There is some advantage to
dry mixing the 4BS seeds with the oxide, since these powders mix
together rapidly if not wet, and the 4BS seeds can start growing
4BS crystals during the paste mixing. Less preferred is to mix by
adding water first, then oxide and wet mixing this, then adding the
4BS seeds, then adding the acid and then final mixing. The BF
material should be added after acid addition is complete, but
before final mixing. It may be possible to add BF at an earlier
point in the process, for example by absorbing borax onto the Fluff
component in an unseparated PCA-Fluff mixture, but conditions for
this have not yet been established. As mentioned above, PCA is
still used, as described in prior application Ser. No. 11/234,077,
as part of the battery paste. After pasting, the use of any 4BS
nucleant will allow faster curing, or equal curing at lower
temperature, either way requiring less thermal energy. During the
cure, the borax dissolves slowly out of the Fluff carrier, reacts
with some lead oxide or basic lead sulfate, and modifies the growth
of 4BS. Further, the basic lead sulfate carrier (fluff) is
incorporated into the crystalline matrix in the cured paste and is
available for electrochemical utilization.
[0063] More specifically, at curing temperatures >70.degree. C.,
secondary nucleation of 4BS can occur within the paste. Both
primary and secondary nucleated 4BS crystals are limited in growth
by the modifying additive. At curing temperatures <70.degree.
C., the primary nucleated 4BS crystals (that is, grown from seeds)
are smaller in size and/or more numerous than would be grown
without the microporous BF carrier and the contained growth
modifying additive. Accordingly, the curing temperature can be
selected to control the amount of 4BS grown, in accordance with the
desired use of the battery; as above, SLI batteries generally are
preferred to have more 3BS than 4BS, while the converse is true
with deep-cycle batteries.
[0064] Laboratory processing tests have shown that less than the
calculated 0.5-1 wt % may be effective, in that a residue of borax
is found in the sump of the curing chamber after curing a batch of
plates containing BF. No similar residue was observed in the plant
trials, but may occur during factory production of plates using BF.
If so, the amount of BF can be reduced or the residue can be
discarded or recycled.
[0065] Those of skill in the art will recognize that batteries are
typically "formed", that is, initially charged, using either a
"1-shot" or a "2-shot" formation technique. In 1-shot formation,
the unformed battery is filled with a more concentrated aqueous
sulfuric acid electrolyte; as the battery undergoes initial
(formation) charge, the sulfates from the various basic lead
sulfates in the plate increase the electrolyte concentration
modestly.
[0066] For example, a filling acid added at a specific gravity of
1.220 to 1.250 is increased to the range 1.270 to 1.300, and the
battery is shipped and sold at this concentration. 2-shot formation
involves initial use of an electrolyte of very low concentration,
1.005 to perhaps 1.030, which after formation increases to the
range 1.050 to 1.180. This acid is dumped out and the battery is
refilled with a more concentrated electrolyte solution slightly
higher in density than the desired final/shipping specific gravity.
For example, a final specific gravity of 1.285 might be achieved by
adding acid of specific gravity 1.290 to a residue inside the
battery of 1.100. Most SLI batteries are formed using the 1-shot
process, while most deep discharge batteries (truck/bus, golf cart,
forklift, stationary) are formed by the 2-shot process.
[0067] Any further residual borax (or similar
crystal-growth-modifying additive) within the plates will be
retained within the cell if the battery had been formed by the
1-shot process, but neither borax, sodium nor any of the other
anions will be deleterious to battery performance and life. Such
further residue within the plates might be washed out during the
acid-dumping step if the battery had been formed by the 2-shot
process. The residue can be separated from the dumped acid,
allowing its reuse.
[0068] There may be some application for using the growth-modifying
additive, for example sodium borate in any of its commercial forms,
to deliberately decrease the viscosity of a battery paste which is
otherwise difficult to paste, and for which the paste formulation
cannot be changed by adding more liquid.
[0069] Table I below summarizes the results of tests performed at
four different battery manufacturer's facilities. The notes and
column headings explain the test circumstances and conditions. For
example, test 1-1 was carried out on Jun. 9, 2010; the grinding was
done using a ball mill (BM); the paste mix included 550 kg of
oxide, 12.5 kg of PCA, 7 kg of BF, 61 kg of water, and 53.5 kg of
sulfuric acid; and an Oxmaster-type mixer was used. The curing
conditions were that the plates were stacked, the curing
temperature was <50.degree. C., the relative humidity was
>95%, and the cure time was 23.55 hours for plates containing
PCA and the BF additive. In several cases, control plates were also
made, using compositions without BF, and the data as to these is
reported as well.
[0070] As indicated, where CCA values were applicable, improvements
of typically 12-15% were noted where the BF was added to the paste
mix.
TABLE-US-00001 TABLE 1 SUMMARY OF PLANT TESTS Paste mix
composition: Kg Company-test no. Date Oxide/PCA/BF/H.sub.2O/acid
1.400 sp.gr. Mixer type Recommendation -
BP1000/22.7/12-18/104/128.8 Oxmaster 1-1 June 2009 BM
550/12.5/7/61/53.5 Oxmaster 1-2 July 2009 BM 500/12.5/7/61/53.5
Oxmaster 2-1 July 2010 BP1000/15/7/126/119 Oxmaster 3-1 April 2010
BM 550/11.5/8/65/50 Eirich 3-2 July 2010 BM 550/11.5/8/62.5/60.4
Eirich Curing variables (C)ure: temp/RH/Time End of Cure/Dry
Co.-test no (D)ry: paste H.sub.2O 5%-8% for >/= 4 hrs % Pb %
H.sub.2O Notes Recommendation: racked, C > 80.degree.C., >95%
RH, </= 48 hrs <2 <1 1-1 stacked, C < 50.degree. C.,
>95% RH, 23.55 hrs BF, 39 hrs control <2 <1 D not
specified 1-2 same as 1-1 same as 1-1 2-1 stacked, C 70.degree. C.,
100% RH, 14 hrs 2.5 3.0 note 5 D 75.degree. C., 0% RH 20 hrs not
measured 3-1 stacked C < 70.degree. C. 3-2 stacked C/D profile
NOTE 6 stacked >2 >2 Racked <2 >2 4-1 stacked C >
80.degree. C., >95% RH, 48 hrs <2 <1 D > 80.degree. C.
unknown decreasing RH Co.-test no CCA improvement Other
processing/performance SEM/porosity/XRD Recommendation: predicted
>10% Predict: notes 1, 2, 3, 4 >80% 4BS with >80% Crystals
<20 microns 1-1 12% notes 1, 2, 3, 4 mostly 4BS, some large xtal
1-2 15% notes 1, 2, 3, 4 BF: very good 3BS, 4BS, Excellent porosity
Control: poor 3BS, unreacted oxide Poor porosity 2-1 15% not yet
reported BF: excellent 4BS < 15 microns Excellent porosity C:
mostly 3BS, unreacted oxide Poor porosity 3-1 >12% not yet
reported BF: <10% large 4BS > 20 microns Predominantly
smaller 4BS Excellent porosity C: mostly unreacted oxide, poor
porosity 3-2 not yet reported not yet reported BF: not yet examined
C: see test 3-1 4-1 not applicable Notes 1, 2, 3, 4 BF: small 4BS
< 20 microns Excellent porosity C: (PCA but no BF) 4BS 10 to 40
microns, excellent porosity NOTES: Paste mix composition: oxide
preparation BM = ball mill, BP = Barton pot (1) faster formation:
tests 1-1 and 1-2 28 hrs BF vs. 30 hrs control (2) less gassing on
formation (3) less water loss on formation (4) mechanically
stronger plate: tests 1-1 and 1-2 mechanical drop test, test 4,
puncture (5) unknown procedures for % Pb, % H.sub.2O, poor
equipment and technique, unknown calibration (6) Profile:
80.degree. C./95% RH/8 hrs but actual 70.degree. C./93% RH/8 hrs,
then 80.degree. C./70% RH/2 hrs, then 80.degree. C./50% RH/4 hrs,
then 80.degree. C./20% RH/2 hrs, then 80.degree. C./0% RH/4hrs,
then 50.degree. C./0% RH/2 hrs, then 40.degree. C./0% RH/2 hrs.
[0071] While several examples of the invention have been provided
above, the invention is not to be limited thereto.
* * * * *