Note: Descriptions are shown in the official language in which they were submitted.
'O 94/07272 2 1 ~ 4 6 5 7 PCI`/US93tO8803
ENERGY STORAGE DEVICE AND
METHODS OF MANUFACTURE
.
BACKGROUND OF THE INVENTION
5 Cross-Reference To Related APplications
The present application is a continuation-in-part of U.S. Application Serial No.07/947,414, filed September 18, 1992, U.S. Application Serial No. 07/947,294, filed
September 18, 1992, and U.S. Application Serial No. 07/958,506, filed October 7,1992, all of which are incorporated by reference in their entirety.
10 Field of the Invention
The present invention generally relates to an energy storage device, and more
particularly to a bipolar double layer capacitor-type energy storage device, and to
methods for manufacturing the same.
Description of the Related Art
15 Ener~y Stora~e Devices -- There has been significant research over the years, relating
to useful reliable electrical storage devices, such as a capacitor or a battery. Large
energy storage capabilities are common for batteries; however, batteries also display
low power densities. In contrast, electrolytic capacitors possess very high power
densities and a limited energy density. Further, carbon based double-layer capacitors
20 have a large energy density; but, due to their high equivalent series resistance (ESR),
carbon electrodes have low power capabilities. It would therefore be highly desirable
to have an electrical storage device that has both a high energy density and a high
power density.
A recent review by B.E. Conway in J. Electrochem. Soc., vol.138 1#6), p.1539
25 (June 1991) discusses the transition from "supercapacitor" to "battery" in
electrochemical energy storage, and identifies performance characteristics of various
capacitor devices.
D. Craig, Canadian Patent No. 1,196,683, in November 1985, discusses the
usefulness of electric storage devices based on ceramic-oxide coated electrodes and
30 pseudo-capacitance. However, attempts to utilize this disclosure have resulted in
capacitors which have inconsistent electrical properties and which are often unreliable.
WO 94/07272 ~ PCl /US93/088--
2i~4~
These devices cannot be charged to 1.0 V per cell, and have large, unsatisfactory
leakage currents. Furthermore, these devices have a very low cycle life. In addition,
the disclosed packaging is inefficient.
M. Matroka and R. Hackbart, US Patent No.5,121,288, discusses a capacitive
5 power supply based on the Craig patent which is not enabling for the present invention.
A capacitor configuration as a power supply for a radiotelephone is taught; however,
no enabling disclosure for the capacitor is taught.
J. Kalenowsky, US Patent No. 5,063,340, discusses a capacitive power supply
having a charge equalization circuit. This circuit allows a multicell capacitor to be
10 charged without overcharging the individual cells. The present invention does not
require a charge equalization circuit to fully charge a multicell stack configuration
without overcharging an intermediate cell.
H. Lee, et al. in IEEE Transactions on Ma~netics, Vol. 25 ~#1), p.324 (January
1989), and G. Bullard, et al., in IEEE Transactions on Ma~netics, Vol. 25 (#1) p. 102
15 (January 1989) discuss the pulse power characteristics of high-energy ceramic-oxide
based double-layer capacitors. In this reference various performance characteristics are
discussed, with no enabling discussion of the construction methodology. The present
invention provides a more reliable device with more efficient packaging.
Carbon electrode based double-layer capacitors have been extensively developed
20 based on the original work of Rightmire, U.S. Patent No.3,288,641. A. Yoshida et al.,
in IEEE Transactions on ComPOnentS, HYbrids and Manufacturin~ Technoloqv, Vol.
CHMT-10, #1,P-100 - 103 (March 1987) discusses an electric double-layer capacitor
composed of activated carbon fiber electrodes and a nonaqueous electrolyte. In
addition, the packaging of this double-layer capacitor is revealed. This device is on the
25 order of 0.4-1 cc in volume with an energy storage capability of around 1 -10 J/cc.
T. Suzuki, et al., in NEC Research and DeveloPment, No.82, pp.118 - 123, July
1986, discloses improved self-discharge characteristics of the carbon electric double-
layer capacitor with the use of porous separator materials on the order of 0.004 inches
thick. An inherent problem of carbon based electrodes is the low conductivity of the
30 material resulting in a low current density being delivered from these devices. A second
difficulty is that the construction of multicell stacks is not done in a true bipolar
~0 94/07272 PCI/US93/08803
21~4`6~7
electrode configuration. These difficulties result in inefficient packaging and lower
energy and power density values.
Additional references of interest include, for example:
The state of solid state micro power sources is reviewed by S. Sekido in Solid
5 State lonics, vol. 9, 10, pp. 777-782 (1983).
M. Pham-Thi et al. in the Journal of Materials Science Letters, vol. 5, p. 415
(1986) discusses the percolation threshold and interface optimization in carbon based
solid electrolyte double-layer capacitors.
Various disclosures discuss the fabrication of oxide coated electrodes and the
10 application of these electrodes in the chlor-alkali industry for the electrochemical
generation of chlorine. See for example: V. Hock, et al. US Patent No. 5,055,169issued October 8,1991; H. Beer US Patent no.4,052,271 issued October 4,1977; andA. Martinsons, et al. US Patent no. 3,562,008 issued February 9, 1971. These
electrodes, however, in general do not have the high surface areas required for an
15 efficient double-layer capacitor electrode.
It would be useful to have a reliable long-term electrical storage device, and
improved methods to produce the same. It would also be desirable to have an improved
energy storage device with energy densities of at least 20-90 J/cc.
Packaqinq of Enerqv Storaqe Devices -- As mentioned above, there has been significant
20 research over the years regarding electrical storage devices of high energy and power
density. The efficient packaging of the active materials, with minimum wasted volume,
is important in reaching these goals. The space separating two electrodes in a capacitor
or a battery is necessary to electronically insulate the two electrodes. However, for
efficient packaging, this space or gap should be a minimum. It would therefore by
25 highly desirable to have a method to create a space separator or gap that is
substantially uniform and of small dimension (less than 5 mil (0.0127 cm).
A common way to maintain separation between electrodes in an electrical
storage device with an electrolyte present (such as a battery or capacitor) is by use of
an ion permeable electrically insulating porous membrane. This membrane is commonly
30 placed between the electrodes and maintains the required space separation between the
two electrodes. Porous separator material, such as paper or glass, is useful for this
WO 94/07272 PCI /US93/088--
2i~657
application and is used in aluminum electrolytic and double layer capacitors. However,
for dimensions below 1 or 2 mil (0.00254 to 0.00508 cm) in separation, material
handling is difficult and material strength of the capacitor is usually very low. In
addition, the open cross-sectional areas typical of these porous membrane separators
5 are on the order of 50-70%.
Polymeric ion permeable porous separators have been used in carbon double
layer capacitors as discussed by Sanada et al. in IEEE, pp.224-230, 1982, and bySuzuki et al. in NEC Research and DeveloPment, No. 82, pp. 118-123, July 1986.
These type of separators suffer from the problem of a small open area which leads to
10 increased electrical resistance.
A method of using photoresist to fill voids of an electrically insulating layer to
prevent electrical contact between two electrode layers for use as a solar cell is
disclosed by J. Wilfried in US Patent No. 4,774,193, issued September 27, 1988.
A process of creating an electrolytic capacitor with a thin spacer using a
15 photosensitive polymer resin solution is disclosed by Maruyama et al in US Patent No.
4,764,181 issued August 16,1988. The use of solution application methods described
in a porous double-layer capacitor electrode would result in the undesirable filling of the
porous electrode.
Additional references of general interest include U.S. Patents 3,718,551;
20 4,052,271; and 5,055,169. All of the applications, patents, articles, references,
standards, etc. cited in this application are incorporated herein by reference in their
entirety.
In view of the above, it would be very useful to have a method to produce a
reliable small space separation between electrodes in electrical storage devices with a
25 large open cross-sectional area.
SUMMARY OF THE INVENTION
The present invention relates to a novel electrical storage device that has
both a high energy density and a high power density.
It is an object of the present invention to provide new methods for
30 manufacturing the storage device.
~0 94/07272 PCI /US93/08803
21;4~657
It is also another object of the present invention to provide a reliable long-
term electrical storage device, and improved methods to produce the same.
It is a further object of the present invention to provide efficient packaging of
an electrical storage device by reducing the gap between the anode and cathode,
5 which reduces the electrical resistance of the ionically conducting electrolyte.
Briefly, the foregoing and other objects are attained by an energy storage
device such as a capacitor, which includes a plurality of cells in a bipolar
configuration. The cells are stacked and bonded together, to impart to the device an
integral and unitary construction.
Each cell includes two electrically conductive electrodes that are spaced apart
by a predetermined distance. The cell also includes at least one dielectric gasket
that is interposed, in relation to each other, between the electrodes, for separating
and electrically insulating these electrodes.
When the electrodes, and the gaskets are bonded together, at least one fill
15 gap is formed for each cell. Each cell also includes a high surface area (porous)
electrically conductive coating layer that is formed on one (or, more) surface of each
electrode. This coating layer optionally includes a set of closely spaced-apart
peripheral microprotrusions, and a set of distally spaced-apart central
microprotrusions. These microprotrusions are formed by novel screen printing or
20 photolithographic printing methods. These microprotrusions impart structural
support to the cells, and provide additional insulation between the electrodes.
An ionically conductive medium fills the cell gap and pores of the high surface
area coating.
BRIEF DESCRIPTIQN OF THE FIGURES
The above and other features of the present invention and the manner of
attaining them, will become apparent, and the invention itself will be best
undzrstood, by reference to the following description and the accompanying
drawings, wherein:
Figure 1 is a perspective view of the preunit 10 a dry energy storage device
which is constructed according to the present invention;
WO 94/07272 PCI/US93/088--
23i44G~7
Figure 1 A is a perspective view of the electrolyte-filled energy storage device1 OA of the present invention;
Figure 2 is a cross-sectional view of the storage device of Figure 1 showing a
removable cord 1 1 7A within the storage device, taken along line 2-2 thereof;
Figure 2A is another cross-sectional view of the storage device of Figure 1,
taken along line 2A-2A thereof;
Figure 3 is a schematic representation of an exploded view of ~he preunit of
Figure 1, illustrating three cells;
Figure 4 is a block diagram of the manufacture steps of the storage device
10A;
Figure 5 is a top plan view of a porous coating layer with microprotrusions
which forms a part of the storage device of Figures 1 through 4;
Figure 6 is a diagrammatic illustration of a capacitive circuit, which is
equivalent to the device 1 OA;
Figure 7 is a schematic representation of a screen printing method to produce
microprotrusions on a coating layer used with the energy storage device according to
the present invention;
Figure 8 is a schematic representation of an electrode holder used in the
manufacture method of Figure 7;
Figure 9 is a schematic representation of a method to photolithographically
produce the microprotrusions according to the present invention;
Figure 10 is a schematic isometric view of a pair of hot rollers used for
laminating a photoresist to an electrode prior to photolithography;
Figure 11 is a schematic isometric view of a mask placed over the photo
resist of Figure 10;
Figure 12 is a schematic isometric view illustrating the exposure of
unprotected portions of the photo resist of Figures 10 and 1 1;
Figure 13 is a cross-sectional view of an electrode which forms a part of the
energy storage device, taken along line 13-13 of Figure 3; and
0 94~07272 PCI/US93/08803
~ 2~ S;7
Figure 14 is a schematic cross-sectional view of two bipolar electrodes with
the high surface area porous coating layer on the electrically conducting substrate
forming one cell.
Figure 15 is a schematic view of a frame used to hold thin support materials
5 during the dip coating process;
Figure 1 5A is a schematic view of wire used in the frame of Figure 15.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
Definitions
The definitions of the following terms are not intended to be exclusive:
"Cord" refers to the thin strips of material included in the method of
manufacture of the dry preunit. After initial heating, the removal of the cord
produces the open fill ports.
"Electrically conducting support material" refers to any electrically conductingmetal or metal alloy, electrically conducting polymer, electrically conducting ceramic,
15 electrically conducting glass, or combinations thereof. Metals and metal alloys are
preferred for producing stock units. The support material should have a conductivity
of greater than about 10 4 S/cm.
"Second electrically conducting material" (having a high surface area) refers
to a porous electrode coating which may be of the same or different composition on
20 each side of the support material. Preferred metal oxides of the present invention
include those independently selected from tin, lead, vanadium, titanium, ruthenium,
tantalum, rhodium, osmium, iridium, iron, cobalt, nickel, copper, molybdenum,
niobium, chromium, manganese, lanthanum, or lanthanum series metals or alloys orcombinations thereof, and possibly containing additives like calcium to increase25 electrical conductivity.
"Electrolyte" refers to an ionically conductive aqueous or non-aqueous
solution or material, which enables the dry preunit to be electrically charged.
"Cab-O-Sil~"' refers to silica filler available from Cabot Corporation of Tuscola,
Illinois. A variety of sizes are available.
WO 94/07272 PCI /US93/088
21~6S7
"Epoxy" refers to the conventional definition of the product which is an
epoxy resin mixed with a specific curing agent, usually a polyamine. or polyepoxide
mi~ced with a polyamine curing agent.
MYLAR~ refers to a polyester of polyethylene terephthalate of DuPont, Inc. of
5 Wilmington, Delaware. It is usually commercially available in sheet form of varying
thicknesses .
"Metal oxide" refers to any electrically conducting metal oxide.
"Mixed metal oxide" refers to an electrically conducting oxide compound of
two or more metal oxides.
"Photoresist" is any photo curable material. Usually, it is an epoxide or
acrylate or combinations thereof.
"ConforMASK" is a negative working photopolymer available commercially
from Dynachem of Tustin, California. This polymer should be used at 50% or less
relative humidity.
15 Drv Preunit Ener~v Stora~e Device
Referring now to the drawings, and more particularly to Figures 1, 2 and 3
thereof, there is illustrated a dry preunit of energy storage device 10 which isconstructed according to the present invention. The energy storage device is first an
assembled dry preunit 10. After filling the present cells with an electrolyte, the
20 surface is heated to close and to fuse the exterior surface, to form device 1 OA
which is then electrically charged.
The device preunit 10 generally includes a plurality of cells, such as the cells1 10, 1 12 and 1 14, which are formed, prepared, and stacked according to the
teaching of the present invention. Figure 1 A illustrates an assembled view of the
25 electrical storage device preunit 1 OA, formed of twelve superposed cells. It should
however be understood to those skilled in the art, after reviewing the present
specification that any different number of cells can be used.
For simplicity of illustration, Figure 3 is an exploded view of the preunit 10,
showing only three exemplary cells 1 10, 1 12 and 1 14. The cells have generally30 similar design and construction, and therefore, only the cells 1 14 and 1 12 will be
described in detail, in relation to Figures 2, 2A, 3 and 13.
~ 94/07272 PCI~/US93~08803
21~6~7
The cell 1 14 includes a first electrically conductive external electrode or end
plate 111 A, and a second internal, electrically conductive bipolar electrode 111 B.
Both electrodes 111 A and 111 B are spaced apart at the edges by means of two
dielectric or electrically insulating gaskets 121 and 123.
When the first and second electrodes 111 A and 111 B, and the insulating
gaskets 121 and 123 and the electrically conducting porous material (oxide) layers
119 and 120 are bonded together to form the cell 1 14, a central air filled gap 130
(Figure 2A) is formed by these elements. When the preunit 10 is ready to be used,
the gap 130 is filled with an electrolyte (not shown) to produce device 1 OA.
For this purpose, an exemplary access or fill port 122, is shown in Figure 2A
for illustration purpose only, and is formed between the gaskets 121 and 123, inorder to allow the electrolyte to fill the gap 130. The fill port 122 is formed by
means of a tab or cord 117A, which is inserted between the gaskets 121 and 123,
prior to fusing or bonding the gaskets 121 and 123. When the gaskets 121 and
123 are heated, the cord 117A becomes surrounded by the reflow gasket material,
which causes the outline of fill port 122 to be formed. The two gaskets become afused polymer mass covering a minimum of the active electrically conducing coating
layers 119 and 120.
Considering now the electrodes 111 A and 111 B in greater detail, the
methods of manufacturing them will be described later. One difference between the
electrodes 111 A and 111 B is that the electrode 111 A optionally includes a tab160A, for connection to a power source (not shown).
A further, but optional distinction between the electrodes 111 A and 111 B, is
that the electrode 111 A includes one porous electrically conductive coating layer
119, which is deposited on a support material or structure 116, while the bipolar
electrode 111 B includes two porous coating layers 120 and 131, which are
deposited on either or both sides of the support material or structure 111 B. Assuch, the electrode 111 B is a true bipolar electrode. It should be understood that
both sides of the electrode 111 A could be coated with porous electrically conductive
layers.
WO 94/07272 2 1 4 4 6 ~ 7- PCI/US93/08~
,
Yet another optional distinction between the electrodes 1 1 1 A and 1 1 1 B liesin the rigidity of the support structures 1 1 1 A and 1 1 1 B. The electrode 1 1 1 A, acting
as an external end plate, should preferably have a more rigid construction, so that it
imparts sufficient rigidity to the overall structure of the energy storage device 1 OA.
5 The electrode 111 B and other similar internal electrodes do not necessarily need to
be as rigid as the external electrode 1 1 1 A. Nonetheless, when the device 1 OA is
large, additional support structure is required, and the internal electrodes, i.e. 111 B,
are used as additional support structure. In this case, it is desirable to rigidify the
internal electrodes, i .e . 111 B.
As a result, the support material 1 16 is thicker than the support material 1 18.
In the preferred embodiment, the support material 1 16 has a thickness of about 10
mils (0.0254 cm),while the support material 118 has a thickness of about 1 mil
(0.00254 cm). Other values could alternatively be selected.
The electrodes 1 1 1 A, 1 1 1 B and the remaining electrodes of the storage
15 device 1 OA, are sized and dimensioned according to the desired application, without
departing from the scope of the invention. For instance, in one application, thedevice 1 OA is miniaturized, e.g. for a cardiac defibrillator. While in another
application, the Gverall volume of the device is one cubic meter or even greater, e.g.
for an electric vehicle. The dimensions of the electrodes determine the overall
20 capacitance of the storage device 1 OA.
In the preferred embodiment, the electrodes, i.e. 1 1 1 A and 1 1 1 B, are
rectangularly shaped. However, these electrodes and consequently the preunit 10
could assume various other shapes, such as circular, square, etc. An important
feature of the preunit 10 is the flexibility of its design, which enables it to be used in
25 various applications.
Considering now the coating layers 1 19 and 120 in greater detail, the
methods of forming them will be described later. In the preferred embodiment, the
coating layer 119 includes a plurality of microprotrusions, while the coating layer
120 does not include such microprotrusions. It should be understood, however, that
30 the coating layer 120 could alternatively be designed similarly to the coating layer
1 1 g, without departing from the scope of the invention.
1~o 94/07272 PCI /US93/08803
2144657
1 1
Figure 5 is a top plan view of the coating layer 119, which includes an array
of microprotrusions, and which is deposited on the inner face or flat side of the
support material 116. The coating layer 119 is porous with high surface area,
electrically conductive, and relatively thin. The array includes two sets of
microprotrusions. The first set includes a plurality of peripheral microprotrusions
125, and the second set includes a plurality of centrally located microprotrusions
127.
In the preferred embodiment, the peripheral and the central protrusions 125
and 127 are similarly designed, and are generally semi-spherically shaped. However,
other shapes, for example a rectangular shape, are contemplated within the scope of
the present inventiGn. The diameter of each protrusion 125 or 127 is about 6 mill0.01524 cm). Different applications of the device 10 might require that the
microprotrusions 125 and 127 be differently designed. The center-to-center
separation of the peripheral microprotrusions 125 is about 20 mil (0.0508 cm), while
the center-to-center separation of the central microprotrusions 127 is about 40 mil
(0.1016 cm).
One reason for the higher density of the peripheral microprotrusions 125, is
to prevent edge shorting. One reason for the lower density of the central
microprotrusions 127, is to provide separation between the electrodes 111A and
111 B, with minimal masking of the electrode surfaces. For this purpose, the gasket
121 is allowed to cover at least part of the microprotrusions 125, but preferably not
the microprotrusions 127.
The peripheral microprotrusions 125 are adjacently disposed along an outer
periphery of the coating layer 119. While only four rows of microprotrusions areillustrated, it shou!d be understood to those skilled in the art, that additional rows
may added, depending on the size and application of the device 10. The central
microprotrusions 127 are similarly adjacently disposed, in an arrayed arrangement,
within a central section 132 of the coating layer 119. As illustrated in Figure 5, the
central microprotrusions 127 are surrounded by the peripheral microprotrusions 125.
The microprotrusions 125 and 127 are formed on the coating layer 119 to
provide added structural support to the first and second electrodes 111 A and 111 B.
WO 94/07272 PCI /US93/088--
2 ~ ~ ~ 6 5 7
For instance, if, the second electrode 111 B starts to sag or bow toward the first
electrode 111 A, the microprotrusions 127 will prevent contact between these
electrodes 111 A and 111 B.
Figure 5 further shows that the coating layer 119 further includes a plurality
5 of spacings, i.e. 133A through 133G, where the cord, i.e., 117A, are placed, in
order to ultimately form the fill port, i.e.122. As illustrated for large eiectrode sizes
the coved only extends partway into the central section 132. For smaller electrode
sizes the coved across the electrode surface with the two ends protruding out
opposite sides, thus forming simultaneously fill ports 113C and 133D. In this case
10 the width of the cord is smaller than, or equal to the center-to-center separation
between the central microprotrusions 127. However, the cord is larger than the
center-to-center separation between the peripheral microprotrusions 125, to prevent
the peripheral microprotrusions from pinching the cord, and preventing it from being
removed the spacings are increased in the peripheral microprotrusions. Alternatively,
15 the cord may be similar in width to the peripheral microprotrusions separation and no
acc:ommodation in the microprotrusion pattern needs to be done.
Considering now the coating layer 120, it serves a similar function as the
coating layer 119, and is deposited on the side of the electrode 111 B, which faces
the inner side of the first electrode 111 A. In the preferred embodiment, the coating
20 layer 120 does not include microprotrusions. In an alternative embodiment of the
preunit 10, the coating layers 119 and 120 are similarly constructed, and include
microprotrusion layers.
Considering now the gaskets 121 and 123, the methods of producing them
will be described later. The gaskets 121 and 123 are generally identical, and are
25 arranged in registration (adjacent and superposable) with each other. For brevity,
only the gasket 121 will be described in greater detail. The gasket 121 includes a
solid peripheral section 143 and a hollow central section 144.
In the preferred embodiment, the cord 117A, or a part thereof, is placed
between the gaskets 121 and 123, and extends across the hollow section, i.e. 144,
30 of the gaskets, and protrudes outside the peripheral section, i.e. 143. In another
embodiment, the cord does not extend across the entire width of the gaskets, and
~0 94/07272 PCl /US93/08803
q-6~7
only a part of the cord is sandwiched between the gaskets and extends beyond both
edges of one side of the gasket.
Turning now to Figure 1, the next adjacent cell 112 will now be briefly
described. The cell 112 is generally similar in design and construction to the cell
114. The cell 112 includes the bipolar electrode 111 B, as its first electrode, and a
second bipolar electrode 111 C. The electrodes 111 B and 111 C are generally
identical, and are spaced-apart, in registration with each other.
A porous coating layer 131, which is identical to the coating layer 119, is
deposited on the surface of the support material 118, facing the electrode 111 C. A
coating layer 133, which is similar to the coating layer 120, is deposited on a
support material or structure 140, which forms a part of the electrode 111 C.
The cell 112 further includes two gaskets 135 and 137 that are identical to
each other and to the gaskets 121 and 123 of the cell 114. A cord 117B forms a
fill port 142 between the gaskets 135 and 137.
The cell 110 is substantially similar to the cell 114, and includes a first bipolar
electrode 111 Y, a second electrode 111 Z, two gaskets 157 and 159, a cord 117C,a tab 160, and a fill port 162. It should be noted that Figure 3 does not show the
inner electrode 111 Y.
Turning now to Figure 6, there is illustrated a diagrammatic view of a
capacitive circuit 200 which is representative of, and generally has an equivalent
function to the device 1 OA. The circuit 200 illustrates the cell 114 as two
capacitors C1 and C2; the cell 112 as two capacitors C3 and C4; and the cell 110as two capacitors C5 and C6. As a result, the device 10 is generally equivalent to a
plurality of capacitors connected in series.
The porous electrically conducting coating 119 in conjunction with an
ionically conducting medium (not shown) within the cell 114, form the capacitor C1.
The ionically conducting medium and the coating 120 form the capacitor C2. The
coating 131 and the ionically conducting medium within the cell 112 form the
capacitor C3. The ionically conducting medium within the cell 112 and the coating
133 form the capacitor C4. Similarly, the cell 110 is represented by the capacitors
C5 and C6.
~/'O 94/07272 PCI/US93/08~
-?~ 65 7
14
An important aspect of the present invention, is the bipolar configuration of
the energy storage device. The use of a single electrode, such as the electrode
111 B to form two back-to-back capacitors, such as the capacitors C2 and C3, result
in a bipolar electrode B. This design significantly reduces the overall size of the
5 device 10A.
While not wanting to be bound by theory, an explanation of the operation of
the capacitive energy storage device, at the molecular level is helpful to understand
the enormous value of the electric double layer. For simplicity, to describe Figure
14, Figure 3 can be used for reference where the same reference numbers are used10 (and the porous material is a mixed metal oxide).
Figure 14 is a schematic cross-sectional representation of the magnified edge
of the magnified edge of the support 118 & 148A and electrically conducting
coating layers (120,131,133,133B).
The center support 188 is depicted as a metal but can be any material which
15 is electrically conducting and provides the support for the coating. The coating
which has high surface area provides the structure and geometry for the energy
storage. As can be seen in Figure 14, layer 120, etc. has a discontinuous surface
with many fissures, micropore and mesopore which create the high surface area.
Thus, the porous coatings 120 and 131 are coated onto support 118 to
20 form bipolar electrode 111 B and coatings 133 and 133B are coated onto support
148A to form bipolar electrode 111 C. After the preunit 10 is assembled, the pull
tabs are removed creating the fill ports and the preunit 10 is charged with electrolyte
190, the fill ports, e,g.117D, are sealed creating device 1 OA.
The device 1 OA is then charged electrically producing the following results at
25 the same time:
Coating 120 becomes negatively charged. Electrically conducting support
118 conducts electrons accordingly. Thus, porous coating 131 becomes positively
charged. The ionically conducting electrolyte ions accordingly. An electric double
layer is formed at the electrode-electrolyte interface forming the individual capacities
30 in circuit 200. Thus, the surface of coating 133 becomes negatively charged, and
the surface of coating 133B becomes positively charged. Because the porous high
~!0 94/07272 PCr/US93/08803
2l~6
1 5
surface area oxide allows the effective surface area of the electrode to become very
high, the corresponding electrical storage capacity of the device increases
dramatically.
Methods of Manufacturinq the EnerqY Storaqe Device
Referring to Figures 1 to 5, a general description for the preferred method to
produce the dry pre-unit 10 of the energy storage device 1 OA, is as follows:
(A) Support Material Preparation
The support material are optionally etched or cleaned by a variety of
conventional pickling and cleaning procedures.
In some experiments, if the metal surface is not etched it is too smooth. This
smooth surface sometimes causes inadequate adhesion of the porous coating. The
etch creates a suitable rough surface.
1. Wet Etching - A preferred procedure is to contact the metal support
with an aqueous inorganic strong acid, e.g. sulfuric acid, hydrochloric acid,
hydrofluoric acid, nitric acid, perchloric acid or combinations thereof. The etching is
usually performed at elevated temperatures of 50 to 95C (preferably 75C) for
about 0.1 to 5 hr (preferably 0.5 hr) followed by a water rinse. Room temperature
acid etching is possible. An alkaline etch or an oxalic acid etch may also used.2. Dry Etching - The roughened support surface is obtained by sputtering,
plasma treatment, and/or ion milling. A preferred procedure is Ar RF sputter etching
at between around 0.001 and 1 torr with about 1 KeV energy at 13.5 Mhz.
Commonly, 0.1-10 watts/cm2 power densities for about 1-60 min. are used to cleanand roughen the surface. Another procedure is to plasma etch the support with a
reactive gas such as oxygen, tetrafluoromethane, and/or sulfurhexafluoride at around
0.1-30 torr for about 1-60 min.
3. Electrcchemical Etching - The roughened surface is obtained by
electrochemical oxidation treatment in a chloride or fluoride solution.
(B) Coating of Support Material
The coating (e.g. oxide) is porous and composed of mostly micropores
(diameter< 1 7A). Large 0.1-1 ,um wide cracks are present on the surface protruding
to depths as thick as the coating. However, greater than 99% of the surface area
WO 94/07272 - PCI/US93/088~
214~657
1 6
arises from these micropores. The average diameter of these micropores are around
6-12 A.
After various post-treatments the pore structure can be altered to increase
the average pore size. For example, the steam post-treatment creates a bimodal
5 pore distribution. In addition to the micropores, a narrow distribution of mesopores
(diameter< 17-1 000A) having a diameter of about 35 A is created. These treated
electrode coatings have 85-95% of the surface area arising from the micropore
structure.
With alternate electrode construction methods this pore size distribution can
10 be varied. The effective high surface area of the coating is 1,000 to 10,000 to
100,000 times larger than the projected surface area of the electrode as a monolith.
The pore size, distribution, and surface area controlled with the temperature ofpyrolysis and/or high temperature water treatment. In addition, the use of
surfactants to create micelles or other organized structure in the coating solution
increases the average pore size up to the values around 100-200 A with only 5-10%
of the surface area coming from micropores.
As illustrated in Figure 13, the electrode 1 1 1 A includes a porous and
conductive coating layer 119, which is formed on at least one surface of the support
material 1 16. The support material 1 16 is electrically conductive, and sufficiently
20 rigid to support the coating layer 119 and to impart sufficient structural rigidity to
the device 10.
One goal of the present invention, is to optimize the energy density and
power density of the device 10. The object is achieved by reducing the thickness of
the support material 116, and maximizing the surface area of the coating layer 119.
25 The power density of the device 10 is further optimized, by maintaining a low resistance .
The surface area of the coating layer 119 is determined by the BET
methodology, which is well known in the art. The surface enhancement, which is
an indication of the optimization of the surface area of the coating layer 119, is
30 determined according to the following equation:
Surface enhancement = (BET Surface Area / Projected Surface Area)
'094/07272 . 6S7 PCI`/US93/08803
In the present invention, the surface enhancement values can be as large as 10,000
to 100,000.
The coating layer 1 19 is porous, and its porosity could range between about
five percent (5%) and ninety-five percent (95%). Exemplary porosity range for
5 efficient energy storage is between about twenty percent (20%) and twenty-five percent (25%).
In conventional double-layer capacitors, the main device resistance is due to
the carbon coating layer. In the present invention, most of the device resistance is
due to the electrolyte, which has a higher resistance than that of the porous
10 conductive coating layer.
When the preunit device 10 is filled with an electrolyte, it becomes ready to
be charged to become device 1 OA. The main criterion for the electrolyte is to be
ionically conductive and have bipolar characteristics. The boundary or interfaceregion between the electrode and the electrolyte is referred to in the field, as the
15 "double layer", and is used to describe the arrangement of charges in this region. A
more detailed description of the double layer theory is found in "Modern
Electrochemistry", by Bockris et al, volume 2, sixth print, chapter 7 (1977).
The surface area of the coating layer affects the capacitance of the device
1 OA. If for instance, the surface enhancement factor is between 1,000 to 20,000,
20 and the double layer capacitance density is between about 10 to 500 microfarad per
cm2 of the interfacial surface area (i.e. the BET surface area), then surface
enhancement capacitance densities of about 0.1 to 10 farads/cm2 electrode are
obtained .
While the double layer theory is described herein, it should be understood that
25 other theories or models, such as the proton injection model, could alternatively be
used.
The high surface area (porous) electrically conducting coating material is
applied onto the support material.
1. Solution Methods - The porous coating material may originate from
30 various reactive precursors in a solution or a sol-gel composition. Numerous
methods of application of these precursor compositions are feasible; but not limited
WO 94/07272 2 1 ~ ~ ~ 5 7 PCr/US93~088~
,
to the following. A curing, hydrolysis and/or pyrolysis process usually is performed
to form the coating on the support. Pyrolysis of the metal salts is usually done in a
controlled atmosphere (nitrogen, oxygen, water, and/or other inert and oxidativegasses) by means of a furnace and/or an infrared source.
(a) Dip Coating - The electrode or support, is dipped into a solution or sol-gel,
coating the support with a precursor coating, and subsequently cured by pyrolytic
and other methods. Optionally, this process may be repeated to increase layer
thickness. A preferred procedure is to dip the support material in a metal
chloride/alcohol solution followed by pyrolysis at between about 250 and 500 C for
5-20 min in a 5-100% oxygen atmosphere.
This process is repeated until the desired weight of coating is obtained. A
final pyrolysis treatment at 250-450C is done for 1-10 hr. Typically about 1-30mg/cm2 of coating is deposited onto a support for a capacitance density of around 1-
10 F per square centimeter electrode cross-sectional area. Another procedure is to
create a sol-gel solution with ruthenium, silicon, titanium and/or other metal oxides
and coat the support as above. By adjusting the pH, water concentration, solvent,
and/or the presence of additives like oxalic acid, formamide, and/or surfactants the
discharge frequency characteristics of the coating may be adjusted.
High relative humidity during the pyrolysis step can be used to complete the
conversion of starting material to oxide at lower temperatures, a procedure is to
pyrolyze at about 300C without control of humidity. However, an additional
procedure is to maintain the relative humidity above about 50% during this pyrolysis
at temperatures below 350C or below.
A preferred method for dip coating thin (e.g. 1 mil) support structures is to
use a wire frame structure 300 to keep the support material 118 under tension
(Figures 15 and 15A).
The wire frame structure 300 includes at least two (2) wires 301 and 301A
of lengths larger ~han the width of the support material 118. Each wire 301 and
301A includes a single length of wire which is tightly coiled at each end about 360
to form two coils 302 and 303. The coils are wrapped so the ends of the coil arearound 1 cm above the plane of the wire. The coils 302 and 303 are placed through
~0 94/07272 - PCI/US93/08803
~2l~6s7
1 9
holes 304 and 305, respectively, in the support materials. The holes 304 and 305are located at two corners on an adjacent side of the support material.
Two additional wires 301 B and 301 C could be similarly used on the
remaining two sides of the support material to provide additional support.
(b~ Spray Coating - The coating solution is applied to the support by a spray
method, cured, and optionally repeated to increase the thickness. A preferred
procedure is to apply the coating solution to the substrate at a temperature of 0-
150 C by means of an ultrasonic or other spray nozzle with a flow rate of around
0.1-5 ml/min in a carrier gas composed of nitrogen, oxygen and/or other reactive and
iner' gases. The coating characteristics can be controlled by the partial pressure of
oxygen and other reactive gasses.
(c) Roll Coating - The precursor coating is applied by a roll coating
methodology, cured, and optionally repeated to increase the thickness. The coatings
described above for dip coating are usable here.
(d) Spin Coating - A spin coating methodology in the conventional art is used
to apply the precursor coating, and optionally repeated.
(e) Doctor Blading - A doctor blading methodology is used to apply the
precursor coating, and optionally repeated.
2. Electrophoretic Deposition - The porous coating or precursor coating is
applied to the support by electrophoretic deposition techniques, and optionally
repeated .
3. Chemical Vapor Deposition - The porous coating or precursor coating may
be applied by chemical vapor deposition techniques known in the art.
(C) Electrode Pretreatment
It has been found that a number of pretreatments (conditioning) or
combinations thereof are useful to improve the electrical characteristics of thecoating (e.g. electrochemical inertness, conductivity, performance characteristics,
etc.). These treatments include for example:
1. Steam - High temperature water or steam treatment controlled in
atmospheres can be used to decrease the leakage current. A method procedure is to
W0 94/07272 2~ PCI /US93/08
contact the coated electrode with water saturated steam in a closed vessel at
between 1 bO and 325 C for between 1 to 6 hr. under autogenic pressure.
2. Reactive Gas The coated electrode is contacted one or more times
with a reactive gas such as oxygen, ozone, hydrogen, peroxides, carbon monoxide,nitrous oxide, nitrogen dioxide, or nitric oxide at between ambient temperature and
300C at a reduced pressure or under pressure. A preferred procedure is to contact
the coated electrode with flowing ozone at between about 5-20 weight percent in air
at between ambient and 100 C and 0.1-2000 torr pressure for 0.1-3 hr.
3. Supercritical Fluid - The coated electrode is contacted with a supercritical
fluid such as carbon dioxide, organic solvent, and/or water. A preferred procedure is
treatment with supercritical water or carbon dioxide for 0.1-5 hrs by first raising the
pressure then the temperature to supercritical conditions.
4. Electrochemical-The coated electrode is placed in a sulfuric acid
electrolyte and contacted with an anodic current sufficient to evolve oxygen gas and
subsequently with a cathodic current. In one embodiment the electrode is contacted
with 10mA/cm2 in 0.5M sulfuric acid for about 5 min, to evolve oxygen gas. The
electrode is then switched to a cathodic current and the open circuit potential is
driven back to a po~ential of between about 0.5V - 0.75V, preferably between 0.5and 0.6 and more preferably about 0.5 V (vs. NHE) with out hydrogen gas evolution.
5. Reactive liquid-The coated electrode is contacted with an oxidizing liquid
such as aqueous solutions of hydrogen peroxide, ozone, sulfoxide, potassium
permanganate, sodium perchlorate, chromium (Vl) species and/ or combinations
thereof at temperatures around ambient to 100 C for 0.1-6 hr. A preferred
procedure uses a 10-100 mg/l aqueous solution of ozone at 20-50 C for around
0.5-2 hr. followed by an aqueous wash. An additional procedure is to treat the
coated electrode in a chromate or dichromate solution.
(D) Spacing between Electrodes
A number of methods are available to obtain electrical insulation and properly
defined spacing between the electrodes. These methods include, for example:
1. Microprotrusions - The separator 125 and 127 between the coatings
1 19 and 120, includes a matrix of small (in area and height) protrusions, i.e. 125
0 94/07272 PCI /US93/08803
21~4657
and 127, on the surface of at least one side of the electrode. These
microprotrusions may be composed of thermosets, thermoplastics, elastomers,
ceramics, or other electrically insulating materials.
Several methods of applying these microprotrusions are included, but not
5 limited to:
(a) Screen Printing - The microprotrusions are placed on the electrode surface
by conventional screen printing, as described below, in greater detail, under the
heading "SCREEN PRINTING". Various elastomers, thermosets, photo curable
plastics, and thermoplastics are applied in this way. A preferred procedure is to use
10 an acid resistant epoxy or VITON~ solution.
(b) Chemical Vapor Deposition - Microprotrusions are also placed on the
electrode surface by depositing silica, titania and/or other insulating oxides or
materials through a mask.
(c) Photolithography - Microprotrusions are also produced by means of a
15 photolithographic methods, as is described later, in greater detail, under the heading
"PHOTOLITHOGRAPHIC PRODUCTION OF MICROPROTRUSIONS".
2. Physically thin separator sheet - The separator between the electrodes
is a thin, substantially open structure material such as glass. A preferred material is
0.001 -0.005 in (0.00254 to 0.01270 cm) porous glass sheet available from
20 Whatman Paper Ltd located in Clifton, NJ.
3. Casting a separator - The separator between the porous material is
also obtained by casting a thin, substantially open structure film such as for example
NAFION~, polysulfones, or various aero- and sol-gels.
4. Air space - The separator between the electrodes is also an air space
25 which is subsequently occupied by the non-aqueous or aqueous electrolyte.
(E) Gasketing
The materials used for the gaskets, such as the gaskets 121, 123, 135, 137,
157 and 159, at the edge of the active electrode surface include any organic
polymer which is stable in the electrical/chemical environment, and to the processing
30 conditions. Suitable polymers include, for example polyimide, TEFZEL~, polyethylene
(high and low density), polypropylene, other polyolefins, polysulfone, KRATON~
WO 94/07272 PCI/US93/088
21~5~
other fluorinated or partly fluorinated polymers or combinations thereof. The gasket
may be applied as a preformed material, screen printed, or by other methods.
(F) Cord for Fill Port
The cord (11 7A, 1 1 7B and 11 7C) for the creation of the fill ports, such as
the fill ports 122 and 142, is of any suitable material having some specific
properties, e.g., it is different from the gasket materials, has a higher melting
temperature (Tm) than the gasket material, and does not melt, flow or adhere to the
gasket material under the heating conditions described herein. Generally, glass,metal, ceramic, and organic polymers or combinations thereof are used.
(G) Stacking
A stack is created by starting with an endplate and alternating gasket
material, cord, electrode, gasket, cord electrode until the desired number of cells are
created finishing with a second endplate, and optionally with a gasket material on
the top outside of the stack.
(H) Assembling (heating and cooling)
The stack is heated under pressure to cause reflow of the gasket material,
adhering and sealing the perimeter of the electrode materials to the adjacent
electrode in the stack; thereby, creating isolated cells and an assembled stack unit.
This is done in an inert atmosphere.
(a) Radio Frequency Induction Heating (RFIH) is used to heat the stack to
cause reflow of the gasket material.
(b) Radiant Heating (RH) is used to uniformly heat the stack to cause reflow
of the gasket material. A preferred method is to use 1-100,um radiation at 0.5-10
watts/cm2 for 1-20 min.
(c) Conductive and/or convective heating in a furnace, optionally in an inert
atmosphere, is used to heat the stack to cause reflow of the gasket material.
(I) Creating the fill port
The cords are pulled to remove them from the assembled unit to
create a dry preunit having at least one fill port per a cell.
(J) Post-Conditioning
0 94/07272 PCI/US93/08803
' 2.1~
1. A number of post-conditioning reactive gas treatments of the stack or
assembled stack or combinations thereof are useful to improve the overall and long
term electrical characteristics of the electrode and resulting device. These include
either before step (H) and/or after step (I) treatment with hydrogen, nitric oxide,
5 carbon monoxide, ammonia, and other reducing gasses or combinations thereof atbetween ambient temperature and the Tm of the gasket material at a reduced
pressure or under pressure.
2. A second post conditioning commonly done is to adjust the open
circuit potential of the electrode after step (F) and stack the electrode in an inert
10 atmosphere (e.g. Ar, N2). This is done by using a cathodic current without hydrogen
evolution .
(K) Filling of Dry Preunit
The dry preunit is filled with an ionically conducting aqueous or non-aqueous
electrolyte .
A preferred electrolyte is approximately 30% sulfuric acid in water due to the
high conductivity. Non-aqueous electrolytes based on propylene and ethylene
carbonate are also used to obtain larger than 1 .2V/cell potentials.
A preferred procedure for filling the dry preunit with liquid electrolyte is to
place the preunit in a chamber, evacuate the chamber between about 1 torr to 1
microtorr, preferably about 250 mtorr to less than 1 torr, and introduce the
electrolyte; thereby, filling the cell gaps with electrolyte through the fill ports.
Alternatively, the preunit may be placed in the electrolyte and a vacuum pulled;thereby causing the gas in the cell gaps to be removed and replaced by the
electrolyte .
In addition, non liquid based electrolytes (e.g. solid and polymer) may be
used. In those cases the electrode is coated with the electrolyte before reflow and a
fill port is not required.
(L) Sealing of Fill Ports
The fill ports are sealed by reflowing an additional film of polymer the same ordifferent over the openings to create a sealed device. This is commonly done with
an induction heater, which locally heats the film over the fill port opening.
WO 94/07272 'i ~ 4A ~' PCIrt US93/088
24
~M) Burn-ln
The device is brought to full charge usually by charging the device in 0.1
V/cell steps at a charging current of about 4 mA/cm2.
(N) Testing
Termination Methods -- Several methods are used to make electrical
connections to the ultracapacitor endplates, and are described below.
1. Endplate Tabs (160 and 1 60A) - The endplates (1 1 1 A and 1 1 1 Z)
themselves have been cut to extend out beyond the normal gasket perimeter. Theseextensions allow attachment of a wire or ribbon. Typically, the extension is a stub
from which all oxide material is removed down to the bare support material; 5 mil
(0.0127 cm) thick nickel ribbon is spot welded to the stub.
2. Silver Epoxy - The oxide coating is removed from the exposed faces of
the endplates or the endplates may be coated only on one side. Clean nickel foilleads or copper plates make electrical connection to the exposed faces by bonding
them together with a conductive silver epoxy. Optionally, the oxide coating is
present .
3. Lugs - Threaded titanium nuts are welded to the thick titanium
endplates before coating. Electrical connection to the titanium nuts is achieved by
screw attachment.
4. Press Contacts - The oxide is removed or the endplates may be coated
only on one side from the exposed side of the endplates before assembly into thedevice stack. The bare support material e.g. titanium, is reverse sputtered to clean
the surface, being careful not to overheat the substrate. The clean surface is then
sputtered with titanium to lay down a clean adhesion layer, followed by gold. The
gold acts as a low contact resistance surface to which electrical contact can bemade by pressing or by wire bonding.
5. Deposition of a compatible medium such for example aluminum, gold,
silver, etc. outside by CVD or other means.
The device resistance is measured at 1 kHz. The device capacitance is
determined by measuring the coulombs needed to bring the device to full charge at a
0 94/07272 PCI /US93/08803
2i~4657
charging rate of around 4 mA/cm2 of electrode area. Leakage current is measured as
the current needed to maintain a full charge after 30 min. of charging.
These devices may be made in various configurations depending on the
desired application. By adjusting the device voltage, cell voltage, electrode area,
5 and/or coating thickness in a rational manner, devices made to fit defined and predetermined specifications are constructed.
The electrode capacitance density (C' in units of F/cm2) is roughly 1 F/cm2 for
every 10,um of coating. Therefore, for large capacitance values a thicker coat is
used. The device capacitance (C) is equal to the electrode capacitance density times
10 the electrode area (A in units of cm2) divided by two times the number of cells (n)
(equation 1).
The leakage current (i") is proportional to the electrode area, while the
equivalent series resistance (ESR) is inversely proportional to the electrode area (eqn.
2). Typical values for i" are less than 20,uA/cm2.
The total number of cells in a device (n) is equal to the cell voltage (V')
divided by the total device voltage (V) (eqn. 3). Cell voltages up to about 1.2 V can
be used with aqueous based electrolytes.
The device height (h), based on a cell gap (h') and a support thickness (h"), isdetermined from the number of cells and the electrode capacitance density in units
20 of cm by equation 4.
The device ESR is a function of the number of cells times the cell gap (h')
times the resistivity of the electrolyte (r) times a factor of about 2 divided by the
area (equation 5).
eqn. 1 C=C'A/2n
eqn. 2 i" a A a 1 /ESR
eqn. 3 n = V/V'
eqn. 4 h/cm = n(0.002C'+ h' + h")
eqn. 5 ESR 2 2nh'r/A
Devices are constructed to meet the requirements of various applications by
30 considering the voltage, energy, and resistance requiremen~s. The following
examples are not meant to be limiting in any way:
WO 94/07272 PCI`/US93/088--
' 2~ 'S~
26
For electric vehicle applications about a 100 KJ to 3 MJ device is used. A
large voltage (about 100 to 1000 V) large energy (1-5 F/cm2) storage device is used
with an electrode area of about 100 to 10,000 cm2.
For electrically heated catalyst applications for reduction of automobile cold
start emissions about a 10 to 80 KJ device is used. This device is about 12 to 50 V
constructed with around 100 to 1000 cmZ area electrodes of 1 -5 F/cm2. Optionally,
a device consisting of several devices in parallel can be constructed to meet the
electrical requirements.
For defibrillator applications about a 200-400 V device with 0.5 to 10 cm2
area electrodes of 1 -3 F/cm2 are used.
For uninterruptable power source applications various series/parallel device
configurations may be used.
Screen Printin~
Considering now a screenprinting method 250, with respect to Figures 7 and
8, the focus of the method 250, is to produce a series of microprotrusions 125 and
127 on the surface of the coating layers, to act as a space separator in electrical
storage devices, such as a capacitor or a battery, in general, and in the dry preunit
energy storage device 10, in particular.
The substrate is usually a thin metal such as titanium, zirconium, or alloys
thereof. The substrate is usually in the shape of a thin metal plate as is conventional
in the capacitor art.
The substrate is coated on one or both sides with a porous carbon compound
or a porous oxide coating. This step is accomplished by methods conventional in the
art. The oxide coating serves as the charge storage area for the device.
Alternately, a stacked set of battery electrodes (e.g., lead for lead acid) or
electrolytic capacitor electrodes (e.g., alumina and tantalum) may be fabricated.
It is important that the flat surfaces of adjacent coated substrates or
electrodes do not contact each other and further be of a uniform separation. Theepoxy microprotrusions accomplish the desired uniform separation.
Sampie Holding -- The coated thin flat substrate needs to be secured (or
held), so that the formation of the microprotrusions is precise and accurate on the
0 94/07272 PCI/US93/08803
21446s7
flat surface of the substrate. For thin metal sheets (0.1 to 5 mil (0.000254 to
0.0127 cm), especially about 1 mil (0.00254 cm)) an electrode holder 275 is
particularly important. If a strong vacuum is pulled on a thin sheet, often reverse
dimples are formed in the thin sheet which cause significant undesirable changes in
the physical and electrical properties of the final device.
The electrode holder 275 includes a porous ceramic holder 276, which is
useful because the pore size is small enough that the dimples do not appear when a
mild or stronger vacuum is pulled. The flat ceramic surface of the ceramic holder
276 must be in intimate contact with the surface of the electrode 111 A, under
conditions which do not deform the metal or disrupt the coating present. The
vacuum used with the porous ceramic is at least 25 in mercury. Preferably the
vacuum is between about 25 and 30 in., especially 26 and 29 in.
Further, the ceramic substrate needs to be flush with the surface of any
mechanical holder to assure that uniform extrusion of the epoxy through the screen
openings occurs. Flush in this context means that the flat surface of the holder and
the surface of the coating for electrical storage differ from each other by between
about + 5 mil (0.0127 cm) deviation or less from level per 6 linear in.
The electrode holder 275 further includes a metal frame 277, which should
also be as flush (flat) as possible so that uniformly sized protrusions are formed from
one end of the electrode to the other.
The electrode holder 275 can be purchased from a number of commercial
sources for example from Ceramicon Designs, Golden, Colorado. Alternatively, thesample holder 276 can be manufactured using commercially available metals, alloys
or ceramics.
Usually, a 5 in (12.7 cm) by 7 in (17.78 cm) coated sheet electrode is
formed.
The metal holder 277 has a plurality of strategically located pins, such as the
three pins 278, 279 and 280, which are used to align and position the electrode
111A, using a plurality of corresponding holes 281, 282 and 283, respectively. The
holes 281, 282 and 283 are usually as close to the peripheral edges of the electrode
WO 94/07272 PCr/US93/08~
"2f~6~7
. . ~
28
1 1 1 A, as possible to conserve useful electrode surface. Alternatively, no aiignment
holes are used, and the pins are used to align the electrode edges.
A stencil (not shown) having a predetermined open pattern, is stretched and
secured in a conventional screen printing frame (not shown). The screen mesh is
removed.
The epoxy components are mixed and the fluid epoxy is placed on the surface
of the stencil, then spread to obtain an even applied coat. This can be accomplished
using a pressure bar, doctor bar or a squeegee.
Usually, constant temperature and humidity are important to obtain an even
1 0 coat.
The stencil is then carefully removed leaving the fluid epoxy protrusions on
the surface of the oxide. The epoxide protrusions are then cured using ambient,
accelerated heat at from between 100 to 1 50C. or light.
The electrode having microprotrusions is then combined with other
electrodes, and assembled in a wet process or a dry process. If a dry process isused, the dry unit 10 is then back filled with electrolyte, when it is to be charged.
It is important that the cured epoxy does not react with the liquid electrolyte
eventually used in the fabrication of the capacitor having multiple layers of
electrodes .
The cured microprotrusions then perform their function by keeping the
spacing between the electrodes uniform.
As can be seen from Figure 6, the edges of the flat surface of the electrode
have protrusions 125 that are closer together than those protrusions 127 in the
active or central portion of the electrode. These protrusions 125 increase the
support at the edges to maintain uniform separations. Alternatively, bars may beused.
It is apparent that from these teachings the following are possible:
Increasing or decreasing the substrate electrode thickness will allow an
increase or decrease in the microprotrusion spacing due to changes in support
rigidity.
0 94/07272 PCl /US93/08803
57
29
Other thermosets, thermoelastomers, or photo curable epoxies or epoxy
derivatives conventional in the art can be used.
- Other microprotrusion pattern elements can be used such as squares, lines,
crosses, etc. Specifically, bars on the perimeter can add mechanical support.
Optionally the screen may be heated, if necessary to bring the resin flowable
epoxy to a temperature when its viscosity becomes suitable for printing for a short
time.
This heating step followed by screen printing of the flowable epoxy resin
must be performed quickly because the working time for the epoxy is significantly
1 0 reduced.
The electrical storage devices produced having the microprotusions 125 and
127 are useful as batteries, capacitors and the like.
Photolitho~raPhic Production of MicroProtrusions
The focus of the present method is to produce a series of microprotrusions on
the surface, or alloys of the electrode substrate, using photolithography, with
respect to Figures 10, 1 1 and 12. The substrate is usually in the shape of a thin
metal plate as is conventional in the capacitor art.
A photo resist film 381 is applied to the surface of the electrode 111A, either
by vacuum lamination using the commercially available Dynachem ConforMASK film
applicator, and Dynachem vacuum applicator Model 7241730, or by passing the
photo resist film 381 and electrode 111A through a pair of heated rollers 384 and
385.
Exposure is done using a standard 1-7kW UV exposure source, such as
mercury vapor lamps 389.
The ConforMask film applicator is developed using standard conditions such
as 0.5-1.0% sodium or potassium carbonate monohydrate in either a developing
tank or a conveyorized aqueous developer. Optionally, after developing the electrode
with microprotrusions may be neutralized in a dilute 10% sulfuric acid solution. This
removes all the unwanted unreacted film to leave the reacted microprotrusions
adhered to the electrode surface.
W094/07272 44~j5~ PCI/US93/08~
To obtain optimum physical and electrical performance properties the
resulting material is put through a final curing process involving both UV irradiation
and thermal treatment utilizing conventional UV curing units and convection air
ovens.
The multiple electrodes are assembled to produce for instance a capacitor, as
described above. The microprotrusions accomplish the desired uniform separation.COMMERCIAL APPLICATIONS
The energy storage device 10A has a multitude of applications, as a primary
or back up power supply, and/or as a capacitor. The size is from 0.1 volt to
1 0C,000 volts or 0.1 cm3 to 1 o6 cm3. Typical voltage ranges may include
combinations of uses in automotive and other applications.
Among these applications are the following:
TYPI CAL TYPI CAL
Automobile APPlications Volt RANGE SlZE(cm3)
Airbags & Seat Restraints 1-100 1-1000
Seat Warmers 1-100 1-100
Electronically Heated Catalyst 1 -1 000 1 -1 ,000,000
Electric Vehicle Propulsion 100-1000 100-1,000,000
Hybrid Electric Vehicle Propulsion 1-1000 10-100,000
Internal Combustion/Ultra
Capacitor Propulsion 1- 1000 100- 100,000
Power Steering 1-1000 1-100
Regenerative Braking/Shock Absorption 1-1000 5-100
Starting Lighting and Ignition with battery 1-1000 2-100
Starting Lighting and Ignition stand alone 1-100 1-100
Medical APplications
Cardiac Defibrillators 10-500 0.1-100
Pacemakers 1-300 0.1-300
Neuro stimulators and like 0.1-300 0.1-300
Implantable and external devices 0.1-300 0.1-300
Surgical Power Tools 10-700 1-10
0 94/07272 PCI /US93/08803
21~65~
TYPICAL TYPICAL
Medical APPlications (cont) Volt RANGE SlZE(cm3)
Ambulatory Monitoring Equipment 1-100 1-100
Automatic Liquid Chromatography 1-100 1-20
Automated Clinical Lab Analysis 1-100 1-20
Computerized Tomography (CT) Scanners 1-1000 1-100
Dental Equipment 1-200 1-10
Digital Radiography Equipment 1-500 1-1000
Electrosurgical Instruments 1-200 1-10
Fiberoptics 1-100 1-100
Examination Scopes 1-100 1-10
Hearing Aids 1-10 0.1-1.0
Infusion Devices 1-100 0.1-10
Magnetic Resonance Imaging (MRI) 1-1000 1-1000
Nuclear Medical Diagnostic Equipment 1-1000 1-100
Electric Patient Monitoring Systems 1-200 1-100
Respiratory Therapy Equipment 1-500 1-100
Surgical Lasers 1-1000 1-1000
Electric Surgical Support Systems 1-100 1-1000
Ultrasonic Diagnostic Equipment 1-100 1-100
Mobile ProPulsion SYstems
Fork lifts 1-1000 100-10,000
Golf carts 1-1000 100-10,000
Farm Implements Trains Subways 1-1000 100-100,000
Regenerative Braking 1-1000 1-100
Business/Commercial Electronics APPlications
Calculators 1-120 0.5-10
Cellular Telecommunications 1- 120 1- 100
Commercial Audio Amplifiers 1-1000 1-10
Commercial Flash/Strobe Lights 1-1000 1-10
Commercial Power Tools 1 -1 000 1 -1 00
WO 94/07272 2 1 4 4 ~ 5 7 PCI/US93/08~
TYPICAL TYPICAL
Business/Commercial Electronics (cont) Volt RANGE SlZE(cm3)
Commercial Video Cameras 1 -120 1 -10
Computers 1 -120 1 -10
Copiers 1 -120 1 -10
Dictation Equipment 1-100 1-1000
Electric Motors 1-1000 1-1000
Electronic Locks 1-120 1-10
Electronic Organizers/PDAs 1-100 1-5
Emergency Lighting Systems 1-440 1-1000
Facsimile Equipment 1-120 1-10
Microphones 1 -120 1 -3
Pagers 1 -120 1 -2
Printers 1 -120 1 -10
Security Systems 1 -120 1 -100
Slide Projectors 1-120 1-100
Uninterruptible Power Supplies 1 - 1000 1 -100,000
Surge Protectors 1-1000 1-100,000
Wireless Networks 1-1000 1-1000
20 Consumer Electronics APPlications
Audio Systems:
Compact/Home 1 -120 1 -10
Portable Tape/CD 1 -120 1 -5
Walkman/Personal Stereo 1-120 1-5
CB Radios 1-120 1-10
HAM Radios 1-120 1-100
Camcorders 1 -120 1 -10
Home Satellite Dishes 1-120 1-10
Microphones 1-120 1-3
Monitors and Cathode Ray Tubes 1 -1000 1 -100
Photo Flash 1-1000 1-3
094/07272 2144~-~7 PCI/US93/08803
33
TYPICAL TYPICAL
Consumer Electronics Ap~lications (cont) Volt RANGE SlZE(cm3)
Receivers, Transceivers 1 -1000 1 -10
Telephone answering devices 1 -120 1 -5
Cellular, cordless phones 1-120 1-3
Toys & Games 1-120 1-10
Television sets 1 -1000 1 -10
Home 1 -1000 1 -10
Portable 1 -1000 1 -10
VCRs 1 -120 1 -10
Video Disk Players 1 - 120 1 - 10
Video Games 1-120 1-10
Watches/Clocks 1 -120 1 -100
Consumer Electric Housewares A~lications
Air Purifiers 1-120 1-100
Bag Sealers 1-500 1-100
Blenders 1 -120 1 -10
Clocks-Total 1 -120 1 -100
Alarm & Desk 1-120 1-10
Coffee Grinders 1-120 1-10
Coffee Makers 1-120 1-10
Convection Ovens 1-1000 1-1000
Corn Poppers 1 - 120 1 - 10
Curling Irons/Brushes 1-120 1-5
Deep Fryers 1-230 1-100
Electric Blankets 1-120 1-10
Flashlights 1 -100 1 -10
Floor Polishers 1-220 1-100
Food Processors 1-120 1-10
Hair Dryers 1-120 1-5
Heating Pads 1-120 1-5
WO 94/07272 ~ PCr/US93/08~
21~46~7
34
(Cont) TYPICAL TYPICAL
Consumer Electric Housewares APPiications Volt RANGE SlZE(cm3)
Home Security Systems 1-120 1-100
Irons 1-120 1 -5
Knives 1 -120 1 -3
Massagers 1-120 1 -5
Mixers 1-120 1-5
Microwave Ovens 1-230 1-10
Power Tools 1 -230 1 -100
Security Systems 1-230 1-100
Shavers 1 -120 1 -3
Smoke Detectors 1-120 1-5
Timers 1-120 1 -3
Toasters/Toaster Ovens 1 -120 1 -5
Toothbrushes (Electric) 1-120 1-3
Vaporizers 1 -120 1 -10
Water Pulsators 1 -120 1 -10
Whirlpools (Portable) 1-120 1-100
Consumer Maior APPliances
Compactors 1 -120 1 -10
Dishwashers 1 -220 1 -100
Dryers 1 -120 1 -100
Freezers 1 -220 1 -100
Ranges 1 -220 1 -1000
Refrigerators 1 -120 1 -100
Washers 1 -220 1 -100
Water Heaters 1-220 1-100
Outdoor APPliances
Bug Killers 1-120 1-10
Outdoor Grills 1-120 1-100
Power Mowers 1-220 1-100
~0 94/07272 2 1 1 ~ 6 5 7 PCr/US93/08803
TYPICAL TYPICAL
Outdoor APPliances (cont) Volt RANGE SlZE~cm3)
Riding Mowers 1-1000 1-1000
Riding Tractors 1 -1000 1 -10,000
Rotary Tillers 1 -1000 1 -10,000
Snow Plows/Blowers 1-220 1-1000
Weed Trimmers 1-220 1-100
Other APPlications
Electro-expulsive Deicing 1-1000 1-100
Electronic Fuses 1 - 1000 1 - 10
Lasers 1 -1000 1 -100
Phased-Array radar 1 -1000 1 -1000
Rail Gun 1-1000 1-10,000
Multiple devices are placed is series and/or parallel for specific applications to
achieve desired performance.
Fabrication of Dry Preunit
The following examples are presented to be descriptive and explanatory only.
They are not to be construed to be limiting in any manner.
EXAMPLE 1
Fabrication of A Drv Preunit
(A) Coatin~ Method
The support structure is prepared by etching a 1 mil (0.00254 cm) titanium
sheet with 35% HNO3/1.5 % HF at 60C for 5 min. The end plates are 5 mil
25 (O .0127 cm) titanium .
The oxide coating solution is 0.2 M ruthenium trichloride trihydrate and 0.2 M
niobium pentachloride in tert-butanol (reagent grade).
The etched Ti sheets are dip-coated by immersion into the solution at ambient
conditions. The coated sheet is submerged into the solution, held for about 1 sec
30 and then removed.
WO 94/07272 ~ PCI/US93/08~
2i44657
36
After each coating, the oxide is dried at 70- C for 10 min, pyrolyzed at
350C for 10 min and removed to cool to ambient temperature all in ambient
atmosphere .
The dip-coating steps are repeated for 10 coats (or any desired number)
5 rotating the Ti sheet so as to dip with alternate sides down. A thickness of about
ten microns is achieved.
The fully coated sheet is final annealed at 350-C for 3 hrs in ambient
atmosphere .
(B) Electrode Pretreatment
The coated electrode is contacted with saturated steam in a closed vessel at
280 C for 3 hrs under autogenic pressure.
(C) SPacin~
Microprotrusions are screen printed on one side of the electrode, as described
below, in greater detail, under the heading "SCREEN PRINTING". The epoxy
compound is EP21AR from Masterbond, of Hackensack, New Jersey.
The epoxy protrusions are cured at 150 C for 4 hr. in air. The coated
electrodes are next die-stamped to the desired shape.
(D) Gasket
A modified high density polyethylene (HDPE, improved puncture resistance
and adhesion) 1.5 mil (0.00381 cm) thick by 30 mil (0.0762 cm) wide with outsideperimeter the same as that of the electrode is placed on the electrodes on the same
side as the microprotrusions and impulse heat laminated. The HDPE is grade PJX
2242 from Phillips-Joanna of Ladd, lllinois.
(E) Cord
One cord (Dupont T2 TEFZEL~ film 90ZM slit in machine direction) 0.9 mil
(0.00229 cm) thick by 10 mil (0.0254 cm) wide is placed across the narrow
dimension of the gasket and electrode surface and aligned between
microprotrusions. The location of the cord is one of three positions centered, left of
center, or right of center.
A second HDPE gasket is placed on the first gasket sandwiching the cord
between the two gaskets.
~0 94/07272 21 ~ 4 6 ~ 7 PCI/US93/08803
37
The second gasket is impulse heated to adhere to the first gasket and to fix
the cord in place.
- (F) Stackin~
Electrode/microprotrusion/gasket/cord/gasket units are stacked in a non-
5 metallic ~ceramic) alignment fixture beginning with a 5 mil (0.0127 cm) end plateunit to the desired number of cells and ending with a plain 5 mil (0.0127 cm) end
plate with the cords arranged such that the location is staggered-left, center, right in
a three unit repeating cycle (end perspective). Light pressure is applied to the top of
the stack through a ceramic piston block to maintain uniform alignment and contact
10 throughout the stack.
(G) Reflow
A radio frequency induction heater (2.5 kW) is used to heat the stack. The
stack was placed centrally in the three turn, 3 in (7.62 cm) diameter coil and heated
for 90 seconds at a power setting of 32 %. The fused unit is allowed to cool to
15 ambient temperature.
(H) Cord Removal
The cords are removed by carefully pulling the exposed ends of the cord to
leave the open fill ports.
EXAMPLE 2
Alternative Fabrication of DrY Preunit
(A) Coatin~ Method
The support structure is prepared by etching a 1 mil (0.00254 cm) titanium
sheet with 50% HCI at 75C for 30 min. The end plates are 2 mil (0.00508 cm)
titanium.
The oxide coating solution is 0.3 M ruthenium trichloride trihydrate and 0.2 M
tantalum pentachioride in isopropanol (reagent grade).
The etched Ti sheets are dip-coated by immersion into the solution at ambient
conditions. The coated sheet is submerged into the solution, held for about 1 sec
and then removed.
After each coating, the oxide is dried at 70 C for 10 min. in ambient
atmosphere, pyrolyzed at 330 C for 15 min in a 3 cubic feet per hrs. flow of 50 vol.
WO 94/07272 2 1 4 4 ~ ~ 7 Pcr/usg3/o88
% oxygen and 50 % nitrogen, and removed to cool to ambient temperature in
ambient atmosphere.
The dip-coating steps are repeated for 30 coats (or any desired number)
rotating the Ti sheet so as to dip with alternate sides down.
The fully coated sheet is final annealed at the above conditions for 3 hr.
(C) sPacinq
VITON~ microprotrusions are screen printed on one side of the electrode, as
described below, in greater detail, under the heading "Vll. SCREEN PRINTING".
The VITON~ protrusions are cured at 150C for 30 min. in air. The coated
electrodes are next die-stamped to the desired shape.
(D) Gasket
A modified high density polyethylene (HDPE, improved puncture resistance
and adhesion) 1.0 mil (0.00254 cm) thick by 20 mil (0.0508 cm) wide with outsideperimeter the same as that of the electrode is impulse heat laminated to both sides
of the electrode. The HDPE is grade PJX 2242 from Phillips-Joanna of Ladd, lllinois.
(E) Cord
One cord, 1 mil (0.00254 cm) diameter TEFLON~ coated tungsten wire is
placed across the narrow dimension of the gasket and electrode surface and aligned
between microprotrusions. The location of the cord is one of three positions
centered, left of center, or right of center.
(F) Stackin~
Electrode/microprotrusion/gasket/cord/gasket units are stacked beginning with
a 2 mil (0.00508 cm) end plate unit to the desired number of cells and ending with a
plain 2 mil (0.00508 cm) end plate with the cords arranged such that the location is
staggered-left, center, right in a three unit repeating cycle (end perspective). (G) Reflow
The HDPE gasket is reflowed in nitrogen at 125 - C for 120 min to reflow the
thermoplastic. The unit is cooled in nitrogen to ambient temperature.
(H) Cord Removal
The cords are removed by pulling the exposed ends to leave the open fill
ports.
~ 094/07272 . ~. PCT/US93/08803
~ `, ~ .
2~44657
39
EXAMPLE 3
Alternative Fabrication of Dry Preunit
(A) Coatin~ Method
The support structure is prepared by etching a 1 mil (0.00254 cm) titanium
sheet with 50% Hcl at 75C for 30 min. The end plates are 10 mil (0.0254 cm)
titanium .
The oxide coating solution is 0.2 M ruthenium trichloride trihydrate and 0.2 M
tantalum pentachloride in isopropanol (reagent grade).
The etched Ti sheets are dip-coated by immersion into the solution at ambient
conditions. The coated sheet is submerged into the solution, held for about 1 sec
and then removed.
After each coating, the oxide is dried at 70 C for 10 min, pyrolyzed at
300C for 5 min and removed to cool to ambient temperature all in ambient
atmosphere.
The dip-coating steps are repeated for 10 coats (or any desired number)
rotating the Ti sheet so as to dip with alternate sides down.
The fully coated sheet is final annealed at 300C for 3 hrs in ambient
atmosphere .
(B) Electrode Pretreatment
The coated electrode is contacted with saturated steam in a closed vessel at
260-C for 2 hrs under autogenic pressure.
(C) SDacin~
Microprotrusions are screen printed on one side of the electrode, as described
below, in greater detail, under the heading "SCREEN PRINTING". The epoxy
compound is grade EP21AR from Masterbond, Hackensack, NJ.
The epoxy protrusions are cured at 150 C for 4 hr. in air. The coated
electrodes are next die-stamped to the desired shape.
(D) Gasket
A modified high density polyethylene (HDPE, improved puncture resistance
and adhesion) 1.5 mil (0.00381 cm) thick by 30 mil (0.0762 cm) wide with outsideperimeter the same as that of the electrode is placed on the electrodes on same side
WO 94/07272 4~5~ PCr/US93/08~
as the microprotrusions and impulse heat laminated. The HDPE~ is grade PJX 2242
from Phillips-Joanna of Ladd, lllinois.
(E) Cord
One cord ( l tt~tL~) 1 mil (0.00254 cm) thick by 10 mil (0.0254 cm) wide is
placed across thei narrow dimension of the gasket and electrode surface and aligned
between microprotrusions. The location of the cord is one of three positions
centered, left of center, or right of center.
A second HDPE~ gasket is placed on the first gasket sandwiching the cord
between the two gaskets.
The second gasket is impulse heated to adhere to the first gasket and to fix
the cord in place.
(F) Stackinq
Electrode/microprotrusion/gasket/cord/gasket units are stacked beginning with
a 10 mil (0.0254 cm) end plate unit to the desired number of cells and ending with a
plain 10 mil (0.0254 cm) end plate with the cords arranged such that the location is
staggered-left, center, right in a three unit repeating cycle (end perspective). (G) Reflow
The gasket is reflowed in nitrogen at 160 C for 45 min to reflow the
thermoplastic. Tne unit is cooled in nitrogen to ambient temperature.
(H) Cord Removal
The cords are removed by carefully pulling the exposed ends to leave the
open fill ports.
EXAMPLE 4
Alternative Fabrication of Dry Preunit
(A) Coatinq Method
The support structure is prepared by etching a 1 mil (0.00254 cm) titanium
sheet with 50% HCI at 75C for 30 min. The end plates are 5 mil (0.0127 cm)
titanium .
The oxide coating solution is 0.2 M ruthenium trichloride trihydrate and 0.2 M
Ti(di-isopropoxide)bis 2,4-pentanedionate in ethanol (reagent grade).
`.'094/07272 - PCT/US93/08803
21~657
- The etched Ti sheets are dip-coated by immersion into the solution at ambient
conditions. The coated sheet is submerged into the soiution, held for about 1 sec
and then removed.
After each coating, the oxide is dried at 70C for 10 min, pyrolyzed at
5 350 C for 5 min in oxygen and removed to cool to ambient temperature ali in
ambient atmosphere.
The dip-coating steps are repeated for 30 coats (or any desired number)
rotating the Ti sheet so as to dip with alternate sides down.
The fully coated sheet is final annealed at 350C for 3 hrs in an oxygen
10 atmosphere.
(C) SPacin~
Microprotrusions are thermally sprayed through a mask on one side of the
electrode. The thermal spray material is TEFLON~ from E.l. DuPont de Nemours &
Co., Wilmington, Delaware.
The TEFLON~ protrusions are cured at 300 C for 0.5 hr. in air. The coated
electrodes are next die-stamped to the desired shape.
(D) Gasket
A modified high density polyethylene (HDPE, improved puncture resistance
and adhesion) 1.5 mil (0.00381 cm) thick by 30 mil (0.0762 cm) wide with outsideperimeter the same as that of the electrode is placed on the electrodes on same side
as the microprotrusions and impulse heat laminated. The HDPE is grade PJX 2242
from Phillips-Joanna of Ladd, lllinois.
(E) Cord
One cord ( I tF/tL~) 1 mil (0.00254 cm) thick by 10 mil (0.0254 cm) wide is
placed across the narrow dimension of the gasket and electrode surface and aligned
between microprotrusions. The location of the cord is one of three positions
centered, left of center, or right of center.
A second HDPE~ gasket is placed on the first gasket sandwiching the cord
between the two gaskets.
The second gasket is impulse heated to adhere to the first gasket and to fix
the cord in place.
WO 94/07272 2 1 ~ 4 6 57 PCI'/US93/088
:, .
42
(F) Stackinq
Electrode/microprotrusion/gasket/cord/gasket units are stacked beginning with
a 5 mil (0.0127 cm) end plate unit to the desired number of cells and ending with a
plain 5 mil (0.0127 cm) end plate with the cords arranged such that the location is
5 staggered-left, center, right in a three unit repeating cycle (end perspective).
(G) Reflow
The gasket is reflowed in nitrogen at 190C for 30 min. to reflow the
thermoplastic. The unit is cooled in nitrogen to ambient temperature.
(H) Cord Removal
The cords are removed by carefully pulling the exposed ends to leave the
open fill ports.
EXAMPLE 5
Alternative Fabrication of DrY Preunit
(A) Coatinq Method
16 The support structure is prepared by etching a 0.8 mil (0.002032 cm) zirconium
sheet with 1 %HF/20% HN03 at 20 C for 1 min. The end plates are 2 mil (0.00508
cm) zirconium.
The oxide coating solution is 0.2 M ruthenium trichloride trihydrate and 0.1 M
tantalum pentachloride in isopropanol (reagent grade).
The etched Ti sheets are dip-coated by immersion into the solution at ambient
conditions. The coated sheet is submerged into the solution, held for about 1 sec
and then removed.
After each coating, the oxide is dried at 85C for 10 min, pyrolyzed at
310 C for 7 min and removed to cool to ambient temperature all in ambient
atmosphere.
The dip-coating steps are repeated for 10 coats (or any desired number)
rotating the Ti sheet so as to dip with alternate sides down.
The fully coated sheet is final annealed at 310C for 2 hrs in ambient
atmosphere.
(C) Spacinq
0 94/07272 PCr/US93/08803
21~657
43
Microprotrusions are thermally sprayed through a mask on one side of the
electrode. The thermal spray material is TEFLON~ from E.l. DuPont de Nemours &
Co., Wilmington, Delaware.
The TEFLON~ protrusions are cured at 310C for 1.0 hr. in air. The coated
electrodes are next die-stamped to the desired shape.
(D) Gasket
A polypropylene gasket 1.5 mil (0.00381 cm) thick by 30 mil (0.0762 cm)
wide with outside perimeter the same as that of the electrode is placed on the
electrodes on same side as the microprotrusions and impulse heat laminated.
(E) Cord
One cord, 1 mil (0.00254 cm) diameter TEFLON~ coated tungsten wire, is
placed across the narrow dimension of the gasket and electrode surface and aligned
between microprotrusions. The location of the cord is one of three positions
centered, left of center, or right of center.
A second polypropylene gasket is placed on the first gasket sandwiching the
cord between the two gaskets.
The second gasket is impulse heated to adhere to the first gasket and to fix
the cord in place.
(F) Stackin~
Electrode/microprotrusion/gasket/cord/gasket units are stacked beginning with
a 2 mil (0.00508 cm) end plate unit to the desired number of cells and ending with a
plain 2 mil (0.00508 cm) end plate with the cords arranged such that the location is
staggered-left, center, right in a three unit repeating cycle (end perspective). (G) Reflow
The gasket is reflowed in nitrogen at 195 C for 60 min. to reflow the
thermoplastic. The unit is cooled in nitrogen to ambient temperature.
(H) Cord Removal
The cords are removed by pulling the exposed ends to leave the open fill
ports.
EXAMPLE 6
Fillin~ of the Cell GaP SPace
.
W094/07272 - PCT/US93/08~
214465~
44
A dry preunit 10 may be filled with an electrolyte with the following
procedure. Any of many possible dry preunit configurations may be used.
(H) Back Fill
The cords are removed manually to open the fill port. The stacked unit is
placed into an evacuation chamber and evacuated to < 35 mtorr for 5 to 60 min.
The liquid electrolyte 3.8 M H2S04 de-airated with nitrogen is introduced into the
chamber and fills the evacuated space between the electrodes.
ll) Seal Fill Port OPenin~s
The electrolyte filled preunit is removed from the chamber. It is rinsed with
deionized water to remove excess electrolyte and dried. HDPE film (1.5 mil
(0.00381 cm) thick) is placed over the fill port openings and impulse heat sealed
over the ports.
(J) Conditionin~
The device is charged up to full charge beginning at 0.1 V/cell increasing by
0.1 V/cell until 1 V/cell is obtained.
(K) Testin~
The device is tested in the conventional manner, having 1 V/cell with leakage
current of less than 25,uA/cm2, and a capacitance density per a cell of greater than
about 0.1 F/cm2. A 10 V device has a height of no rnore than 0.05", a 40 V device
has a height of no more than 0.13", and a 100 V device has a height of no more
than 0.27".
Performance characteristics for various device geometries and configurations
based on a sulfuric acid electrolyte are presented in Table 1.
Table 1
Ultraca~acitor Device Performance Characteristics
Area/cm2 2 2 2 2 25 25
volt 10 40 100 100 100 100
C/mF 26 6.7 2.6 10 150 753
ESR/mohm 100 330 780 780 62 70
vol/cc 0.29 0.73 1.6 1.6 11 32
J/cc 4.5 7.4 8.1 31 69 111A
~'094/07272 21 ~ ~ 6S 7 PCT/US93/08803
watt/cc 860 1660 2000 2000 3670 1100
EXAMPLE 7
Alternative Backfill of Dry Preunit
A dry preunit 10 may be filled with an electrolyte with the following
procedure. Any of many possible dry preunit configurations may be used.
(H) Back Fill
The cords are removed to open the fill port. The stacked unit is placed into
an evacuation chamber and evacuated to < 35 mtorr for 5 to 60 min. The liquid
non-aqueous electrolyte 0.5 M KPF6 in propylene carbonate de-airated with nitrogen
is introduced intG the chamber and fills the evacuated space between the electrodes.
(I) Seal Fill Port Openinqs
The electrolyte filled preunit is removed from the chamber and excess
electrolyte is removed. HDPE film (1.5 mil (0.00381 cm) thick) is placed over the fill
port openings and impulse heat sealed over the ports.
(J) Conditionin~
The device is charged up to full charge beginning at 0.1 V/cell increasing by
0.1 V/cell until 1.5 V/cell is obtained.
(K) Testinq
The device is tested in the conventional manner, having 1.5 V/cell with
leakage current of around 100,uA/cm2, and a capacitance density of around 4
Mf/cm2 for a 10 cell device.
EXAMPLE 8
DEVICE POST-TREATMENT CONDITIONS
The following is a list of the electrical properties (Table 3) of devices using
various gas postconditioning techniques to adjust the electrode rest potential so that
charging to at least 1 V/cell on multicell devices filled with 4.6 M sulfuric acid
electrolyte is possible and reduced leakage currents are observed. This treatment is
done before, during, and/or after reflow of the gasket material. For gas treatment at
temperatures below that used for gasket reflow the atmosphere was exchanged withan inert gas such as nitrogen or argon during reflow. For treatment after reflow of
WO 94/07272 2i 4465~ PCI/US93/088
46
the gasket material the tabs were removed before treatment. During treatment theatmosphere is evacuated and filled with the reactive gas periodically.
TABLE 3
Device characteristics for various postconditioning.
qas T/ C t/min. i"/LIA/Cm2 V/cell
H2 50 20 8 1.0
CO 100 170 40 1.0
CO 90 103 12 1.0
CO 90 165 20 1.0
CO 80 120 25 1.1
NO 75 20 27 1.0
NO 95 140 21 1.1
NH3 85 30 26 1.0
FORMATION OF MICROPROTRUSIONS BY SCREEN PRINTING
EXAMPLE 9
Application of EPOXV MicroProtrusions bY Screen Printinq onto a Porous Coatinq on a
Thin Substrate
(A) Screen Preparation - A 325 mesh stainless steel screen is stretched on a
standard screen printing frame. To this screen is edge glued (Dexter Epoxy 608 clear)
20 to a smaller 1-1.5 mil (0.00254 to 0.00381 cm~ thick brass sheet which has holes (6.3
mil (0.016 cm) diameter) drilled or etched to the desired pattern. The screen mesh is
removed from the area covered by the brass sheet leaving the brass sheet edge glued
to the screen mesh attached to the frame.
(B) Sample Holding - A vacuum is pulled on a porous alumina holding plate of 10
25 ,um average pore diameter is used to hold the 1 mil (0.00254 cm) thick porous oxide
coated material during the printing.
(C) Epoxy - A two component epoxy Master Bond EP21 AR is modified to the
desired viscosity (thixotropic, 300,000 to 400,000 cps) by the addition of a silica filler.
The filled epoxy having the desired viscosity is available by purchase order from Master
30 Bond, Inc. of Hackensack, New Jersey. The epoxy is prepared as per instructions.
The useful lifetime as a flowable fluid is about 30 min.
0 94/07272 PCI/US93/08803
21~4~57
47
(D) Screen printer parameters --
squeegee speed: 1-2 in/s
snap off: 20-30 mil (0.0508 to 0.0762 cm)
Constant temperature and humidity of the epoxy are important to assure an
5 even applied coat. Typical conditions are about 40-70% relative humidity and a temperature of about 20-25C.
(E) Printed epoxy pattern -- An array of epoxy bumps essentially 1 mil (0.00254
cm) in height and about 7.5 mil (0.019 cm) in diameter are produced. A typical pattern
on an electrode consists of an array of microprotrusions deposited on 40 mil (0.1016
10 cm) center-to-center spacing. In addition, the density of microprotrusions at the
perimeter of the electrode is increased by decreasing their center-to-center spacing to
20 mil (0.508 cm). The screen printed epoxy configuration is cured at 150C for a
minimum of 4 hr.
EXAMPLE 10
Screen Print Formation of EPOXY MicroProtrusions
(A) Screen Preparation -- A 230 or 325 mesh screen (8x10 in stainless steel)
without an emulsion on the surface, mounted on a standard printing frame, is used as
the base piece. An etched, drilled or punched stencil (6.0 x 8.5 molybdenum) is edge
glued using Dexter Epoxy 608 Clear from Dexter located to the back side of the screen.
20 MYLAR~ is placed over the stencil-screen unit and pressure applied to smooth the epoxy
into a uniform layer.
The screen is then flipped, epoxy applied to the top side of the screen, a
MYLARæ sheet placed over the area and the epoxy smoothed. The MYLAR~ sheet on
the top side of the screen is then removed. The screen-stencil assembly is then placed
25 into a 120C oven with ambient atmosphere for 5 min to cure the epoxy. Alternatively,
the epoxy can be cured by setting at ambient temperature for 30-60 min.
After removal of the screen-stencil from the oven, the MYLAR~ on the back side
is carefully peeled away immediately. The mesh screen on the top side is then cut
away using a sharp edge, with care being taken to prevent cutting of the stencil. Upon
30 removal of the mesh over the stencil pattern, any heat stable thermoset adhesive (e.g.
an epoxy resin) is applied to the cut mesh-stencil perimeter, covered with MYLAR~, and
WO 94/07272 PCI/US93/088--
21~657
48
the epoxy smoothed to ensure edge attachment of the screen to the stencil. The epoxy
is cured in the oven for 5 min. The resulting item is a stencil stretched taut by the
screen, ready for printing.
(B) Sample Holding - A porous ceramic holding (e.g. Fig. 8) plate (Ceramicon
5 Designs, Golden, Colorado, P-6-C material) of 4.5-6,u pore diameter with a porosity of
36.5% (30-60% porosity is acceptable) is used to hold the 1 mil (0.00254 cm) thick
porous oxide coated material during the printing by pulling a vacuum through the porous
ceramic plate. The ceramic plate is cut to the appropriate dimensions (the size and
shape of the substrate to be printed). This ceramic plate then is inserted into an
10 aluminum (steel, etc) frame 277 and epoxy or other adhesive) that can be mounted to
a screen printer. The ceramic plate is then carefully ground flush to the metal frame as
flat as possible. Locating pins 278, 279 and 280 are then added to hold the substrate
11 IA in appropriate location using holes 281, 282 and 283.
(C) Epoxy - The Master Bond EP 21 ART3 (a two component epoxy (of
15 polyamine hardener, 33 weight percent and a liquid epoxy resin, 67 weight percent)
about with a viscosity of 150,000 to 600,000 cps). The epoxy is prepared as per the
product instructions. The useful lifetime as a flowable fluid is about 30 min.
(D) Screen Printing Parameters -
Squeegee Speed 1-2 in/s (depends
upon epoxy viscosity)
Snap Off 20-30 mil (0.0050 to 0.0076 cm) (Related to
screen tension; and
adjusted accordingly)
(E) Printed epoxy pattern - An array of epoxy bumps essentially about 1 to
1.25 mil (0.00254 to 0.00316 cm) in height and about 7.5 mil (0.019 cm) in diameter
are produced. A typical pattern on an electrode consists of an array of microprotrusions
deposited on 40 mil (0.1 cm) center-to-center spacing. In addition, the density of
microprotrusions around the perimeter of the electrode is increased by decreasing their
center-to-center spacing to 20 mil (0.0508 cm). The screen printed epoxy configuration
is cured at 150C for 4 to 12 hrs. in an ambient atmosphere.
~ 0 94/07272 PCI'/US93/088~3
2~6S7
49
EXAMPLE 11
ALTERNATIVE SCREEN PRINTING PARAMETERS
(A) Separator Bumps -- Separator bumps range in height from .001 to .004 in.
(.00254 to 0.01016 cm) and widths of .006 to .012 in. (.01524 to .03038 cm).
5 Separator bumps may take the form of dots, squares, rectangles or a composite of
these shapes. The widths of the bumps increase as the bump height is increased.
(B) Separator Pattern -- Two patterns are utilized on an electrode substrate, the
Active Field Area and the Bounding Border Area. The AFA has the separator bumps
located on .040 by .040 in. (.1016 by .1016 cm) center to center spacing and are10 normally dots. The BBA has an increased bump density with a .020 by .020 in. (.0508
by .0508 cm) center to center spacing. Rows of rectangles alternate between arrays
of dots.
(C) Screen Preparation -- Design of the separator configuration is performed on
a CAD (Computer Aided Drafting) system. The CAD electronic data is converted to a
15 Gerber plot file. This plot data is used by the screen manufacture to create the artwork
to produce the desired thickness stencil for the screen printer. The screen is sent ready
to use by SMT (Screen Manufacturing Technologies of Santa Clara, California).
(D) Electrode Vacuum Plate (Workholder) -- A porous ceramic plate (Ceramicon
Designs, Golden, Colorado, P-6-C material) trimmed .050 smaller than the electrode
20 peri,neter is fitted and epoxied into an aluminum plate designed to fit the screen printer.
The top and bottom surfaces are ground flush and parallel. Multiple pins are inserted
around the centered electrode edge creating a corner stop for placement of the
electrode substrate.
(E) Epoxy -- A two component epoxy Master Bond EP21AR is modified to the
25 desired viscosity (thixotropic, 300,000 to 400,000 cps) by the addition of a silica filler.
The filled epoxy having the desired viscosity is available by purchase order from Master
Bond, Inc. of Hackensack, New Jersey. The epoxy is prepared as per instructions.The useful lifetime as a flowable fluid is about 30 min.
(F) Screen Printing Parameters
Thick Film Screen Printer
Squeegee Durometer 45 to 100 Type A
WO 94/07272 PClr/US93/088~
2144q.~
Squeegee Speed 1-2 in/s
Squeegee Pressure 10 to 15 Ibs .
Squeegee Down Stop .010 to max. in.
Snap Off .010 to .030 in. (.0254 to .0762 cm.)
FORMATION OF MICROPROTRUSIONS BY PHOTOLITHOGRAPHY
EXAMPLE 12
Hot Roller Photolitho~raPhic Production of Microprotrusions
(A) The ConforMASK~' 2000 high conformance solder mask of 1.5 mil
(0.0038 cm) in thickness is cut to the same size as the electrode.
(B) The photo resist film 381 is applied by placing the ConforMASK'!9 film on
the electrode material surface 111 A, after removing a release sheet 382 between the
pho;o resist film 381 and the electrode 111 A, and passing the laminate through heated
rollers (384 and :385), at 150F, to adhere the photoresist film 381 to the electrode
surface 111 A. A polyester cover sheet 382A on the outside of the photo resist film
381 is then removed.
(C) A dark field mask 387 containing rows of transparent holes (openings
388) is placed on the photo resist 381. A typical pattern consists of an array of holes
6 mil (0.0212 cm) in diameter 40 mil (0.1 cm) center-to-center spacing with a higher
density (20 mil (0.0508 cm) center-to-center) for three rows on the perimeter of the
electrode.
(D) The film 381 is exposed through the holes 388 and the mask 387, for
about 20 seconds, to a conventional UV light source, i.e. mercury vapor lamps 389.
The mask is then removed.
(E) The unexposed area of the photo resist is developed or stripped by placing
it in a tank with 1 % potassium carbonate for 1.5 min.
(F) The electrode surface with the microprotrusions ~standoffs) are then
washed with de-ionized water, placed in a tank with 10% sulfuric acid, for 1.5 min and
a final de-ionized water rinse.
(G) First, the microprotrusions 13 are exposed to UV light. A final cure of the
microprotrusions (standoffs) is done in a convection air oven at 300F for 1 hr.The finished electrode 111 A is used directly, or is treated, as described above.
~0 94/07272 PCI /US93/08803
~$ 7
EXAMPLE 13
Vacuum Lamination of Photo resist
(A) The ConforMASK~ 2000 high conformance solder mask of 2.3 mil
(0.0058 cm) in thickness is cut slightly larger than the electrode.
(B) The photo resist film is 381 vacuum laminated to the electrode 111 A, and
onto a supporting backing plate using standard operating conditions (160C,0.3 mbars)
using a Dynachem vacuum applicator model 724 or 730. The polyester cover sheet
382A is removed.
(C) The dark field mask 387 containing rows of transparent holes 388 is
placed on the photo resist film 381. A typical pattern includes an array of holes 6 mil
(0.0015 cm) in diameter 40 mil (0.102 cm) center-to-center spacing with a higherdensity (20 mil (0.0054 cm) center-to-center) for three rows on the perimeter of the
electrode.
(D) The film is exposed for 20 to 40 seconds to a non-collimated UV light
source of 3-7 KW power.
(E) The unexposed area of the photo resist film is developed or stripped by
using 0.5% potassium carbonate in a conveyorized spray developing unit, followed by
a de-ionized water rinsing and turbine drying.
(F) A final cure of the microprotrusion standoffs is done in a two step
process. First, the microprotrusions are exposed to UV light in a Dynachem UVCS 933
unit and then placed in a forced air oven at 300-310F for 75 min.
The finished electrode is used directly or further treated as described above.
EXAMPLE 14
Surfactants for Porosity Control
329 of cetyltrimethyl ammonium bromide was added to 1 ~ of iso-propanol with
stirring and slight heat. After approximately one hour 73 9 of TaCI5 and
47 9 of RuCI3 H2O was added to the clear solution. The standard coating procedure
was performed with interim pyrolysis at 300C or 5 min. and a final pyrolysis at 300C
for 3 hours. The average pore diameter of the coating increased to around 45 A. After
W094/07272 PCT/US93/088 ~
~21~ 57
52
post-treatment in steam at 260C at 680 psi for 2 hr. the average pore diameter
increased to 120 A.
A 25 wt % cetyltrimethyl ammonium chloride in water solution may also be used
to modify the pore diameter of the resulting coating.
EXAMPLE 1 5
Thermal Elastomeric Gasket
An alternative construction methodology is to sandwich a thermal elastomer
gasket (e.g. KRATON~) between the two HDPE gaskets. Device characteristics are
similar to those previously described.
EXAMPLE 16
Inclusion of Second Material to Accommodate Electrolvte Volume Increases
A porous hydrophobic material is added to each cell to accommodate any volume
increase of the electrolyte due to an increase in temperature.
This material is placed in the cell as either a gasket material inside the perimeter
HDPE gasket, or as a disk replacing part of the separator material.
A common material used is a PTFE material from W.L. Gore & Associates, Inc.
1-3 mil thick. Preferably, the PTFE material has water entry pressures from 20 to 100
psi.
EXAMPLE 17
Alternate Electrode Pretreatment
After the electrodes have microprotrusions, gaskets, and pull cards (after step
E), the electrodes are placed in 1M sulfuric acid and the open circuit potential is
adjusted to about 0.5V(vs HHE) using cathodic current with no hydrogen evolution.
The electrodes are transferred submerged in deionized water to an inert atmosphere
(e.g. Ar) where they are dried and assembled.
While only a few embodiments of the invention have been shown and described
herein, it will become apparent to those skilled in the art that various modifications and
changes can be made in the improved method to produce an electrical storage device
such as a battery or a capacitor having improved lifetime and charge/recharge
characteristics and low leakage current, and the device thereof without departing from
~0 94/07272 PCI /US93/08803
?l~6s7
the spirit and scope of the present invention. All such modifications and changes
coming within the scope of the appended claims are intended to be carried out thereby.
