U.S. patent application number 11/580400 was filed with the patent office on 2010-12-02 for material and device properties modification by electrochemical charge injection in the absence of contacting electrolyte for either local spatial or final states.
This patent application is currently assigned to Board of Regents of University of Texas System. Invention is credited to Ray Henry Baughman, Dong-Seok Suh, Anvar Abdulahadovic Zakhidov.
Application Number | 20100304215 11/580400 |
Document ID | / |
Family ID | 34919567 |
Filed Date | 2010-12-02 |
United States Patent
Application |
20100304215 |
Kind Code |
A1 |
Suh; Dong-Seok ; et
al. |
December 2, 2010 |
Material and device properties modification by electrochemical
charge injection in the absence of contacting electrolyte for
either local spatial or final states
Abstract
In some embodiments, the present invention is directed to
processes for the combination of injecting charge in a material
electrochemically via non-faradaic (double-layer) charging, and
retaining this charge and associated desirable properties changes
when the electrolyte is removed. The present invention is also
directed to compositions and applications using material property
changes that are induced electrochemically by double-layer charging
and retained during subsequent electrolyte removal. In some
embodiments, the present invention provides reversible processes
for electrochemically injecting charge into material that is not in
direct contact with an electrolyte. Additionally, in some
embodiments, the present invention is directed to devices and other
material applications that use properties changes resulting from
reversible electrochemical charge injection in the absence of an
electrolyte.
Inventors: |
Suh; Dong-Seok; (Seoul,
KR) ; Baughman; Ray Henry; (Dallas, TX) ;
Zakhidov; Anvar Abdulahadovic; (Richardson, TX) |
Correspondence
Address: |
Matheson Keys Garsson & Kordzik PLLC
7004 Bee Cave Rd., Bldg. 1, Suite 110
Austin
TX
78746
US
|
Assignee: |
Board of Regents of University of
Texas System
Austin
TX
|
Family ID: |
34919567 |
Appl. No.: |
11/580400 |
Filed: |
October 13, 2006 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
10591730 |
Feb 26, 2007 |
|
|
|
PCT/US05/07084 |
Mar 4, 2005 |
|
|
|
11580400 |
|
|
|
|
60550289 |
Mar 5, 2004 |
|
|
|
Current U.S.
Class: |
429/209 |
Current CPC
Class: |
H01M 8/1004 20130101;
H01G 11/22 20130101; H01G 11/06 20130101; H01G 11/34 20130101; H01M
8/1007 20160201; H01M 4/92 20130101; H01L 29/4908 20130101; H01M
4/8605 20130101; B82Y 30/00 20130101; Y02E 60/50 20130101; H01L
29/45 20130101; H01G 9/155 20130101; H01M 4/926 20130101; H01G
11/42 20130101; Y02E 60/13 20130101 |
Class at
Publication: |
429/209 |
International
Class: |
H01M 4/02 20060101
H01M004/02 |
Goverment Interests
[0002] This invention was made with partial Government support
under Contract No. MDA972-02-C-0044 awarded by the Defense Advanced
Research Projects Agency. The Government has certain rights in this
invention.
Claims
1-200. (canceled)
201. A supercapacitor/battery hybrid energy storage device
comprising at least one first sheet of a high surface area material
that is predominately non-faradaically charged and a second sheet
of a material that is predominately faradaically charged, wherein
these sheets are laterally joined together to make a device
electrode, and wherein the gravimetric surface area of the first
sheet is at least about 10 times that of the second sheet.
202. The supercapacitor/battery hybrid energy storage device of
claim 201, wherein the device is in a dry state, and wherein there
is no complete ion path in an electrolyte between said electrode
and a counter electrode.
Description
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application is a 371 US National Phase of
PCT/US2005/007084 filed on 4 Mar. 2005, which claims priority to
U.S. Provisional Application Ser. No. 60/550,289 filed on 5 Mar.
2004.
FIELD OF THE INVENTION
[0003] This invention relates to tuning the electronic, magnetic,
and optical properties of materials and devices by non-faradaic
electrochemical charge injection, wherein such tuned properties are
developed either in the presence or absence of contacting
electrolyte, and necessarily maintained in the absence of directly
contacting electrolyte.
BACKGROUND OF THE INVENTION
[0004] It is well known that charge injection can change the
magnetic, electronic and optical properties of materials. Prior-art
methods for changing these material properties by charge injection
either (1) involve electrostatic gate-based charge injection across
a dielectric (so charge injection is limited by dielectric
breakdown), (2) use an electrolyte that contacts the transformed
material (thereby limiting device applicability), or (3) use dopant
intercalation (thereby limiting applicable materials and providing
problematic structural changes). These three methods are called
dielectric-based charge injection, non-faradaic electrochemical
charge injection, and faradaic electrochemical charge injection,
respectively.
[0005] Dielectric-based charge injection is used for field effect
transistor (FET) devices that are critical for both today's
electronic circuits and those proposed for the future. In these FET
transistor devices, current is carried through a semiconductor
channel between source and drain electrodes. The current through
this semiconductor (channel) is controlled by charge injection into
the semiconductor channel by application of a voltage between the
source electrode and the gate electrode, which is separated from
the semiconductor channel by a dielectric. This charge injection is
that of an ordinary dielectric capacitor, so the amount of charge
injection that can be achieved is limited by the breakdown strength
of the dielectric. While enormously useful for submicroscopic
electronic devices, this dielectric-based charge injection is
unsuitable for macroscopic charge injection in materials having
macroscopic external dimensions. X. Xi et al., (Applied Physics
Letters 59 3470 (1991)) have demonstrated dielectric-based
switching of the superconducting transition temperature (T.sub.c)
of films of YBa.sub.2Cu.sub.3O.sub.7-x over a 2 K range. The
achieved resistance modulation in the normal state can be as much
as 20% in the normal state and 1500% near T.sub.c. Using a similar
method of dielectric-based charge injection in the oxygen deficient
YBa.sub.2Cu.sub.3O.sub.7-x superconductor, J. Mannhart et al.
(Applied Physics Letters 62, 630 (1993)) demonstrated that T.sub.c
can be changed by up to 10 K. However, the dielectric-based method
of T.sub.c switching used by Xi et al. and by J. Mannhart at al. is
not applicable for a macroscopically thick superconducting
material. In addition, J. A. Misewich et al. (Science 300, 783-786
(2003)) have used dielectric-based charge injection to make an
electrically driven light source from a single nanotube. Also, Y.
S. Choi et al. (Diamond and Related Materials 10, 1705-1708 (2001))
have used under-gate dielectric-based charge injection to modulate
electron emission for field emission displays, but find
disadvantage in this application as a result of field-induced
electron beam spreading and restrictions on the anode voltage.
[0006] Non-faradaic electrochemical charge injection uses
nanostructured materials having very high surface area and is
applicable for materials ranging from nanoscale materials to bulk
materials. However, unless the material is a metal or metal oxide
catalyst, the electrolyte is a ceramic held at high temperatures
(P. E. Tsiakaris, et al., Solid State Ionics 152-153, 721-726
(2002)) prior-art technologies teach that this charge injection can
only be accomplished by developing and maintaining contact of the
electrolyte with regions of the material where charge injection is
desired, which for macroscopic nanoporous materials includes
internal surfaces. In other words, the prior art teaches that
non-faradaic electrochemical charge injection into non-catalytic
materials generally requires maintained contact of that material
with the electrolyte. This non-faradaic electrochemical charge
injection has been used to provide electrochemical
electromechanical actuators (artificial muscles, see R. H. Baughman
et al., Science 284, 1340 (1999), R. H. Baughman et al., U.S. Pat.
No. 6,555,945) and liquid-ion-gated FETs (field-effect transistors,
see M. Kruger, Applied Physics Letters 78, 1291-1293 (2001)).
However, the maintained contact between the electrode (including
both internal and external surfaces) and the electrolyte limits
applicability of prior-art methods of non-faradaic electrochemical
charge injection. For example, the surrounding electrolyte for the
above described liquid-ion-gated FETs limits their applicability
for gas state sensing--since a sensed gas must first dissolve in
the electrolyte before it can be detected, which decreases both
device response rate and sensitivity and limits detection
capabilities to gases that can significantly dissolve in the
electrolyte. In addition, non-faradaic electrochemical charge
injection provides the basis for supercapacitors having much larger
charge storage capabilities than ordinary dielectric
supercapacitors. In the prior art (K. H. An et al., Adv. Funct.
Mater. 11, 387 (2001) and C. Niu et al., Appl. Phys. Lett. 70, 1480
(1997)) these supercapacitors are kept in the charged state as a
result of maintained contact between the nanostructured electrodes
and the electrolyte. Since the electrolyte provides mechanisms for
self-discharge, long term energy storage in such a supercapacitor
is not possible. Also, the possibility of charging non-faradaic
supercapacitors, removing the electrolyte, and then storing energy
in the dry-state supercapacitors has heretofore not been
conceived.
[0007] Faradaic electrochemical charge injection involves the
intercalation of ions into a solid electronically conducting
electrode material. This method is limited to the types of
materials that can incorporate dopant by a reversible process,
preferably at room temperature. For example, elemental metals and
metal alloys cannot undergo charge injection by this method.
Similarly, this method of charge injection is not useable for
non-porous materials having three-dimensional covalent bonding.
Also, substantial dopant intercalation fundamentally changes the
structure of the material and can introduce gross structural
defects. As a consequence, de-doping does not completely return the
material to the original state. Nevertheless, the faradaic
electrochemical method of charge injection has great value, as
indicated by the year 2000 award of a Nobel prize for the discovery
that dopant intercalation (either chemically or electrochemically)
into semiconducting conjugated polymers can convert these
semiconductors into metallic conductors. Faradaic electrochemical
doping (for conducting polymers and other materials) is used for
both primary and rechargeable batteries (Y. Gofer et al., Applied
Physics Letters 71, 1582-1584 (1997)), conducting polymer actuators
(R. H. Baughman, Synthetic Metals 78, 339 (1996)), electrochromic
displays (W. Lu et al., Synthetic Metals 135-136, 139-140 (2003)),
the control of membrane ion permeability (P. Burgmayer and R. W.
Murray, J. Phys. Chem. 88, 2515-2521 (1984)), the release of drugs
and other biochemically active agents (H. Shinohara et al.,
Chemistry Letters, 179-182 (1985), L. L. Miller et al., U.S. Pat.
No. 4,585,652), and electrochemical light emitting displays (G. Yu
et al., Science 270, 1789-1791 (1995)). For such devices, dramatic
structural changes are typically associated with dopant
intercalation, and these charges are not fully reversed on
de-intercalation--which limits cycle life. The required dopant
insertion and de-insertion processes (called intercalation and
de-intercalation) result in slow device response, short cycle life,
hysteresis (leading to low energy conversion efficiencies), and a
device response that depends on both rate and device history.
[0008] The embodiments of the present invention eliminates key
problems of these prior-art technologies, by showing that
non-faradiac electrochemical charge injection can be maintained,
and even developed, at room temperature for regions of the
electrode that are not in direct physical contact with the
electrolyte. The present discoveries enable materials and device
applications that would not be possible, or would be less
advantageous, in the presence of locally contacting
electrolyte.
BRIEF DESCRIPTION OF THE INVENTION
[0009] A first object of the invention is to provide processes for
the combination of injecting charge in a material electrochemically
via non-faradaic (double-layer) charging and retaining this charge
and associated desirable properties changes when the electrolyte is
removed.
[0010] A second object is to provide compositions and applications
using material property changes that are induced electrochemically
by double-layer charging and retained during subsequent electrolyte
removal.
[0011] A third object is to provide reversible processes for
electrochemically injecting charge into material that is not in
direct contact with an electrolyte.
[0012] A fourth object is to provide devices and other material
applications that use properties changes resulting from reversible
electrochemical charge injection in the absence of an electrolyte.
Examples of such application of charge injection and associated
magnetic, optical, and electronic properties changes are for
optically transparent electronic conductors; spintronic devices;
information storage devices; nanostructured magnets; chemical and
mechanical sensors; electromechanical actuators; the control of
thermal and electrical energy transport; the tuning of surface
energy and friction; the switching, phase shift, and attenuation of
electromagnetic radiation; tuning magnetoresistive materials; and
drug delivery.
[0013] Invention embodiments include processes for improving
material properties by non-faradaic charge injection and retaining
these switched properties in the absence of electrolyte that
contacts charge-switched electrode regions.
[0014] More specifically, in some embodiments, the present
invention provides for processes whose overall effect is to
provide, retain and employ charge injection to substantially change
the properties of a material A, material A being either a largely
electrolyte-free porous material region, or a particle, the process
comprising the steps of: (a) immersing material A into an
electrolyte E; (b) providing an ion conducting and substantially
electronically insulating continuous path between material A and a
counter electrode material B; (c) applying a potential between
material A and the counter electrode B for sufficient time that the
desired charge is injected into material A; and (d) substantially
removing the electrolyte E from contact with material A, wherein
both material A and counter electrode B have an electronically
conducting charged or uncharged state and material A has an
achievable capacitance for non-faradaic charging of above about 0.1
F/g.
[0015] Compositions of matter resulting from the above processes
are also provided in the invention embodiments. An example is a
composition of matter containing non-faradaically injected charge
and substantially no electrolyte that maintains, in a suitable
environment, a potential that deviates from the potential of zero
charge by at least 0.1 volt.
[0016] Devices that utilize non-faradaic charge injection in the
absence of locally contacting electrolyte are also provided in
invention embodiments. These devices, having a tunable response,
comprise: (a) a nanostructured electrode component C of a first
electrochemical electrode and an electrode component D of a second
electrochemical electrode; (b) an ionically conducting material
that is substantially electronically non-conducting that connects
said first and said second electrochemical electrodes; and (c) a
means of providing a voltage between said first and said second
electrochemical electrodes, wherein the electrode component C is
not in direct contact with an electrolyte, the electrode component
C has an achievable capacitance of above about 0.1 F/g for
substantially non-faradaic charging, and wherein properties changes
of the electrode component C in response to injected charge are
used to achieve device performance. In some embodiments it is
advantageous if said first electrochemical electrode and said
second electrochemical electrode are both porous electrodes having
a capacitance of at least about 0.1 F/g, and wherein said ionically
conducting material that is substantially electronically
non-conducting at least partially penetrates both said first and
said second electrochemical electrodes.
[0017] The foregoing has outlined rather broadly the features of
the present invention in order that the detailed description of the
invention that follows may be better understood. Additional
features and advantages of the invention will be described
hereinafter which form the subject of the claims of the
invention.
BRIEF DESCRIPTION OF THE DRAWINGS
[0018] For a more complete understanding of the present invention,
and the advantages thereof, reference is now made to the following
descriptions taken in conjunction with the accompanying drawings,
in which:
[0019] FIG. 1 shows the observed tunability of four-point
electrical conductivity as a function of applied potential (versus
Ag/Ag+) for a sheet of single-wall carbon nanotubes immersed in an
organic electrolyte (0.1M tetrabutylammonium hexafluorophosphate in
acetonitrile). The different curves are for three successive cycles
of electrochemical potential change (using squares, circles, and
triangles for successive cycles).
[0020] FIG. 2 shows the dependence of four-point electrical
conductivity upon the amount of injected charge (per carbon) for
the nanotube sheet used for the FIG. 1 experiment (black data
points are for experiments using 0.1M tetrabutylammonium
hexafluorophosphate/acetonitrile electrolyte and near-white data
points are for related measurements using 1 M aqueous NaCl
electrolyte). The origin of the charge scale is arbitrary.
[0021] FIG. 3 shows measured cyclic voltammetry for the SWNT sheet
of FIGS. 1 and 2 when immersed in tetrabutylammonium
hexafluorophosphate electrolyte. These results show that the
charging is predominately non-faradaic by double-layer charge
injection.
[0022] FIG. 4 shows that the dramatic hole-injection-induced
increase in electrical conductivity of the nanotube sheet in FIGS.
1 and 2 is largely retained when the hole-injected electrode is
dried in a flowing dry nitrogen atmosphere to remove the
electrolyte. The insert to this figure shows results for the
initial four-hour period on an expanded time scale.
[0023] FIG. 5 shows the retention of conductivity enhancement when
the nanotube sheet is removed from the electrolyte and held in air
for the investigated five-day period.
[0024] FIG. 6 depicts cyclic voltammetry measurements (20 mV/sec
using 1 M aqueous NaCl electrolyte) for nanostructured platinum
electrodes, showing that charging is non-faradaic by double-layer
charge injection.
[0025] FIG. 7 shows potential measurements before and after removal
of the nanostructured Pt electrodes of FIG. 6 from the 1 M NaCl
aqueous electrolyte (and subsequent reimmersion into the
electrolyte). The measured electrode potentials of both positive
and negative electrode indicate that non-faradaically injected
charge is partially retained even when the electrodes are removed
from the electrolyte.
[0026] FIG. 8 shows a scanning electron microscope (SEM) picture of
a nanostructured inverse-opal carbon electrode that was found to
store charge without the need for a surrounding electrolyte.
[0027] FIG. 9 uses carbon nanotube electrodes to schematically
illustrate a means used in invention embodiments for injecting
charge into regions of a high-surface-area electrode that is not
directly contacted with electrolyte.
[0028] FIG. 10 schematically illustrates a prior-art device
technology for using electrochemical double-layer charge injection
to switch the conductivity of a semiconducting channel material.
This device uses the prior art technology of liquid-ion gating,
which means that the gate is immersed in a liquid electrolyte that
provides the necessary ions as liquid state species.
[0029] FIG. 11 schematically illustrates a first electrochemical
transistor device of the present invention that does not use
liquid-ion gating, which could be used for information storage,
electronic switching, or gas sensing.
[0030] FIG. 12 schematically illustrates a second electrochemical
transistor device of the present invention that does not use
liquid-ion gating, which could be used for information storage,
electronic switching, or gas sensing.
[0031] FIG. 13 schematically illustrates an optical gas sensor,
based on the surface-enhanced Raman effect, that uses
electrochemically controlled charge injection in a
metallo-dielectric photonic crystal to optimize sensitivity and
species selectivity.
[0032] FIG. 14 schematically illustrates a fuel cell of invention
embodiments. Unlike the case of prior-art fuel cells, the fuel cell
redox reactions predominately occur in the gas phase (on surfaces
of carbon nanotubes that are exposed to the H.sub.2 and O.sub.2
gas) without directly contacting electrolyte.
[0033] FIG. 15 schematically illustrates a tunable nanotube device
in which tunability results from an electrochemically-induced
insertion of ions inside a nanotube, and insertion of associated
counter electronic charges onto the nanotube.
[0034] FIG. 16 schematically illustrates a supercapacitor of
invention embodiments that can be charged, drained of electrolyte,
partially or completely evacuated then reactivated for subsequent
discharge in a remote location by refilling with electrolyte. In
the illustrated case (showing a cross-section of the device normal
to the supercapacitor electrode sheets), the electrolyte for device
refill is carried in a compartment of the device. In an alternative
invention embodiment, the electrolyte (which can be salt water) is
injected into the supercapacitor device from an electrolyte
container that is separate from the device.
[0035] FIG. 17 schematically illustrates a filtration device that
uses non-faradaic charging to dynamically and selectively control
the transport of material though a membrane having discreet
pores.
[0036] FIG. 18 schematically illustrates an electromechanical
actuator device of invention embodiments that can provide much
larger actuation strains than any prior art device of any type.
This device uses a carbon multiwalled nanotube (MWNT) which
telescopes outward in order to decrease free energy by increasing
the surface area available for charge injection.
[0037] FIG. 19 schematically illustrates an electrochemically-gated
device that can be used for atomic probe imaging and electron
emission.
[0038] FIG. 20 schematically pictures a device that operates like
the device of FIG. 19, except that the nanotube probe tip base
makes electrical contact to the nanotube. Such design facilitates
device construction.
[0039] FIG. 21 schematically illustrates a device in which a
high-surface-area nanostructured material functions as an
electrochemically-gated ion beam source, that can be modulated at
will by the application of low voltage potential between
electrochemically-active electrodes.
[0040] FIG. 22 schematically illustrates a cross-sectional view of
a device in which individual nanoscale electrodes (pixels) are
electrochemically charged either positively or negatively using a
focused electron beam. In contrast with the case of other device
illustrations, electrochemical charging does not require a wire
lead to the electrode, so extremely small pixels can be
conveniently addressed within a densely packed pixel array.
[0041] FIG. 23 schematically illustrates one type of device of
invention embodiments that provides tunable thermal
conductivity.
[0042] FIG. 24 illustrates an invention embodiment in which
predominately non-faradaic charge injection is used to optimize the
figure of merit (ZT) of thermoelectric elements interconnected by
electrolyte.
[0043] FIG. 25 illustrates a device embodiment of the present
invention where electromagnetic radiation propagates within the
device either at least approximately parallel to the electrode
layers or predominately along the lengths of element 2500 (and like
elements), along the lengths of element 2502 (and like elements),
or along both of these types of elements (in contrast with the
possibly approximately perpendicular propagation of the
electromagnetic radiation for the device of FIG. 13).
[0044] FIG. 26 schematically illustrates an EBIG (Electrolyte-Bare
Ion Gated) device that provides electroluminescence by using a
semiconducting nanostructured material--in this case a
semiconducting carbon nanotube.
[0045] FIG. 27 schematically illustrates another EBIG that is light
emitting, wherein the light-emitting element is a semiconducting
inverse-opal photonic crystal.
[0046] FIG. 28 schematically illustrates a back-gated
electrolyte-bare ion gate (EBIG) field effect transistor that has
an air gap.
[0047] FIG. 29 illustrates a device embodiment of the present
invention, wherein the advantageous combination of non-faradaic
electrochemical and gas-gap-based electrostatic charge injection
for a chemical sensor that can replace conventional Chem-FETs, is
depicted. This illustrated embodiment depicts a Chem-FET sensor
device that is at the same time a EBIG sensor.
DETAILED DESCRIPTION OF THE INVENTION
[0048] Invention embodiments are directed to processes, materials,
and devices that utilize non-faradaically injected charge and
associated changes in either magnetic, electronic, optical, or
chemical properties, wherein the non-faradaically injected charge
is retained in the absence of direct or maintained contact with the
electrolyte.
[0049] Differentiation between faradaic and non-faradaic charging
processes is important for understanding the advantages and novelty
of these invention embodiments. Significant amounts of electronic
charge can be injected (i.e., stored) in an electrode typically
only if counter ions of opposite charge are available in close
proximity to the electronic charge injected into the electrode.
These counter ions can compensate the electrostatic repulsion of
the electronic charges on the electrode, thereby enabling the
electronic charge injection process to proceed to high levels. A
faradaic process for an electrode material is one in which
electronic charge injected into the electrode is predominately
compensated by ions that are inserted (i.e., intercalated) into the
volume of the charge-injected component of the electrode. A
non-faradaic process is one in which compensation of electronic
charge on the electrodes is by ions that do not enter the solid
volume of the charge-injected component of the electrode. A
well-known example of a non-faradaic process is one in which the
counter ions (to the electronic charge on the electrodes) are
located in the electrolyte (within the so-called charge double
layer). Evidence of faradaic charging processes is provided by the
existence of well-defined peaks in cyclic voltamograms
(current-versus-potential plots at constant potential scan rates).
In contrast, the current in a cyclic voltamogram depends only
weakly on potential in a potential range where electrode charging
is non-faradaic and the electrolyte is neither reduced nor
oxidized. Correspondingly, for the purposes of this invention, the
existence of potential range, where electrode potential increases
substantially linearly with injected charge under some charging
conditions, is sufficient (but not necessary) evidence for
predominately non-faradaic charging in this potential range for
these charging conditions. If ions corresponding to injected
electronic charges are predominately located on or near the closest
surface of an electrode material, including the internal surfaces
arising from material porosity and the interior volume of hollow
fibers, the charging process is herein defined as non-faradaic,
independent of the shape of cyclic voltammetry curves. The pore
size in a charge injected electrode structure can be at any size
that is larger than that of the incorporated ions (together with
possible solvating species), since this is the situation needed in
order to avoid harmful dimensional charges as a result of the
volume of the ion (together with possible solvating species).
Charging by ion incorporation within a nanotube, such as a carbon
nanotube or a carbon scroll is specifically designated as being
non-faradaic for the purposes of this invention.
[0050] Prior-art understanding of non-faradaic charge injection is
that the electrode capacitance (and therefore the non-faradaic
charge injection capability) for an electrode is given by the
following equation: C.sub.e=A.sub.eC.sub.s. In this equation,
C.sub.s is the capacitance per unit surface area of the conducting
component of the electrode when this surface area is in contact
with electrolyte. A.sub.e is the total surface area of the
electronically conducting component of an electrode that is in
contact with a suitably thick layer of electrolyte. This electrode
surface area includes all electrolyte-coated internal and external
surface area of the electronically conducting component of the
electrode. Ignoring the potential dependence of C.sub.s, the amount
of injected electrode charge is q.sub.e=(V-V.sub.o)C.sub.e, where
V.sub.o is the potential at which there is zero electronic charge
injected in the electrode and V is the potential applied to the
electrode, measured with respect to the same reference electrode as
V.sub.o. Applicants have discovered that these equations are not
generally valid when only part of a nanostructured electrode is in
contact with the electrolyte. The processes that lead to violation
of these equations are the basis for certain embodiments of this
invention.
[0051] More specifically, Applicants have discovered that dopant
ions (together in some instances with solvating species from the
electrolyte) can migrate at room temperature from the electrolyte
(solid-state or liquid) to regions of the electrode that are not
coated with electrolyte. This dopant ion migration enables
electronic charge to be injected in regions of the electrode that
are not in contact with the electrolyte, thereby increasing the
injected charge (q.sub.e) above the values given by the equation
given in the preceding paragraph. Most important for device
applications, such as field emission displays, Applicants'
discovery means that electrochemical double-layer charge injection
can be used to inject charge in regions of the electrode that are
in a gas or vacuum, rather than in an electrolyte.
[0052] This enabling and unexpected discovery resulted from
Applicants surprising observation that charge can be
non-faradaically injected into a nanoporous electrode, and then
retained in this electrode when this electrode is removed from a
volatile electrolyte and dried by pumping in dynamic vacuum. This
observation means that the optical, magnetic, and electronic
properties of an electrode material can be tuned, and that these
tuned properties can be substantially maintained for materials
applications in ambient environments, where the problematic
electrolyte is absent.
[0053] Also generally important for invention embodiments, the
stability observed (in the absence of electrolyte) for injected
charge on surfaces of nanostructured materials (internal and
external) indicates that one can continuously vary the degree of
charge injection in nanostructured material elements that are not
in direct contact with the electrolyte and maintain this injected
charge when the power source is disconnected. This stability and
the continuous tunability of charge injection in the absence of
direct electrolyte contact enables many of the materials and device
applications in the device embodiments. These embodiments require
that the continuous tuned material elements are electronically
connected to an electrode, so that the electronic part of charge
injection can occur. Additionally, these embodiments require that
ion conduction path(s) exist or can self form to the continuously
tuned elements--which need not be in direct contact with the
electrolyte.
[0054] Processes of the present invention generally require at
least two electrodes, called working and counter electrodes. For
the cases where the desired charge-injection-produced properties
changes are for only one of these electrodes, the working electrode
is defined as this electrode. In addition, one can optionally
employ a reference electrode, whose function is to place the
potentials of the working and counter electrodes on an absolute
scale. The working electrode should typically have a capacitance of
at least 0.1 F/g when fully immersed in electrolyte. More
typically, the capacitance of the working electrode fully immersed
in the electrolyte should exceed 1 F/g. Most typically, the
capacitance of the working electrode should exceed 10 F/g when
fully immersed in electrolyte. The specific capacitance of an
electrode can be derived using the conventional method from the
dependence of gravimetric current on the scan rate of electrode
potential (versus a reference potential). (See J. Li et al., J.
Phys. Chem. B. 106, 9299-9305 (2002) for a description of this
method.) The above-mentioned specific capacitances for the typical,
more typical, and most typical invention embodiments are the
maximum values that can be measured within the stability range of
the electrolyte.
[0055] Before discussing particular embodiments of the present
invention, some surprising observations that provide basis for
these embodiments will be described. These discoveries are further
elaborated in the examples section. Following this, elaboration
will be provided on (1) specialized invention embodiments and (2)
material components and processing methods for the practice of
these embodiments.
[0056] The results depicted in FIGS. 1 and 2 show that a potential
change and corresponding electrode charge injection can
dramatically change the properties of a nanostructured electrode
material. FIG. 1 shows the presently observed continuous tunability
of four-point electrical conductivity as a function of applied
potential (versus Ag/Ag+) for a sheet of single-wall carbon
nanotubes (SWNTs) immersed in an organic electrolyte (0.1M
tetrabutylammonium hexafluorophosphate in acetonitrile). These
results demonstrate that the electrical conductivity of the
nanotube sheet can be increased by about an order of magnitude by
electrochemical charge injection. There is slight hysteresis
evident for the curves in FIG. 1, with the conductivity .sigma. on
the extreme left side of the potential minimum being slightly
higher for hole injection (increasingly positive applied potential)
and the conductivity slightly lower on the extreme right side of
the minimum for electron injection (increasingly negative applied
potentials). The different curves are for three successive cycles
(using squares (101), circles (102), and triangles (103) for
successive cycles). The density of these nanotube sheets is about
0.3 g/cm.sup.3, versus the density of about 1.3 g/cm.sup.3 for
densely packed nanotubes having close to the observed average
nanotube diameter. Hence, the void volume in these nanotube sheets
is about 76.9 volume percent. This high void volume, and the
correspondingly high accessible surface area, is generally
important for achieving high degrees of non-faradaic charge
injection at modest applied potentials. Supporting this conclusion,
the measured BET surface area determined from nitrogen gas
adsorption for these nanotube sheets is approximately 300
m.sup.2/g.
[0057] FIG. 2 shows the dependence of four-point electrical
conductivity upon the amount of injected charge for both the above
experiment with 0.1 M tetrabutylammonium
hexafluorophosphate/acetonitrile electrolyte (black data points
(201)) and for other experiments using 1 M NaCl electrolyte
(near-white data points (202)). Although not indicated here in this
plot of conductivity versus charge (because charge measurements are
not reliable for potentials exceeding the redox stability of
aqueous electrolyte), reversible conductivity increases from about
100 S/cm to about 1000 S/cm were also observed at positive
potentials for experiments using the 1 M aqueous NaCl electrolyte.
The origin of the charge scale is arbitrary. The charge per carbon
at the minimum in electrical conductivity can be used to place
origin of the charge axis, since theory suggests that conductivity
should be minimized at the potential of zero charge (pzc), where
there is no charge on the carbon nanotube.
[0058] FIG. 3 shows measured cyclic voltammetry during charge
injection for the above SWNT sheet when immersed in the above
tetrabutylammonium hexafluorophosphate electrolyte. The absence of
major peaks in this cyclic voltammetry curve (measured versus
Ag/Ag+) indicates that charging is predominately non-faradaic for
this electrolyte and potential range.
[0059] FIG. 4 shows that the dramatic increase in electrical
conductivity of the nanotube sheet in FIGS. 1 and 2 (obtained by
hole injection for the nanotube sheet and the organic electrolyte
of these examples) is largely retained when the hole injected
electrode is dried in flowing dry nitrogen atmosphere to remove the
electrolyte. The insert to this figure (describing results over a
four-day period) shows first a conductivity increase and then a
conductivity decrease during the first few hours of this four day
study, which might be a result of volatilization of the
acetonitrile component of the organic electrolyte. FIG. 5 shows the
retention of conductivity enhancement when the nanotube sheet is
removed from the electrolyte and held in air for the investigated
five-day period. These results show that the enhancement of
electrical conductivity of the hole-doped nanotube sheet is
relatively stable even in atmospheric air. The electrical
conductivity enhancement (and retained charge) degrades more
rapidly in air for the hole-injected electrode than for the
electron-injected electrode.
[0060] Applicants have found equally unexpected results when these
room temperature measurement results were extended to generic
nanostructured materials, and, in particular, to a nanostructured
metal--which has no possibility of accommodating ions by faradaic
processes (intercalation). These results are for a platinum
electrode made by compaction of 30 nm diameter Pt nanoparticles,
using the method described by J. Weissmuller et al. in Science 300,
312 (2003). The cyclic voltammetry results in FIG. 6 (using 1 M
aqueous NaCl electrolyte) show that these electrodes provide the
classical dependence of current on applied potential that arises
for double-layer charging. There are no current peaks due to
Faradaic processes and the current at constant voltage scan rate
(20 mV/sec) varies little with potential. From plots of current
versus potential scan rate in 1 M aqueous NaCl electrolyte), the
electrode capacitance is about 14.5 F/g. The high volume fraction
of void space observed in these pellets (between 81.6 and 87.2
volume percent for compaction pressures between 0.6 MPa and 2.1
MPa), together with the corresponding high gravimetric surface
area, explains the high degree of non-faradaic charge injection
that results for modest applied potential for the nanostructured Pt
electrode.
[0061] Perhaps most importantly, Applicants have found that the
nanoporous Pt electrode remains charged when disconnected from the
power source and removed from the electrolyte. Initial results
indicating this stability are shown in FIG. 7. As shown in FIG. 7,
the charge on the positive electrode (i.e., the hole-doped
electrode) is retained when the electrode is removed from the
electrolyte, and then held in air. Indication of this retained
charge is provided by reimmersion of the nanoporous hole-injected
Pt electrode in the 1 molar NaCl electrolyte, and finding that the
electrode potential is substantially unchanged. Just like the case
for the carbon nanotube electrode, the potential of the negatively
charged electrode is less stable than for the positively charged
electrode, as indicated by the results shown in the lower part of
FIG. 7.
[0062] Since it is possible that some electrolyte is still retained
inside the pores of the Pt electrode during the experiment depicted
in FIG. 7, this experiment was repeated using much longer time
periods when the electrodes are not in contact with the 1 M NaCl
electrolyte, and dynamically pumping on the nanoporous Pt during
this time period after the pellet electrodes have been disconnected
from the power source and the electrolyte was removed from the
electrochemical cell. The electrode potentials (versus Ag/AgCl)
before and after this two day exposure of the electrodes to dynamic
pumping were +0.58 V and +0.45 V for the hole-injected electrode
and -0.04 V and +0.02 for the electron-injected electrode. The
potential between the two electrodes changed from the initial 0.62
V to a final 0.43 V after removal of the electrodes from the
electrolyte and dynamically pumping on these electrodes for two
days.
[0063] To further evaluate the stability of charge storage, the
time period in which the platinum pellets were exposed to dynamic
vacuum was extended to a week. After this, the nanoporous Pt
electrodes were reimmersed in the 1 M NaCl electrolyte to determine
their charge state by electrochemical potential measurements
(naturally, without applying any external potential). High charge
storage was again indicated for the positively charged Pt electrode
(indicated by retention of a 0.28 V potential, versus Ag/AgCl,
compared with the initial potential before electrolyte removal of
0.33 V). The negatively charged electrode had lower stability, as
indicated by a potential change from the initial -0.68 V (before
removal of the electrolyte and the week-long process of drying the
electrode in dynamic vacuum) to a final potential on initial
reimmersion into the electrolyte of -0.32 V.
[0064] Electron and hole injection in the nanostructured platinum
samples does not significantly change electrical conductivity (in
contrast with the case for the nanostructured nanotube sheets),
since this charge injection causes only a small fractional change
in the already high total electron concentration. While it is known
that charge injection in a liquid electrolyte can provide changes
in unit cell volume, the prior art did not anticipate that these
charge-induced volume changes can be retained in electrolyte-free
materials (J. Weissmuller et al., Science 300, 312 (2003) and R. H.
Baughman, Science 300, 268-269 (2003)).
[0065] The implications of these results for practical applications
in invention embodiments are profound, since diverse properties
(including superconductivity, magnetism, and magneto-resistance)
can be very sensitive to material volume.
[0066] Applicants have also shown experimentally that inverse opals
(also called inverse-lattice photonic crystals) of conducting
materials are another type of composition that can be used in some
invention embodiments. In these experiments, a carbon inverse opal
was synthesized by infiltrating phenolic resin into an ordinary
SiO.sub.2 opal, pyrolizing this phenolic resin, and then removing
the SiO.sub.2 template material by dissolution in HF solution. A
scanning electron microscope (SEM) image of this carbon inverse
opal is shown in FIG. 8. Experiments done in both 1 M aqueous
electrolyte and tetrabutyl ammonium
hexafluorophosphate/acetonitrile electrolyte show that
electrochemically-injected charge is partially retained when the
electrodes are removed from the electrolyte and dried, and that the
stability of injected charge is much higher than for the same
electrodes immersed in the electrolyte with the power source
disconnected.
[0067] This retention of electrochemical double-layer-injected
charge and charge-injection-induced properties changes, when the
charge-injected material is not in contact with the electrolyte, is
more broadly important for invention embodiments. Also important,
Applicants have found that dry-state-retained injected charge is
highly mobile, as evidenced by Applicants' observation that the
charge on dry negatively and positively charged nanostructured
electrodes, rapidly occurs when these electrodes are contacted in
the dry state.
[0068] Processes, materials, and device applications stemming from
the above-described discovery of a means for
electrochemically-injecting charge in nanostructured materials and
retaining this injected charge when the nanostructured material is
not in contact with the electrolyte, is described next. These
applications utilize Applicants' discoveries that non-faradaically
injected charge and associated properties changes are retained in
the absence of electrolyte. Use is also made of Applicants'
discovery that non-faradaically injected charge (ion and
corresponding electronic counter charge) of an electrolyte-free
electrode can be highly mobile. Because of this mobility of the
ions and electronic charge, electronic charge and counter ions can
be reversibly and controllably electrochemically inject into
nanostructured material elements that are not placed in direct
contact with an ion source (for example, an electrolyte or
intercalated material, such as a doped conducting polymer).
[0069] Both here and elsewhere herein the terms "free of
electrolyte," "not directly contacted with electrolyte," "not
contacted by the electrolyte," "largely electrolyte free," and like
terms have a specific meaning that is now defined. These terms
apply for a material element a if any of the following cases
applies: (1) material component .alpha. was at no point placed in
physical contact with a bulk electrolyte composition employed for
either processing, device fabrication, or device operation; (2)
either anions, cations, and/or solvating species are present only
in about 10 nm or thinner surface layers on or within material
component .alpha. or as salt crystals on or within material
component .alpha. that cannot serve as an effective electrolyte
under application conditions; (3) electrolyte-derived chemical
species in materials component .alpha. that can be almost entirely
removed from the bulk electrolyte by pumping in dynamic vacuum at
room temperature are essentially stable in materials component
.alpha. under the same conditions; (4) the ratio of free anions to
free cations that are present in materials component .alpha. (i.e.,
those that are not crystallized as salt crystals) is either lower
than 0.9 or above 1.1; and (5) major components in the utilized
electrolyte are substantially absent in materials component
.alpha.. Defining the meaning of these terms and like terms is
pertinent, since the devices of invention embodiments operate by
the surface diffusion of ions from the electrolyte and such ion
diffusion can be accompanied by co-diffusion of ion-solvating
species from the electrolyte.
[0070] First to be considered is a process of the present
invention, wherein the properties of a nanostructured material are
tuned by non-faradaic charge injection in an electrolyte, and
wherein these properties changes and associated injected charge are
retained when the electrode is free of electrolyte. In perhaps the
simplest embodiment of this process, the material to be processes
by electrochemical charge injection is immersed into a liquid
electrolyte that is based on a solvent that can be volatized (like
water or acetonitrile). This material component is used as the
working electrode of an electrochemical cell that comprises this
working electrode and a counter electrode. A reference electrode
can also be present in this electrochemical cell, since measurement
of the potential of this working electrode with respect to this
reference electrode can be used to regulate the charge injection
process. This nanostructured material, containing electrolyte in
its pore structure, is charged by application of a potential
difference between the working and counter electrodes. Satisfactory
completion of the desired degree of charge injection can be
monitored by measuring either the total charge passed through the
electrochemical cell or the evolution of electrode potential versus
a reference electrode. After charge injection, the nanostructured
electrode (or electrodes) in which charge injection is needed can
be disconnected from the power source and then dried or otherwise
processed to remove liquid electrolyte (such as by washing in the
solvent component of the electrolyte, followed by drying). In order
to retain injected charge in the dried electrode, it is generally
important that the electrode material is not contacted with
materials that can undergo redox reactions with charge on the
electrode. Avoiding such degradative reactions is typically more
difficult for electron-injected electrodes than it is for
positively injected electrodes, although this problem is decreased
in difficulty when the degree of charge injection in the electrode
is not high.
[0071] A useful method for substantially eliminating charge
degradation of electron-injected materials (by reaction with
hole-donating impurities) is to enclose these materials in an
environment that comprises "getter" substances that are easier to
oxidize than the electron-injected material. Likewise, a useful
method for substantially eliminating charge degradation of
hole-injected material (by reaction with electron-donating
impurities) is to enclose these materials in an environment that
comprises getter substances that are easier to reduce than the
hole-injected material. A host of materials can be used as getter
substances, such as Li metal or n-doped conducting organic polymers
for electron-injected materials or device components and heavily
p-doped conducting polymers for hole-injected materials or device
components. In some invention embodiments where neither increases
or decreases in charge injection are desirable, it is helpful if
redox active components of the getter material do not contact the
charge-injected material. For this purpose, the getter material,
and components thereof, should ideally be immobile with respect to
diffusion and volatilization at the normal operating temperature of
the device or the charge injected material. An important exception
to this is the case where contact of the getter material is used to
maintain the electrochemical potential (and therefore the degree of
charge injection) at a value set by the getter material. In such
embodiments, the getter material is likely to be an alkali metal,
an alkali metal alloy, or a donor or acceptor intercalated
material.
[0072] In some embodiments, processes of invention embodiments can
be described as providing, retaining, and employing charge
injection to substantially change the properties of a largely
electrolyte-free porous material region A. Such processes comprise
the steps of: (a) placing an electrolyte within the interior of A,
(b) providing an ion conducting and substantially
electronically-insulating continuous path between A and a
counter-electrode material B, (c) applying a potential between A
and counter electrode B for sufficient time that the desired charge
is injected into A, and (d) substantially removing the electrolyte
from contact with A, wherein both A and B have an electronically
conducting charged or uncharged state and A has an achievable
capacitance for non-faradaic charging of above about 0.1 F/g. A
material produced by this process is referred to herein as an
"Electrolyte-Bare Ion-Gated" material (EBIG material), since charge
injected electrochemically by double-layer ion gating is retained
and the electrolyte is substantially absent in the final state.
[0073] The electrolytes used for process step (a) can be any of the
many aqueous and organic electrolytes later-described. While
solid-state electrolytes can be employed, they are generally less
desirable than liquid electrolytes having electrolyte components
that are easily volatilized in step (d) of the process.
Electrolytes having high redox stability (such as the below-listed
ionic liquids) are especially useful when high degrees of electrode
charge injection are needed. The composition of material A
generally depends upon the intended use of the charge-injection
modified material, and these compositions for various applications
will be later-elaborated. The composition of electrode B is
relatively unimportant unless the charging of this electrode is
desirably used for the production of material that is oppositely
charge injected to that of electrode A (or unless specific
properties changes are needed for electrode B in a device
application). If the function of electrode B is merely to provide a
counter-electrode to material A, the capacitance of this electrode
is typically at least twice that of electrode A, so that most of
the applied potential between electrode A and electrode B is
applied across electrode A. The potential applied during step (c),
the time evolution of this potential, and the duration of this
potential depend upon the redox stability window of the electrolyte
and whether or not redox processes occurring in the electrolyte
will harmfully degrade either the electrolyte or the electrodes. It
is simplest to use either constant current or a constant
inter-electrode potential between electrode A and electrode B.
However, it is useful for maximizing rate for batch materials
processing that resistance compensation be used to control the
inter-electrode potential, in order to insure that both rapid
charge injection occurs and damage to either the electrodes or
electrolyte due to excessive applied potentials is avoided. J. N.
Barisci et al. (Journal of Smart Materials and Structures 12,
549-555 (2003)) describe the use of resistance compensation for
another use (electrochemical electromechanical actuators), and the
concepts described by these authors are applicable for the present
application.
[0074] One measure of the degree of charge injection is the
magnitude of the deviation of electrode potential from the
potential of zero charge. This deviation for an electrode element
that is not substantially contacted with electrolyte is typically
above about 0.1 V for selected invention embodiments. More
typically, this deviation should be above about 0.4 V for these
invention embodiments.
[0075] The process step of charging an electrode immersed in an
electrolyte can be accomplished under conditions that optimize the
degree of infiltration of the electrolyte within the pore space of
the electrolyte, since this optimizes the degree of charge
injection achieved at a particular potential (via increasing the
realized capacitance of the electrode material in the electrolyte).
For the purpose of increasing electrolyte infiltration by improving
electrolyte wetting of the electrolyte, a series of potentials can
be applied whose effect is to remove charge from the electrode and
then to re-inject charge into the electrode. For example, a
constant voltage scan rate can be used to cycle the applied
electrode between minimum and maximum potentials one or more times,
so as to leave the electrode in a desired state at the end of the
cycling process. The point is that, while charging can be usefully
accomplished for regions of the electrode that are not directly
contacted by the electrolyte, the degree of charging achieved at a
particular potential is generally less than that achieved if the
electrode is in contact with the electrolyte. For electrodes that
are difficult to infiltrate with electrolyte, and situations where
multiple cycling does not undesirably increase process cost or
degrade electrode structure, this process of cycling, for the
purpose of enhancing charge injection, can typically be
accomplished at least about three times after the initial charge
injection. This electrochemical cycling process includes the steps
of applying a series of potentials whose effect is to remove charge
from the electrode and then to re-inject charge into the electrode,
so as to thereby increase the realized gravimetric capacitance of
the electrode.
[0076] Processes of invention embodiments for employing charge
injection to substantially change the properties of a largely
electrolyte-free porous material can be applied to various forms of
these materials, such as sheets, fibers, and powders. Various
methods can be used for convenient non-faradaic electrochemical
charge injection into powders, such as nanofiber powders. A first
method is to assemble these powders into a solid electrode form,
such as a sheet, use this sheet as an electrode for predominantly
non-faradaic charge injection, and then to break these sheets into
powders.
[0077] A second method is to disperse the powder in electrolyte, in
at least one compartment of a separated-compartment electrochemical
cell (where the anode compartment and cathode compartments are
separated by either an ion conducting membrane or a porous frit).
Agitation of the dispersion of powder in the electrochemical cell
(such as by stirring or other electrolyte flow process) brings the
powder particles or fiber into intermittent contact with the
electrode of this compartment, thereby permitting non-faradaic
charging of the powder. The opposite compartment of this
electrochemical cell can have a conventionally-shaped electrode
(such as a sheet) and can, optionally, also contain an agitated
electrolyte-dispersed powder that can be either the same or
different from the powder in the other cell compartment. An
advantage of this arrangement is that two powder samples can be
charged (albeit with opposite charge) during the charging step of
the process.
[0078] A third method is to use a moving conducting belt to carry
the powder into and out of the electrochemical cell, wherein this
moving belt is part of the electrode. The opposite electrode
(typically in a separate electrode compartment) can either include
a conventionally configured electrode (such as a sheet), or can
contain the same type as the belt-electrode powder delivery system
as for the other mentioned electrode.
[0079] A fourth method is to disperse the powder in the electrolyte
of an electrode compartment, and to configure this compartment so
that the effect of gravity is to provide contact between the powder
and the electrode of this compartment (which is typically a planar
electrode). In this method the electrolyte is selected to be either
substantially heavier or substantially lighter than the powder. In
the former case the electrode of the compartment is in the lower
most region of the electrolyte compartment, and in the latter case
the electrode is in the upper most region of this compartment.
[0080] After predominately non-faradaic charge injection, the
charge injected particles can be removed from the electrolyte and
optionally washed and dried without completely losing the injected
charge. Subsequently, the charge injected particles can be
redispersed in another carrier that is useful for applications,
such as normal saline solution for medical applications.
[0081] Nanoparticles typically aggregate together, and this
aggregation can limit properties obtainable before and after
subsequent processing. One benefit of charge injection, as well as
the retention of this charge injection in substantially dried
states, is a decrease in the degree of this aggregation and the
ability to retain this decreased aggregation during subsequent
processing, such as in the formation of nanoparticle/polymer
composites (like carbon nanotube/polymer composites).
[0082] Because of the free energy increase due to the contribution
of high surface area of nanoparticles and nanoparticles aggregates,
these compositions can be used as high energy explosives and
propellants. Methods of invention embodiments provide a way to
appropriately adjust (and increase) the energy release capabilities
of explosives and propellants based on nanoparticles. The energy in
taking a non-capacitively charged nanoparticle from the potential
of the charged state (V) to the potential of zero charge
(V.sub.pzc) is 1/2C.sub.aA(V-V.sub.pzc).sup.2, where C.sub.a is the
capacitance per unit area and A is the effective surface area of
the nanoparticle. This energy can contribute to the energy released
during explosions and propellant operation. Also, the release of
capacitively-stored energy can be used to initiate explosion upon
electronic contact of nanoparticles containing either opposite
charge or differing amounts of charge of the same sign. Such
explosion can be initiated by mechanically contacting (and thereby
electronically contacting) two or more solids comprising particles
having differing signs or differing extents of charge injection.
One application for such mechanical initiation of explosion is in
safety air bags for vehicles. In order to maximize the gravimetric
energy release potential of the charged explosives and charged
propellants of present invention embodiments, it is useful that the
counter ions used for charge injection have high free energy with
respect to reaction products produced by explosion and propellant
burning. In addition, to this application of non-capacitive
charging of nanoparticles to increase the energy release during
explosions or during the burning of a propellant, the additional
heating effect associated with charging can be used to increase the
effectiveness of chemical warming elements based on chemical
reactions involving nanoparticles, and to regulate this warming
process by control of charge release from nanoparticles. Like the
warming devices of the prior art, these warming devices can be used
in shoes and other articles of clothing to increase the comfort
level of individuals exposed to freezing temperatures. One or more
components of the above-mentioned charged energy release materials
can usefully comprise electronically-conducting fiber, such as spun
carbon nanotube fibers.
[0083] The process of charge injection changes the surface tension
of particles, which can aid in dispersing particles in various
materials, such as paint manufacture and generic polymer composite
manufacture from either solution or melt phases. Also, this
tunability in surface tension can aid in achieving appropriate
dispersion of biologically-active charged particles for medical
applications. Examples are the delivery of biochemically active
agents that are either the counter-ions to the injected charge or
the electronically charge injected particle, or a combination
thereof. In addition to more direct biochemical effects, the
biochemical activity can result from heating the charge-injected
particle using actinic radiation, such as infrared radiation or
microwave radiation. Biochemically-active ions or ion-solvating
species can either be delivered to the particles during the
original charge injection process or during ion exchange or ion
transformation processes like those described in subsequent
paragraphs. A sometimes achievable benefit of using these latter
processes is that valuable biochemically-active agents can be
accommodated with less waste of these agents during processing.
These species can be any of the various biochemically-active
species that can serve as either anions, cations, or ion solvating
species, like DNA, RNA, polypeptides (such as an enzyme, antibody,
or aptamer).
[0084] After predominately non-faradaic charge injection (either
for the particles or a nanostructured solid), the charge-injected
material can optionally be contacted to a second material, called
an "ion modification material," that either (1) provides ions that
replace the function of the initial ions that are originally the
counter charges to the injected electronic charge, or (2) reacts
with the initial ions to provide ionic species, which then serve as
counter ions to the injected electronic charge. The contacting
material is here called an "ion exchange material" or an "ion
transformation material," depending upon whether the result of
contact with the charge-injected material is the ion exchange
process of (1) or the ion transformation process of (2). The
benefit of such exposure can be to replace the original ions with
ions that provide either increased environmental stability or
increased response for the detection of agents in sensing
applications. Alternatively, the result of this contact can be to
provide ions that can be released as drugs or other chemicals
during electronic discharge of the charge injected material. The
above mentioned contact can be by exposure of the charge-injected
material to gaseous, liquid, or solid states of either the ion
exchange material or the ion transformation material. One example
of such processes is by immersing a hole-injected material
containing BF.sub.4.sup.- into a salt solution containing the more
environmentally stable PF.sub.6.sup.-. This exposure will result in
at least partial replacement of the ambient unstable BF.sub.4.sup.-
with the more stable PF.sub.6.sup.-.
[0085] Another example, useful for medical applications, involves
immersion of the charge-injected material into an aqueous solution
containing DNA or a polypeptide, such as an enzyme, antibody, or
aptamer. This exposure can result in the exchange of the original
ion with an ion of the biological material and/or interaction of
the original ion (such as H.sup.+) with the biological material to
produce a new ion that comprises the chemical compositions of both
the original ion and the biological material. This biological
material is now tightly bound to the charge injected material until
decay of this charge injection causes release of the biological
agent, which can be used for drug delivery. This decay of charge
injected, and subsequent drug release, can result from exposure of
the charge-injected material to redox active materials (such as
those present in the body), by contact between hole injected and
electron injected materials, the application of an electrical
potential that causes current flow that decreases charge injection,
and by heating the charge injected material (such as by exposure of
the charge injected material to infrared radiation in the infrared
transparency region of mammalian tissues or exposure to microwave
radiation), or by exposure to other actinic radiation (such as
.gamma.-ray, x-ray, beta particle, or alpha particle radiation).
Most generally, the ion modification material can comprise a
biologically active component, or one that becomes biologically
active as a result of exposure to the charge-injected material. In
some invention embodiments it is useful for this biologically
active component to be a radioactive component, so that delivery of
the charge-injected material to the mammalian body provides the
ability to deliver radiation to targeted tissues. Especially for
these biological applications, after the contact processes with the
ion modification material, excess ion modification material can be
usefully removed from the charge-injected material (such as by
drying). Prior to exposure to the mammalian body, the
charge-injected material thereby obtained can be immersed in
another agent that facilitates delivery of the biologically active
material to the mammalian body. This agent is typically
substantially biocompatible, like normal saline solution. While
diverse materials can be used as the charge-injected material for
drug delivery, carbon nanotubes and shell-core metal particles are
especially useful for some application embodiments, since these
materials provide infrared absorption in the transparency region of
the mammalian body. For the preparation of suitable shell-core
metal particles that can be non-faradaically charge injected using
the present art, see C. L. Loo et al., Technology in Cancer
Research & Treatment 3, 33-40 (2004) and references
therein.
[0086] Semiconducting nanoparticles, especially those employed for
color-based biochemical sensing, are useful for the practice of
invention embodiments. Examples include, but are not limited to,
nanoparticles of ZnS, ZnSe, CdS, and CdSe. The particles for these
invention embodiments are typically less than 200 nm in their
smallest dimension. Using methods that are above-described, such
nanoparticles (or functionalized derivatives thereof) can be
conveniently charged electrochemically in an electrolyte (either
containing the desired counter ion, or a counter ion that can be
replaced by the desired counter ion during subsequent processing
steps). This charging can be used to modify the dispersability of
the charged, dried particles in liquids or melts, including polymer
solutions or melts used to form nanoparticle composites. In one
application mode, the semiconductor nanoparticles (charge injected
using electrochemical processes of this invention) are used in
composites that are employed for light emitting diodes and
photovoltaic cells. In another application mode, the
charge-injected nano particles are used as color-based sensors,
wherein the electronic charge injection and associated counter ions
effect the aggregation of these particles in response to contact
with an analyte (which can be biological). Such aggregation (or the
interaction of the analyte with single charged nanoparticles)
provides the spectroscopic response used for sensing, which is
typically a color change or a change in luminescence.
[0087] The charge-injected materials of invention embodiments can
be used as scaffolds for the growth of tissue in either culture
media or in organisms, including humans. Also, this charge
injection can be used to increase biocompatibility for devices
implanted in the body. These invention embodiments utilize
Applicants' surprising discovery that electrochemically-injected
charge is substantially retained when a material is removed from
the electrolyte and optionally washed and or dried.
[0088] Biocompatibility of the surfaces of devices implanted in the
body is improved in some invention embodiments by overcoating these
surfaces with a porous material having a high gravimetric surface
area, such as glassy carbon. Using the device with the over coated
porous material as an electrode in an electrochemical cell, charge
is injected predominantly non-faradaically in the porous coating.
The type (holes or electron) and degree of charge injection, as
well as the nature of the counter ion to the injected charge, can
affect biocompatibility. The injected ions can be either those of
the original electrolyte or those that substitute for the original
ions during subsequent processes, such as by immersion of the
charge injected device into a solution containing the replacement
ions. As an alternative to this charge injection into porous
coatings that are preformed on the article to be implanted, such as
a pacemaker or an artificial heart, the electronic charge injection
can be in a powder (such as nanofibers) that are first charge
injected and then used to overcoat the article. The sign of
electronic charge injection is typically positive, although
negative electronic charge injection can also be used (with some
decrease in the lifetime on injected charge in the body). The
counter ions for the electronically injected charge can include
various inorganic, organic, and biochemically derived species.
Examples are Na.sup.+, Cl.sup.-, proteins (especially enzymes that
are cellular growth factors), antibiotics, DNA, and RNA. Nanofibers
(particularly nanofibers configured as porous sheets and
macrofibers) are particularly useful as substrate materials for the
growth of tissue either in culture media or in animal or human
bodies. Considerations on the choice of counter ions and the sign
of charge injection are similar to those above-recited for surface
coatings for implanted devices.
[0089] Applicants' discovery that electrochemically injected charge
is stable in the absence of electrolyte also enables the tuning of
nanostructured materials properties for materials absorption and
desorption, such as for hydrogen storage. The key point here is
that charge injection (and associated ion migration) selectively
changes the interaction energy of materials with a nanostructured
substrate (such as nanostructured fibers, sheets, and powders) and
therefore changes the absorption capabilities of the substrate
material. Discharge of this injected charge (such as by contacting
oppositely charge-injected substrate materials) can aid in the
release of absorbed materials.
[0090] Materials used for properties tuning by non-faradaic charge
injection (including tuning for the above-mentioned chemical and
drug release processes) generally have an achievable gravimetric
capacitance for predominately non-faradaic charging that is
typically above about 0.1 F/g, more typically above about 1 F/g,
and most typically above about 10 F/g. By achievable gravimetric
capacitance, it is meant the capacitance measured when using an
electrolyte and potential range that maximizes non-faradaic
charging and still yields predominately non-faradaic charging. The
gravimetric surface area of the materials used for properties
tuning by non-faradaic charge injection have an achievable
gravimetric surface area (measured in nitrogen gas using the
standard Brunauer-Emmett-Teller, BET, method) of typically above
about 0.1 m.sup.2/g, more typically above about 5 m.sup.2/g, and
most typically above about 50 F/g.
[0091] Volatile electrolytes can be removed from the charged
electrode material most simply by just evaporating the volatile
component of the electrolyte. Alternatively or additionally, the
electrolyte can be removed by washing with a second liquid that is
typically not an electrolyte for certain invention embodiments. For
these invention embodiments this second electrolyte is typically
substantially free of any salt. Ideally, this second liquid should
be either miscible with the electrolyte or capable of dissolving
ions of said electrolyte. This washing step or a sequence of
washing steps using either the same or different liquids can be
usefully accomplished either before or after an optional step in
which the original electrolyte is wholly or partially removed by
volatilization.
[0092] Rather than removing the original electrolyte from the
charge injected electrode, charge injection can be accomplished at
a higher temperature where this electrolyte serves as an effective
electrolyte and then this original electrolyte can be cooled to a
lower material use temperature where the original electrolyte
material has such low ionic conductivity that it does not
effectively serve as an electrolyte. This low ionic conductivity is
preferably below 10.sup.-6 S/cm. This higher temperature is
typically one at which the electrolyte is substantially liquid and
this lower use temperature is typically one in which the
electrolyte is substantially solid. As an alternative to using
temperature change to convert a liquid electrolyte to a
substantially non-conducting material, this transformation can be
accomplished by polymerizing the liquid electrolyte, for example,
using thermal annealing, introduction of a catalyst, or exposure to
actinic radiation (typically visible, ultraviolet, or higher energy
radiation).
[0093] The liquid that is used to wash the electrolyte from the
electrode is typically removed by volatilization. This liquid used
for washing can optionally and usefully be the solvent base of the
electrolyte (such as the acetonitrile of an original 0.1M
tetrabutylammonium hexafluorophosphate/acetonitrile electrolyte or
the water in an original 1 M aqueous NaCl electrolyte). However, in
other useful embodiments of the present invention, the liquid that
displaces the electrolyte is polymerized while in contact with the
electrode material. This polymerization can be accomplished, for
example, using thermal annealing, introduction of a catalyst, or
exposure to actinic radiation (typically visible, ultraviolet, or
higher energy radiation).
[0094] Diverse properties can be tuned, and in many cases
dramatically changed, as a consequence of the above process for
electrochemically injecting charge non-faradaically and retaining
this charge in the electrolyte-free state. Some of the effects by
which charge injection can change material properties are: (1) the
direct effect of changing band filling, (2) the effect of volume
expansion and associated dimensional change due to charge
injection, and (3) the effect of creating surface dipoles involving
the injected electronic charge and correlated ions on the material
surface. Material properties tuned by the processes of invention
embodiments include, for example, electrical conductivity,
absorption and reflectivity (including color), thermopower, thermal
conductivity, surface energy, and dielectric constant. Materials
with low conductivities in the non-charge-injected state (like
forms of carbon, conducting polymers, and doped and undoped
semiconductors, exemplified by metal oxides and metal sulfides) can
be used for invention embodiments that benefit from large
charge-injection-induced changes of conductivity, absorption and
reflectivity (including electrode color), thermopower, thermal
conductivity, surface energy, or dielectric constant.
[0095] Physical properties, such as electrical conductivity,
optical absorption, magnetization, magnetoresistance,
electromagnetic shielding properties, and the critical temperature
of superconducting transition (T.sub.c) depend very strongly on the
number of charge carriers at the Fermi level. Optional and useful
materials of invention embodiments include those that maximize the
tunability of these properties that can be achieved by
predominately non-faradaic electrochemical charge injection. Such
materials include those providing (1) a low concentration of charge
carriers (electrons or holes) at the Fermi level for the uncharged
state, and (2) a strong dependence of Fermi energy on the amount of
charge injection. This strong dependence of Fermi energy on charge
injection is typically characteristic of the singularities in
density of states found for low dimensional conductors. Hence, 1-D
materials (like nanofibers and conjugated polymers like
polythiophene) and 2-D layered materials (like cuprates of
chalcogenides) are included in some compositions. However,
materials that can be intercalated (and thereby charge-injected
faradaically) are optionally charged predominately
non-faradaically. This can be done by either (1) choosing the
potential used for charge injection to avoid proximity to the
potentials where faradaic redox reactions occur or (2) by employing
anions and cations that have unsuitable size for intercalation
under the kinetic conditions used for charge injection. For
example, bundled single-wall carbon nanotubes can be intercalated
if the applied potential is either too high or too low. This
intercalation can be avoided by suitably choosing the potential
range so that intercalation does not occur or by using ions in the
electrolyte that are too large for facile intercalation (M. Stroll
et al., Chem. Phys. Lett. 375, 625-631 (2003)).
[0096] Materials used for electronic conductivity tuning by
non-faradaic electrochemical charge injection are optionally and
preferably semiconductors when the goal is to provide the maximum
dynamic range of tunability. Materials having singularities in
density of states near the Fermi level can be used as materials
having tunable electronic conductivity. These include the various
well-known nanofibers, like single-wall and multi-wall carbon
nanotube fibers.
[0097] The non-faradaically induced increase in electrical
conductivity of essentially electrolyte-free materials can be
usefully employed for making materials that combine high electrical
conductivity with high optical transparency. These transparent
conducting electrodes are of major importance for such applications
as liquid crystal displays, light emitting displays (both organic
and inorganic), solar cells, switchable transparency windows, solar
cells, micro lasers, optical modulators, and optical polarizers.
Inorganic electrodes like ITO (indium tin oxide) degrade on bending
and require costly vacuum based deposition methods. The embodiments
of this part of the invention combines Applicants' discovery of
processes for dramatically enhancing the electrical conductivity of
nanostructured carbon nanotubes with prior-art discoveries related
to the application of uncharged nanostructured materials for making
transparent conductors.
[0098] The prior art invention of A. Rinzler and Z. Chen
(International patent number WO 2004/009884 A1 and
PCT/US2003/022662 on "Transparent Electrodes from Single wall
carbon nanotubes") describes methods for making electronically
conducting, transparent SWNT-containing sheets. The described
process enables the achievement of percolation between SWNTs in
uniform coatings or films. The achieved sheet resistance is in the
range of 200 ohm/sq for an optical transmission of 30%, which can
be decreased to 50 ohm/sq for very thin films less than a hundred
nanometers thick). However, the sheet resistance should be less
than 10 ohm/sq for applications in light emitting displays, solar
cells, and other current dependent devices.
[0099] Presented here are invention embodiments in which the
electrical conductivity of single-wall carbon nanotubes is
increased by over an order of magnitude, and retain much or all of
this conductivity enhancement when the electrolyte is removed. An
improvement that Applicants bring is an increase in the level of
electrical conductivity that can be obtained, while still
maintaining the desired degree of optical transparency. Carbon
nanotubes that are non-faradaically injected and largely
electrolyte free are especially useful for invention embodiments
directed to highly conducting, optically transparent conductors.
These conductors can be in the form of sheet composites or sheet
coatings. Also, the carbon nanotubes can be single-wall nanotubes,
wherein the nanotubes can be either unbundled or having a small
bundle diameter. These single-wall carbon nanotubes can be hole
injected, since Applicants have found that hole-injected carbon
nanotubes provide much more stable electrical conductivity
enhancements than do electron-injected carbon nanotubes. In order
to insure long lifetimes for these non-faradaically hole-injected
carbon nanotubes, it is generally important to use binder materials
that do not contain electron donating impurities having redox
potentials that enable reaction with the holes on the nanotubes. To
insure that this is the case, any binder composition should have
sufficiently high ionization potential that electron transfer does
not occur to the hole-injected nanotubes. Many conventional
polymers, like polyethylene and polypropylene, have this desired
characteristic. Conventional methods can be used to treat binder
compositions so as to remove trace reactive electron donor
impurities that might react with the holes in the hole-injected
carbon nanotubes.
[0100] The work function of electrode materials generally is very
important for applications involving either hole or electron
injection in organic light emitting displays and for charge carrier
collection for solar cells, as well as for related devices. The
non-faradaic charge injection of invention processes provides Fermi
level and work function shifts that can be used to optimize this
charge injection. The prior art work on multi-wall carbon nanotubes
(M. Kruger et al., Applied Physics Letters 78, 1291 (2001)) has
shown that the Fermi level increases by up to 0.3-0.5 eV upon
electron injection, and decreases by up to 1 eV for hole injection,
which will provide corresponding charges in the work function.
These prior-art results are from electrochemical charge injection
in liquid electrolyte. The enabling improvement provided by the
present invention embodiments is that it is shown that the
typically required charge injection can be retained in the absence
of contacting electrolyte--either solid or liquid state.
[0101] The electron-injected electrolyte-free nanotubes (with
decreased work function) can be used as electron-injecting
electrodes for such devices as OLEDs. On the other hand, the
hole-injected electrolyte-free nanotubes (with increase work
function) can be used as hole-injecting electrodes. These
hole-injected nanotube electrodes can be used to replace ITO, since
they also have large achievable work function and can be used as an
effective hole injector for OLEDs.
[0102] These device applications involving hole-injected and
electron-injected electrolyte-free nanostructured materials as
charge-injecting electrodes typically utilize device configurations
in which an electrochemical potential is applied to the
nanostructured material during device operation. This is
particularly important for the electron-injected nanostructured
electrodes because charge can only be stabilized on these
electron-injected electrodes in an inert environment, and oxygen
exposure can cause degradation if the degree of electron injection
is high. On the other hand, hole-injected nanostructured electrodes
can easily function without the need for either a continuously or
intermittently applied electrochemical potential to refresh charge
injection. Hence, these hole-injected nanotubes can be dispersed in
a suitable unreactive binder and used as electrodes for devices,
without any need for electrochemical charging after the initial
charge injection. Transparent carbon nanotube fiber composites have
already been used as replacements for low work function Al or Ag
electrodes for plastic donor-acceptor solar cells ("Organic
Photovoltaics" Eds. C. Brabec, V. Dyakonov, J. Parisi and N. S.
Sariciftci, Springer Series in Materials Science Vol. 60, 2003).
These nanotubes were not substantially charge injected. The ability
provided by the present invention embodiments is to
electrochemically tune the work function of the nanotubes for this
application, without compromising desired performance by the need
of the prior art for either dopant intercalation or an imbibed
electrolyte.
[0103] Processes in accordance with some invention embodiments can
be used for changing the properties of largely electrolyte-free
nanostructured superconductors by substantially non-faradaic
electrochemical charge injection. High temperature superconductors
are especially useful, such as members of the YBCO family. The most
preferred compositions include, for example, LaSrCuO.sub.2,
YBa.sub.2Cu.sub.3O.sub.7-x, GdBa.sub.2Cu.sub.3O.sub.7-x,
BiSr.sub.2CaCu.sub.2O.sub.8+x, and related cuprates. The
compositional parameter x is typically in the range of 0.4 to 0.5
for YBa.sub.2Cu.sub.3O.sub.7-x. Further guidance for the
superconductor compositions that are most suitable for
electrochemical non-faradaic charge injection of invention
embodiments can be found in C. L. Lin et al., Applied Physics
Letters 71, 3284 (1997). Although these authors use short-lived
photo-injection of charge to modify the superconducting properties,
Applicants have found that those compositions that undergo the
greatest changes in superconducting properties upon photo-injection
will also undergo the greatest changes when using the non-faradaic
electrochemical charge invention of the present invention
embodiments.
[0104] These superconducting oxides have low electron density at
the Fermi, so non-faradaically injected charge carriers will
significantly change their properties. While the prior art has used
charge injection to modify the T.sub.c of superconductors, this
prior-art work has not used predominately non-faradaic
electrochemical charging of a porous superconductor having a high
gravimetric surface area. See, for example, X. Xi et al. in Applied
Physics Letters 59, 3470 (1991), who use a non-electrochemical
process for switching the superconducting transition temperature of
films of YBa.sub.2Cu.sub.3O.sub.7-x, over a 2 K range. The achieved
resistance modulation in the normal state can be as much as 20% and
1500% near T.sub.c. Unlike the switching of superconducting
properties in the present invention embodiments, the T.sub.c
switching observed by Xi et al. is not practically useable for a
macroscopic bulk superconducting material and is unstable over the
long term if connection to an electrical power source is not
maintained.
[0105] The non-faradaic charge injection of invention embodiment
can be used for tuning superconductors for electronic transport and
for electromagnetic wave shielding and propagation for ultraviolet,
visible, infrared, radio frequency, and microwave frequencies. A
particular application embodiment is as superconducting elements
for active filtering, attenuation, phase shifting, and inter-line
coupling for microwave transmission lines (such as micro strip
line, strip line and co-planar wave guides). Methods for using
superconductors for microwave lines and modulated wave guides are
well known in the art for the microwave and radio frequency bands,
and this technology can be used for application of the
non-faradaically charge-injected superconductors of the present
invention. Superconductors of the invention embodiments may also be
used with photonic crystal arrays involving superconducting and
semiconducting elements to provide switching of photonic crystal
properties in the optional and preferred EM wavelength bands, such
as modulation of either the photonic band gap or the width and
cut-off frequency of metallicity gap. Other major applications for
materials made by processes of invention embodiments are as
superconducting transmission lines, and magnets.
[0106] Perovskite manganites with the general formula
R.sub.1-xA.sub.xMnO.sub.3, where R is a trivalent rare earth
element and A is a divalent alkali earth element, are included
among the preferred compositions, where optionally and most
preferably R is La, Pr, Nd, or Sm, A is Ba, Ca, or Sr, and
0.3<x<0.5. The reason for this preference is the sensitivity
of the properties of these materials to EBIG (Electrolyte-Bare
Ion-Gated) charge injection, which makes them especially useful for
EBIG materials and devices. In fact, EBIG charge injection into
these materials can cause transformation between insulating
anti-ferromagnetic and metallic ferromagnetic states.
[0107] The prior art has shown that the subtle balance between the
anti-ferromagnetic insulating state and the ferromagnetic metallic
state can be shifted by application of external perturbations, like
magnetic field (A. J. Millis, Nature 392, 147 (1998) and Y. Tomioka
et. al., Phys. Rev. B 53, R1689 (1996)); electric field (A.
Asamitsu et al., Nature 388, 50 (1997)); high pressure (Y. Morimoto
et. al., Phys. Rev. B 55, 7549 (1997)); exposure to X-rays (V.
Kiryukhin, et. al., Nature 386, 813 (1997)); or exposure to visible
light (K. Miyano, et al., Phys. Rev. Lett. 78 (1997) 4257 and M.
Fiebig, et al., Science 280 1925 (1998)).
[0108] In contrast with these prior-art approaches for the
perovskite manganites, Applicants either induce or sensitize
transitions between the insulating antiferromagnetic state and the
conducting ferromagnetic state by using non-faradaic
electrochemical charge injection that is either tuned or maintained
in the absence of contacting electrolyte. The non-faradaically
electrochemically-induced phase transition of invention embodiments
can be used advantageously to replace the dielectric-based electric
field effects of the prior art (S. Q. Liu et. al., Appl. Phys.
Lett. 76 2749 (2000) and J. Sakai et. al., J. Appl. Phys. 90, 1410
(2001)) in high density, nonvolatile memory devices. The colossal
magnetoresistance of the perovskite manganites can be tuned by the
EBIG (electrolyte-bare ion-gated) methods of invention embodiments.
This tuning can be used for applications where a perovskite
manganites acts as a device channel, which is switched by
non-faradaic electrochemical charge injection from insulating and
anti-ferromagnetic to metallic and ferromagnetic. Such tuning is
usefully employed for the spintronic devices of invention
embodiments, especially where a perovskite manganite is between a
magnetic source and drain. EBIG tuning changes the sensitivity of
the perovskite manganites to transitions induced by either light or
magnetic fields, and these effects can be usefully employed in
devices. Additionally, EBIG materials can also be used for the
control of electromagnetic wave propagation for ultraviolet,
visible, infrared, radio frequency, and microwave frequency
regions, since modulation of electrical conductivity also changes
the refractive index, dielectric constant, absorption, and optical
reflectivity.
[0109] Devices of invention embodiments generally include at least
three elements: a working electrode, a counter electrode and a
possibly multi-component electrolyte material that helps provide an
ion conducting path between the working and counter electrodes.
Electrochemical charge injection in an electrolyte-free electrode
component is accomplished by applying a voltage between at least
two electrodes (a working electrode and a counter electrode) that
are both in partial contact with an electrolyte, where an
uninterrupted path for ionic transport exists between these two
electrodes. Charge injection into regions of the electrode that are
not contacted with electrolyte occurs, in most of the invention
embodiments, by diffusion of the dopant ions on the surface of a
nanostructured material, such as a carbon nanofiber used as the
channel of a field-effect transistor.
[0110] For invention embodiments where very high electrochemical
charge and discharge rates are desirable, the rate of surface
migration of ions in an electrochemical electrode can be increased.
Using at least two electronic contacts (typically at close to
opposite ends of the electrochemical electrode), a suitably high
current is applied along the electrochemical electrode in the
direction where ion migration is desirable. This current is
referred to herein as the intra-electrode ion-migration enhancement
current. Two processes produce enhanced ion migration rate from
this applied intra-electrode electronic current. The first is
electrode resistive heating, which increases ion mobility. The
second is the "electron wind force". This electron wind force is
well known to cause failure of small cross-section metal wires on
circuit boards (through causing migration of atoms in the wire by
the combined effects of electrostatic interaction with the electric
field and momentum transfer with electronic carriers due to
scattering). Reversing the direction of intra-electrode
ion-migration enhancement current reverses the direction of the
electron wind force for a given ionic species (together with
possible solvation sphere), so increased migration rates can be
obtained for both electrode electrochemical charging and
discharging. To achieve this beneficial effect of rate enhancement
on both electrode charging and discharging, the direction of the
intra-electrode ion-migration enhancement current can be reversed
in direction when transitioning between electrochemical charge and
discharge processes. Use of these physical processes to enhance
charge and discharge rates for an electrochemical device is quite
useful and has not been previously shown in the prior art.
[0111] FIG. 9 schematically illustrates a means used in invention
embodiments for injecting charge into a high-surface-area electrode
that is only partially contacted with electrolyte. The charged
state is pictured, where 900 is the electrode with positive
injected electronic charge (electrostatically balanced by pictured
anions that are in close proximity, symbolized by the white
spheres) and 901 is the counter electrode with negative injected
charge (electrostatically balanced with pictured cations,
symbolized by black spheres). The nanostructured electrode
materials for both 900 and 901 are single-wall nanotubes, although
virtually any type of nanostructured conductor can be used (as long
as this conductor does not undergo degradative intercalation in the
potential range of device operation). As will be later described,
material selection depends upon the device type and performance
needs. Component 902 is an electrolyte, which only partially
contacts electrodes 900 and 901. Element 903 is the variable
voltage power source and associated leads that electrically connect
to the two nanotube electrodes of 900 and 901. An applied potential
from 903 injects charge of opposite sign in the two pictured
single-wall carbon nanotube electrodes 900 and 901. The required
counter ions diffuse along the nanotube surfaces to enable this
electronic charge injection. Reversing the direction of current
flow, by changing the applied potential, causes the ions to diffuse
from the nanotube surfaces back to the electrolyte 902.
[0112] FIG. 9 pictures an arrangement where the ions used for
electrode charging are stored in the uncharged state in the
electrolyte. Alternatively, ions can be shuttled between the
electrode elements 900 and 901 (or from and to optional additional
electrodes) during the charge and discharge processes.
Additionally, a combination of these electrolyte storage and
inter-electrode shuttle processes can be employed. As will be later
elaborated, one or more of the electrodes in an electrochemical
device can be predominately faradaically charged and discharged
during device operation. Typically, a region of at least one device
electrode is predominately non-faradaically charged and discharged
during normal device operation.
[0113] For comparison with the electrochemical semiconductor device
of the present invention shown in FIG. 11, FIG. 10 illustrates a
prior-art electrochemical semiconductor transistor device. The
illustrated prior-art device of FIG. 10 is referred to in the
literature (M. Kruger, Applied Physics Letters 78, 1291-1293
(2001)) as being a "liquid-ion gated" device. The device is
separated from the substrate 1000 by an insulating layer 1001
(typically SO.sub.2). The device channel and an electrochemical
electrode (1002) is a semiconducting carbon nanotube, which is
contacted by metal source and drain electrodes (1007 and 1008,
respectively). The device is liquid-ion gated by using the variable
potential source 1006 and associated wiring to apply a potential
between the micropipette (1004) enclosed Pt wire electrode 1005 and
the carbon nanotube electrode 1002, which is analogous to the
channel of a conventional field effect transistor. The electrolyte
1003 covers both the nanotube channel electrode and the Pt wire
counter-electrode 1005. The device operates by using the power or
signal source of 1006 to electrochemically inject charge into the
nanotube electrode/channel 1002. This injected charge changes the
electrical conductivity of the nanotube channel 1002, thereby
varying the current that flows through the channel in response to a
voltage difference applied between source electrode 1007 and drain
electrode 1008. This device is called a liquid-ion gated FET, since
the nanotube channel 1002 is immersed in a liquid electrolyte and
the ions in this electrolyte are needed for the charge injection
(i.e., gating) process. This or other prior art does not recognize
that electrochemical double-layer charge injection can result for
electrode regions that are not in electrolyte, which is the reason
for the illustrated complete immersion of the nanotube in the
electrolyte. This immersion of the channel 1002 in a liquid
electrolyte is clearly problematic for ordinary transistor
applications. The device of FIG. 10 could be used like a Chem-FET
(chemical sensor based on a FET) for detecting materials dissolved
in the electrolyte, by using the effect of these materials on
double-layer charge injection. However, if the material to be
detected is a gas, this material must first dissolve in the
electrolyte--which decreases device response rate and
sensitivity.
[0114] FIG. 11 schematically illustrates an electrochemical
transistor of the present invention embodiments that is not
liquid-ion gated. This is an EBIG device since the electrolyte is
not deposited to completely overlap the device channel, and the
region of the device that is bare of electrolyte facilitates device
function. The device is built over a trench (1108) in an insulating
substrate (1107). There are two device channels, and neither of
these channels is in contact with a liquid electrolyte. More
generally important, the component of each channel that largely
determines gate resistivity is not surrounded by either a liquid
electrolyte or a solid electrolyte--although at least an ion
component of this electrolyte must be contacting (together,
optionally, with solvating species). There are two source and drain
electrodes (1100 and 1101, respectively) for the first leg of the
device and two source and drain channels (1102 and 1103,
respectively) for the second leg of the device. Likewise there are
two semiconductor channels. The channel for the first leg of the
device (1104) and the channel for the second leg of the device
(1105) can be, for example, nanofibers (such as carbon SWNTs). A
solid-state electrolyte (1106) lays over part of the source and
drain electrodes and the channel for each leg of the device.
[0115] Device operation is as follows: Application of a voltage
difference between electrode 1100 and 1102 (or gate electrode 1100
and 1103) causes predominately non-faradaic charge injection of
opposite sign in channels 1104 and 1105. This charge injection is
enabled by the surface diffusion of cations (to electrostatically
compensate for motion of electrons) for the more negatively charged
channel and by the surface diffusion of anions (to
electrostatically compensate hole motion) for the more positively
charged channel. This charge injection in the active channel
lengths (predominately the channel lengths that are suspended over
trench 1108) changes in a controllable way the electrical
conductivity of the channels, which is indicated by a change in the
current passing between electrodes 1100 and 1101 in response to an
applied potential between these source and drain electrodes (and
between 1102 and 1103 in response to an applied potential between
these electrodes).
[0116] The device of FIG. 11 (and a related device of FIG. 12) can
be operated as a replacement for a field-effect transistor, or in
such applications as information storage or gas sensing. Because of
the dual-leg nature of the FIG. 11 device, this device provides two
transistor elements, two information storage elements, and two
sensors that are controlled in a correlated manner.
[0117] When used as a chemical sensor for gas state species, a gas
phase species interacts with the channel in a way dependent on the
charged state of this channel, to thereby provide a change in the
resistivity of this channel. Importantly, this and related devices
of invention embodiments can be used to do something that has
heretofore been impossible--to do the equivalent of cyclic
voltammetry for substances that are directly delivered to the
device channel (or channels) in a gaseous state. This unprecedented
device capability results from the fact that materials in the
environment of the channel will undergo redox reaction with charges
on the channel.
[0118] Like for ordinary liquid state cyclic voltammetry, the
existence and rate of such reaction for a particular species in the
gaseous state will depend upon the redox potentials for these
species. In standard liquid-state cyclic voltammetry, one scans
electrode potential at a constant rate and then plots the resulting
inter-electrode current flow versus the applied potential. This
process can be used for the present cyclic voltammetry, with the
important difference that the sensed material is in the gas phase.
For the device of FIG. 11 and devices of related invention
embodiments, additional sensor information can be collected that
helps uniquely characterize gas phase species. This information
includes the effect of such gas-state-delivered species on the
channel resistance as a function of channel potential, which is
typically measured with respect to a reference electrode. Instead
of detecting redox processes by using voltage scans and detecting
the resulting current flow, a desired current can be caused to flow
between electrodes, and the resulting potential of working and
reference electrodes can be monitored versus time. Additionally,
resistive or optical heating of the channel can be used to cause
desorption of materials from the channel, and thereby provide
another means for characterizing an analyte.
[0119] The device pictured in FIG. 11 does not provide a reference
electrode, which is typically used in liquid state cyclic
voltammetry. Such a reference electrode, or more than one reference
electrode, can be usefully incorporated in the device of FIG. 11 by
including a reference electrode material (such as a platinum wire
or a platinum film) in electrolyte 1106. This reference electrode
(or a multiplicity of such reference electrodes) should not be in
electrical contact with other electronically conductive elements of
the device. Measurement of the potential of a channel with respect
to a reference electrode (which can be located in close proximity)
enables placement of channel potentials on an absolute scale, so
that the redox potentials of detected gas phase species can be most
reliably determined. The two channel materials in FIG. 12 need not
be identical or even comprise a nanofiber. Specifically, the use of
film strips for one or more of these channels is also included in
some invention embodiments. However, these channel materials should
typically have both semiconducting and highly conducting states and
the possibility of transitioning between these states as a result
of charge injection. One of these channel materials can optionally
also be a material that is predominately charged faradaically in
the potential range of device operation, such as a conducting
organic polymer or vanadium pentoxide nanoribbons (see G. Gu et
al., Nature Materials 2, 316-319 (2003) and provided references for
description of the synthesis, properties, and self-assembly of
these vanadium pentoxide nanoribbons).
[0120] FIG. 12 schematically illustrates a second electrochemical
transistor device of the present invention that does not use
liquid-ion gating, and which could be used for information storage
or gas sensing. Like the device of FIG. 11, this is also a EBIG
device--but unlike the device of FIG. 11 the present device has
only one channel. The device is configured over a trench (1207) in
an insulating substrate (1206). The operation and benefits of this
device are similar to that of the device in FIG. 11. A potential
applied between gate electrode 1202 and source electrode 1201
controls the amount of charge injection in the channel 1203, which
can be a semiconductor when there is no charge injection. This
charge injection charges the electrical conductivity of the
channel, which is measured by applying a voltage between source
electrode 1201 and drain electrode 1200 and measuring the resulting
current flow through channel 1203. A material 1204 capable of
charge injection overlies the gate electrode and is electronically
part of this gate electrode. This material can undergo charge
injection either predominately non-faradaically or predominately
non-faradaically in the gate-source voltage operation range of the
device. The solid-state electrolyte 1208 contacts the material 1204
and the channel material 1203, and provides an ion-conducting path
between these elements. However, the measured conductivity of the
channel is in large part determined by regions of the channel that
do not contact the electrolyte, and the charging of this channel
can be predominately non-faradaic in the typically utilized
operation range of the device.
[0121] Well-known prior-art methods can be used for the fabrication
of the devices of FIGS. 11 and 12, and these methods will be later
provided in a general discussion of methods generically useful for
the fabrication of devices from material elements that can have
nanoscale dimensions. Examples of the many dozens of papers in the
literature that teach one skilled in the art to fabricate
individual nanofiber devices are J. A. Misewich et al., Science
300, 783-786 (2003), L. Gangloff et al., Nano Letters 4, 1575-1579
(2004), and G. S. Duesberg et al., Nano Letters 3, 257-259
(2003).
[0122] Devices like that shown in FIG. 12 can also be used a
chemical sensors of materials that are delivered in liquid states.
A variety of processes can be used for the sensing. One process is
ion exchange between the channel material and ions of the analyte.
Another process is reaction of ions of the analyte with ions
originally present as counter ions to the electronically injected
charge. A third process is redox reaction of injected charge with
the analyte. Selectivity of response can be achieved for sensor
applications by incorporating biochemically selective species in
the device channel material, such as DNA, RNA, and polypeptides
(including, for example, antibodies, enzymes, and aptamers). This
incorporation can optionally be accomplished for the device channel
by exposure of the device channel in the non-faradaically charge
injected state to large biological molecules, like DNA, RNA, and
polypeptides. Such exposure can result in either the replacement of
the original ions with such large molecules in the charged state or
reaction of the original ions with these large molecules to produce
new ion types.
[0123] These biological molecules (like DNA, RNA, and polypeptides)
are normally charged, so they are suitable counter-ions to
electronic materials injected in an electrode material. Also, they
can provide recognition functions that are very useful for sensing.
However, because of large molecular size they do not have high
mobility on surfaces. Hence, these materials are most preferably
delivered to a surface (such as a device channel) from a liquid
media. This delivery (such as from normal saline solution) can be
optionally accomplished during device fabrication, and this liquid
and unneeded salt can be optionally removed later during device
fabrication by evaporation and washing processes. Because of the
low mobility of these large biological molecules on surfaces,
switching of the degree of ionic charge on these molecules during
device operation is typically accomplished by using electrolytes
that provide ionic species having high mobility, like the H.sup.+
ion in Nafion. In such case, the electronically controllable degree
of incorporation of the H.sup.+ ion in the biomolecules determines
the amount of ionic charge on the biomolecules, which in part
determines the sensitivity of a channel comprising such a
biomolecules to analytes, such as a structurally matched strands of
DNA. Hybridization of a DNA strand on a device channel with an
analyte DNA strand is an especially useful sensing mode.
[0124] Is well known in the prior art that faradaic charge
injection can profoundly affect the optical properties of material
and that these faradaically-changed optical properties can be
maintained in the absence of an electrolyte. However, prior-art
investigators have not discovered that optical properties changes
can result from non-faradaic charge injection that is accomplished
without direct contact of an ionically conducting material (such as
an electrolyte or electronically intercalated conducting
polymer).
[0125] The optical device schematically illustrated in FIG. 13,
which is illustrative of many related devices of the invention
embodiments, utilizes the above-described discoveries of
Applicants'. Devices of this and related types of the invention
embodiments provide the switching or optical properties of a
material region that is not directly contacted with an electrolyte
or an intercalated material that provides ion transport. Typical
applications of these devices are gas sensors based on surface
enhanced Raman (SERS) or fluorescence, infrared camouflage layers,
and electronically switchable photonic crystal mirrors.
[0126] FIG. 13 schematically illustrates an optical gas sensor,
based on the surface-enhanced Raman effect, that uses
electrochemically controlled charge injection in a
metallo-dielectric photonic crystal to optimize sensitivity and
species selectivity. The desired device performance enhancement
originates from a number of possible effects, including (1) the
charge-injection-tuned pickup of gas phase components that are to
be sensed, (2) the optimization of the resonant effect of SERS by
charge-injection-based tuning of plasma frequency (and therefore
the resonance enhancement of the SERS effect), and (3)
concentration of the gas-phase components on the high surface area
of the electrode used for sensing.
[0127] Element 1300 of FIG. 13 is an inverse-lattice photonic
crystal, which also functions as a working electrochemical
electrode. This element is preferably a conducting photonic crystal
having a void volume of greater than about 50%. This element is
optionally an inverse-lattice photonic crystal that is comprised of
a high reflectivity metal, like silver (which is especially
suitable). Element 1302 is a solid-state electrolyte that contacts
the photonic crystal. The solid-state electrolyte preferably has
low electronic conductivity and an ionic conductivity of above
10.sup.-4 S/cm at room temperature. Element 1301 is a counter
electrode to the working electrode 1300, which is also a photonic
crystal. This counter electrode can be one that operates
predominately faradaically or predominately non-faradaically.
Intercalated conducting polymers that undergo predominately
faradaic charging during device operation or very high surface area
non-intercalated materials (like nanoporous Pt or nanofibers) are
especially suitable for use as this counter-electrode. This counter
electrode element 1301 can optionally be a second photonic crystal
that can be predominately charged non-faradaically during device
operation. The benefit of using two photonic crystals is that one
obtains two materials for SERS sensing (one with negatively
injected charge and the other with positively injected charge),
which can be simultaneously probed optically during device
operation. Element 1303 is a variable voltage or variable current
power supply, and associated electrical wires to the working
electrode (1304) and the counter electrode (1301). Item 1304 in
FIG. 13 indicates the input and output of light to the photonic
crystal, which need not be in the pictured orthogonal direction to
the photonic crystal surface. The electrolyte element 1302
typically includes a reference electrode (not shown) for placing
measured potentials on an electrochemical scale--so that the degree
of charging can be determined from the measured potential of 1300
with respect to this reference electrode, and used to control the
degree of charge injection for 1300.
[0128] Various methods well known in the art could be used to make
the device of FIG. 13 and related devices, as either devices that
are macroscopic or macroscopic in lateral area, and optionally
probed either above the diffraction limit of light or using well
known photon tunneling methods (that avoid the diffraction
limitation of light). Methods for fabricating a metallo-dielectric
photonic crystal 1300 (which can be from a high reflectivity metal'
like silver) are described, for example, by A. A. Zakhidov et al.,
U.S. Pat. No. 6,261,469 and U.S. Pat. No. 6,517,762, L. Xu et al.,
Advanced Materials 15, 1562-1564 (2003), L. Xu et al., J. Am. Chem.
Soc. 123, 763 (2001), and O. D. Velev et al., Nature 401, 548
(1999). Fabrication of the device can proceed, for instance, by
using methods of these references (and optionally using
conventional lithographic methods) to make photonic crystals 1300
or arrays of photonic crystals (if an array of many devices are
required) on a substrate. Subsequent deposition of electrolyte 1302
and counter electrode 1301 (for instance, a conducting organic
polymer) can, for instance, be accomplished by spin coating 1301
and 1302 (again with optional usage of conventional lithographic
methods to define device size) on 1300. The thereby fabricated
devices can be transferred to a second substrate using conventional
wafer bonding methods, followed by either chemical or mechanical
removal of the first mentioned substrate, so that each photonic
crystal is obtained as the top layer on the second substrate. The
electrical connections to the photonic crystal 1300 and the
conducting polymer in each device can then be lithographically
defined. More simply, employing a slab of photonic crystal as the
substrate and depositing the electrolyte 1302 and counter electrode
1301 on one side of this substrate can eliminate the need for these
first and second substrates.
[0129] Application of the device of FIG. 13 as a SERS based gas
sensor is as follows: Applying a potential between the working
electrode 1300 and the counter electrode 1301 injects charge into
the photonic crystal electrode (1300), thereby modifying the
adsorption of targeted materials onto the external and internal
surfaces of 1300 and appropriately shifting the frequency of
surface plasmons. Different gas components can be selectively
detected by measuring the SERS spectra during scanning the degree
of charge injection in 1300. Either this device scanning is at a
sufficiently slow rate that surface adsorption and desorption can
be accomplished at ambient temperature or heating of component 1300
can be used to accelerate these surface absorption and desorption
processes. This heating can be accomplished either electrically by
resistance heating or by the heating effect of radiation
adsorption. Monitoring the SERS spectra during heating and cooling
processes is usefully employed for obtaining additional information
about the composition of the sensed gas. Additionally, measurement
of current flow between the working and counter electrode as a
function of the potential of these electrodes (or the electrode
potential versus applied inter-electrode current and time) can be
used in combination with the SERS spectra to characterize
analytes.
[0130] Like for most of the devices of invention embodiments, the
device of FIG. 13 can be operated either in a "rocking chair
device" mode or in an "electrolyte ion storage" mode, or as a
combination thereof. In the rocking chair mode, ions are shuttled
between working and counter electrodes during device operation, and
the function of the electrolyte (1302) is just to electronically
insulate the working and counter electrodes and to enable ionic
transport between them. In the electrolyte ion storage mode, the
electrolyte stores the ions that are injected into opposite
electrodes during device operation (typically anions for one
electrode and cations for the opposing electrode for a given change
in inter-electrode operation voltage). If there is only one mobile
ion in the device system, then the device operation will be by the
rocking chair mode, which requires only sufficient electrolyte 1302
to insure that the working electrode (1300) and the counter
electrode (1301) are electronically insulated with respect to each
other and that these electrodes are intimately contacted by the
electrolyte. Although device operation in either of these modes can
be usefully employed for this and other devices of invention
embodiments, device operation in predominately the electrolyte
storage mode can be advantageous. The reason is that operation in
strictly a rocking chair device mode requires that at least one of
these electrodes is charge injected during device fabrication.
Operation in the electrolyte ion storage mode requires electrolytes
in which both anions and cations are mobile. This contrasts with
the case of fuel cell devices of invention embodiments (based on
either hydrogen or hydrocarbon fuels), where electrolytes providing
predominately H.sup.+ ionic conduction are optional and more
preferred, because of the need for H.sup.+ ion conduction for these
fuel cells and the relatively high electronic conductivities of
many H.sup.+ transporting electrolytes.
[0131] Devices having the basic configuration shown in FIG. 13 can
also be used to switch other properties of the working electrode of
this figure, especially magnetic properties, electrical
conductivity, microwave absorption, surface energy, thermal
diffusivity and thermal conductivity, thermopower, the existence
and characteristics of superconductivity, and surface friction
coefficients. For all of these application modes except switching
surface friction and switching surface energy, it is advantageous
that the working electrode material comprises at least 20% void
volume. More advantageously, the void volume is at least about 50%,
and most advantageously, this void volume is at least about 75% for
the working electrode. Specific compositions suitable for these
types of devices are listed elsewhere in this section.
[0132] The presently discovered ability to electrochemically inject
charge into electrode regions that are not in direct contact with
electrolyte is especially useful for fuel cell devices. According
to conventional thinking, fuel cell redox reactions (oxidation or
reduction of fuel cell reactants) occur at locations where the
catalyst, fuel cell reactant, electrolyte, and electrode come in
joint contact. The device embodiments in the fuel cell area utilize
Applicants' discovery that redox reactions (either faradaic or
non-faradaic) can occur for electrode regions that are not in
contact with electrolyte. One consequence of this discovery are
devices providing greater catalyst utilization, since efforts to at
least partially contact the electrode with electrolyte results in
many catalyst particles being buried in electrolyte, thereby making
these catalyst particles inactive--since they are not exposed to
the fuel cell reactant (which is often a gas). The result of
ineffective use of catalyst is high catalyst cost for the fuel
cell. Another consequence of the discoveries used in invention
embodiments is high achievable gravimetric current densities, since
a high gravimetric surface area electrode is used and most of the
catalyst on this surface area is active.
[0133] FIG. 14 schematically illustrates a fuel cell in accordance
with invention embodiments, which is enclosed by element 1400.
Elements 1401 and 1402 are carbon nanotube forests along with
associated H.sup.+ cations (1407) and O.sup.-2 anions (1408)
resulting from redox reaction of the fuel components (H.sub.2 and
O.sub.2). By carbon nanotubes forests it is meant an array of
vertically aligned nanotubes. These forests contain catalyst
particles 1406, such as Pt nanoparticles. Elements 1403 and 1404
are electronically conducting sheets that are H.sup.+ ion
conductors (such as partially protonated, conducting polyanaline)
that are electrically contacted on the left to withdraw the
electrical energy produced by the fuel cell. These sheets (1403 and
1404), which make electrical contact to the nanotubes, are
separated by a sheet (1405) of a proton-conducting electrolyte of
the type used for conventional fuel cells, such as hydrated Nafion.
The large spheres 1409 symbolize the negatively charged polymer in
this electrolyte. This fuel cell works by the oxidation of H.sub.2
to produce protons (H.sub.2.fwdarw.2H.sup.++2 electrons) on the top
nanotube forest 1401 and by the reduction of O.sub.2 to produce
O.sup.-2 (O.sub.2+4 electrons.fwdarw.2O.sup.-2) on the bottom
nanotube forest (1402). Unlike in the case of prior art fuel cells,
these redox reactions predominately occur in the gas phase (on
surfaces of the nanotube forest that are exposed to the H.sub.2 and
O.sub.2 gas). H.sup.+ ions transverse the electronically and
ionically conducting sheet 1403, the electrolyte sheet 1405
(ionically conducting), and the electronically and ionically
conducting sheet 1404 to react with the O.sup.-2 ions to produce an
overall cell reaction (O.sub.2+2H.sub.2.fwdarw.2H.sub.2O). The
reduction at electrode 1401 and the reduction at electrode 1402
provides a cell potential of approximately a volt that produces a
current through electrically powered devices that are attached to
the indicated + and - electrodes (shown on the left hand side of
the figure).
[0134] Various methods well known in the art can be used to
fabricate the fuel cell device of FIG. 14. For example, the
synthesis of either single walled or multiwalled nanotube forests
(like the pictured nanotube arrays) by chemical vapor deposition
(CVD) and by plasma-enhanced CVD is described in the literature by
various authors (see S. Fan et al., Science 283, 512-(1999), A. M.
Casell et al., Langmuir 17, 260-(2001), and L. Delzeit et al., J.
Appl. Lett. 91, 6027-(2002)). A conducting polymer (like conducting
polyaniline) can be chemically polymerized, deposited from
solution, or electrochemically polymerized on top of the nanotube
array that is on the growth substrate. Then the conducting polymer
with attached nanotube forest can be stripped from the growth
substrate to free the combined elements 1401 and 1403. The nanotube
forest 1402 and contacting conducting polymer 1404 can be
analogously fabricated. Both of these nanotube forests and
supporting conducting polymer layers can be connected on opposite
sides of a freestanding polymer electrolyte membrane 1405 to
thereby complete the complicated aspects of device fabrication.
This connection can be made, for example, by using a solvent or
swelling agent to soften either the electrolyte layer or the
conducting polymer layers 1403 and 1404 (or both), so lamination of
the electrolyte and conducting polymer layers is facilitated. The
nanotube forest electrode 1401 can be optionally replaced with
various other nanostructured materials, such as the porous carbon
networks loaded with catalyst. See A. A. Zakhidov et al. in Science
282, 897 (1998), U.S. Pat. No. 6,261,469, and U.S. Pat. No.
6,517,762 and J.-S. Yu et al., J. Am. Chem. Society 124, 9382-9383
(2002) for details on how these porous carbon networks can be
synthesized by opal templating, and filled with catalyst
particles.
[0135] While the fuel cell reactants shown for the device are
H.sub.2 and O.sub.2 (which can be oxygen in air), other fuel cell
fuel couples (in either gaseous or liquid states or a combination
thereof) can be more generically used for invention embodiments.
These include, for example, methanol and hydrazine and the oxidant
hydrogen peroxide. The fuel component that is oxidized is
optionally and most preferably in gaseous form and is optionally
and most preferably either hydrogen; a hydrocarbon such as
CH.sub.4, C.sub.2H.sub.6, or C.sub.3H.sub.8 at preferably
100-200.degree. C.; an alcohol such as methanol or
C.sub.2H.sub.4(OH).sub.2 at preferably 20-80.degree. C.; H.sub.2S
at preferably 20-90.degree. C.; a nitrogen derivative such as
NH.sub.2NH.sub.2 at preferably 20-60.degree. C.; or ammonia at
preferably 200-400.degree. C. The fuel component that is reduced is
optionally and most preferably oxygen.
[0136] Examples of proton-conducting electrolytes that are useful
for the fuel cell device of FIG. 14 (and in related devices) are
Nafion, S-PEEK-1.6 (a sulfonated polyether ether ketone), S-PBI (a
sulfonated polybenzimidazole), and phosphoric acid complexes of
nylon, polyvinyl alcohol, polyacryamide, and polybenzimidazole
(poly[2,2'-(m-phenylene)-5,5'-bibenzimidazole]. These and other
useful proton conductors for use at either ambient or higher
temperatures are described in G. Alberti et al., J. Membrane
Science 185, 73-81 (2001); G. Alberti and M. Casciola, Solid Sate
Ionics 145, 3-16 (2001); P. L. Antonucci et al., Solid State Ionics
125, 431-437 (1999); L. Jorissan et al., J. Power Sources 105,
267-273 (2002); A. Bozkurt at al., Solid Sate Ionics 125, 225-233
(1999); and T. Norby, Solid Sate Ionics 125, 1-11 (1999). As an
alternative to Pt, various other fuel cell catalysts can be used,
including Ru and such Pt alloys as Pt--WO.sub.3 and Pt--Ru and
other catalysts given in B. Rajesh et al., Fuel 81, 2177-2190
(2002). This catalyst can be added to the nanotube arrays either
after initial synthesis of the nanotube forest, or after subsequent
process steps by using well-known methods (see, for example, B.
Rajesh et al., Fuel 81, 2177-2190 (2002); C. Wang et al., Nano
Letters 4, 345-348 (2004); W. Li et al., J. Phys. Chem. B 107,
6292-6299 (2003); X. Sun et al., Chem. Phys. Lett. 379, 99-104
(2003); G. Che et al., Nature 393, 346-349 (1998); and J.-H. Han,
Diamond and Related Materials 12, 878-883 (2003)).
[0137] While carbon nanotube forests are used here to illustrate
the device methodology, any of a number of nanoporous conductors
can provide the function of elements 1401 and 1402, such a sheets
of nanofiber paper (such as carbon nanotube paper) and membranes
made by the growth of carbon nanotubules in the pores of alumina
membranes (see G. Che et al., Nature 393, 346-349 (1998) and B.
Rajesh et al., Fuel 81, 2177-2190 (2002)). The electrode elements
1401 and 1402 need not be made of the same material. In fact, one
of these electrode elements can be a fuel cell element of the prior
art that is essentially fully infiltrated with electrolyte.
[0138] Devices based on electrochemically tunable charge injection
on the inside of hollow nanofibers are also fall within the scope
of the present invention embodiments. Such devices are especially
useful for applications where the presence of ions on nanofibers is
problematic. For example, it is sometimes useful to tune the work
function (and therefore the electron emission properties) of
nanotube fibers by charge injection for applications where these
nanotubes are used for the field emission of electrons. These
nanotube field emitters operate in vacuum and the presence of
mobile ions on the external surface of nanotubes or nanotube
bundles can cause undesirable fluctuations in electron emission
characteristics, and stability in electron emission is critically
important for such applications as cold cathode emitters for high
resolution electron microscopes. In these invention embodiments,
the inside of the hollow nanotube fibers can be contacted with all
of the components of an electrolyte. However, in states accessed
during normal device operation, the material inside the nanotube
will differ from conventional electrolytes in that these is a net
excess or either anions or cations.
[0139] FIG. 15 schematically illustrates a tunable nanotube device
in which tunability results from electrochemically induced
insertion of ions inside a nanotube, and insertion of associated
counter electronic charges onto the nanotube. One particularly
important tunable property is electronic work function. An
important type of device having the FIG. 15 configuration is a
electron field emitter that can be used, for example, as an
electron emitter for field-emission displays, electron microscopes,
and x-ray sources. Increased stability of electron emission
intensity is one advantage of this particular invention embodiment,
as compared with other invention embodiments of this invention
where ions are electrochemically placed on either the inside of the
nanotube or on both the inside and outside of the nanotubes. The
case is especially useful when the ions electrochemically inserted
into the interior of said nanotube are at least 10 times more
numerous than those electrochemically inserted on the exterior
surface of said nanotube
[0140] The device shown in FIG. 15 contains
electrochemically-active electronic conductors 1500, 1502, and
1506. Elements 1500 and 1502 can optionally be materials that can
be intercalated, so they electrochemically charge predominately
faradaically. Element 1500 is the counter electrode to a working
electrode that comprises the electronically interconnected elements
1502 and 1506. Element 1501 is an electrolyte that separates these
working and counter electrodes. Element 1506 is an electronically
conducting nanotube that is closed at one end and open at the
opposite end. Element 1504 (and like elements in electrolyte 1505
and inside the nanotube element 1506 are mobile cations. Elements
1505 in the electrolyte 1501 are anions, which can be either mobile
or immobile. The total charge of on these anions matches the total
charge of the cations in the electrolyte 1501. In contrast, the
total ion charge in elements 1500, 1502, and 1506 is matched in
large part by electronically injected charge in these elements.
Element 1503 is an ion permeation barrier material, such as a
deposited metal, which restricts ion placement during charging to
filling of the nanotube 1506. This element 1503 can optionally be
omitted. One disadvantage of this omission is that the resulting
partial placement of ions on the exterior surface of the nanotube
can decrease the stability of electron emission. An advantage of
omitting element 1503 is that higher amounts of charge can be
injected into element 1506 (at a given applied potential between
1500 and 1502) than would be the case if ion placement during
charging were restricted to only the inside of the nanotube element
1506. Element 1506 is a variable potential power source and
associated electronic connections that enables the application of a
potential between the counter electrode 1500 and the working
electrode that comprises 1502 and 1506. Element 1508 is a power
source that applies a potential between nanotube 1506 and a counter
electrode (1509) to thereby cause electron emission.
[0141] The device of FIG. 15 operates as follows. Ions shuttle
between the counter electrode (1500) and the working electrode
(comprising 1502 and 1506) depending upon the potential applied
between these electrodes by 1507. The degree of charge injection
varies in a controllable manner with the work function of nanotube
element 1506. Electron emission occurs (predominately from near the
tip of nanotube element 1506) in response to the application of an
applied voltage between 1506 and 1509.
[0142] The choice of materials for the device of FIG. 15 can be
made in various useful ways depending upon performance
considerations for a particular application. Also, variations on
this device design are useful. For example, there is an excess of
cations in the device of FIG. 15, and this excess of cations is
stored in the working electrode (comprising 1502 and 1506) and
counter-electrode (1500). The counter charges are injected
electrons in these electrodes. In normal operation, this pictured
device operates as an above mentioned rocking chair mode, where
ions are shuttled between working and counter electrodes during
device operation, and the function of the electrolyte (1505) is
just to electronically insulate the working and counter electrodes
and to enable ionic transport between them. While the ions shuttled
in FIG. 15 are cations, other invention embodiments involve the
inter-electrode shuttling of cations. Also, devices of invention
embodiments can be constructed analogously so that operation is in
the electrolyte ion storage mode. The only basic difference between
the device of FIG. 15 and a related device of invention embodiments
that operates in the electrolyte storage mode is that both anions
and cations used for device operation predominately come from the
electrolyte, so these ions must both be mobile. The electrode
elements 1500 and 1502 can be either a nanoporous material that
charges predominately non-faradaically or an electrode material
that charges predominately faradaically, such as low surface area
conducting polymer. The nanotube element 1506 can optionally be a
multiwall nanotube, instead of the pictured single walled
nanotubes. However, this multiwall nanotube should typically be
closed on one end for at least one of the nanotubes in the
multiwall assembly (in order to enhance the stability of electron
emission). The nanotube element 1506 can either have an open end
that is located in element 1502 or an open end that opens into the
electrolyte 1505 (as pictured in FIG. 15), as long as this nanotube
makes electronic contact with element 1502. These nanotubes can be
carbon nanotubes or any of the enormous variety of conducting
materials that are known to form either single walled nanotubes or
multiwall nanotubes (including double walled nanotubes). Also
nanotube element 1506 can comprise a bundle of single walled
nanotubes, a bundle or multiwall nanotubes, or a mixture thereof.
The aspect ratio of the nanotube or nanotube bundle element (ratio
of diameter to exposed length) is important for determining the
field enhancement factor, as is the absence of proximity similarly
high neighboring conductors. Specific compositions useful for the
practice of invention embodiments are described later, when the
various possible compositions and comparative advantages and
disadvantages of these compositions for various application
embodiments are described.
[0143] Devices of FIG. 15, and related devices of invention
embodiments, contrast with those of the prior art. Consider the
application of the FIG. 15 device for field emission of electrons.
Carbon nanotubes are currently widely studied as a source for
electric-field-induced electron emission (so-called cold cathode
electron emission). They are already used as a cold cathode for
numerous applications, such as Field Emission Displays (FEDs) and
x-ray sources. Invention embodiments enable gate-controlled
electrochemical charge injection in carbon nanotubes for the
optimization and control of the electronic work function of the
carbon nanotubes, and therefore their emission characteristics.
When electrons are injected into the nanotubes the work function
decreases, and when holes are injected the work function increases.
Invention embodiments provide electronically tunable cold cathodes
for flat panel displays, lighting fixtures, electron sources for
electron microscope, and discharge tubes for over-voltage
protection.
[0144] The next described devices of invention embodiments are
supercapacitors having greatly reduced self-discharge and the
ability to be dry-shipped. Prior art supercapacitors are based on
continuously maintained contact of electrolyte with the capacitor
electrodes. This electrolyte contact causes two major problems. The
first problem is that the presence of this electrolyte means that
charge cannot be stored in supercapacitors over long time periods.
This self-discharge cannot be eliminated as long as electrolyte
interconnects the two electrodes, since (a) self-discharge results
from the small electronic component of electrolyte conductivity and
(b) redox mediators can exist in the electrolyte, which undergo
oxidation at one electrode and reduction at the opposite
electrode--to thereby shuttle electronic charge between electrodes.
If the first problem could be solved then it might be possible to
ship charged supercapacitors for later use in the field. A second
problem then arises, since the continuously maintained electrolyte
in the supercapacitor increases the supercapacitor shipping weight.
The experimental discoveries involving this invention show that
injected charge in electrodes can be maintained even when the
electrolyte is absent, and that self-discharge is inherently lower
that for the case where the supercapacitor electrodes are immersed
in electrolyte.
[0145] FIG. 16 schematically illustrates a supercapacitor of
invention embodiments that can be charged, drained of electrolyte,
partially or completely evacuated then reactivated for subsequent
discharge in a remote location by refilling with electrolyte. In
the illustrated case (showing a cross-section of the device normal
to the supercapacitor electrode sheets (1601 and 1602)), the
electrolyte for device refill is carried in a compartment of the
device. In an alternative invention embodiment, the electrolyte
(which can be salt water) is injected into the supercapacitor
device from an electrolyte container that is separate from the
device. A benefit of this alternative design is that the
supercapacitor can be dry shipped in charged state without the need
to simultaneously transport the electrolyte.
[0146] While batteries can provide much higher energy storage
densities than supercapacitors, the power densities of
supercapacitors can be much higher than for batteries, which is
what makes this invention embodiment important. Also, Applicants
have discovered that injected charge is much more stable in the
absence of contacting electrolyte than it is for the
electrolyte-containing supercapacitor. In the device of FIG. 16 (in
container 1606), non-faradaically injected charge is stored in the
hole-injected electrode (1601) and in the electron-injected
electrode (1602) in the absence of contacting electrolyte.
Subsequent release of this stored energy is enabled when
electrolyte (1608) in the device is pushed into the supercapacitor
compartment that contains the sheet electrodes 1601 and 1602 and
the porous insulator sheet (1603), which prevents short circuiting
of these electrodes. Application of a force to the moveable element
1607 breaks a membrane (not shown) separating the supercapacitor
electrodes and separator sheet from the electrolyte and pushes the
electrolyte into the active cell volume, as shown in 1609 (which is
the supercapacitor device after release of the electrolyte into the
region of the electrochemically active cell elements). Before this
electrolyte release, the electrical contacts (1604 and 1605) to the
electrochemical electrodes are electrically floating, but after the
release of electrolyte these electrical contacts acquire the
relative potentials shown in 1609.
[0147] The invention embodiments also include a
supercapacitor/battery hybrid device in which at least one (and in
some cases both) of the device anode and cathode comprise elements
that that are predominately faradaically charged and predominately
non-faradaically charged. The component that is predominately
non-faradaically charged is typically exterior to the component
that is predominately faradaically charged, such as being a
covering a sheet on one or both sides of a sheet of the
predominately faradaically charged material. These predominately
faradaically charged electrode components should be in electrical
contact and advantageously in close physical contact. This
supercapacitor/battery hybrid device has the advantages of devices
like that shown in FIG. 16 in that the device can be stored and
shipped in a dry state without contacting electrolyte, and then
electrolyte can be placed in contact with the electrodes
immediately prior to usage, thereby increasing charge storage life
over the life that would result if the device electrodes were in
maintained joint contact with the electrolyte. In this "dry state"
of the supercapacitor/battery hybrid energy storage device there is
no complete ion path in an electrolyte between said electrode and a
counter electrode, so redox mediator impurities in the electrolyte
and trace electronic conductivity the electrolyte cannot cause
charge degradation. Moreover, this supercapacitor/battery hybrid
energy storage device has the optional benefit of being refilled
with electrolyte (or the liquid component of said electrolyte, such
as water) at usage site--thereby providing a reduced shipping
weight. Though at a cost of increased device weight and volume
(needed to accommodate the weight and volume of the predominately
faradaic component of the hybrid device), the supercapacitor
component can self-charge after capacitive discharge, as a result
of ion and charge flow from the predominately faradaic electrode
component to the predominately non-faradaic electrode component.
The gravimetric surface area of the predominately non-faradaically
charged electrode component can advantageously be at least 10 times
that of the predominately faradaically charged electrode component.
More advantageously, the gravimetric surface area of the
predominately non-faradaically charged electrode component can be
at least about 100 times that of the predominately faradaically
charged electrode component. Various materials can be used for the
predominately non-faradaically charged electrode component and for
the predominately faradaically charged electrode component, and
these different electrode components can be the same material
having two quite different degrees of porosity. Advantageous
electrode compositions for the predominately faradaically charged
electrode compositions are alkali metals and alkali metal alloys,
intercalated forms of carbon, and intercalated conducting polymers.
Porous carbon nanotube sheets are particularly suitable for the
predominately non-faradaically charged electrode composition. As
for conventional batteries, there can be more than electrodes in
the supercapacitor/battery hybrid energy storage devices, and these
electrodes can be connected either in-series or in-parallel
arrangements (or a combination thereof) depending upon the required
device output voltage. Both the flat plate and spiral would
electrode arrangements used for conventional batteries can be
usefully employed for the present supercapacitor/battery hybrid
devices.
[0148] Alternatively, from a view of optimizing discharge rate by
minimizing diffusion distances for a supercapacitor/battery hybrid
energy storage device, the predominately faradically charged
electrode component can interpenetrate within a predominately
non-faradaically charged electrode component. For example, a high
discharge rate and high charge storage capacity electrode could be
produced by dispersing particles or nanofibers of an doped
electronically conducting organic polymer (such as
alkali-metal-doped poly(p-phenylene, a conducting polyaniline, or
doped V.sub.2O.sub.5 nanofibers) within a matrix comprising
uncoated carbon nanofibers. Of the various alternative ways to
accomplish this dispersion, filtration of the faradaically
conducting component together with dispersed carbon nanotube fibers
to form a composite sheet is especially simple. Excepting the
co-addition of the faradaically charged component, this is the
conventional process used to form nanotube sheets (J. Liu at al.,
Science 280, 1253 (1998); A. G. Rinzler et al., Appl. Phys. A 67,
29 (1998)).
[0149] Devices of invention embodiments can use dynamically varied
non-faradaic charge injection to control the movement of materials
through filters containing nanoscale and/or microscale pores. These
embodiments can use predominately non-faradaic double-layer charge
injection to dynamically and selectively control the flow of
materials (in gases, liquids, or melts) through porous
membranes.
[0150] One such invention embodiment (using the concept of discrete
pores) is schematically illustrated in FIG. 17. The invention
embodiments of this figure and that of alternative invention
embodiments (where membrane pores are uniformly distributed) are
most generically applicable for dynamic and selective control of
the permeability of porous membranes for materials in dissolved
states; particles in solutions, melt, or supercritical states; and
molten materials. In the device of FIG. 17, 1701 is a filter
material containing pores 1702. The walls of these pores, which are
substantially in the thickness direction of the filter material
1702, are coated (at least in part) with an electronic conductor.
The pictured coating on the top surface of the filter material is
also an electronic conductor (1701), which substantially
electronically contacts the electronic material on the walls of the
1702 pores of the filter material. The substantially interconnected
electronically conducting materials on the top surface of membrane
1700 and the walls of pores 1702 electronically interconnects with
material 1700, and this total interconnected structure serves as
the working electrode of the device. Element 1704 is the counter
electrode of the device. This counter electrode element contacts
the electrolyte 1703, but does not substantially contact the
working electrode element. Element 1703 (the electrolyte) and 1704
(the counter-electrode) either do not extend over the pores in the
membrane, or are porous themselves--so as not to impede the
operation of the membrane. The power source used to obtain membrane
channel charging, together with electrical leads to the electrode
element 1700 and counter electrode, is element 1704. This power
source can be of variable voltage, so that the membrane filtration
properties can be dynamically tuned. Discharge of the
non-faradaically injected charge removes the ions in the pores of
the filter, thereby enabling convenient cleaning of the filter when
needed.
[0151] The device of FIG. 17 operates in the following way.
Application of a potential between the said working electrode and
the counter electrode 1704 causes predominately non-faradaic
charging of the electronically conducting material on the walls of
pores 1702. This involves essentially simultaneous charging of
these walls of 1702 by electronic charge injection and the
migration of ions into the pores (preferably from the electrolyte
1703) to electrostatically compensate these electronic charges.
These ionic counter charges selectively retard transport of
materials down the pores by effectively reducing the size of pores
1702. Also, the effectiveness of this pore filling on transport
through the membrane depends on the degree of charge injection, the
sign of charge injection versus that of components in the material
being filtered, as well as the geometry of the ions associated with
charge injection versus those in the material being filtered. These
dependencies provide additional features that enable the selective
tuning of transport through the membrane by materials having
different sizes, geometries, charges, and charge distributions.
[0152] The conducting electrode material in pores 1702 need not
extend along the entire length of the pores. In fact, a major
fraction of the membrane pore volume can be used to store a drug,
which is released at a rate determined by the applied potential
between the working electrode and the counter electrode 1704. Also,
a drug reservoir can be on one side of the membrane so that drug
delivery (for example, through the skin) is determined by membrane
transport that depends upon the inter-electrode potential and the
corresponding degree of non-faradaic charging.
[0153] Devices of invention embodiments also provide
electromechanical actuation and electrochemical tunable chemical
actuation. It is well know that charge injection, either faradaic
or non-faradaic, can provide electrochemical electromechanical
actuation. This electrochemical actuation is fundamentally
different from that for magnetostrictive, electrostrictive,
ferroelectric, electrostatic, and shape-memory actuation. For
examples of such faradaic and non-faradaic actuation, see R H.
Baughman, Synthetic Metals 78, 339-353 (1996); R. H. Baughman et
al., Science 284, 1340 (1999); R. H. Baughman et al., U.S. Pat. No.
6,555,945; G. Gu et al., Nature Materials 2, 316-319 (2003); and R.
H. Baughman, A. A. Zakhidov, and W. A. de Heer, Science 297,
787-792 (2002). A problem with prior art technologies of actuation
using electrochemically-induced dimensional changes is that an
electrolyte must be used, and both solid-state and liquid
electrolytes provide disadvantages. First, liquid electrolytes are
generally problematic because of the need for electrolyte
containment, and on the microscale (for anything other than
microfluidic applications) incompatibility with conveniently
employable device fabrication methods. Second, while the use of
liquid electrolytes can avoid these problems, others appear. Most
importantly, the mechanical modulus of the electrolyte acts to
constrain the achievable actuation in the best case, and in the
worst case (for electrolyte ion storage operation modes) the
electrolytes provide dimensional changes themselves and these
electrolyte-induced dimensional changes work in opposition to the
dimensional changes of at least one of the actuator devices. Hence,
these all-solid-state electrochemical electromechanical actuators
have been cantilever devices. These cantilever devices, when based
on a porous actuator electrode, contain electrolyte in the volume
of this electrode, as well as a separator between the two needed
electrodes. Hence, these prior art all-solid-state electrochemical
electromechanical actuator operate by bending, thereby utilizing
the mismatch in electrochemically induced dimensional changes of
mechanically coupled opposite electrodes. This presents a problem,
since such bending actuators do not provide a very efficient way to
convert electrical energy to mechanical energy. Unlike faradaic
electrochemical actuators, the non-faradaic actuators of some
preferred embodiments do not require dopant intercalation and
de-intercalation during the actuator cycle, so they do not suffer
from cycle life and cycle rate limitations from such partially
irreversible processes.
[0154] FIG. 18 schematically illustrates an electromechanical
actuator device of invention embodiments that operates by a
different mechanism, and can provide much larger actuation strains
than any prior art device of any type. This device uses a carbon
multiwall nanotube (MWNT) that telescopes outward in order to
decrease free energy by increasing the surface area available for
charge injection. The picture on the left (elements 1800, 1802, and
1804) shows the MWNT before non-faradaic charge injection, and the
picture on the right shows the same MWNT that is partially extended
by non-faradaic charge injection. Element 1800 is the MWNT in fully
contracted state and element 1801 is the MWNT in partially extended
state. This MWNT is a working electrode for a electrochemical cell
whose counter electrode is element 1804 and 1805 (which is the same
element for the two indicated states of the device). The electrode
elements are electronically separated by the electrolyte
(designated 1802 on the left and 1803 on the right). The (-) sign
on element 1805 and the (+) sign on element 1801 indicates that a
potential has been applied between the working electrode 1805 and
the counter electrode 1801. This applied potential results in the
injection of electrons into the working electrode 1801, and the
associated migration of positively charged ions 1806 to the surface
of this MWNT 1801.
[0155] The device operates because the actuator is a supercapacitor
with an energy associated with the capacitance C of
E.sub.c=1/2q.sup.2/C, where q is the amount of charge injection.
Since charge injection is largely to nanotube walls that are
externally exposed (rather than internal to the multiwall
nanotube), the capacitance C of the multiwall carbon nanotube
(MWNT) increases approximately proportionally to the external wall
area of the MWNT (A.sub.e). Hence, C=A.sub.eC.sub.a, where C.sub.a
is the capacitance per exposed surface area of the MWNT. At fixed
degree of charge injection q, C can be decreased by increasing
A.sub.e by the telescope-like extension of the MWNT. This provides
the driving force for actuation, which is opposed by the energy
cost of creating new external surface area.
[0156] A device of FIG. 18 can be fabricated, for example, using
methods already employed for making carbon nanotube tips for atomic
force microscopy (AFM) and scanning tunneling microscopy (STM). For
example, a MWNT can be grown on the tip of a conducting doped
silicon nanoprobe tip, which provides electrical connection to the
MWNT electrode. Thereafter, the electrolyte layer can be deposited
on the Si nanoprobe tip either by electrochemical polymerization,
chemical reaction, or solution deposition. Then a conducting
polymer layer can be deposited on top of the electrolyte layer by
chemical reaction or solution deposition. This conducting polymer
layer serves as the counter electrode of the device. Thereafter,
the electrolyte coating and the conducting polymer coating over
most (but not all) of the MWNT can be removed by conventionally
employed methods, such as the application of a voltage pulse.
Finally, a nanoprobe manipulator can be used to make contact with
the conducting polymer counter electrode. The thereby obtained MWNT
electromechanical actuator can be used for manipulations on the
nanoscale. This actuating nanoprobe tip can be used in gaseous,
vacuum, or liquid environments, including biological fluids.
[0157] The mechanical actuators of the present invention
embodiments can be run in reverse to convert mechanical energy to
electrical energy for mechanical sensor and energy conversion
devices. The benefits of using electrode elements that are not in
direct contact with the electrolyte is the same as for the
above-described actuators that are used to convert electrical
energy to mechanical energy. For cases in which the working and
counter electrodes are identical, the generation of electrical
energy requires that mechanical stress be applied differently to
these electrodes. Typically, a tensile stress is applied to one
electrode while a compressive stress (or a decrease in tensile
stress) is applied to the other electrode. These electrochemical
devices for sensing mechanical stress and strain and for converting
mechanical energy to electrical energy generate high currents at
low voltages, which provides advantages for some applications over
ferroelectric mechanical-to-electrical energy converters, which
generate low currents at high voltages. This performance of
electrochemical energy converters of the present invention
embodiments is desirable for minimizing the effect of lead
capacitances for remotely located sensors, so that sensor-response
amplifiers need not be located down-hole when doing seismology for
oil exploration. The ability to operate these energy-harvesting
devices at low frequencies could be usefully exploited for the
conversion of mechanical energy of ocean waves to electrical
energy. An array of such devices can be electrically interconnected
in series to provide an increased output voltage. In another
embodiment, the electrodes of the electrochemical electromechanical
energy-harvesting device are electrically biased during device
operation using an applied voltage. An advantage of such biasing is
that the electrical energy generated by mechanical stress can be
increased. However, when using such biasing, the stress generated
voltage changes should ideally be electrically isolated from the
biasing voltage. This can be accomplished by using a capacitor in
series with the bias circuit. These devices used to convert
mechanical energy to electrical energy typically utilize either a
uniaxial or biaxial applied stress. Device polarity can be achieved
using different materials for opposing electrodes, an applied bias
voltage, or the application of differing stresses to the opposing
electrodes.
[0158] FIG. 19 is illustrative of one generically applicable device
configuration of invention embodiments. This device is a nanoprobe
that can be used for such diverse applications as atomic probe
imaging and electron emission. Element 1900 is a conducting
nanoprobe tip base (such as an elemental metal), which mechanically
supports the device and provides electronic connection to element
1903, which is an electrochemical counter electrode for the device.
Element 1901 is an electrolyte that separates the counter-electrode
element 1903 from element 1902, which is part of the working
electrode. Element 1904 is a nanofiber, such as a carbon single
wall nanotube, a multiwall carbon nanotube, or a bundle of these
nanotubes that electronically interconnects only with element 1902.
Element 1905 is a power source (and associated electronic
interconnects) that provides an adjustable potential between the
counter electrode 1903 and the working electrode, which comprises
elements 1902 and 1904.
[0159] In the operation of this device, a change in the potential
applied between working electrode and counter electrode (by power
source element 1905) causes a change in the degree of electronic
charge injection in element 1904, as well as the migration of ions
on element 1904 to compensate these electronically injected
charges.
[0160] Consider first the application of the device of FIG. 19 for
scanning probe imaging, such as either atomic force microscopy or
atomic tunneling microscopy. The dynamically controllable charge
injection into 1904 (via the potential applied by 1905) changes the
work function of 1904 and therefore can dynamically tune the
properties of this probe tip for atomic tunneling microscopy.
Equally important, and sometimes more important for some
applications, the device of FIG. 19 is configured so that
electrochemical charge injection in nanotube tip 1904 is
accompanied by the transport of counter ions onto the nanotube tip
1904. These ions can substantially modify in a tunable way the
tunneling characteristics of the nanotube tip 1904 for both atomic
force and atomic tunneling microscopes. Use of the device of FIG.
19 as a field emission tip can utilize both the charge in work
function of the tip (as a result of electronic charge injection)
and the effect or counter ions on this tip on field emission.
Moreover, the device of FIG. 19 can be just one of an array of like
field emission tips that can be independently controlled
electronically.
[0161] The devices like shown in FIG. 19 can be constructed by
applying presently know technologies for nanofabrication. For
example, the nanoprobe base 1900 (which is an electrically
conducting contact) can be coated in the tip region by the counter
electrode material 1903, which can be a conducting organic polymer
that is deposited from polymer solution (or deposited from monomer
solution by electrochemical polymerization). The electrolyte 1901
can be then over coated on the counter electrode element (again,
for example, by either solution deposition or
electro-polymerization), as can be the subsequent coating of
element 1902, which is part of the working electrode. The nanotube
tip 1904 can be attached to the electronically and ionically
conducting element 1902 by various methods. For example, until the
electrolyte 1902 loses the solvent used for deposition, this layer
can act as an adhesive for harvesting a nanotube or small nanotube
bundle from a pre-synthesized nanotube array, such as a nanotube
forest. Electrical contact to the counter-electrode 1903 is by the
conducting nanoprobe base 1900. The other electrical contact (to
the working electrode component 1902 can be made using a
nanomanipulator during an inspection process to assess whether or
not the nanotubes element 1904 is correctly adhering to 1902. For
the purpose of conveniently making said contact (as well as for
convenience in making suitable depositions on 1901 and 1902) it
should be understood that these layers can extend much further back
from the tip of 190 than is shown in FIG. 19.
[0162] FIG. 20 shows a device that operates like the device of FIG.
19, except that the nanotube probe tip base makes electrical
contact to the nanotube. Such design facilitates device
construction. The nanoprobe base is the conducting cylinder 2000,
the power source with associated electrical interconnects is 2005,
the counter-electrode is 2001 (which can optionally be a doped
conducting polymer), the electrolyte separating working and counter
electrodes is 2002, the working electrode is comprised of 2003
(which can optionally be a conducting polymer) and element 2004.
Although element 2004 is shown here as a carbon nanotube, various
conducting probe tip materials can serve this function. In fact,
element 2004 can be just a sharpened part of the nanoprobe base
2000, such as an electrochemically-thinned tungsten wire. Also,
element 2003 can optionally be eliminated.
[0163] As an example, the fabrication of probes of the type shown
in FIG. 20, is described, but with element 2003 eliminated (so
diffusion of ions along the thereby revealed tip surface of the
nanoprobe base 2000 is required for electrochemical charging of
2004 during device operation. The nanoprobe base (2000) with
attached nanotube 2004 is commercially produced, and these are
typically made by depositing catalyst on the tip of nanoprobe base
2000, and using chemical vapor deposition to grow the nanotube on
2004 on 2000. (For several methods that can be used for such
nanotube growth on a probe tip, see J. H. Hafner et al., Nature
398, 761-762 (1999) and J. H. Hafner et al., J. Physical Chem. 105,
743-746 (2001)). After making electrical connection to nanotube
base 2000, nanotube base 2000 and attached 2004 can be dipped into
an electrolyte, so that an insulating protective coating can be
deposited onto the tip of nanoprobe base 2000 and all of element
2004. (See J. K. Campbell et al., J. Am. Chem. Soc. 121, 3779-3780
(1999) for methods for monitoring the depth of immersion into the
electrolyte, so only the tip region is coated with an insulating
protective coating.) The electrolyte 2002 can be then
electrochemically polymerized onto the desired region of the device
(since the insulating character of the protective coating will
prohibit deposition on top of this coating. Next, element 2001 can
be deposited from solution (via a dip/dry process) and electrical
connection is made to element 2001. Deposition of the material of
element 2001 on top of the protective coating is avoided by having
selected the coating material so that it is not wet by the
mentioned solution. Finally, either chemical etching or dissolution
in a solvent removes the protective coating.
[0164] FIG. 21 illustrates an invention embodiment in which a
high-surface-area nanostructured material (2110) functions as an
electrochemically gated ion beam source. This device uses our
surprising experiment observation that we can non-faradaically
inject charge in a nanostructured element, and retain this charge
in the absence of a contacting electrolyte. While this figure
pictures a carbon nanotube as the emission element (2110), this
emissive element can be any of a variety of nanostructured
materials whose surfaces (interior, exterior, or a combination of
interior and exterior) can contain counter ions that can be field
emitted. Element 2101 is an electrical insulator. Element 2102 is a
vacuum or other medium in which the source of field emission
species (2103) are dispersed. For example, this medium can be
liquid, gaseous, or solid (such as an alkali metal that is
evaporated into the gaseous state using an heater; a solution of an
alkali metal, or a complex of an organic solid with an alkali
metal). Element 2100 in this figure serves as an electronically
conducting electrode contact, as well as a confinement wall for
2102 and 2103 (although the containment and electrode contact
element of 2100 can optionally be separate materials). The field
emission precursor species (element 2103) can be either charged or
neutral. If this species is neutral, should be capable of
undergoing at least partial charge transfer with counter electrode
element material 21004 for some applied potential to this element.
If this field emission species is charged, with the counter ion to
this charge residing in 2102, counter-electrode element 2104 is
usefully capable of undergoing charge transfer with this counter
ion charge. For example, a neutral species 2103 can be an alkali
metal atom that undergoes oxidation or reduction with counter
electrode element 2104--thereby simultaneously transferring
electronic charge to 2104, and being converted to an ion. If
electron emission species 2103 is charged, the counter-ion to this
charge within 2102 should ideally be able of transferring
electronic charge to counter-electrode element 2104, thereby
enabling the migration of the ion emission species into 2104. The
electron emission species 2103 becomes the ionic electron emission
species 2105 within counter-electrode 2104, as well as the same ion
emission species that is pictured within electrolyte 2107, the
working electrode components 2108 and 2110, and the field emitted
ion beam 2111. The counter charges to the ionic emission species in
2104, 2108, and 2110, are, respectively, the electronically
injected charge (holes or electrons) in these elements. The
counter-ion species in electrolyte 2106 to the type ions found in
2105 are the counter ions 2106 found within 2107. Element 2112 is
the electrostatic counter electrode to the ion emission source
2110, wherein 2112 is at a negative potential with respect to 2110
if the ions emitted by the ion source are positive ions. Elements
2113 and 2114 are voltage sources (and associated electronic
interconnects), which are able to apply a variable voltage. The
voltage supply element 2113 applies a voltage between the electrode
(comprising 2108 and 2110) and the counter-electrode element
comprising 2104 (along with the associated ions 2104 and 2106) via
the electrode conducting elements 2100 and 2109.
[0165] The operation of the device of FIG. 21 is described in the
following. Element 2100 helps confine a source of material for ion
emission (2103), which can be in either charged or neutral. Either
as a result of (1) electronic charge transfer from neutral forms of
this field emission species to counter-electrode element 2104 or
(2) electronic charge transfer from counter ions (within 2102) to
counter-electrode element 2104, the ion emission species derived
from 2103 are able to migrate into the counter-electrode element
2104 (where they become 2105). In the absence of parasitic reaction
processes, migration of the ion emission species 2105 across the
electrolyte 2107 and into working electrode components 2008 and
2110 is controlled by the potential applied by 2113. The
application of a potential with suitable sign and magnitude between
2110 and 2112 causes the emission of the ion beam 2111.
[0166] The preceding figures showing invention embodiments have
used electrical wires to connect electrochemical electrodes to the
power source. FIG. 22 shows an invention embodiment in which
electrochemical charge injection results from charging via an
electron beam, so there is no direct contact of an electrical lead
to the electrochemically charged element. The major advantage of
such electron-beam-induced charging is that the electron beam
serves as an electrical contact that can be conveniently changed at
will and, more importantly, used to individually address electrode
elements (called pixels) that are on the nanometer or sub-nanometer
scale. Interestingly, electron beam irradiation can be used to
charge nanoscale pixels either negatively or positively--low
electron energies result in negative charging, and high electron
energies result in positive charging (since the loss of secondary
electrons can exceed the gain of electrons from the primary
beam).
[0167] FIG. 22 provides a schematic cross-sectional view of a
device in which individual nanoscale pixels can be
electrochemically charged either positively or negatively using a
focused electron beam. Element 2203 comprises a electronically
conducting support 2200, electronically non-interconnecting
wire-like pixels 2202 and 2210 that are a parallel array, and a
solid-state electrolyte 2201 that interpenetrates the pixel array.
Element 2204 is an enclosure that separates the vacuum (2205) on
the inside of 2204 from both the ambient atmosphere and an
atmosphere that can be arbitrarily chosen for the enclosed region
2209, which contains samples that are placed on top of the exposed
pixels 2210. Element 2211 is a voltage source that provides the
electron-beam-induced electrochemical charging of pixel elements
like 2202 and 2210. The electron beam, used for charging, is 2206.
This focused electron beam is produced by the electron beam source
2207. Sheet 2208, which completes enclosure 2209, can optionally be
used as an easily removable entry port for placing samples in
contact with pixels 2210 within enclosure 2209.
[0168] The device of FIG. 22 operates as follows. Charging of a
particular pixel in the pixel array is accomplished by focusing the
electron beam 2206 on that pixel. Depending upon the voltage used
to produce the electron beam, the nanotube becomes either
negatively or positively charged. For example, if electrons are
injected into a particular nanotube, holes will be injected into
the counter-electrode pixels (i.e., those contacting 2200). The
existence of electrolyte 2201 enables the charging of individual
nanotubes to much higher extents than would be possible if 2201
were a dielectric, since ions from electrolyte 2201 can locally
electrostatically compensate for the electronic charge on the
pixels.
[0169] The device of FIG. 22 can be used for various useful
processes--including, for example, the assembly of nanoscale
objects on the platform comprising the pixel array. The assembly
process utilized here is the switching of surface tension of a
pixel as a result of predominately non-faradaic electrochemical
charging of the pixel. The thereby self-assembled nanostructured
objects, which could be the components for a nanoscale circuit
board, can optionally be transferred to another substrate for
subsequent fabrication steps. This process of pixel-based assembly
and transfer of assembled nanoscale objects can be repeated as
needed for a manufacturing process. Surface tension changes induced
by temperature gradients, selective area deposition, or by
photo-induced reactions are well known in the prior art. However,
this use of electron-beam-induced pixel charging for
electrochemically switching surface energy is unique to the present
invention embodiment.
[0170] In another application mode, the device of FIG. 22 is used
for a new type of nanoscale microscopy, which we call Electron-Beam
Electrochemical microscopy (EBE microscopy). This application made
uses the well-known fact that detection of scattered electrons can
be used to measure the potential of an element in an electron
microscope. While the choice of investigated material is virtually
unlimited, biological materials (like cells and cells arrays and
DNA) provide important application opportunities. The microscopy
uses oppositely charges pixels as opposite electrodes for doing
spatially resolved electrochemical characterization, including
spatially resolved cyclic voltammetry (curves of current versus
potential at constant voltage scan rate, or related curves of
potential versus charge at constant current). Such spatial
information on the oxidation and reduction processes can be used to
probe chemicals within cells or to sequence DNA. The device of FIG.
22 employs focused electron beam generation and electron energy
detection capabilities that are well know, and widely used in
scanning electron microscopes and scanning transmission
microscopes. Numerous papers in the prior art describe methods for
making vertically aligned nanotube arrays (nanotube forests) and
manipulating these nanotube forests to make devices. These methods
can be usefully employed for the present invention embodiments that
use nanotube forests. Examples of these methods are described by P.
Soundarrajan et al. in J. Vac. Sci. Technology A 21, 1198-1201
(2003), J. Li et al., J. Phys. Chem B 106, 9299-9305 (2002), and A.
M. Cassell et al., Nanotechnology B15, 9-15 (2004).
[0171] The discovery that electrochemically injected charge is
stable in the absence of electrolyte also enables the dynamic
tuning of nanostructured materials properties for materials
absorption and desorption, such as hydrogen storage. Charge
injection (and associated ion migration) selectively changes the
interaction energy of materials with a nanostructured substrate
(such as nanostructured fibers, sheets, and powders) and therefore
changes the absorption capabilities of the substrate material.
Changing the degree of charge injection by changing the applied
potential can aid in the release of absorbed materials.
[0172] Invention embodiments also provide the use of predominately
non-faradaic electrochemical charge injection for dynamically
tuning thermal conductivity. These invention embodiments use the
unexpected observations described herein that non-faradaically
injected electronic charge and associated counter ions are stable
on nanostructured materials even in the absence of a contacting
electrolyte. Use is also made here of the fact that phonon
scattering lengths are long in low-dimensional nanostructured
materials (like carbon nanotubes), so these materials can have
large thermal conductivities (see P. Kim et al., Phys. Rev. Lett.
87, 215502-1 to 215502-4 (2001)). The large thermal conductivities
can be usefully employed, for example, for connecting hot
circuit-board components to a material that is at lower
temperature. However, for both the purpose of regulating
temperature and avoiding the heating effect of electrical pulses
sometimes used in thermoelectric cooling, it is desirable to be
able to electrically tune thermal conductivity. This tunability of
thermal conductivity is achieved here as a result of the combined
effects of electronic charge injection and the associated migration
of counter ions on the surface of the nanostructured material
(which can change thermal conductivity because of a change in
charge carrier density, can increase conductivity because of phonon
transport in the ion layer, and can decrease thermal conductivity
because of ion-based scattering of phonons associated with a highly
thermally conducting nanostructured material, like carbon
nanotubes).
[0173] FIG. 23 schematically illustrates one type of device of
invention embodiments that provides tunable thermal conductivity.
The blocks in FIG. 23 (2300, 2302, and like elements) are arrays of
multiwall nanotubes that have been grown on a substrate by
patterned deposition of catalyst on the substrate, followed by
growth of a nanotube forest (comprising parallel nanotubes that are
orthogonal to the substrate). While this sample is not yet
infiltrated with electrolyte, 2301 shows the inter-block spaces
where electrolyte can be infiltrated during device fabrication. The
blocks of nanotubes in FIG. 23 are 100 microns in lateral
dimensions, although either larger or smaller dimensions can be
employed for these blocks. An advantage of using small lateral
dimensions is in an enhanced rate of tunability, and a disadvantage
is an increased cost of device fabrication. The substrate (upon
which these blocks rest) contains conducting pads, so alternating
blocks provide working and counter-electrodes for an
electrochemical cell. The application of a potential between these
working and counter electrodes causes charge injection and
associated ion injection into the nanoporous blocks, and thereby
charges thermal conductivity. This changed thermal conductivity
varies in an electronically controllable manner thermal transport
along the nanotubes between the substrate and a material that is
positioned on top of the blocks of nanotubes.
[0174] The electrolyte 2301 can fill or partially fill the regions
between the nanotube blocks in FIG. 23. It is required that
electrolyte connects at least one working electrode bock with one
counter electrode block. Pervasive presence of the electrolyte 2301
in the structure is problematic, since the extensive presence of
this electrolyte can act to reduce the dynamic range of thermal
conductivity tuning (since the electrolyte can also contribute to
thermal conductivity). When a large dynamic tunability range is
needed (albeit with some sacrifice to tunability rate), the
electrolyte should preferably not substantially infiltrate the
blocks (2300, 2302, and like blocks). From this view of maximizing
dynamic range of tunability, the electrolyte can optionally be
deposited so as to interconnect blocks in only one direction. Also,
the electrolyte need not fill the cavity between any two adjacent
blocks. While this invention embodiment is illustrated using an
array of identical nanotube blocks, non-block arrays can also be
used and the material used for electrodes need not be comprised of
carbon nanotubes. Also, the material used for working and
counter-electrode elements can be different. In addition, the
counter electrode element need not be porous and need not be
charged predominately non-faradaically during device operation. For
example, the working electrode elements can be a metal mesh made by
a known opal templating process (see A. Zakhidov et al. in Science
282, 897 (1998), U.S. Pat. No. 6,261,469, and U.S. Pat. No.
6,517,762, as well as O. D. Velev et al. in Nature 401, 548 (1999))
and the counter-electrode element can be a doped conducting
polymer. Another useful method for making nanoporous metals for
such device embodiments is by dealloying alloys (see J. Erlebacker
et al., Nature 410, 450-453 (2001)).
[0175] FIG. 24 illustrates an invention embodiment in which
predominately non-faradaic charge injection is used to optimize the
figure of merit (ZT) of thermoelectric elements (2402 and 2405).
These thermoelectric elements are interconnected by electrolyte
2401. The indicated applied potentials on electrodes 2400, 2403,
2404, and 2406 results in predominately non-faradaic charge
injection into these electrodes (electrons in 2402 and holes in
2405), which optimizes the ZT of 2402 and 2405, by changing the
thermal power (S), thermal conductivity (.sigma..sub.T) and
electrical conductivity (.sigma..sub.e) of these elements (ZT,
which is an efficiency index, is
S.sup.2.sigma..sub.e/.sigma..sub.T). These thermoelectric elements
can be interconnected electronically like for conventional
thermoelectric devices to provide either a cooling capability
(pictured in FIG. 24), or the capability of generating electrical
energy from a temperature difference. In these cases, the top
electrodes (2400 and 2404) are at a different temperature than the
bottom electrodes (2403 and 2406) during device operation.
[0176] At least one of the electrodes 2405 and 2402 should be
porous. In the illustrated case, both of these electrodes are
porous, and made by the opal templating process of A. Zakhidov et
al., Science 282, 897 (1998) and U.S. Pat. No. 6,261,469 and U.S.
Pat. No. 6,517,762). A direct photonic crystal (also called a
direct-lattice photonic crystal or a opal photonic crystal)
typically comprises spheres. An inverse photonic crystal (also
called an inverse-lattice photonic crystal or inverse-opal photonic
crystal) is made by templating the void space of a direct photonic
crystal. Like for the device of FIG. 23, there are advantages in
minimizing the amount of electrolyte in the device. The reason for
doing so is that any electrolyte that provides a substantial path
for thermal transport between hot and cold regions of the device
can degrade ZT, since it increases the effective .sigma..sub.T.
[0177] The unexpected discoveries involving this invention show
that it is not necessary for electrolyte to substantially fill
nanoporous electrode of FIG. 24 (2402 or 2405) in order to obtain
electrochemical tunability of thermoelectric properties. This is in
contrast with the prior art, such as earlier teachings of R. H.
Baughman et al. in U.S. Pat. No. 6,555,945. The benefit of largely
eliminating the electrolyte from these porous elements is to
decrease the effective contribution of electrolyte thermal
conductivity to .sigma..sub.T in the equation for ZT. In fact, ZT
can be maximized by using the minimal amount of electrolyte that is
needed to provide a continuous ion path between the working and
counter electrodes (2402 and 2405).
[0178] Devices of invention embodiments can be used for the control
of electromagnetic wave propagation and reflectivity for
ultraviolet, visible, infrared, radio frequency, and microwave
frequency regions. Such switching can use changes in electrical
conductivity, refractive index, dielectric constant, absorption,
and reflectivity as a consequence of electronically controlled
predominately no-faradaic charge injection. It is known in the
prior art that such changes can be induced in porous materials by
electrochemical non-faradaic charge injection. However, the prior
art has not understood that such electrochemical switching of
properties can occur for material regions that are not contacted by
an electrolyte. One benefit of invention embodiments is the
decrease or total elimination of undesirable properties
contributions from an electrolyte.
[0179] The sensor device of FIG. 13 provides one invention
embodiment in which optical properties are tuned without employing
electrolyte contact to tuned regions that are accessible to the
electromagnetic radiation. This embodiment utilizes an electrolyte
that is on the reverse side of the optical element (and obscured by
it), so that presence of the electrolyte did not adversely affect
achieved properties. The device of FIG. 25 is one in which the
electromagnetic radiation propagates either at least approximately
parallel to the electrode layers or predominately along the lengths
of element 2500 (and like elements), along the lengths of element
2502 (and like elements), or along both of these types of elements
(in contrast with the possibly approximately perpendicular
propagation of the electromagnetic radiation for the device of FIG.
13). Element 2500 is the negatively charge-injected electrode and
2501 is an electrical contact for this electrode. Element 2502 is a
positively charge-injected electrode, element 2503 is an electrical
contact to this electrode, and element 2504 is an electrolyte that
separates elements of the type 2500 from elements of the type 2502.
Charge injection in 2500 and in 2502 (and like elements in the
array) provides the desired charge in transmission and reflection
properties of the device. The device operates by the changes in
properties for electromagnetic wave propagation that result from
charge injection into 2500 and 2502 (and like elements). The
benefit of this device arrangement is that the electrolyte 254 need
not penetrate into the elements (2500 and 2502) that are switched
by charge injection, so the effects of charge injection on
electromagnetic wave propagation are not obscured by possible
absorption of electromagnetic radiation by 2504. These benefits
result from the discovery of the inventors that predominately
non-faradaic charge injection can occur without the need for
contacting electrolyte.
[0180] FIG. 26 schematically illustrates an EBIG (Electrolyte-Bare
Ion Gated) device that provides electroluminescence by using a
semiconducting nanostructured material--in this case a
semiconducting carbon nanotube. Elements 2600 and 2601 are
electrical contacts to a semiconducting nanotube, which are charged
negatively and positively, respectively, by the battery and
associated electrical connections that comprise element 2604.
Element 2602 is an electrolyte that provides a continuous path
between opposite ends of the nanotube (element 2603). This device
operates by hole injection on one end of the nanotube and electron
injection on the opposite nanotube end. These injected charges are
compensated by ions of opposite sign that migrate along the
nanotube to coulombically compensate for the injected charge. Light
(indicated by the arrow 2605 is emitted as a result of
recombination of injected holes and electrons. Unlike other
invention embodiments, the working and counter electrodes of this
device are electronically interconnected, albeit with a
semiconducting nanotube.
[0181] FIG. 27 schematically illustrates another EBIG that is light
emitting, wherein the light-emitting element is a semiconducting
inverse-opal photonic crystal (2700). Elements 2702 and 2703 are
source and drain electrodes, respectively, and are simultaneously
working and counter electrodes. These electrodes contact opposite
sides of the pictured photonic crystal 2700, which is also
contacted by the electrolyte 2701, which can be a solid-state
electrolyte that is highly transparent in the spectral range where
light emission is required. Element 2704 is an insulating substrate
material, which can optionally contain an array of the EBIG light
emitters, as well as semiconductor devices for independently
varying light emission from each EBIG light emitter. Hole and
electrons are injected on opposite ends of the semiconducting
crystal 2700 as a result of the EBIG gating, which are
electrostatically compensated by the diffusion of counter-ions from
the electrolyte 2701. Charge carriers of opposite sign are injected
by the source and drain electrodes and their recombination in 2700
results in the emission of light (illustrated by element 2705). The
benefit of having the electrolyte substantially non-penetrating in
2700 is that the width of the photonic bandgap is unaffected by the
refractive index of the electrolyte (although it can be effected to
a lesser extent by ions that diffuse into 2700 during
electrochemical gating). The benefit of using a photonic crystal as
the light-emitting element is that the emission from the photonic
crystal is directional. Also, by using teachings of the prior art
for non-gated photonic crystal lasers, light emission by lasing can
be obtained.
[0182] FIG. 28 schematically illustrates a back-gated
electrolyte-bare ion gate (EBIG) field effect transistor that has
an air gap. This device differs from air-gap field-effect
transistors in that the electrolyte 2804 is present, which can
supply counter ions of opposite sign for electronic charges
injected into channel 2800 by the source electrode (2801) and drain
electrode (2802). Element 2805 is the gate electrode, element 2806
is an insulator, and element 2803 is an air gap. For application
modes in which the device of FIG. 28 is used like a Chem-FET (field
effect transistor chemical sensor) for gas sensing, this air gap
can optionally be used as a gas flow path for gas sensing and
analysis. As for other EBIG sensor devices of the invention
embodiments, devices of FIG. 28 can be used for detailed gas
sensing by the equivalent of gas cyclic voltammetry, in which case
it is useful to have one or more reference electrode in electrolyte
2804.
[0183] This device of FIG. 28 fundamentally differs from the
ion-gated devices of the prior art in that the channel electrode
2800 is not fully contacted with electrolyte and non-contacted
regions substantially contribute to device functionality. The
entire device of FIG. 28 can optionally be encapsulated in a
protective material, such as a barrier polymer, using known art of
device coating. Various device geometries orthogonal to this
cross-sectional view can be conveniently used. For example,
application modes of this device as a gas sensor can profitably use
a configuration in which the air gap 2803 is used for gas delivery
and is elongated orthogonal to the plane of the drawing. On the
other hand, using either a square or cylindrical geometry for this
air gap can facilitate miniaturization. The left and right parts of
channel 2800 (together, respectively, with the contacting
electrodes 2801 and 2802) can be viewed as two current-carrying
electrochemical electrodes. Element 2805 is a third
current-carrying electrochemical electrode. Charge injection into
the left and right sides of the channel will be identical if the
potential difference between 2801 and 2802 is negligible compared
with that between 2801 and 2805. In the case where 2805 is
electronically floating, a change in the potential between 2801 and
2802 will result in hole injection in one side of the channel 2800
and hole injection in the opposite side of 2800. This channel
ideally has a semiconducting state, so that changes in the degree
of charge injection can provide large changes in the conductivity
of the channel, which is measured by the source-drain current
(between 2801 and 2802) that results from the application of a
small change of potential between source and drain. The channel
2800 can be either nonporous or solid, and element 2805 can be
either faradaically or non-faradaically charged during device
operation. Also, ions can either be stored in the electrolyte in
one state of the device or these ions used for the charge-discharge
process of 2800 and 2805 can be shuttled between 2805 and 2800 (and
the reverse) during device operation. A combination of these ion
processes can also be profitably used. In one device embodiment,
the gate 2805 is optionally a metallic conductor. In another
invention embodiment, 2805 is a semiconducting material, electrical
contacts are provided between opposite ends of 2805, and
measurement of current flow between these electrical contacts (in
response to a potential difference between these contacts) provides
a second type of device response.
[0184] Various metal disulfides are especially useful as channels
for the devices like that shown in FIG. 28, like the semiconducting
WSe.sub.2, MoS.sub.2, MoSe.sub.2, HfS.sub.2, SnS.sub.2 layer
phases. The related phase of NbSe.sub.2 is useful for other
invention embodiments, since it is metallic at ambient and
superconducting at low temperatures.
[0185] Note that the device configuration of FIG. 28 can also be
used for an electroluminescent light source if element 2800 is a
semiconductor. In this case, electrode 2805 is optional and
functions to assist in electrochemical charge injection in the
light-emitting element 2800. The connecting electrolyte path
between the left and right sides of the channel 2800 need not be
present. In such case, transport of ions between opposite ends of
2800 can occur either within 2805 or on the surface of 2805.
[0186] The device of FIG. 29 shows the advantageous combination of
non-faradaic electrochemical and gas-gap-based electrostatic charge
injection for a chemical sensor that can replace conventional
Chem-FETs. This example is for a Chem-FET sensor device that is at
the same time a EBIG sensor. An analyte gas is passed through
conduit 2906. Element 2909 is either an air gap or, if elements
2906 and 2909 are optionally connected, a gap that is filled with
the analyte gas. Element 2907 is an electrolyte; element 2903 is a
semiconducting channel; elements 2904, 2905, and 2910 are source,
drain, and gate electrodes for dry-state electrochemical charge
injection; element 2908 is an insulator; and elements 2901, 2902,
and 2900 are, respectively, source, drain, and gate electrodes for
the gas-gap-based electrostatic charge injection. The benefit of
this hybrid Chem-FET is that either dielectric-based or
electrochemical dry-state charge injection can be used (or a
combination thereof). Only the electrochemical charge injection
results in the movement of ions onto the channel, and these ions
modify device response for a given degree of charge injection.
Hence, the availability of these two charge injection methods
enables improved selectivity for analyte detection.
[0187] The technologies for fabricating nanoscale materials,
nanoscale devices and nanoscale materials assembled within
macroscopic devices have rapidly advanced in the last decade, and
these technologies can be exploited for the fabrication of devices
of invention embodiments. For example, numerous methods have been
developed that are applicable for positioning nanofibers within
devices, orienting these nanofibers in desired directions, and
providing appropriate electronic interconnections. Issues such as
determining the most appropriate contacting metal and the
deposition methods applicable for these metals in nanoscale devices
are resolved to a satisfactory state in the prior art literature
(see, for example, Y. F. Hsiou et al., Applied Physics Letters 84,
984-986 (2004) for methods for applying low resistance contacts to
carbon nanotubes).
[0188] Well known and widely practiced micro-lithographic
techniques can be used to obtain patterned depositions and to
anchor nanofibers (such as carbon nanotubes) on a substrate. A
useful solution to the problem of firmly mechanically and
electrically contacting individual single-wall nanotubes or
single-wall nanotubes bundles (while leaving much of the nanotube
or nanotube bundle free of constraining contacts) has been
described by D. A. Walters et al. in Applied Physics Letters 74,
3803-3805 (1999). Using micro-lithographic techniques, these
authors demonstrated that single-wall nanotubes could be rigidly
attached on opposite ends (both electrically and mechanically), so
that they are suspended over a microscopic trench.
[0189] Various other methods well known in the prior art can be
used to provide nanofibers that are positioned in desired locations
on substrates, and appropriately oriented. One method is by
"scanning tip electrospinning", which has been described by J.
Kameoka et al. in Nanotechnology 14, 1124-1129 (2003). This method
achieves patterned deposition of continuous individual nanotube
fibers by electrostatic spinning from a microfabricated spinning
tip. Electrostatic spinning is typically of a solution of a polymer
(such as polyethylene oxide, polyacrylonitrile, DNA, a conducting
polymer like polyaniline, polyethylene, and various copolymers),
which can optionally include particles such as carbon nanotube
fibers (see D. H. Reneker and I. Chun in Nanotechnology 7, 216-223
(1996)). Depending upon the final state needed for the patterned
spun nanofibers, the polymer can be pyrolized (such as to convert a
polyacrylonitrile nanofiber to a carbon nanofiber), dissolved, or
chemically etched array (so as to reveal the nanoparticles, such as
carbon nanotube fibers). Probably the most commonly used method for
starting nanotube growth from a desired position on a substrate is
to deposit the catalyst nanoparticles used for growth at that
position, for example using either contact printing or
photo-lithographic methods (H. Dai, Accounts Chemical Research 35,
1035-1044 (2002)). As shown in the preceding reference, contact
printing on posts within a post array can be used to grow nanotubes
between these posts--like the wires spanning high-tension towers.
Orientation of nanofibers in a desired direction on substrates can
involve such methods as dielectrophoresis (R. Krupke et al., Nano
Letters 3, 1019-1023 (2003)), the orientating effects of liquid
crystals or fluid flow fields (S. Huang et al., J. Am. Chem. Soc.
125, 5636 (2003)), and magnetic or electric fields (Y. Zhang et
al., Applied Physics Letters 79, 3155-3157 (2001) and H. Dai,
Accounts Chemical Research 35, 1035-1044 (2002)). Also important
for device fabrication, nanofibers (such as carbon nanotubes)
self-orient during microwave plasma-enhanced chemical vapor growth
to be locally perpendicular to the growth substrate even when this
substrate is highly non-planar, such as the surface of an optical
fiber (see C. Bower et al., Applied Physics Letters 77, 830-832
(2000)). Like the case of carbon nanotubes, ZnO conveniently
self-organizes into useful forest-like arrays (with the nanowires
orthogonal to the substrate) during simple vapor transport and
condensation processes (M. H. Huang et al., Science 292, 1897-1899
(2001)). Forms of elemental carbon, especially carbon nanofibers or
graphite are included in some compositions as one or more
electrodes for invention embodiments. Like for the case of
conducting polymers most of these carbon compositions can be used
as either predominately non-faradaic or predominately faradaic
electrodes, depending upon the gravimetric surface area of the form
of carbon and the potential of device operation. For use as a
non-faradaic electrode or non-faradaic electrode component,
graphite is especially suited to be in the form of largely
exfoliated graphite. Suitable carbon fibers include multi-wall
nanotubes (which comprise concentric graphite sheets), single-wall
nanotubes (which comprise a single cylindrical graphite sheet),
carbon fiber scrolls (a spirally wound graphite sheet), and carbon
fibers with radial alignment (in which graphite planes extend
radially about the fiber direction). Since multi-wall carbon
nanotubes, single-wall carbon nanotubes (SWNTs), and carbon
nanotube scrolls have hollow interiors, these hollow materials can
optionally be filled by insulating, semiconducting, or metallic
compositions by known methods, so as to thereby modify charge
injection characteristics. In order to maximize surface area for
predominately non-faradaic charging, the number-average diameter of
single-wall and multi-wall carbon nanotubes used as predominately
non-faradaic electrodes is preferably below about 10 nm. The term
number-average diameter means the ordinary average of the diameters
of the nanotubes, without any special weighting according to the
size of the diameter.
[0190] Single-wall carbon nanotubes (SWNTs) are especially
well-suited for use as electrodes. The dual laser method, the
chemical vapor deposition (CVD) method, and the carbon-arc method
are suitable methods for making the carbon nanotubes, especially
single-wall carbon nanotubes, and these methods are well known in
the literature (R. G. Ding et al., Journal of Nanoscience and
Nanotechnology 1, 7 (2001) and J. Liu et al., MRS Bulletin 29,
244-250 (2004)). Carbon single-wall nanotubes can have armchair,
zig-zag, or chiral arrangements of carbon atoms. These nanotubes
are differentiated in that the armchair nanotubes have a
circumference of para-connected hexagonal rings (like found in
poly(p-phenylene)), the zig-zag nanotubes have a circumference of
linearly side-connected hexagonal rings (like found in linear
acenes), and the chiral nanotubes differ from the armchair and
zig-zag nanotubes in that they have a sense of handedness. The
geometry of carbon nanotubes is specified using conventional
nomenclature using the indices (n, m). Depending on the appearance
of a belt of carbon bonds around the nanotube diameter, the
nanotube is the armchair (n=m), zig-zag (n or m=0), or chiral (any
other n and m) variety. All armchair SWNTs are metals; those with
n-m=3k, where k is a nonzero integer, are semiconductors with a
tiny bandgap; and all others are semiconductors with a bandgap that
inversely depends on the nanotube diameter. Since methods are
available for selecting carbon nanotubes according to conductivity
type, the choice of nanotubes for particular applications can be
optimized. The optimal SWNT type depends upon the device or
materials application. Metallic nanotubes are generally suitable
for applications where maximum conductivity is desired, such as for
transparent conductors or as counter electrodes where rapid
response rate is desired. On the other hand, semiconducting
nanotubes are desired for the channel material for semiconductor
devices of the present invention.
[0191] Various methods of separating single wall and multiwall
nanotubes (by type, length, diameter, etc.) are useful for
invention embodiments. Examples of known methods for such
separation involve (1) use of charge transfer agents that complex
most readily with metallic nanotubes, (2) complexation with
selected types of DNA, and (3) dielectrophoresis (R. Krupke et al.,
Nano Letters 3, 1019-1023 (2003) and R. C. Haddon et al., MRS
Bulletin 29, 252-259 (2004)). These and other methods can be used
to provide nanotube materials for the practice of this invention,
such as the nanofiber transistors of FIGS. 11 and 12. The various
dielectrophoretic methods are especially useful for depositing
nanotubes of a desired conducting type (metallic (or small bandgap)
or semiconducting nanotubes) on a device substrate between two
electrodes (which are used to apply the alternating current
potential used for the dielectrophoresis). Depending upon the
frequency of the voltage applied for dielectrophoretic deposition,
either metallic (and near metallic nanotube having very small
bandgaps) or semiconducting nanotubes (having much larger band
gaps) can be preferentially deposited on semiconductor substrates.
If the nanotubes in solution are partially bundled, and the chosen
frequency for the dielectrophoretic process favors deposition of
metallic nanotubes, the deposited nanotubes can be bundles
containing a metallic nanotube (or very small bandgap
semiconducting nanotube) together with semiconducting nanotubes.
More generally, if it is desirable to remove metallic nanotubes
from a single walled nanotube bundle (or remove a metallic nanotube
from being the outermost wall on a multiwall nanotube, well-known
high voltage pulse methods can be used (see P. Collins et al.,
Science 292, 706-709, (2001) and A. Javey et al., Nano Letters 2,
929-932 (2002)).
[0192] Synthetic methods generally result in mixtures of nanotubes
having different diameters. Use of catalyst for nanotube synthesis
that is close to monodispersed in size (and stable in size at the
temperatures used for synthesis) can dramatically decrease the
polydispersity in nanotube diameters, and having this narrower
range of nanotube diameters can be useful for invention
embodiments. S. M. Bachilo et al. describe such a method in Journal
of the American Chemical Society 125, 11186-11187 (2003).
[0193] Nanotubes and nanoscrolls filled with agents that provide
charge transfer to the nanotubes are included in preferred
electrode compositions, including those used as device channels.
One reason for this preference is that these agents effectively
bias the degree of charge transfer to the nanotubes and
nanoscrolls, thereby shifting the Fermi level of the nanotubes and
consequently the density of states at the Fermi level. The filled
volume of the nanotubes should have a net non-zero charge and the
counter charge to this net non-zero change can predominately reside
on or near the external surface of the nanotube or nanoscroll.
Examples of filling agents that can have a net non-zero charge when
inside the nanotubes or nanoscrolls are ionic salts and salt
solutions, partially ionized elements (such as alkali metals and
divalent elements such as calcium), and organic transfer agents.
For the purposes of this invention embodiment, ionic salts and
ionic salt solutions inside nanotubes and nanoscrolls are not
considered to be contacting electrolytes. Other suitable filling
agents for nanotubes and nanoscrolls are fullenenes (especially
C.sub.60) and fullerenes complexed with charge transfer agents
(like monovalent, divalent, and trivalent elemental metals). The
complexed fullerenes that are included within the volume of the
nanotube or nanoscroll can optionally comprise endohedral
fullerenes that contain n species Z that are within the C.sub.m,
which are designated using the symbols Z.sub.n@C.sub.m. Typical
examples are Gd@C.sub.82, La@C.sub.82, La.sub.2@C.sub.82,
Sm@C.sub.82, Dy@C.sub.82, Ti.sub.2C.sub.80, La.sub.2C.sub.80, and
like compositions.
[0194] Various methods are particularly useful in invention
embodiments for filling or partially filling nanotubes. These
methods typically include a first step of opening nanotube ends,
which is conveniently accomplished using gas phase oxidants, other
oxidants (like oxidizing acids), or mechanical cutting. The opened
nanotubes can be filled in various ways, like vapor, liquid phase,
melt phase, or supercritical phase transport into the nanotube.
Methods for filling nanotubes with metal oxides, metal halides, and
related materials can be like those used in the prior art to fill
nanotubes with mixtures of KCl and UCl.sub.4; KI; mixtures of AgCl
and either AgBr or AgI; CdCl.sub.2; Cdl.sub.2; ThCl.sub.4;
LnCl.sub.3; ZrCl.sub.3; ZrCl.sub.4, MoCl.sub.3, FeCl.sub.3, and
Sb.sub.2O.sub.3. In an optional additional step, the thereby filled
(or partially filled) nanotubes can be optionally treated to
transform the material inside the nanotube, such as by chemical
reduction or thermal pyrolysis of a metal salt to produce a metal,
such as Ru, Bi, Au, Pt, Pd, and Ag. M. Monthioux, Carbon 40,
1809-1823 (2002) provides a useful review of these methods for
filling and partially filling nanotubes, including the filling of
nanotubes during nanotube synthesis. The partial or complete
filling of various other materials useful for invention embodiments
is described in J. Sloan et al., J. Materials Chemistry 7,
1089-1095 (1997).
[0195] Conducting nanofibers need not contain carbon in order to be
useful for invention embodiments, and a host of processes are well
known in the art for making suitable nanofibers. Some examples are
the growth of superconducting MgB.sub.2 nanowires by the reaction
of single crystal B nanowires with the vapor of Mg (Y. Wu et al.,
Advanced Materials 13, 1487 (2001)), the growth of superconducting
lead nanowires by the thermal decomposition of lead acetate in
ethylene glycol (Y. Wu et al., Nano Letters 3, 1163-1166 (2003)),
the solution phase growth of selenium nanowires from colloidal
particles (B. Gates et al., J. Am. Chem. Soc. 122, 12582-12583
(2000) and B. T. Mayer et al., Chemistry of Materials 15, 3852-3858
(2003)), and the synthesis of lead nanowires by templating lead
within channels in porous membranes or steps on silicon substrates.
The latter methods and various other methods of producing metal and
semiconducting nanowires suitable for the practice of invention
embodiments are described in Wu et al., Nano Letters 3, 1163-1166
(2003), and are elaborated in associated references. Y. Li et al.
(J. Am. Chem. Soc. 123, 9904-9905 (2001)) has shown how to make
bismuth nanotubes. Also, X. Duan and C. M. Lieber (Advanced
Materials 12, 298-302 (2000)) have shown that bulk quantities of
semiconductor nanofibers having high purity can be made using
laser-assisted catalytic growth. These obtained nanofibers are
especially useful for invention embodiments and include single
crystal nanofibers of binary group III-V materials (GaAs, GaP,
InAs, InP), tertiary III-V materials (GaAs/P, InAs/P), binary II-VI
compounds (ZnS, ZnSe, CdS, and CdSe), and binary SiGe alloys. Si
nanofibers, and doped Si nanofibers, are also useful for invention
embodiments. The preparation of Si nanofibers by laser ablation is
described by B. Li et al. (Phys. Rev. B 59, 1645-1648 (1999)).
Various methods for making nanotubes of a host of useful materials
are described by R. Tenne in Angew. Chem. Int. Ed. 42, 5124-5132
(2003). Also, nanotubes of GaN can be usefully made by exitaxial
growth of thin GaN layers on ZnO nanowires, followed by the removal
of the ZnO (see J. Goldberger et al., Nature 422, 599-602 (2003)).
Nanofibers having approximate composition MoS.sub.9-xI.sub.x, which
are commercially available from Mo6 (Teslova 30, 1000 Ljubljana,
Slovenia) are included as useful compositions (most especially for
x between about 4.5 and 6). Related compositions are also described
by D. Vrbani et al. in Nanotechnology 15, 635-638 (2004). Among
other applications embodiments, these latter fibers are useful for
electron emission tips and as a tunable superconductor.
[0196] Another way to make the high-surface-area materials used in
invention embodiments is by templating a self-assembled structure
that has high surface area. For example, a periodic template
crystal (called an opal) can be obtained by the sedimentation of
spheres that are substantially monodispersed in diameter. These
spheres typically have an average sphere diameter of between 500 nm
and 10 nm. In most cases, these spheres are from either (a) an
inorganic oxide, such as SiO.sub.2, which can be removed by
chemical processes such as exposure to aqueous acid or base, or (b)
an organic polymer that can be removed by pyrolysis, chemical
reaction, or dissolution. The template crystal, after an optional
sintering process to provide inter-sphere necks, is infiltrated
with either the electrode material or a material that can be
converted to the electrode material. This sintering process in
described by Zakhidov et al. in Science 282, 897 (1998) and U.S.
Pat. No. 6,261,469 and U.S. Pat. No. 6,517,762. Thereafter, the
template material is removed to provide an inverse lattice, which
is a structural replica of the original template crystal. As an
example, Zakhidov et al. (Science 282, 897 (1998)) used
plasma-enhanced CVD to make a very high surface area graphitic
carbon. Millimeter thick opal plates based on monodispersed
SiO.sub.2 spheres were infiltrated with carbon from a
hydrogen/methane plasma created by microwave excitation. Extraction
of the SiO.sub.2 spheres from the carbon infiltrated opal (using
aqueous HF) resulted in a very high surface area, nanoporous foam
in which carbon layers as thin as 40 .ANG. make the internal
surface of the foam. As another suitable method for fabrication
conducting sphere arrays having high surface area for use as
nanostructured electrodes, conducting spheres that are nearly
monodispersed in diameter can be directly assembled from sphere
dispersions using conventional methods. These spheres are typically
less than about 200 nm in average diameter, more typically less
than 100 nm in average sphere diameter, and most typically less
than 50 nm in average sphere diameter.
[0197] The devices of this invention may comprise more that two
current-carrying electrodes that can be operated at different
voltages. Advantages of using more than two current carrying
electrodes are that additional flexibility is achieved with respect
to the degree of charge injection in the individual electrodes,
which can be useful for optimizing device performance.
[0198] In some processes of the present invention, electrode
charging for at least one electrode is predominately non-faradaic,
meaning that over 50% of the initially injected charge is injected
non-faradaically. It should be emphasized that this definition
pertains to the nature of initial charge injection from the
electrolyte, since initially non-faradaically injected charge can
later transform to charge that is retained faradaically during
electrode drying processes. Such is the case if dopant ions that
are initially stored in an electrochemical double layer later
intercalate into the electrode material. In the non-faradaic
process of certain embodiments, the ions from the electrolyte
(which compensate the electronic charges injected into the
electrodes) are located close to the surface of at least one of
these electrode elements. This is in contrast to the faradaic
processes where the ions that compensate the injected electronic
charges penetrate inside the electrode material and change its
structure, usually expanding it. In some invention embodiments,
typically over 50% of the stored charge is stored non-faradaically
in the charged material, meaning that at least 50% of this stored
charge is associated with ions on or near the surface of the
material. In some invention embodiments, typically over 75% of the
stored charge is stored non-faradaically in the electronically
charged material, meaning that at least 75% of this stored charge
is associated with ions on or near the surface of the material.
Because of this location of charge on material surfaces in the
non-faradaic processes of invention embodiments, ions from the
electrolyte need not penetrate the electrode material and need not
cause phase changes within the electrode material. This use of
non-faradaic charge injection is generally most important for
device applications where the charge injected electrode material
must be repeatedly charged and discharged during device operation,
since non-faradaic charge injection generally provides longer
device cycle life than does faradaic charge injection. It should be
recognized that charge injection processes can be non-faradaic over
one potential range, and then become faradaic when this potential
range is extended. Consequently the definitions of predominately
non-faradaic and predominately non-faradaic indirectly signify the
potential range used for hole or electron charging. A device is
called a predominately non-faradaic as long as there is a device
operation range where a useful device response can be obtained from
predominately non-faradaic charging of at least one electrode.
While charge that is initially inserted predominately
non-faradaically can later produce materials intercalation (i.e.,
ions insertion into solid material volume), it is advantageous if
less than 50% of counter ion charge associated with maximum
injected electronic charge is intercalated in a solid material
during material use or normal device operation.
[0199] The achievement of very high electrode capacitances requires
the use of nanostructured materials that have small sizes in at
least one dimension, and such small dimensions can affect the
properties of the nanostructured material in both charge-injected
and non-charge-injected states. Consequently, the use of materials
with larger dimensions (sheet thicknesses, fiber diameters, or
particle sizes) can in some cases be suitable (with a corresponding
decrease in suitable electrode capacitances)--especially for the
tuning of highly-scale-sensitive properties, such as
ferromagnetism.
[0200] The case where only the working electrode operates
predominately non-faradaically is also included in invention
embodiments. Devices in which ions predominately insert on the
internal and external surface of one electrode and in the material
volume of the second electrode are included here. This can be a
useful embodiment in cases where cycle life limitations are not
problematic for the desired application mode or where the
potentially higher charge storage densities of faradaic processes
provide needed benefits of reduced device size or weight. However,
predominately non-faradaic operation of both electrodes is useful
when very long device cycle lifetimes are needed, when
charge-injection modified properties of both electrodes are
utilized, or when dopant intercalation and associated structural
changes degrade needed properties.
[0201] The device types of some embodiments have at least one
electrode (typically the working electrode) that has high
gravimetric surface area, since this high surface area is typically
required to obtain high degrees of double-layer (non-faradaic)
charge injection. In fact, material selection to provide either
predominately non-faradaic or predominately faradaic performance is
made according to either surface area measurements or structural
results (from typically scanning electron microscopy, SEM, electron
transmission microscopy, TEM, or atomic probe microscopies). The
gravimetric surface area is conveniently taken as the surface area
measured in nitrogen gas by the standard Brunauer-Emmett-Teller
(BET) method. The gravimetric surface area of the working electrode
is advantageously above about 1 m.sup.2/g, more advantageously
above about 10 m.sup.2/g, and most advantageously above about 100
m.sup.2/g. In some instances where the electrodes must be
repeatedly charged and discharged during device operation and long
cycle life is needed, the surface area of both working and counter
electrodes is advantageously above 1 m.sup.2/g, more advantageously
above about 10 m.sup.2/g, and most advantageously above about 100
m.sup.2/g.
[0202] It is advantageous for some device embodiments that either
the working electrode or both working and counter electrodes
comprise a mixture of electronically conductive materials that
serve as electrode components. This mixture of materials in the
electrodes can include materials that are non-faradaically charged
and faradaically charged.
[0203] The working and counter-electrodes can be made of either the
same or different materials. Suitable examples of electrode
materials include (a) high surface area metallic compositions
obtained by the degenerate doping of semiconductors (such as Si,
Ge, n-doped or p-doped cubic boron nitride, mixtures of Si and Ge,
and GaAs), (b) conducting forms of conjugated organic polymers
(such as polyacetylene, poly(p-phenylene), or poly(p-phenylene
vinylene) and copolymers thereof), (c) carbonaceous materials
obtained by the pyrolysis and surface area enhancement of polymers,
(d) graphite, carbon nanotubes, and less ordered forms of carbon
formed by pyrolysis, (e) elemental metals and alloys of these
metals, and (f) electrically conducting metal oxides and metal
chalcogenides (such as CdS and CdSe). Doped diamond is another
suitable electrode material, especially hole-doped diamond (which
is a conductor and even a superconductor--see E. A. Ekimov et al.,
Nature 428, 542-545 (2004)). Use of this degeneratively doped
diamond as a predominately non-faradaically charge injected
material, this diamond should be configured as a high surface area
material. Methods for making nanoporous diamond having high surface
area are described by A. A. Zakhidov et al. in Science 282, 897
(1998), U.S. Pat. No. 6,261,469, and U.S. Pat. No. 6,517,762.
[0204] Organic conducting polymers are among the suitable
compositions for use as predominately faradaic electrodes. Very
high surface area conducting polymers are included as suitable
compositions for predominately non-faradaic electrodes. Various
methods can be used to obtain conducting polymer electrodes having
high surface areas. For example, known methods can be used for the
electrostatic spinning of conducting polymers into nanometer
diameter fibers. For these conducting polymers the predominately
non-faradaic behavior is obtained as a result of this high surface
area and operation of the electrodes in potential ranges where the
major faradaic process do not substantially occur. Especially
suitable organic conducting polymers are those with planar or
nearly planar backbones, such as poly(p-phenylene),
poly(p-phenylene vinylene), and polyacetylene. Other suitable
conducting polymers are various conducting polypyrroles,
polyanilines, polyalkylthiophenes, and polyarylvinylenes. The
synthesis of conducting polymers suitable for such embodiments is
well known, and is described, for example, in the Handbook of
Conducting Polymers, Second Edition, Eds. T. A. Skotheim et al.
(Marcel Dekker, New York, 1998).
[0205] The nanostructured conductor (such as a nanotube) used as a
charge injected electrode can optionally be coated with another
material. If fact, benefits can result even if this over-coated
material is a poor electronic conductor. However, if the
over-coated material is a poor conductor, provision should be made
to insure that charge injection into the nanostructured material
can occur. This is accomplished, for example, by either making
direct electrical contact to the nanostructured conductor or
insuring that the over-coated poorly conducting material is
sufficiently thin that tunneling or other electronic transport is
possible (on the desired time scale) across this poorly conducting
material.
[0206] Various benefits can result for over-coating the
nanostructured electrode material with a second material. As a
first benefit, this over-coated material can serve to protect the
charge injected state of this conducting nanostructured material
against undesired reaction with redox-active impurities in the
environment. This can be the case, for example, when electronic
charge injection into a nanostructured material is enabled by ion
motion internal to this nanostructured material, such as on the
inside of a nanotube or in one of the two possible
non-interconnected labyrinths of an inverse lattice photonic
crystal. A second achievable benefit is most effectively realized
when the ions for the electronically injected charge are on the
exterior surface of the over-coated material, so that this over
coated material is between the ions and the nanostructured base
material. This second benefit is that the enormous electric field
generated by charge injection is applied across the over coated
material, which can provide useful electronic, magnetic, or optical
properties for this material by causing some of the charge to be
injected into the over-coated material.
[0207] A host of methods can be used for providing this
over-coating layer, such a vapor state, liquid state,
super-fluid-state coating of the nanostructured electrode material.
Also the nanostructured material can be formed inside an insulating
over coating material after this over coating material if formed,
such as by the filling of insulating boron nitride nanotubes with
C.sub.60 and the subsequent coalescence of arrays of these C.sub.60
molecules to form carbon nanotubes as the conducting nanostructured
electrode material. (See W. Mickelson et al., Science 300, 467-469
(2003) for this process for forming carbon nanotubes inside BN
nanotubes.)
[0208] Aerogels, and especially carbon aerogels and aerogels based
on conducting organic polymers, are included in the list of
suitable electrode materials. Resorcinal-formaldehyde-derived
carbon aerogels are especially suitable carbon aerogels. These
carbon aerogels can be conveniently produced using the sol gel
method, supercritical drying using liquid CO.sub.2, and pyrolysis
in nitrogen at about 1000.degree. C. The synthesis of these carbon
aerogels is described by Salinger et al. in Journal of
Non-Crystalline Solids 225, 81 (1998) and by Wang et al., in
Journal of the Electrochemical Society 148, D75-D77 (2001). Other
aerogels that are useful for invention embodiments are described by
J. L. Mohanan et al. in Science 307, 397-399 (2005).
[0209] Various methods are well known in the literature for
assembling nanostructured fibers and particles into forms well
suited for the practice of the present invention. For example,
sheet shaped electrodes of single-wall nanotubes can be
conveniently formed by filtering an aqueous suspension of such
purified carbon tubes through poly(tetrafluoroethylene) filter
paper, as described by Lui et al., in Science 280, 1253 (1998).
Peeling the resulting paper-like sheet from the filter results in a
free standing sheet of carbon nanotube bundles. This sheet, which
can conveniently range in thickness from 0.1-100 microns, possesses
mechanical strength, which is derived from the micro-scale
entanglement of the nanotube bundles. In order to increase the
mechanical properties of these sheets for the applications, it is
advantageous for the nanotube sheets to be annealed at a
temperature of at least 400.degree. C. for 0.5 hour or longer prior
to use. More advantageously, these nanotube sheets should be
annealed at a temperature of at least 1000.degree. C. for 0.5 hour
or longer in either an inert atmosphere or a hydrogen-containing
atmosphere. In order to preserve the nanotube structure, this
anneal temperature is preferably below 2000.degree. C.
Alternatively, carbon nanotubes can be deposited on a surface by
deposition from a dispersion of nanotubes in a liquid, such as
dichloroethane or water, which contains a surfactant (such as
Triton X-100 from Aldrich, Milwaukee, Wis.).
[0210] The relative and absolute sizes of working and
counter-electrodes are important for determining optimal device
design. Consider first the case where both working and counter
electrodes operate predominately non-faradaically. In order to
obtain rapid switching of the properties of the working electrode
(without the necessity of faradaic processes in the electrolyte
that can degrade cycle life), the total surface area of the counter
electrode should generally be at least about twice as large as the
working electrode. If a more rapid device response is required, the
total surface area of the counter electrode can be at least about
ten times larger than that of the working electrode. For
microdevices having extremely fast switching rates, it is most
advantageous that the total surface area of the counter electrode
is at least about a hundred times larger than that of the working
electrode. The conditions for high rate switching of the properties
of the properties of the working electrode can also be expressed in
terms of electrode capacitance, where the ratio of
counter-electrode to working-electrode capacitances is
advantageously at least about 2, more advantageously at least about
ten times, and most advantageously (for certain high rate devices)
about at least a hundred times larger than that of the working
electrode.
[0211] In addition to optimizing rate performance, these relative
surface areas and relative electrode capacitances determine the
fraction of total applied inter-electrode potential that is applied
across each individual electrode. Since the total potential that
can be applied is typically determined by the stability of the
electrolyte, the selection of increasingly large values for these
ratios is also desirable for increasing the fraction of the applied
potential that is applied across the working electrode, and
therefore increasing the amount of charge injection in this
electrode and the corresponding degree of properties change for
this electrode. On the other hand, the selection of relative
working and counter-electrode capacitances and surface areas close
to unity can be desirable if substantial properties switching is
needed for both electrodes or if there is a need to minimize device
size or weight. Device response rate decreases with increasing size
and increasing thickness electrodes, and the lowest capacitance
electrode typically determines response rate. Hence, for
applications were high charge and discharge rates are needed, the
thickness of the lowest capacity electrode should be advantageously
below about five millimeters (5000 microns), more advantageously
below about 1000 microns, even more advantageously below about 100
microns, and most advantageously below 50 microns. However, it
should be understood from the present teachings that much thinner
electrodes can be desirably used when very fast rate response is
not a performance issue and much thinner electrodes can be used in
microscopic devices where very small sizes and very high
charge/discharge rates are required. Charge injected materials that
are larger than a micron in the smallest external dimension are
useful for applications where the substantial material volumes are
required. In order to achieve large degrees of charge injection,
such materials should ideally contain at least about 50% void
volume.
[0212] With further regard to the rate of device response, the rate
response (as a fraction of the maximum achievable response)
increases with increasing values of (R.sub.SC.sub.S).sup.-1, where
R.sub.S is the effective resistance of the electrochemical system
and C.sub.S is the effective capacitance of this system. Key
contributions to R.sub.S can come from the resistivities of the
working and counter electrodes and the resistivity of the
electrolyte for ionic conduction. Consequently, it is advantageous
that the device contains at least one electrode having an
electronic conductivity at room temperature that is above 1 S/cm.
More advantageously at least one electrode has an electronic
conductivity at room temperature that is above 100 S/cm. Even more
advantageously at least one electrode has an electronic
conductivity at room temperature that is above 1000 S/cm. Most
advantageously, the device contains at least two electrodes that
each has an electronic conductivity at room temperature that is
above 1000 S/cm. In general, the electrical conductivities of
nanotube assemblies will be anisotropic and depend upon the degree
of charge injection. In such cases, the conductivities referred to
above correspond to the highest conductivity direction of the
conducting material in the most conducting state obtained by charge
injection.
[0213] From a viewpoint of having fast charge and discharge rates,
the ionic conductivity of the electrolyte is advantageously above
10.sup.-4 S/cm, more advantageously above about 10.sup.-3 S/cm, and
most advantageously above about 10.sup.-1 S/cm. In addition, device
response rates are enhanced for the more poorly conducting
electrolytes by minimizing the average thickness of electrolyte
that separates the at least two typically required device
electrodes. The maximum electrolyte thickness that separates the
two typically required electrodes is preferably less than 1
millimeter when these electrolytes are solid-state inorganic or
organic electrolytes. More preferably this average electrolyte
thickness is less than 0.1 millimeters for solid-state
electrolytes. However, for highly conducting ionic fluids (such as
aqueous salts like aqueous NaCl, aqueous bases like aqueous KOH,
and aqueous acids like aqueous sulfuric acid) much larger
inter-electrode separations can be used without adversely effecting
charge and discharge rates.
[0214] Various inorganic and organic liquid, gel, and solid
electrolytes can be used for preferred invention embodiments.
Generally, liquid electrolytes are ideally suitable for the
processes of this invention in which charge is electrochemically
induced non-faradaically in a material, and the charge and
associated properties changes are retained when the electrode
material is disconnected from the power source and the electrolyte
is removed from the electrode material. The reason for this
suitability is that the convenient means are available for removal
of liquid electrolytes from the charge-injected material without
eliminating the charge injection (such as simple evaporation of the
solvent component of the electrolyte). On the other hand,
solid-state electrolytes are suitable for devices that retain the
used electrolyte, since the use of solid-state electrolytes
eliminates the problems of liquid electrolyte containment and
incompatibility with the generic strategies conventionally used for
device fabrication. There are tradeoffs between these different
electrolyte types with respect to the allowable temperature and
voltage operating ranges and the obtainable electrical
conductivities. An electrolyte that is suitable (because of its low
cost and high ionic conductivity) is water containing simple salts,
like 1 M NaCl or 1 M KCl. Other very high ionic conductivity
electrolytes (like concentrated aqueous KOH and sulfuric acid) are
also suitable for providing rapid charging and discharge. Aqueous
electrolytes comprising at least about 4 M aqueous H.sub.2SO.sub.4,
or 4 M aqueous KOH, are especially suitable for application
embodiments where the electrolyte is used only for materials
processing by charge injection. Aqueous electrolytes comprising
about 38 weight percent H.sub.2SO.sub.4 and electrolytes comprising
above 5 M aqueous KOH are most especially suitable. For
applications where a large degree of charge injection is needed,
electrolytes with large redox windows are suitable. Especially
suitable organic electrolytes include propylene carbonate, ethylene
carbonate, butylene carbonate, diethyl carbonate, dimethylene
carbonate, and mixtures thereof with salts such as LiClO.sub.4,
LiAsF.sub.6, LiPF.sub.6, LiBF.sub.4, LiCF.sub.3SO.sub.3,
Li(CF.sub.3SO.sub.2).sub.2N, and Li(CF.sub.3SO.sub.2).sub.3C. Ionic
liquid electrolytes (like 1-butyl-3-methyl imidazolium
tetrafluoroborate) and ionic liquids in polymer matrices are
especially suitable because of the achievable wide redox stability
range and the cycle life that they provide for redox cycling
[0215] Solid-state electrolytes can also be used advantageously,
since such electrolytes enable all-solid-state devices. Suitable
organic-based solid-state electrolytes are polyacrylonitrile-based
solid polymer electrolytes (with salts such as potassium, lithium,
magnesium, or copper perchlorate, LiAsF.sub.6, and
LiN(CF.sub.3SO.sub.2).sub.2). Suitable organic solvents for these
solid-state and gel electrolytes include propylene carbonate,
ethylene carbonate, .gamma.-butyrolactone, and mixtures thereof.
Suitable gel or elastomeric solid electrolytes include lithium
salt-containing copolymers of polyethylene oxide (because of high
redox stability windows, high electrical conductivities, and
achievable elastomeric properties), electrolytes based on the
random copolymer poly(epichloridrin-co-ethylene oxide), phosphoric
acid containing nylons (such as nylon 6,10 or nylon 6), and
hydrated poly(vinyl alcohol)/H.sub.3PO.sub.4. Other suitable gel
electrolytes include polyethylene oxide and polyacrylonitrile-based
electrolytes with lithium salts (like LiClO.sub.4) and ethylene and
propylene carbonate plasticizers. The so called "polymer in salt"
elastomers (S. S. Zhang and C. A. Angell, J. Electrochem. Soc. 143,
4047 (1996)) are also suitable for lithium-ion-based devices, since
they provide very high lithium ion conductivities, elastomeric
properties, and a wide redox stability window (4.5-5.5 V versus
Li.sup.+/Li). Suitable electrolytes for high temperature device
applications include ionic glasses based on lithium ion conducting
ceramics (superionic glasses), ion exchanged .beta.-alumina (up to
1,000.degree. C.), CaF.sub.2, La.sub.2Mo.sub.2O.sub.9 (above about
580.degree. C.) and ZrO.sub.2/Y.sub.2O.sub.3 (up to 2,000.degree.
C.). Other suitable inorganic solid-state electrolytes are AgI,
AgBr, and Ag.sub.4RbI.sub.5. Suitable inorganic molten salt
electrolytes for high temperature devices include alkali metal
halides (such as NaCl, KCl, and mixtures of these salts) and
divalent metal halides (such as PbCl.sub.2). Some of the
proton-conducting electrolytes that are useful in invention
embodiments as the solid-state electrolyte include, among other
possibilities, Nafion, S-PEEK-1.6 (a sulfonated polyether ether
ketone), S-PBI (a sulfonated polybenzimidazole), and phosphoric
acid complexes of nylon, polyvinyl alcohol, polyacryamide, and a
polybenzimidazole (such as
poly[2,2'-(m-phenylene)-5,5'-bibenzimidazole].
[0216] The devices of some embodiments can use either one
electrolyte or more than one electrolyte. For example, the
electrolyte that contacts part of a porous nanostructured electrode
can be different from the electrolyte that further provides an ion
conducting path between electrodes. Also, different electrolytes
can be used as contacting materials for different electrodes.
Employing more than one electrolyte can be used to optimize device
operation. For instance, a particular electrolyte (or electrolytes)
can be chosen for optimizing either double-layer formation or
electrode ionic conductivity. While the electrolyte that separates
electrodes is substantially electronically insulating, the ion
conductor that contacts an individual electrode can have a
significant degree of electronic conductivity. In fact, conducting
polymers are used as the electrode conducting element for certain
described invention embodiments. Like electrolytes, these
conducting polymers can provide ion conduction and serve as an ion
source. However, unlike electrolytes, these conducting polymers are
electronically conducting. Hence, it is advantageous that these
conducting polymers do not provide an uninterrupted electronic path
between opposite electrodes. For this reason, conducting polymers
(or other ion-intercalated electronic conductors) are
advantageously used in combination with one or more electronically
insulating electrolyte to form a inter-electrode pathway that is
largely interrupted for inter-electrode electronic transport, but
maintained for inter-electrode ion transport.
[0217] The following examples are provided to demonstrate
particular embodiments of the present invention. It should be
appreciated by those of skill in the art that the methods disclosed
in the examples which follow merely represent exemplary embodiments
of the present invention. However, those of skill in the art
should, in light of the present disclosure, appreciate that many
changes can be made in the specific embodiments described and still
obtain a like or similar result without departing from the spirit
and scope of the present invention.
Example 1
[0218] This example shows that the electrical conductivity of
carbon nanotube electrode sheets can be continuously varied by
about an order of magnitude using predominately non-faradaic
electrochemical charge injection in 1 M NaCl electrolyte
[0219] These nanotube sheets are fabricated analogously to a
previously described method (A. G. Rinzler et al., Appl. Phys. A
67, 29 (1998)) by dispersing the carbon nanotubes in
surfactant-containing aqueous solution and filtering this nanotube
dispersion though through a 47 mm diameter
poly(tetrafluoroethylene) filter sheet (Millipore LS) under house
vacuum. A sheet of nanotubes collected on the filter paper, which
was washed using water and methanol, dried, and then lifted from
the filter paper substrate to provide a free-standing SWNT sheet.
The density of these nanotube sheets is about 0.3 g/cm.sup.3,
versus the density of about 1.3 g/cm.sup.3 for densely packed
nanotubes having close to the observed average nanotube diameter.
Hence the void volume in these nanotube sheets is about 76.9 volume
percent. This high void volume, and the correspondingly high
accessible surface area, is generally important for achieving high
degrees of non-faradaic charge injection. Supporting this
conclusion, the measured BET surface area determined from nitrogen
gas adsorption for these nanotube sheets is approximately 300
m.sup.2/g.
[0220] Two carbon nanotube electrodes (one small: .about.10 mm wide
and .about.30 mm long, and the other large: .about.1 mm wide and
.about.30 mm long) were cut from free-standing carbon nanotube
sheets made by the above described method. For the purpose of
making four-probe electrical conductivity measurement, four
in-series electrical contacts were made to the small carbon
nanotube electrode. Gold wire (0.0127 mm diameter) was attached
using heat cured conductive epoxy (H20E, Epoxy Technology), which
was cured at 70.degree. C. for four hours. The area of electrical
contact was covered by chemically resistive epoxy (Eccobond A 316,
Emerson & Cuming) in order to protect the electrical contact
from exposure to subsequently used electrolytes. One electrical
contact was similarly made to the large carbon electrode.
[0221] The small carbon nanotube electrode with four-probe
contacts, the large carbon nanotube electrode having one electrical
contact, and the Ag/AgCl reference electrode were immersed in 1M
aqueous NaCl. This selected configuration (with the larger
electrode about ten times smaller than the larger electrode) keeps
the potential of the bigger electrode almost constant as a
potential difference is applied between the small electrode
(working electrode) and the large electrode (working electrode), to
thereby change the potential of the of the small electrode relative
to reference electrode. The four-point electrical conductivity of
the small electrode was measured as a function of the potential of
this electrode with respect to the Ag/AgCl reference electrode.
[0222] Multiple experiments using the above-described method
provided the following results. When the potential of the carbon
nanotube paper electrode (versus Ag/AgCl) was in the range of about
-0.35 to about -0.50V, the conductivity showed the minimum of
60-100 S/cm. This conductivity for an uncharged carbon nanotube
sheet monotonically increased up to about 1000 S/cm when the
applied potential was increased to above +0.9V, and it
monotonically increased up to about 250 S/cm when the potential was
decreased to below -0.9V. The dependence of conductivity change
upon potential (versus Ag/AgCl) was quite symmetric with respect to
the potential at which the conductivity minimum occurs (at between
about -0.35 about -0.50V).
[0223] Cyclic voltammetry in this potential range shows that there
are no noticeable peaks due to Faradaic processes in the utilized
potential range, indicating that the major electrical conductivity
increases (up to an order of magnitude) are the result of
predominately non-faradaic charging.
Example 2
[0224] This example demonstrates that tuning the electrical
conductivity of single-wall nanotube sheets over an
order-of-magnitude range can also be accomplished using an organic
electrolyte, instead of the aqueous electrolyte of Example 1.
[0225] The experimental procedure was exactly the same as for
Example 1, except that the 1M NaCl aqueous electrolyte was replaced
with either 0.1M TBAPF.sub.6 (tetrabutyl ammonium
hexafluorophosphate) or 0.1M TBABF.sub.4 (tetrabutyl ammonium
tetrafluoroborate) dissolved in acetonitrile and the Ag/AgCl
reference electrode was replaced with the Ag/Ag.sup.+ reference
electrode. There were no significant differences noted between
charge-injection tuning of the nanotube sheet conductivity in 0.1M
TBAPF.sub.6/acetonitrile and in 0.1 M TBABF.sub.4/acetonitrile.
[0226] Because the potential window of non-faradaic reaction in
non-aqueous electrolyte is much wider than for aqueous
electrolytes, the potential can be changed up to +1.0V (versus
Ag/Ag.sup.+) in the positive potential direction, resulting in a
maximum electrical conductivity of above 1000 S/cm at these
positive potentials. In the negative potential direction,
non-faradaic charging was possible down to -1.5V (versus
Ag/Ag.sup.+), resulting in a maximum electrical conductivity of
about 600-700 S/cm at these negative potentials (where electrons
are being injected into the nanotubes). The minimum value was
.about.100 S/cm near -0.3V (versus Ag/Ag.sup.+).
[0227] FIG. 1 shows the presently observed continuous tunability of
four-point electrical conductivity as a function of applied
potential (versus Ag/Ag.sup.+) for a sheet of single-wall carbon
nanotubes immersed in one of these organic electrolytes (0.1M
tetrabutylammonium hexafluorophosphate in acetonitrile). These
results show that the electrical conductivity can be increased by
over an order of magnitude by electrochemical charge injection.
There is slight hysteresis evident for the curves in FIG. 1, with
the conductivity a on the extreme left side of the potential
minimum slightly higher for hole injection (increasingly positive
applied potential) and the conductivity slightly lower on the
extreme right side of the minimum for electron injection
(increasingly negative applied potentials). The different curves
are for three successive cycles (using squares (101), circles
(102), and triangles (103) for successive cycles).
[0228] FIG. 3 shows measured cyclic voltammetry during charge
injection for the above SWNT sheet when immersed in the above
tetrabutylammonium hexafluorophosphate electrolyte. The absence of
major peaks in this cyclic voltammetry curve (measured versus
Ag/Ag.sup.+) indicates that charging is predominately non-faradaic
for this electrolyte and potential range.
Example 3
[0229] This example shows that the hysteresis in properties tuning
can be significantly reduced if electrical conductivity is varied
by charging the amount of injected charge, as opposed to being
controlled by changing applied voltage. This point is illustrated
for the 0.1M tetrabutylammonium hexafluorophosphate/acetonitrile
electrolyte by comparing the hysteresis in the electrical
conductivity versus potential (FIG. 1 and Example 1) with those in
FIG. 2, where electrical conductivity is plotted versus charge per
carbon in the nanotube working electrode. The decrease that is
obtained in hysteresis in going from voltage control of
conductivity to charge control is even larger for the aqueous 1 M
NaCl electrolyte of Example 1. The data in FIG. 2 also shows the
dependence of four-point electrical conductivity upon the amount of
injected charge (per carbon) for the nanotube sheet is nearly
identical for the 1 M NaCl electrolyte of Example 1 and the 0.1M
tetrabutylammonium hexafluorophosphate/acetonitrile electrolyte of
Example 2. For the results in FIG. 2, the black data points are for
experiments using 0.1M tetrabutylammonium
hexafluorophosphate/acetonitrile electrolyte and near-white data
points are for measurements using 1 M aqueous NaCl electrolyte. The
origin of the charge/carbon scale in FIG. 2 is arbitrary.
Example 4
[0230] The unexpected results described in this example show that
charge non-faradaically injected into carbon nanotube sheets is
retained to a large extent when the carbon nanotube sheet is
disconnected from the power source, and then dried in either air or
in flowing nitrogen gas. This retention of injected charge is
indicated by substantial retention of the electrical conductivity
enhancement caused by charge injection.
[0231] The nanotube sheet preparation and the method of charge
injection and conductivity measurement is the same as in the above
examples. The electrolyte used is the 0.1M tetrabutylammonium
hexafluorophosphate/acetonitrile of Examples 2 and 3.
[0232] FIG. 4 shows that the dramatic hole-injection-induced
increase in electrical conductivity of the nanotube sheet in FIGS.
1 and 2 is largely retained when the hole-injected electrode is
dried in flowing dry nitrogen atmosphere to remove the electrolyte.
The insert to this figure shows results over a four-hour period on
an expanded time scale. FIG. 5 shows the retention of conductivity
enhancement when the nanotube sheet is removed from the electrolyte
and held in air for the investigated five day period.
Example 5
[0233] This example demonstrates the generality of non-faradaically
injecting charge in a nanostructured material in accordance with
embodiments of the present invention, and maintaining this injected
charge and associated properties changes when the charge injected
material is removed from the electrolyte and dried. More
specifically, it is shown that charge non-faradaically injected
into platinum nanoparticle pellets is substantially maintained even
after disconnecting the platinum pellets from the power source,
their removal from the electrolyte, and exposure of these pellets
to a dynamically pumped vacuum for a week.
[0234] The platinum nanoparticle pellets investigated here were
made at room temperature by compaction of 30 nm platinum
nanoparticles. This method used for making the investigated pellets
is like that described by J. Weissmuller et al. in Science 300, 312
(2003). The method used for attaching electrodes to the platinum
pellets is the same as that described in Example 1 for the carbon
nanotube sheets.
[0235] The density of these Pt pellets were low (2.74 to 2.96
g/cm.sup.3 for an applied compaction pressure of about 0.6 to 1 MPa
and 3.71 to 3.75 g/cm.sup.3 for an applied compactions pressure of
about 2.1 MPa), as compared with the density of solid Pt (21.45
g/cm.sup.3). These densities correspond to a volume void space in
the Pt pellets of between 81.6 and 87.2 volume percent. This high
volume fraction of void space and the corresponding high
gravimetric surface area explains the high degree of non-faradaic
charge injection that results for modest applied potential for the
Pt electrode.
[0236] FIG. 6 provides cyclic voltammetry measurements (20 mV/sec
using 1 M aqueous NaCl electrolyte) for nanostructured platinum
electrodes, showing that charging is predominately by double-layer
charge injection, which is a non-faradaic process. There are no
current peaks due to Faradaic processes and the current at constant
voltage scan rate (20 mV/sec) varies little with potential. From
plots of current versus potential scan rate in 1 M aqueous NaCl
electrolyte), the electrode capacitance is about 14.5 F/g.
[0237] Most importantly, it is found that the nanoporous Pt
electrode remains charged when disconnected from the power source,
removed from the electrolyte, and dried. Indication of this
retained charge is provided by reimmersion of the nanoporous Pt
electrode in the 1 M NaCl electrolyte, and finding that the
electrode potential is substantially unchanged. Initial results
indicating this stability are shown in FIG. 7. Just like the case
for the carbon nanotube electrode, the potential of the negatively
charged electrode is less stable than for the positively charged
electrode, as indicated by the results shown in the lower part of
this figure.
[0238] Since it is possible that some electrolyte is still retained
inside the pores of the Pt electrode during the experiment of FIG.
7, this experiment was repeated using much longer time periods
after the pellet electrodes have been disconnected from the power
source and the electrolyte was removed from the electrochemical
cell. The electrode potentials (versus Ag/AgCl) before after this
two days exposure of the electrodes to dynamic vacuum pumping were
+0.58 V and +0.45 V for the hole-injected electrode and -0.04 V and
+0.02 for the electron-injected electrode. The potential between
the two electrodes changed from the initial 0.62 V to a final 0.43
V after removal of the electrodes from the electrolyte and applying
a dynamic vacuum on these electrodes for two days.
[0239] To further evaluate the stability of charge storage, the
time period in which the platinum pellets were exposed to dynamic
vacuum was extended to a week. After this, the Pt electrodes were
reimmersed in the 1 M NaCl electrolyte to determine their charge
state by electrochemical potential measurements (naturally, without
applying any external potential). High charge storage is again
indicated for the positively charged Pt electrode (indicated by
retention of a 0.28 V potential, versus Ag/AgCl, compared with the
potential immediately before electrolyte removal of 0.33 V). The
negatively charged electrode had lower stability, as indicated by a
potential change from the initial -0.68 V (before removal of the
electrolyte and the week-long process of drying the electrode in
dynamic vacuum) to a final potential on initial reimmersion into
the electrolyte of -0.32 V.
[0240] All patents and publications referenced herein are hereby
incorporated by reference. It will be understood that certain of
the above-described structures, functions, and operations of the
above-described embodiments are not necessary to practice the
present invention and are included in the description simply for
completeness of an exemplary embodiment or embodiments. In
addition, it will be understood that specific structures,
functions, and operations set forth in the above-described
referenced patents and publications can be practiced in conjunction
with the present invention, but they are not essential to its
practice. It is therefore to be understood that the invention may
be practiced otherwise than as specifically described without
actually departing from the spirit and scope of the present
invention as defined by the appended claims.
* * * * *