Restoration of conductivity with TTF-TCNQ charge-transfer salts

Article abstract of DOI:10.1002/adfm.201000159

The formation of the conductive TTF-TCNQ (tetrathiafulvalenetetracyanoquinodimethane) charge-transfer salt via rupture of microencapsulated solutions of its individual components is reported. Solutions of TTF and TCNQ in various solvents are separately in

Full text of DOI:10.1002/adfm.201000159

www.afm-journal.de
www.MaterialsViews.com
Restoration of Conductivity with TTF-TCNQ
Charge-Transfer Salts
By Susan A. Odom, Mary M. Caruso, Aaron D. Finke, Alex M. Prokup,
Joshua A. Ritchey, John H. Leonard, Scott R. White, Nancy R. Sottos, and
Jeffrey S. Moore*
strategy.[7–9] One of the few attempts to
demonstrate electronic self-healing was
reported by Bielawski and co-workers in
the examination of reversible bond forma-
tion between N-heterocyclic carbenes and
transition metals.[7]
The development of electronic devices
capable of autonomic repair requires self-
healing components that are transported to
the site of damage and subsequently
activated. Self-healing polymers[10–13] that
utilize microencapsulated polymer precur-
sors or reagents and efficiently heal damage
from external stimuli are excellent exam-
ples that display requisite mass transport
and activation behavior. Recently, Caruso
et al. encapsulated suspensions of conduc-
The formation of the conductive TTF-TCNQ (tetrathiafulvalene–
tetracyanoquinodimethane) charge-transfer salt via rupture of
microencapsulated solutions of its individual components is reported.
Solutions of TTF and TCNQ in various solvents are separately incorporated
into poly(urea-formaldehyde) core–shell microcapsules. Rupture of a mixture
of TTF-containing microcapsules and TCNQ-containing microcapsules
results in the formation of the crystalline salt, as verified by FTIR spectroscopy
and powder X-ray diffraction. Preliminary measurements demonstrate the
partial restoration of conductivity of severed gold electrodes in the presence of
TTF-TCNQ derived in situ. This is the first microcapsule system for the
restoration of conductivity in mechanically damaged electronic devices in
which the repairing agent is not conductive until its release.
tive carbon nanotubes in the core of poly(urea-formaldehyde)
(PUF) microcapsules and successfully released the nanotubes
upon capsule rupture.[9] For the present study, we explore an
approach in which the conductive healing agent is generated upon
mechanical damage. The major advantage of this approach is the
greater mobility of the precursor solutions compared to suspen-
sions of conductive particles; here it is possible that both core
solutions may travel by capillary action to the relevant damage site
before forming the conductive salt.
The conductive tetrathiafulvalene–tetracyanoquinodimethane
(TTF-TCNQ) charge-transfer salt[14,15] is a promising system for
restoration of conductivity that can be generated from solutions of
its precursors. Individually, TTFand TCNQ are soluble in a variety
of organic solvents, commercially available in bulk quantity, and
are non-conductive. The formation of the relatively insoluble
charge-transfer salt (Scheme 1) occurs rapidly at room temperature,
1. Introduction
Mechanical durability is an important concern for integrated
microelectronic devices.[1–4] Thermomechanical failures result
from the mismatch of thermal expansion coefficients, Young’s
modulus, and Poisson’s ratio of the package materials.[5] These
mismatches are the most common reasons for failures in
integrated circuits that are caused by discontinuities, or ‘‘electrical
opens.’’[6] Despite the importance of mechanical failures to the
lifetime of electronic circuitry, the development of electronic
devices that are capable of autonomously repairing mechanical
damage in electronic devices remains a relatively unexplored
[*] Prof. J. S. Moore, Dr. S. A. Odom, M. M. Caruso, A. D. Finke,
A. M. Prokup, J. A. Ritchey, J. H. Leonard
Department of Chemistry and The Beckman Institute
University of Illinois at Urbana-Champaign
405 N. Mathews Ave. Urbana, IL 61801 (USA)
E-mail: jsmoore@uiuc.edu
Prof. N. R. Sottos
Department of Materials Science & Engineering and
The Beckman Institute
University of Illinois at Urbana-Champaign
405 N. Mathews Ave. Urbana, IL 61801 (USA)
Prof. S. R. White
Department of Aerospace Engineering and The Beckman Institute
University of Illinois at Urbana-Champaign
405 N. Mathews Ave. Urbana, IL 61801 (USA)
DOI: 10.1002/adfm.201000159
Scheme 1. Formation of TTF/TCNQ charge-transfer salt.
Adv. Funct. Mater. 2010, 20, 1721–1727
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1721

www.afm-journal.de
www.MaterialsViews.com
and preferentially forms the TTF-TCNQ salt in a 1:1 molar ratio,
even when TTFand TCNQ are not combined in equal amounts.[16]
Rapid and spontaneous generation of conductive TTF-TCNQ via
microcapsule delivery of solutions of its constituents could enable
autonomous conductivity restoration.
Single-crystal TTF-TCNQ shows a high conductivity of ca. 400–
500 S ꢀ cmꢁ1 [17,18]
,
making it a useful material for electronic
devices.[19–21] The conductivity of polycrystalline TTF-TCNQ has
also been investigated, fabricated through chemical vapor
deposition,[22] thermal evaporation,[23] and from crystals grown
from combined solutions of TTF (the electron donor) and TCNQ
(the electron acceptor).[24] These films generally have conductiv-
itiesof 5–30 S ꢀ cm
.
ꢁ1 [19,22,25] In arecent exampleof solution-based
salt formation, conductive films and circuits were deposited onto
substrates using inkjet printing of the donor and acceptor
solutions, forming the charge-transfer salt (in ca. 10 s) before
complete solvent (dimethyl sulfoxide) evaporation. Without post-
processannealing, thismethodproducedfilmswithconductivities
Figure 1. Optical microscope images from A) an attempt to encapsulate
crystalline TTF-TCNQ salt in PA, B) MCs containing powdered TTF-TCNQ
salt suspended in PA; inset: ruptured MCs containing powdered TTF-
TCNQ salt in PA, C) TTF-PA MCs, and D) TCNQ-PA MCs. All scale bars are
200 mm.
of 5–10 S ꢀ cmꢁ1 [24]
,
comparable to that of vacuum-deposited
films.[19,22,25]
In this article, we report the encapsulation of individual
solutions of TTFand TCNQ, as well as characterization of the core
materials. We additionally demonstrate the formation of the TTF-
TCNQ charge-transfer salt from the rupture of a mixture of TTF-
and TCNQ-containing microcapsules. We analyze the micro-
capsules upon rupture and report the electrical characterization of
ruptured microcapsules to determine their potential for the
restorationofelectricalconductivity. Thesemeasurementsinclude
current–voltage (I–V) sweeps of the microcapsules between
tungsten probe tips and over severed gold circuits.
more than other encapsulatable solvents. The solubility of each
compound and the minimum solution concentrations required
forprecipitationofthecharge-transfersaltfrommixedsolutionsof
equal volumes and concentrations were determined for each
solvent (Table S1). The minimum concentration required for salt
precipitation for all solvents was one-half to one-third of the
maximum solubility of TCNQ. For this test the minimum
concentration required for precipitation was given for precipita-
tion of the salt within 5 min after mixing room temperature
solutions and the salt formed in all cases.
2. Results and Discussion
TTF and TCNQ were individually incorporated into micro-
capsule cores as solutions in PhCl, EPA, and PA at the maximum
solution concentrations, although TCNQ suspensions were
sonicated at 40 8C to allow for higher concentrations of TCNQ
into the core solutions (Table S2). PUF core–shell microcapsules
were prepared using an in situ emulsification polymerization
following our reported procedure.[26,28] Higher concentrations of
TTF and TCNQ were encapsulated in PA by modifying the
encapsulation procedure through addition of 3.0 g of a poly-
urethane prepolymer to the core solution prior to addition to the
reaction mixture.[29] Scanning electron microscope (SEM) and
optical microscope images confirm microcapsule formation for all
three solvent systems (Fig. 1C and D, and Supporting
Information).
Electron impact mass spectra of the dried microcapsule core
solutions confirmed the presence of TTF and TCNQ in the
microcapsules (Fig. S8 and S9). Quantitative analysis of TTF and
TCNQ was performed by taking the absorption spectra of diluted
solutions of the microcapsule cores (Fig. 2). Beer’s law plots of TTF
and TCNQ in solvent provided quantitative calibration (see
Supporting Information). The wt% of each compound was then
calculated after correcting for solvent dilution based on spectra of
the microcapsule core solutions (Table 1).
Unlike the straightforward encapsulation of conductive carbon
nanotubes,[8] suspensions of the TTF-TCNQ salt were not readily
encapsulated using an emulsification polymerization procedure.
In a preliminary attempt to encapsulate polycrystalline TTF-
TCNQ, we found that the crystals, regardless of size, were not
incorporated within the microcapsules (abbreviated MCs in
figures and tables) but instead adhere to the shell wall or are in
partial contact with the microcapsule exterior (Fig. 1A). These
resulting capsules, which contained a solvent core, collapsed upon
drying and sieving. In a second approach, powdered TTF-TCNQ
salt was encapsulated when the particle size was much smaller
than the microcapsule diameter (Fig. 1B); however, microcapsules
ruptured by manual crushing by mortar and pestle (the technique
used for rupturing microcapsules throughout this publication) did
not release the salt from the interior of the capsules (Fig. 1B, inset).
Due to the difficulty of encapsulating the TTF-TCNQ salt directly,
we investigated the encapsulation of individual solutions of each
compound.
TTF and TCNQ were qualitatively screened in various solvents
that have been shown to form high-quality microcapsules and
promoteself-healing in epoxypolymers.[26,27] WhileTTFdisplayed
good solubility in most solvents, the solubility of TCNQ was much
lower and proved to be the limiting factor in solvent choice. Of
the solvents investigated, chlorobenzene (PhCl), ethyl phenylace-
tate (EPA), and phenyl acetate (PA) dissolved TCNQ significantly
The amount of TTF in each type of microcapsule (by wt%) was
similar to that of the initial core solutions. Sulfur elemental
analysis of the isolated shell walls showed 1%–2% sulfur by mass,
Adv. Funct. Mater. 2010, 20, 1721–1727
1722
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.afm-journal.de
www.MaterialsViews.com
Figure 3. Microcapsules crushed between two glass slides: A) 50 mg PA-
MCs; B) 50 mg TTF-PA MCs; C) 50 mg TCNQ-PA MCs; D) 50 mg each TTF-
PA and TCNQ-PA MCs.
Figure 2. UV-vis absorption spectra of filtered suspensions of ruptured
TTF-PA MCs (black), supernatant of TTF-PA MCs soaked for 60 min in PA
(red), ruptured TCNQ-PA MCs (blue), and supernatant of TCNQ-PA MCs
soaked for 60 min in PA (green), all spectra obtained from samples
containing 2.0 mg MCs per mL PA.
When mixtures of TTF and TCNQ microcapsules were
ruptured,adark-browncolorwasimmediatelyobserved,indicative
of the TTF-TCNQ charge-transfer salt formation (Fig. 3).
Previously, the TTF-TCNQ charge-transfer salt has been char-
acterized by UV-vis-NIR absorption spectra in acetonitrile.[31] In
this study, UV-vis analysis was not useful for determining of the
formation of the TTF-TCNQ salt; the characteristic bands of the
TCNQ radical anion (including one band with lmax at 744 nm) that
are used to identify thecharge-transfer salt formation werepresent
in the absorption spectrum of TCNQ in PA alone. This effect is
presumably dueto theelectronicinteraction ofTCNQ with PA(see
Supporting Information and Fig. S14 and S15 for more details).
Because UV-vis analysis was insufficient to determine the
formation of the TTF-TCNQ salt, IR spectroscopy was used to
verify charge-transfer salt formation. The C–H stretching region
(Fig. 4) and C–N stretching region (Fig. S16) of neat TTF (3102,
3081, 3063 cmꢁ1), TCNQ (3050, 2220 cmꢁ1), and TTF-TCNQ
(3093, 3073, 2203 cmꢁ1) were compared to spectra of the isolated
core materials. In all three cases, the spectra of the core materials
matched those of the neat compounds, consistent with incorpora-
tion of TTF and TCNQ into the microcapsules, as well as TTF-
TCNQ salt formation upon microcapsule rupture and mixing.
When a mixture of ruptured TTF-PA and TCNQ-PA micro-
capsules was dispersed in mineral oil immediately after rupture
and suspension in oil, optical micrographs revealed small particles
that are not present in ruptured TTF-PA or TCNQ-PA micro-
capsules alone (Fig. 5 and S5). Powder X-ray diffraction (XRD) was
Table 1. Weight percentages of TTF and TCNQ incorporated into the
microcapsules based on UV-vis absorption spectra of diluted solutions
of crushed, filtered microcapsules.
core solvent
TTF [wt%]
TCNQ [wt%]
PhCl
EPA
PA
0.22
0.27
0.75
1.1
0.043
0.044
0.34
PA [a]
0.70
[a] Polyurethane prepolymer added to microcapsule core.
which corresponds to 0.6–1.2 wt% TTF in the shell wall (Table S5).
In all three solvents, the wt% of TCNQ in the microcapsule cores
was significantly lower than the initial amount in the core solution.
It is possible that TCNQ is incorporated into the shell wall or
precipitated from solution, given its lower solubility in PA,
resulting in the lower relative core concentrations. Consistent with
this possibility for shell-wall incorporation is the green color of the
TCNQ-PA microcapsule shell walls, presumably due to the
presence of reduced TCNQ.[30] Importantly, the core solutions for
TCNQ-PA (and TTF-PA) capsules are yellow, which is evident
upon filtration of ruptured microcapsules.
DespitetherelativelyminordifferenceinsolubilitiesofTTFand
TCNQ in the three core solvents, the amounts of each compound
incorporated into the microcapsule cores were ca. 3ꢂ higher and
10ꢂ higher (by wt%), respectively, for PA microcapsules than for
PhCl and EPA microcapsules. In PA, the amount of TCNQ
encapsulated was ca. 50% that of TTF. Attempts to increase the
amount of encapsulated TCNQ by increasing the concentration
of TCNQ in the core solvent by preparing supersaturated core
solutions led to poor quality capsules that leached core content
rapidly. PA microcapsules with higher concentrations of TTF
or TCNQ were prepared with polyurethane prepolymer added
to the core. The modified capsules showed increased thermally
stability compared to the TTF-PA microcapsules (as measured
by thermogravimetric analysis (TGA), see Supporting
Information), but also showed significant leaching of core content
after 2 weeks.
Figure 4. Selected region of the IR spectra of KBr pellets of TTF, TCNQ,
and TTF-TCNQ (solid lines) and spectra from ruptured, filtered, and dried
PA-based MCs (dotted lines).
Adv. Funct. Mater. 2010, 20, 1721–1727
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1723

www.afm-journal.de
www.MaterialsViews.com
Figure 6. Experimentally observed diffraction pattern of the precipitate
isolated from combining the TTF-PA and TCNQ-PA core materials (black)
and simulated pattern (red).
Figure 5. Optical micrograph of 1:1 (wt%) of crushed of TTF-PA:TCNQ-PA
microcapsules dispersed in mineral oil.
used to determine if the particles were crystalline TTF-TCNQ.
Because the presence of shell wall and polymer results in broad
diffraction bands that interfere with the analysis of the precipitate
itself, for this experiment, TTF-PA and TCNQ-PA microcapsules
were individually ruptured and shell walls removed with a syringe
filter; then a 1:1 v/v ratio of the cores was combined. A black solid
immediately precipitated, which was filtered and washed with
acetonitrile. Powder XRD analysis of the black solid gave a pattern
that matched a simulated powder pattern based on the previously
reported single crystal structure of TTF-TCNQ (Fig. 6 and
Supporting Information), thus indicating the production of
crystalline TTF-TCNQ.
After verifying the charge-transfer salt formed upon rupture
and combination of TTF-PA and TCNQ-PA microcapsule cores,
electrical properties of the ruptured microcapsules were examined
by I–V measurements. Before analyzing ruptured microcapsules,
comparative solutions of TTFin PA, another of TCNQ in PA, and a
suspension of TTF-TCNQ in PA were analyzed by a procedure for
collecting I–V data similar to that used for the analysis of
microcapsules containing carbon nanotube suspensions.[8] For
each case, the analyte was deposited onto a glass slide and two
tungsten probe tips were immersed in the suspension, spaced
approximately 100 mm apart. The potential was varied from ꢁ10 to
þ10 Vand the current was recorded. TTFin PA alone showed little
measurable response to applied voltage, while TCNQ in PA and
TTF-TCNQ in PA showed linear I–V responses, with the I–V ratio
ofTTF-TCNQbeingmoreintensethanthatofTCNQ(Fig. S21a). It
is possible that TCNQ in PA shows a I–V response due to partial
reductionupon interaction with PA. Measurements of TTF-TCNQ
in PA at multiple concentrations were also recorded (Fig. S21b);
the intensity of the response was dependent on the concentration
ofTTF-TCNQ. Observation byopticalmicroscopyrevealedthatthe
TTF-TCNQ crystals aligned between the probe tips, similar to the
behavior observed for carbon nanotube suspensions.[8]
Figure 7. I–V measurements of analytes on glass slides measured between
two tungsten probe tips spaced approximately 100 mm apart for A) PA and
suspensions of ruptured TTF-PA and TCNQ-PA microcapsules in a 1:1 ratio
(wt%) in PA, and B) neat ruptured TTF-PA, TCNQ-PA, and TTF-PA:TCNQ-
PA in a 1:1 ratio (wt%) microcapsules.
The same measurements were performed for suspensions of
crushed TTF-PA and TCNQ-PA microcapsules in a 1:1 ratio (wt%)
at multiple concentrations in PA (Fig. 7a). The I–V response was
also linear for these cases and dependent on microcapsule wt%.
Adv. Funct. Mater. 2010, 20, 1721–1727
1724
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.afm-journal.de
www.MaterialsViews.com
I–V measurements were also obtained for neat ruptured TTF-PA
and TCNQ-PA microcapsules, both individually and combined
(Fig. 7b). In these measurements, the current reached a maximum
value for the combination of TTF-PA and TCNQ-PA micro-
capsules after 5–7 scans, potentially due to increased alignment of
the charge-transfer salt in subsequent scans (data is shown for the
last scan). This phenomenon was previously observed in related
work when suspensions of ruptured microcapsules containing
carbon nanotubes were analyzed.[8] Without dilution, the neat
ruptured microcapsule samples were too opaque to see individual
crystals/particles and prevented optical confirmation of this effect.
The optimal ratio of TTF-PA:TCNQ-PA microcapsules for the
restoration of conductivity was determined from I–V measure-
ments performed on substrates containing model damage of an
electrical circuit. Well-defined gap distances (ꢃ100 mm) in
conductive lines were fabricated using standard photolithography
techniques and patterned masks. Gold lines were deposited onto
glass slides with gaps of varying distances at the center of each line
(30, 50, and 100 mm), including one gold line with no gap as a
control (Fig. 8, right). To ensure that the gaps were bridged with a
consistent sample, an excess of ruptured microcapsules was
deposited onto the substrate over the gaps in the gold lines.
Tungsten probe tips were used to make contact with the gold lines,
and the I–V response was measured (see image of the set-up in
Fig. 8, left).
Initially, gold lines with 50 mm gaps were used to compare neat
crushed microcapsules. The ratios of TTF-PA:TCNQ-PA micro-
capsules were varied to determine which combination of micro-
capsules gave the largest I–V response (Fig. 9a). The TTF-PA and
TCNQ-PA microcapsules alone were also analyzed and showed no
measurable current. Of the ratios of TTF-PA:TCNQ-PA micro-
capsules analyzed, the largest current response occurred with a 1:2
ratio. This ratio of capsules is consistent with delivering the optimal
1:1 ratio (by mass and molarity) of TTF:TCNQ, which is expected to
yield the highest concentration of charge-transfer salt.
Once the ideal ratio of microcapsules was determined, this
optimized ratio was used to examine the change in I–V response
over multiple gap sizes to determine the potential for conductivity
restoration for various damage states. In these experiments, a 1:2
ratio of neat crushed TTF-PA:TCNQ-PA microcapsules was
applied to substrates over the 30, 50, and 100 mm gaps. As
expected for a polycrystalline material, the I-V responses were
larger for shorter gaps (Fig. 9b). All gap sizes showed conductivity
restoration, indicating that the TTF-TCNQ charge-transfer salt
bridges at least 100 mm distances upon microcapsule rupture.
Lastly, we examined the effect of solvent removal on the
restoration of conductivity. To achieve a successful autonomic
system, conductivity restoration must be maintained upon solvent
evaporation. For this experiment, neat ruptured microcapsules
were used to bridge the 30 mm gap. Cycling was performed until
the maximum I–V was obtained (5–7 scans). The microcapsules
were then allowed to air dry on the substrate. After drying, I–V
scans were performed on the same substrate. For the dried
microcapsules, the I–Vcurve not only became more linear, but the
magnitude of the I–V response also increased 20-fold, that is, at
ꢁ10 V,themagnitudeofthecurrentincreasedfromꢁ2.1 ꢂ 10ꢁ5 to
ꢁ4.4 ꢂ 10ꢁ4 A (Fig. 9c). Based on these experimental results, we
expect that solvent removal will assist in the autonomic restoration
of conductivity.
Figure 8. A) Photograph of probe station set-up showing crushed micro-
capsules bridging breaks in gold lines on a glass substrate. B) Pattern
design for gold lines with controlled gap sizes on glass substrate optical
micrograph of a 30 mm gap.
3. Conclusions
The feasibility of using charge-transfer salt precursors in organic
solvents to restore conductivity was demonstrated. Each compo-
nent of the charge-transfer salt was separately encapsulated as
solutions in core–shell microcapsules, and the charge-transfer salt
was formed upon combination of the cores of both types of
microcapsules after rupture. Both precursor solutions are low
viscosity liquids, avoiding delivery problems of viscous solutions
or suspensions. This system has the potential to serve as a useful
model for a two-part electronic self-healing system using liquid
precursors by comparing the degree of restoration of conductivity
of one- and two-part microcapsule systems. Investigations are
currently underway to study this system in the autonomic healing
Adv. Funct. Mater. 2010, 20, 1721–1727
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1725

www.afm-journal.de
www.MaterialsViews.com
of damaged electronic systems, including comparison to alter-
native, single-capsule systems containing suspensions of con-
ductive materials.
4. Experimental
EPA, PA, PhCl, urea, ammonium chloride, and resorcinol were purchased
from Aldrich and used as received. TCNQ was purchased from Aldrich and
was purified by recrystallization from acetonitrile [32]. For PA-based
microcapsules used in electrical measurements, TTF was recrystallized
from hexanes as previously described [33]. Ethylene-maleic anhydride
copolymer (Zemac-400) powder with an average molecular weight of
400 kDa (Vertellus) was used as a 2.5 wt% aqueous solution. The
commercial polyurethane prepolymer Desmodur L 75 was purchased from
Bayer MaterialScience and used as received.
Microcapsules were prepared with slight modifications using our
previously published procedure [26] in which the shell wall materials and
aqueous phase were reduced by half from the initial urea-formaldehyde
emulsion procedure [28]. Similarly, 100 mL of deionized water at
room temperature was placed in a 600 mL beaker, along with 25 mL of
2.5% (wt/v) ethylene co-maleic anhydride as a surfactant. The beaker was
placed in a temperature controlled water bath equipped with a mechanical
stirring blade (40 mm diameter), which was brought to 400 RPM. To the
aqueous solution was added the solid wall-forming materials: urea (2.50 g),
ammonium chloride (0.25 g), and resorcinol (0.25 g). Afterward, the pH
was adjusted from 2.7 to 3.5 by addition of aqueous NaOH solution. The
core solutions, which were previously sonicated at 40 8C for 30 min to
ensure dissolution of the solids, were added to the stirring solution,
creating an emulsion. After 10 min, 6.33 g of formalin solution was added
and the temperature was increased to 55 8C at 10 8C minꢁ1. The reaction
proceeded under continuous stirring for 4 h after which the reaction
mixture was allowed to cool to room temperature. The microcapsule
solution was filtered the next day, washing with water, and then dried under
air for at least 6 h before sieving.
For the encapsulation of crystalline TTF-TCNQ, the charge-transfer salt
was formed from the combination of TTF in acetonitrile with TCNQ in
acetonitrile, then recrystallized from acetonitrile. The crystals were ground
such that the length of the longest axis of the majority of the crystals was
100 mm or less. During the encapsulation, the TTF-TCNQ salt was
suspended in PA and used as the core material following our previously
described procedure for encapsulation [26]. In the case of the
encapsulation of powdered TTF-TCNQ, the TTF-TCNQ salt was generated
in situ by pouring individual solutions of TTF and TCNQ (0.30 g TTF in
30 mL PA, 0.30 g TCNQ in 30 mL PA, each sonicated at 40 8C for 30 min
beforehand) directly into the emulsification polymerization reaction vessel.
Higher concentrations of TTF and TCNQ were prepared by adding 3.0 g
polyurethane prepolymer to the core solutions before sonication; these
microcapsules are labeled TTF-PAꢄ and TCNQ-PAꢄ.
Size distributions for capsules prepared were obtained from multiple
optical micrographs of dried capsules in mineral oil on glass slides taken
using a Leica DMR Optical Microscope at various magnifications (see
Supporting Information for size distributions). Capsule diameter
measurements were obtained from the micrographs using ImageJ analysis
software. A minimum of 100 measurements was used for each analysis of
each batch of capsules produced. Microcapsules were sieved to collect
those with diameters ranging from 125–180 mm. Images of dried capsules
were obtained using SEM (FEI/Philips XL30 ESEM-FEG) after sputter
coating with a gold-palladium source.
Figure 9. A) I–V measurements for neat ruptured microcapsules on gold
line on glass with 50 mm gap, with ratios of MCs noted. B) I–V measure-
ments for neat ruptured 1:2 TTF:TCNQ MCs deposited onto gold lines on
glass with 30, 50, and 100 mm gaps. C) I–V measurements for neat
ruptured 1:2 TTF:TCNQ microcapsules on gold lines with 30 mm gaps
immediately after crushing (green) and after drying (red).
TGA was performed on a Mettler-Toledo TGA851, calibrated by indium,
aluminum, and zinc standards. A heating rate of 10 8C ꢀ minꢁ1 was used,
and experiments were performed under nitrogen atmosphere from 25–
650 8C. For each experiment, approximately 5 mg of sample were accurately
weighed (ꢅ0.02 mg) into an alumina crucible.
Mass spectra were recorded on a 70-VSE C in EIþ mode through the
University of Illinois Mass Spectrometry Laboratory, SCS. The samples
were obtained by crushing microcapsules with a mortar and pestle. The
residues were dried under vacuum for ca. 24 h (120 mTorr) to remove the
Adv. Funct. Mater. 2010, 20, 1721–1727
1726
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.afm-journal.de
www.MaterialsViews.com
majority of the core solvent. Samples of the remaining core and shell wall
were submitted for mass spectroscopy.
Elemental analyses were performed by the University of Illinois
Microanalysis Laboratory. Samples of microcapsule shell walls were
prepared by crushing the microcapsules with a mortar and pestle, then
washing the shell walls with dichloromethane and collecting the walls with
[1] M. L. Minges, Electronic Materials Handbook: Packaging, ASM Inter-
national, Handbook Committee, 1989.
[2] A. W. Sleight, Endeavour 1995, 19, 64.
[3] M. Ohring, Reliability and Failure of Electronic Materials and Devices,
Academic Press, San Diego 1998.
¨
a Buchner funnel. The shell walls were dried at room temperature (ca.
22 8C) under vacuum (120 mTorr) for one week before submission for
elemental analysis.
[4] M. Zhao, L. Jiao, H. Yuan, Y. Feng, M. Zhang, Solid State Ionics 2007, 178,
387.
Absorption spectra were recorded using
a Shimadzu UV-vis
[5] R. B. R. van Silfhout, Y. Li, W. D. van Driel, Proc. EuroSIME (Ed.: L. J. Ernst),
Shaker Publishing BV, Maastricht 2001, pp. 207–301.
[6] P. Viswanadham, P. Singh, Failure Modes and Mechanisms in Electronic
Packages, Chapman & Hall, New York 1998.
Spectrophotometer, model number UV-1601PC. Before analysis, all
microcapsule samples were soaked for 5 min in dichloromethane, then
were rinsed with dichloromethane to remove any TTF or TCNQ from the
microcapsule surface. For soaking experiments, the microcapsules of
known masses (generally 50–100 mg) were immersed in a known volume
of dichloromethane (generally 2–5 mL). The samples were then filtered
gently using syringe filtration (pore diameter 45 mm). UV-vis absorption
spectra were recorded of the filtered solutions.
[7] K. A. Williams, A. J. Boydston, C. W. Bielawski, J. R. Soc. Interface 2007, 4,
359.
[8] M. M. Caruso, S. R. Schelkopf, A. C. Jackson, A. M. Landry, P. V. Braun, J. S.
Moore, J. Mater. Chem. 2009, 19, 6093.
[9] M. M. Caruso, D. A. Davis, Q. Shen, S. A. Odom, N. R. Sottos, S. R. White,
J. S. Moore, Chem. Rev. 2009, 109, 5755.
For absorption spectra of ruptured microcapsules, a known mass of
microcapsules (generally 50–100 mg) was ruptured by grinding in a mortar
and pestle. The slurry of ruptured microcapsules was transferred to a vial
using a known volume of solvent (generally 2–5 mL), and the suspension
was filtered to remove the shell wall. The absorption spectra of the filtered
solution was recorded, which was further diluted quantitatively to obtain
absorbance values between 0.5–1.5.
Powder XRD measurements were performed on neat crushed
microcapsules, dried versions thereof, or precipitates from the micro-
capsule cores using a Bruker General Area Detector Diffraction System.
I–V measurements were performed using a two-point measurement
technique at a Signatone table-top probe station customized with a
standard S725 manipulator. Tungsten probe tips were immersed in
suspensions of microcapsules (crushed by mortar and pestle) approxi-
mately 100 mm apart. The voltage was swept from ꢁ10 to þ10 V
continuously 5–7 times for each sample using an Agilent 4155C
Semiconductor Parameter Analyzer.
[10] S. R. White, N. R. Sottos, P. H. Geubelle, J. S. Moore, M. R. Kessler,
S. R. Sriram, E. N. Brown, S. Viswanathan, Nature 2001, 409, 794.
[11] S. H. Cho, H. M. Andersson, S. R. White, N. R. Sottos, P. V. Braun, Adv.
Mater. 2006, 18, 997.
[12] J. M. Kamphaus, J. D. Rule, J. S. Moore, N. R. Sottos, S. R. White, JRS
Interface 2008, 5, 95.
[13] S. R. White, M. M. Caruso, J. S. Moore, MRS Bull. 2008, 33, 766.
[14] L. B. Coleman, M. J. Cohen, D. J. Sandman, F. G. Yamagishi, A. F. Garito,
A. J. Heeger, Solid State Commun. 1973, 12, 1125.
[15] J. Ferraris, D. O. Cowan, V. Walatka, J. H. Perlstein, J. Am. Chem. Soc. 1973,
95, 948.
[16] S. M. Rosenfeld, R. G. Lawler, H. R. Ward, J. Am. Chem. Soc. 1973, 95,
946.
´
[17] D. Jerome, Chem. Rev. 2004, 104, 5565.
[18] E. M. E. Conwell, Semiconductors and Semimetals, Highly Conductive
Quasi-One Dimensional Organic Crystals, Vol. 27, Academic Press, London
1998.
Substrates for conductivity measurements were prepared as follows.
Glass slides were cleaned using Piranha solution (3:1 v/v conc. H2SO4:30%
H2O2) for 1 h and then rinsed with Milli-Q water (>18 MV.cm) and dried
under a stream of nitrogen. Shipley 1805 was spin-coated onto the cleaned
wafers at 3000 RPM for 30 s. The photoresist was patterned through a
transparency mask using a Karl Suss mask aligner (ꢆ16 mW cmꢁ2) for 10 s.
After development of the photoresist using AZ 351 developer, a 10 nm
chromium adhesion layer followed by a 50 nm Au layer was deposited by
electron beam evaporation. Sonication of the substrates in acetone
removed any remaining photoresist and gold overlayer. Gold gate
dimensions were confirmed by optical microscopy.
[19] Y. Takahashi, T. Hasegawa, Y. Abe, Y. Tokura, G. Saito, Appl. Phys. Lett.
2006, 88, 073504.
[20] Y. Takahashi, T. Hasegawa, S. Horiuchi, R. Kumai, Y. Tokura, G. Saito,
Chem. Mater. 2007, 19, 6382.
[21] H. Wada, T. Taguchi, B. Noda, T. Kambayashi, T. Mori, K. Ishikawa,
H. Takezoe, Org. Electron. 2007, 8, 759.
[22] A. Figueras, J. Caro, J. Fraxedas, V. Laukhin, Synth. Met. 1999, 102,
1611.
[23] K. Yase, N. Ara, A. Kawazu, Mol. Cryst. Liq. Cryst. 1994, 247, 185.
[24] M. Hiraoka, T. Hasegawa, T. Yamada, Y. Takahashi, S. Horiuchi, Y. Tokura,
Adv. Mater. 2007, 19, 3248.
Acknowledgements
[25] T. Sumimoto, K. Kudo, T. Nagashima, S. Kuniyoshi, K. Tanaka, Synth. Met.
1995, 70, 1251.
This work was partially supported by the National Science Foundation for
funding through a National Science and Engineering Initiative (Award
Number DMR-0642573). Additionally, funding for electrical measurements
and device fabrication was supported by the Center for Electrical Energy
Storage, an Energy Frontier Research Center funded by the U.S.
Department of Energy, Office of Science, Office of Basic Energy Sciences.
S. Odom thanks the National Science Foundation for an American
Competitiveness in Chemistry Fellowship funded by the American
Recovery and Reinvestment Act. M. Caruso thanks the Department of
Defense for an NDSEG Fellowship. We thank Audrey Bowen for assistance
with photolithography, Scott Robinson in the Imaging Technology Group at
the Beckman Institute for assistance with SEM imaging, and Danielle Gray
for powder XRD collection and analysis. We also thank Vertellus for their
donation of Zemac-400. Supporting Information is available online from
Wiley InterScience or from the author.
[26] B. J. Blaiszik, M. M. Caruso, D. A. McIlroy, J. S. Moore, S. R. White,
N. R. Sottos, Polymer 2009, 50, 990.
[27] M. M. Caruso, B. J. Blaiszik, S. R. White, N. R. Sottos, J. S. Moore, Adv.
Funct. Mater. 2008, 18, 1898.
[28] E. N. Brown, M. R. Kessler, N. R. Sottos, S. R. White, J. Microencapsulation
2003, 20, 719.
[29] M. M. Caruso, B. J. Blaiszik, H. Jin, S. R. Schelkopf, D. S. Stradley,
N. R. Sottos, S. R. White, J. S. Moore, ACS Appl. Mater. Interfaces.
2010, in press.
[30] L. R. Melby, R. J. Harder, W. R. Hertler, W. Mahler, R. E. Benson,
W. E. Mochel, J. Am. Chem. Soc. 1962, 84, 3374.
[31] T. Tomkiewicz, J. B. Torrance, B. A. Scott, D. C. Geen, J. Chem. Phys. 1974,
60, 5111.
[32] W. L. F. Armarego, C. L. L. Chai, Purification of Laboratory Chemicals, 5th ed.
Butterworth Heinimann, Elsevier Science, Oxford 2003.
[33] R. McIntyre, H. Gerisher, Ber. Bunsen-Ges. 1984, 88, 963.
Received: January 27, 2010
Published online: May 3, 2010
Adv. Funct. Mater. 2010, 20, 1721–1727
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1727