Polythiophene Doping of the Cu-Based Metal-Organic Framework (MOF) HKUST-1 Using Innate MOF-Initiated Oxidative Polymerization

Article abstract of DOI:10.1021/acs.inorgchem.8b03465

The copper-based metal-organic framework (MOF) HKUST-1 adsorbs organic molecules into its pores. When loaded with electron-rich oligothiophenes, the resulting system reacts under heat to initiate oxidative polymerization without the use of any other oxidant or catalyst. This reaction is not observed in the non-redox-active MOF MIL-100(Al). We have characterized the composites by optical and nanoscale microscopy, vibrational and UV-vis spectroscopy, X-ray photoelectron spectroscopy, N₂ sorption analysis, and thermogravimetric analysis/residual gas analysis. Unsubstituted oligothiophenes polymerize within MOF pores, while 3,4-ethylenedioxythiophene forms a coating on the MOF surface. MOF composites with conjugated polymer dopants trapped inside their pores undergo profound shifts in the composite electronic structure. Reasoning from time-dependent density functional theory calculations of an HKUST-1 model system bound to monomers, we rationalize the observed reactivity and propose an initiation mechanism based on a ligand-to-metal charge-transfer state.

Full text of DOI:10.1021/acs.inorgchem.8b03465

Article
pubs.acs.org/IC
Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX
Polythiophene Doping of the Cu-Based Metal−Organic Framework
(MOF) HKUST‑1 Using Innate MOF-Initiated Oxidative
Polymerization
Nicholas Marshall,*,† William James,† Jeremy Fulmer,† Scott Crittenden,‡ Anthony B. Thompson,§
Patrick A. Ward,§ and Gerard T. Rowe†
†Department of Chemistry and Physics, University of South Carolina Aiken (USC Aiken). 471 University Parkway, Aiken, South
Carolina 29801, United States
‡Department of Physics and Astronomy, University of South Carolina. 712 Main Street, Columbia, South Carolina 28208, United
States
§Applied Research Center, Savannah River National Laboratory, 301 Gateway Drive, Aiken, South Carolina 29803, United States
S
* Supporting Information
ABSTRACT: The copper-based metal−organic framework
(MOF) HKUST-1 adsorbs organic molecules into its pores.
When loaded with electron-rich oligothiophenes, the resulting
system reacts under heat to initiate oxidative polymerization
without the use of any other oxidant or catalyst. This reaction
is not observed in the non-redox-active MOF MIL-100(Al).
We have characterized the composites by optical and
nanoscale microscopy, vibrational and UV−vis spectroscopy,
X-ray photoelectron spectroscopy, N2 sorption analysis, and
thermogravimetric analysis/residual gas analysis. Unsubsti-
tuted oligothiophenes polymerize within MOF pores, while 3,4-ethylenedioxythiophene forms a coating on the MOF surface.
MOF composites with conjugated polymer dopants trapped inside their pores undergo profound shifts in the composite
electronic structure. Reasoning from time-dependent density functional theory calculations of an HKUST-1 model system
bound to monomers, we rationalize the observed reactivity and propose an initiation mechanism based on a ligand-to-metal
charge-transfer state.
to a number of highly efficient devices3,13,14 but suffers from
various degradation processes that place inherent limits on
BHJ cells. Many of the degradation modes of BHJ cells involve
INTRODUCTION
■
Polythiophene (PT) and its derivatives may constitute the
most well-studied group of conjugated polymers, in close
competition with polyanilines. PTs are electron-rich p-type
alteration of the nanoscale phase separation due to heat cycling
or diffusion.15 Inspired by the structure of these cells, we
semiconductors and can be readily doped by oxidation to
sought to design a composite PT material that could have
similarly sized conjugated polymer domains but in which the
other phase would be composed of a crystalline lattice rather
increase their conductivity.1 PTs find commercial applications
in organic electronics where flexible conductors or semi-
conductors are desirable, with a well-known commercial
application in antistatic coatings2 and a number of emerging
than hard-to-control precipitated fullerene domains. The size
uses in photovoltaics,3 electrochemical sensors or indica-
and shape of the organic domains would therefore be fixed by
tors,4−6 and electrochromic devices.7,8 One of the best-known
the lattice void structure, determined by the choice of the
lattice material. The final structure of the resulting composite
uses of PTs is in bulk-heterojunction (BHJ) photovoltaic
devices, in which a solution containing both a PT and an
would be enforced by the lattice structure, with polymer
domains separated by rigid lattice features rather than relatively
electron acceptor is deposited (usually by spin coating) onto a
weak van der Waals forces present in organic blends. For
substrate. The resulting thin film can exhibit nanoscale phase
separation of the PT and electron acceptor, effectively
instance, a typical fullerene (PCBM) used in BHJ blends
becomes unstable above 155 °C,16 while crystalline materials
constituting a nanostructured material, which can separate
excitons into dissociated charge carriers.9−11 This nanoscale
structure allows PT blends with an electron acceptor [most
often fullerenes such as phenyl-C61-butyric acid methyl ester
(PCBM), with many others reported12] to serve as the core of
efficient thin-film solar cells, especially poly(3-alkylthiophene)-
s. This self-assembling “soft matter” BHJ architecture has led
based on ligand−metal interactions are commonly stable at
250 °C even with adsorbed guest materials,17 with much
higher stabilities of metal−organic frameworks (MOFs)
Received: December 14, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.8b03465
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. Radical cation generated by a metal-to-ligand charge-transfer excitation in the MOF/BT complex, initiating oxidative polymerization.
reportedly reaching well above 500 °C.18 The design of
crystalline lattice materials, including optimization for a
particular pore shape or structure, has been extensively studied
and recently reviewed for zeolites and covalent frameworks as
well as MOFs.19−21 A PT-in-lattice material produced by the
formation of PTs in a lattice pore could serve as a starting
point in our search for composites with useful properties.
Other important applications of conjugated polymers,
including electrochromic coatings and organic light-emitting
devices, also require electrical cycling of the polymer. PTs are
often used in both of these applications.8,22 Therefore, the
synthesis of PT-in-lattice materials might enable improvements
in the durability of these devices as well.
MOFs are a natural choice as the host lattice for this sort of
investigation because it is well-known that many MOFs can
adsorb organic molecules into their large, regular pores.23,24
Furthermore, the pores of many MOFs are a few nanometers
in diameter25 and therefore on the order of the desirable
domain size for polymers in heterojunctions, i.e., an order of
magnitude below the diffusion length of an exciton in a
conjugated polymer.14 Inspired by the need to create a stable
nanoscale PT architecture, we sought to make a hybrid
material consisting of conjugated polymers integrated into
MOFs. Our initial experiments in this area involved a two-step
process of filling the MOF pores with monomer followed by
the initiation of oxidative polymerization.
Several MOF/conjugated polymer host−guest structures
have been reported, including key work by Yanai et al.26 and
Uemura et al.27 in the late 2000s, and a flurry of recent
reports,28,29 including poly(3,4-ethylenedioxythiophene)
(PEDOT) formed in a Zn-based MOF,30 PT nanorods
templated by a coordination polymer,31 and a Ti-based
MOF/terthiophene (TT) hybrid with high photoconductiv-
ity.32 Jiao et al.’s work in forming polypyrrole (PPy) in a Ni/
Zn-based MOF using only atmospheric oxygen as an oxidant
was of particular interest because Jiao et al. constructed devices
using their technique using the conductivity of the polymer to
make electrochemical contact between PPy and carbon
nanotubes.33 Several other groups prepared PPy/MOF
composites by a direct reaction between pyrrole and MOFs
in the presence of oxygen. Haldar et al. have also very recently
reported a TT/MOF composite in thin-film form,34 and Qiu et
al. used HKUST-1 as a template for electropolymerization of
aniline.35 Our preliminary MOF/PT hybrid syntheses revealed
that the CuII-based MOF HKUST-1 can induce the oxidative
polymerization of oligothiophenes innately, without the
application of an external oxidant such as ferric chloride or
iodine (Figure 1). Using this method, we prepared polymeric
composites of HKUST-1 with 2,2′-bithiophene (BT), 3,4-
ethylenedioxythiophene (EDOT), and TT. These composites
are referred to in the text as HKUST-1/poly(BT), HKUST-1/
PEDOT, and HKUST-1/poly(TT) respectively.
This finding resembles an older work of Millar et al. that
found that Cu-exchanged zeolites could induce oxidative
polymerization.36 The reaction takes place with equal facility in
air or under an ultrapure nitrogen atmosphere. To our
knowledge, the present work is the first example of a MOF/
conjugated polymer composite preparation that does not rely
upon an external oxidant. This discovery provides a convenient
and useful route to accessing MOF/conjugated polymer
hybrids in bulk and thin-film form. Furthermore, in light of
increasing interest in this type of composite, the discovery that
the reaction takes place as a result of innate oxidation of guest
molecules by MOF metal centers must be taken into account
in efforts to prepare new MOF hybrids, especially if such an
initiation would be undesirable or if the MOF is likely to
collapse or rearrange in response to the formation of reduced
centers.
RESULTS AND DISCUSSION
■
Composite Characterization: Is a Host−Guest Struc-
ture Formed? Our initial approach to composite synthesis
followed a two-step model: first thiophene monomers are
adsorbed into a MOF sample, followed by the application of an
external oxidant to generate the polymer-in-MOF architecture.
After an initial screening of several well-known MOFs, we
settled on the copper(II)/trimesic acid structure HKUST-1.
This material can be readily synthesized using a solvothermal
procedure at moderate temperatures, and thin films can be
prepared using the Eddaoudi group’s convenient spin-coating
procedure.37 Both solvothermal (i.e., bulk) and spin-coated
materials are visibly crystalline to optical and nanoscale
microscopy, making it straightforward to distinguish polymer
deposits from the MOF host material. To our surprise, we
noticed during our first experiments that, upon heating BT-
treated HKUST-1 or letting it stand at ambient temperature
for approximately 1 day, a dramatic color change occurred
without the addition of any other reactants. The HKUST-1
thin film developed a very deep-blue-green color (appearing
nearly black), and the edges of the film displayed a red tint
characteristic of free PT. We inferred that an oxidative
polymerization reaction was occurring under these conditions.
Similarly, bulk samples of activated HKUST-1 suspended in
liquid BT, TT, or EDOT spontaneously reacted to form a
durable green-black composite (Figure 2) over the course of
several days. Surprisingly, this reaction occurred both in air and
under an ultrapure nitrogen atmosphere.
It was relatively easy to establish the presence of PTs in the
resulting composites using vibrational spectroscopy. The
B
DOI: 10.1021/acs.inorgchem.8b03465
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
broadening and a high baseline, likely because of the innate
fluorescence of PTs.
In the IR spectra of the composites (Figure 4), the C−H
out-of-plane bend of PT appears at 875 cm−1, a blue shift of
approximately 50 cm−1 from the corresponding mode in BT.39
The out-of-plane bend is the most intense peak in many
published spectra of oligothiophenes and is intense in our
material. A mode involving the C−H in-plane bend can be
seen distinctly at 1042 cm−1; conveniently, this peak does not
overlap any intense IR peaks in HKUST-1. CC symmetric
stretching modes distinguishable from those due to the host
MOF can be seen at 1588 and 1640 cm−1 as well. The narrow
CC asymmetric stretching mode at 1420 cm−1, seen as a
shoulder on the MOF peak at 1446 cm−1, is identified by
Sauvajol et al. as being due to PT.40 The related peak at 1470
cm−1, particularly characteristic of the extended conjugation in
the PT system, appears near 1480 cm−1 in our samples, similar
to other reports of oxidatively polymerized BT.41,42 The peak
at 1047 cm−1, characteristic of PT.40
Having established the presence of PTs in the prepared
composites, we examined the morphology of the composites
with scanning electron microscopy (SEM) and atomic force
microscopy (AFM) to determine whether the polymers were
true guest materials (inside MOF pores) or simply coated on
the MOF surface. The large facets of crystalline HKUST-1
served as a molecularly flat substrate for AFM,43 on which the
surface polymer could be readily seen. Nanoscale microscopy
of HKUST-1 samples before and after treatment with BT and
TT (Figures 5b,c and 6) showed that poly(BT) and poly(TT)
composites were indistinguishable from pristine HKUST-1
(Figure S9), ruling out mere physisorption of the highly
insoluble PT. The polymer formed by the HKUST-1/
oligothiophene reaction, therefore, must be present inside of
the MOF pores rather than coated on the surface.
In contrast to the unsubstituted oligothiophenes, PEDOT
formed on the surface of the HKUST-1 materials (Figures 5d,e
and 6b). SEM and AFM show that the distinctive triangular
facets of HKUST-1 crystals are still apparent in the PEDOT
composite. In this composite, the HKUST-1 crystals are
covered with a coating of the insoluble, rigid polymer PEDOT
in a core−shell architecture.
In one case, the reaction of excess BT with a ca. 1 μm spin-
cast thin film, the reaction between the MOF and monomer
destroys the parent MOF crystal lattice, leaving behind a
tortuous layer of irregular nanostructures a few tens of
nanometers in width (Figure 5f). We inferred that these
structures were PT aggregates that remained after disintegra-
tion of the MOF structure, similar to those observed by SEM
in other work where MOF composites were deliberately
etched.38
In addition to the nanoscale techniques, composites were
imaged using bright-field microscopy (Figure 2). Microscopic
imaging of the striking blue-green crystals reveals no particles
of physisorbed polymer but only the distinctive triangular
facets of the HKUST-1 MOF. Therefore, the observed spectra
are the product of the materials visible in nanoscale
microscopy (Figures 5d,e and 6) rather than an artifact of a
microscopic-scale polymer/MOF blend.
Composition and Structure of Composites from X-
ray and Physical Techniques. Elemental composition
measurements by energy-dispersive spectroscopy (EDS) and
X-ray photoelectron spectroscopy (XPS) are consistent with
the presence of PTs (Tables 1 and 2). Artifacts of elemental
Figure 2. HKUST-1 showing a dramatic change in color upon heating
with oligothiophene and EDOT monomers, without detectable
changes in the morphology or visible accretions of free conjugated
polymers.
Raman spectrum of PT (substituted and unsubstituted) is
distinctive, containing a characteristic intense CC doublet
often used to identify this type of polymer.38 In the present
work, all Raman spectra of the composites are dominated by
the spectrum of the conjugated polymer (Figure 3), with the
peaks due to the MOF still visible but weak in intensity. This
change in the Raman spectrum (with the MOF peaks effaced)
is similar to that observed by Wang et al.30 for their MOF/
PEDOT composite. The most clearly visible peaks due to PTs
are the CC stretching modes centered around 1500 cm−1
and the in-plane C−H mode at 1048 cm−1. Many features in
the PEDOT composite are similar to those in the PT-based
material, save that the Raman spectrum of PEDOT obviously
lacks the C−H in-plane aromatic modes near 1067 cm−1 and
presents a new C−O stretch at 1120 cm−1. The PEDOT
composite also shows unique IR modes at 1189 and 1058 cm−1
corresponding to C−O−C deformations and a C−S−C
deformation at 890 cm−1, higher than that of plain PT.38 In
all of the Raman spectra, treated materials show peak
C
DOI: 10.1021/acs.inorgchem.8b03465
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. Raman spectra of composites dominated by the PT and PEDOT guest materials, with the distinctive CC stretching doublet clearly
visible in all spectra near 1500 cm−1 and the faint but characteristic C−S−C ring deformation appearing at 700 cm−1. Spectra are not offset except
for the reference spectrum of PT, which is placed above the composite spectra for readability.
Figure 4. IR spectroscopy of HKUST-1 before and after treatment with oligothiophenes and EDOT shows the appearance of key peaks
corresponding to the conjugated polymer after heating.
composition typical for each technique appear in these results.
XPS overestimates C in all samples because of the high surface
specificity of the technique in detecting adventitious C on
surfaces. EDS, by contrast, detects O, Si, and Na from the glass
substrates because of the greater penetration depth of the
technique. Neither technique is quantitative of the bulk unless
the sample is perfectly homogeneous, but because the S
content is a good proxy for PT in the material, both techniques
show consistent trends reflecting an increasing degree of MOF
modification with polymer in the order poly(BT) < PEDOT <
poly(TT), with the HKUST-1/poly(TT) composition con-
sistent with the maximum possible conversion of adsorbed
monomer to polymer. This trend is consistent with the peak
intensities from vibrational spectroscopy (Figures 3 and 4).
In high-resolution 2p S XPS of the poly(BT) composite, the
2p S binding energy is 164.2 eV, consistent with literature
reports for PT (Figure 7).36 In the film derived from
decomposition of the poly(BT)/HKUST-1 composite, a
second, lower-binding-energy peak appears at 161.9 eV. A
similar peak is reported in XPS of a vacuum-deposited metallic
copper layer on a PT thin film after heating and is attributed to
a sulfide-like state.44 This result leads us to infer that the
etched material contains Cu in a reduced state interacting with
S atoms in poly(BT).
A SEM survey of the etched HKUST-1/poly(BT) composite
film turned up a defect that resembles a smiley face (Figure
S2). EDS image mapping of this defect shows the Si/Na/O
signature of glass in the exposed background of the “face”, and
Cu and S in the material of the thin film. The correlation
between the distribution of S and Cu is consistent with our
identification of the remaining material as PT that retains a
reduced Cu species as a byproduct of MOF decomposition.
D
DOI: 10.1021/acs.inorgchem.8b03465
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
completely filled composite. The primary LMCT band appears
at 460 nm in the unmodified composite, in addition to a lower-
energy peak due to the d−d transition in the distorted
octahedral environment of the CuII dimer.45,46 The red shift of
these transitions is related to the proportion of guest polymer
within the HKUST-1 pores. This red shift is minimal in the
PEDOT composite compared to the unsubstituted thiophene
derivatives. On the other hand, free PEDOT, with a broader,
red-shifted absorption centered at 550 nm,38 is not visible in
the spectrum of the PEDOT composite. Therefore, some
degree of internal modification in HKUST-1/PEDOT,
accounting for the small observed red shift, is consistent with
the observed absorption spectrum.
As previously noted, HKUST-1/poly(BT) and HKUST-1/
poly(TT) show no evidence by SEM and AFM of physisorbed
PT on the surface, while PEDOT is visibly coated on HKUST-
1/PEDOT (Figures 5 and 6). However, Raman spectra of
these three composites are comparable in the intensity of the
conjugated polymer bands, with the EDOT composite actually
having the weakest overall signal of the three samples. While
not conclusive, the lower intensity of the PEDOT Raman peak
bolsters our hypothesis that HKUST-1/PEDOT consists of a
PEDOT coating on the HKUST-1 crystal rather than a
completely filled composite.
Figure 5. SEM imaging of thin-film HKUST-1 samples: (a) pristine
HKUST-1 thin film; (b) HKUST-1/poly(BT) composite; (c)
HKUST-1/poly(TT) composite; (d) HKUST-1/PEDOT composite.
(e) High-magnification imaging of part d, showing visible PEDOT
features on a crystal facet. (f) PT features remaining after etching of
the HKUST-1/poly(BT) composite.
Composite surface area and volume analysis by N2 sorption
data fit to a Brunauer−Emmett−Teller (BET) isotherm gives
results consistent with the proposed structures of the
composites (Table 3). The pore volume and surface area
values for HKUST-1 are a reasonable match for the original
and subsequent literature reports for this material.47
Composites of both poly(BT) and PEDOT retain the majority
of their free volume. The large remaining free volume is
unsurprising in the case of HKUST-1/poly(BT), which shows
no visible external polymer in nanoscale microscopy and a
relatively weak PT signature in vibrational spectroscopy.
However, PEDOT is plainly detected in micrographs and IR
spectra of HKUST-1/PEDOT yet fills less than half of the
HKUST-1 pore volume. This observation is consistent with a
“core−shell” or coated morphology in HKUST-1/PEDOT,
where the interior of the MOF host is not fully modified. The
pore-size distribution of the composites (Figures S23 and S24)
reinforces this interpretation. HKUST-1/poly(BT) retains
(partially occluded) the pore structure of the parent
HKUST-1, which is known to be trimodal in pore size;
HKUST-1/PEDOT is heavily blocked (likely because of the
visible surface coating) with only small pores available. In
HKUST-1/poly(TT), nearly the entire free volume of the
HKUST-1 structure is filled.
Figure 6. Single facet of a HKUST-1/poly(BT) composite crystal (a)
flat to AFM, with an root-mean-square (RMS) roughness of 0.28 nm
and a maximum excursion of 2.9 nm on an individual facet. The edge
of the facet is seen at the top left. (b) HKUST-1/PEDOT composite
displaying a rough morphology typical of a surface layer of insoluble
conjugated polymer, with an RMS roughness of 5 nm.
EDS element mapping supports the assignment of the
nanostructures from Figure 5 as PEDOT-coated HKUST-1;
features of the film are high in O and S relative to the
background by EDS (Figure S7). This heavy deposition of
PEDOT makes it difficult to easily determine whether the
nanostructures are merely coated with polymer, are filled with
polymer, or are some combination of the two possibilities.
However, on the basis of the UV−vis spectrum of the PEDOT
composite (Figure 8), we believe that the HKUST-1/PEDOT
material is PEDOT coated on HKUST with incomplete or
partial modification of the pore interiors rather than a
Thermogravimetric analysis/residual gas analysis (TGA/
RGA) measurements support the conclusion of a higher degree
of polymer inclusion for the HKUST-1/PEDOT and HKUST-
1/poly(TT) composites compared to HKUST-1/poly(BT).
All composites display a weight loss at 120 °C (Figure 9),
which corresponds to the release of water as shown by the
Table 1. Predicted Mass Percent Compositions of Materials versus Observed Values by XPS
element
HKUST-1 (theor)
HKUST-1/20% PT (theor)
HKUST-1 (obs)
HKUST-1/poly(BT)
C
O
S
36
32
40
27
7
46
28
45
27
4
Cu
32
27
25
24
E
DOI: 10.1021/acs.inorgchem.8b03465
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 2. MOF and Composite Element Percent Compositions Measured by EDS
element
HKUST-1
HKUST-1/poly(BT)
HKUST-1/PEDOT
HKUST-1/poly(TT)
C
O
S
37
31
26
29
1
27
28
3
22
31
7
Cu
25
24
22
17
Figure 7. S 2p XPS of the HKUST-1/poly(BT) composite (red) and the same composite etched after 24 h of reaction time (black), which match
the literature values for PT S 2p binding energies.
Table 3. BET N2 Adsorption Characterization of HKUST-1
and Composites
a
b
sample
HKUST-1
HKUST-1/poly(BT)
HKUST-1/PEDOT
SBET (m2/g)
Vp (cm3/g)
1124 45
900 36
619 24
0.553
0.488
0.335
0.077
HKUST-1/poly(TT)
154
7
a
b
BET surface area. Single-point pore volume at 760 mmHg.
compared to that of pure HKUST-1, but a clear distinction
between two different thermal decomposition steps around
350 °C was not observed. RGA of the MOF decomposition
step clearly shows the release of CO2 and CO gas (Figures
S25−S28), but monomers and polymer decomposition
products were not observed, likely because of the condensation
of larger molecules before the heated capillary transfer line to
the RGA is reached.
The Raman spectrum of the extracted HKUST-1/poly(BT)
composite retains the poly(BT) signature from the as-prepared
composite (Figure S20). This result indicates that while the
degree of modification of HKUST-1 by poly(BT) is lower than
that of the more reactive monomers, the formed composite
does contain polymer rather than mere dimers or trimers,
which would be soluble in toluene.
Matrix-assisted laser desorption/ionization time-of-flight
(MALDI-TOF) analysis of composites also confirms the
presence of higher thiophene oligomers in the composite,
although the observed ions only give a lower bound on the
material’s degree of polymerization because of the extreme
difficulty of dissociating individual PT chains from the bulk
(Figures S10−S13). The largest oligomer ion observed in any
Figure 8. Reflectance UV spectra of spin-cast HKUST-1 thin films
treated with oligothiophene monomers showing progressive red-
shifting of both HKUST-1 absorptions, in contrast with the EDOT-
treated film, which shows little perturbation. Dashed vertical lines
indicate the absorption maxima of the pristine film. Spectra are offset
in the y axis for clarity.
TGA/RGA data (Figures S25−S28). Weight loss at this
temperature in HKUST-1 has been reported in previous work
and is typically far greater than the weight loss observed in our
composites.48,49 The HKUST-1/PEDOT and HKUST-1/
poly(TT) composites clearly show a thermal decomposition
step before the MOF decomposition observed near 350 °C.
The HKUST-1/poly(BT) composition displays a slight
reduction of the onset MOF decomposition temperature
F
DOI: 10.1021/acs.inorgchem.8b03465
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
In the C 1s XPS spectrum of the etched film, the peak
corresponding to carboxylate C is completely missing, and the
main C sp2 peak shifted, consistent with a film comprised
primarily of PT in which none of the original MOF survives.
This result is surprising considering that remnants of the
distinctive triangular structure of HKUST-1 survive in the
treated thin film after the MOF is etched away (Figure 5f), and
these structures are seen by XPS (Figure 11) and EDS imaging
(Figure S1) to contain Cu. The most straightforward
interpretation of these results is that Cu ions remain in the
PT as a dopant after destruction of the MOF structure.
We attribute this surprising decomposition of the MOF to
reduction of the CuII centers by a reaction with monomer. CuI
has a relatively small affinity for carboxylate ligands,50 as is
shown by a straightforward ligand-field argument. While
HKUST-1 has been found to stably contain a high fraction
of CuI centers,51 if a critical fraction of CuII units are reduced
in the relatively small crystals that comprise the HKUST-1 thin
film, the strain induced by the CuI site52 could decompose the
structure. Uemura et al. reported a similar decomposition
caused by pyrrole.27 The Cu remaining in the etched film is
largely in the CuI form based on the energy of the Cu LMM
peak (Figure 11), an observation consistent with the CuII
centers in the MOF acting as the oxidizing agent in
polymerization.
Figure 9. TGA curves of HKUST-1 (blue), HKUST-1/PEDOT
(orange), HKUST-1/poly(TT) (dark red), and HKUST-1/poly(BT)
(green) from 30 to 600 °C with a ramp rate of 5 °C/min. HKUST-1/
PEDOT and HKUST-1/poly(TT) show visible features near 280 °C
distinct from the general MOF decomposition at 350 °C.
composite is an n = 15 peak (five linked TT units) in HKUST-
1/poly(TT) composite. These chain lengths are comparable to
those of other MOF/PT composites reported recently, such as
Haldar’s composite displaying a n = 12 peak (four linked TT
units) prepared using molecular iodine as the external
oxidant.34
Experimental Insights into the Reaction Mechanism:
XPS and Effects of Varied Reaction Conditions. As
previously noted, under particularly extreme conditions
(heating at 100 °C for 24 h in the presence of a large excess
of BT), a thin film of HKUST-1 is etched by the reaction. This
process yields a layer of PT nanostructures still bound to the
oxide surface. Traces of the original triangular silhouettes of
the MOF surface features can still be seen in the SEM image of
the residual PT film after etching (Figure 5). C 1s XPS
confirms that the MOF structure is eliminated because no
carboxylate from the trimesate linker remains (Figure 10).
CuI may be directly formed by the reaction between the
MOF and guest molecules or formed as a result of air oxidation
of metallic Cu produced by the reaction. Because CuI can
disproportionate into Cu metal and CuII, we cannot make an
inference about the form of Cu directly produced in this
reaction from the data, but the simplest interpretation of the
XPS data is that CuI is produced and remains metastable in the
composite matrix on the order of days. Traces of this effect are
seen in the Cu 2p XPS spectrum of the bulk material, where
the CuI peak is more intense after treatment with BT (Figures
12 and 13). Other than this small but detectable enhancement
in the CuI peak, the XPS spectrum of BT-modified bulk
solvothermal HKUST-1 shows little qualitative difference from
that of the pretreatment MOF.
In light of Hu et al.’s relevant preparation of a MOF/PPy
hybrid using O oxidation and because of a possible role of O in
our own system due to the potential air oxidation of CuI
centers produced in the MOF, we tested the effects of
excluding O from the HKUST-1 system during the reaction.
HKUST-1/poly(BT) and poly(TT) composites form as
readily in a well-purged ultrapure nitrogen environment as in
air, without signs of decomposition of the underlying MOF.
Optical micrographs of HKUST-1/poly(BT) prepared under
nitrogen show that the material is still crystalline (Figure S18).
The Raman spectrum of HKUST-1/poly(BT) prepared in
oxygen-free conditions contains a strong PT CC stretching
doublet, while the powder X-ray diffraction (PXRD) patterns
of composites prepared both in air and under an inert
atmosphere show retention of the HKUST-1 peak structure
(Figure 14).
This finding is important because as noted above an
extremely large excess of BT can reductively etch a thin film of
HKUST-1, leaving behind Cu and PT. We attribute the
survival of the framework in bulk versus thin film form to the
greater crystallinity of the bulk, solvothermally synthesized
material. As discussed earlier, Uemura et al. found that pyrrole
decomposed HKUST-1 under an inert atmosphere.27 In this
case versus pyrrole, the durability of HKUST-1 reacting with
Figure 10. HKUST-1 thin film reacted with a large excess of BT at
high temperature etched by the reaction. This process forms a layer of
PT nanostructures that show no traces of carboxylate C in the C 1s
XPS spectrum.
G
DOI: 10.1021/acs.inorgchem.8b03465
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 11. (a) Cu 2p XPS of the BT-treated etched film showing that all remaining Cu in the thin film is in the Cu0 or CuI state, while the
unmodified HKUST-1 film is primarily in the CuII state. (b) Cu LMM peak in the etched film appearing at 917.4 eV. This value yields an Auger
parameter of 1849.7 eV for the Cu 2p3/2 peak, consistent with CuI in the sample.
oxidation steps required to cross-link BT or TT compared to
pyrrole.
Etching of the poly(TT) composite prepared under nitrogen
with 1 M HCl, followed by repeated centrifugation and
resuspension of the pellet in acetone, yielded isolable PT
constituting 13% of the starting HKUST-1 mass (Figure S14).
IR and Raman spectroscopy of the isolated material was a good
match for bona fide PT prepared by ferric chloride oxidation of
thiophene (Figure 15), although the lower ratio of the
terminal-to-backbone C−H out-of-plane bending intensities
in the composite material indicates a lower degree of
polymerization, and the blue shift in the CC stretching
mode near 1350 cm−1 suggests that poly(TT) produced in the
composite has a lower doping level than PT prepared with
Figure 12. XPS of the Cu 2p region of bulk HKUST-1 before and
FeIII.53 Clearly, O is not required for the polymerization
after treatment with BT showing enhancement of the CuI peak after
reaction to occur, although O may react with CuI centers in the
lattice after polymerization is initiated.
reaction.
Raman spectroscopy and PXRD of composites made from a
MIL-100(Al) bulk solvothermal material, which has a high
band gap and a relatively inaccessible conduction band,54
the arene monomer is probably due to the oligothiophenes’
lower nucleophilicity, as well as the smaller number of
reveal partial decomposition of the MOF but no detectable
H
DOI: 10.1021/acs.inorgchem.8b03465
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 13. Fitting of the Cu 2p3/2 peak in the HKUST-1/poly(BT) composite (top) and unmodified HKUST-1 (bottom) revealing that
postreaction HKUST-1 contains more than double the CuI peaks of the pristine MOF.
formation of the polymer (Figure S29). This result is also
consistent with the hypothesis of the redox-active HKUST-1
serving as the oxidant in the polymerization because MIL-
100(Al) cannot serve in this role.
insight into the likely first step of the polymerization
mechanism.
The excited state produced by the transition generates a
radical cation centered on the oligothiophene monomer
(Figure 18), a reactive intermediate that will initiate polymer-
ization similarly to well-known processes initiated by oxidizing
agents such as persulfates or ferric chloride.34,56
Computational Insights into the Reaction Mecha-
nism. Polymer doping dramatically affects the absorption
spectrum of HKUST-1 (Figure 8). We attribute the shift in the
primary visible absorption peak near 400 nm to a direct
The putative initiation step is consistent with the observed
trend in monomer reactivity, in which the more electron-rich
monomers TT and EDOT react to a greater extent relative to
BT. In the more electron-rich monomer ligands, the energy of
the transition is relatively low and more easily accessed
thermally. Supposing that radical-cation generation is the rate-
limiting step, which is easily argued by Hammond’s postulate,
the rate of polymerization initiation should be directly
determined by the energy of the transition in Figure 17.
The magnitude of changes with the chain length in the
vertical excitation energies in the present study (decreasing by
0.98 eV from thiophene to BT) shows a reasonable similarity
to similar changes in the vertical ionization energies (IEs) of
the monomers, with a decrease of 1.3 eV in IEs from thiophene
to BT (Figure 19). Camarada et al. reported similar IE
differences between thiophene and BT performed at a
comparable level of theory, which ranged from 1.3 eV in the
gas phase to 0.7 eV using a polarized continuum model
(PCM) of acetonitrile.57
interaction of the guest polymer with the CuII unit in the
2
center of the HKUST-1 paddlewheel structure. This
interaction has precedent with small-molecule ligands in
HKUST-1, and it is known that guest molecules in the
HKUST-1 pore complex at the Cu2 site55 perturb the MOF’s
absorption spectrum. Even in unmodified HKUST-1, the most
intense excitation is a ligand-to-metal charge-transfer (LMCT)
band occupying a state built from the Cu 3d orbitals.45 It is
reasonable that the presence of an electron-rich guest perturbs
this excitation, shifting it entirely out of the visible spectrum in
the TT composite. This perturbation is seen in the composites
(Figure 8) with the red shifts of both HKUST-1 excitations
increasing with the oligothiophene chain length but relatively
little red shift from an external coating of PEDOT. Time-
dependent density functional theory (TD-DFT) calculations of
excitation energies between monomers and a HKUST-1 model
system (Figure 16) are also consistent with this interpretation.
For instance, an IR-accessible (ca. 1000 nm) excitation
moves the electron density from both BT and TT ligands onto
the Cu center, effectively constituting an example of the
proposed LMCT band (Figure 17). This effect gives critical
Oxidation of the growing polymer chain is the rate-limiting
step in the polymerization reaction,58 and (as seen above) we
expect the polymer oxidation potential as a function of the
chain length to closely follow the change in the polymer
I
DOI: 10.1021/acs.inorgchem.8b03465
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 14. (a) PXRD pattern of the HKUST/BT composites prepared both under nitrogen and in air showing retention of the HKUST-1
structure. (b) Raman spectrum of the HKUST-1/BT composites showing the characteristic CC modes of PT.
ionization potential with the chain length. The ionization
potential of a growing PEDOT chain is always substantially
lower than that of a comparable PT chain (Figure 19). This
result explains the formation of the coated structure in the case
of PEDOT but the filled composite in the case of
unsubstituted oligothiophenes. From Figure 19, we expect
the rate of polymerization of PEDOT to be particularly rapid,
always outstripping that of unsubstituted thiophenes. In a
hypothetical limiting case where polymer growth is extremely
rapid compared to the rate of diffusion into the MOF, the
polymer will be deposited exclusively on the crystal surface,
similar to reactions reported with solid (nonporous) organic
oxidizing agents.59 This simple analysis implies that, in general,
monomers comprised of easily oxidized single arene units such
as EDOT will tend to form surface coatings of polymer rather
than host−guest structures in redox-active MOFs. The analysis
also implies that it may be possible to switch a monomer from
surface-initiated polymerization to interior polymerization by
absorbing the monomer into the MOF at a low temperature,
followed by a temperature jump to induce polymerization, if
the energy of adsorption of the monomer on the MOF is
significantly lower than the barrier to the excitation
represented in Figure 17. Zybaylo et al.60 report a 18 kcal/
mol binding energy for pyridine on HKUST-1 based on a fit of
the quartz crystal microbalance sorption data, which is likely an
upper bound for small arenes because the pyridine N atom will
coordinate strongly to open Cu sites. This value is substantially
lower than our ca. 30 kcal/mol estimated transition energy
(Figure 17) for the initiation of EDOT polymerization,
implying that such control may be possible even in the case
of EDOT. We will explore initiation methods, including
temperature effects on the composite structure, in more detail
in a future study.
CONCLUSIONS
■
The redox-active MOF HKUST-1 initiates the oxidative
polymerization of thiophene monomers, including BT, TT,
and EDOT. This effect can be exploited to yield conjugated
polymers entrapped within MOF hosts. The reaction occurs
more readily with more electron-rich monomers. The non-
redox-active Al-based MOF MIL-100(Al) does not initiate
polymerization despite its Lewis acidity. Electronic transition
energies predicted by TD-DFT calculations to yield radical-
cation excited states on oligothiophene ligands are good
predictors of the observed reactivity series of oligothiophenes.
Depending on the polymer ionization potential, the reaction
J
DOI: 10.1021/acs.inorgchem.8b03465
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 16. Our model system for the HKUST-1/monomer system
consisting of a single paddlewheel element built from benzoate ligands
on a CuII center, complexed to an oligothiophene ligand (shown:
2
thiophene).
Figure 15. IR (a) and Raman (b) spectra of poly(TT) isolated from
the HKUST-1 composite prepared under nitrogen that match well
with the spectra of PT prepared by ferric chloride oxidation of
thiophene. The ratio of peaks (starred) at 695 and 785 cm−1,
representing out-of-plane bends of the terminal and backbone C−H
bonds, respectively, indicates that poly(TT) has a lower degree of
polymerization than the material made with traditional oxidative
polymerization.
Figure 17. (Left) Energy diagram showing the TD-DFT-determined
vertical excitation energies corresponding to charge transfer between
the oligothiophene highest occupied molecular orbital and Cu lowest
unoccupied molecular orbital with the expected dx −y character.
Energies are reported relative to the respective ground state. (Right)
Difference density plots for the respective electronic transitions
(green, negative electron density change; purple, positive electron
density change). We attribute the initiation of polymerization by
HKUST-1 to the excited state populated by this transition.
2
2
may result in PT-filled MOF or a PT-coated MOF. Both
composite types are novel nanomaterial architectures and have
potential applications in organic electronics.
oligothiophene complexes were predicted with Tamm−Dancoff
approximation TD-DFT calculations at the B3LYP/def2-SVP level
of theory using PBE-optimized geometries.69 Visualization of selected
electronic excitations was achieved through the use of TD-DFT
difference densities to identify those transitions with the thiophene
highest occupied molecular orbital based LMCT character.
Spectroscopy and Microscopy. XPS. XPS measurements were
taken at the Center for Engineering and Computing, University of
South Carolina, using a Kratos Axis Ultra DLD instrument equipped
with a monochromated Al Kα source and hemispherical analyzer. On
indium−tin oxide and glass substrates, the C 1s line was used as a
binding energy reference and set to a value of 284.6 eV. Plotting was
performed with the SciDAVis software package.
EXPERIMENTAL SECTION
■
Computational Methods. All calculations were carried out using
Orca 4.0.1.2 for Linux.61 GaussView 5.062 was used for visualization of
orbitals and electron/spin densities. The MOF was simulated with a
simple Cu2 paddlewheel complex formed with four benzoate groups
built in the molecular editor Avogadro 1.2.0.63 The appropriate
thiophene-based molecule was placed on the Cu−Cu bond axis,
bound through its S atom. Geometry optimizations were carried out
at the PBE/def2-SVP level of theory64,65 with Grimmes dispersion
correction (D3BJ).66,67 To correct for basis set superposition error, all
calculations employed the geometrical counterpoise correction.68 For
each optimized structure, the assignment of an energetic minimum
was confirmed through numerical frequency calculations (i.e., no
imaginary frequencies). The electronic spectra of the paddlewheel
PXRD. PXRD measurements were taken at the Department of
Chemistry and Biochemistry, University of South Carolina, facility on
K
DOI: 10.1021/acs.inorgchem.8b03465
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 18. Map of unpaired spin densities in (a) the triplet ground state and (b) the excited state populated by the low-lying LMCT transition of
the oligothiophene/paddlewheel complex showing that the transition confers radical-cation character on the TT ligand, providing an initiator for
oxidative polymerization.
spectrophotometer equipped with the manufacturer’s standard single-
source reflectance probe.
Mass Spectrometry. MALDI spectra were acquired at the Mass
Spectrometry Center, Department of Chemistry and Biochemistry,
University of South Carolina, core facility on a Bruker Ultraflex II
MALDI-TOF/TOF instrument.
TGA/RGA. TGA/RGA measurements were carried out on a
PerkinElmer TGA 8000 analyzer located in an Ar-filled glovebox
and attached to a SRS 300 RGA analyzer via a heated capillary
transfer line. Measurements were carried out with an initial 1 min
equilibration at 30 °C, followed by heating to 600 °C at a ramp rate of
5 °C/min. Ar was used as the carrier gas, and the heated transfer line
was held at a temperature of 130 °C.
N2 Physisorption Analysis. N2 physisorption isotherms were
collected on a Micromeritics Tristar 3020 analyzer at 77 K. Samples
were degassed under a dynamic vacuum of <30 mTorr for 12 h at 110
°C prior to analysis.
ASSOCIATED CONTENT
* Supporting Information
■
S
Figure 19. Plot of IEs of growing PT (circles) and PEDOT (squares)
The Supporting Information is available free of charge on the
polymer chains demonstrating that oxidation of a PEDOT chain
ACS Publications website at DOI: 10.1021/acs.inorg-
remains more favorable than that of an unsubstituted PT chain by ca.
chem.8b03465.
25 kcal/mol at any degree of polymerization. This result rationalizes
the faster growth rate of PEDOT films, which leads to the coated
composite morphology.
Full synthetic procedures, additional SEM, AFM, and
optical images, additional EDS maps and spectra, N2
sorption data, TGA/RGA data, H NMR, TLC images
1
of etched composites, and MALDI characterization of
materials (PDF)
Output files for DFT calculations (ZIP)
Output files for DFT calculations (ZIP)
a Rigaku Ultima IV PXRD system (Cu K radiation) with a diffracted-
beam graphite monochromator.
AFM. All AFM measurements were performed on a Dulcinea/
Cervantes atomic force microscope from Nanotec Electronica using a
silicon tip and cantilever operated in a noncontact tapping mode with
an approximately 70 kHz resonance frequency. The freely available
software package WSxM was used for visualization and analysis of the
images.70
SEM. SEM images were acquired and EDX spectra/element
mapping was performed on an Hitachi Cold-Field-Emission Series
8200 microscope at the Applied Research Center facility (Aiken, SC).
A 9 kV accelerating voltage was used for the typical images shown.
IR and NMR Spectroscopy. IR spectra were acquired on a Nicolet
380 spectrometer with a Smart Orbit diamond ATR attachment. A
total of 64 scans were summed at a 4 cm−1 resolution to produce a
typical spectrum. Routine 1H and 13C NMR of monomers were taken
using a Anasazi EFT 60 MHz FT-NMR spectrometer referenced to
tetramethylsilane at 0 ppm.
Raman Spectroscopy and Microscopy. Raman spectra were
acquired on a Renishaw InVia Raman microscope equipped with four
independent laser lines (488, 514, 633, and 785 nm). After an initial
testing to maximize the signal, the 488 nm laser was used for all
samples shown, typically at 1% maximum power and a 1 s exposure
time. Selected spectra were repeated using the 785 nm laser to check
for artifacts. Spectra shown are the sum of five individual scans, save
for isolated PTs, which are reported as single scans.
AUTHOR INFORMATION
Corresponding Author
*E-mail: nicholasm@usca.edu. Phone: +1-803-641-3409.
ORCID
Nicholas Marshall: 0000-0001-8048-0857
Scott Crittenden: 0000-0001-8982-0184
Anthony B. Thompson: 0000-0003-2843-9675
Gerard T. Rowe: 0000-0002-7935-8034
■
Author Contributions
This manuscript version has been approved by all coauthors.
N.M. initiated the project and wrote the manuscript except as
noted below. W.J. performed all synthetic work, IR, and UV−
vis spectroscopy and wrote the experimental sections of the
manuscript pertaining to the synthesis of materials. J.F.
obtained and validated all Raman spectra and optical
micrographs. S.C. proposed SPM characterization techniques
and obtained and validated AFM measurements. A.B.T.
proposed and performed BET measurements and analyzed
data. P.A.W. proposed and performed TGA and MS
Absorbance Spectroscopy. Absorbance spectra of MOF compo-
sites were acquired using an SI Photonics model 440 UV−vis
L
DOI: 10.1021/acs.inorgchem.8b03465
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(12) Zhang, G.; Zhao, J.; Chow, P. C. Y.; Jiang, K.; Zhang, J.; Zhu,
Z.; Zhang, J.; Huang, F.; Yan, H. Nonfullerene Acceptor Molecules for
Bulk Heterojunction Organic Solar Cells. Chem. Rev. 2018, 118,
3447−3507.
(13) Zhang, J.; Zhu, L.; Wei, Z. Toward Over 15% Power
Conversion Efficiency for Organic Solar Cells: Current Status and
Perspectives. Small Methods 2017, 1, 1700258.
measurements and analyzed data. G.R. helped to propose and
devise the experiments, performed DFT calculations and
analyzed the results, and helped in the interpretation of the
spectral data.
Notes
The authors declare no competing financial interest.
(14) Tamai, Y.; Ohkita, H.; Benten, H.; Ito, S. Exciton Diffusion in
Conjugated Polymers: From Fundamental Understanding to
Improvement in Photovoltaic Conversion Efficiency. J. Phys. Chem.
Lett. 2015, 6, 3417−3428.
(15) Rafique, S.; Abdullah, S. M.; Sulaiman, K.; Iwamoto, M.
Fundamentals of Bulk Heterojunction Organic Solar Cells: An
Overview of Stability/Degradation Issues and Strategies for Improve-
ment. Renewable Sustainable Energy Rev. 2018, 84, 43−53.
(16) Zhao, J.; Swinnen, A.; Van Assche, G.; Manca, J.; Vanderzande,
D.; Mele, B. V. Phase Diagram of P3HT/PCBM Blends and Its
Implication for the Stability of Morphology. J. Phys. Chem. B 2009,
113, 1587−1591.
(17) Wang, R.; Wunder, S. L. Thermal Stability of Octadecylsilane
Monolayers on Silica: Curvature and Free Volume Effects. J. Phys.
Chem. B 2001, 105, 173−181.
(18) Cavka, J. H.; Jakobsen, S.; Olsbye, U.; Guillou, N.; Lamberti,
C.; Bordiga, S.; Lillerud, K. P. A New Zirconium Inorganic Building
Brick Forming Metal Organic Frameworks with Exceptional Stability.
J. Am. Chem. Soc. 2008, 130, 13850−13851.
(19) Farha, O. K.; Hupp, J. T. Rational Design, Synthesis,
Purification, and Activation of MetalOrganic Framework Materials.
Acc. Chem. Res. 2010, 43, 1166−1175.
(20) Li, Y.; Yu, J. New Stories of Zeolite Structures: Their
Descriptions, Determinations, Predictions, and Evaluations. Chem.
Rev. 2014, 114, 7268−7316.
(21) Mercado, R.; Fu, R.-S.; Yakutovich, A. V.; Talirz, L.; Haranczyk,
M.; Smit, B. In Silico Design of 2D and 3D Covalent Organic
Frameworks for Methane Storage Applications. Chem. Mater. 2018,
30, 5069−5086.
(22) Armstrong, N. R.; Wang, W.; Alloway, D. M.; Placencia, D.;
Ratcliff, E.; Brumbach, M. Organic/Organic’ Heterojunctions:
Organic Light Emitting Diodes and Organic Photovoltaic Devices.
Macromol. Rapid Commun. 2009, 30, 717−731.
(23) Zhou, H.-C.; Long, J. R.; Yaghi, O. M. Introduction to Metal−
Organic Frameworks. Chem. Rev. 2012, 112, 673−674.
(24) Long, J. R.; Yaghi, O. M. The Pervasive Chemistry of Metal−
Organic Frameworks. Chem. Soc. Rev. 2009, 38, 1213−1214.
(25) Chen, Z.; Gu, Z.-G.; Fu, W.-Q.; Wang, F.; Zhang, J. A Confined
Fabrication of Perovskite Quantum Dots in Oriented MOF Thin
Film. ACS Appl. Mater. Interfaces 2016, 8, 28737−28742.
(26) Yanai, N.; Uemura, T.; Ohba, M.; Kadowaki, Y.; Maesato, M.;
Takenaka, M.; Nishitsuji, S.; Hasegawa, H.; Kitagawa, S. Fabrication
of Two-Dimensional Polymer Arrays: Template Synthesis of
Polypyrrole between Redox-Active Coordination Nanoslits. Angew.
Chem., Int. Ed. 2008, 47, 9883−9886.
(27) Uemura, T.; Kadowaki, Y.; Yanai, N.; Kitagawa, S. Template
Synthesis of Porous Polypyrrole in 3D Coordination Nanochannels.
Chem. Mater. 2009, 21, 4096−4098.
(28) MacLean, M. W. A.; Kitao, T.; Suga, T.; Mizuno, M.; Seki, S.;
Uemura, T.; Kitagawa, S. Unraveling Inter- and Intrachain Electronics
in Polythiophene Assemblies Mediated by Coordination Nanospaces.
Angew. Chem., Int. Ed. 2016, 55, 708−713.
(29) Le Ouay, B.; Boudot, M.; Kitao, T.; Yanagida, T.; Kitagawa, S.;
Uemura, T. Nanostructuration of PEDOT in Porous Coordination
Polymers for Tunable Porosity and Conductivity. J. Am. Chem. Soc.
2016, 138, 10088−10091.
(30) Wang, T.; Farajollahi, M.; Henke, S.; Zhu, T.; Bajpe, S. R.; Sun,
S.; Barnard, J. S.; Lee, J. S.; Madden, J. D. W.; Cheetham, A. K.;
Smoukov, S. K. Functional Conductive Nanomaterials via Polymer-
isation in Nano-Channels: PEDOT in a MOF. Mater. Horiz. 2017, 4,
64−71.
ACKNOWLEDGMENTS
■
This project was supported by funding provided by the College
of Sciences and Engineering Summer Scholars Institute, USC
Aiken. The Summer Scholars Institute is supported by the
Workforce Opportunities in Regional Careers (WORC) grant,
ADP, and the South Carolina IDeA Networks of Biomedical
Research (SC INBRE). Funding for this project provided
through USC Aiken INBRE was supported by grants from the
National Center for Research Resources (Grant 5 P20
RR016461) and the National Institute of General Medical
Sciences (Grant 8 P20 GM103499) from the National
Institutes of Health. Support was also provided through a
RISE grant from the Office of the Vice President for Research,
University of South Carolina. The authors thank Prof. Chad
Leverette (USC Aiken) for Raman microscope time and
training and Dr. Steve Xiao (SRNL) for N2 physisorption
apparatus access. The authors also thank core facilities experts
Drs. Gregory Morrison, Mike Walla, and Stavros Karakalos
(University of South Carolina) and Patrick Woodell (Applied
Research Center, Savannah River National Laboratory) for
their skilled work in collecting data on the materials described
in this report.
REFERENCES
■
(1) Soos, Z. G. Theory of π-Molecular Charge-Transfer Crystals.
Annu. Rev. Phys. Chem. 1974, 25, 121−153.
(2) Reuter, K.; Kirchmeyer, S.; Elschner, A. Handbook of Thiophene-
Based Materials; Wiley-Blackwell, 2009; pp 549−576.
(3) Hedley, G. J.; Ruseckas, A.; Samuel, I. D. W. Light Harvesting
for Organic Photovoltaics. Chem. Rev. 2017, 117, 796−837.
(4) Marshall, N.; Sontag, S. K.; Locklin, J. In Functional Polymer
Films; Knoll, W., Advincula, R. C., Eds.; Wiley-VCH Verlag GmbH &
Co. KGaA, 2011; pp 319−377.
(5) Kaloni, T. P.; Giesbrecht, P. K.; Schreckenbach, G.; Freund, M.
S. Polythiophene: From Fundamental Perspectives to Applications.
Chem. Mater. 2017, 29, 10248−10283.
(6) Krikstolaityte, V.; Kuliesius, J.; Ramanaviciene, A.; Mikoliunaite,
L.; Kausaite-Minkstimiene, A.; Oztekin, Y.; Ramanavicius, A.
Enzymatic Polymerization of Polythiophene by Immobilized Glucose
Oxidase. Polymer 2014, 55, 1613−1620.
̈
(7) Dyer, A. L.; Osterholm, A. M.; Shen, D. E.; Johnson, K. E.;
Reynolds, J. R. Electrochromic Materials and Devices; Wiley-Blackwell,
2015; pp 113−184.
(8) Neo, W. T.; Ye, Q.; Chua, S.-J.; Xu, J. Conjugated Polymer-
Based Electrochromics: Materials, Device Fabrication and Application
Prospects. J. Mater. Chem. C 2016, 4, 7364−7376.
(9) Rita Narayan, M.; Singh, J. Study of the Mechanism and Rate of
Exciton Dissociation at the Donor-Acceptor Interface in Bulk-
Heterojunction Organic Solar Cells. J. Appl. Phys. 2013, 114,
No. 073510.
(10) Whaley, K. B.; Kocherzhenko, A. A.; Nitzan, A. Coherent and
Diffusive Time Scales for Exciton Dissociation in Bulk Heterojunction
Photovoltaic Cells. J. Phys. Chem. C 2014, 118, 27235−27244.
(11) Chen, W.; Xu, T.; He, F.; Wang, W.; Wang, C.; Strzalka, J.; Liu,
Y.; Wen, J.; Miller, D. J.; Chen, J.; Hong, K.; Yu, L.; Darling, S. B.
Hierarchical Nanomorphologies Promote Exciton Dissociation in
Polymer/Fullerene Bulk Heterojunction Solar Cells. Nano Lett. 2011,
11, 3707−3713.
M
DOI: 10.1021/acs.inorgchem.8b03465
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(31) Kitao, T.; MacLean, M. W. A.; Le Ouay, B.; Sasaki, Y.;
Tsujimoto, M.; Kitagawa, S.; Uemura, T. Preparation of Poly-
thiophene Microrods with Ordered Chain Alignment Using Nano-
porous Coordination Template. Polym. Chem. 2017, 8, 5077−5081.
(32) Wang, S.; Kitao, T.; Guillou, N.; Wahiduzzaman, M.;
Martineau-Corcos, C.; Nouar, F.; Tissot, A.; Binet, L.; Ramsahye,
N.; Devautour-Vinot, S.; Kitagawa, S.; Seki, S.; Tsutsui, Y.; Briois, V.;
Steunou, N.; Maurin, G.; Uemura, T.; Serre, C. A Phase Trans-
formable Ultrastable Titanium-Carboxylate Framework for Photo-
conduction. Nat. Commun. 2018, 9, 1660.
Physisorbent Materials. Angew. Chem., Int. Ed. 2015, 54, 14372−
14377.
(49) Mu, B.; Walton, K. S. Thermal Analysis and Heat Capacity
Study of MetalOrganic Frameworks. J. Phys. Chem. C 2011, 115,
22748−22754.
(50) Sevryugina, Y.; Rogachev, A. Y.; Petrukhina, M. A. The First
Hexanuclear Copper(I) Carboxylate: X-Ray Crystal Structure and
Reactivity in Solution and Gas-Phase Reactions. Inorg. Chem. 2007,
46, 7870−7879.
(51) Szanyi, J.; Daturi, M.; Clet, G.; Baer, D. R.; Peden, C. H. F.
Well-Studied Cu−BTC Still Serves Surprises: Evidence for Facile
Cu2+/Cu+ Interchange. Phys. Chem. Chem. Phys. 2012, 14, 4383−
4390.
(52) Todaro, M.; Alessi, A.; Sciortino, L.; Agnello, S.; Cannas, M.;
Gelardi, F. M.; Buscarino, G. Investigation by Raman Spectroscopy of
the Decomposition Process of HKUST-1 upon Exposure to Air. J.
Spectrosc. 2016, 2016, 1−7.
(53) Akimoto, M.; Furukawa, Y.; Takeuchi, H.; Harada, I.; Soma, Y.;
Soma, M. Correlation between vibrational spectra and electrical
conductivity of polythiophene. Synth. Met. 1986, 15, 353−360.
(54) Ling, S.; Slater, B. Unusually Large Band Gap Changes in
Breathing MetalOrganic Framework Materials. J. Phys. Chem. C 2015,
119, 16667−16677.
(33) Jiao, Y.; Chen, G.; Chen, D.; Pei, J.; Hu, Y. Bimetal−Organic
Framework Assisted Polymerization of Pyrrole Involving Air Oxidant
to Prepare Composite Electrodes for Portable Energy Storage. J.
Mater. Chem. A 2017, 5, 23744−23752.
(34) Haldar, R.; Sen, B.; Hurrle, S.; Kitao, T.; Sankhla, R.; Ku
Welle, A.; Heissler, S.; Brenner-Weiß, G.; Thissen, P.; Uemura, T.;
̈
hl, B.;
̈
Gliemann, H.; Barner-Kowollik, C.; Woll, C. Oxidative Polymer-
ization of Terthiophene and a Substituted Thiophene Monomer in
Metal-Organic Framework Thin Films. Eur. Polym. J. 2018, 109, 162−
168.
(35) Lu, C.; Ben, T.; Xu, S.; Qiu, S. Electrochemical Synthesis of a
Microporous Conductive Polymer Based on a Metal−Organic
Framework Thin Film. Angew. Chem., Int. Ed. 2014, 53, 6454−6458.
(36) Millar, G. J.; McCann, G. F.; Hobbis, C. M.; Bowmaker, G. A.;
Cooney, R. P. Spectroscopic investigation of the polymerisation of
pyrrole and thiophene within zeolite channels. J. Chem. Soc., Faraday
Trans. 1994, 90, 2579−2584.
(37) Chernikova, V.; Shekhah, O.; Eddaoudi, M. Advanced
Fabrication Method for the Preparation of MOF Thin Films:
Liquid-Phase Epitaxy Approach Meets Spin Coating Method. ACS
Appl. Mater. Interfaces 2016, 8, 20459−20464.
(55) Prestipino, C.; Regli, L.; Vitillo, J. G.; Bonino, F.; Damin, A.;
Lamberti, C.; Zecchina, A.; Solari, P. L.; Kongshaug, K. O.; Bordiga, S.
Local Structure of Framework Cu(II) in HKUST-1 Metallorganic
Framework: Spectroscopic Characterization upon Activation and
Interaction with Adsorbates. Chem. Mater. 2006, 18, 1337−1346.
̈
(56) Niemi, V. M.; Knuuttila, P.; Osterholm, J. E.; Korvola, J.
Polymerization of 3-Alkylthiophenes with FeCl3. Polymer 1992, 33,
1559−1562.
(38) Wang, M.; Jamal, R.; Wang, Y.; Yang, L.; Liu, F.; Abdiryim, T.
Functionalization of Graphene Oxide and Its Composite with
Poly(3,4-Ethylenedioxythiophene) as Electrode Material for Super-
capacitors. Nanoscale Res. Lett. 2015, 10, 370.
(39) Matharu, A. S.; Cowling, S. J.; Wright, G. Laterally Fluorinated
Liquid Crystals Containing the 2,2′-bithiophene Moiety. Liq. Cryst.
2007, 34, 489−506.
(40) Sauvajol, J. L.; Poussigue, G.; Benoit, C.; Lere-Porte, J. P.;
Chorro, C. Sample Dependence of Raman Spectrum on Poly-
thiophene Films. Synth. Met. 1991, 41, 1237−1242.
(57) Camarada, M. B.; Jaque, P.; Díaz, F. R.; del Valle, M. A.
Oxidation Potential of Thiophene Oligomers: Theoretical and
Experimental Approach. J. Polym. Sci., Part B: Polym. Phys. 2011,
49, 1723−1733.
(58) Higashimura, H.; Kobayashi, S. Encyclopedia of Polymer Science
and Technology; American Cancer Society, 2016; pp 1−37.
(59) Sato, K.; Masaki, H.; Arayasu, M.; Oaki, Y.; Imai, H.
Conductive Polymer Nanosheets Generated from the Crystal Surface
of an Organic Oxidant. ChemPlusChem 2017, 82, 177−180.
(60) Zybaylo, O.; Shekhah, O.; Wang, H.; Tafipolsky, M.; Schmid,
̈
(41) Zhu, L.; Niu, Y.; Cao, Y.; Lei, A.; Ai, X.; Yang, H. N-Type
Redox Behaviors of Polybithiophene and Its Implications for Anodic
Li and Na Storage Materials. Electrochim. Acta 2012, 78, 27−31.
(42) Zhang, Z.; Wang, F.; Chen, F.; Shi, G. Preparation of
Polythiophene Coated Gold Nanoparticles. Mater. Lett. 2006, 60,
1039−1042.
R.; Johannsmann, D.; Woll, C. A Novel Method to Measure Diffusion
Coefficients in Porous Metal−Organic Frameworks. Phys. Chem.
Chem. Phys. 2010, 12, 8092−8098.
(61) Neese, F. The ORCA Program System. WIREs Comput. Mol.
Sci. 2012, 2, 73−78.
(62) Dennington, R.; Kieth, T. A.; Milliam, J. M. GaussView;
Semichem, Inc., 2009.
̂
̀
̈
(43) Shoaee, M.; Agger, J. R.; Anderson, M. W.; Attfield, M. P.
Crystal Form, Defects and Growth of the Metal Organic Framework
HKUST-1 Revealed by Atomic Force Microscopy. CrystEngComm
2008, 10, 646−648.
(63) Hanwell, M. D.; Curtis, D. E.; Lonie, D. C.; Vandermeersch, T.;
Zurek, E.; Hutchison, G. R. Avogadro: An Advanced Semantic
Chemical Editor, Visualization, and Analysis Platform. J. Cheminf.
2012, 4, 17.
(64) Weigend, F.; Ahlrichs, R. Balanced Basis Sets of Split Valence,
Triple Zeta Valence and Quadruple Zeta Valence Quality for H to Rn:
Design and Assessment of Accuracy. Phys. Chem. Chem. Phys. 2005, 7,
3297−3305.
(44) Lachkar, A.; Selmani, A.; Sacher, E.; Leclerc, M.; Mokhliss, R.
Metallization of Polythiophenes I. Interaction of Vapor-Deposited Cu,
Ag and Au with Poly(3-Hexylthiophene) (P3HT). Synth. Met. 1994,
66, 209−215.
̈
(45) Gu, Z.-G.; Heinke, L.; Woll, C.; Neumann, T.; Wenzel, W.; Li,
Q.; Fink, K.; Gordan, O. D.; Zahn, D. R. T. Experimental and
Theoretical Investigations of the Electronic Band Structure of Metal-
Organic Frameworks of HKUST-1 Type. Appl. Phys. Lett. 2015, 107,
183301.
(46) Borfecchia, E.; Maurelli, S.; Gianolio, D.; Groppo, E.; Chiesa,
M.; Bonino, F.; Lamberti, C. Insights into Adsorption of NH3 on
HKUST-1 Metal−Organic Framework: A Multitechnique Approach.
J. Phys. Chem. C 2012, 116, 19839−19850.
(47) Chui, S. S.-Y.; Lo, S. M.-F.; Charmant, J. P. H.; Orpen, A. G.;
Williams, I. D. A Chemically Functionalizable Nanoporous Material
[Cu3(TMA)2(H2O)3]N. Science 1999, 283, 1148−1150.
(48) Kumar, A.; Madden, D. G.; Lusi, M.; Chen, K.-J.; Daniels, E. A.;
Curtin, T.; Perry, J. J.; Zaworotko, M. J. Direct Air Capture of CO2 by
(65) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient
Approximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865−3868.
(66) Grimme, S.; Ehrlich, S.; Goerigk, L. Effect of the Damping
Function in Dispersion Corrected Density Functional Theory. J.
Comput. Chem. 2011, 32, 1456−1465.
(67) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A Consistent and
Accurate Ab Initio Parametrization of Density Functional Dispersion
Correction (DFT-D) for the 94 Elements H-Pu. J. Chem. Phys. 2010,
132, 154104.
(68) Kruse, H.; Grimme, S. A Geometrical Correction for the Inter-
and Intra-Molecular Basis Set Superposition Error in Hartree-Fock
and Density Functional Theory Calculations for Large Systems. J.
Chem. Phys. 2012, 136, 154101.
N
DOI: 10.1021/acs.inorgchem.8b03465
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(69) Becke, A. D. Density-functional Thermochemistry. III. The
Role of Exact Exchange. J. Chem. Phys. 1993, 98, 5648−5652.
́
́
(70) Horcas, I.; Fernandez, R.; Gomez-Rodríguez, J. M.; Colchero,
́
J.; Gomez-Herrero, J.; Baro, A. M. WSXM: A Software for Scanning
Probe Microscopy and a Tool for Nanotechnology. Rev. Sci. Instrum.
2007, 78, No. 013705.
O
DOI: 10.1021/acs.inorgchem.8b03465
Inorg. Chem. XXXX, XXX, XXX−XXX