Contents lists available atScienceDirect
European Polymer Journal
journal homepage:www.elsevier.com/locate/europoljMacromolecular Nanotechnology
Tuning charge transport across junctions of ferrocene-containing polymer
brushes on ITO by controlling the brush thickness and the tether lengths
Lu Gan
a, C.S. Suchand Sangeeth
a, Li Yuan
a, Dominik Ja
ńczewski
b,c, Jing Song
b,
Christian A. Nijhuis
a,d,e,⁎aDepartment of Chemistry, National University of Singapore, 3 Science Drive, Singapore 117543, Singapore
bInstitute of Materials Research and Engineering A*STAR (Agency for Science, Technology and Research), 2 Fusionpolis Way, Innovis, #08-03, Singapore 138634,
Singapore
cLaboratory of Technological Processes, Faculty of Chemistry, Warsaw University of Technology, Noakowskiego 3, 00-664 Warsaw, Poland dCentre for Advanced 2D Materials and Graphene Research Centre, National University of Singapore, 6 Science Drive 2, Singapore 117546, Singapore eNUSNNI-Nanocore, National University of Singapore, Singapore 117411, Singapore
A R T I C L E I N F O
Keywords: Ferrocene Polymer brushes Redox-active Charge transport Conduction mechanism Molecular electronicsA B S T R A C T
This paper describes the electrical characteristics of junctions of ferrocene (Fc)-containing polymer brushes (PBs) grafted on indium tin oxide (ITO) bottom electrodes contacted with eutectic gallium indium (EGaIn) top elec-trodes. We studied the charge transport phenomena across these junctions as a function of the PB thickness as well as the length of the tether (i.e., an alkyl chain side linker) between the Fc units and the polymethyl me-thacrylate backbone. Junctions with PBs with a thickness of < 20 nm do not rectify while those junctions with PBs with a thickness of > 20 nm do rectify. Temperature dependent charge transport measurements revealed that the charge transport mechanism is dominated by space charge limited conduction (SCLC).
1. Introduction
One of the major goals of molecular electronics is to understand, control, and design, electronic circuits using molecules as the active components[1–3], such as self-assembled monolayers (SAMs)[4–9], conjugated molecular wires[10–12], or single (bio)molecules or oli-gomers[13–18]. Polymer brushes (PBs) are interesting alternatives as one end of the chain can be covalently anchored to a substrate and the chemical structure as well as the PBs thickness can be precisely con-trolled[19–23]. Redox-active PBs are considered smart materials since their properties can be changed by altering the redox state of the redox-active groups. PBs have been used as smart surfaces in applications such as sensing[24,25], allowing to change the thermo-responsivity[26], wettability[27], swelling[28], catalytic activity[29]and bioactivity [30,31]. For example, Vancso’s group fabricated chemically modified electrodes (gold surface) with covalently tethered poly(ferrocenylsi-lane) chains [25]. These PBs were 10 nm thick, robust, and showed reversible redox behaviour in organic solvents. In the reduced form, these PBs showed a higher adhesion with hydrophobic atomic force microscopy (AFM) tips than in the oxidized form. In addition, the PBs worked as electrochemical sensors to detect ascorbic acid. Gallei’s group synthesized well-defined ferrocene (Fc)-containing PBs on
polystyrene particles[28]. The redox behaviour of the Fc-containing brush shell was investigated by dynamic light scattering and cyclic voltammetry (CV). An obvious swelling after oxidation of the brush shell was observed which almost doubled the hydrodynamic volume of the shell and therefore they are promising for applications in separa-tion, or sensing. Our group reported Fc-containing polymethyl metha-crylate PBs with different lengths of the tether connecting the Fc units to the PB backbone[22]. These redox-responsive PBs were character-ized by CV, their thicknesses in air were measured by AFM and ellip-sometry, and their mechanical properties were evaluated by AFM-based nanoindentation. The results indicated that the stiffness and packing structure of the PBs were affected by the length of the alkyl chain te-thers (-(CH2)n- with n = 1, 4, and 9). Despite these advances, PBs have been only occasionally used in nano-electronics and examples include the use of PBs as hole/electron transport layers or dielectric elements in organic electronics[32–38], memory devices[39,40], or charge storage [41]. It is important to improve our understanding of the mechanisms for charge transport across PBs.
Charge transport properties across polymers are mostly studied in so-called polymer thinfilms where a polymer is typically spin coated on an electrode with the polymer chain aligned in a random manner with respect to the surface. In principle, PBs are immobilized on a bottom
http://dx.doi.org/10.1016/j.eurpolymj.2017.10.009
Received 6 September 2017; Accepted 5 October 2017
⁎Corresponding author at: Department of Chemistry, National University of Singapore, 3 Science Drive, Singapore 117543, Singapore.
E-mail address:chmnca@nus.edu.sg(C.A. Nijhuis).
Available online 10 October 2017
0014-3057/ © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/). T
electrode and can be contacted with a top-electrode to yield electrode-PBs-electrode junctions, which makes it possible to study charge transport phenomena across polymers aligned in parallel with respect to the surface normal of the electrodes. A few examples of such junc-tions have been reported before. Huck and Friend et al. have presented a method to obtain order in molecular semiconductors through PBs generated by surface-initiated polymerization[37], their brushes have shown characteristics of high mobility for hole transport. They also have shown that relative to spin-coatedfilms, polymer chains in PBs have a high degree of stretching resulting in a three orders of magni-tude increase in the current density[38]. However systematic charge transport studies on PBs with varying length and as a function of temperature are lacking.
Charge transport measurements across metal-thin polymer film-metal junctions have been conducted by many groups[42–46]. These studies show that the mechanism of charge transport in these organic semiconducting devices can be divided mainly into 2 classes: interface dominated injection limited conduction (ILC) and bulk controlled space-charge limited conduction (SCLC)[47,48]. ILC dominates when the barrier between the organic layer and the electrodes limits the charge injection rate into the organic layer. Upon applying an electric field, charge carriers either hop or tunnel across the metal-polymer interface which has a large barrier height (i.e., resistive contacts). In devices where one (or both) of the electrodes forms (form) an Ohmic contact with the organic layer, the conduction is limited by the prop-erties of the organic layer. In devices with low mobility, the injected charge carriers will create a space-charge region near the interfaces, if the number of injected carriers is higher than the number of intrinsic carriers presented in the sample [49]. Once the space-charges have been formed, the electrostatic potential of these space charges will limit the injection of additional injection of carriers. By varying the dopant concentration and thickness of the polymerfilm[43,50], the transport mechanism in organic semiconductor devices can be tuned.
Depending on the mobility and concentration of charge traps, the SCLC mechanism can be classified as either free (TFSCLC) or trap-filling (TFLSCLC)[49,51]. In a perfect insulator without intrinsic car-riers or traps[52–53], TFSCLC model obeys the Mott-Gurney transport expression[55]: = J ε μV d 9 8 SCLC r 2 3 (1)
whereεris the dielectric constant of the material, µ is the mobility, d is
the thickness of thefilm, V is the applied voltage, and J is the current density. Here, if the mobility is independent of the electricfield, the current will vary according to the well-known V2law. If either trapping or the effect of field on the mobility is important, the current varies as Vm, where m > 2. This is known as TFLSCLC[42,45,46,56–58]. In normal cases, trap sites will be present due to intrinsic disorder in polymeric materials, which results from intentionally doping[43,59]or intrinsic doping, or structural defects and impurities[60–63].
In contrast, when charge traps with varying energy are present, the traps will befilled gradually with increasing applied bias. Until all traps have beenfilled, the current will increase faster than V2with increasing bias and the current can be approximated with the modified Mott-Gurney transport expression[55]:
⎜ ⎟ = ⎛ ⎝ + + ⎞⎠ ⎛ ⎝ + ⎞ ⎠ − + + + J q μN l l l l ε ε H V d 2 1 1 1 l v l l l l 1 1 r 0 t 1 2 1 (2) where Nνis the carrier density, Htis the trap density, q is the electron charge,ε0is permittivity of free space. The parameter l is defined as l = Tc/T = Et/kBT, where Etis the characteristic trap energy, T is the temperature, kB is the Boltzman constant, and Tc is a characteristic temperature[42]. The value of Etcan be determined from the slope of the log-log plot of J(V) and is related to the activation energy Eaas given by Eq.(3). Here, Eais the activation energy for hopping which can be obtained from temperature dependent J(V) measurements fol-lowed byfitting of the data to the Eq.(4) [64]. The value of Htcannot be determined directly and is usually treated as afitting parameter.
⎜ ⎟ = ⎛ ⎝ ⎞ ⎠ E E k qH d ε ε V ln 2 a t B t 2 r 0 (3) = ⎛⎝ − ⎞ ⎠ J J e E k T 0 a B (4)
Here, we report a detailed charge transport study of Fc-containing PBs as function of (1) the thickness of the polymer brush (dPB) over the range of 2–50 nm, (2) applied bias (V) of −1.5 V to +1.5 V (or −3.0 V to +3.0 V for thickerfilms), (3) temperature T = 250–340 K, and (4) the length of the tether that links the Fc units to the backbone of the brush (n = 1, 4, and 9;Scheme 1).Scheme 1shows the structure of the redox-active Fc-containing PBs of polyferrocenylmethyl methacrylate (PFMMA, n = 1), polyferrocenylbutyl methacrylate (PFBMA, n = 4), and polyferrocenylnonyl methacrylate (PFNMA, n = 9). All PBs were formed on indium-tin-oxide (ITO) bottom-electrode. We used the
Scheme 1. Schematic illustration of the ITO-PBs//GaOx/EGaIn junction. PFMMA ≡
polyferrocenylmethyl methacrylate, PFBMA ≡ polyferrocenylbutly methacrylate, PFNMA ≡ polyferrocenylnonyl methacrylate. dPBstands for the brush thickness, n indicates the
“EGaIn (eutectic alloy of Ga and In) technique” to contact the PBs be-cause this method results in good electrical contact without causing damage to the organic layer, yields data in statistically large numbers with good yields in working devices, and allows for measurements of J vs. V as a function of temperature T[5]. EGaIn is a liquid-metal with a melting point of 15 °C. It is non-toxic (unlike Hg) and coated with a 0.7 nm thin layer of conductive GaOx[65], which gives this alloy non-Newtonian properties (itflows under shear pressure but behaves as a solid at rest) so it can be shaped into tips or stabilized in microchannels in polydimethylsiloxane (PDMS, a transparent rubber) [66]. These features of GaOx/EGaIn make it possible to form soft electrical contact to soft materials without causing damage to the material of interest surface[65]. Typically, junction areas are 300 µm2. We found that the electrical characteristics of our redox-active brushes were determined by brush thickness but not by the length of the tether. Temperature dependent measurements demonstrated that the mechanism of charge transport is dominated by TFLSCLS for junctions and that the rectifi-cation ratio and direction of rectifirectifi-cation depends on the length of PBs. 2. Experimental procedures
2.1. Chemicals and materials
PFMMA, PFBMA, and PFNMA brushes were generated by surface initiated atom transfer radical polymerization (SI-ATRP), following the same procedures as described in detail elsewhere [22]. Absolute ethanol (AR EtOH, EMSURE), toluene (95%, EMSURE), were used for substrates cleaning. Diluted (with deionized water, 18 MΩ cm) per-chloric acid (HClO4, 75%, Alfa) was used as electrolyte in wet elec-trochemistry. Indium-tin-oxide (ITO, a mixture of In2O3in 90 wt% and SnO2 in 10 wt%, was obtained from Singapore Optics, 101 × 101 × 1.1 mm, layer thickness ∼180 nm, sheet resistance ∼10 Ω sq−1) and silicon wafer (Si/SiO2, 100, p-type, University Wa-fers, USA) were used as substrates.
2.2. Determination of the thickness of PBs
AFM was performed with a NanoWizard III instrument (JPK in-strument AG, Berlin, Germany) to determine the brush thickness by measuring the height difference across scratches, i.e., AFM profilometry (tapping mode in air, standard silicon probes). Ellipsometry (Variable Angle Spectroscopic Ellipsometer, V.A.S.E., M190-1700, Digipol) was used as the supplementary technique to verify the thickness as de-scribed elsewhere[22].
2.3. Cyclic voltammetry (CV)
AUTOLAB PGSTAT302N with NOVA 1.10 software was used to record the cyclic voltammograms in an aqueous (aq.) solution 1.0 M HClO4, between−0.5 V to 1.5 V at a scan rate of 1.00 V s−1[22]. We used a Pt disk as the counter electrode, a Ag/AgCl reference electrode, and the polymer brush attached on ITO substrate was used as the working electrode. The area of the working electrode exposed to the electrolyte was 0.24 cm2.
2.4. EGaIn cone-shape tip junction
A home-built“EGaIn-setup” set-up was used to form the top-con-tacts with the eutectic metal alloy EGaIn (75.5% Ga and 24.5% In by weight, Sigma-Aldrich), see details in Ref.[4]. We used cone-shape tips of EGaIn to fabricate large area junctions (∼300 μm2) and collected statistically large numbers of J(V) data (around 300 J(V) scans) fol-lowing previously reported procedures[65]. We recorded 15–20 J(V) traces (0 V→ +1.5 V → 0 V → −1.5 V → 0 V) with a 0.075 V step size from 14–16 junctions. We calculated the average values of log10|J| for each applied bias to construct the log-average J(V) curves and to
determine the values of log10(R) using previously reported methods [65].
2.5. EGaIn microfluidic-device and temperature dependent measurement The J(V, T) measurements were conducted using devices with the EGaIn top-electrode stabilized in a through-hole in a transparent rubber of PDMS. We collected 10 traces (0 V→ +3.0 V → 0 V → −3.0 V → 0 V) at a 0.15 V step size for each sample. We used a voltage range higher than the usual cone-shape tip junction since the changes in J vs. T were more pronounced at higher applied bias. In all the measure-ments, we biased the GaOx/EGaIn top electrode and grounded the ITO bottom electrode. The J(V, T) data were measured over a range of temperature of 250–340 K at intervals of 10 K using probe station (model Lakeshore CRX-VF) under vacuum (0.1 Pa).
3. Results and discussions
3.1. Characterization of PBs
Recently, we have reported the preparation and characterization of the PBs shown inScheme 1in detail[22]. These PBs were derived from a methacrylate monomer which have Fc units linked via a short –(CH2)n– tether with n = 1 (polyferrocenylmethyl methacrylate, PFMMA), n = 4 (polyferrocenylbutly methacrylate, PFBMA), or n = 9 (polyferrocenylnonyl methacrylate, PFNMA). The PBs were prepared by SI-ATRP on substrates of both ITO substrates and Si/SiO2wafers and their thickness were controlled by adjusting the polymerization time. The AFM morphologies recorded on PBs (supported by Si/SiO2 sub-strates) indicate uniformfilms[22](Fig. S1, Page S2–S3). The surface coverage of the Fc groups,ΓFcin mol cm−2, was estimated by CV using Eq.(5)(Fig. S2, Page S4), where Qtotis the total charge, n is the number of electrons per mole of reaction (here n = 1), F is the Faraday con-stant, and A is the surface area of the electrode exposed to the elec-trolyte solution[67].
= Q nAF
ΓFc tot/ (5)
As we have reported before, the value ofΓFcincreases linearly as a function of polymerization time for the PFMMA brushes and the short tether results in stiff brushes that can stand up in air up to 40 nm tall (determined by AFM profilometry and ellipsometry[22]). The PFBMA and PFNMA brushes, on the other hand, form more disordered struc-tures where the individual PBs chains tend to collapse and/or form interdigitated structures. These effects have been described in detail in reference[22]and here we used PBs that had the same characteristics (confirmed by AFM profilometry, ellipsometry, and CV).
3.2. Electrical characteristics of the junctions
We fabricated junctions of ITO–PBs//GaOx/EGaIn with brush thicknesses of 2–50 nm to investigate their electronic properties as function of the tether length (n = 1, 4, and 9, with n is the number of CH2 units; Scheme 1). We measured and analyzed statistically large numbers of data, following previously reported procedures[65]. Ty-pically, we measured 15–20 J(V) curves per junction and 14–16 junc-tions for each type of PB to yield a total of about 300 J(V) curves per polymer brush. The log10|J| values for each measured bias (one curve is 0 V→ +1.5 V → 0 V → −1.5 V → 0 V, recorded in steps of 0.075 V) were plotted in histograms to which Gaussians werefitted to determine the Gaussian mean of log10|J| and the log-standard deviation. These values were then used to construct the log-average J(V) curves. The Gaussian mean of the values of log10(R) were determined by plotting all values of log10(R) determined from the individual J(V) curves at ± 1.5 V (R (=|J (−1.5 V)|/|J (+1.5 V)|) in histograms to each of which a Gaussian wasfitted. These histograms along with Gaussian fits are shown in Figs. S4–S6 (pages S10–S12). Tables S1–S3 (pages S7–S9)
summarize the statistical results of the junctions, including the number of total junctions, the number of shorts, the number of traces, the yield in non-shorts, and the values of R with their log-standard deviations.
The stability of log10|J| of two representative PFMMA PBs junctions (one thin brush with dPB≈ 8 nm and one thick brush with dPB≈ 27 nm) with the cone-shape tip EGaIn top electrodes were tested over 100 J(V) cycles. For junctions with thin PBs, log10|J| at positive bias (V = +1.5 V) decreased as the scan number increased (Fig. 1A and B), which resulted in a decrease of log10(R) (Fig. 1C). Thus, these junctions were not stable and therefore we could not investigate their J(V) characteristics as a function of temperature. In contrast, junctions with thick PBs, the values of log10|J| at both positive and negative bias were stable and so were the log10(R) values (Fig. 1D–F). Therefore, only junctions with thick PBs were selected for temperature dependent J(V) analysis which are described below. The durability of a thick PBs sample was measured (Fig. S3, Page S5–S6), and the sample was stable over four months.
Fig. 2shows representative log-average J(V) curves and histograms of the rectification ratios log10(R) with Gaussianfits to these histograms recorded from junctions with PFMMA, PFBMA, and PFNMA PBs with two different thicknesses.Fig. 3shows plots of the values of log10(R) and log10|J| at +1.5 V and−1.5 V for the three types of junctions as a function of dPB. The value of J decreases, as expected, with increasing value of dPB. For instance, the value of log10|J| is typically about 3–4 orders of magnitudes lower for dPB≈ 40 nm than for those junctions when dPB≈ 8 nm at |V| = 1.5 V. Interestingly, the decay of the current is steep for junctions with thin PBs film (dPB< 15 nm) but the rate changes abruptly to much a slower decay. In addition, junctions with thin PBs have small values of R (i.e., the J(V) curves are nearly sym-metrical) while those junctions with thick PBs (dPB> 20 nm) rectify currents with values of R of > 100. The direction of rectification changes as a function of dPB: the value of log10(R) is < 0 for junctions with dPB< 15 nm and increases for junctions with 15 nm < dPB< 20 nm until log10(R) saturates around 2.5–3 for junctions with dPB> 20 nm. Interestingly, these trends are similar for all junctions, i.e., clear side chain effects are not visible[22]. These sharp transitions of the values of J and R vs. dPB clearly indicate changes in charge transport mechanism for thin and thick PBs. The effects of the side
chains are discussed in more detail in Section3.5. The small hysteresis is only visible for junctions with very low currents (on the order of pA) which are most likely caused by capacitive currents from the contacts, but may involve small charging effects of the Fc units of the PBs[68]. 3.3. The mechanism of charge transport vs. PB thickness
As described above, rectification of currents was observed at ne-gative bias for junctions with thick PBs. To determine whether this asymmetric shape of the J(V) curves resulted from a Schottky barrier or was caused by injection limited conduction (ILC), we examined the electronic characteristics for a series of PFMMA brushes as a function of dPBat room temperature. In the case of ILC, J depends on the interface barrier and is independent of dPB. We measured the J(V) characteristics of junctions where the EGaIn was stabilized in a through-hole in PDMS with PFMMA as a function of six different values dPBin the range of 22–58 nm.Fig. 4A shows the log-log plots of the values of J in the high bias regime (V =−2.55 to −3.0 V) vs. dPBand that J follows a power law dependence with dPB. The data gives a bestfit to Eq.(1)with the power law relationshipJ∝dk
PB(k < 0) with a slope k of∼−2.7 which
indicates that the transport mechanism is SCLC.
We also analyzed the log-log plots of the J(V) data for each junction (Fig. 4B) to distinguish between TFSCLC and TFLSCLC. All junctions follow a power low dependence in the high bias regime of 1.2 V < |V| < 3.0 V (Fig. S8shows an example of a complete J(V) curve measured at 0 to−3 V). According to Eq.(1), a power law de-pendence ofJ∝Vmwith m > 2 is indicative of the presence of charge
carrier traps. A bestfit of Eq. (1)to our data indicates a power de-pendence with m is 4–5 from which we conclude that the mechanism of charge transport is TFLSCLC with an exponential trap distribution [51–58].
For all three kinds of thin PBs (dPB< 20 nm), we observed that the value of log10|J| decreases exponentially with increasing value of dPB, at both positive and negative bias (Fig. 3B, D, and F). This trend follows the general tunneling equation[6,64](Eq.(6)),
= −
J J e βd
0 (6)
where J0(A cm−2) is the pre-exponential factor and β (Å−1) is the
Fig. 1. (A) 100 J(V) curves of cone-shape tip junction measured with a junction with a thin PBs of 8 nm, (B) log10|J| values measured at +1.5 V and−1.5 V, and (C) log10(R) values as
the function of scan number. (D) 100 J(V) curves of cone-shape tip junction measured with a PBs of 27 nm, (E) log10|J| values measured at +1.5 V and−1.5 V, and (F) log10(R) values as
tunneling decay constant[69].
Fitting the data to Eq. (6) gives an average value of β = 0.57 ± 0.05 Å−1for both positive and negative bias, which is in the same range of aliphatic monolayers[2,3,70,71](seeFig. S7for all details). This value indicates that through-bond tunneling dominates the mechanism of charge transport. Similar long range tunneling phe-nomena have been recently observed in other systems including junc-tions with ferritin[72], molecular wires crafted on surface[11,12], or single oligomers [15,73]. We were not able, however, to record tem-perature dependent charge transport measurements for junctions with thin PBs as these junctions were not stable enough to do so as explained above.
3.4. Temperature dependent charge transport measurements
Many factors can contribute to formation of charge traps including defects in the organicfilms or at the interfaces, or induced by physical (e.g., conformational disorder) and chemical defects (e.g., broken bonds and impurities)[60–62]. Here we believe that the Fc moieties act as the trap site because of the redox activity of Fc. Temperature dependent measurements make it possible to identify the nature of traps in our junctions. To investigate the mechanism of charge transport in more detail and to determine values of Eaand other transport parameters, we carried out temperature dependent J(V) measurements using junctions
with PFMMA, PFBMA, and PFNMA PBs. Table 1 and Figs. S9–S12 summarize all data. Unfortunately, we were only able to obtain stable junctions that could withstand full J(V,T) characterization for junctions with thick PBs with values of dPB≥ 36 nm. We believe that the major reason is that we had to use the conductive ITO surfaces which have a fairly high surface roughness of 3.5 nm measured over 2 × 2μm2[22]. We have shown before that the junction properties of thin organicfilms (in the form of self-assembled monolayers) are highly sensitive to the topography of the bottom-electrode and junctions with rough surfaces are prone to defects and junction failure resulting in shorts[5]. Here, we believe a similar explanation holds on the rough ITO electrode, which causes defects and lowers the stability of the junction. This al-lows us only to measure the properties of thick PBs vs. T reliably. To determine whether the length of tether is important, we measured the temperature dependent J(V) characteristics of three junctions with PFMMA, PFBMA, and PFNMA PBs with the same thickness of 42–43 nm. In addition, we also recorded J(V,T) curves from three junctions with PFMMA brushes with dPBvalues of 36, 42, and 47 nm to examine the effect of dPBon the charge transport characteristics.
Fig. 5A shows the J(V,T) curves on a linear scale of a representative junction with PFMM with a thickness of 47 nm for T = 250–340 K in the bias window of ± 3.0 V. The value of J at negative bias decreases with decreasing T indicating that hopping dominates. The values of J at positive bias dependent weakly on T with a small value of
Fig. 2. Two representative semi-log plots of the log-average J(V) curves recorded from junctions with (A) PFMMA (black squares, dPB= 43 nm; red dots, dPB= 8 nm), (C) PFBMA
(black squares, dPB= 38; red dots, dPB= 5 nm), and (E)
PFNMA (black squares, dPB= 36 nm; red dots, dPB= 8 nm)
PBs. The error bars (-) represent the log-standard deviation. Histograms of R with Gaussianfits (solid lines) to these histograms for junctions with (B) PFMMA, (D) PFBMA, and (F) PFNMA PBs determined at V = ± 1.5 V.
Ea= 0.14 ± 0.04 eV (Fig. S13, page S19).Fig. 5B shows power law plots of the |J|(|V|,T) data at negative bias (1.9 V < |V| < 3.0 V) along withfits to Eq.(2). The slope l changes as function of T and is plotted inFig. 5C. The linear relationship between l and 1/T indicates that traps are present with exponential distribution in energy. A value of Tcof 2185 ± 47 K was determined from the slope of the 1/T vs. l using Eq.(2). This value of Tc, gives a value of Etof 0.19 ± 0.01 eV. Finally, the values of J follows Eq. (4) from which we obtained Ea= 0.45 ± 0.01 eV (Fig. 5D). A similar analysis was performed for the other junctions and the results are summarized in Table 1. The values of Et= 0.19 ± 0.01 eV are lower than the values of Ea= 0.45 ± 0.01 eV, this extra energy of on average of∼0.26 eV is the energy used for hopping between sites and results in carrier trans-port. Interestingly, the trap density Htis close to the density of Fc units (ρFc, estimated from the surface coverage of brushes on ITO surface) determined with CV (Fig. S2). Using Eq.(5)and a value ofεrof 2.8 [74], we found a value of Htof 4.8 × 1018cm−3at−3.0 V. This value is similar to other reported values [42] and about three orders of magnitude lower than Fc density of the PBsρFc(∼1021). As we men-tioned before, normally the trap sites will be present due to the intrinsic disorder in polymeric materials, such as intentional doping, intrinsic doping, or structure defects/impurities. Here, the Fc units can be con-sidered as a type of intrinsic disorder of the polymer brush. Hence, the
above observations demonstrate that Fc units contribute to the trap sites.
3.5. Effect of brush thickness and tether length
Table 1shows that the differences in the transport parameters (Et, Ea and Ht) between junctions with different values of dPB and tether lengths are small. InFig. 6,firstly, we compare the transport parameters obtained from junctions with PFMMA polymer brushes with different values of dPB(circles inFig. 6). The value of Etdecreases with increasing value of dPB, while the value of Eadid not change within error. This observation can be explained by an increase of the disorder in the polymer brush as the chain length increases due to intercalation of neighboring polymer chains. This intercalation reduces the distances between individual Fc units and gives the system more conformational degree of freedom which may lower the values Etand the corresponding values of Tc. Disorder and intercalation also explains the increase inρFc and Ht, which increases with increasing dPB. This observation indicates that these two parameters (ρFcand Ht) have a positive correlation which is consistent with our hypothesis that the Fc groups act as the charge traps in the charge transport process. The Eadetermines the overall charge transport in the polymer brush with temperature. Since the value of Etis about 0.2 eV (on average) and lower than the value of Ea,
Fig. 3. The values log10(R) determined at V = ± 1.5 V and
log10|J| at +1.5 V and−1.5 V vs. dPBfor junctions with
PFMMA (A and B), PFBMA (C and D), and PFNMA (E and F) PBs. Error bars are log-standard deviations and the dashed lines are guides to the eyes.
the variation in the Etwith dPBdoes not affect the Easignificantly. Also at high electricfields of > 0.6 MV/m, most likely field assisted hopping dominates and the activation from temperature is nearly independent of thefilm thickness[75].
The effect of the tether length can be determined by comparing the transport parameters junctions with PFBMA (triangles in Fig. 6), PFNMA (squares inFig. 6), and PFMMA PBs with a similar thickness (42–43 nm).Table 1shows that the stiff PFMMA brush has a slightly larger trap density than the junctions with the other two polymer brushes which may result from the different packing configuration [22]: PFBMA and PFNMA brushes have longer tethers than the PFMMA brushes, which makes them less rigid so that the polymer chains are prone to collapse and/or form intercalated structures increasing the
density of Fc units. With this higher trap density, the corresponding values of Tcand Etare also higher. Also in this series the value of Eais constant because the trapping energy Etis much lower than the acti-vation energy Ea, the variation in the Etwith polymer brush thickness does not affect the Easignificantly.
4. Conclusions
We described temperature dependent charge transport measure-ments of redox-active polymer brushes sandwiched between two elec-trodes as a function of brush thickness (2–50 nm) and the length of the tether (1, 4, or 9 CH2 units) connecting the redox-active site to the polymer backbone. The junctions showed distinct charge transport re-gimes and junctions with thin PBs (dPB< 20 nm) tunneling dominates the mechanism of charge transport while junctions with thick PBs (dPB> 20 nm) hopping dominates via trapfilling space charge limited conduction (TFLSCLC). Temperature dependent measurements allowed us to further understand the essential charge transport parameters (i.e., the characteristic temperature Tc, trapping energy Et, activation energy Ea, and trap density Ht) as function of the PBs structure. The results indicate that the Fc units act as the main trapping site. Although the charge transport mechanism depends strongly on the thickness of the brush, the tether length had no significant influence because Ethas a much lower value than Ea.
The junctions with thick brushes rectify currents, likely because the charge transport mechanism is TFLSCLC and the PBs form a Schottky type of contact and rectify[76–79]. The mechanism of charge transport across thin junctions is dominated by tunneling which indicates charge traps do not dominate charge transport. This observation suggest that for these thin PB junctions almost no charge transfer between the PB and the electrode took place and hence a Schottky barrier is not present. These results indicate that PBs can be used to fabricate junctions whose electrical properties can be controlled via the chemical structure of PBs. The transition from tunneling to hopping has been only observed in few systems till date, all of which were long conjugated molecules[11,12] and for a ferritin (an iron storing protein)[71]. The fact that here such a transition could be observed despite a non-conjugated backbone in-dicates that the close spaced Fc units provide tunneling sites to which and from charge carriers can tunnel efficiently. To investigate these phenomena in more detail, temperature dependent measurements are required, but the thin PBs on ITO were too defective to do so. Based on these results, we believe that PBs are an interesting class of molecules to study charge transport phenomena and are potentially useful to induce electronic function at the nano-scale.
Acknowledgements
Prime Minister’s Office, Singapore under its Medium sized centre program is acknowledged for supporting this research. We also ac-knowledge the Ministry of Education (MOE) for supporting this
Fig. 4. (A) Power law plot of J vs. dPBof junctions with PFMMA brushes in the high bias
regime show. The dashed lines arefits to Eq.(1). (B) Power law plot of J vs. V at negative bias (1.2 V < |V| < 3.0 V) for junctions with PFMMA with different values of dPB. The
dashed lines arefits to Eq.(1).
Table 1
Characteristics of the PBs junctions.
dPBa(nm) Tc(×103K) Et(eV) Ea(eV) Ht(×1018cm−3) ρFc(×1021cm−3)b PFMMA 47 ± 1 2.22 ± 0.05 0.19 ± 0.01 0.45 ± 0.01 4.8 ± 0.01 2.0 ± 0.01 42 ± 1 2.68 ± 0.12 0.23 ± 0.01 0.46 ± 0.01 3.9 ± 0.01 1.9 ± 0.01 36 ± 1 3.55 ± 0.19 0.31 ± 0.02 0.45 ± 0.01 3.3 ± 0.01 1.5 ± 0.01 PFBMA 43 ± 1 2.39 ± 0.05 0.21 ± 0.01 0.43 ± 0.01 4.2 ± 0.01 1.9 ± 0.01 PFNMA 42 ± 1 2.43 ± 0.76 0.21 ± 0.01 0.43 ± 0.01 4.2 ± 0.01 1.9 ± 0.01 ad
PBwas determined by AFM profilometry (on Si/SiO2substrates). bρ
research under award No. MOE2015-T2-1-050. The authors are also grateful to the Agency for Science, Technology, and Research (A∗STAR) for providing research support.
Appendix A. Supplementary material
Characterisation of the polymer brushes by atomic force micro-scopy, ellipsometry, and cyclic voltammetry; electrical characterization
of the junctions and temperature dependent charge transport data. Supplementary data associated with this article can be found, in the online version, athttp://dx.doi.org/10.1016/j.eurpolymj.2017.10.009.
References
[1] R.L. McCreery, A.J. Bergren, Progress with molecular electronic junctions: meeting experimental challenges in design and fabrication, Adv. Mater. 21 (2009) 4303–4322.
Fig. 5. Thermally activated charge transport across ITO–PBs//GaOx/EGaIn junctions with
PFMMA PBs of a thickness of 47 nm. (A) J(V) plots as a function of T. Inset shows the values of |J| measured at−3.0 and +3.0 V as a function of T. (B) Power law plot of same |J|(|V|) data at negative bias (1.9 V < |V| < 3.0 V). Colored dash lines indicatefits to Eq.(2). (C) The value of l plotted against 1/T. The grey dashed line is afit to l = Tc/T. Error bars represent the error of the
fits shown in panel (B). (D) The value of ln|J| at |V| = 3.0 V plotted as the function of 1/T, grey dashed line indicates afit to Eq.(4).
Fig. 6. Correlation plots of (A) dPBvs. Ht, (B) dPBvs.ρFc(C) dPBvs.Ea,
(D) dPBvs. Et. Circle (●) indicates the PFMMA brush, triangle (▴)
indicates the PFBMA brush, and the square (■) indicates the PFNMA brush. Note that the symbols of 42-nm PFMMA and 42-nm PFNMA for ρFcvalues in panel (B) are overlap.
[2] A. Vilan, D. Aswal, D. Cahen, Large-area, ensemble molecular electronics: motiva-tion and challenges, Chem. Rev. 5 (2017) 4248–4286.
[3] A. Vilan, D. Cahen, Chemical modification of semiconductor surfaces for molecular electronics, Chem. Rev. 5 (2017) 4624–4666.
[4] N. Nerngchamnong, L. Yuan, D.C. Qi, J. Li, D. Thompson, C.A. Nijhuis, The role of van der Waals forces in the performance of molecular diodes, Nat. Nanotechnol. 8 (2013) 113–118.
[5] D. Thompson, C.A. Nijhuis, Even the odd numbers help: failure modes of SAM-based tunnel junctions probed via odd-even effects revealed in synchrotrons and super-computers, Acc. Chem. Res. 49 (2016) 2061–2069.
[6] M.M. Thuo, W.F. Reus, C.A. Nijhuis, J.R. Barber, C. Kim, M.D. Schulz, G.M. Whitesides, Odd-even effects in charge transport across self-assembled monolayers, J. Am. Chem. Soc. 133 (2011) 2962–2975.
[7] M. Carlotti, A. Kovalchuk, T. Wachter, X. Qiu, M. Zharnikov, R.C. Chiechi, Conformation-driven quantum interference effects mediated by through-space conjugation in self-assembled monolayers, Nat. Commun. 7 (2016) 13904–13911. [8] G. Wang, T.-W. Kim, T. Lee, Electrical transport characteristics through molecular
layers, J. Mater. Chem. 21 (2011) 18117–18136.
[9] D.K. Aswal, S. Lenfant, D. Guerin, J.V. Yakhmi, D. Vuillaume, Fowler-Nordheim tunnelling and electrically stressed breakdown of 3-mercaptopropyltrimethox-ysilane self-assembled monolayers, Nanotechnology 16 (2005) 3064–3068. [10] A. Nitzan, Electron transmission through molecules and molecular interfaces, Annu.
Rev. Phys. Chem. 52 (2001) 681–750.
[11] H. Yan, A.J. Bergren, R.L. McCreery, M.L.D. Rocca, P. Martin, P. Lafarge, J.C. Lacroix, Activationless charge transport across 4.5 to 22 nm in molecular electronic junctions, Proc. Natl. Acad. Sci. 110 (2013) 5326–5330.
[12] S.H. Choi, B. Kim, C.D. Frisbie, Electrical resistance of long conjugated molecular wires, Science 320 (2008) 1482–1486.
[13] S.V. Aradhya, L. Venkataraman, Single-molecule junctions beyond electronic transport, Nat. Nanotechnol. 8 (2013) 399–410.
[14] L. Venkataraman, J.E. Klare, C. Nuckolls, M.S. Hybertsen, M.L. Steigerwald, Dependence of single-molecule junction conductance on molecular conformation, Nature 442 (2006) 904–907.
[15] L. Lafferentz, F. Ample, H. Yu, S. Hecht, C. Joachim, L. Grill, Conductance of a single conjugated polymer as a continuous function of its length, Science 323 (2009) 1193–1197.
[16] I. Diez-Perez, J. Hihath, T. Hines, Z.-S. Wang, G. Zhou, K. Mullen, N. Tao, Controlling single-molecule conductance through lateral coupling of [π] orbitals, Nat. Nanotechnol. 6 (2011) 226–231.
[17] N. Amdursky, D. Marchak, L. Sepunaru, I. Pecht, M. Sheves, D. Cahen, Electronic transport via proteins, Adv. Mater. 26 (2014) 7142–7161.
[18] N. Amdursky, I. Pecht, M. Sheves, D. Cahen, Electron transport via cytochrome C on Si–H surfaces: roles of Fe and heme, J. Am. Chem. Soc. 135 (2013) 6300–6306. [19] S. Minko, Responsive polymer brushes, J. Macromol. Sci Polym. Rev. 46 (2006)
397–420.
[20] S.T. Milner, Plymer brushes, Science 251 (1991) 905–915.
[21] X. Sui, S. Zapotoczny, E.M. Benetti, P. Schӧn, G.J. Vancso, Characterization and molecular engineering of surface-grafted polymer brushes across the length scales by atomic force microscopy, J. Mater. Chem. 20 (2010) 4981–4993.
[22] L. Gan, J. Song, S. Guo, D. Jańczewski, C.A. Nijhuis, Side chain effects in the packing structure and stiffness of redox-responsive ferrocene-containing polymer brushes, Eur. Polym. J. 83 (2016) 517–528.
[23] P.M. Mendes, Stimuli-responsive surfaces for bio-applications, Chem. Soc. Rev. 37 (2008) 2512–2529.
[24] A. Akhoury, L. Bromberg, T.A. Hatton, Interplay of electron hopping and bounded diffusion during charge transport in redox polymer electrodes, J. Phys. Chem. B 117 (2013) 333–342.
[25] X. Sui, X. Feng, J. Song, M.A. Hempenius, G.J. Vancso, Electrochemical sensing by surface-immobilized poly(ferrocenylsilane) grafts, J. Mater. Chem. 22 (2012) 11261–11267.
[26] A. Alkan, C. Steinmetz, K. Landfester, F.R. Wurm, Triple-stimuli-responsive ferro-cene-containing PEGs in water and on the surface, ACS Appl. Mater. Interfaces 7 (2015) 26137–26144.
[27] B. Nagel, N. Gajovic-Eichelmann, F.W. Scheller, M. Katterle, Ionic topochemical tuned biosensor interface, Langmuir 26 (2010) 9088–9093.
[28] M. Mazurowski, M. Gallei, J. Li, H. Didzoleit, B. Stühn, M. Rehahn, Redox-re-sponsive polymer brushes grafted from polystyrene nanoparticles by means of surface initiated atom transfer radical polymerization, Macromolecules 45 (2012) 8970–8981.
[29] V. Bui-Thi-Tuyet, G. Trippé-Allard, J. Ghilane, H. Randriamahazaka, Surface and electrochemical properties of polymer brush-based redox poly(ionic liquid), ACS Appl. Mater. Interfaces 8 (2016) 28316–28324.
[30] B. Nagel, A. Warsinke, M. Katterle, Enzyme activity control by responsive redox polymers, Langmuir 23 (2007) 6807–6811.
[31] T.K. Tam, M. Ornatska, M. Pita, S. Minko, E. Katz, Polymer brush-modified elec-trode with switchable and tunable redox activity for bioelectronic applications, J. Phys. Chem. C 112 (2008) 8438–8445.
[32] S.H. Park, H.S. Lee, J.-D. Kim, D.W. Breiby, E. Kim, Y.D. Park, D.Y. Ryu, D.R. Lee, J.H. Cho, A polymer brush organic interlayer improves the overlying pentacene nanostructure and organicfield-effect transistor performance, J. Mater. Chem. 21 (2011) 15580–15586.
[33] K. Park, S.H. Park, E. Kim, J.D. Kim, S.Y. An, H.S. Lim, H.H. Lee, D.H. Kim, D.Y. Ryu, D.R. Lee, J.H. Cho, Polymer brush as a facile dielectric surface treatment for high-performance, stable, soluble acene-based transistors, Chem. Mater. 22 (2010) 5377–5382.
[34] J.C. Pinto, G.L. Whiting, S. Khodabakhsh, L. Torre, A. Rodríguez, R.M. Dalgliesh,
A.M. Higgins, J.W. Andreasen, M.M. Nielsen, M. Geoghegan, W.T.S. Huck, H. Sirringhaus, Organic thinfilm transistors with polymer brush gate dielectrics synthesized by atom transfer radical polymerization, Adv. Funct. Mater. 18 (2008) 36–43.
[35] I.M. Rutenberg, O.A. Scherman, R.H. Grubbs, W. Jiang, E. Garfunkel, Z. Bao, Synthesis of polymer dielectric layers for organic thinfilm transistors via surface-initiated ring-opening metathesis polymerization, J. Am. Chem. Soc. 126 (2004) 4062–4063.
[36] S.H. Kim, M. Jang, H. Yang, J.E. Anthony, C.E. Park, Physicochemically stable polymer-coupled oxide dielectrics for multipurpose organic electronic applications, Adv. Funct. Mater. 21 (2011) 2198–2207.
[37] H.J. Snaith, G.L. Whiting, B. Sun, N.C. Greenham, W.T.S. Huck, R.H. Friend, Self-organization of nanocrystals in polymer brushes. Application in heterojunction photovoltaic diodes, Nano Lett. 5 (2005) 1653–1657.
[38] G.L. Whiting, H.J. Snaith, S. Khodabakhsh, J.W. Andreasen, D.W. Breiby, M.M. Nielsen, N.C. Greenham, R.H. Friend, W.T.S. Huck, Enhancement of charge-transport characteristics in polymericfilms using polymer brushes, Nano Lett. 6 (2006) 573–578.
[39] Y. Wei, D. Gao, L. Li, S. Shang, Memory effect in polymer brushes containing pendant carbazole groups, Polymer 52 (2011) 1385–1390.
[40] Y. Liu, S. Lv, L. Li, S. Shang, Conductivity switching and memory effect in polymer brushes with carbazole pendant moieties, Synth. Met. 162 (2012) 1059–1064. [41] X. Ma, Z. Xie, Z. Liu, X. Liu, T. Cao, Z. Zheng, Polymer brush electrets, Adv. Funct.
Mater. 23 (2013) 3239–3246.
[42] V. Kumar, S.C. Jain, A.K. Kapoor, J. Poortmans, R. Mertens, Trap density in con-ducting organic semiconductors determined from temperature dependence of J-V characteristics, J. Appl. Phys. 94 (2003) 1283–1285.
[43] P. Anjaneyulu, C.S.S. Sangeeth, R. Menon, Space-charge limited conduction in doped polypyrrole devices, J. Appl. Phys. 107 (2010) 0937161–0937165. [44] A.J. Campbell, D.D.C. Bradley, D.G. Lidzey, Space-charge limited conduction with
traps in poly(phenylene vinylene) light emitting diodes, J. Appl. Phys. 82 (1997) 6326–6342.
[45] M. Taunk, S. Chand, Bias and temperature dependent charge transport inflexible polypyrrole devices, J. Appl. Phys. 115 (2014) 0745071–0745074.
[46] S. Diaham, M.-L. Locatelli, Space-charge-limited currents in polyimidefilms, Appl. Phys. Lett. 101 (2012) 2429051–2429054.
[47] A.J. Campbell, M.S. Weaver, D.G. Lidzey, D.D.C. Bradley, Bulk limited conduction in electroluminescent polymer devices, J. Appl. Phys. 84 (1998) 6737–6746. [48] A. Rose, Space-charge-limited currents in solids, Phys. Rev. 97 (1955) 1538–1544. [49] K. Weiser, Current injection in solids, in: Murray A. Lampert, Peter Mark (Eds.),
Electrical Science Series, vol. xiv, Academic Press, New York, 1970, pp. 354 (illus.). [50] R. Valaski, S. Ayoub, L. Micaroni, I.A. Hümmelgen, Influence of film thickness on
charge transport of electrodeposited polypyrrole thinfilms, Thin Solid Films 415 (2002) 206–210.
[51] J.M. Montero, J. Bisquert, G. Garcia-Belmonte, E.M. Barea, H.J. Bolink, Trap-lim-ited mobility in space-charge limTrap-lim-ited current in organic layers, Org. Electron. 10 (2009) 305–312.
[52] J. Shao, G.T. Wright, Characteristics of the space-charge-limited dielectric diode at very high frequencies, Solid-State Electron. 3 (1961) 291–303.
[53] P. Mark, W. Helfrich, Space-charge-limited currents in organic crystals, J. Appl. Phys. 33 (1962) 205–215.
[54] K.C. Kao, W. Hwang, Electrical Transport in Solids: with Particular Reference to Organic Semiconductors,first ed., New York: Pergramon Press, vol. 44, 1981, pp. 731–733.
[55] N.F. Mott, R.W. Gurney, Electronic Processes and Ionic Crystals, Oxford Press, 1940.
[56] S.C. Jain, W. Geens, A. Mehra, V. Kumar, T. Aernouts, J. Poortmans, R. Mertens, M. Willander, Injection-and space charge limited-currents in doped conducting organic materials, J. Appl. Phys. 89 (2001) 3804–3810.
[57] P.W.M. Blom, M.J.M. de Jong, J.J.M. Vleggaar, Electron and hole transport in poly (p-phenylene vinylene) devices, Appl. Phys. Lett. 68 (1996) 3308–3310. [58] S.C. Jain, A.K. Kapoor, W. Geens, J. Poortmans, R. Mertens, M. Willander, Trap
filled limited of conducting organic materials, J. Appl. Phys. 92 (2002) 3752–3754. [59] W. Bantikassegn, O. Inganäs, The electrical properties of junctions between
alu-minium and doped polypyrrole, J. Phys. D: Appl. Phys. 29 (1996) 2971–2975. [60] J.A. Anta, G. Marcelli, M. Meunier, N. Quirke, Models of electron trapping and
transport in polyethylene: current-voltage characteristics, J. Appl. Phys. 92 (2002) 1002–1008.
[61] M. Meunier, N. Quirke, Molecular modeling of electron trapping in polymer in-sulators, J. Chem. Phys. 113 (2000) 369–376.
[62] M. Meunier, N. Quirke, A. Aslanides, Molecular modeling of electron traps in polymer insulators: chemical defects and impurities, J. Chem. Phys. 115 (2001) 2876–2881.
[63] M. Lagos, J. Mahanty, V. Slusarenko, Subsurface traps for impurities, Surf. Sci. 191 (1987) 806–812.
[64] R.L. McCreery, Molecular electronic junctions, Chem. Mater. 16 (2004) 4477–4496. [65] W.F. Reus, C.A. Nijhuis, J.R. Barber, M.M. Thuo, S. Tricard, G.M. Whitesides,
Statistical tools for analyzing measurements of charge transport, J. Phys. Chem. C 116 (2012) 6714–6733.
[66] A. Wan, L. Jiang, C.S.S. Sangeeth, C.A. Nijhuis, Reversible soft top-contacts to yield molecular junctions with precise and reproducible electrical characteristics, Adv. Funct. Mater. 24 (2014) 4442–4456.
[67] A.J. Bard, L.R. Faulkner, Electrochemical Methods-fundamentals and Applications, second ed., Wiley Electrochemistry, 2001.
[68] C.A. Nijhuis, W.F. Reus, G.M. Whitesides, Molecular rectification in metal-SAM-metal oxide-metal-SAM-metal junctions, J. Am. Chem. Soc 131 (2009) 17814–17827.
[69] C. Joachim, M.A. Ratner, Molecular electronics: some views on transport junctions and beyond, Proc. Natl. Acad. Sci. USA 102 (2005) 8801–8808.
[70] L. Jiang, C.S.S. Sangeeth, A. Wan, A. Vilan, C.A. Nijhuis, Defect scaling with contact area in EGaIn-based junctions: impact on quality, joule heating, and apparent in-jection current, J. Phys. Chem. C 119 (2015) 960–969.
[71] L. Yuan, L. Jiang, B. Zhang, C.A. Nijhuis, Dependency of the tunneling decay coefficient in molecular tunneling junctions on the topography of the bottom electrodes, Angew. Chem., Int. Ed. 13 (2014) 3377–3381.
[72] K.S. Kumar, R.R. Pasula, S. Lim, C.A. Nijhuis, Long-range tunneling processes across ferritin-based junctions, Adv. Mater. 18 (2016) 1824–1830.
[73] M. Koch, F. Ample, C. Joachim, L. Grill, Voltage-dependent conductance of a single graphene nanoribbon, Nat. Nanotechnol. 7 (2012) 713–717.
[74] Z. Ahmad, Dielectric materials: Chapter 1 Polymer dielectric materials, InTechOpen, 2012.
[75] M. Jaiswal, C.S.S. Sangeeth, R. Menon, Electroconductance in single-wall carbon nanotubes, Appl. Phys. Lett. 95 (2009) 0321111–0321113.
[76] C. Sentein, C. Fiorini, A. Lorin, J.-M. Nunzi, Molecular rectification in oriented polymer structures, Adv. Mater. 9 (1997) 809–811.
[77] P.S. Abthagir, R. Saraswathi, Junction properties of metal/polypyrrole Schottky barriers, J. Appl. Polym. Sci. 81 (2001) 2127–2135.
[78] G. Horowitz, Organicfield-effect transistors, Adv. Mater. 10 (1998) 365–377. [79] H. Sirringhaus, Device physics of solution-processed organicfield-effect transistors,
