Enhanced doping and structure relaxation of unsubstituted polythiophene through oxidative chemical vapor deposition and mild plasma treatment

We report on the enhancement of electrical properties of unsubstituted polythiophene (PT) through oxidative chemical vapor deposition (oCVD) and mild plasma treatment. The work function of p-type oCVD PT increases after the treatment, indicating the Fermi level shift toward the valence band edge and an increase in carrier density. In addition, regardless of initial values, nearly the same work function is obtained for all the plasma-treated oCVD PT films as high as ∼5.25 eV, suggesting the pseudo-equilibrium state is reached in the oCVD PT from the plasma treatment. This increase in carrier density after plasma treatment is attributed to the activation of initially not-activated dopant species (i.e. neutrally charged Br), which is analogous to the release of trapped charge carriers to the valence band of the oCVD PT. The enhancement of electrical properties of oCVD PT is directly related to the improvement of the thin film transistor performance such as drain current on/off ratio, ∼103 and field effect mobility, 2.25 × 10−2 cm2 Vs−1, compared to untreated counterparts of 102 and 0.09 × 10−2 cm Vs−1, respectively.


Introduction
Recent rapid advancements in wearable, flexible, and portable electronics increase attention in π-conjugated polymers due to their mechanical flexibility, electrical conductivity, and tunable functionalities such as surface properties, selective permeability, and stimuli responses [1][2][3][4].Thus far, the majority of the studies regarding conjugated polymers have focused on substituted polymers to enhance the chemical reactivity and hence facilitate solution-based polymerization processing [5][6][7].However, the enhanced reactivity of substituted polymers also promotes reactions with water and oxygen, leading to significant performance degradation over time [8][9][10][11].These molecules of water and oxygen react with dopant anions and residual oxidants in some conjugated polymers, during which negatively charged carriers are donated and recombined with positively charged carriers, leading to a reduction in conductivity [10][11][12][13].Therefore, it has been a long-time challenging issue to mitigate the performance degradation over time, which hinders the implementation of this class of materials into real-world devices.
Despite promising features such as intrinsic properties, stable performance, and simpler structures without side chains, the synthesis of unsubstituted polymers has been limited in conventional solution-based processing due to its rigid backbone structure and low reactivity.In addition to the benefits above, unsubstituted polythiophene (PT), as one of the unsubstituted polymers, is also known for its reasonable dopability that leads to a semiconducting range of conductivity for promising electronic and optoelectronics applications in such as organic thin film transistors and organic photovoltaics.We previously reported the facile synthesis of unsubstituted PT through oxidative chemical vapor deposition (oCVD) at room temperature with mild vacuum conditions [14][15][16].The effect of processing pressure in the oCVD chamber on the PT conjugation was investigated.The PT conjugation increases with increasing chamber pressure while the carrier density increases from 4 × 10 16 cm −3 to 3 × 10 17 cm −3 in the investigated chamber pressures, ranging from 1 to 300 mTorr [14].Excellent air stability was demonstrated for the oCVD PT with nearly consistent electrical properties during the monitoring period of longer than three months, which results from the rigid nature of the unsubstituted backbone structure [15].Although there were other efforts to process unsubstituted PT including electrochemical polymerization, the applications of these methods were limited due to harsh processing conditions and additional requirements such as the inevitable use of conductive substrate [17][18][19].
The next challenge for the implementation of PT, as well as for many other typical organic and polymeric semiconductors, is to enhance their prevalent disordered states and defects.Since these disordered states and defects work as charge carrier trap sites and/or scattering centers, the carrier transport of these organic and polymeric semiconductors is quite restrained and the resulting carrier mobility is typically as low as less than 1 × 10 −4 cm 2 Vs −1 while that of the inorganic counterparts typically much higher, >1 cm 2 Vs −1 .Therefore, it is required to treat or passivate these states so that higher carrier mobilities can be achieved.Particularly, despite its superior environmental stability to other substituted polymers, research on the defect states of unsubstituted PT is limited due to its processing difficulties (i.e., insoluble nature in solvents).
In this report, the effect of mild plasma treatment on the defect state in oCVD PT and the consequent electrical properties is discussed.The electronic states, chemical environments, and electrical properties were evaluated through x-ray photoelectron spectroscopy (XPS), UV-Vis, and Raman spectroscopy and the characteristics are compared before and after plasma treatment.In order to investigate the impact of the defect state treatment on electronic device applications, TFT devices were fabricated and the performance is compared before and after the treatment.

Experimental details
A custom-designed oCVD reactor was used to synthesize and deposit unsubstituted PT films.Vapors of thiophene monomers (>99%, Sigma-Aldrich) were introduced to the oCVD chamber from a temperature-controlled glass jar and then the monomer vapors reacted with oxidants of iron (III) chloride (FeCl 3 , 97%, Sigma-Aldrich), which was sublimated in the chamber.Due to the high vapor pressure of thiophene monomer in vacuum, no heating was required to vaporize the monomer while the oxidants were sublimated from a crucible encased by a resistive heater at a temperature of 180 • C. In the oCVD chamber, PT films were deposited on Si and glass substrates at room temperature.The substrate stage was located at the top of the reactor chamber and about 20 cm from the oxidant source crucible, of which the inverted face-down configuration was intended to provide directionality of the oxidant vapor to the substrate and also to minimize the surface adsorption of the unreacted species during the polymerization process.To ensure the film homogeneity, the substrate stage was rotated at 5 rpm.A chamber base pressure of 1 × 10 −4 Torr or below was achieved by a turbomolecular pump and a dual stage rotary vane pump.During the oCVD depositions, working pressures in the chamber was automatically controlled by a pneumatically controlled throttle valve within a range of 10 −3 -10 −2 Torr.As reported previously, the oxidizing agent and its associated pressure in the chamber are the primary factors in altering the doping level in the resulting oCVD PT films [15,16].To manipulate the doping level and consequently the carrier density, oCVD PT films were deposited at five different pressures of 1, 10, 75, 150, and 300 mTorr.These pressures were adjusted by varying the FeCl 3 heating temperature (ranging from 170 • C to 225 • C), while maintaining a consistent thiophene monomer flow rate was fixed at approximately 30 sccm throughout the oCVD process.These samples were denoted as PT-1, PT-2, PT-3, PT-4, and PT-5, respectively, based on the deposition pressure.Resulting oCVD PT films were rinsed with 5 M HBr acid and methanol for 10 mins each to remove residual oxidants and unreacted monomers.Some of the oCVD films were treated with mild Ar plasma in a plasma cleaner (Glow Research) at 10 W with ultra-high purity Ar (99.999%) at 30 sccm for 60 s.Before the plasma treatment, the chamber was pumped down to a base pressure below 4 × 10 −3 Torr, and the treatment was conducted at a working pressure of 1 × 10 −1 Torr without any heating of the substrate.
Thickness of oCVD PT films was measured using an ellipsometer (FS-1, Film Sense) at a light incident and detection angle of 65 • .Optical properties and chemical bonding in oCVD PT were evaluated with a Cary50 UV-Vis spectrophotometer (Varian/Agilent), Fourier-Transform Infrared Spectroscopy (FTIR, Nexus 670 ThermoNicolet Spectrometer) with an attenuated total reflection accessory, and Horiba LabRam Evolution multiline Raman spectrometer with a 532 nm laser.Surface morphologies and root mean square (RMS) roughness of the synthesized thin films were measured using atomic force microscopy (AFM, Bruker Innova) in tapping mode with a Bruker Si cantilever (tip frequency of 330 kHz).Microstructures of the oCVD PT films and devices were analyzed and imaged using a FEI Nova NanoSEM system.
The chemical environment and valence state of major elements in oCVD PT were investigated in a Thermo Fisher NEXSA XPS with soft Al Kα x-ray radiation at an incident energy of 1486.6 eV.XPS measurements were conducted at ultra-high vacuum at a pressure of 5 × 10 −8 Torr or below with an electron flood gun to prevent charging on the sample surface and mild Ar cluster ion (Ar n + ) cleaning to exclude any surface contaminants.Photoelectrons were collected at an angle of 55 • from the surface normal.The work function of oCVD PT was determined using the same XPS spectrometer with a conductive sample stage by which the oCVD PT films were biased with respect to earth at a voltage of −30 V, and the photo-emitted electrons were accelerated and the onset of emission was detected.The spectrometer and the binding energy (BE) scale were calibrated in a typical way with a reference gold specimen to ensure accurate determinations of onset energy and Fermi level position.
Bottom-gated TFT devices based on oCVD PT channel were fabricated on thermally oxidized SiO 2 /Si substrates where PT (∼30 nm), SiO 2 (50 nm), and heavily-doped Si (0.003-0.005Ω cm) were used as the channel, the gate dielectric, and the gate electrode, respectively.For TFT source and drain electrodes, Ti/Au (∼5/50 nm) was thermally evaporated on top of the channel oCVD PT.Channel and source/drain regions were defined using in-situ shadow masks with six different aspect ratios (width/length, W/L) of 2000/200, 2000/150, 2000/100, 1000/200, 1000/150, and 1000/100 in µm/µm.The TFT device performance was evaluated with an Agilent 4155B semiconductor parameter analyzer in a light-tight probe station at room temperature in ambient air.The output and transfer characteristics were evaluated with more than 18 devices for each set of devices.The fabricated devices show approximately ±5% deviations in the TFT performance parameters such as field effect mobility, which demonstrates good reproducibility of the oCVD and other employed fabrication process.

Results and discussion
Thiophene underwent oxidative polymerization with an oxidizing (i.e.doping) agent of sublimated iron(III) chloride on a substrate, leading to the formation of a solid polymer film of which the detailed synthetic route of PT is shown in figure 1 [20, 21].The commonly accepted mechanism for this oxidative polymerization process in thiophene entails the creation of radical cations [21,22].Subsequent oxidation processes give rise to polarons (figure 1, x) and bipolarons (dimers; figure 1, y), where the dopant counter anions scavenge two protons to stabilize the dimer (figure 1, z) The oxidizing agent further acts on the dimer, which these stepwise reactions are continuously repeated to form the PT polymer (figure 1, {).In a doped state, the resulting conductivity is attributed to the polarons and bipolarons due to the excess oxidation where the cations are charge-balanced by dopant counter anions [21,23].
In order to identify the effect of mild plasma treating on the electronic structure of oCVD PT, work function measurements were performed on a series of oCVD PT films (PT-1 through 5 grown at different chamber pressures as detailed in the Experimental section) using photoelectron spectroscopy.The work function (Φ) is related to the photon energy (hν = 1486.6eV in this study), the onset cut-off kinetic energy (E onset ) which is the free electron energy due to the photoelectric effect and the Fermi edge kinetic energy (E F,kin ) from which the Fermi energy location is estimated [24].The work function is determined by the equation below [24,25]: ( The photoelectron spectroscopic results are plotted in figures 2  Since no additional extrinsic doping is made during the Ar plasma treatment (4 eV), the origin of the increase in carrier density is likely attributed to that the plasma treatment works as an additional energy source, further activating initially present but inactivated dopants (i.e.neutrally charged doping agents).This process can be equivalently understood by that carriers trapped at defect states within the bandgap are released in the presence of additional energy source (i.e.plasma treatment).The increase in the carrier density (figure 2(c), lower) was further verified by TFT/TLM measurements as shown in figure S4 (supporting information) [14,15].The overall trend of the carrier density concurs with that of the work function (figure 2(c, upper)), which supports the notion that the pseudo-equilibrium carrier density is achieved after the plasma treatment.
The oCVD PT-4 sample was selected for further investigation since TFT devices with PT-4 demonstrated the highest figure-of-merit performance such as drain current on/off ratio and field effect mobility.Figure 2(d) shows typical UV-Vis absorption spectra of oCVD PT-4 before and after a mild plasma treatment where a slight red shift is observed after the treatment and the peak maximum is found at approximately 520 nm due to the interband transition with the bandgap [26].The optical bandgap (E g ) is determined from the UV-Vis spectra via the equation, E g = hc/λ onset , where h is the Planck constant, c is the speed of light, and λ onset is the onset wavelength at which the illuminated light energy is sufficient to excite valence electrons to the conduction band [27].The bandgap of oCVD PT is found to be approximately the same as 1.97 eV for both untreated and treated films.
Valence band (E V ) measurements of oCVD PT films before and after mild plasma treatment are shown in figure 2(e).The valence band location (E F -E V ), with respect to the Fermi level (E F ), of oCVD PT is estimated to be approximately 0.59 and 0.31 eV for the untreated and plasma-treated PT films, respectively, by the linear extrapolation of the linear regime of the spectra.The E V identification further verifies the p-type conduction behavior of oCVD PT since the E F is located below the midgap within the bandgap for both cases before and after the plasma treatment.In addition, the reduced (E F -E V ) location of 0.31 eV after the mild-plasma treatment indicates that the E F shifted toward the valence band, which further supports the increase in carrier density after the treatment.According to the measured electronic and optoelectronic characteristics and energy levels, the band structure of oCVD PT is constructed in figure 2(f) with the work function of Au used for the source/drain metallization for TFT application in this study.
Raman spectroscopy was exploited to further compare doping levels and ordered/disordered structures in oCVD PT-4 films before and after plasma treatment.Figure 3(a) shows the Raman spectra for the films before (lower) and after (upper) the mild plasma treatment in the range of Raman shift from 500 to 2800 cm −1 , which are similar to each other.The strong intensity of the peak at approximately 1459 cm −1 is attributed to the symmetric C α =C β stretching band.Peaks at the lower Raman shift regime (∼650-750 cm −1 ) are ascribed to the combined contributions from the coplanar thiophene structure (at ∼700 cm −1 ) and chain distortion (at ∼680 cm) −1 [28].It should be noted that, therefore, the intensity ratio between these two components allows for the comparison of the relative amounts of distortion in the chain.In figure 3(b), the areal intensity ratio of A 680 /A 700 is found to be 0.46 for the pristine PT, which is 4.6 times higher than that of the plasma-treated sample (A 680 /A 700 = 0.10) [14].In addition, a minor peak due to the C=C antisymmetric stretching vibration at ∼1521 cm −1 appears on the right side (higher Raman shift) of the symmetric C α =C β stretching band, of which the antisymmetric vibration indicates shorter PT chains or distortion of thiophene rings confining π electrons [14,28].In the extended view near 1521 cm −1 (figure 3(c)), the weaker antisymmetric intensity of the plasma-treated oCVD PT, therefore, is another indicative of increased ordered structure and enhanced conjugation of the plasma treated film.These Raman analyses reveal that the plasma treatment significantly enhances the chain structures with more ordered coplanar structure.This structural relaxation, which is the observed enhanced conjugation and ordered structure, is known to release the confined π electrons at the disordered/defect state (i.e., become delocalized electrons) [29][30][31].These analyses regarding the PT chain conjugation and the level of distortion support the work function investigations and enhanced p-type doping levels (i.e., conjugation) shown in figure 2.
In order to further investigate the optoelectronic properties of oCVD PT-4 before and after plasma treatment, particularly the difference in energy and defect states within the bands, PL spectroscopy measurements were performed.Figure 4 shows the normalized PL spectra of the oCVD PT-4 films before and after plasma treatment, from which the bandgap of oCVD PT is determined by the peak maximum  located at hν = ∼1.968for both untreated and plasma-treated samples [26,32].These PL bandgap investigations agree the UV-Vis measurements shown in figure 2(d).However, prominent differences are observed in the PL spectrum after the treatment.The PL intensity attributed from the side bands (i.e.rather than the main band-to-band transition at the main peak energy), which are related to defect states, is significantly decreased in the treated PT [32].Particularly in the inset spectra (not normalized), the peak maximum PL intensity of the plasma-treated PT is significantly increased, which indicates that the band-to-band transition becomes much dominant while in the as-deposited (untreated) sample, a significant amount of light energy (i.e.photon) was consumed to photo-excite trapped electrons in the side bands (i.e.defect states).The overall contribution the side bands to the PL excitation the untreated PT film is comparable to the amount of the band-to-band excitation, indicating a significant portion of disordered phases present in the untreated sample.In addition, the full width at half maximum is considerably reduced (∼0.005eV) after treatment, compared to ∼0.011 eV of untreated counterpart (figure S2 in supporting information).These changes after the mild plasma treatment indicate that the oCVD PT films become a more ordered phase with less defect states and more uniform HOMO and LUMO energy states within the molecules.This notion can be further understood by a relaxation process of the disordered structure of the pristine PT, which is equivalent to the detrapping process that liberates trapped charge carriers from localized trap states [33].This relaxation and additional doping after the mild plasma treatment are supported by the changes in Raman information near 690 and 1520 cm −1 as depicted in figure 3, as well as the FTIR spectra shown in figure S1, which demonstrate enhanced doping after the plasma treatment.
The PL results well support the work function and Raman measurements, in which the increased work function after treatment is attributed to the reduced defect states and the associated delocalization of trapped  carriers through structural relaxation [29][30][31]33].A plausible mechanism for the structural relaxation, trapping and releasing free carriers, is suggested through XPS investigations later in figure 5 that determine the valence state of the major elements and hence the related doping state of the oCVD PT films.
AFM and SEM analyses were employed to determine the effect of the plasma treatment on the microstructure of oCVD PT films.These investigations indicated that the reduced defect state, resulted from the treatment is of direct relevance to the microstructure of oCVD PT films.AFM topographic images obtained from the plasma-treated sample (figures 4(e) and (f)) denser and more networked structure that that of as-deposited untreated counterparts (figures 4(b) and (c)).The RMS surface roughness of the treated PT sample is ∼9.5 nm while the RMS roughness of the untreated sample is ∼17.2 nm.SEM micrographs (figures 4(d) and (g)) concur with the AFM images, exhibiting smoother surface microstructures of the plasma treated PT samples.The enhanced ordered structure, identified by Raman (figure 3) as well as the smoother surface morphology in AFM and SEM images are expected to facilitate the charge transport and enhance the TFT performance.This will be further detailed later in figure 6.
In order to investigate the chemical environment and the valence state of the oCVD PT films and also to evaluate the effect of mild plasma treatment on the doping state, the BE of major elements was measured by XPS survey and core-level high resolution (HR) scans before and after the plasma treatment.Figure 5(a) shows full survey scans in the BE range of 0-850 eV of the treated and untreated oCVD PT films.These spectra are similar to each other in the survey scans, showing considerable peaks from O, C, S and Br.No Fe-and Cl-related information was detected in the XPS survey analysis (figure 5(a)) and HR spectra (figure S3 in supporting information).Fe-related reaction products and residual species were removed by acid and MeOH rinsing.In addition, Cl -ions were exchanged with Br -ions through a dopant exchange process as reported previously during HBr treatment through the equilibrium reaction where a complex of the doped monomer unit (EDOT + ) and its counter-anion (i.e.Br − ) can be represented as [EDOT + Br − ] [8,12]: Core-level HR Br 3d XPS spectra are shown in figure 5(b) before and after treatment.The locations of the two major peaks are identical to each other.A peak located at lower binding energies of ∼67.4 eV is attributed to non-oxidized bromine (Br) species [34,35].Another peak positioned at higher binding energies are due to the oxidized Br phases at ∼70.3 eV through the dopant exchange as described in equation ( 2) [8,12,34,35].It should be noted that after the treatment, the peak from the oxidized Br is considerably increased where the peak ratio of oxidized/non-oxidized is 89.1% after the mild plasma treatment while the ratio before the treatment is much lower, 52.9%.The increased amount of oxidized contribution observed after treatment provides clear evidence that a greater quantity of bromine has reacted with the monomers (i.e.EDOT + Br − ), and is present in the polymer's backbone structure, likely forming bonds between carbon in EDOT and bromine (C-Br).This can be understood by (i) physiosorbed Br-species (e.g.HBr) in the resulting oCVD PT were further doped into the PT; and/or (ii) initially incorporated but not activated Br were activated as practical dopants that accept an electron and leave a hole in the polymer backbone of which the process is facilitated by the additional energy source during the plasma treatment.This process is well supported by the work function analysis in figure 1 that proves the enhanced doping (i.e. carrier density increase) after treatment.
The core-level HR C 1 s spectra are shown in figure 5(c) where each spectrum consists of two major peaks due to C-C at lower BE ∼284.6 eV (peak 1) and bond between carbon and oxygen (e.g.C-O & C-O-C) at higher BE ∼286.1 eV (peak 2) [36].Other minor intensities ascribed to double bonds between carbon and oxygen (C=O) at BE = ∼287.7 eV and shake-up features of PT at 289.2 eV are also observed.A major difference after plasma treatment is the reduced intensity of the C-C contribution of which the peak ratio (peak 1/peak 2) is 0.68, while the peak ratio before the treatment is 0.97.The reduced C-C intensity may be accounted by the formation of bonds with carbon and other species such as Br during the treatment, which is supported by the HR Br3d XPS spectra (figure 5(b)), further supporting the higher oxidized Br content after the mild plasma treatment.
To investigate the effect of the enhancement of optoelectronic properties on potential device applications, test TFTs were fabricated and the performances were compared before and after the plasma treatment.The only difference between the two device sets is the application of mild plasma treatment to the channel PT for the treated TFTs.In figure 6(a), a bottom-gated device schematic and a SEM image of the fabricated TFTs are depicted.As provided in the experimental section, six different channel width/channel length (W/L) aspect ratios were used for the TFT performance evaluations.Drain current (I D ) was recorded at drain bias (V D ) scanned from 0 to 25 V as the primary sweep and gate bias (V G ) applied from 12 to −18 V as the secondary sweep.In order to directly compare the drain current in response to the drain and gate biases, the same y-axis scale was used on both output plots in figures 6(b) and (c).Within these measurement ranges, both pristine and plasma-treated TFT devices demonstrate solid saturation behaviors at all gate biases applied.In the inset in figure 6(b) of the pristine TFT, the saturation I D is approximately 4 × 10 −8 A at the gate bias of −18 V.However, after plasma treatment (figure 6(c)), the drain current is much enhanced, 6.4 × 10 −7 A which is approximately 15-fold higher than that of the pristine TFT.The enhanced current behavior is mainly attributed to the increased carrier density and ordered state (i.e.enhanced carrier transport) after the plasma treatment, which was verified with the work function (figure 2) and PL (figure 4) investigations in the previous paragraphs.This is in agreement with an earlier study that reports that order/disorder structures and morphology of conducting polymers play an important role in governing transport properties [37].Device transfer characteristics were also measured and compared in figure 6(d).After plasma treatment, drain current on/off ratio increased by about an order of magnitude, which concurs with other measurement results that validate an increase in carrier density (i.e.doping level) and reduced defect (disordered) states.The field effect mobility (µ FE ) of the devices (figure 6(e)) was estimated in the saturation regime using the equation below, by which the effect of plasma treatment on the carrier transport was investigated where C ox is the oxide capacitance, and the V T is the threshold voltage.In order to accurately extract the µ FE , the (I D ) 1/2 vs V G curves are corrected (dashed lines in figure 6(e)) as we reported previously [15,38].The field effect mobility of the treated TFT, obtained from the corrected curve, is estimated to be ∼2.25 × 10 −2 cm 2 Vs −1 which is more than 20 times higher than 0.09 × 10 −2 cm 2 Vs −1 of the untreated counterpart.The enhanced carrier mobility after the mild plasma treatment is attributed to the reduced scattering events of the free carriers by the reduced disordered state and the increased carrier density that augments the charge screening effect [15,38].

Conclusions
We demonstrated mild plasma treatment as an effective way to reduce defect states of oCVD PT within the gap and enhance the carrier density and the carrier transport.A series of characterizations such as work function and valence band measurements evidenced an increase in carrier density after plasma treatment.In addition, XPS investigations identified the increased oxidized Br ion content after the treatment, which is the dopant counter anions of the oxidatively polymerized PT and therefore a direct indication of enhanced doping.Raman shift and PL analyses exhibited much ordered states in the mild plasma-treated PT, which suggests a structural relaxation mechanism for the enhanced carriers after the mild plasma treatment.
During the phase transitioning from disordered to ordered states, free carriers that were initially trapped in the defect states are released and contribute to the conduction process.Much enhanced TFT performance was achieved with the plasma-treated PT channel due to the reduced defect states (i.e., enhanced carrier transport) and the enhanced doping.This study may have significant relevance to studies that aim to enhance the structural properties of conducting polymers and improve the device performance by integrating organic and polymeric semiconductors.
(a) and (b) where the E onset and the E F.kin are shown in figures 2(a) and (b), respectively, as a function of carrier density exhibited in figure 2(c).The scatter plots indicate the results before the plasma treatment while the solid line curves show the plasma-treated spectra.The E onset (figure 2(a)) is determined as the point where the linear regime of each curve intersects with a baseline.The position of E F,kin (figure 2(b)) is estimated by extrapolating the linear portion of kinetic energy.The extracted E onset and E F,kin are tabulated in table S1 (supporting information).The determined work function of the samples before and after plasma treatment is summarized in figure 2(c, upper).The work function of as-deposited oCVD PT increases with increasing carrier density (figure 2(c, lower)): PT-1 with the carrier density of 0.4 × 10 17 cm −3 (lowest) exhibits the work function of approximately 4.54 eV and monotonically increases to 5.23 eV of PT-5 with the highest carrier density of 2.5 × 10 17 cm −3 .The significance of the work function analysis is two-fold: First, the work function of oCVD PT increased after the mild plasma treatment, which indicates that the Fermi level of PT shifted toward the

Figure 1 .
Figure 1.Schematic of the polymerization process of polythiophene.

Figure 2 .
Figure 2. (a)-(c) work function measurements of oCVD PT before and after mild plasma treatment: (a) Eonset, (b) E F.kin , and (c) extracted work function and related carrier density of oCVD PT films; (d) UV-Vis absorption spectra of oCVD PT films, where the Eg is extracted; (e) Valence band of oCVD PT which is located approximately 0.59 and 0.31 eV below the Fermi level (i.e.EF-EV) for the as-deposited and mild plasma-treated oCVD PT, respectively; and (f) constructed energy-level diagram according to the work function, Eg and valence band measurements along with Au for the channel/metallization contact in TFT application.

Figure 3 .
Figure 3. (a) Raman spectra of oCVD PT-4 films before (blue; bottom) and after (orange; upper) plasma treatment, along with extended views of Raman shift regimes ranging (b) 620-760 cm −1 and (c) 1500-1540 cm −1 .The spectra reveal enhanced ordered chain states with coplanar structures and reduced asymmetric C=C stretching after plasma treatment, indicating significantly improved π-conjugation and higher doping concentrations in the plasma-treated PT sample.

Figure 4 .
Figure 4. (a) Normalized PL spectra before and after plasma treatment from which the bandgap of oCVD PT is determined to be approximately 1.968 eV for both as-deposited and plasma treated samples.The PL spectrum of plasma treated sample evidently presents much reduced side bands, indicating enhanced ordered states resulted from the plasma treatment.Inset (not-normalized) spectra show that the plasma treated oCVD PT displays significantly stronger band-to-band transition, further supporting the notion of enhanced ordered states.(b)-(g) microstructure investigations of AFM topographic images on (b), (e) a 10 µm × 10 µm area and (c), (f) a 1 µm × 1 µm area, and (d), (g) SEM micrographs, obtained from as-deposited (untreated) and treated oCVD PT samples, respectively.

Figure 5 .
Figure 5. XPS spectra of (a) survey scans and core-level HR scans of (b) Br 3d and (c) C 1s obtained before and after mild plasma treatment, supporting the plasma-assisted enhanced doping state in the oCVD PT films.

Figure 6 .
Figure 6.(a) Schematic of a bottom-gated oCVD PT-based TFT device with various channel aspect ratios (W/L) of 2000/200, 2000/150, 2000/100, 1000/200, 1000/150, and 1000/100 in µm/µm, and an SEM micrograph of fabricated devices; TFT output characteristics obtained from (b) as-deposited and (c) plasma treated PT-based TFTs where the drain current of the treated TFT devices increases approximately one order of magnitude after plasma treatment, which is indicative of the enhanced carrier density in the PT channel.Inset in (b) is an extended plot of the output curve (b).; and transfer curves of (d) ID vs VG and (e) (ID) 1/2 vs VG where the device on/off ratio and the field effect mobility are extracted before and after plasma treatment.The TFT transfer characteristics were measured at a saturation drain bias at −25 V.