Heavily N-doped monolayer graphene electrodes used for high-performance N-channel polymeric thin film transistors

Abstract

Recently graphene attracted much attention as a promising electrode material for organic field effect transistors (OFETs). However the electrodes used in most of the graphene-based OFETs were prepared from pristine graphene which suffers from relatively low conductivity and poor controlling of work function. In this work, we report the N-doping by Cs₂CO₃ of the CVD-grown single-layer graphene (SLG) via a facial spin-coating process. The Cs₂CO₃-engineered SLGs exhibit a heavy and stable N-doping, as well as significantly decreased work function (3.9 eV) compared to pristine graphene. The doped graphene was used as the source/drain electrodes in the bottom-contact top-gated OFETs based on a good electron transporter poly{[N,N′-bis(2-octyldodecyl)-1,4,5,8-naphthalenedicarboximide-2,6-diyl]-alt-5,5′-(2,2′-bithiophene)} (P(NDI2OD-T2)). The polymeric FETs show an enhancement of electron mobility by a factor of 10, as well as a reduction of contact resistance compared to the devices using pristine graphene. It is attributed to a remarkable lowering of electron injection barrier at the polymer/graphene electrode due to the decreased work function of graphene. In addition, the microstructural observations reveal that the face-on molecular packing and morphological feature do not change for the P(NDI2OD-T2) films coated on the doped graphene electrodes compared to those on the SiO₂ dielectric, in spite of a highly hydrophobic surface of the graphene.

---

1. Introduction

Carbon-based nanomaterials synthesized by different methods have potential applications in numerous areas including supercapacitors, electronic devices, catalysis and so on.^1–11 Graphene, a honeycomb-like sheet of atomic carbons,^12,13 has recently attracted much attention as a promising electrode material for organic devices such as light-emitting diodes,^14,15 solar cells^16,17 and organic field-effect transistors (OFETs),^18–22 due to its high transparently, excellent mechanical flexibility and environmental stability. Large area single-layer graphene (SLG) synthesized via chemical vapor deposition (CVD) method has been utilized as source/drain (S/D) electrodes for the OFETs.^18–22 The devices exhibit good hole- or ambipolar transport properties using the active materials such as pentacene or the low band-gap polymers, as well as higher charge injection capability than the commonly used Au electrode, which is attributed to the favorable interaction between graphene and the active layer.^20,22 Furthermore, for the OFETs with bottom-contact structure, the physical steps form at the interface of the metal S/D electrodes and the organic active layer, which generates a poor electrical contact and different growth of the semiconducting layer on the electrodes and within the channel region. Atomic-layer thin and atomically flat graphene enables the formation of low steps at the interface and continuous growth of the organic films over the electrode and channel region. However, till now the graphene electrodes used in the OFETs were mostly prepared from pristine graphene. Carrier density in pristine SLG graphene is low, which will result in a low electrode conductivity. Pristine graphene also shows the poorly-defined work function (in the range of 4.2–4.9 eV).^18,20–22 Since high performance OFETs requires low contact resistance and efficient charge injection from electrode to organic active layers, it is crucial to perform effective doping to enhance the conductivity and control Fermi level of graphene.

Various chemical doping methods have been applied²³ such as atomic substitution,^24,25 molecular absorption,^26,27 covalent functionalization via self-assembly monolayers,^28,29 and the use of polymer layers or nanoparticles.^30–32 The substitution of carbon atoms with B or N, and covalent functionalization can achieve stable doping, however result in the damage of carbon network of graphene via base plane reaction, as well as significant increase of defect density.^24,25,28,29 On the other hand, in spite of more favorable, non-destructive and noncovalent doping by chemical species has an instability problem. Such issue becomes more challenging for N-type doping due to intrinsic susceptibility of N-dopants to water and oxygen molecules in air. Till now, the reports on high performance flexible electronic device utilizing the N-doped graphene electrodes are still scare. Therefore developing stable, non-destructive yet efficient N-doping approaches is highly dispensable.

Cesium carbonate (Cs₂CO₃) was shown to act as a good interfacial layer between cathode and the organic semiconductors, which efficiently enhances electron injection in such devices as light-emitting diodes and solar cells.^33–35 In this paper, we demonstrate heavy n-doping of the CVD-grown single-layer graphene (SLG) by spin-coating of the Cs₂CO₃ solution on the graphene surface. Compare to the doping of graphene by thermal evaporation deposition of Cs₂CO₃ (ref. 36) or by soaking in alkali metal carbonate salts solution,³⁷ this doping method is more simple and favorable to the fabrication processes of organic electronic devices. We utilize the Cs₂CO₃-doped graphene as the bottom source/drain electrodes of the N-channel OFETs. The active layer used is poly{[N,N′-bis(2-octyldodecyl)-1,4,5,8-naphthalenedicarboximide-2,6-diyl]-alt-5,5′-(2,2′-bithiophene)} (P(NDI2OD-T2), see Fig. 1), a representative of recently-developed low band-gap donor–acceptor polymers.^38–40 It is an air-stable N-type transporter, exhibiting excellent electron mobility as high as 0.85 cm² V⁻¹ s⁻¹ on the top-gated OFETs.³⁹ Our results show that high performance FETs of P(NDI2OD-T2) are achieved using the N-doped graphene S/D electrodes with low work function. To best of our knowledge, it is the first report on performance improvement of the N-channel OFETs via the application of the N-doped SLG electrodes.

Fig. 1 Schematic illustration of the fabrication process for a bottom-contact/top-gate OFET of P(NDI2OD-T2) using the Cs₂CO₃-doped graphene S/D electrodes. The inset is chemical structure of P(NDI2OD-T2).

2. Experimental section

Single layer graphene was synthesized on a copper foil (25 μm, Alfa Aesar) by low-pressure chemical vapor deposition (CVD) in a quartz tube furnace.⁴¹ The copper foil was pre-treated by 5% nitric acid briefly and then cleaned consequently in deionized (DI) water, acetone and isopropyl alcohol. Before growth, the foil was annealed at 1000 °C for 30 min in a flow of 50 sccm H₂ (pressure: 1 Torr). Then a mixture of 10 sccm CH₄ and 50 sccm H₂ flow was introduced for 30 min. As-grown graphene was spin-coated with 4% poly(methyl methacrylate) (PMMA) (M_w ∼ 950k, Sigma-Aldrich) in chlorobenzene, and the copper was etched in Marble's Reagent for 1 h. Once fully etched, the graphene was placed in a 10% HCl solution to remove the Cu residues. Finally the graphene was transferred into the DI water and then placed onto a SiO₂ (230 nm)/Si substrate. The PMMA layer was removed by soaking in warm (50 °C) acetone for 2 hours. The remaining residues were completely removed through annealing under 300 °C in high vacuum (10⁻⁵ Pa) for 3 hours. The graphene FET structure was fabricated by depositing Cr/Au (5/50 nm) source/drain electrodes on the clean graphene through a shadow mask. The doping of the SLG graphene was performed in the N₂ atmosphere by spin-coating of 0.2% Cs₂CO₃ solution (in 2-ethoxylethanol) on the graphene at 3000 rpm for 40 s. Then the doped graphene was annealed in situ at 120 °C for 30 min to remove the solvent.

The OFETs of P(NDI2OD-T2) using the graphene S/D electrodes were prepared with the processing steps shown schematically in Fig. 1. The 70 nm Au strips were evaporation deposited on the graphene through a shadow mask to define the electrode pattern. Uncovered parts of the graphene were etched way under O₂ plasma, and the exposed SiO₂/Si surface was then treated with HMDS. The Au patterns were etched in an iodide solution (Au TFA Etchant, Transene) to obtain the graphene source/drain electrodes with the defined channel lengths and widths (L = 50, 70, 100, 200 μm; W = 200 μm). The graphene electrode areas were doped as described above. A film of P(NDI2OD-T2) was deposited onto the substrate via spin-coating of the 1% solution in dichlorobenzene and subsequently annealed at 120 °C for 1 h. A 900 nm PMMA dielectric layer was spin-coated from a 5% PMMA solution in n-butyl acetate and baked in the N₂ atmosphere at 110 °C for 3 h. Finally an Al (20 nm) strip was deposited through a mask as the gate electrode.

All electrical measurements were performed on a probe station using a Keithley 2612A source meter in a N₂ glove box. The pristine and doped graphene samples were characterized by scanning electron microscopy (SEM, Helios Nanlab 600i), transmission electron microscope (TEM, JEM2010), micro-Raman spectroscopy (T6400, Horiba Jobin Yvon) under at an excitation wavelength of 532 nm, and X-ray photoelectron spectroscopy (XPS) using monochromatized Al (Kα) X-ray (hν = 1486.6 eV). Electronic structure of graphene sheets was acquired by photoemission spectra using a synchrotron light source (photon energy of 170 eV) at National Synchrotron Radiation Laboratory of Hefei. All spectra were measured at an applied bias of −10 V at the sample. For structural characterizations of the P(NDI2OD-T2) films on graphene, 2-dimensional (2D) grazing incidence X-ray diffraction (GIXRD) measurement was performed at Shanghai Synchrotron Radiation Facility (SSRF) on the beam line BL14B with the photon energy of 10 keV. The incidence angle of X-ray beam is 0.2°. Specular scan X-ray diffraction experiment was performed on a Rigaku-TTR3 X-ray diffractometer with Cu Kα (λ = 1.5406 Å) radiation. The morphology of the P(NDI2OD-T2) films was characterized with a Veeco MultiMode (Nanoscope V) atomic force microscope (AFM) under ambient conditions.

3. Results and discussion

The structure of the CVD grown graphene was characterized by STM, via transferring the graphene sheet on a Cu foil onto the TEM lacey carbon-coated grid. Fig. 2a shows the high-magnification bright-field TEM image of a graphene sheet at the edge region, indicating a single-layer structure. Selected area electron diffraction (SAED) patterns of a domain in the graphene shown in Fig. 2b, reveal only one set of sharp hexagonal diffraction spots, manifest of high crystalline nature. Surface changes of graphene after doping were inspected by SEM (in Fig. 2c and d) and energy dispersive spectroscopy (EDS) (in Table S1 and Fig. S1 in ESI†), respectively. The PMMA-transferred and Au-layer patterned graphene exhibits a clean surface, with only some wrinkles observable which is typical for the CVD grown graphene. After the Cs₂CO₃ treatment (in Fig. 2d), some amount of sub-micrometer-size dark particles are found to uniformly disperse on the graphene surface. The EDS element analysis (see Table S1 and Fig. S1†) confirms the existence of the homogenously distributed Cs trace, indicating small particles on graphene should be attributed to the adsorbates of Cs₂CO₃. However, we didn't observe any metal particles which form on the graphene surface doped by alkali metal chlorides.⁴²

Fig. 2 (a) High-magnification TEM image of a CVD-grown graphene, the arrow denotes the edge of the graphene sheet; (b) SEAD pattern of a graphene domain. (c) and (d) SEM images of pristine graphene and the graphene sheet coated by Cs₂CO₃ solution on the SiO₂/Si substrate, respectively.

The doping effect of Cs₂CO₃ on graphene was characterized by comparing Raman spectra of the as-transferred graphene with those of the doped graphene (in Fig. 3a). The G-band and 2D-band position of pristine graphene is ∼1584 cm⁻¹ and ∼2688 cm⁻¹, respectively. The intensity ratio of I_2D/I_G is ca. 2.0, indicating a single-layer graphene.⁴³ Previous works have reported that the position of the G peak upshifts for both electron- and hole-doping, while the 2D peak shifts to higher wavenumber for p-type doping but to lower wavenumber for n-type doping.^27,44,45 In present case, the G peak upshifts by 16 cm⁻¹ and 2D peak downshifts by 5 cm⁻¹ after the Cs₂CO₃ treatment, indicating n-doping of graphene. Furthermore, as shown in Fig. 3b the full width at half maximum (FWHM) of the G peak (11 cm⁻¹) exhibits a narrowing while the intensity ratio of 2D- to G-peak (I_2D/I_G) decreases to 0.6 from 2.0 for pristine samples, consistent with previous observation which is manifest for n-doped graphene. It is also noticed from the inset of Fig. 3a that there is absent of the D peak for the Cs₂CO₃ doped graphene. It indicates that the basal plane reactions and formation of substitution impurities do not take place by the Cs₂CO₃ dopants. Such reactions usually damage the carbon network of graphene and introduce the defects in graphene, leading to an irreversible effect. In our work, the doping effect of Cs₂CO₃ fades away slightly when the sample was exposed in air, as manifested from the downshift of the G peak and the enhanced intensity ratio of I_2D/I_G in Fig. 3c.

Fig. 3 (a) Raman spectra of pristine graphene (P-G) and the Cs₂CO₃-doped graphene. The CVD-grown monolayer graphene sheets were transferred on the SiO₂/Si substrates. The inset shows the spectra around the D-band (1300–1450 cm⁻¹). (b) Comparisons on the FWHM of the G-band (left) and the intensity ratio of 2D to G peak (I_2D/I_G) (right) for pristine graphene and the doped graphene. (c) The changes of Raman spectra of the Cs₂CO₃-doped graphene with the exposing time in air.

The doping level can be roughly estimated from the variation in the I_2D/I_G intensity ratio.⁴⁵ The chemical doping induces the similar intensity suppression of the 2D peak as the doping of graphene by electrochemical gate. A doping level of 1.40 ± 0.3 × 10¹³ cm⁻² is estimated for the Cs₂CO₃ doped graphene, according to the variation of the I_2D/I_G ratio vs. charge concentration measured by Das et al.⁴⁵ Such high electron doping concentration will give rise to the shift of Fermi level of graphene by ca. 0.70 eV. The variation of doping level estimated from the I_2D/I_G intensity ratio is generally consistent with the magnitude of the shift in the G peak position (ca. 6 cm⁻¹ in maximum) upon Cs₂CO₃ treatment. It should be noted that some degree of inhomogeneity in doping level may exist over the graphene flakes.

To have a deeper understanding on n-doping mechanism of graphene by Cs₂CO₃ surface functionalization, the XPS measurements were performed to study the interfacial interaction between Cs₂CO₃ and graphene. Fig. 4a show the C 1s peaks of the CVD graphene before and upon the doping. The C 1s peak of pristine graphene can be separated into four components of the sp² hybridized carbon bond (C=C) at 284.5 eV, sp³ hybridized carbon bond (C–C) at 285.4 eV, C–O bond at 286.3 eV, and a carbonyl group (C=O) near 288.6 eV.⁴⁶ The presence of the oxidized carbon peaks should originate from the wet transfer process of the CVD synthesized graphene, which uses organic reagents such as PMMA and acetone. Due to annealing process after spin-coating Cs₂CO₃ solution, the intensity of C=O and C–O decreased. The C 1s peak shift takes place after the Cs₂CO₃ treatment, with a shift of the C=C peak position to higher binding energy by 0.2 eV from 284.5 eV, a hint of N-doping. Furthermore the intensity ratio of the C=C bonds to C–C bonds decreases after the doping treatment, suggesting the formation of strong chemical bonds between graphene and the adsorbed Cs₂CO₃. Fig. 4b shows the Cs 3d core-level spectra of the Cs₂CO₃-coated graphene. The high intensity of Cs 3d peaks verifies the pronounced adsorption of cesium on graphene by spin-coating process. Previous work has proposed that Cs₂CO₃ tends to combine the oxide functional group (e.g. C=O and C–O) of the carbon atoms with the cesium ions to form the C–O–Cs complexes on the surface of graphene.^37,47 The interfacial monolayer of complex has a low work function and further provides additional dipole to reduce the work function of graphene. Our XPS measurement results support this mechanism by which n-doping originates from covalent bonding between carbon atoms and Cs in Cs₂CO₃. It will be noted that, the concentration of Cs₂CO₃ (0.01 M) applied in our case is much lower compared to 0.1–1.0 M Cs₂CO₃ coated on graphene in previous reports,³⁷ however heavy n-doping effect is still achieved. It will be relevant to efficient adsorption of Cs₂CO₃ on graphene surface as well as the post-annealing to facilitate the formation of the C–O–Cs complexes.

Fig. 4 XPS spectra of pristine graphene (P-G) sheet and the graphene after the Cs₂CO₃ doping, respectively. (a) C 1s core level spectra. The C 1s peak was fitted by four components: sp² carbon (C=C), sp³ carbon (C–C), C–O bond and C=O bond. Small peak shift is evident in the doped graphene; (b) Cs 3d spectra.

To examine the effect of the Cs₂CO₃-doping on electrical performance of graphene, we have fabricated the graphene field-effect transistors (GFETs) with Au/Cr source/drain electrodes, as schematically shown in Fig. 5a. Fig. 5b shows representative drain current–gate voltage (I_D–V_g) curves of pristine graphene and the graphene doped by Cs₂CO₃. The V-shape curve for the as-prepared graphene exhibits typical ambipolar transport properties. The charge neutral point V_CNP (coinciding with the Dirac point in pristine graphene), where the graphene conductivity is minimum, is 30 V, indicating p-doping at the condition without gate bias.⁴⁸ Upon coating Cs₂CO₃ on graphene, the graphene sample shows a monotonic I_D–V_G curve, as well as a huge negative shift of V_CNP (the V_CNP is beyond the maximum negative gate voltage (−110 V) which approaches the breakdown voltage of the SiO₂ dielectric). It indicates a heavy n-doping of graphene as discussed above. After exposing such a doped graphene in ambient air for 30 min, the ambipolar transport appears again however the charge neutral point V_CNP is shifted to −20 V compared to the pristine sample, manifest of a weaker n-doping. The crossover from heavy n-doping into ambipolar properties of graphene should be attributed to the adsorption of water from air which degrades electron conduction, in consistent with the variation of the Raman spectra.

Fig. 5 (a) Schematic of the graphene field-effect transistors (GFETs) prepared in this work. (b) Transfer characteristics of the GFETs based on pristine single-layer graphene (SLG, black), the SLG doped by Cs₂CO₃ (red) as well as the same doped graphene followed by exposing in air for 30 min (blue). The channel length and channel width of the GFETs is 100 μm and 1.5 mm, respectively.

The hole and electron mobilities are derived from the slope of the I_D–V_G curves in linear regime using the equation,

where L and W is the channel length and width, respectively, C_i is the gate capacitance, V_DS is the drain-source voltage. The electron and hole mobility is estimated as 500 and 2050 cm² V⁻¹ s for pristine graphene, respectively. After Cs₂CO₃ doping, electron mobility keeps nearly the same value (500 cm² V⁻¹ s⁻¹), which may be due to relatively low amount of the Cs₂CO₃ dopants on graphene. On the other hand, the Fermi energy of graphene changes with carrier number density n as(where Fermi velocity V_F is 1.1 × 10⁶ ms⁻¹), which can further be inferred by the charge neutral point V_CNP of graphene.⁴⁹ The number density of excess electrons induced by the Cs₂CO₃ doping, estimated from the relation n = η|V_n|, is more than 1.0 × 10¹³ cm⁻², where V_n is the shift of V_CNP (vs. V_CNP of 0 V) and η is 9.4 × 10¹⁰ cm⁻² V⁻¹ in this case. Therefore the Fermi energy position of the doped graphene is upshifted by more than 0.4 eV compared to intrinsic graphene with a V_CNP of 0 V (whereas Fermi energy of the as-prepared graphene downshifts by 0.22 eV). Since the work function of intrinsic graphene was measured as about 4.6 eV,⁵⁰ it can be deduced that the work-function of the heavily doped graphene by Cs₂CO₃ is lowered than 4.2 eV. Furthermore, the work function of graphene sheets was measured by synchrotron radiation photoemission spectra as shown in Fig. S2 (ESI†), and determined to be 4.77 eV and 3.94 eV for pristine graphene and the Cs₂CO₃ doped graphene, respectively. The values are generally consistent with those estimated from above GFET measurements. Such large shift of Fermi level up to ca. 1.0 eV evidences the effectiveness of the Cs₂CO₃ doping of graphene via simple spin coating process.

Since the Cs₂CO₃-doping remarkably reduces work function of graphene, it is therefore expected that enhanced performance of n-type or ambipolar organic FETs is achieved using such graphene electrode. P(NDI2OD-T2) (see Fig. 1) was utilized as the active layer in the OFET as it is an excellent air-stable electron transporter. The polymer films were spin-coated on both the graphene and the SiO₂ dielectric and followed by annealing. Surface morphology of the films was examined by AFM as shown in Fig. 6 and S3.† The film coated on the doped graphene surface (in Fig. 6) displays the interconnected fiber-like domains, with surface root mean squared (RMS) roughness of only 0.40 nm. The similar morphology is also observed on the films coated on pristine graphene and the SiO₂/Si substrates (Fig. S3, in the ESI†), in consistent with previous work.³⁹

Fig. 6 AFM height (a) and phase (b) image of the annealed P(NDI2OD-T2) film spin-coated on the Cs₂CO₃ doped graphene surface. The inset shows the profile of the line highlighted in (a).

X-ray diffraction (XRD) measurements were performed to identify the effects of the electrode and dielectric surface on polymer chain stacking and crystallinity in the P(NDI2OD-T2) films (Fig. S4 in the ESI†). Specular XRD pattern in Fig. S4a† shows no distinct diffraction peaks for the P(NDI2OD-T2) film on graphene. However synchrotron-based GIXRD patterns (in Fig. S4b†) display a wide π-stacking (010) reflection along the out-of-plane direction, as well as multiple orders of lamellar stacking (h00) peaks and chain backbone repeat peaks (00k) along the in-plane direction, manifest of face-on packing and high degree of in-plane ordering in the films. The films coated on the SiO₂ surface show the similar XRD pattern but a higher diffraction intensity.⁵¹ The similar structure and morphology of the films on both the graphene and dielectric surface indicates continuous growth of the ordered domains of P(NDI2OD-T2) over the electrodes and channel region.

The role of graphene electrodes on electrical properties of the P(NDI2OD-T2)-based OFETs were examined, via using the bottom-contact/top-gate FET structure as schematically shown in Fig. 1. Fig. 7b–e show typical transfer and output characteristics of the P(NDI2OD-T2) OFETs using the S/D electrodes of the Cs₂CO₃ doped SLG and pristine graphene, respectively. The field-effect mobility μ was calculated in the saturation regime according to the equation

where C is the capacitance of PMMA dielectric and the V_th is the threshold voltage. The P(NDI2OD-T2) devices using the doped SLG electrodes, exhibits a significantly larger drain current as well as high electron mobility of 0.40 cm² V⁻¹ s⁻¹ (among one of the best values for the P(NDI2OD-T2) based OFETs reported previously^38–40), compared with the devices using pristine SLG electrodes where the FET mobility of 0.04 cm² V⁻¹ s⁻¹ is achieved. It is noted that a mobility of ca. 0.024 cm² V⁻¹ s⁻¹ was achieved on the control devices utilizing the Au S/D electrode.⁵¹ Furthermore the threshold voltage V_T of the doped-SLG based OFET decreases to a low value of 4 V from 10 V for the device utilizing pristine graphene electrodes. The performance enhancement should be attributed to the lowering of work function (to 3.94 eV) of graphene electrode by the Cs₂CO₃ doping, which reduces the height of electron-injection barrier (that is the energetic difference between Fermi level of graphene and the LUMO level of P(NDI2OD-T2)) at the polymer–graphene interface, as illustrated from the diagram of the energy level alignment in Fig. 7a.

Fig. 7 (a) Schematic diagrams of energy level alignment at the interface between P(NDI2OD-T2) and pristine graphene and the n-doped graphene, respectively. (b) and (c) Transfer and output characteristics of the OFETs of P(NDI2OD-T2) prepared with pristine graphene S/D electrodes; (d) and (e) transfer and output characteristics of OFETs of P(NDI2OD-T2) with the n-doped graphene electrodes. The channel length and width is 200 μm and 200 μm, respectively.

The role of graphene electrodes on electron injection was further investigated in term of contact resistance which can be obtained by measuring the device characteristics as a function of channel length using the transmission line method. Fig. S5 (in ESI†) shows a series of the I_D–V_D curves at the fixed V_G (50 V) of the OFETs with different channel lengths (50–200 μm). Fig. 8(a) shows the linear dependence of total device resistance R_total on the channel lengths, where R_total is extracted from the inverse slope of linear parts of the I_D–V_D curves in the lower V_D region (in Fig. S5†). The contact resistance R_c were obtained as the intercept of the extrapolation fit of R_total versus the channel length data, according to the equation

where μ_i is intrinsic field effect mobility. Fig. 8b summaries the R_c data as a function of the gate voltages V_G, which were obtained by performing the same measurement at different V_G's. It is revealed that the contact resistance decreases with the V_G values for the P(NDI2OD-T2) OFETs using both the doped-graphene and pristine graphene as electrodes. Such behavior is correlated to the higher carrier density at the polymer–graphene interface. More importantly, the devices with the Cs₂CO₃ doped SLG show smaller R_c values compared to the OFETs with pristine graphene at all V_G values (for example, 5.31 MΩ μm vs. 8.79 MΩ μm at a V_G of 50 V). It is verified further that the efficiency of electron injection is enhanced via the Cs₂CO₃ doping of graphene electrodes.

Fig. 8 (a) Total resistance R_total versus channel length at a fixed V_G of 50 V for the OFETs of P(NDI2OD-T2) using pristine-and the doped-graphene as the S/D electrode, respectively. The R_total values were calculated from the I_D–V_D curves at different channel length. In this plot, the y-intercept is the contact resistance R_c. (b) R_c versus V_G curves of the OFETs using pristine graphene and the Cs₂CO₃ doped graphene electrodes.

4. Conclusions

In summary, we have described a method by which the CVD grown single-layer graphene can be effectively doped via spin-casting the Cs₂CO₃ solution. Such N-doping process is found to result in a remarkable reduction of work function of graphene and meanwhile not introduce the damages in graphene sheet. The aspects on the process and mechanism of the Cs₂CO₃ doping were understood from comprehensive structural and electronic characterizations. The doped graphene was also utilized as the drain/source electrodes to investigate the impacts on devices performance of the N-type P(NDI2OD-T2) based FETs. The devices using the Cs₂CO₃ doped graphene electrode exhibit a remarkably improved performance compared to those using pristine graphene electrodes, which originates from the decreased electron injection barrier and better contact properties at the interface between P(NDI2OD-T2) and the doped graphene. This work clearly demonstrates that the doping of the graphene electrodes is facile and effective approach to control the operation and improve the performance of organic electron devices.

Acknowledgements

This work was supported financially by Chinese Academy of Sciences (CAS) and National Nature Science Foundation of China (NSFC, Grant No. 11074256, 11574314, 21301177). The authors thank Dr Ranran Zhang, Dr Haifeng Du and Dr Lei Zhang at HMFL for assistance on the Raman spectra, SEM and XRD measurements. The authors also appreciates the technique assistances on the GIXRD experiment from the staffs in the BL14B station at Shanghai Synchrotron Radiation Facilities, as well as on the photoemission spectra measurement from the staffs at National Synchrotron Radiation Laboratory of Hefei.

Notes and references

1. A. Lekawa-Raus, K. Walczak, G. Kozlowski, M. Wozniak, S. C. Hopkins and K. K. Koziol, Carbon, 2015, 84, 118–123.
2. M. Hassan, K. R. Reddy, E. Haque, A. I. Minett and V. G. Gomes, J. Colloid Interface Sci., 2013, 410, 43–51.
3. K. R. Reddy, B. C. Sin, C. H. Yoo, W. Park, K. S. Ryu, J. S. Lee and Y. Lee, Scr. Mater., 2008, 58(11), 1010–1013.
4. S. H. Choi, D. H. Kim, A. V. Raghu, K. R. Reddy, H. I. Lee, K. S. Yoon and B. K. Kim, J. Macromol. Sci., Part B: Phys., 2012, 51(1), 197–207.
5. K. R. Reddy, M. Hassan and V. G. Gomes, Appl. Catal., A, 2015, 489, 1–16.
6. M. Hassan, K. R. Reddy, E. Haque, S. N. Faisal, S. Ghasemi, A. I. Minett and V. G. Gomes, Compos. Sci. Technol., 2014, 98, 1–8.
7. K. R. Reddy, H. M. Jeong, Y. Lee and A. V. Raghu, J. Polym. Sci., Part A: Polym. Chem., 2010, 48(7), 1477–1484.
8. K. R. Reddy, B. C. Sin, K. S. Ryu, J. C. Kim, H. Chung and Y. Lee, Synth. Met., 2009, 159(7), 595–603.
9. Y. R. Lee, S. C. Kim, H. I. Lee, H. M. Jeong, A. V. Raghu, K. R. Reddy and B. K. Kim, Macromol. Res., 2011, 19(1), 66–71.
10. K. R. Reddy, B. C. Sin, C. H. Yoo, D. Sohn and Y. Lee, JJ. Colloid Interface Sci., 2009, 340(2), 160–165.
11. M. Hassan, E. Haque, K. R. Reddy, A. I. Minett, J. Chen and V. G. Gomes, Nanoscale, 2014, 6(20), 11988–11994.
12. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. Dubonos, I. V. Grigorieva and A. A. Firsov, Science, 2004, 306, 666–669.
13. A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6, 183–191.
14. N. Li, S. Oida, G. S. Tulevski, S.-J. Han, J. B. Hannon, D. K. Sadana and T.-C. Chen, Nat. Commun., 2013, 4, 2294.
15. Y. Han, L. Zhang, X. Zhang, K. Ruan, L. Cui, Y. Wang, L. Liao, Z. Wang and J. Jie, J. Mater. Chem. C, 2014, 2, 201–207.
16. Z. Liu, P. You, S. Liu and F. Yan, ACS Nano, 2015, 9, 12026–12034.
17. S. B. Jo, H. H. Kim, H. Lee, B. Kang, S. Lee, M. Sim, M. Kim, W. H. Lee and K. Cho, ACS Nano, 2015, 9, 8206–8219.
18. W. H. Lee, J. Park, S. H. Sim, S. Lim, K. S. Kim, B. H. Hong and K. Cho, J. Am. Chem. Soc., 2011, 133, 4447–4454.
19. C. A. Di, D. Wei, G. Yu, Y. Liu, Y. Guo and D. Zhu, Adv. Mater., 2008, 20, 3289–3293.
20. S. Lee, G. Jo, S. J. Kang, G. Wang, M. Choe, W. Park and T. Lee, Adv. Mater., 2011, 23, 100–105.
21. S. J. Kang, G.-H. Lee, Y.-J. Yu, Y. Zhao, B. Kim, K. Watanabe, T. Taniguchi, J. Hone, P. Kim and C. Nuckolls, Adv. Funct. Mater., 2014, 24, 5157–5163.
22. J. Y. Choi, W. Kang, B. Kang, W. Cha, S. K. Son, Y. Yoon, H. Kim, Y. Kang, M. J. Ko, H. J. Son, K. Cho, J. H. Cho and B. Kim, ACS Appl. Mater. Interfaces, 2015, 7, 6002–6012.
23. J. S. Oh, K. N. Kim and G. Y. Yeom, J. Nanosci. Nanotechnol., 2014, 14, 1120–1133.
24. L. S. Panchokarla, K. S. Subrahmanyam, S. K. Saha, A. Govindaraj, H. R. Krishnamurthy, U. V. Waghmare and C. N. R. Rao, Adv. Mater., 2009, 21, 4726–4730.
25. D. C. Wei, Y. Q. Liu, Y. Wang, H. L. Zhang, L. P. Huang and G. Yu, Nano Lett., 2009, 9, 1752–1758.
26. Y. Kim, J. Ryu, M. Park, E. S. Kim, J. M. Yoo, J. Park, J. H. Kang and B. H. Hong, ACS Nano, 2014, 8, 868–874.
27. X. C. Dong, D. L. Fu, W. J. Fang, Y. M. Shi, P. Chen and L. J. Li, Small, 2009, 5, 1422–1426.
28. B. Lee, Y. Chen, F. Duerr, D. Mastrogiovanni, E. Garfunkel, E. Y. Andrei and V. Podzorov, Nano Lett., 2010, 10, 2427–2432.
29. C.-J. Shih, Q. H. Wang, Z. Jin, G. L. C. Paulus, C. D. Blankschtein, P. Jarillo-Herrero and M. S. Strano, Nano Lett., 2013, 13, 809–817.
30. S. K. Lee, J. W. Yang, H. H. Kim, S. B. Jo, B. Kang, H. Bong, H. C. Lee, G. Lee, K. S. Kim and K. Cho, ACS Nano, 2014, 8, 7968–7975.
31. D. Wang, X. Liu, L. He, Y. Yin, D. Wu and J. Shi, Nano Lett., 2010, 10, 4989–4993.
32. J. Meyer, P. R. Kidambi, B. C. Bayer, C. Weijtens, A. Kuhn, A. Centeno, A. Amaia Pesquera, A. Zurutuza, J. Robertson and S. Hofmann, Sci. Rep., 2014, 4, 5380.
33. T. Hasegawa, S. Miura, T. Moriyama, T. Kimura, I. Takaya, Y. Osato and H. Mizutani, Dig. Tech. Pap. - Soc. Inf. Disp. Int. Symp., 2004, 35, 154.
34. J. Huang, Z. Xu and Y. Yang, Adv. Funct. Mater., 2007, 17, 1966–1973.
35. H.-H. Liao, L.-M. Chen, Z. Xu, G. Li and Y. Yang, Appl. Phys. Lett., 2008, 92, 173303.
36. J. D. Lin, C. Han, F. Wang, R. Wang, D. Xiang, S. Qin and W. Chen, ACS Nano, 2014, 8, 5323–5329.
37. K. C. Kwon, K. S. Choi, B. J. Kim, J. L. Lee and S. Y. Kim, J. Phys. Chem. C, 2012, 116, 26586–26591.
38. Z. Chen, Y. Zheng, H. Yan and A. Facchetti, J. Am. Chem. Soc., 2009, 131, 8–9.
39. H. Yan, Z. Chen, Y. Zheng, C. Newman, J. R. Quinn, F. Dotz, M. Kastler and A. Facchetti, Nature, 2009, 457, 679–686.
40. M. Caironi, M. Bird, D. Fazzi, Z. Chen, R. Di Pietro, C. Newman, A. Facchetti and H. Sirringhaus, Adv. Funct. Mater., 2011, 21, 3371–3381.
41. X. Li, W. Cai, J. An, S. Kim, J. Nah, D. Yang, R. Piner, A. Velamakanni, I. Jung, E. Tutuc, S. K. Banerjee, L. Colombo and R. S. Ruoff, Science, 2009, 324, 1312–1314.
42. K. C. Kwon, K. S. Choi, C. Kim and S. Y. Kim, J. Phys. Chem. C, 2014, 118, 8187–8193.
43. A. C. Ferrari, J. C. Meyer, V. Scardaci, C. Casiraghi, M. Lazzeri, F. Mauri, S. Piscanec, D. Jiang, K. S. Novoselov, S. Roth and A. K. Geim, Phys. Rev. Lett., 2006, 97, 187401.
44. R. Lv, Q. Li, A. R. Botello-Méndez, T. Hayashi, B. Wang, A. Berkdemir, A. L. Hao, Q. Elías, R. Cruz-Silva, H. R. Gutiérrez, Y. A. Kim, H. Muramatsu, J. Zhu, M. Endo, H. Terrones, J.-C. Charlier, M. Pan and M. Terrones, Sci. Rep., 2012, 2, 586.
45. A. Das, S. Pisana, B. Chakraborty, S. Piscanex, S. K. Saha, U. V. Waghmare, K. S. Novoselov, H. R. Krishnamurthy, A. K. Geim, A. C. Ferrari and A. K. Sood, Nat. Nanotechnol., 2008, 3, 210–215.
46. A. Benayad, H.-J. Shin, H. K. Park, S.-M. Yoon, K. K. Kim, M. H. Jin, H.-K. Jeong, J. C. Lee, J.-Y. Choi and Y. H. Lee, Chem. Phys. Lett., 2009, 475, 91–95.
47. J.-H. Huang, J.-H. Fang, C.-C. Liu and C.-W. Chu, ACS Nano, 2011, 5, 6262–6271.
48. A. Nourbakhsh, M. Cantoro, A. Klekachev, F. Clemente, B. Soree, M. H. van der Veen, T. Vosch, A. Stesmans, B. Sels and S. De Gendt, J. Phys. Chem. C, 2010, 114, 6894–6900.
49. Y. B. Zhang, V. W. Brar, F. Wang, C. Girit, Y. Yayon, M. Panlasigui, A. Zettl and M. F. Crommie, Nat. Phys., 2008, 4, 627–630.
50. Y. J. Yu, Y. Zhao, S. Ryu, L. E. Brus, K. S. Kim and P. Kim, Nano Lett., 2009, 9, 3430–3434.
51. G. Pan, F. Chen, L. Hu, K. Zhang, J. Dai and F. Zhang, Adv. Funct. Mater., 2015, 25, 5126–5133.

---

Footnote

† Electronic supplementary information (ESI) available: EDS element analysis on graphenes, SR photoemission spectra of graphenes, AFM images of P(NDI2OD-T2) films on pristine graphene and SiO₂/Si, XRD patterns of P(NDI2OD-T2), and I_D–V_D curves of the OTFTs at various channel lengths. See DOI: 10.1039/c6ra20496a

---