Research Article www.acsami.org
Controlled Doping in Graphene Monolayers by Trapping Organic Molecules at the Graphene−Substrate Interface Pawan Kumar Srivastava, Premlata Yadav, Varsha Rani, and Subhasis Ghosh* School of Physical Sciences, Jawaharlal Nehru University, New Delhi 110067, India ABSTRACT: We report controlled doping in graphene monolayers through charge-transfer interaction by trapping selected organic molecules between graphene and underneath substrates. Controllability has been demonstrated in terms of shifts in Raman peaks and Dirac points in graphene monolayers. Under field effect transistor geometry, a shift in the Dirac point to the negative (positive) gate voltage region gives an inherent signature of n- (p-)type doping as a consequence of charge-transfer interaction between organic molecules and graphene. The proximity of organic molecules near the graphene surface as a result of trapping is evidenced by Raman and infrared spectroscopies. Density functional theory calculations corroborate the experimental results and also indicate charge-transfer interaction between certain organic molecules and graphene sheets resulting p- (n-)type doping and reveals the donor and acceptor nature of molecules. Interaction between molecules and graphene has been discussed in terms of calculated Mulliken charge-transfer and binding energy as a function of optimized distance. KEYWORDS: graphene, Raman spectroscopy, doping, DFT, charge transfer evidenced from Raman spectra.16 In particular substitution of boron in a graphitic framework has been studied14 extensively, but the intrinsic instability of these compounds against moisture and oxygen remains the key issue to be addressed. Moreover, we argue that these routes suffer from lack of controllability and tunability in terms of p-type and n-type doping. Several chemical procedures have also been used to obtain graphene by reducing graphene oxide (GO) thermally or chemically.8 These methods result in only p-type doping due to unwanted functionalities present in the ambient. Basically, there are two ways of covalent functionalization of MLG:8 (i) covalent bond between free radicals and carbon bonds of graphene and (ii) bond formation between oxygen (in GO) and organic functional groups. In this work, we present a feasible method to control the doping in MLG. Motivation of this work is based on several theoretical/experimental studies which suggest that graphene due to its monatomic thickness, could act as a very selective and permeable filtration membrane17−20 which blocks the entities with radii larger than 4.5 Å. Hence, if somehow any molecule can be trapped at the graphene−substrate interface, it would be difficult to remove that molecule even after annealing the samples at elevated temperatures. We have shown that such trapping of selected organic molecules between graphene and underneath the substrate could assist charge-transfer interaction between graphene and a trapped molecule. This interaction results in controlled n-type and p-type doping in MLG depending on whether the trapped molecule behaves as a donor or an
1. INTRODUCTION Monolayer graphene (MLG),1 a two-dimensional sp2-hybridized carbon, arranged in a honeycomb lattice, is the most intensively studied material. Due to several extraordinary properties2−4 numerous potential applications have been anticipated in fields of transparent,5 flexible,6 and ultrafast electronics.2 In spite of its several merits, the absence of band gap and difficulty to achieve controllable and reproducible doping in graphene, weaken its competitive strength in the area of electronic and sensing applications. Generally, graphene field effect transistors (FETs) show unwanted and unrestrained ptype doping in the ambient. Several pathways have been utilized to modulate the doping type in graphene such as adsorption of molecules,7 covalent functionalization,8 and by carbon atom replacement in a graphene matrix with dopant atoms.9 However, graphene doped by these methods are usually sensitive to the changes in the environment.7−9 Recently, Chang et al.10 doped graphene with nitrogen to get n-type behavior but MLG was heavily damaged as the highest mobility they could achieve was ∼1.0 cm2/(V s). Several attempts have been made to achieve controlled n-type doping by ion implantation of nitrogen11 and adsorption of hydrogen atoms12 on the graphene surface. The nitrogen atoms incorporated in graphene act as strong scattering centers, resulting in deterioration of the carrier mobility. Ho et al.9 recently demonstrated n-type doping of graphene with extended air stability; however electronic properties of graphene were severely modified due to the presence of traps at the interface. Several other methods13−16 which involve substitutional metallic (B, Sb, and so on) doping in graphene, have been employed to dope graphene, but these methods are prone to induce severe structural disorder in graphene layers as © 2017 American Chemical Society
Received: October 17, 2016 Accepted: January 17, 2017 Published: January 17, 2017 5375
DOI: 10.1021/acsami.6b13211 ACS Appl. Mater. Interfaces 2017, 9, 5375−5381
Research Article
ACS Applied Materials & Interfaces
function (631G-basis set). Three different graphene configurations were used as (i) 19 rings (54 carbon atoms), (ii) 37 rings (96 carbon atoms), and (iii) 61 rings (150 carbon atoms), where valences of the edge carbon atom were satisfied with hydrogen atom. The input was given with an organic molecule at the center of the graphene sheet without any bond formation (covalent or ionic). In addition to this trial structure, DFT calculations were also performed on some different geometry, whose details are provided in text.
acceptor. Controllability of the doping in MLG has been demonstrated by a shift in respective Raman peak positions and a shift in Dirac points (VD) which are due to charge-transfer interaction between organic molecules and graphene. To assist the charge transfer interaction, trapped molecules should be close enough to graphene surface. This concept has been evidenced by gate capacitance and infrared (IR) spectroscopic measurements and further supported by density functional theory (DFT) calculations which reveal the donor and acceptor nature of different solvent molecules resulting in n-type and ptype doping. Based on our results, we would also like to highlight the versatility of our method in terms of controllability. Recently21,22 deposition of thin atomic layers at the graphene−substrate interface was employed to achieve ntype or p-type doping in graphene layers. Dissanayake et al.23 have shown n-type doping in graphene using soda-lime glass as substrate via surface-transfer doping from sodium atoms which result in only n-type fixed doping. All these attempts are not suitable for controlled and reproducible doping in graphene. In this aspect, the method presented here is tunable and suitable for controlled p-type and n-type doping in MLG. Moreover, this approach is also highly suitable for FET geometry where nor p-type doping could be achieved just by trapping molecules between MLG and substrate. This work is segmented into three parts: the first part provides evidence of doping in MLG; the second part provides evidence of close proximity of solvent molecules to the graphene surface; and the third part demonstrates that the adsorption and donor/acceptor natures of organic molecules are revealed by DFT calculations.
3. RESULTS AND DISCUSSION To achieve MLG flakes, we have used chemical exfoliation method, which could assist trapping of molecules when graphene dispersion is transferred to substrate for further examinations (see Experimental Section). Raman spectroscopy has been used to characterize defects in as exfoliated graphene layers.24,25 There are several attempts to probe disorder in MLG by analyzing intensity ratios and shift in energetic positions of Raman peaks.24−27 Figure 1 shows Raman spectra
2. EXPERIMENTAL SECTION 2.1. Synthesis and Characterization. Graphene layers were isolated by sonicating HOPG purchased from Sigma-Aldrich, in various organic solvents as mentioned in the text with initial concentration and total sonicated volume of 0.06 mg/mL and 30 mL, respectively. In this study we have used toluene, chlorobenzene, acetonitrile, acetone, N,N- dimethylformamide or DMF, and propylene carbonate or PC for exfoliation of HOPG. This exfoliation technique contains two stages: (i) sonication of HOPG in solvents (8−12 h) followed by (ii) centrifugation (2 h) in order to separate the thicker flakes. Sonication was done using ultrasonic bath (500 W). After sonication, subsequent dispersion was centrifuged using a swinging head centrifuge. After centrifugation, a solution comprising thin graphene layers was dropped onto a SiO2/Si substrate. Preliminary sample drying was done at 300 K at a base pressure of ∼10−3 mbar; then dried samples were subsequently annealed at 200 °C for 30 min under vacuum. Molecular trapping as discussed in the Introduction is facilitated by the transfer of graphene layers from the solution to solid substrates (SiO2) where the residual solvent from the exfoliation process remains at the graphene−substrate interface even after annealing at higher temperatures (will be discussed later). Raman spectroscopy was performed using a WITec GmbH Alpha 300 confocal Raman microscope (excitation wavelength, 532 nm; laser power, 5 mW; spatial resolution, 300 nm). FETs were fabricated on SiO2 (300 nm)/Si (n+2) substrates in bottom gate geometry with lithographically defined top metallic (Au/45 nm) electrodes (source/ drain) with aspect ratio (W/L) of 3. A vast majority of devices contain monolayer graphene flakes, with sizes 2−20 μm. Electrical characterizations were performed in vacuum (10−3 mbar). IR spectroscopy (excitation wavelength ∼ 1064 nm) on graphene samples supported on SiO2/Si substrates was carried out using a Varian-7000 UMA-600 IR microscope by inspecting individual graphene flakes. 2.2. Computation of Minimum Energy Structure. DFT calculation has been employed to obtain energetically optimized structure and interaction between graphene and trapped molecules. Calculations were performed using Lee−Yang−Parr correlation
Figure 1. Raman spectra of MLG exfoliated in (from bottom to top) toluene, chlorobenzene, acetone, DMF, and PC. In toluene and chlorobenzene, a small D peak intensity has been observed. In contrast, considerable D band intensity in the cases of acetone, DMF, and PC can be seen. A monotonic red shift in 2D Raman peak positions can be observed with an increase in dielectric constant (k) of the solvents used for exfoliation. This red shift in Raman peak positions in the cases of g-acetone, g-DMF and g-PC is an indication of doping.
of MLG exfoliated in toluene, chlorobenzene, acetone, DMF, and PC. It is observed that Raman spectra of graphene exfoliated in toluene (g-toluene) and chlorobenzene (gchlorobenzene) show negligible D band (∼1350 cm−1) intensity with G and 2D peaks around 1580 and 2695 cm−1, respectively. On the contrary, graphene exfoliated in acetone (g-acetone), DMF (g-DMF), and PC (g-PC) exhibit significant D band intensities with G and 2D peak positions around 1570 and 2640 cm−1, respectively. We have consistently observed this shift in Raman peaks in more than 50 MLG samples prepared in different dielectric environments. It will be shown that the red shift in Raman peak and considerable Raman D peak intensity in case of the doped graphene are interrelated. It is worth mentioning that Coulomb impurities away from the graphene plane such as intercalants and substrate charged impurities have a small contribution to disorder induced Raman D peak.26,27 Hence, substantial D peak intensity corroborates the presence of lattice perturbations that would result in lattice relaxation and subsequently hardening/softening of the phonons. This phenomenon can be seen as the shift in Raman G and 2D modes of MLG samples. We have extensively studied the comparative shift in various MLG samples prepared 5376
DOI: 10.1021/acsami.6b13211 ACS Appl. Mater. Interfaces 2017, 9, 5375−5381
Research Article
ACS Applied Materials & Interfaces in various solvents, and examples are given in Figure 2. It is apparent that maxima of the G peak fall around 1580 and 1570
Figure 3. Transfer characteristics of field effect transistors fabricated using g-acetonitrile, g-chlorobenzene, g-toluene, g-acetone, g-DMF, and g-PC (in the sequence from p-region to n-region). In g-acetonitrile, characteristics show a Dirac point in the positive gate bias (+5 V) region; g-toluene and g-chlorobenzene characteristics show Dirac points close to zero gate bias. In g-acetone, g-DMF, and g-PC, Dirac points shifted toward the higher negative gate bias region indicating a transition from p-type to n-type doping in MLG. Model schematic and optical image of one of the graphene FETs are also given.
DMF, and g-PC exhibit asymmetry around VD which is shifted to negative gate voltage indicating n-type doping in MLG. Hence, we have convincingly shown that graphene samples with red/blue shift in Raman 2D peaks also exhibit VD in the negative/positive gate bias region, and this demonstrates excellent agreements in results obtained from Raman spectra and electrical measurements. In this case, asymmetry in electron−hole conduction due to doping leads to a large shift in VD.29 This kind of asymmetry around a Dirac point and its correlation with doping in MLG has been theoretically proposed by Novikov,30 suggesting that the difference in the scattering cross-section for relativistic electrons and holes in graphene is responsible for asymmetric conduction in MLG. This asymmetry should not be observed in systems in which electrons or holes are governed by nonrelativistic quantum mechanics. However, in the case of systems in which electrons or holes are governed by relativistic quantum mechanics, the scattering cross section depends on the sign of the scattering potential (donor or acceptor) or sign of the carriers (electrons or holes). In this scenario, the presence of donor/acceptor impurities which scatter electrons/holes more efficiently resulting in asymmetric conduction around the Dirac point. Essentially, it is possible to decide the nature of the dopants from the asymmetry. Moreover, the dependence of electrical and Raman signatures on the choice of the solvents has been further explored by examining the evolution in carrier concentration (n) and Fermi energy (EF). Figure 4 shows the evolution in n for different MLG samples. In MLG, n has been calculated using the relation n = CgVD/e, where Cg and e are the gate capacitance and electronic charge, respectively. In particular, g-toluene and g-chlorobenzene show n ∼ 1011 cm−2, whereas, in g-acetonitrile, g-acetone, g-DMF, and g-PC, n increases up to ∼1012−1013 cm−2. Now let us discuss the
Figure 2. Histograms representing occurrences of Raman G and 2D peak positions of graphene flakes prepared in (a, b) PC, (c, d) DMF, (e, f) acetone, (g, h) chlorobenzene, and (i, j) toluene. It is evident that the maxima of the G peak fall near 1580 and 1570 cm−1 for graphene prepared in low-k and high-k solvents, respectively. Similarly, maxima for 2D peak falls near 2690 cm−1 and 2665 ± 5 cm−1 for graphene prepared in low-k and high-k solvents, respectively. Here k (dielectric constant) is given just to substantiate different organic molecules (low-k solvents, chlorobenzene and toluene; high-k solvents, acetone, DMF, and PC).
cm−1 for graphene prepared in g-toluene/g-chlorobenzene and g-DMF/g-PC, respectively. Maxima for Raman 2D peak are around 2690 cm−1 and 2665 ± 5 cm−1 for MLG prepared in gtoluene/g-chlorobenzene and g-DMF/g-PC, respectively. Monotonic shift in the 2D peak position toward lower wavenumber suggests softening of phonon modes which is attributed to ntype doping in MLG.24,27 To corroborate the Raman results discussed in the previous section, we have carried out electrical measurement on graphene based FETs. Figure 3 shows the transfer characteristics of FETs based on MLG. Two main observations should be pointed out here: (i) shift in VD and (ii) deviation from symmetry around VD, in MLG prepared in different solvents. In particular, g-toluene, g-chlorobenzene, and g-acetonitrile show symmetry around VD which varies between +1.0 and +5.0 V. Positions of the VD and symmetric electron/hole conduction in MLG provide a clear signature that these MLG are either pristine or extremely low doped.28,29 In contrast, g-acetone, g5377
DOI: 10.1021/acsami.6b13211 ACS Appl. Mater. Interfaces 2017, 9, 5375−5381
Research Article
ACS Applied Materials & Interfaces
functionalization of MLG, one would expect poor μ similar to graphene oxide10,32 but this is not the case as we have achieved μ ∼ 104 cm2/(V s).33,34 Covalent attachment of molecules (with a graphene surface) is feasible only when CC bonds in graphene are disrupted resulting in poor μ. Hence chargetransfer interaction due to adsorption of solvent molecules on MLG surface is most probable mechanism for doping in MLG samples. Now we point out that trapping of these organic molecules at the graphene−substrate interface is the only way to have these molecules in close proximity to the MLG surface;otherwise, in the absence of covalent attachment, molecules will evaporate completely at annealing temperature of 200 °C (see Experimental Section). To elucidate the close proximity of molecules to the graphene surface experimentally, we have acquired IR spectra of MLG prepared in different solvents. A particular example is given in Figure 5 where IR
Figure 4. (a) Variation in carrier (electron and hole) density for MLG samples prepared in various solvents (where, for example, 1.00E+14 represents 1.00 × 1014). Red and blue bars denote electron and hole densities, respectively. (b) ID/IG as a function of electron and hole density. Empty circles (in blue) and filled circles (in red) denote hole and electron densities, respectively.
Figure 5. Infrared spectra of chlorobenzene/g- chlorobenzene and DMF/g-DMF.
spectra of chlorobenzene/g-chlorobenzene and DMF/g-DMF are compared. In chlorobenzene, four IR peaks could be observed at 1035, 1490, and 1600 cm−1 and between 2900 and 3000 cm−1. The peak at 1033 cm−1 resembles to C−H bending vibration of the alkyl group attached to the graphitic ring; peaks at 1485 and 1608 cm−1 signify in plane graphitic C−C stretching and other peaks suggest in-plane alkyl and aromatic C−H vibration, respectively.35−37 In g- chlorobenzene, only CC vibration at 1635 cm−1 can be observed and all other vibrational modes are absent. Slight shifts in CC vibrations have also been observed in IR spectra of g- chlorobenzene as compared to chlorobenzene. This could be due to the different aromatic structures of chlorobenzene and graphene (hydrogen less aromatic ring). It can be concluded from IR spectra that gchlorobenzene is undoped. In DMF, four significant peaks were observed at 1200 cm−1, 1350 cm−1, 1600 and 3000 cm−1. All these vibrational modes are inherent features of a DMF molecule,35−38 whereas the peak at 2350 cm−1 corresponds to the vibrational signature of the CO2 molecule. In the case of gDMF, out of plane C−H bending and wagging modes (peaks between 1200 and 1400 cm−1) were entirely absent and could be attributed to the supporting substrate, which would restrict these vibrations. Moreover, CC vibrational mode (signature of planar vibration of carbon atoms in graphitic lattice) at 1600 cm−1 in g-DMF could be seen and as expected is absent in bare DMF. In addition to in plane vibration of carbon atoms in graphitic lattice, vibrations of CO, C−OH, and so on are also present in g-DMF. A substantial shift in CO vibrational mode (Δω ∼ 45−50 cm−1) has been observed which is attributed to the close proximity of DMF molecules to the graphitic surface. This tenability (n-type or p-type) can be explained if we could find out the donor/acceptor nature of the solvent molecules. In order to settle this issue, we have performed DFT calculations.39
evolution of Raman peaks in the perspective of doping in MLG. Especially in MLG, this evolution could be defined in terms of G, 2D, and D peaks. Figure 4 shows the monotonic evolution in intensity ratios of D to G peaks (ID/IG) as a function of n for MLG samples. For g-chlorobenzene and g-toluene, we get ID/IG ∼ 0.002, indicating negligible defect density in MLG samples, whereas we have observed an increase in ID/IG ∼ 0.4 for g-PC. This increase in ID/IG has already been observed by several groups24,27,31 and accounts for doping in MLG samples. A logarithmic plot between ID/IG and n also indicates that marginal disorder in MLG varies linearly with doping concentration. It corroborates that trapped molecules are close to the MLG surface so that it has substantial effect on MLG resulting Raman D peak. EF was calculated using the relation26EF = ℏυf πn , where υf is the Fermi velocity of carriers (∼10 6 m/s) in MLG. For g-toluene and gchlorobenzene, EF has been estimated to be ∼50 meV whereas for g-acetone, g-DMF, and g-PC, it increases beyond ∼200 meV. Thus, relative shift in Fermi level is attributed to the choice of the appropriate solvents used for exfoliation of MLG. Raman analysis in connection with observed FET behavior (position of VD) unambiguously demonstrates the modulation in carrier type and scaling of doping level in MLG grown in different dielectric environments. In view of our finding discussed in previous sections, apparently the role of the solvents is an extremely important factor in deciding the doping level in MLG. In this context, we have two possible routes to substantiate the doping mechanism in MLG: (i) covalent attachment of molecules on the MLG surface and (ii) charge-transfer interaction due to selective adsorption of molecules on MLG surface. Doping due to covalent functionalization of MLG with molecules has been ruled out due to the following reasons: in the case of covalent 5378
DOI: 10.1021/acsami.6b13211 ACS Appl. Mater. Interfaces 2017, 9, 5375−5381
Research Article
ACS Applied Materials & Interfaces
Figure 6. (a) Trail and (b) optimized geometries for adsorption of various solvent molecules onto the graphene sheet (top view). From top to bottom: g-acetonitrile, g-toluene, g-chlorobenzene, g-DMF, and g-PC. (c) Charge density maps corresponding to optimized geometries (scale bar is given in terms of Milliken charge transfer). Equilibrium distance and amount of charge transferred between various adsorbed molecules and graphene are also given. Red and blue (artificial overlaid) surfaces of graphene in the optimized geometry signify n-type and p-type doping, respectively. Color coding of atoms: dark gray atom, carbon; light gray atom, hydrogen; red atom, oxygen; green atom, chlorine; blue atom, nitrogen.
To find an energetically optimized geometry of graphene with different organic molecules, interaction between graphene and the molecules were estimated using DFT (for method details, see Experimental Section) with a graphene layer with 150 carbon atoms (Figure 6). Our simulation suggests that chlorobenzene stabilizes at a distance d > 4.5 Å, whereas toluene, acetonitrile, acetone, DMF, and PC stabilize at 3.9, 3.6, 3.9, 3.5, and 3.3 Å, respectively, in the optimized geometry. On the basis of simulation results shown in Figure 6, we now have two categories of solvent molecules: (i) acceptor molecules (toluene, chlorobenzene, and acetonitrile) which attract an electron cloud from graphene and (ii) donor molecules (acetone, DMF, and PC) which donate electron cloud to graphene. In both cases (p-type and n-type doping), the fraction of transferred charge depends monotonically on the equilibrium distance between molecules and graphene. Molecules which are closer the graphene’s basal plane will interact more strongly with graphene, and it would be reflected in the resultant charge transfer. Figure 7 summarizes the variation in the amount of charge transferred between
Figure 7. Amount of charge transferred between adsorbed molecules and graphene as a function of their respective equilibrium distance. Binding energy of adsorbed molecules with graphene is also given. Red and blue bars denote n-type and p-type doping, respectively.
molecules and graphene as a function of equilibrium distance. We have also calculated the adsorption energy (EAD) of 5379
DOI: 10.1021/acsami.6b13211 ACS Appl. Mater. Interfaces 2017, 9, 5375−5381
Research Article
ACS Applied Materials & Interfaces
of Raman peaks for various graphene samples provide preliminary evidence of doping, whereas a shift in the energetic position of the Raman 2D peak in addition to substantial shift in the Dirac point to positive/negative gate voltage formally represent the p- or n-type doping. The doping mechanism is further reconciled with DFT results which reveal the donor and acceptor nature of different solvent molecules resulting in ntype and p-type doping. We also believe that in context of versatility and control over p-type and n-type doping, this method is invulnerable as compared to many other used techniques.41−44 We believe that observed controllability of doping in MLG would open up a new pathway for graphene based electronics.
energetically optimized geometry (methodology is described in ref 33 and39) to further substantiate the interaction of organic molecules with graphene. It is worth mentioning that, for thermodynamic equilibrium, negative adsorption energy is required. In the case of toluene, chlorobenzene, and acetonitrile, EAD turned out to be positive (0.071−0.1 eV) and for acetone, DMF, and PC, it turned out to be negative (−1.6 eV < EAD < −2.48 eV) for energetically relaxed geometries. It suggests that acetone, DMF, and PC molecules form a stable geometry with graphene and would be close enough to the graphitic surface to have substantial interaction with graphene’s basal plane, whereas molecules with positive EAD would not be in thermodynamic equilibrium and will exhibit little interaction due to their relatively large distance from graphene. These theoretical findings are in agreement with experimental results and corroborate the p-type and n-type doped MLG prepared in different solvents. We have seen that trapping of a particular solvent leads to por n-type doping in graphene. We emphasize that this type of doping is basically attributed to the electrophilic and nucleophilic character of trapped molecules.40 As shown in Figure 8, acetone, DMF, and PC have a dense electron cloud
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Subhasis Ghosh: 0000-0002-0825-3921 Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We thank AIRF, JNU for providing IR facility. Hitesh Mamgain, WITec GmbH, Germany is kindly acknowledged for providing help in confocal Raman measurements. P.K.S., P.Y., and V.R. thank CSIR and UGC, Government of India for financial support through a fellowship.
■
Figure 8. Charge distribution map for various organic molecules as labeled in the figure.
REFERENCES
(1) Geim, A. K.; Novoselov, K. S. The Rise of Graphene. Nat. Mater. 2007, 6, 183−191. (2) Xia, F.; Mueller, T.; Lin, Y. M.; Valdes-Garcia, A.; Avouris, P. Ultrafast Graphene Photodetector. Nat. Nanotechnol. 2009, 4, 839− 843. (3) Nair, R. R.; Blake, P.; Grigorenko, A. N.; Novoselov, K. S.; Booth, T. J.; Stauber, T.; Peres, N. M. R.; Geim, A. K. Fine Structure Constant Defines Visual Transparency of Graphene. Science 2008, 320, 1308− 1308. (4) Ranjbartoreh, A. R.; Wang, B.; Shen, X.; Wang, G. Advanced Mechanical Properties of Graphene Paper. J. Appl. Phys. 2011, 109, 014306. (5) Kim, K. S.; Zhao, Y.; Jang, H.; Lee, S. Y.; Kim, J. M.; Kim, K. S.; Ahn, J. H.; Kim, P.; Choi, J. Y.; Hong, B. H. Large-Scale Pattern Growth of Graphene Films for Stretchable Transparent Electrodes. Nature 2009, 457, 706−710. (6) Xu, Y.; Bai, H.; Lu, G.; Li, C.; Shi, G. Flexible Graphene Films via the Filtration of Water-Soluble Noncovalent Functionalized Graphene Sheets. J. Am. Chem. Soc. 2008, 130, 5856−5857. (7) Schedin, F.; Geim, A. K.; Morozov, S. V.; Hill, E. W.; Blake, P.; Katsnelson, M. I.; Novoselov, K. S. Detection of Individual Gas Molecules Adsorbed on Graphene. Nat. Mater. 2007, 6, 652−655. (8) Georgakilas, V.; Otyepka, M.; Bourlinos, A. B.; Chandra, V.; Kim, N.; Kemp, K. C.; Hobza, P.; Zboril, R.; Kim, K. S. Functionalization of Graphene: Covalent and Non-Covalent Approaches, Derivatives and Applications. Chem. Rev. 2012, 112, 6156−6214. (9) Ho, P. H.; et al. Self-Encapsulated Doping of n-Type Graphene Transistors with Extended Air Stability. ACS Nano 2012, 6, 6215− 6221. (10) Chang, D. W.; Lee, E. K.; Park, E. Y.; Yu, H.; Choi, H. J.; Jeon, I. Y.; Sohn, G. J.; Shin, D.; Park, N.; Oh, J. H.; Dai, L.; Baek, J. B. Nitrogen-Doped Graphene Nanoplatelets from Simple Solution EdgeFunctionalization for n-Type Field-Effect Transistors. J. Am. Chem. Soc. 2013, 135, 8981−8988.
surrounding the oxygen atom. Hence due to the nucleophilic character of these molecules, a small fraction of this electron cloud would be transferred to graphene to make equilibrium geometry, whereas, due to the electrophilic character of the rest of the three molecules, an electron cloud from graphene would be transferred to the molecules to create equilibrium. For instance, acetone, DMF, and PC have a lone pair of electrons (on oxygen), which they can donate to graphene, making graphene n-type doped. Moreover, optimized geometries of these organic molecules with graphene rings suggest that the oxygen atom (of organic solvents) is pointing toward graphene, and this scenario is favorable for electron transfer from molecules to graphene. In contrast, due to the electrophilic character40 of toluene, chlorobenzene, and acetonitrile, charge is being transferred from graphene to molecules, leaving graphene marginally p-type doped as evidenced in experiments and DFT calculations as well.
4. CONCLUSION We have demonstrated an efficient and controlled method to dope monolayer graphene. It has been shown that doping of MLG can be achieved by charge-transfer interaction between organic molecules and graphene. This has been possible by trapping selective organic molecules between graphene and the underlying substrate. Hence, charge-transfer interaction between molecules and graphene which is the key factor of controllable doping is facilitated by close proximity of molecules to the graphene surface. Evolution in intensity ratios 5380
DOI: 10.1021/acsami.6b13211 ACS Appl. Mater. Interfaces 2017, 9, 5375−5381
Research Article
ACS Applied Materials & Interfaces (11) Geng, D.; Yang, S.; Zhang, Y.; Yang, J.; Liu, J.; Li, R.; Sham, T. K.; Sun, X.; Ye, S.; Knights, S. Nitrogen Doping eEfects on the Structure of Graphene. Appl. Surf. Sci. 2011, 257, 9193−9198. (12) Kim, B. H.; Hong, S. J.; Baek, S. J.; Jeong, H. Y.; Park, N.; Lee, M.; Lee, S. W.; Park, M.; Chu, S. W.; Shin, H. S.; Lim, J.; Lee, J. C.; Jun, Y.; Park, Y. W. N-type Graphene Induced by Dissociative H2 Adsorption at Room Temperature. Sci. Rep. 2012, 2, 690. (13) Jeon, I.-Y.; Choi, M.; Choi, H.-J.; Jung, S.-M.; Kim, M.-J.; Seo, J.M.; Bae, S.-Y.; Yoo, S.; Kim, G.; Jeong, H. Y.; Park, N.; Baek, J.-B. Antimony-Doped Graphene Nanoplatelets. Nat. Commun. 2015, 6, 7123. (14) Kawai, S.; Saito, S.; Osumi, S.; Yamaguchi, S.; Foster, A. S.; Spijker, P.; Meyer, E. Atomically Controlled Substitutional Borondoping of Graphene Nanoribbons. Nat. Commun. 2015, 6, 8098. (15) Liu, Y.; Shen, Y.; Sun, L.; Li, J.; Liu, C.; Ren, W.; Li, F.; Gao, L.; Chen, J.; Liu, F.; Sun, Y.; Tang, N.; Cheng, H. M.; Du, Y. Elemental Super Doping of Graphene and Carbon Nanotubes. Nat. Commun. 2016, 7, 10921. (16) Yeom, D.-Y.; Jeon, W.; Tu, N. D. K.; Yeo, S. Y.; Lee, S. S.; Sung, B. J.; Chang, H.; Lim, J. A.; Kim, H. High-Concentration Boron Doping of Graphene Nanoplatelets by Simple Thermal Annealing and Their Supercapacitive Properties. Sci. Rep. 2015, 5, 9817. (17) Surwade, S. P.; Smirnov, S. N.; Vlassiouk, I. V.; Unocic, R. R.; Veith, G. M.; Dai, S.; Mahurin, S. M. Water Desalination Using Nanoporous Single-Layer Graphene. Nat. Nanotechnol. 2015, 10, 459− 464. (18) Joshi, R. K.; Carbone, P.; Wang, F. C.; Kravets, V. G.; Su, Y.; Grigorieva, I. V.; Wu, H. A.; Geim, A. K.; Nair, R. R. Precise and Ultrafast Molecular Sieving Through Graphene Oxide Membranes. Science 2014, 343, 752−754. (19) Jiang, D.; Cooper, V. R.; Dai, S. Porous Graphene as the Ultimate Membrane for Gas Separation. Nano Lett. 2009, 9, 4019− 4024. (20) Sint, K.; Wang, B.; Kral, P. Selective Ion Passage through Functionalized Graphene Nanopores. J. Am. Chem. Soc. 2008, 130, 16448−16449. (21) Kim, H.; Kim, H. H.; Jang, J. I.; Lee, S. K.; Lee, G. W.; Han, J. T.; Cho, K. Doping Graphene with an Atomically Thin Two Dimensional Molecular Layer. Adv. Mater. 2014, 26, 8141−8146. (22) Zheng, L.; Cheng, X.; Wang, Z.; Xia, C.; Cao, D.; Shen, L.; Wang, Q.; Yu, Y.; Shen, D. Reversible n-type Doping of Graphene by H2O-Based Atomic-Layer Deposition and Its Doping Mechanism. J. Phys. Chem. C 2015, 119, 5995−6000. (23) Dissanayake, D. M. N. M.; Ashraf, A.; Dwyer, D.; Kisslinger, K.; Zhang, L.; Pang, Y.; Efstathiadis, H.; Eisaman, M. D. Spontaneous and Strong Multi-Layer Graphene n-Doping on Soda-Lime Glass and Its Application in Graphene-Semiconductor Junctions. Sci. Rep. 2016, 6, 21070. (24) Ferrari, A. C.; Robertson, J. Interpretation of Raman Spectra of Disordered and Amorphous Carbon. Phys. Rev. B: Condens. Matter Mater. Phys. 2000, 61, 14095−14107. (25) Ferrari, A. C.; Meyer, J. C.; Scardaci, V.; Casiraghi, C.; Lazzeri, M.; Mauri, F.; Piscanec, S.; Jiang, D.; Novoselov, K. S.; Roth, S.; Geim, A. K. Raman Spectrum of Graphene and Graphene Layers. Phys. Rev. Lett. 2006, 97, 187401. (26) Chen, C. F.; Park, C. H.; Boudouris, B. W.; Horng, J.; Geng, B.; Girit, C.; Zettl, A.; Crommie, M. F.; Segalman, R. A.; Louie, S. G.; Wang, F. Controlling Inelastic Light Scattering Quantum Pathways in Graphene. Nature 2011, 471, 617−620. (27) Maultzsch, J.; Reich, S.; Thomsen, C. Double-resonant Raman Scattering in Graphite: Interference Effects, Selection Rules, and Phonon Dispersion. Phys. Rev. B: Condens. Matter Mater. Phys. 2004, 70, 155403. (28) Mayorov, A. S.; Mayorov, A. S.; Elias, D. C.; Mukhin, I. S.; Morozov, S. V.; Ponomarenko, L. A.; Novoselov, K. S.; Geim, A. K.; Gorbachev, R. V. How Close Can One Approach the Dirac Point in Graphene Experimentally? Nano Lett. 2012, 12, 4629−4634.
(29) Chen, J. H.; Jang, C.; Adam, S.; Fuhrer, M. S.; Williams, E. D.; Ishigami, M. Charged-Impurity Scattering in Graphene. Nat. Phys. 2008, 4, 377−381. (30) Novikov, D. Numbers of Donors and Acceptors from Transport Measurements in Graphene. Appl. Phys. Lett. 2007, 91, 102102. (31) Das, A.; Pisana, S.; Chakraborty, B.; Piscanec, S.; Saha, S. K.; Waghmare, U. V.; Novoselov, K. S.; Krishnamurthy, H. R.; Geim, A. K.; Ferrari, A. C.; Sood, A. K. Monitoring Dopants by Raman Scattering in an Electrochemically Top-Gated Graphene Transistor. Nat. Nanotechnol. 2008, 3, 210−215. (32) Kobayashi, T.; Kimura, N.; Chi, J.; Hirata, S.; Hobara, D. Channel-Length-Dependent Field-Effect Mobility and Carrier Concentration of Reduced Graphene Oxide Thin-Film Transistors. Small 2010, 6, 1210−1215. (33) Srivastava, P. K.; Yadav, P.; Ghosh, S. Dielectric Environment as a Factor to Enhance the Production Yield of Solvent Exfoliated Graphene. RSC Adv. 2015, 5, 64395−64403. (34) Srivastava, P. K.; Ghosh, S. Defect Engineering as a Varsatile Route to Estimate Various Scattering Mechanisms in Graphene on Solid Substrates. Nanoscale 2015, 7, 16079−16086. (35) Xu, Y.; Liu, Z.; Zhang, X.; Wang, Y.; Tian, J.; Huang, Y.; Ma, Y.; Zhang, X.; Chen, Y. A Graphene Hybrid Material Covalently Functionalized with Porphyrin: Synthesis and Optical Limiting Property. Adv. Mater. 2009, 21, 1275−1279. (36) Sahoo, S.; Hatui, G.; Bhattacharya, P.; Dhibar, S.; Das, C. K. One Pot Synthesis of Graphene by Exfoliation of Graphite in ODCB. Graphene 2013, 02, 42−48. (37) Hsu, C.-P. S. In Handbook of Instrumental Techniques for Analytical Chemistry; Settle, F. A., Ed.; Prentic Hall: Upper Saddle River, NJ, USA, 1997; pp 247−284. (38) Galande, C.; Mohite, A. D.; Naumov, A. V.; Gao, W.; Ci, L.; Ajayan, A.; Gao, H.; Srivastava, A.; Weisman, R. B.; Ajayan, P. M. Quasi-Molecular Fluorescence from Graphene Oxide. Sci. Rep. 2011, 1, 85. (39) Li, W.; Wang, Y. B.; Pavel, I.; Ye, Y.; Chen, Z. P.; Luo, M. D.; Hu, J. M.; Kiefer, W. DFT and HF Studies of the Geometry, Electronic Structure, and Vibrational Spectra of 2-Nitrotetraphenylporphyrin and Zinc 2-Nitrotetraphenylporphyrin. J. Phys. Chem. A 2004, 108, 6052− 6058. (40) Clayden, J.; Greeves, N.; Warren, S. Organic Chemistry (Oxford University Press: Oxford, U.K., 2012; Chapters 5 and 6. (41) Kawai, S.; Saito, S.; Osumi, S.; Yamaguchi, S.; Foster, A. S.; Spijker, P.; Meyer, E. Atomically Controlled Substitutional Boron Doping of Graphene Nanoribbons. Nat. Commun. 2015, 6, 8098. (42) Sforzini, J.; Hapala, P.; Franke, M.; van Straaten, G.; Stöhr, A.; Link, S.; Soubatch, S.; Jelínek, P.; Lee, T.-L.; Starke, U.; Švec, M.; Bocquet, F. C.; Tautz, F. S. Structural and Electronic Properties of Nitrogen-Doped Graphene. Phys. Rev. Lett. 2016, 116, 126805. (43) Joucken, F.; Tison, Y.; Le Fèvre, P.; Tejeda, A.; Taleb-Ibrahimi, A.; Conrad, E.; Repain, V.; Chacon, C.; Bellec, A.; Girard, Y.; Rousset, S.; Ghijsen, J.; Sporken, R.; Amara, H.; Ducastelle, F.; Lagoute, J. Charge Transfer and Electronic Doping in Nitrogen-Doped Graphene. Sci. Rep. 2015, 5, 14564. (44) Willke, P.; Amani, J. A.; Sinterhauf, A.; Thakur, S.; Kotzott, T.; Druga, T.; Weikert, S.; Maiti, K.; Hofsass, H.; Wenderoth, M. Doping of Graphene by Low-Energy Ion Beam Implantation: Structural, Electronic, and Transport Properties. Nano Lett. 2015, 15, 5110− 5115.
5381
DOI: 10.1021/acsami.6b13211 ACS Appl. Mater. Interfaces 2017, 9, 5375−5381
